Sudden Death in Epilepsy: Forensic and Clinical Issues

  • 49 176 5
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

Sudden Death in Epilepsy: Forensic and Clinical Issues

SUDDEN DEATH in EPILEPSY FORENSIC AND CLINICAL ISSUES SUDDEN DEATH in EPILEPSY FORENSIC AND CLINICAL ISSUES EDITED BY

1,533 175 11MB

Pages 1020 Page size 492 x 744.2 pts Year 2012

Report DMCA / Copyright

DOWNLOAD FILE

Recommend Papers

File loading please wait...
Citation preview

SUDDEN DEATH in EPILEPSY FORENSIC AND CLINICAL ISSUES

SUDDEN DEATH in EPILEPSY FORENSIC AND CLINICAL ISSUES EDITED BY

CLAIRE M. LATHERS PAUL L. SCHR AEDER MICHAEL W. BUNGO JAN E. LEESTMA

Boca Raton London New York

CRC Press is an imprint of the Taylor & Francis Group, an informa business

CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 © 2011 by Taylor and Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S. Government works Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1 International Standard Book Number-13: 978-1-4398-0223-6 (Ebook-PDF) This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint. Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http:// www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com

To (DS) a 4-year old boy with a history of nocturnal seizures who found hidden Christmas presents, including the one he wanted most of all, a remote control car two weeks before Christmas. The next morning he was found dead in bed. He and other victims of SUDEP challenge all of us to find preventive measures as quickly as possible. (CML) To Carol and Richard Lathers and to Marcel J. Lajoy for their love, support, and words of wisdom while completing this endeavor. (CML) To two university students (MB) and (AG) who were victims of SUDEP and to their courageous families who helped to found Epilepsy Bereaved in the United Kingdom. (PLS) To my patient wife Barbara and daughters Maria and Ellen who dealt with my many hours of editorial distraction with forbearance, encouragement and love. (PLS) To my grandson, Lucas, who missed quality time, so that this endeavor could be completed. (MWB)

Table of Contents

Foreword by Samuels Foreword by Schachter Preface Acknowledgments Editors Contributors

xv xvii xxi xxv xxvii xxxi

Section I Forensics of Sudden Death

1

Neurocardiologic Mechanistic Risk Factors in Sudden Unexpected Death in Epilepsy

3

Claire M. Lathers, Paul L. Schraeder, and Michael W. Bungo

2

Forensic Considerations and Sudden Unexpected Death in Epilepsy

37

Jan E. Leestma

3

Omega-3 Fatty Acids in Sudden Unexpected Death in€Epilepsy: Guardian of the Brain–Heart Connection

57

Fulvio A. Scorza, Esper A. Cavalheiro, Ricardo M. Arida, Vera C. Terra, Carla A. Scorza, Eliza Y. F. Sonoda, and Roberta M. Cysneiros

4

Unanswered Questions: SUDEP Studies Needed

67

Claire M. Lathers, Paul L. Schraeder, and Michael W. Bungo

5

Medullary Serotonergic Abnormalities in Sudden Infant Death Syndrome: Implications in SUDEP David S. Paterson

vii

77

viii Table of Contents

6

Forensic Case Identification

95

Paul L. Schraeder, Elson L. So, and Claire M. Lathers

7

Sudden Unexpected Death in Epilepsy: Future Research Directions

109

Simona Parvulescu-Codrea

8

Forensic Postmortem Examination of Victims of Sudden Unexpected Death in Epilepsy

131

Claire M. Lathers, Paul L. Schraeder, Steven A. Koehler, and Cyril H. Wecht

9

One-Year Postmortem Forensic Analysis of Deaths in Persons with Epilepsy

145

Steven A. Koehler, Paul L. Schraeder, Claire M. Lathers, and Cyril H. Wecht

10

Drug Abuse and SUDEP

159

Steven B. Karch

11

Cocaine-Induced Seizures, Arrhythmias, and Sudden Death

169

Claire M. Lathers, Michelle M. Spino, Isha Agarwal, Laurie S. Y. Tyau, and Wallace B. Pickworth

12

Risk Factors for Sudden Death in Epilepsy

187

Thaddeus S. Walczak

13

EEG Findings in SUDEP

201

Maromi Nei and Nicole Simpkins

14

Severity of Seizures as a Forensic Risk and Case Reports

209

Edward H. Maa, Michael P. Earnest, Mark C. Spitz, and Jacquelyn Bainbridge

15

Intractable Epilepsy in the Setting of Malformations of Cortical Development as a Mechanism for SUDEP

221

Lara Jehi and Imad Najm

16

Neurogenic Cardiac Arrhythmias Howan Leung and Anne Y. Y. Chan

235

Table of Contents

17

Stress and SUDEP

ix

253

Claire M. Lathers and Paul L. Schraeder

18

Genetics of Sudden Death in Epilepsy

267

Neeti Ghali and Lina Nashef

19

Cardiac Channelopathies and Sudden Death

285

Benito Herreros

20

Sodium Channel Dysfunction: Common Physiopathologic Mechanism Associated with Sudden Death ECG Abnormalities in Brugada Syndrome and Some Types of€Epilepsy: Case Histories

303

Claire M. Lathers, Paul L. Schraeder, and Michael W. Bungo

21

Not Seizure but Syncope

311

Saumya Sharma, Trieu Ho, and Bharat K. Kantharia

22

Syncope, Seizures, and SUDEP: Case Histories

325

Claire M. Lathers, Paul L. Schraeder, and Michael W. Bungo

23

Sudden Death in Epilepsy: Relationship to the Sleep–Wake Circadian Cycle and Fractal Physiology

333

John D. Hughes and Susumu Sato

24

SUDEP: Medicolegal and Clinical Experiences

347

Braxton B. Wannamaker

Section II SUDEP Animal Models: MECHANISMS OF RISKS

25

Sudden Death: Animal Models to Study Nervous System Sites of Action for Disease and Pharmacological Intervention Claire M. Lathers

363

x Table of Contents

26

Synaptic Plasticity of Autonomic Ganglia: Role of Chronic Stress and Implication in Cardiovascular Diseases and Sudden Death

395

Karim A. Alkadhi and Karem H. Alzoubi

27

Animal Model for Sudden Cardiac Death: Autonomic Cardiac Sympathetic Nonuniform Neural Discharge

427

Claire M. Lathers

28

Animal Model for Sudden Unexpected Death in Persons with Epilepsy

437

Claire M. Lathers and Paul L. Schraeder

29

A Characterization of the Lockstep Phenomenon in Phenobarbital-Pretreated Cats

465

Jeffrey M. Dodd-O and Claire M. Lathers

30

Relationship of the Lockstep Phenomenon and Precipitous Changes in Blood Pressure

481

Amy Z. Stauffer, Jeffrey M. Dodd-O, and Claire M. Lathers

31

Interspike Interval Histogram Characterization of€Synchronized Cardiac Sympathetic Neural Discharge and€Epileptogenic Activity in the Electrocorticogram of€the€Cat

495

Daniel K. O’Rourke and Claire M. Lathers

32

Power Spectral Analysis: A Procedure for Assessing Autonomic Activity Related to€Risk Factors for Sudden and Unexplained Death in€Epilepsy

513

Animal Model for Sudden Cardiac Death: Sympathetic Innervation and Myocardial Beta-Receptor Densities

539

Stephen R. Quint, John A. Messenheimer, and Michael B. Tennison

33

Claire M. Lathers and Robert M. Levin

34

Antiepileptic Activity of Beta-Blocking Agents

551

Claire M. Lathers, Kam F. Jim, William H. Spivey, Claire Kahn, Kathleen Dolce, and William D. Matthews

Table of Contents

35

Arrhythmias Associated with Epileptogenic Activity Elicited by Penicillin

xi

567

Claire M. Lathers and Paul L. Schraeder

36

Role of Neuropeptides in the Production of Epileptogenic€Activity and Arrhythmias

577

Claire M. Lathers

37

Sudden Epileptic Death in Experimental Animal Models

591

Ombretta Mameli and Marcello Alessandro Caria

38

Sympathetic Nervous System Dysregulation of Cardiac Function and Myocyte Potassium Channel Remodeling in Rodent Seizure Models: Candidate Mechanisms for SUDEP 615 Steven L. Bealer, Cameron S. Metcalf, Jason G. Little, Matteo Vatta, Amy Brewster, and Anne E. Anderson

39

The Urethane/Kainate Seizure Model as a Tool to Explore€Physiology and Death Associated with Seizures

627

Mark Stewart

40

Acute Cardiovascular Response during Kindled Seizures

645

Jeffrey H. Goodman, Richard W. Homan, and IsAac L. Crawford

41

DBA Mice as Models of Sudden Unexpected Death in€Epilepsy

659

Carl L. Faingold, Srinivasan Tupal, Yashanad Mhaskar, and Victor V. Uteshev

Section III Clinical Issues of Sudden Death

42

Cardiac and Pulmonary Risk Factors and Pathomechanisms€of Sudden Unexplained Death in Epilepsy€Patients Josef Finsterer and Claudia Stöllberger

679

xii Table of Contents

43

Neurocardiac Interactions in Sudden Unexpected Death in€Epilepsy: Can Ambulatory Electrocardiogram-Based Assessment of Autonomic Function and T-Wave Alternans Help to Evaluate Risk?

693

Richard L. Verrier and Steven C. Schachter

44

Arrhythmogenic, Respiratory, and Psychological Risk Factors for Sudden Unexpected Death and Epilepsy: Case Histories

711

Claire M. Lathers

45

Sudden Arrhythmic Death Syndrome: Underlying Cardiac Etiologies, Their Implications, and the Overlap with SUDEP

721

Paramdeep S. Dhillon and Elijah R. Behr

46

Odds Ratios Study of Antiepileptic Drugs: A Possible Approach to SUDEP Prevention?

743

Claire M. Lathers, Paul L. Schraeder, and H. Gregg Claycamp

47

Antiepileptic Drugs Benefit/Risk Clinical Pharmacology: Possible Role in Cause and/or Prevention of SUDEP

755

Claire M. Lathers and Paul L. Schraeder

48

Clinical Pharmacology and SUDEP

789

Claire M. Lathers and Paul L. Schraeder

49

Experience-Based Teaching of Therapeutics and Clinical€Pharmacology of Antiepileptic Drugs: Sudden Unexplained Death in Epilepsy: Do Antiepileptic Drugs Have a Role?

801

Claire M. Lathers and Paul L. Schraeder

50

Clinical Pharmacology of Antiepileptic Drug Use: Clinical Pearls about the Perils of Patty

827

Paul L. Schraeder and Claire M. Lathers

51

Compliance with Antiepileptic Drug Treatment and the€Risk of Sudden Unexpected Death in Epilepsy Torbjörn Tomson

845

Table of Contents

52

SUDEP Clinical Case Histories: Typical and Atypical

xiii

853

Paul L. Schraeder

53

Cardiac Antiarrhythmic Agents: Pharmacological Basis for Their Antiarrhythmic and Proarrhythmic Effects

861

Saumya Sharma, Trieu Ho, and Bharat K. Kantharia

54

Could Beta–Blocker Antiarrhythmic and Antiseizure Activity Help Prevent SUDEP?

877

Claire M. Lathers

55

Decision Analysis and Risk Management

887

H. Gregg Claycamp

56

Epilepsy Surgery and the Prevention of SUDEP

905

Ryan S. Hays and Michael R. Sperling

57

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP

915

Jane Hanna and Rosemary Panelli

58

Bereavement and Sudden Unexpected Death in Epilepsy

937

Lina Nashef and Lene Sahlholdt

59

SUDEP: A Clinical and Communicative Conundrum

943

Paul L. Schraeder and Claire M. Lathers

60

Epilepsy and SUDEP: Lessons Learned: Scientific and Clinical Experience

953

Claire M. Lathers and Paul L. Schraeder

61

SUDEP: A Mystery Yet to Be Solved

967

Claire M. Lathers and Paul L. Schraeder

62

Forensic Evidence and Expert Witnesses: Scientific Evidence: Getting It in and Keeping It Out

973

Thomas L. Bohan

Index

983

Foreword by Samuels

Epilepsy is one of the most prevalent neurological diseases in the world. Known for ages and clearly described in some of the oldest medical treatises, it was greatly feared in the ancient world because of the belief that it could cause sudden death. As reasonable treatments arose—ἀrst bromides, then phenobarbital, and, in the mid-twentieth century, phenytoin, the ἀrst modern antiepileptic drug—the mainstream of the medical community came to believe that epileptic seizures themselves were relatively harmless. During my own training in internal medicine in the 1970s, it was widely taught that the best treatment for a convulsion was to simply place the patient in a safe environment and let the seizure take its course. With the development of newer, safer treatments that include drugs and nonpharmacological approaches such as surgery and, more recently, vagal and deep brain stimulation, it became progressively more apparent that the ancients were, in fact, correct. Epilepsy, a state of recurrent unprovoked seizures, itself indeed carried with it an increased risk for sudden unexpected death and the term SUDEP (sudden unexpected death in epilepsy) was born. SUDEP now joins a long list of sudden death syndromes, including sudden death in middle aged men, sudden unexpected nocturnal death syndrome (SUNDS), sudden death from fright, sudden infant death syndrome (SIDS), sudden death in young athletes, sudden death associated with drug use, sudden death from heart disease, sudden death during sedative drug (including alcohol) withdrawal, sudden death during delirium, sudden death from stroke (including subarachnoid hemorrhage), and sudden death from head injury. With the advent of long-term monitoring of various physiological parameters, including the electrocardiogram, arterial oxygen saturation, and the electroencephalogram, it became apparent that some of these sudden death syndromes had in common a capacity to produce malignant cardiac arrhythmias, respiratory arrest, or both. In his landmark paper of 1942, “Voodoo” Death, the eminent physiologist Walter B. Cannon recounted stories of sudden death from the anthropological literature and posited an autonomic storm as the unifying hypothesis. In the past half-century, much has been learned about the capacity of the brain to damage the visceral organs, but many features of SUDEP remain an enigma. Despite all of the advances in the diagnosis and treatment of seizure disorders, the threat of sudden death still hovers over the epileptic patient, much as it did in ancient times. In Sudden Death in Epilepsy: Forensic and Clinical Issues, four of the most eminent experts in SUDEP, Claire Lathers, Paul Schraeder, Michael Bungo, and Jan Leestma, have put together an impressive tome representing the state of this art and science. The book’s three sections, Forensics of Sudden Death, SUDEP Animal Models: Mechanisms of Risk, and Clinical Issues in Sudden Death, are written by a veritable who’s who in the ἀeld. The interested reader can ἀnd chapters on the history, diagnosis, phenomenology, mechanisms, genetics, animal models, pathology, physiology, and even the social repercussions of this devastating phenomenon. Books written by so many authors inevitably create a challenge with regard to organization and thematic coherence, but the four editors, themselves xv

xvi

Foreword by Samuels

each major contributors to the literature on SUDEP, have done an admirable job pulling together the disparate array of experts into a volume that holds together and is readable. Sudden Death in Epilepsy: Forensic and Clinical Issues should be of great interest to neurologists, psychiatrists, neurosurgeons, internists, cardiologists, neuroscientists, and cardiovascular specialists. Epileptologists and their trainees will ἀnd the content invaluable in their day-to-day lives of counseling and treating people with epilepsy. Taken as a whole, the content acts as a roadmap to those who hope to someday fully understand and prevent this dramatic and tragic event. Martin A. Samuels, MD, FAAN, MACP, DSc (hon) Chairman, Department of Neurology Brigham and Women’s Hospital and Professor of Neurology Harvard Medical School Boston, Massachusetts

Foreword by Schachter

If sudden death in epilepsy is the most feared and serious consequence of epilepsy, why is it seldom discussed and woefully under-researched? The editors point to many reasons for this in the Preface—inadequate animal models and basic understanding, lack of clinical recognition from treating physicians and medical examiners, inability to eliminate the possibility of sudden death in the nearly one in three patients with epilepsy whose seizures are drug-resistant, and continuing reluctance among physicians to discuss sudden death with patients and their families. Compounding these issues is the silo-style infrastructure of academic medicine, which creates intrinsic barriers to establishing clinical and research collaborations across relevant disciplines, such as epidemiology, neurology, cardiology and pulmonology, as well as between physicians and applied scientists, including electrical, mechanical, and computer engineers. Despite these and other problems, there are reasons to be optimistic that research will begin to solve the mysteries posed by Drs. Lathers and Schraeder at the end of this book. First, creative and passionate researchers are dedicated to eradicating sudden death in epilepsy, perhaps most of all the editors of this volume, who have put together the most comprehensive and current treatise on the topic. The like-minded authors span numerous disciplines and their chapters challenge current paradigms and suggest new ways of thinking about sudden death in epilepsy. Second, the National Institute of Neurological Disorders and Stroke (NINDS) is actively engaged. Curing Epilepsy 2007: Translating Discoveries into Therapies, organized by the NINDS, and attended by more than 400 researchers, health care professionals, patients, and family members, affirmed sudden death in epilepsy as a major target for research, with the development and validation of at least one prevention strategy to decrease its occurrence as a short-term goal, and identiἀcation of the responsible mechanisms, including effects of seizures on autonomic functioning, particularly cardiac and respiratory, as a longer-term goal. The NINDS also sponsored a workshop on sudden death in epilepsy, held in November, 2008, which brought together researchers, clinicians, and patient advocacy groups to discuss strategies and to make plans for research and outreach. Third, professional epilepsy societies have recently established committees and taskforces with patient advocacy organizations, and have committed resources to work together to educate epilepsy professionals, doctors in training, and patients and families about sudden unexpected death in epilepsy (SUDEP), and to chart research agendas. The recommendations of one such task force include “convening a multidisciplinary workshop to reἀne current lines of investigation and to identify additional areas of research for mechanisms underlying SUDEP; performing a survey of patients and their families and caregivers to identify effective means of education that will enhance participation in SUDEP research; conducting a campaign aimed at patients, families, caregivers, coroners, and medical examiners that emphasizes the need for complete autopsy examinations for patients with suspected SUDEP; and securing infrastructure grants to fund a consortium xvii

xviii

Foreword by Schachter

of centers that will conduct prospective clinical and basic research studies to identify preventable risk factors and mechanisms underlying SUDEP” [1]. Fourth, numerous patient advocacy groups and family survivors are talking openly and urgently about sudden death in epilepsy at conferences and on Web sites. In addition, these groups are aggressively funding research projects with the hope of uncovering answers that will move the ἀeld forward. Indeed, these and other developments will likely enable critically important progress to be made in the understanding of sudden death in epilepsy and its prevention. These answers cannot come soon enough. And yet answers alone are not sufficient. If lack of seizure control is a signiἀcant risk factor for sudden death in epilepsy, then we must also urgently address the treatment gap in epilepsy, both by making treatments available to the tens of millions of persons with epilepsy in developing countries who currently have no access to anticonvulsants, and by ensuring that all patients everywhere with epilepsy have access to state-of-the-art care, including the full range of pharmacological and nonpharmacological treatment options. Even if researchers identify the underlying mechanism(s) of sudden death in epilepsy and patients have full access to comprehensive care, there will still remain one major barrier—lack of effective communication between physicians and their patients. Here are the agonizing reflections of a physician about whether to inform patients about sudden death€[2]: As a doctor and an epileptologist, I experience the deepest frustration when I am notiἀed of the sudden and unexpected death of one of my patients. Often these patients have been found dead in bed, lying dead on the floor of their apartment, or drowned in the bathtub. They are invariably young and are often at the beginning of their lives. I never forget them, and I feel that I have truly and ultimately failed them. . . . one of my young patients, a 25-year-old man, died last year while in bed. His parents had encouraged him to live alone so that his life would be as normal as possible. They had visited him that evening and had eaten dinner with him. He had a girlfriend, but she had not been there that night. The next morning he was found lying face down on his bed, fully dressed. No one was sure whether or not he had experienced a seizure. When his parents called me, we talked about the problem of sudden death and epilepsy. They wondered why his other doctors and I had not told them or their son that this could occur. I asked them if they thought he would have lived any differently if he had known. If we told every patient with epilepsy about this possibility, some might not dare live independent lives or might be burdened by anxiety, knowing that they might not awake in the morning. To this day, I do not know if my answer was appropriate.

This challenge must be addressed and solved before the discoveries of research can be translated to patient care if we are ever to end the scourge of sudden death. As physicians caring for patients with epilepsy, we should not wait for all the answers. We must accept the imperfect state of knowledge and inform our patients and, where appropriate, their families in a meaningful and compassionate way about sudden death, and work with them as much as possible to reduce the risk factors, especially by completely controlling generalized convulsive seizures. The editors and authors of Sudden Death in Epilepsy: Forensic and Clinical Issues have produced a landmark book that holds extraordinary promise for meaningful progress, bringing us closer to the day when sudden death will be fully preventable. Until then, my

Foreword by Schachter

xix

hope is that this book will inspire researchers, and give some measure of comfort to the bereaved and hope to those living with epilepsy and to their loved ones. Steven C. Schachter, M.D. Departments of Neurology Beth Israel Deaconess Medical Center Harvard Medical School Chief Academic Officer Center for Integration of Medicine and Innovative Technology Boston, Massachusetts

References 1. So EL, Bainbridge J, Buchhalter JR, et al. Report of the American Epilepsy Society and the Epilepsy Foundation Joint Task Force on sudden unexplained death in epilepsy. Epilepsia 2009;50: 917–922. 2. Schachter SC, ed. Epilepsy in our experience: Accounts of health care professionals. Oxford: Oxford University Press; 2008, pp. 7–8.

Preface

This book should be considered a sequel to, rather than a substitute for, the 1990 book Epilepsy and Sudden Death, edited by Lathers and Schraeder. Much of the material in the 1990 book, especially the discussion of animal research data, is still timely, as little additional animal-based work is extant. In addition, the epidemiological, behavioral, and drug-abuse-related data is also currently relevant. This current volume is an expansion rather than an updated sequel to the previous book on sudden unexpected death in epilepsy (SUDEP). These statements are based on our extensive review of the literature for this current book, which indicates that most of the research questions regarding evaluation of mechanisms and prevention of SUDEP have, to date, not been adequately addressed. When the previous book was published, the phenomenon of sudden unexpected/unexplained death in persons with epilepsy was considered to be a controversial topic or a very rare phenomenon by clinicians and pathologists alike. The public had little or no knowledge of the phenomenon. However, one positive consequence of the work done in SUDEP over the ensuing years is that now the phenomenon is widely referred to as SUDEP and it is accepted as a complication of epilepsy worldwide. SUDEP is one of the most common causes of death in young adults with a history of epilepsy, and presents a spectrum of dilemmas to forensic experts, clinicians, and researchers. At a most basic level, as determined in a national survey of coroners and medical examiners and discussed in this book, even in self-evident cases, SUDEP is infrequently used on death certiἀcates as a ἀnal diagnosis or medical cause of death. Ironically, this survey found that the majority of those same coroners and medical examiners who did not routinely use SUDEP on death certiἀcates, nonetheless acknowledged its validity as a diagnosis in a theoretical case in which no cause of death in a person with epilepsy could be found on postmortem examination. This disconnect between intellectual acknowledgment of its existence, and actual use of SUDEP as a death certiἀcate diagnosis, in all likelihood, results in signiἀcant underreporting of the prevalence of this tragic phenomenon. It is hoped that this book will prove to be an important resource to improve the knowledge of coroners and medical examiners about the use of the term SUDEP in appropriate cases. Since a portion of this new book addresses forensic issues, it is intended to be a resource for forensic pathologists, attorneys, coroners, and medical examiners as they struggle to determine the cause of death in persons with epilepsy. Because most other clinical and research aspects of SUDEP are also addressed, neurologists, experts in epilepsy, cardiologists, clinical pharmacologists, pharmacists, nurses, students, and persons with epilepsy or with a family member so diagnosed should all ἀnd sections of interests. Over the past two decades, many more basic and clinical scientists doing research in the ἀeld of epilepsy are focusing on the problem of SUDEP, as it is one of the most common causes of death associated with having epilepsy. However, to date, most research has focused on clinical and epidemiological data, with relatively few investigators using xxi

xxii

Preface

animal models. In addition, there has been little effort to determine whether there are common links between the risk of SUDEP and sudden cardiac deaths. While cardiologists are focusing increasingly on genetic risk factors for potentially life-threatening arrhythmias in young persons without coronary artery disease, there has been little effort to establish collaboration between neuroscience investigators interested in SUDEP and cardiology researchers interested in unraveling the genetically determined risk factors for potentially fatal arrhythmias. This book will be of help in directing neurological and cardiologic investigators to common areas of interest, especially as relates to the unanswered question of why some persons with epilepsy seem to be at risk for neurogenically induced fatal cardiac events. The basic science of epilepsy as relates to SUDEP is updated and expanded in this book. The role of alcohol and other drugs as seizure enhancing or producing conditions are discussed. The role of medication, compliance with prescribed antiepileptic drugs, or lack thereof, is an even more important problem today and is addressed in several chapters. The relevance of epidemiological studies of SUDEP is also presented. Sudden death in the pediatric population is also reviewed. The neurocardiological aspects of SUDEP are addressed in some detail. While SUDEP remains a major unsolved problem, thanks to the efforts of those who have conducted animal and epidemiological studies, what was once denied and called a myth is an acknowledged reality that has to be dealt with as a multidisciplinary issue. To solve the mystery of SUDEP, a global focus is required. Persons at risk must be identiἀed and preventive treatment regimens developed to decrease the occurrence of SUDEP. Clinical chapters emphasize that sophisticated simultaneous ambulatory EKG and EEG telemetry and respiratory function monitoring of patients at risk for sudden death will help identify cardiac, respiratory, and epileptogenic interactions in order to decrease the risk of SUDEP. Basic scientiἀc research programs and clinical and epidemiology studies are needed. Multidisciplinary teams working in clinical settings and with laboratories must address the global issues of SUDEP. Forensic chapters in this book discuss the fact that if the correct term “SUDEP” is used on autopsy reports, and if postmortem verbal autopsies are conducted when needed, the true incidence of SUDEP well may be determined to be much higher than previously thought. Drug development can justify the need for new antiepileptics and drugs in other pharmacological classes that address the reduction of the risk of SUDEP. Animal model chapters discuss new data gleaned by building on the previously utilized models and the lessons learned during the last quarter century. The use of animal models continues to be one of the most useful approaches to better understanding SUDEP. Discussion in various chapters may be summarized by stating that it is important to “think out of the box” when evaluating an established animal model that has the potential, with modiἀcation(s), to investigate possible mechanisms of SUDEP. Various authors of the animal model chapters presented in this book emphasize that multiple models are needed to investigate the pathophysiology of SUDEP, to hypothesize about effective treatments, to develop pilot studies in persons with epilepsy, and to conduct conἀrmatory large-scale clinical trials. The ἀelds of pharmacology, clinical pharmacology, and cardiology have much to offer as we work to improve compliance, develop new antiepileptic drugs, and apply different categories of drugs that hopefully attenuate the chances of occurrence of SUDEP. Authors address the possible overlapping mechanisms that may apply to the risk of sudden

Preface

xxiii

unexpected death occurring in epilepsy and in cardiac disease. Several chapters explore the interaction between the central and peripheral autonomic nervous systems and the cardiopulmonary systems. Included is a discussion of the potential interactive role of genetically determined subtle cardiac risk factors for arrhythmias, with a predisposition for seizurerelated cardiac arrhythmias. The possible mechanisms that are operant in producing both epileptogenic and cardiogenic arrhythmias are addressed. Several chapters examine proposed mechanistic factors in SUDEP, listing risk categories of arrhythmogenic, respiratory, and hypoxia, and psychological factors and discussing mechanisms for risks associated with each category. Several chapters discuss patients with Brugada syndrome and an interesting, interpretive presentation of a hypothesis to explain a common pathophysiologic mechanism associated with sodium channel dysfunction that may be common to clinical electrophysiological abnormalities and some types of epilepsy. Clariἀcation of risk factors and establishment of the mechanism of SUDEP are important to establish preventative measures for SUDEP and emphasize the need to strive for full seizure control. Several chapters discuss the importance of encouraging patients with epilepsy to receive nonmedical, common sense, lifestyle-modifying interventions that have generally accepted public health beneἀts, even though there is as yet no consensus that they may or may not prevent sudden death. Cardiac patients, psychiatric patients, and certain ethnic groups experiencing acute stressful circumstances are at risk for unexpected sudden death. The impact of adverse emotional states on the autonomic control of cardiac rhythm is an established factor leading to cardiac dysrhythmias in humans and other species. Although stress is associated with changes in autonomic neural function, its role as a potential risk factor for SUDEP has not been investigated. The association of epilepsy with depression and anxiety indicates that emotional stress should be evaluated as a potential risk factor for SUDEP. The interactions between emotional factors and the arrythmogenic potential of epileptiform discharges, and the possibility of beneἀt from stress management intervention need investigation. Prospective studies of patients are needed to determine how we can identify which persons with epilepsy are at risk for SUDEP. In a number of chapters, the authors speculate about common potential preventive measures to minimize the risk of both sudden unexpected death in epilepsy and sudden cardiac death. Several chapters address the issue of clinicians who treat persons with epilepsy, manifesting reluctance to discuss the possibility of SUDEP with their patients. This reluctance seems to be the result of concern that even introducing the topic to the patient and family would be stressful for them. However, for the most part, families of SUDEP victims express disappointment that they had not known of this possibility and call for widespread acknowledgment of the potential for occurrence of sudden death in association with epilepsy. Epilepsy Bereaved in the United Kingdom, an organization founded by families of SUDEP victims, has been particularly successful in raising the level of awareness of SUDEP in persons with epilepsy and their families, and within the medical profession and the general public. This kind of advocacy and dissemination of information will serve to increase the availability of resources used to solve the tragic mystery of SUDEP. A primary purpose of this book is to provide clinicians with the knowledge necessary to improve their comfort level in discussing SUDEP with patients and families and thereby to allow for freer dissemination of information about minimizing the known risk factors for SUDEP (e.g., erratic compliance in taking prescribed antiepileptic drugs). Committed investigators in research must solve the mystery of SUDEP using a leadership philosophy foundation that provides innovative vision and approaches for SUDEP

xxiv

Preface

research and teaching programs. The interaction of teaching and research is essential. While a student is learning how to conduct research, that person must simultaneously learn to become a teacher. Academic fellowships and competitions and grant funding for medical students, postdoctoral fellows, residents, and faculty will attract medical and graduate trainees to work on SUDEP and move the SUDEP knowledge base forward. As self-learning exercises, we have incorporated a variety of case studies of sudden death within chapters and as standalone chapters as practical teaching exercises. Clinical and basic science investigators must provide vision and a fertile environment to teach students to become tomorrow’s leaders in the struggle to solve the mystery of SUDEP. Claire M. Lathers Paul L. Schraeder Michael W. Bungo Jan E. Leestma

Acknowledgments

I wish to acknowledge the dedication and hard work of co-editor Claire Lathers. (JEL) We would not have been able to complete this book in a timely manner without the diligent secretarial assistance of Marie Faiola. (CML, PLS, MWB, JEL).

xxv

Editors

Claire M. Lathers, PhD, FCP, has been credentialed as a Senior Biomedical Research Scientist by the U.S. Food and Drug Administration (FDA) for international recognition of her work in the two areas of cardiovascular autonomic dysfunction associated with sudden death in persons with epilepsy and with space flight. The primary focus of her international cardiovascular pharmacology research career has centered on autonomic peripheral and central mechanisms involved in the control and regulation of blood pressure, heart rate and rhythm, and the electroencephalogram. Dr. Lathers and Dr. Schraeder have collaborated and published numerous studies and two books focusing on epilepsy and sudden unexplained death. Dr. Lathers served the FDA for a total of 11 years, including four years as the Senior Advisor for Science to the director in the Center for Veterinary Medicine and director of the Office of New Animal Drug Evaluation and ἀve years in the Center for Drug Evaluation and Research as a pharmacology reviewer. Claire also served as a special government expert for 2 years. Dr. Lathers spent 14 years working as a visiting scientist at NASA/ USRA, collecting data from subjects in ground-based studies and from astronauts and cosmonauts before, during and after space flight. Dr. Lathers earned a BS in pharmacy from Albany College of Pharmacy, Union University and her PhD in pharmacology from the State University of New York at Buffalo School of Medicine. She completed an NIH funded two-year postdoctoral fellowship at the Medical College of Pennsylvania. Her academic faculty experience includes working at the Medical College of Pennsylvania (15 years), Albany College of Pharmacy (two years as president, dean, and tenured professor); Uniformed Services University of the Health Sciences (three years part time); and Gwynedd Mercy College (11 years part time). Dr. Lathers is currently collaborating with a number of academicians on scientiἀc issues of the nexus between human and veterinary medicine clinical pharmacology, antimicrobial resistance, food safety, and biodefense measures. In addition to her academic and government service, Dr. Lathers has worked in the pharmaceutical industry. She served as chief scientiἀc officer of Barr Pharmaceuticals for three years and worked part time with four other pharmaceutical companies during a 15-year period. Dr. Lathers has authored or co-authored over 300 publications and abstracts, has edited three books, and has presented data at over 140 international meetings. She is an emeritus fellow, an honorary member of the Board of Regents, and a past president of the American College of Clinical Pharmacology, having served as regent, treasurer, and president. Dr. Lathers also served as the section editor of the educational series entitled: “Innovative Teaching Methods in Clinical Pharmacology” for the Journal of Clinical Pharmacology for 17 years. Claire served as a member of the Board of the Annapolis Center, charged to evaluate risk assessments, and worked on the epidemiology, toxicology, and food safety workshops and accords. In recognition of her work, Dr. Lathers has been the recipient of numerous awards and honors. xxvii

xxviii Editors

Paul L. Schraeder, MD, FAAN, is professor emeritus of neurology at Drexel University College of Medicine, former chief of neurology at the Medical College of Pennsylvania Hospital, Philadelphia, Pennsylvania and former professor of medicine and neurology at the Robert Wood Johnson School of Medicine in Camden, New Jersey; head of the Division of Neurology at Cooper Hospital/University Medical Center, Camden, New Jersey; and former associate professor of neurology at the Medical College of Pennsylvania. He is a member of the Philadelphia Neurological Society, the American Epilepsy Society, and a fellow of the American Academy of Neurology. He has served on the professional advisory board of the Epilepsy Foundation of Southeastern Pennsylvania and the Epilepsy Foundation of America and as medical advisor to Epilepsy Bereaved, a support organization for surviving friends and family of victims of sudden unexplained death in epileptic persons (SUDEP) in the United Kingdom. Dr. Lathers and Dr. Schraeder co-edited the ἀrst book addressing the topic of Epilepsy and Sudden Death (Marcel Dekker, 1990). Dr. Lathers and he have collaborated for over three decades studying and investigating the mystery of SUDEP and developed the ἀrst experimental animal models of this fatal phenomenon. Dr. Schraeder organized a collaborative nationwide survey of how coroners and medical examiners evaluate the deaths of persons with a history of epilepsy. Dr. Schraeder received the AB€ degree from Bucknell University, Lewisburg, Pennsylvania and the MD degree from€Jefferson Medical€College, Philadelphia, Pennsylvania. He completed his residency in neurology and fellowship in electroencephalography and experimental epilepsy at the University of Wisconsin. Michael W. Bungo, MD, FACC, FACP, has agreed to serve as the third co-editor of this new book. After residency and fellowship training in cardiology at Harvard Medical School programs in Boston, Dr. Bungo worked full time for NASA as director of the Space Biomedical Research Institute at the Johnson Space Center. During his time at NASA, Dr. Bungo ἀrst cataloged arrhythmias occurring in spaceflight crews and hypothesized that the unique physiological and psychological environments of space flight may be arrhythmogenic. The Aerospace Medical Association awarded him the Louis H. Bauer Founders Award and NASA awarded him the NASA Medal for Exceptional Scientiἀc Achievement for his pioneering research work concerning the heart’s adaptation to zero gravity. While at NASA, Dr. Bungo served on the Joint U.S.–U.S.S.R. Working Group that developed the now combined space station science program for these prior competitors. He left NASA to assume the positions of director of the Heart Station, Division of Cardiology and vice chair for Inpatient Affairs, Department of Internal Medicine at the University of Texas Medical Branch in Galveston. Dr. Bungo subsequently moved to the University of Texas Medical School in Houston and served as chief of staff at the LBJ General Hospital, CEO of UT-Physicians, and vice dean for Clinical Affairs. He is currently a professor of medicine in the Division of Cardiology at that same institution. Dr. Bungo’s current research project funded by NASA is entitled “Cardiac Atrophy and Diastolic Dysfunction during and after Long Duration Spaceflight: Functional Consequences for Orthostatic Intolerance, Exercise Capacity, and Risk of Cardiac Arrhythmias.” In addition to numerous publications and presentations, Dr. Bungo and Dr. Lathers have collaborated since 1989 and have co-authored seven published papers. In 2008, Dr. Bungo has co-authored three papers and one book chapter with Drs. Lathers and Schraeder on the mystery of SUDEP. Dr.€Bungo has delivered scientiἀc presentations at three international symposiums organized by Dr.€Lathers.

Editors

xxix

Jan E. Leestma, MD, MM, is the lead author of the second edition of Forensic Neuropathology. He received the MD degree from the University of Michigan School of Medicine in 1964, and a Masters of Management (MM) degree from the J.L. Kellogg Graduate School of Management of Northwestern University, Evanston, Illinois in 1986. He completed residency training in anatomic pathology and neuropathology at the University of Colorado Medical Center, Denver, Colorado and a neuropathology fellowship at the€Albert Einstein College of Medicine, Bronx, New York. He is certiἀed in both anatomic pathology and neuropathology by the American Board of Pathology (1970). He served in the United States Air Force Medical Corps at the Armed Forces Institute of Pathology, Washington, D.C. (1968–1971) and was honorably discharged with the rank of major, USAF MC. He was an assistant and associate professor of pathology and neurology at Northwestern University School of Medicine (1971–1986) and served as chief of neuropathology at both Northwestern Memorial Hospital and the Children’s Memorial Hospitals, Chicago, Illinois. He was professor of pathology and neurology, and dean of students for the Division of the Biological Sciences and the Pritzker School of Medicine at the University of Chicago, Chicago, Illinois (1986–1987). He was an assistant medical examiner and neuropathology consultant to the Office of the Medical Examiner, Cook County, Illinois (1977–1987). He was a guest researcher at the Karolinska Institutet, Huddinge University Hospital, Pathology Institute, Stockholm, Sweden (1981–1982). He was associate medical director and chief of neuropathology at the Chicago Institute of Neurosurgery and Neuroresearch in Chicago (1987–2003). He has had a private consulting practice in forensic neuropathology since the early 1970s which continues to the present time. He has given expert testimony in more than 30 U.S. states, Canada, and the United Kingdom. He is the author of more than 100 professional publications including numerous book chapters in texts. He was the author of Forensic Neuropathology (ἀrst edition), Raven Press, New York, 1988. He is a member of the American Association of Neuropathologists, and of the American Academy of Forensic Sciences. Dr. Leestma wrote some chapters in Epilepsy and Sudden Death, edited by Drs. Lathers and Schraeder. Thus it is a natural extension of this collaboration for these three editors to work with co-editor Dr. Bungo and CRC Press is to expand the focus of forensics and clinical issues. The diverse scientiἀc expertise and endeavors of each of the four editors, working in the different ἀelds of forensics, neurology, cardiology, and clinical pharmacology, have united in this edition to produce a book with a special emphasis on the forensics and clinical issues associated with neurocardiology, epilepsy, arrhythmias, and sudden death.

Contributors

Isha Agarwal

Medical College of Pennsylvania Philadelphia, Pennsylvania

Karim A. Alkadhi, PhD

Department of Pharmacological and Pharmaceutical Sciences University of Houston Houston, Texas

Karem H. Alzoubi

Department of Clinical Pharmacy Jordan University of Science and Technology Irbid, Jordan

Anne E. Anderson, MD

Departments of Pediatrics, Neurology, Neuroscience Baylor College of Medicine Houston, Texas

Ricardo M. Arida

Departamento de Fisiologia Universidade Federal de São Paulo/Escola Paulista de Medicina (UNIFESP/EPM) São Paulo, Brasil

Jacquelyn Bainbridge, PharmD, FCCP School of Pharmacy University of Colorado Health Sciences Center Denver, Colorado

Steven L. Bealer, PhD

Department of Pharmacology and Toxicology University of Utah Salt Lake City, Utah

Elijah R. Behr, MD

Cardiac and Vascular Division St. George’s University of London London, United Kingdom

Amy Brewster

Department of Pediatrics Baylor College of Medicine Houston, Texas

Michael W. Bungo, MD, FACC, FACP,€CPE

Division of Cardiology University of Texas Medical School at Houston Houston, Texas

Esper A. Calvalheiro

Disciplina de Neurologia Experimental Universidade Federal de São Paulo/Escola Paulista de Medicina (UNIFESP/EPM) São Paulo, Brasil

Marcello Alessandro Caria

Department of Biomedical Sciences Human Physiology Division Sassari, Italy

Anne Y. Y. Chan

Department of Medicine and Therapeutics Chinese University of Hong Kong and Prince of Wales Hospital Hong Kong SAR, China

H. Gregg Claycamp, PhD

Center for Drug Evaluation and Research Office of Compliance U.S. Food and Drug Administration Silver Spring, Maryland

Isaac L. Crawford

Department of Neurology Southwestern Regional Epilepsy Center Veterans Administration Medical Center University of Texas Southwestern Medical Center Dallas, Texas

Thomas L. Bohan, PhD, JD, F-AAFS, D-IBFES MTC Forensics Peaks Island, Maine

xxxi

xxxii

Roberta M. Cysneiros

Programa de Pós-Graduação em Distúrbios do Desenvolvimento do Centro de Ciências Biológicas e da Saúde da Universidade Presbiteriana€Mackenzie São Paulo, Brasil

Paramdeep S. Dhillon

Cardiac and Vascular Sciences St. George’s University of London London, United Kingdom

Contributors

Benito Herreros, MD, PhD

Cardiology Department Hospital Universitario Rio Hortega Valladolid, Spain

Trieu Ho, MD

Division of Cardiology University of Texas Medical School at Houston Houston, Texas

Richard W. Homan, MD

Department of Physiology Johns Hopkins University Baltimore, Maryland

Department of Neurology University of Texas Southwestern Medical Center Southwestern Regional Epilepsy Center Veterans Administration Medical Center Dallas, Texas

Kathleen Dolce

John D. Hughes, MD

Jeffrey M. Dodd-O, MD

Department of Drug Metabolism Smith Kline & French Laboratories King of Prussia, Pennsylvania

Michael P. Earnest, MD Department of Neurology Denver Health and Hospitals Denver, Colorado

Carl L. Faingold, PhD

Department of Pharmacology Southern Illinois University School of Medicine Springἀeld, Illinois

Josef Finsterer

Krankenanstalt Rudolfstiftung Vienna, Austria

Neeti Ghali

Department of Clinical Genetics Guy’s Hospital London, United Kingdom

Jeffrey H. Goodman, PhD

Department of Physiology and Pharmacology State University of New York Brooklyn, New York

Jane Hanna, OBE, MA, BCL Epilepsy Bereaved Wantage, United Kingdom

Ryan S. Hays

Department of Neurology Thomas Jefferson University Philadelphia, Pennsylvania

National Institute of Neurological Disorders and Stroke National Institutes of Health Bethesda, Maryland

Richard Hughes

Department of Neurology Denver Health and Hospitals Denver, Colorado

Lara Jehi, MD

Department of Neurology Cleveland Clinic Neurological Institute Cleveland, Ohio

Kam F. Jim, PhD

Clinical Documentation Sanoἀ-Aventis Great Valley, Pennsylvania

Claire Kahn, PhD

Department of Drug Metabolism Smith Kline & French Laboratories King of Prussia, Pennsylvania

Bharat K. Kantharia, MD

Division of Cardiology University of Texas Medical School at Houston Houston, Texas

Steven B. Karch, MD, FFFLM, FFDov Berkeley, California

Steven A. Koehler

Office of the Coroner of Allegheny County Pittsburgh, Pennsylvania

Contributors

Claire M. Lathers, PhD, Emeritus FCP Albany, New York

xxxiii

Lina Nashef, MBChB, MD, FRCP

Chicago, Illinois

Departments of Neurology and Clinical Neurosciences King’s College Hospital London, United Kingdom

Howan Leung

Maromi Nei, MD

Robert M. Levin

Daniel K. O’Rourke, MD

Albany College of Pharmacy Albany, New York

Medical College of Pennsylvania Philadelphia, Pennsylvania

Jason G. Little

Rosemary Panelli, PhD

Jan E. Leestma, MD

Department of Medicine and Therapeutics Chinese University of Hong Kong Hong Kong SAR, China

Department of Neurology Jefferson Comprehensive Epilepsy Center Philadelphia, Pennsylvania

Department of Pharmacology and Toxicology University of Utah Salt Lake City, Utah

Joint Epilepsy Council of Australia Seymour, Australia

Edward H. Maa, MD

Graduate School of Nursing University of Virginia Newport News, Virginia

Neurology Denver Health and Hospitals Denver, Colorado

Ombretta Mameli, MD

Department of Biomedical Sciences Human Physiology Division Sassari, Italy

William D. Matthews, PhD

Department of Drug Metabolism Smith Kline & French Laboratories King of Prussia, Pennsylvania

John A. Messenheimer, MD

Department of Neurology University of North Carolina School of Medicine Chapel Hill, North Carolina

Cameron S. Metcalf

Department of Pharmacology and Toxicology University of Utah Salt Lake City, Utah

Yashanad Mhaskar

Department of Pharmacology Southern Illinois University School of Medicine Springἀeld, Illinois

Imad Najm, MD

Department of Neurology Cleveland Clinic Neurological Institute Cleveland, Ohio

Simona Parvulescu-Codrea, MD, PhD

David S. Paterson, PhD Department of Pathology Children’s Hospital Boston Boston, Massachusetts

Wallace B. Pickworth, PhD Addiction Research Center National Institute on Drug Abuse Baltimore, Maryland

Stephen R. Quint, PhD

Department of Neurology University of North Carolina Chapel Hill, North Carolina

Lene Sahlholdt

Danish Epilepsy Centre Dianalund, Denmark

Martin Allen Samuels, MD Brigham and Women’s Hospital Harvard Medical School Boston, Massachusetts

Susumu Sato, MD

Office of Clinical Director National Institute of Neurological Disorders and Stroke Bethesda, Maryland

xxxiv

Steven C. Schachter, MD

Beth Israel Deaconess Medical Center Harvard Medical School Boston, Massachusetts

Paul L. Schraeder, MD, FAAN

Department of Neurology (emeritus) Drexel University College of Medicine Philadelphia, Pennsylvania

Carla A. Scorza

Disciplina de Neurologia Experimental Universidade Federal de São Paulo/Escola Paulista de Medicina (UNIFESP/EPM) São Paulo, Brasil

Fulvio Alexandre Scorza

Disciplina de Neurologia Experimental Universidade Federal de São Paulo/Escola Paulista de Medicina (UNIFESP/EPM) São Paulo, Brasil

Saumya Sharma, MD

Division of Cardiology University of Texas Medical School at Houston Houston, Texas

Nicole Simpkins

Jefferson Comprehensive Epilepsy Center Thomas Jefferson University Philadelphia, Pennsylvania

Elson L. So, MD

Section of Electroencephalography Mayo Clinic College of Medicine Rochester, Minnesota

Eliza Y. F. Sonoda

Disciplina de Neurologia Experimental Universidade Federal de São Paulo/Escola Paulista de Medicina (UNIFESP/EPM) São Paulo, Brasil

Michael R. Sperling, MD

Jefferson Comprehensive Epilepsy Center Thomas Jefferson University Philadelphia, Pennsylvania

Michele M. Spino

Medical College of Pennsylvania Philadelphia, Pennsylvania

Mark C. Spitz, MD Health Science Center University of Colorado Aurora, Colorado

Contributors

William H. Spivey, MD (deceased) Medical College of Pennsylvania Philadelphia, Pennsylvania

Amy Z. Stauffer, MD

Medical College of Pennsylvania Philadelphia, Pennsylvania

Mark Stewart, MD, PhD

Department of Physiology/Pharmacology and Neurology State University of New York Brooklyn, New York

Claudia Stöllberger

Second Medical Department Krankenanstalt Rudolfstiftung Vienna, Austria

Michael B. Tennison

School of Medicine University of North Carolina Chapel Hill, North Carolina

Vera C. Terra

Departamento de Neurologia, Psiquiatria e Psicologia Médica Universidade de São Paulo. Ribeirão Preto São Paulo, Brasil

Torbjörn Tomson

Department of Neurology Karolinska University Hospital Stockholm, Sweden

Srinivasan Tupal

Department of Pharmacology Southern Illinois University School of Medicine Springἀeld, Illinois

Laurie S. Y. Tyau, MD

Medical College of Pennsylvania Philadelphia, Pennsylvania

Victor V. Uteshev

Department of Pharmacology Southern Illinois University School of Medicine Springἀeld, Illinois

Matteo Vatta

Department of Pediatrics Baylor College of Medicine Houston, Texas

Contributors

Richard L. Verrier, PhD, FACC Beth Israel Deaconess Medical Center Harvard Medical School Boston, Massachusetts

ἀ addeus S. Walczak, MD MINCEP Epilepsy Care Minneapolis, Minnesota

xxxv

Braxton B. Wannamaker, MD Epilepsy Services and Research, Inc. Orangeburg, South Carolina

Cyril H. Wecht, MD, JD

Office of the Coroner of Allegheny County Pittsburgh, Pennsylvania

Forensics of Sudden Death

I

Neurocardiologic Mechanistic Risk Factors in Sudden Unexpected Death in Epilepsy

1

Claire M. Lathers Paul L. Schraeder Michael W. Bungo

Contents 1.1 Introduction 1.2 Risk Factors for SUDEP 1.3 SUDEP Animal Models 1.3.1 Therapies as Factors 1.4 Case of SUDEP with a Premorbid Diagnosis of Nonepileptic Seizures 1.5 Discussion 1.5.1 Future Steps References

3 3 15 24 24 26 26 26

1.1â•…Introduction There are at least three categories of factors that may be operative in the mechanisms for sudden unexpected death in epilepsy (SUDEP) (Lathers et al. 2008a): arrhythmogenic, including changes in autonomic neural and cardiac function, respiratory and hypoxia, and psychological. Each of these main risk factor categories, in all likelihood, includes many subcategories. For example, the arrhythmogenic category includes pharmacological drug effects, genetic ion channelopathies, and acquired heart disease. The psychological factors include stress, anxiety, and depression states. This list of three primary areas of mechanism for SUDEP was expanded to include eight topics (Lathers et al. 2008b, 2008c).

1.2â•…Risk Factors for SUDEP The risk factors for SUDEP are as follows:

1. Pathophysiological mechanisms: cardio/autonomic 2. Respiratory/autonomic factors and hypoxia 3. Syncope 4. Genetic/structural mechanistic factors 5. Risk factors for SUDEP 6. Therapies: increased or decreased risk 7. Psychological factors 8. Unusual factors possibly modifying risk 3

4 Sudden Death in Epilepsy: Forensic and Clinical Issues

The pathophysiological mechanisms of SUDEP are not extant (see Table 1.1). From a cardiac standpoint, there is no question that the sympathetic component of cardiac innervation is involved in the production of potentially fatal tachyarrhythmias. It has also been established that postganglionic cardiac sympathetic innervation is altered in association with temporal lobe epilepsy and may be a pathophysiological risk factor for SUDEP (Druschky et al. 2001). These clinical data correlate with animal studies that found epiÂ� leptiform discharge–related postganglionic cardiac sympathetic abnormalities that were associated with cardiac conduction disturbances and arrhythmias (Lathers and Schraeder 1982, 1987, 1990a; Schraeder and Lathers 1983, 1989; Lathers 2010a, 2010b; Lathers et€al. Table 1.1â•… Pathophysiological Mechanisms Model, Symptoms, and/or Changes Nonuniform autonomic cardiac postganglionic neural discharge associated with coronary occlusion of left anterior descending coronary artery and/or ouabain toxicity in cats.

Arrhythmia monitoring of pacemaker patients. Cardiac beta receptor distribution: Beta receptor density right atria signiἀcantly lower than left atria; right ventricle signiἀcantly lower than left ventricle; density in ventricles higher than atria; beta receptor density of distal distribution of left anterior descending coronary artery signiἀcantly higher than proximal distribution. These regional density differences are related to the cardiac contractile strength of different areas of heart. Regional differences in beta adrenoreceptor densities reflect differences in postganglionic cardiac sympathetic innervation of the myocardium (Randall et al. 1977, 1978, 1984). Regulation of cardiac neural discharge is a new paradigm in the management of sudden cardiac death.

Mechanisms of Sudden Death Nonuniform (increases, decreases, and/or no change) autonomic neural postganglionic cardiac sympathetic discharge traveling through the stellate ganglia causes cardiac€arrhythmias, ventricular ἀbrillation, and/or SUD in the manner described by Han and Moe (i.e., nonuniform recovery of excitability in ventricular muscle). Suggested neural discharge site for drugs to modify occurrence of arrhythmia and death (Lathers et al. 1974a, 1974b, 1977a, 1977b, 1978, 1981, 1988; Lathers 1980, 1981, 1982; Lathers and Roberts 1980, 1985; Spivey and Lathers 1985; Han and Moe 1964). Revealed tachycardias, most likely ventricular tachycardia, related to SUD, give insight into terminal event mechanisms (Wichter et al. 2005; Nagele et al. 2007). A correlation with the release of norepinephrine at sympathetic nerve terminals in the heart in a manner to produce arrhythmia. Innervation density is high in subepicardium and central conduction system. Nonuniform postganglionic cardiac sympathetic cardiac innervation is related to nonuniform beta sympathetic receptor locations in the heart and affects cardiac contractility and development of arrhythmias and/or death. In diseased hearts, cardiac innervation density varies and may lead to sudden cardiac death (Lathers et al. 1986, 1986b, 1988, 1990, 2010a, 2010b, 2010c). These data were conἀrmed by Druschky et al. (2007), Ieda et al. (2006, 2007, 2008), and Kimura et al. (2007). These site differences will vary release of norepinephrine in various sites of the heart. This modiἀes cardiac contractile function and may trigger development of arrhythmias and/or sudden death. Site difference, in part, is one component of the mechanism(s) involved in sudden cardiac death. The heart is extensively innervated and its performance is regulated by the balance of discharges within and between the autonomic nervous system divisions (Ieda et al. 2008; Lathers et al. 1977a, 1977b, 1978; Lathers 1980, 1981). (continued)

Neurocardiologic Mechanistic Risk Factors in SUDEP

5

Table 1.1â•… Pathophysiological Mechanisms (Continued) Model, Symptoms, and/or Changes Nonuniform autonomic cardiac postganglionic neural discharge associated with pentylenetetrazol-induced interictal epileptogenic activity in cats. Lockstep phenomenon (LSP): Cardiac postganglionic sympathetic and vagal discharges were synchronized one for one with both ictal and interictal discharges and premature ventricular contractions, ST/T changes, and conduction blocks, and precipitous changes in blood pressure occurring concurrent with interictal spikes.

Blockading GABAergic and glycinergic receptors in medulla slices of newborn rats evoked intermittent seizure-like ἀring of cardiac parasympathetic neurons, suggesting the seizure-like pattern of ἀring during an epileptic attack may cause neurogenic ictal bradyarrhythmias, cardiac asystole, or even sudden death in persons with epilepsy (Wang et al. 2006). Cardiac postganglionic denervation in patients with epilepsy examined to evaluate ictal asystole because tachyarrhythmias are common during epileptic seizures, while bradyarrhythmias or asystoles occur less frequently (Kerling et al. 2009).

Chaos science: Simple systems that manifest periodic activity are easily perturbed, and are less able to return to the preperturbed state.

Mechanisms of Sudden Death Nonuniform autonomic neural discharge, autonomic neural imbalance of postganglionic cardiac sympathetic, and vagal discharge causes cardiac arrhythmias, ventricular ἀbrillation, and/or asystole and/or SUDEP (Lathers and Schraeder 1982, 1987; Schraeder and Lathers 1983, 1988; Lathers et al. 1984, 1990, 1993; Carnel et al. 1985; Tumer et al. 1985). Lockstep phenomenon is considered to be one potential mechanism for SUDEP (Lathers et al. 1987; Lathers and Schraeder 1990a, 1990b; Stauffer et al. 1989, 1990; Dodd-O and Lathers 1990; O’Rourke and Lathers 1990). 1. Spatial and temporal summation of neuronal discharges in a subcortical center producing a stimulus strong enough to overcome the cortical and ganglionic threshold (Dodd-O and Lathers 1990). 2. Increased synaptic recruitment, resulting in ampliἀcation of subcortical stimuli along their path, so when reaching the cortex and sympathetic ganglion, they are capable of causing susceptible neurons in these regions to discharge. 3. Increased irritability of all neurons so that subcortical impulses could stimulate cortical and ganglionic neurons (Dodd-O and Lathers 1990). Provides supporting data for the lockstep phenomenon ἀnding of Lathers et al. (1987) and colleagues. It is to be determined if the predisposition of central PNS ἀring correlates with electrical remodeling of myocardium, possibly secondary to epileptogenic activity related cardiac neural discharges (LSP).

Single-photon emission computed tomography examined I-(123)-meta-iodobenzylguanidine as a marker of postganglionic cardiac norepinephrine uptake. Pronounced reduction in cardiac single-photon emission computed tomography uptake in asystolic patients indicated postganglionic cardiac catecholamine disturbance. Impaired sympathetic cardiac innervation limits adjustment and modulation of heart rate and may increase the risk of asystolic events and, eventually, SUDEP (Kerling et al. 2009). Data of Kerling et al. support the ἀndings of Lathers et al. (1986a, 1986b, 1987, 1990), Lathers and Levin (2010), and those of Han and Moe (1964). A periodic rhythm in the brain, where normally rich complexity exists, implies a susceptibility to failure that may result in death (Gleick 1987). (continued)

6 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 1.1â•… Pathophysiological Mechanisms (Continued) Model, Symptoms, and/or Changes Mode locking: Epileptic focus and medullary cardiac center may be locked one to one during state of LSP in the manner suggested by Gleick (1987).

Power spectral analysis (Quint et al. 1990).

Sympathetic innervation is critical for effective cardiac function (Saffitz 2008). Developmental and regulatory mechanisms determining density and pattern of cardiac sympathetic innervation are unclear, as is role of innervation in arrhythmogenesis. Sema3a establishes cardiac sympathetic innervation patterning. Sema3a is abundantly expressed in the trabecular layer in early-stage embryos but is restricted to Purkinje ἀbers after birth, forming an epicardial-to-endocardial transmural sympathetic innervation pattern. Sema3a(−/−) mice lacked a cardiac sympathetic innervation gradient and exhibited stellate ganglia malformation, leading to marked sinus bradycardia due to sympathetic dysfunction. Sympathetic dysfunction: altered postganglionic cardiac sympathetic innervation found postmortem in patients with chronic temporal lobe epilepsy (Druschky et al. 2001).

Mechanisms of Sudden Death Occurrence of a very regular oscillator in the brain is theoretically dangerous, regardless of mechanism. Fractal processes are ubiquitous in biological systems, include brain electrical depolarizations, and are systems that convey different information. Complexity of a fractal process may at times depreciate into a simple periodic process representing decay of the system and dramatic change (Goldberger et al. 1985; Winfree 1987). Causes of death include failure of the brain and heart. If the brain sends a message to the heart triggering a fatal arrhythmia, or if the heart enters an arrhythmia via its own initiative, sudden death will occur. Perturbations of cardiac electrical depolarization may have several mechanisms, any or several of which may be operations (O’Rourke and Lathers 1990). Time-frequency domain analyses of heart rate variability in patients with epilepsy and time-frequency mapping of R–R intervals during partial seizure may provide procedures to assess autonomic activity related to risk factors for SUDEP (Novak et al. 1999; Everengul et al. 2005; Persson et al. 2007). Cardiac-speciἀc overexpression of Sema3a in transgenic mice (SemaTG) associated with reduced sympathetic innervation and attenuation of the epicardial-to-endocardial innervation gradient. SemaTG mice demonstrated sudden death and susceptibility to ventricular tachycardia, due to catecholamine supersensitivity and prolongation of the action potential duration. Concluded appropriate cardiac Sema3a expression is needed for sympathetic innervation patterning and is critical for heart rate control (Ieda et al. 2007).

Altered postganglionic cardiac sympathetic innervation may increase risk of cardiac abnormalities and/or SUDEP (Lathers et al. 1990, Chapter 22).

(continued)

Neurocardiologic Mechanistic Risk Factors in SUDEP

7

Table 1.1â•… Pathophysiological Mechanisms (Continued) Model, Symptoms, and/or Changes Amygdaloid kindled seizure effect on cardiovascular system was examined in rats. An abrupt 50% increase in mean arterial pressure (BP) lasting 20–30 s after initiation of seizure occurred with profound bradycardia, characterized by a rate about half of that recorded before stimulation. Changes in heart rate and BP observed during amygdaloid kindled seizures are similar to those observed during secondary spontaneous seizures. Effects apparently are independent of kindling stimulus because stimulusinduced cardiovascular changes were not present at the beginning of kindling. Use of electrical stimulation as a therapy for epilepsy is currently being studied in experimental animals and in patients with epilepsy. This study examined the effect of preemptive, low-frequency, 1-Hz sine wave stimulation (LFS) on incidence of amygdala kindled seizures in rats. Amygdaloid kindled seizures in unanesthetized rats induced abrupt elevation of blood pressure accompanied by a signiἀcant decrease in heart rate.

Mechanisms of Sudden Death Results suggest the kindling seizure model is useful to study underlying mechanisms of seizure-induced cardiac arrhythmias and possibly the clinical phenomenon of SUDEP (Vindrola et al. 1984; Goodman et al. 2005).

Dramatic decrease in incidence of stage 5 seizures in fully kindled animals after preemptive LFS suggests low-frequency stimulation may be an effective therapy for prevention of seizures in patients with epilepsy (Gary-Bobo and Bonvalette 1977). Muscarinic receptor blockade with atropine (1 mg/kg, i.v. abolished seizure-induced bradycardia. Seizure-induced hypertension was unaffected by beta-adrenergic blockade with timolol (1 mg/kg, i.v.), but reduced by phentolamine (5€mg/kg, s.c., an alpha-adrenergic receptor antagonist). Chemical sympathectomy was induced with 6-hydroxydopamine (100 mg/kg, i.v.), an agent that does not cross the blood–brain barrier, eliminated the pressor response but did not completely block seizure-induced bradycardia. Effectiveness of 6-hydroxydopamine was tested with tyramine (0.5 mg/kg, i.v.), an agent that releases endogenous catecholamines. Results indicate amygdaloid kindled seizures activate both branches of the autonomic nervous system. Bradycardia was mediated by the parasympathetic system; the pressor response was caused by an increase in peripheral resistance due to alpha-adrenergic receptor activation. Findings show kindling is a useful seizure model for future studies on the effect of seizures on cardiovascular function and possible mechanisms of seizure-related sudden unexplained death (Goodman et al. 1989, 1990).

Source: Lathers, C. M., P. L. Schraeder, M. W. Bungo, in Psychological Factors and Cardiovascular Disorders, ed. L. Sher, Nova Science Publishers, Inc., Hauppauge, NY, 2008. With permission.

8 Sudden Death in Epilepsy: Forensic and Clinical Issues

1986a, 1986b, 1986c, 1987, 1993, 2008a, 2008b, 2008c). In addition to the risk of neurogenically induced arrhythmias, neurogenic apnea could be associated with SUDEP. Table 1.2 summarizes the SUDEP risk factors of respiratory/autonomic changes and hypoxia. Rare clinical case reports describe incidences of apnea associated with epileptiform activity (So et al. 2000). Additionally, a well-known sheep model of epilepsy (Simon et al. 1982; Table 1.2â•…Respiratory Factors and Hypoxia Model, Symptoms, and/or Change Postmortem found multiple areas of pulmonary punctuate hemorrhages and large areas of gross hemorrhage and edema in animals dying after induced epileptogenic activity, asystole, or ventricular ἀbrillation (Lathers and Schraeder 1982; Lathers et al. 1984; Carnel et al. 1985; Schraeder and Lathers 1983). In addition to the general and neurological risk factors, there is increasing evidence that cardiac (Aurlien et al. 2009; Kerling et al. 2009; Pezzella et al. 2009; Strzelczyk et al. 2008) and pulmonary (Scorza et al. 2007; Tavee and Morris 2008) changes additionally predispose a person to SUDEP. As with neurological risk factors, cardiac and pulmonary risk factors have neither been investigated by prospective observational follow-up studies nor by intervention studies. The pathomechanisms are still not established (Johnston and Smith 2007). Finsterer and Stöllberger (2010, this book) discuss recent ἀndings and practical implications concerning potential cardiac and pulmonary risk factors and pathomechanisms of SUDEP.

Changes in cardiac function alter cerebral blood flow, which, in turn, produces central hypoxia resulting in epileptogenic activity.

Mechanisms of Sudden Death Tissue hypoxia, hypercarbia, and alterations in acid–base balance may have contributed to the results in our model of experimental epilepsy. Acid–base balance was maintained only within physiological range before the initiation of epileptogenic activity.

So far, there is minimal evidence that any primary pulmonary disease could be a deἀnite risk factor for SUDEP (Finsterer and Stöllberger, 2010, this book). It is also unknown if patients with muscular respiratory insufficiency are at increased risk not to survive a tonic–clonic seizure. These patients appear particularly endangered because they often also have epilepsy and their epilepsy is often difficult to treat. Despite this uncertainty about pulmonary risk factors, there are frequent reports about patients who develop severe pulmonary problems during or after seizures, such as ictal hypoxemia or hypercapnia (Bateman et al. 2008), apnea (Bell and Sander 2006; Jehi and Najm 2008; Ryvlin et al. 2009), acute neurogenic pulmonary edema (Jehi and Najm 2008), or postictal laryngospasm (Tavee and Morris 2008). In a prospective autopsy series on 52 SUDEP patients, 80% had pulmonary congestion and edema (Leestma et al. 1989). It is uncertain if these abnormalities are due to primary pulmonary, cardiac, or laryngeal mechanisms. It is also unclear if only patients with previous lung disease, as opposed to previously healthy subjects, develop such problems. There is little information available about the effects of generalized tonic–clonic seizures on the respiratory system in general. Do generalized seizures induce bronchospasm or loss of tone of the muscles involved in respiration? Recent investigations have shown that at least the vital capacity, forced vital capacity, and forced expiratory volume within the ἀrst second (FEV1 etc.), are not signiἀcantly different between healthy subjects and epilepsy patients (Scorza et al. 2007) (Finsterer and Stöllberger 2010, this book). Some patients exhibit changes in cardiovascular status preceding the onset of convulsions (Schott et al. 1977; Schraeder et al. 1983). (continued)

Neurocardiologic Mechanistic Risk Factors in SUDEP

9

Table 1.2â•…Respiratory Factors and Hypoxia (Continued) Model, Symptoms, and/or Change The pulmonary edema model of status epilepticus in unanesthetized, chronically instrumented sheep in which sudden death and pulmonary edema occur.

Audiogenic seizures: respiratory arrest. Central alveolar hypoventilation syndrome (Ondine’s curse). Failure of automatic involuntary respiration with preservation of voluntary respiratory drive (Ondine’s curse) is rare, reported following a variety of morphologic lesions near respiratory centers in the lower brainstem. Risk factors include uncontrolled convulsive seizures, as well as respiratory and cardiac factors relating to treatment and supervision.

Epilepsy-related hypoxia. Ictal apnea. Postictal central apnea appears to be one potential mechanism for SUDEP. A 55-s convulsive seizure occurred in a 20-year-old female as she underwent video-EEG monitoring (So et al. 2000). Persistent apnea then developed. Electrocardiogram monitoring rhythm was not altered for the

Mechanisms of Sudden Death Catecholamine levels and seizure type and duration did not differ between animals dying suddenly or those surviving. Benign arrhythmias were generated in all animals; in no case did an arrhythmia account for death of an animal. Striking hypoventilation demonstrated in sudden death group but not in surviving animals. Differences in peak left atrial and pulmonary artery pressures, and in extravascular lung water; pulmonary edema did not account for the demise of sudden death animals. Thus, this model of epileptic sudden death supports a role of central hypoventilation in etiology of sudden unexpected death and shows an association, albeit not fatal, with pulmonary edema. The importance of arrhythmia in its pathogenesis is not conἀrmed (Simon et al. 1982; Johnston et al. 1995, 1997). Respiratory arrest mechanisms, modulated in part by serotonin, may cause SUDEP (Tupal and Faingold 2006a, 2006b). Central alveolar hypoventilation syndrome causes sudden death in a 39-year-old woman with heterotopia of the inferior olive (Matschke and Laas 2007). Neuropathologic examination disclosed preexisting malformation of the lower brain stem and acute local subarachnoid bleeding. Both respiratory and cardiac mechanisms are important. The apparent protective effect of lay supervision supports a role for respiratory factors, in part amenable to intervention by simple measures. Malignant tachyarrhythmias are rare during seizures and sinus bradycardia/arrest, although infrequent, occurs. Both types of cardiac arrhythmias can have a genetic basis as a contributory factor. Authors explore the potential of coexisting liability to cardiac arrhythmias as a contributory factor, but acknowledge that bridging evidence between inherited cardiac gene determinants and SUDEP is lacking (Terrence et al. 1981; Coulter 1984; Schraeder 1987; Nashef et al. 2007). Central apnea with seizures (Schraeder 1987) and neurogenic pulmonary edema and adult distress syndrome (Terrence et al. 1981). (Penἀeld and Jasper 1954; Bobo and Bonvallet 1975) So et al. (2000) note that although epileptic seizures may be associated with arrhythmogenic actions at the heart, in this patient the mechanism of marked central suppression of respiratory activity after seizures was clearly involved and almost resulted in sudden death. This case and Dr. Schraeder’s (1983) case highlight that both respiratory and cardiac changes do occur in persons with epilepsy. The (continued)

10 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 1.2â•…Respiratory Factors and Hypoxia (Continued) Model, Symptoms, and/or Change ἀrst 10 s, then it gradually and progressively slowed and stopped 57 s later. Cardiorespiratory resuscitation was successful. No evidence of airway obstruction or pulmonary edema was noted. One previous cardiorespiratory arrest after a complex partial seizure without secondary generalization had been reported for this patient. Postictal respiratory arrest was induced by serotonin receptor inhibition and prevented by selective serotonin reuptake inhibitor drugs (Tupal and Faingold 2006a, 2006b). Audiogenic seizure in mice. Obstructive sleep-apnea–induced cardiovascular complications.

Likely processes of the sudden infant death syndrome (SIDS) are identiἀed (apnea, failed arousal, failed autoresuscitation, etc.). The way in which epidemiological risk factors, genetics, neurotransmitter receptor defects, and neonatal cardiorespiratory reflex responses interact to lead to sudden death during sleep is unclear. 5-HT SIDS: Role of medullary 5-HT abnormalities in pathogenesis of SIDS, putative in utero origin of these abnormalities, and environmental and genetic factors interact to ultimately result in sudden death of the infant (Patterson 2010, this book).

Mechanisms of Sudden Death timing of events such as seizures, respiratory and/or laryngospasm, and cardiac EKG changes do vary in different patients.

The role of serotonin in SUDEP must be examined in future animal studies, including the DBA mice model of SUDEP (Faingold et al. 2010, this book). Suggested for use to study postictal respiratory arrest (So 2008). Implicated in pathogenesis of various cardiovascular diseases, including systemic hypertension, coronary artery disease, congestive heart failure, pulmonary hypertension, stroke, and cardiac arrhythmias. Mechanisms by which obstructive sleep apnea affects the cardiovascular system may involve mechanical effects on intrathoracic pressure, increased sympathetic activation, intermittent hypoxia, and endothelial dysfunction (Malow et al. 2000; Jain 2007). It is hypothesized that the neurophysiological basis of SIDS resides in a persistence of fetal reflex responses into the neonatal period, such as ampliἀcation of inhibitory cardiorespiratory reflex responses and reduced excitatory cardiorespiratory reflex responses. Explores ways in which multiple subtle abnormalities interact to lead to sudden death and emphasizes difficulty of ante-mortem identiἀcation of infants at risk for SIDS (Leiter and Böhm 2007). Underlying vulnerability involves a developmental defect in brainstem serotonergic (5-HT) systems that results in failure of protective cardiorespiratory responses to potentially life-threatening, but normally occurring events (e.g., hypoxia, hypercapnia), in the infant during sleep (Patterson 2010, Chapter 5, this book).

Source: Lathers C. M., P. L. Schraeder, M. W. Bungo, in Psychological Factors and Cardiovascular Disorders, ed. L. Sher, Nova Science Publishers, Inc., Hauppauge, NY, 2008. With permission.

Johnston et al. 1995, 1997) found that although neurogenic pulmonary edema was commonly observed, the mechanism of death in the animals was central neurogenic hypoventilation. Likewise, syncope is a factor for sudden death (Table 1.3). It is evident that, in all likelihood, there are multiple contributing risk factors that, in unfortuitous combinations (including environmental circumstances) in individuals with epilepsy, may result in unexpected death (Nashef et al. 1998). This chapter consists of summary tables highlighting possible risk categories and mechanisms for risks. The reader is encouraged to use the table

Neurocardiologic Mechanistic Risk Factors in SUDEP

11

Table 1.3â•… Syncope Symptoms and/or Changes Syncope is a transient loss of consciousness and postural tone. Usually due to temporary, self-terminating global cerebral hypoperfusion. Important to differentiate from other nonsyncopal transient loss of consciousness attacks (Chen et al. 2008).

Although arising suddenly from prolonged recumbency or returning from weightlessness to earth’s gravity can result in syncope from orthostatic or vasovagal effects, there are many other possible causes. Cardiac causes are more likely to occur in the elderly; noncardiac causes are more common in the younger population. Cases described illustrate often unexpected mechanisms of syncope in otherwise healthy individuals. Cases of sudden collapse. At times, there are obvious cardiac etiologies or central nervous system etiologies; at other times, there are interactive cause and effects with cardiac disorders leading to central nervous system effects or visa versa. However, all too often there are subtle interplays with genetic predispositions, environmental interaction, therapeutic interventions, or unknown organic disease. As consequences are extreme, a thorough investigation and understanding of the mechanisms leading to syncope are of paramount importance. Psychogenic syncope and psychogenic seizures are common disorders but are difficult to identify. Headupright, tilt-table testing evaluates vasovagally mediated syncope and convulsive syncope. In eight patients with syncope and/or tonic–clonic motor activity, without changes in blood pressure and heart rate, transcranial Doppler cerebral blood flow velocity, and EEG monitoring revealed they were, in all instances, psychogenic or malingering. Combined long term ECG and EEG monitoring. Isolated cardiac rhythm abnormalities were noted in 21 patients, but none were symptomatic and no deἀnitive arrhythmias occurred. Isolated EEG abnormalities were noted in 11 patients, 5 of whom had EEG abnormalities consistent with seizure disorders. Simultaneous EEG and ECG abnormalities were seen in four patients. In two, a previously unsuspected etiology for syncope was found: seizures in one patient with heart disease and sinus pauses in another thought to have a seizure disorder.

Mechanisms of Sudden Death The most important screening tool in identifying mechanism(s) of syncope is a detailed history emphasizing a search for underlying disease, speciἀc associated circumstances, and pre- and post-event symptoms. The type of diagnostic studies (i.e., cardiac or neurologic) undertaken should be based on historical data. Seizures must be considered as a possible mechanism of otherwise unexplained loss of consciousness in nonelderly persons, including air crew members (Schraeder et al. 1994; Williams and Frenneaux 2007). Review recent research relevant to managing syncope in adults: syncope evaluation in the emergency department, effectiveness of a structured and standardized approach to syncope, role of the implantable loop recorder, and efficacy of nonpharmacological physical treatments (Chen et al. 2008; Bungo et al. 2010, this book). Clinical cases briefly explore the interactions of disordered electrical potentials in the brain and disordered electrical potentials in the heart (Lathers et al. 2010b, Chapter 22, this book).

It was concluded that patients who pass out or convulse during head-upright tilt without any change in physiologic parameters can be presumed psychogenic in origin and may be referred for psychiatric evaluation without further expensive diagnostic studies (Grubb et al. 1992). Combined ambulatory EEG/ECG monitoring may prove useful in evaluation of some patients with syncope (Beauregard et al. 1991).

(continued)

12 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 1.3â•… Syncope (Continued) Symptoms and/or Changes Vasovagal syncope. Much of the natural history is unknown. Study determined whether patients presenting for care had a recent increase in worsened syncope frequency. Insufficient cerebral perfusion is a common cause leading to a loss of consciousness with a critical reduction of blood flow to the reticular activating system. In neurally mediated syncope, a paradoxical reflex can occur that induces an increase in cerebrovascular resistance and contributes to the critical reduction of cerebral blood flow.

Mechanisms of Sudden Death Many syncope patients present for care after a recent worsening of their frequency of syncope (Franco 2007; Sheldon et al. 2007). Outlined anatomic structures involved in cerebral autoregulation, its mechanisms in normal and pathologic conditions, and noninvasive neuroimaging techniques used to study cerebral circulation and autoregulation. Emphasis placed on description of autoregulation pathophysiology in orthostatic and neurally mediated syncope (Franco 2007).

Source: Modiἀed and updated from Lathers C. M., P. L. Schraeder, M. W. Bungo, in Psychological Factors and Cardiovascular Disorders, ed. L. Sher, Nova Science Publishers, Inc., Hauppauge, NY, 2008. With permission.

of contents of this book to identify which chapters are pertinent to a given topic within each table. The reader is referred especially to chapters in this book discussing risks for SUDEP by Walczak (2010), arrhythmias (Lathers and Schraeder 1982, 1987; Lathers et al. 1987), LSP (Schraeder and Lathers 1983, 1989), respiratory issues (Scorza et al. 2007; Finsterer and Stöllberger 2010; Lathers and Schraeder 1982; Carnel et al. 1985), syncope (Sharma et al. 2010, Chapter 21, this book; Lathers et al. 2010b, Chapter 22, this book), genetics (Herreros 2010), stress (Lathers and Schraeder 2006), therapies (Lathers and Schraeder 2002, 2010; Lathers et al. 2003a, 2003b, 2003c; Tompson 2008, 2010; Ryvlin et al. 2009), and unusual factors that may modify SUDEP risk (Scorza et al. 2008a, 2008b, 2010a, 2010b; Calderazzo et al. 2009; Terra-Bustamante et al. 2009). All are encouraged to read Brodie and Holmes (2008) and the chapter presenting how to educate persons with epilepsy about their risk factors and how to help families and survivors of SUDEP deal with the unexpected loss of a person with epilepsy (Hanna and Panelli, 2010, Chapter 57). There is general agreement that central nervous and autonomic nervous system/cardiorespiratory interactions include arrhythmias and apnea (Tables 1.1 and 1.2). Complicating the nervous system/cardiac relationship are relatively recent discoveries of genetically determined predispositions to arrhythmias that may result in seizure-like events at the time of the acute cardiac dysfunction (Table 1.4). How these genetic cardiac predispositions interact with the central and autonomic nervous systems is not understood. It is generally accepted that certain associated clinical circumstances [e.g., male gender (Table 1.5), poorly controlled generalized tonic clonic seizures, use of multiple antiepileptic drugs, changing doses or drugs, withdrawal of antiepileptic drugs, and poor compliance with antiepileptic drug use] are associated with an increased risk of SUDEP (Table 1.6). That psychogenic factors associated with life stresses are risk factors for cardiac arrhythmias and sudden death is well recognized by cardiologists (Lathers and Schraeder 2006). Also, certain ethnic groups have an increased incidence for stress-related sudden deaths [e.g., bangungut in healthy Filipino men (Table 1.7)]. There has been almost no organized effort to determine the role of stress as a risk factor for SUDEP. The stress response involves acute or chronic increases in sympathetic neural activity. Furthermore, Scorza and colleagues have identiἀed unusual factors that may modify the risk for SUDEP such as ambient temperature, the lunar phases of the moon,

Neurocardiologic Mechanistic Risk Factors in SUDEP

13

Table 1.4â•… Genetic/Structural Mechanistic Factors Associated Changes Four main inherited arrhythmia syndromes that are thought to be responsible for sudden death. Two clinical cases exhibit both Brugada syndrome and epilepsy produced by sodium channel dysfunction (Lathers et al. 2010b, this book). Brugada syndrome is produced by a mutation in gene SCN5A, which encodes the alpha subunit of the cardiac sodium channel (Antzelevitch et al. 2005a, 2005b; Aurlien et al. 2009. Herreros (2010, Chapter 19) note that some epileptic syndromes (Graves 2006) are due to different mutations in genes encoding alpha subunits of neuronal sodium channels (SCN1A, SCN2A) or in the beta subunit (SCN1B), common for both cardiac and neuronal isoforms (Lehmann-Horn and Jurkat-Rott 1999). The long QT syndrome: A genetically transmitted cardiac arrhythmia due to ion channel protein abnormalities, affecting the transport of potassium and sodium ions across the cell membrane. Patients may present with syncope, seizures, or aborted cardiac arrest. Long QT syndrome is an important cause of unexplained sudden cardiac death in the young.

Cardiac hypertrophy: An independent predictor of cardiovascular morbidity and mortality, predisposition to heart failure, QT interval prolongation, and ventricular arrhythmias.

Mechanisms of Sudden Death Four syndromes: long QT syndrome, Brugada syndrome, short QT syndrome, and catecholaminergic polymorphic ventricular tachycardia reviewed by Herreros (2010, Chapter 19, this book). Additional clinical, electrocardiographic, and genetic studies are needed to improve individual risk stratiἀcation and to determine any relationship among sodium channel dysfunction, Brugada ECG, and idiopathic epilepsy. Two patients presented in the cases appear to support the possibility that a common pathophysiologic mechanism associated with sodium channel dysfunction may be common to ECG abnormalities of Brugada syndrome and some types of epilepsy. Many patients must be screened to conἀrm. Risks factors for sudden death must be deἀned and linked with mechanisms for death (Kornick et al. 2003; Lathers et al. 2008a, 2008b, 2008c; Herreros, 2010, Chapter 19; Lathers et al. 2010a, Chapter 20).

Diagnosis of long QT syndrome depends on an ECG showing a prolonged QT interval (Kiehne and Kauferstein 2007). Establishment of a registry and discovery of genetic mutations causing the syndrome contribute greatly to understanding this condition and impetus to understanding other inherited cardiac arrhythmias. Genotype-phenotype correlation studies allow risk stratiἀcation of long QT syndrome patients. Lifestyle modiἀcation to avoid triggers for malignant cardiac arrhythmias, and use of beta blockers, pacemakers, and implantable deἀbrillators may reduce mortality in these patients (Vohra 2007). Mutations of cardiac ion channel genes affecting repolarization cause the majority of congenital cases. Despite detailed molecular characterizations of mutated ion channels, understanding how individual mutations may lead to arrhythmias and sudden death requires study of intact heart and modulation by autonomic nervous system. Studies of molecularly engineered mice with mutations in genes known to cause long QT syndrome in humans and speciἀc to cardiac repolarization in mice are reviewed (Salama and London 2007). Cardiac angiotensin II overproduction leads to long QT syndrome, resulting from IK1 potassium-dependent prolongation of action potential duration via modulation of channel subunit expression (Domenighetti et al. 2007). (continued)

14 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 1.4â•… Genetic/Structural Mechanistic Factors (Continued) Associated Changes Inherited arrhythmia syndromes, channelopathies.

Short QT syndrome: ECGs show a shortened QT interval and tall and narrow T waves. The cause is mutation of the potassium channel genes. Cardiac hypertrophy: associated with a dramatic change in gene expression proἀle of cardiac myocytes. Many genes important during development of the fetal heart but repressed in adult tissue are reexpressed, resulting in gross physiological changes and arrhythmias, cardiac failure, and sudden death. One transcription factor possibly important in repressing expression of fetal genes in the adult heart is REST (repressor element 1-silencing transcription factor).

Genetic factors in SUDEP: Likely to reflect underlying heterogeneous mechanisms common to the brain and heart, as has been shown to be the case with sudden cardiac death and sudden infant death syndrome (Ghali and Nashef 2010, this book).

Mechanisms of Sudden Death Long QT syndrome (Kiehne and Kauferstein 2007), short QT syndrome, catecholaminergic polymorphic ventricular tachycardia, Brugada syndrome, and overlapping phenotypes; established connections between these syndromes and idiopathic ventricular ἀbrillation (Sarkozy and Brugada 2005). Persons with short QT syndrome have increased familial risk of sudden cardiac death (Gaita et al. 2003, 2004). REST expression prevents increased BNP (Nppb) and ANP (Nppa) gene, encoding brain and atrial natriuretic peptides. Adult rat ventricular myocytes response to endothelin-1 and inhibition of REST results in increased expression of these genes in H9c2 cells. Increased expression of Nppb and Nppa correlates with increased histone H4 acetylation and histone H3 lysine 4 methylation of promoter-proximal regions of these genes. Deletions of individual REST repression domains combined activities of two domains of REST required to efficiently repress transcription of Nppb gene. A single repression domain is sufficient to repress Nppa gene. Data provide insight into molecular mechanisms for changes in gene expression proἀle cardiac hypertrophy (Bingham et al. 2006, 2007). Overlapping mechanisms. The increasing number of mongenically associated epilepsy syndromes raises the question of how epilepsy and arrythmogenic genetic disturbances may be concurrent. The association of uncontrolled epilepsy with risk of SUDEP leads to speculation that in some cases of idiopathic epilepsy, a genetic mutation may also cause a predisposition to fatal cardiac arrhythmias. The recent association of long QT syndrome with epilepsy suggests that ion channel gene mutations may be inherited susceptibility factors for neurogenic cardiac arrhythmia in some persons with epilepsy. Further investigation into overlap of epileptogenic and arrhythmogenic epidemiological and genetic factors is warranted.

and other factors (Table 1.8). As is evident from the listings in the above-mentioned tables, there is a likelihood that common risk factors for SUDEP and arrhythmogenic cardiac disease are related to centrally initiated peripheral autonomic dysfunction in association with epileptiform discharges and stress. Issues that need resolution include a better understanding of the individual risks, the mechanisms of cardiac arrhythmia and arrest in persons without a previously identiἀed structural heart disease, a deἀnition of abnormal interactions between the central nervous system (CNS) and the heart, the role of neurogenic pulmonary edema and central apnea in combination with cardiac autonomic neural and subtle anatomic and genetic factors as risk for SUDEP, and development of primary and secondary preventive measures along with educational programs to disseminate essential information to physicians, patients, and families.

Neurocardiologic Mechanistic Risk Factors in SUDEP

15

Table 1.5â•…Risk Factors for SUDEP SUDEP Risk Factors in a Swedish Population (Nilsson et al. 2001)

Retrospective Study of Risk Factors, United States Reviewed association between several clinical variables and SUDEP to elucidate risk factors. Characteristics of 67 cases correlated with published previous studies. Education of physicians about existence of SUDEP and risk factors is imperative to improving patient education and reduction in mortality (Lear-Kaul et al. 2005). Behavioral Risk Factor Surveillance System Data from South Carolina Compared health insurance coverage, health care visits, health-related behaviors among persons with epilepsy vs. general population (Ferguson et al. 2008).

Position of Patients at Night (Kinney et al. 2009; Monter et al. 2007; McGregor and Wheless 2006).

1. Higher number of seizures per year (relative risk of 10.16 in patients with more than 50 seizures/year compared to no more than 2 years) 2. Increased number of antiepileptic drugs (9.89 for three drugs vs. monotherapy) 3. Early vs. late onset epilepsy (7.72) 4. Frequent changes in antiepileptic drug dosage vs. unchanged dosage (6.08) 5. Risk and early onset and SUDEP risk and seizure frequency was weaker for females 6. Frequent dosage changes had a stronger association in females 7. Early age of onset and male sex 1. Long history of seizure disorder 2. Under medication or poorly controlled seizure activity 3. Male gender 4. Age younger than 40 years 5. Mental or physical stress

Persons with epilepsy: 1. Are more likely to smoke. 2. Have less physical activity. 3. Need better access to health care. 4. Need interventions focused on smoking cessation and increase in physical activity. 5. Lack money to visit a doctor. One-third of respondents with active epilepsy reported in past 12 months needed to see a doctor but could not because of cost. Note: Authors do not relate these behavior risk factors to increased likelihood of SUDEP per se, but one could also consider them to be risk factors for SUDEP. 1. Especially important in persons with uncontrolled epilepsy in institutional settings. 2. Avoid prone position and soft pillows when sleeping. 3. Use respiration monitors during sleep. 4. Train caregivers to be vigilant and to intervene to prevent respiratory compromise for institutionalized and at-home patients.

1.3â•… SUDEP Animal Models The importance of using many different animal models to study SUDEP in order to glean insights into the various mechanisms of risks and their contribution to the initiation of the death event is discussed by Lathers (2010a, Chapter 25) (Table 1.1). So (2008) emphasizes the signiἀcance of using audiogenic seizure mice to study postictal respiratory arrest. Postictal respiratory arrest was induced by serotonin receptor inhibition and prevented by

16 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 1.6â•… Therapies: Increased or Decreased Risk Symptoms and/or Changes

Mechanisms of Sudden Death

Increased risk SUDEP The putative advantage of new drugs is a smaller spectrum of Antiepileptic drugs—New antiepileptic possible adverse events, such as sedation and some may drugs, e.g., topiramate and lamotrigine, minimize noncompliance by reducing side effects of developed for chronic focal and secondarily lethargy and cognitive impairment. Many new antiepileptic generalized epileptic seizures. Therapeutic drugs do have less frequent interactions, leading to efficacy of these drugs does not seem to be improved tolerability with comedication (Lathers et al. superior to traditional anticonvulsants such 2003a, 2003b, 2003c; Walczak 2003). However, difficulty in as phenobarbital. achieving therapeutic dosage with some of the newer antiepileptic drugs because of side effects makes one question whether some newer agents are better than the older antiepileptic drugs. Suggests SUDEP rates reflect population rates and not a Risk of SUDEP rates in patients on speciἀc drug effect. FDA requires warning labels on the risk lamotrigine, gabapentin, topiramate, of SUDEP in association with use of each of the mentioned tigabine, and zonisamide are similar to drugs (Lathers and Schraeder 2002). those on standard antiepileptic drugs. Compliance with antiepileptic drugs In a coroner’s office review of forensic cases in Allegheny important in prevention of SUDEP. County for the year 2001, Lathers et al. (2003a, 2003b) found low, or no, levels of antiepileptic drugs postmortem in persons with epilepsy who died of SUDEP. Hughes (2009) deemed the most important SUDEP risk Noninterventional, single-arm study (Martin factor to be noncompliance with antiepileptic medication. et al. 2009). Explored the effectiveness and Ryvlin et al. (2009) found the risk of SUDEP is increased in behavioral outcomes in intellectually patients who have poor compliance and exhibit nocturnal disabled patients with topiramate for seizures and generalized tonic–clonic seizures. However, epilepsy. compliance is not the only therapeutic question to ask about victims of SUDEP. Others include: Is the correct dose of the antiepileptic drug being used? Was more than one drug prescribed and were there recent dose drug changes? All agree that maintenance of a stable therapeutic drug levels is crucial to avoid SUDEP. Some improvement in nearly all behavioral aspects was observed under concomitant topiramate therapy. In addition, seizure frequency decreased. Two unexpected deaths in patients taking the antiepileptic drug topiramate were attributed to SUDEP (Martin et al. 2009). A prolonged QT interval leads to increased risk for QT-prolonging drugs and risk of SUD. development of ventricular tachyarrhythmias, particularly Torsades de pointes is associated with early polymorphic ventricular tachycardia (torsades de pointes). cardiac depolarization. Drugs that prolong Polymorphic arrhythmia may rapidly develop into QT interval: class III antiarrhythmic agents, ventricular ἀbrillation and cause sudden death antimicrobial agents (fluoroquinolone and (Reingardiene and Vilcinskaite 2007). macrolide antibiotics), antipsychotic and antidepressant drugs, agents used in general anesthesia, and antimycotics (Reingardiene and Vilcinskaite 2007). Adverse cardiovascular effects of antipsychotic treatment: Cardiovascular Effects of Antipsychotics tachycardia, orthostatic hypotension, and rarely, SUD, Older antipsychotic literature was primarily muscarinic cholinergic antagonism, alpha(1)-adrenergic concerned with physiological consequences antagonism, or receptors associated with cardiac of muscarinic cholinergic antagonism, (continued)

Neurocardiologic Mechanistic Risk Factors in SUDEP

17

Table 1.6â•… Therapies: Increased or Decreased Risk (Continued) Symptoms and/or Changes

Mechanisms of Sudden Death

Alpha(1)-adrenergic antagonism, or receptors associated with cardiac conduction, but current literature recognizes that, for most antipsychoticexposed patients, the more signiἀcant cardiovascular burden of treatment is mediated by metabolic adverse effects such as weight gain, dyslipidemia, and diabetes mellitus.

conduction, metabolic adverse effects of weight gain, dyslipidemia, and diabetes mellitus (Michelsen and Meyer 2007). High-dose methadone can induce QT prolongation by hERG inhibition, resulting in QT prolongation, but new evidence shows that QT prolongation can occur at much lower doses, even when the drug is not given IV.

Methadone produces QT prolongation, arrhythmias, and sudden death (Karch 2010, this book, Chapter 10).

May be acting directly on myocardial conduction to produce arrhythmia and death (Lathers and Lipka 1986, 1987; Lathers et al. 1986a; Lipka and Lathers 1987; Lipka et al. 1988).

Chlorpromazine or thioridazine do not appear to produce arrhythmia or death via a central locus in an experimental cat model. Decreased risk SUDEP The question must be asked if persons thought to be at risk for SUDEP should be placed on a beta blocker, in addition to the prescribed anticonvulsant(s). Beta blockers are also used to reduce stress and persons with epilepsy generally are stressed by the disease and associated problems (Lathers and Schraeder 2006). Beta blockers produce anticonvulsant activity whether administered via the intraosseous route or intravenously (Jim et al. 1988, 1989; Spivey et al. 1987a, 1987b; Lathers et al. 1989, 1990, 2008). Note the intraosseous and endotracheal routes of administration of antiepileptic drugs may be used when IV access is not available in status epilepticus. These routes are appropriate for emergency room pediatric seizures or cardiac arrest when unable to establish a traditional IV line (Rusli et al. 1987; Jim et al. 1988, 1989; Spivey et al. 1987a, 1987b; Lathers et al. 1989, 1990, 2008). Postmortem found no serious pathology in the tibia bone after intraosseous administration, other than the expected needle track (Lathers et al. 1989). Selective serotonin reuptake inhibitor drugs. Postictal respiratory arrest was induced by serotonin receptor May be protective in some persons at risk inhibition and prevented by selective serotonin reuptake for SUDEP with a component of apnea. inhibitor drugs. The role of serotonin in SUDEP must be examined in future animal studies (Tupal and Faingold 2006). Vagal nerve stimulation may decrease risk of SUDEP by 50% Vagal nerve stimulation may decrease (Annegars et al. 2000). susceptibility to ventricular tachycardias (Verrier et al. 2009; Verrier and Schachter 2010). Beta blockers exhibited anticonvulsant activity.

Source: Lathers C. M., P. L. Schraeder, M. W. Bungo, in Psychological Factors and Cardiovascular Disorders, ed. L. Sher, Nova Science Publishers, Inc., Hauppauge, NY, 2008. With permission.

18 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 1.7â•… Psychological Factors Symptoms and/or Changes Scared to death. A 21-year-old student had generalized tonic–clonic seizures induced by mental image of human pain. One ictal event occurred while listening to a description of suffering, as read from Fox’s Book of Martyrs. While again listening to the offending passage during EEG and ECG monitoring, had 25 s of asystole terminating in electrocerebral silence and a generalized tonic, tonic–clonic seizure. A 24-hour ambulatory monitor recorded episodes of progressive sinus bradycardia concomitant with P–R interval prolongation and Wenckebach atrioventricular block. Sinoatrial conduction times and sinus node recovery times were normal on atrial pacing (Schraeder et al. 1983). Emotional trauma and psychological stress. Stress response is associated with increased sympathetic activity and catecholamine levels that may be associated with increased risk of cardiac arrhythmias, especially in context of epileptiform cerebral discharges (Pickworth et al. 1990). Stress is associated with changes in autonomic neural function. Its role as a potential risk factor for SUDEP is not known. Association of epilepsy with cardiac abnormalities, such as neurogenic arrhythmias and microscopic perivascular and interstitial ἀbrosis, and with depression and anxiety indicates emotional stress with related increases in catecholamines should be evaluated as a potential risk factor for SUDEP. Stress and anger. Compare morphology and initiation pattern between ventricular arrhythmias that are triggered by anger and those that are not. At the time of shock, patients with implantable cardioverter-deἀbrillators recorded levels of deἀned mood states preceding the shock in a diary. Stress risk factors and triggers of sudden death. An autopsy proven case-control study (Owada et al. 1999).

Mechanisms of Sudden Death Since implantation of a permanent pacemaker, has been asymptomatic. Patient demonstrates advantages of reproducing the circumstances associated with an unexplained loss of consciousness while monitoring EEG and ECG (Schraeder et al. 1983).

Arrhythmogenic effects of efferent sympathetic drive precipitate cardiac arrhythmia and sudden death. Patients with preexisting heart disease are particularly at risk. Generation of proarrhythmic activity patterns within cerebral autonomic centers may be ampliἀed by afferent feedback from a dysfunctional myocardium (Gray et al. 2007). Impact of adverse emotional states on autonomic control of cardiac rhythm is a known important factor leading to cardiac dysrhythmias in humans and other species. Interaction between emotional factors and arrythmogenic potential of epileptiform discharges and possibility of beneἀt from stress management intervention need to be investigated (Lathers and Schraeder 2006). Ventricular arrhythmias occurring in setting of anger are more likely pause dependent and polymorphic. Suggests that in predisposed populations anger may create an arrhythmogenic substrate susceptible to more disorganized rhythms, a possible mechanism linking emotion and sudden death (Stopper et al. 2006). Identiἀed risk factors and triggers of sudden death in cases where causes of death were deἀnitely proven by autopsy in Japan. Legal and medical records for 4 years were investigated. Of 271 cases, 176 patients, 20 to 59 yrs-old were cases of sudden death in working generations. Of these, 91 cases, 52% could be analyzed by telephone interviews from close family members. One examiner undertook all phone interviews with the case subjects. As control subjects, 1167 employed persons who consulted for a health check. Of sudden death cases, ἀnal diagnosis in 29 cases was coronary artery disease (31.9%), 18 acute cardiac dysfunction (continued)

Neurocardiologic Mechanistic Risk Factors in SUDEP

19

Table 1.7â•… Psychological Factors (Continued) Symptoms and/or Changes

Retrospective study. Reviewed association between several clinical variables and SUDEP to elucidate risk factors. Characteristics of 67 cases correlated with published ἀndings in previous studies (Lear-Kaul et al. 2005).

Mechanisms of Sudden Death (19.8%), 6 other cardiac diseases (6.6%), 4 acute aortic dissection (4.4%), 4 cerebrovascular disease (4.4%), and 30 other diseases (32.9%). Through conditional logistic analysis, the following risk factors emerged as candidates: Long-term stress, history of heart disease, hypertension, chest symptoms, autonomic disturbance, short-term stress, and a smoking habit. Short-term stress, autonomic disturbance and a smoking habit increased the risk of sudden death due to coronary artery disease. Long-term stress was associated with an increased risk of sudden death due to acute cardiac dysfunction. Demonstrated that autonomic disturbance and stress were closely related to occurrence of sudden death (Owada et al. 1999). Attributes that deἀne an at-risk group of epileptics include: less than 40 years old, male, long history of seizure disorder, under medication or poorly controlled seizure activity, and mental or physical stress. Education of physicians about the existence of SUDEP and risk factors is imperative in improving patient education and reduction in mortality (Lear-Kaul et al. 2005).

Source: Lathers C. M., P. L. Schraeder, M. W. Bungo, in Psychological Factors and Cardiovascular Disorders, ed. L. Sher, Nova Science Publishers, Inc., Hauppauge, NY, 2008. With permission.

selective serotonin reuptake inhibitor drugs. The possible role of serotonin in SUDEP must be examined in future animal studies (Faingold et al. 2010, Chapter 41). The ἀring pattern of cardiac parasympathetic neurons, as well as cardiac sympathetic postganglionic nerves, during an epileptic attack has been shown to change and has been termed the lockstep phenomenon (Lathers et al. 1987; Stauffer et al. 1989, 1990; Dodd-O and Lathers 1990; O’Rourke and Lathers 1990). The observation of autonomic neural discharges time-locked to cortical epileptiform is evidence that epileptogenic activation of the cardiac parasympathetic nerves, revealed by ictal bradyarrhythmias or cardiac asystole, may be one contributing cause of sudden death of patients with epilepsy. Likewise, epileptogenic activation of the cardiac sympathetic nerves may be another contributing cause of SUDEP. An imbalance between the two systems’ (i.e., the cardiac parasympathetic and the cardiac sympathetic) neural discharge patterns may contribute to SUDEP (Lathers and Schraeder 1982, 1986; Schraeder and Lathers 1983, 1989). Wang et al. (2006) provided supporting data for the LSP ἀnding. They examined blockade of inhibitory neurotransmission evoked seizure-like ἀring of cardiac parasympathetic neurons in brainstem slices of newborn rats. Speciἀcally, blockade of GABAergic and glycinergic receptors in medulla slices evoked intermittent seizure-like ἀring of cardiac parasympathetic neurons, suggesting the seizure-like pattern of ἀring during an epileptic attack may cause neurogenic ictal bradyarrhythmias, cardiac asystole, or even sudden death in persons with epilepsy. Nonuniform autonomic cardiac postganglionic neural discharges are associated with coronary occlusion of the left anterior descending coronary artery and/or ouabain toxicity in cats (Lathers et al. 1974a, 1974b, 1977a, 1977b, 1978; Lathers 1980, 1981). Nonuniform

20 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 1.8â•…Unusual Risk Factors Modify SUDEP Factor Unusual potential risk factors for SUDEP were examined by Scorza and colleagues (2008a, 2008b, 2010a, 2010b), Calderazzo et al. (2009), and Terra-Bustaman et al. (2008).

Physical activity (Scorza et al. 2008a).

SUDEP seems to occur more commonly during sleep (Asadi-Pooya and Sperling 2009). Chronobiology of acute pulmonary edema (Bilora et al. 1998).

Mechanism of Sudden Death Studies are needed to examine the role of unusual factors that are listed below to determine if they are applicable to SUDEP. These include life style modifying interventions with accepted public health beneἀts, but as yet with no consensus that they may or may not prevent sudden death. Animal studies and clinical studies are needed (Lathers et al. 2008b, reply; 2010c, Unanswered question). May modify the role of autonomic effects and cardiorespiratory disturbances. Beneἀts of physical activity may be related to reductions in sympathetic activity. Morbidity and mortality in cardiovascular disease are often associated with elevations in sympathetic activity (Scorza et al. 2008a; Schraeder et al. 1983). Increased physical action does have a beneἀcial cardiovascular effect. The question is, will increased physical activity be of beneἀt if one assumes cardiovascular sympathetic dysfunction or insufficiency in persons with epilepsy? In general, regular exercise does not have a downside, if the patient is cardiovascularly ἀt. The role of regular exercise in prevention of SUDEP is not established (Lathers et al. 2008b reply; 2010c, this book). Autonomic cardiorespiratory disturbances. Alterations in clinical parameters including ECG changes, blood pressure, respiration, and vasomotor tone (Hirsch and Martin 1971; Lathers and Schraeder 1982; Schraeder and Lathers 1983; Wannamaker 1985). ECG changes in experimental epilepsy include heart rate changes, arrhythmias, conduction blocks, altered ECG morphology, and QT interval). See discussion of Bilora et al. (1998) below and Chapter 23 by Hughes and Sato (2010, this book). Found pulmonary edema incidence higher during the night. No signiἀcant weekly or circannual rhythms detected. Examine possible relations with decreased heart output, increased sympathetic tone, or partial baroceptor desensitization occurring at night (Bilora et al. 1998). Evaluation of some cases of SUDEP and ners-SUDEP during long-term electroencephalography monitoring suggest that autonomic instability ending in cardiorespiratory arrest may be triggered by postictal suppression rather than by ictal activation of the autonomic nervous system. More epidemiologic studies on high-risk populations of persons with epilepsy are needed (Schuele 2009). The increase sympathetic tone or desensitization of baroceptors may contribute to occurrence of cardiac autonomic arrhythmias. See discussions of interictal activity and cardiac arrhythmias (continued)

Neurocardiologic Mechanistic Risk Factors in SUDEP

21

Table 1.8â•…Unusual Risk Factors Modify SUDEP (Continued) Factor

Seizure activity and neurogenesis (Scorza et al. 2008b). Omega-3 fatty acid nutritional deἀciency (Scorza et al. 2008a).

Low temperature (Scorza et al. 2008a).

Winter temperatures may lead to cardiac abnormalities and, hence, to sudden death. Are they a risk factor for SUDEP? (Colugnati et al. 2008).

Role of heart rate in rats with epilepsy during low temperature exposure (Sonoda et al. 2008). Minimum external temperature, mean external temperature (Bell et al. 2009). Climate fluctuations (Calderazzo et al. 2009). Sunlight may be a beneἀt in that vitamin D deἀciency may increase risk of SUDEP (Scorza et al. 2010a).

Mechanism of Sudden Death and baroreceptor changes by Lathers (2010a, 2010b) and cardiac and pulmonary risk factors and pathomechanisms of SUDEP by Finisterer and Stollberger (2010, Chapter€42). Aberrant dentate granule cell neurogenesis may influence negatively the cardiovascular system of a patient with epilepsy and lead to cardiac abnormalities and risk of SUDEP. Beneἀcial effect of nutritional aspects of omega-3 fatty acid may decrease cardiac arrhythmias and sudden death (SUD) in patients at risk for cardiac disease. No established answer at this time. See comments by Lathers et al. (2008b). Questions have been raised (Lathers et al. 2008a, 2008b, 2008c, 2010c) that need to be addressed: 1. Exactly how does one deἀne a low temperature? 2. What is the degree range used to deἀne a low temperature? 3. What is the duration at this low degree range? 4. Is there a dose response so that there is an increased risk of sudden death related to the lowest temperature? Mammals hibernate as a strategy to survive cold conditions. Hibernating mammals inherit a stable cardiovascular function. They show resistance to hypothermia at a cellular level, the membrane potential and excitability are more stable in their cardiac cells. Aortic smooth muscle cells maintain ionic gradients upon prolonged exposure to low temperature, cardiac myocytes from the mammals maintain constant levels of intracellular free calcium and forceful contractility at 10°C or lower. Postulate hibernating mammals have cardiovascular particularities that confer heart protection. The relevance to persons with epilepsy is to be determined. Low temperature increased the heart rate of patients with epilepsy. Authors suggest exposure to low temperatures could be a risk factors for cardiovascular abnormalities and, thus, for SUDEP. Found no evidence of association between either mean temperature group or minimum temperature group and SUDEP but there was a slight excess of SUDEP in the coldest (mean temperature) groups. May have an effect on SUDEP occurrence. Some clinical studies suggest low vitamin D levels may be associated with death from heart failure and sudden cardiac death. There also appears to be an association between low vitamin D and seizure occurrence. Scorza et al. suggest vitamin D may exhibit anticonvulsant action. More data are needed. (continued)

22 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 1.8â•…Unusual Risk Factors Modify SUDEP (Continued) Factor The lunar phase (Terra-Bustamante et al. 2009).

Lunar phase, month, season: no association found between these factors and SUDEP (Bell et al. 2009).

Mechanism of Sudden Death An epilepsy unit performed an 8-year analysis, examining possible associations between the phase of the moon and SUDEP. Incidence of SUDEP was highest at full moon (70%), followed by waxing moon (20%), and new moon (10%). No SUDEPs occurred during the waning cycle. Preliminary ἀndings suggest the full moon correlates with occurrence of SUDEP. Dataset of National Sentinel Clinical Audit of EpilepsyRelated Deaths examined all death certiἀcates of those who died in England and Wales between 9/1999 and 8/2000 to see if epilepsy was mentioned. Of 2412 deaths, 409 were identiἀed as probable SUDEP.

(increases, decreases, and/or no change) autonomic neural postganglionic cardiac sympathetic discharge traveling through the stellate ganglia, causes cardiac arrhythmias, ventricular ἀbrillation, and/or sudden cardiac death in the manner described by Han and Moe (1964), i.e., nonuniform recovery of excitability in ventricular muscle. It was suggested that the modiἀcation of neural discharge is a way for drugs to modify the occurrence of arrhythmia and death (Lathers et al. 1974a, 1974b, 1977a, 1977b, 1978; Lathers 1980, 1981; Han and Moe 1964). Nonuniform autonomic cardiac postganglionic neural discharge was also found to be associated with pentylenetetrazol-induced interictal epileptogenic activity in cats. Nonuniform autonomic neural discharge and autonomic neural imbalance of postganglionic cardiac sympathetic and vagal discharge cause cardiac arrhythmias, ventricular ἀbrillation, asystole, and/or SUDEP. Regional differences in the beta adrenoceptor densities reflects differences in postganglionic cardiac sympathetic innervation of the myocardium (Randall 1977, 1984; Randall et al. 1984). These site differences will vary the release of norepinephrine in various sympathetic nerve terminals and sites in the heart to modify cardiac contractile function and may trigger development of arrhythmias and/or sudden death. Innervation density of beta adrenoceptor densities is high in the subepicardium and central conduction system. Nonuniform postganglionic cardiac-sympathetic cardiac innervation is related to the nonuniform beta-sympathetic receptor locations in the heart and affects cardiac contractility and development of arrhythmias and/or death. In diseased hearts, cardiac innervation density varies and may lead to sudden cardiac death (Lathers et al. 1985, 1986a, 1986b, 1988, 1990, 2010; Lathers and Levin 2010, Chapter 33, this book). Seizures exert effects on cardiac function (Schuele, 2009). Since tachyarrhythmias are common during epileptic seizures, whereas bradyarrhythmias or asystoles occur less frequently, Kerling et al. (2009) evaluated cardiac postganglionic denervation in patients with epilepsy to determine if this was responsible for ictal asystole. I-(123)-metaiodobenzylguanidine was used as a marker of postganglionic cardiac norepinephrine uptake, using single-photon emission computed tomography. They concluded that a pronounced reduction in cardiac single-photon emission computed tomography uptake of asystole patients indicates postganglionic cardiac catecholamine disturbance. Impaired sympathetic cardiac innervation limits adjustment and heart-rate modulation and may increase the risk of asystole and, eventually, sudden unexpected death in persons with epilepsy. These data support the ἀnding of Lathers and colleagues described above. The

Neurocardiologic Mechanistic Risk Factors in SUDEP

23

reader is also referred to the chapter by Lathers and Levin (2010) that examines nonuniform sympathetic innervation and beta-sympathetic receptor locations and discusses how these factors affect cardiac contractility and the development of cardiac arrhythmias and/ or sudden death. Cerebral ischemia is associated with neuron degeneration. Accumulation of excess excitatory amino acids in the synaptic clef, activation of excitatory amino acid receptors, and influx of calcium into neurons are involved in the development of ischemia-induced neuronal death. Schwartz et al. (1995) hypothesized that neuroprotection may occur if inhibitory transmission via gamma-aminobutyric acid (GABA) is enhanced to offset excitation. Diazepam, a drug known to increase GABA-induced chloride channel opening, was studied in rats after hippocampal GABA levels had returned to post-transient global ischemia basal levels. The authors concluded that delayed enhancement of GABAergic neurotransmission directly at the site of vulnerability after an ischemic event does protect the vulnerable neurons from death. This model should be adapted to an animal model of seizure activity to study the effect of diazepam on GABA-mediated effects that may prevent ischemia-induced neuronal death, prevent the worsening of central neuronal communication due to epileptogenic activity, and eventually contribute a protective CNS effect that lessens individual at risk for SUDEP. Table 1.2 discusses postmortem multiple areas of punctuate hemorrhages and large areas of gross hemorrhage and edema in animals dying after inducing epileptogenic activity, asystole, or ventricular ἀbrillation (Lathers and Schraeder 1982; Schraeder and Lathers 1983; Lathers et al. 1984; Carnel et al. 1985). Tissue hypoxia, hypercarbia, and alterations in the acid–base balance may have contributed to the results in our model of experimental epilepsy. Acid–base balance was maintained only within physiological range before the initiation of epileptogenic activity (Lathers and Schraeder 1982, 1987; Schraeder and Lathers 1983, 1989; Carnel et al. 1985; Stauffer et al. 1989, 1990). Changes in cardiac function alter cerebral blood flow, which, in turn, produces central hypoxia that results in epileptogenic activity. Some patients exhibit changes in cardiovascular status preceding the onset of convulsions (Schott et al. 1977; Schraeder et al. 1983). So (2008) has emphasized risk factors of postictal apnea and hypoxia, with or without pulmonary congestion, in combination with generalized tonic–clonic seizures and, to a lesser degree, with complex partial seizures and respiratory arrest as a mechanistic cause of SUDEP. So et al. (2000) also noted that, although epileptic seizures may be associated with arrhythmogenic actions of the heart, in their patient, the mechanism of marked central suppression of respiratory activity after seizures was clearly involved and almost always resulted in sudden death. This case, and the case of Schraeder et al. (1983), highlight that both respiratory and cardiac changes do occur in persons with epilepsy. The timing of events such as seizures, respiratory and/or laryngospasm, and cardiac EKG changes does vary in different patients. Ryvlin et al. (2009) note that the pathophysiology of SUDEP is not clear, but postictal central or obstructive apnea is one likely mechanism. However, as noted by Lathers (2010b, Chapter 44, this book): Caution must be exerted when concluding respiratory changes alone are the primary mechanism of death. At the ἀrst onset of the clinical problem cardiac arrhythmias may be felt by the patient but may not be visually detected by a witness. ‘Invisible’ cardiac arrhythmias may be initiated and then followed by ‘visible’ respiratory distress. Therefore, in addition to repositioning of the patient to ensure ease of respiration and/or stimulation of respiration, it is important, if possible, also to simultaneously monitor and medically support cardiac rate and rhythm.

24 Sudden Death in Epilepsy: Forensic and Clinical Issues

The reader is referred to additional discussion of actual case histories focused on arrhythmogenic, respiratory, and psychological risk factors in Chapter 44 of this book (Lathers 2010b). Most likely, different mechanisms and/or a different combination of mechanisms are responsible for death in different persons with epilepsy. 1.3.1â•… Therapies as Factors Data suggest that beta blockers exert a protective effect against seizure induction and/ or the development of cardiac arrhythmias with interictal and ictal activity (Spivey et al. 1987b; Jim et al. 1988, 1989; Lathers et al. 1989a, 1989b, 1990, 2008a) (Table 1.6). The question must be asked whether persons thought to be at risk for SUDEP should be placed on a beta blocker, in addition to the prescribed anticonvulsant(s). Celiker et al. (2008) reported clinical experiences of patients with catecholaminergic polymorphic ventricular tachycardia and concluded that medical treatment with propranolol and verapamil may decrease the incidence of arrhythmia. If patients are still refractory, implantation of intracardiac deἀbrillators should be considered. Obviously, a delay in diagnosis or inadequate treatment can result in sudden cardiac death. The same is true for persons at risk for SUDEP. See the case history of SUDEP by Schraeder (2010). Hughes (2009) addressed the issue of how to predict patients at risk for SUDEP by reviewing published reports of such deaths. With a mean incidence of SUDEP at 1.8/1000, a mean standardized mortality ratio of 6.8, and a mean percentage of SUDEP cases among deaths from epilepsy at 16.6%, the problem of SUDEP and risk factors is an important issue to be resolved. Hughes identiἀed 17 risk factors and concluded, just as Lathers et al. (2008a, 2008b, 2008c) stated, that a cardiac or pulmonary problem may be a primary risk factor in different patients. He deemed the most important risk factor to be noncompliance with antiepileptic medication. Maintenance of therapeutic drug levels is crucial to avoid SUDEP. Ryvlin et al. (2009) also found that the risk of SUDEP is increased in patients who have poor compliance, nocturnal seizures, and generalized tonic–clonic seizures. However, compliance is not the only risk factor to be addressed if a person is a victim of SUDEP. There are several postmortem questions to be asked (Lathers and Schraeder 2009) about a patient on an antiepileptic drug that still becomes a SUDEP victim. One must inquire whether the correct dose of the antiepileptic drug was being used to control seizures. Another clinical pharmacology question that must be asked is whether the patient was on the correct antiepileptic drug to control the particular type or mixture of seizures experienced. When evaluating the role of drugs as protectors of life, clinical pharmacologists (Lathers and Schraeder 2002) caution us to remember that the use of all drugs is a risk/beneἀt ratio evaluation (Lathers et al. 2003a, 2003b, 2003c). Thus, the use of antiepileptic drugs may not provide 100% protection to the patient against sudden death.

1.4â•…Case of SUDEP with a Premorbid Diagnosis of Nonepileptic Seizures Table 1.8 discusses unusual risk factors examined by the laboratory of Scorza to determine if they are risk factors for SUDEP and/or have been speculated to be risk factors. This includes asking whether the lunar phase has an effect on SUDEP (Terra-Bustamante et al. 2009). A retrospective examination of the incidence of SUDEP in an epilepsy unit over an

Neurocardiologic Mechanistic Risk Factors in SUDEP

25

A 35-year-old woman was found dead in bed by her parents. She had a longstanding psychiatric history with the occurrence of seizure-like events. On one occasion,€she had a nonepileptic seizure in her neurologist’s office consisting of sliding to the floor with bizarre asynchronous bilateral motor activity without loss of consciousness. This event occurred consequent to an emotionally tense situation at her home. Past observation of another type of event consisting of some automatisms and postÂ� ictal confusion complicated the history. Multiple routine EEGs over the years were unremarkable, save for one that showed unequivocal isolated left temporal interictal discharges. The patient was placed on carbamazepine and found to have consistent therapeutic levels. The patient recognized that the complex partial events were no longer occurring, but that others continued. After the retirement of her neurologist, the patient was seen at another center and subjected to several days of inpatient EEG monitoring, during which multiple clinical events were observed without any epileptiform activity being documented on any of the EEG recordings. As a result, the patient was informed that her seizures were nonepileptic and was advised to taper her antiepileptic medications. Several weeks later her parents notiἀed her former neurologist of her demise. Co mment s by L at hers, Schraeder, and Bung o Persons with only nonepileptic seizures would, by deἀnition, not be at risk for SUDEP. However, the unfortunate reality is that from 10% to 30% of persons who appear to have long-established nonepileptic seizures also have epilepsy (Betts 1997). Thus, although the majority of persons with nonepileptic seizures do not have concurrent epilepsy, in those who also have a bona-ἀde seizure disorder, as demonstrated in this case, there is a risk of SUDEP associated with withdrawal of antiepileptic drugs based on the observation of only nonepileptic events. The physician must consider all aspects of the history and all prior EEG data before having conἀdence that the antiepileptic medication can be safely withdrawn. One should also keep in mind that rapid discontinuation of medication that was at a therapeutic serum level could induce a withdrawal seizure even in a person without epilepsy, further clouding the issue of diagnosis. From Betts, T. 1997. Psychiatric aspects of nonepileptic seizures. In Epilepsy: A Comprehensive Textbook, ed. J. Engel and T. A. Pedley. Philadelphia, PA: LippincottRaven Publishers.

8-year period was reviewed to look for a possible association. The review found the number of SUDEP cases was highest at the time of a full moon (70%), followed by a waxing moon (20%), and a new moon (10%). There were no SUDEP cases during the waning cycle. The authors concluded that a full moon appears to correlate with SUDEP. This same laboratory has examined other factors that may be involved in SUDEP, from regular imbibing of sardines and salmon to the influence of climate fluctuations (Calderazzo et al. 2009). Scorza et al. (2008b) have raised the question of whether seizure activity also influences dentate granule cell neurogenesis, since neurogenesis persists throughout life in adult mammalian dentate gyrus and is regulated by environmental, physiological, and

26 Sudden Death in Epilepsy: Forensic and Clinical Issues

molecular factors. The presence of hilar ectopic dentate granule cells was studied after status-�epilepticus was induced experimentally and it was determined that these cells migrate aberrantly, are abnormally integrated, and that the resulting hyperexcitability may contribute to seizure generation and/or propagation. Since high seizure frequency is a potential risk factor for SUDEP, the authors hypothesize that cardiac arrhythmias during and between seizures or transmission of epileptic activity to the heart via the autonomic nervous system may play a role. Thus, the aberrant neurogenesis may negatively influence the cardiovascular system of the patient with epilepsy and lead to cardiac abnormalities and then to the unwanted event of SUDEP. Studies are needed to examine the role in SUDEP played by these unusual factors to determine if they are risk factors.

1.5â•…Discussion 1.5.1â•… Future Steps The American Epilepsy Society and the Epilepsy Foundation Joint Task Force on SUDEP (So et al. 2008) assessed knowledge about SUDEP and recommended the following steps: 1. Hold a multidisciplinary workshop to reἀne current lines of investigation and identify additional areas of research for mechanism underlying SUDEP. 2. Conduct a survey of patients, families, and caregivers to identify effective means of education to enhance participation in SUDEP research. 3. Campaign for emphasis of the need for complete autopsy examinations for patients with suspected SUDEP. 4. Secure infrastructure grants to fund a consortium of centers to conduct prospective clinical and basic research studies to identify preventable risk factors and mechanisms underlying SUDEP. During the interim before these steps have been achieved, it is most important to provide prompt and optimal control of seizures, especially generalized convulsive seizures, to prevent SUDEP. A global focus is needed to resolve the risk factors for and mechanisms of epilepsy and sudden death (Lathers 2009).

References Annegers, J. F., S. P. Coan, W. A. Hauser, and J. Leestma. J. 2000. Epilepsy, vagal nerve stimulation by the NCP system, all-cause mortality, and sudden, unexpected, unexplained death. Epilepsia 41: 549–553. Antzelevitch, C. 2005a. Cardiac repolarization. The long and short of it. Eurpace 7 (Suppl 2): 3–9. Antzelevitch, C. 2005b. Modulation of transmural repolarization. Ann N Y Acad Sci 1047: 314–323. Asadi-Pooya, A. A., and M. R. Sperling. 2009. Clinical features of sudden unexpected death in epilepsy. J Clin Neurophysiol Aug 24. [Epub ahead of print]. Aurelien, D., T. P. Leren, E. Taubøll, and L. Gjerstad. 2009. New SCN5A mutation in a SUDEP victim with idiopathic epilepsy. Seizure 18: 158–160. Bateman, L. M., C. S. Li, and M. Seyal. 2008. Ictal hypoxemia in localization-related epilepsy: Analysis of incidence, severity and risk factors. Brain 131: 3239–3245.

Neurocardiologic Mechanistic Risk Factors in SUDEP

27

Beauregard, L. A., R. Fabiszewski, C. H. Black et al. 1991. Combined ambulatory electroencephalographic and electrocardiographic recordings for evaluation of syncope. Am J Cardiol 68: 1067–1072. Bell, G. S., and J. W. Sander. 2006. Sudden unexpected death in epilepsy. Risk factors, possible mechanisms and prevention: a reappraisal. Acta Neurol Taiwan 15: 72–83. Bell, G. S., J. L. Peacock, and J. W. Sander. 2009. Seasonality as a risk factor for sudden unexpected death in epilepsy: A study in a large cohort. Epilepsia Oct 27. [Epub ahead of print]. Betts, T. 1997. Psychiatric aspects of nonepileptic seizures. In Epilepsy: A Comprehensive Textbook, ed. J. Engel and T. A. Pedley, Chapter 199. Philadelphia, PA: Lippincott-Raven. Bilora, F., G. Vettore, C. Maifredini, S. Rocco, M. Pastorello, and F. Petrobelli. 1998. Chronobiology of acute pulmonary edema in an emergency service. Cardiologia 43: 303–307. Bingham, A. J., L. Ooi, L. Kozera, E. White, and I. C. Wood. 2007. The repressor element 1-silencing transcription factor regulates heart-speciἀc gene expression using multiple chromatinÂ�modifying complexes. Mol Cell Biol 27: 4082–4092. Bingham, A. J., L. Ooi, and I. C. Wood. 2006. Multiple chromatin modiἀcations important for gene expression changes in cardiac hypertrophy. Biochem Soc Trans 34 (Pt 6): 1138–1140. Bobo, E. G., and M. Bonvallet. 1975. Amygdala and masseteric reflex: I. Facilitation, inhibition and diphasic modiἀcations of the reflex, induced by localized amygdaloid stimulation. Electroencephalogr Clin Neurophysiol 39: 329–339. Brodie, M. J., and G. L. Holmes. 2008. Should all patients be told about sudden unexpected death in epilepsy (SUDEP)? Pros and Cons. Epilepsia 49 (Suppl 9): 99–101. Calderazzo, L., R. M. Arida, R. M. Cysneiros, E. A. Cavalheiro, and F. A. Scorza. 2009. From sardines to salmon: Influence of climate fluctuations on sudden unexpected death in epilepsy. Epilepsy Behav Jan. 20. Carnel, S. B., P. L. Schraeder, and C. M. Lathers. 1985. The effect of phenobarbital pretreatment on cardiac neural discharge and pentylenetetrazolâ•‚induced epileptogenic activity. Pharmacology 20: 225–240. Celiker, A., I. Erdogan, T. Karagoz, and S. Ozer. 2008. Clinical experiences of patients with catecholaminergic polymorphic ventricular tachycardia. Cardiol Young 23: 1–8. Chen, L. Y., D. G. Benditt, and S. Win-Kuang. 2008. Management of syncope in adults: An update. Mayo Clin Proc 83: 1280–1293. Colugnati D. B., R. M. Arida, S. L. Cravo, G. H. Schoorlemmer, A. C. de Almeida, E. A. Cavalheiro, and F. A. Scorza. 2008. Hibernating mammals in sudden cardiac death in epilepsy: What do they tell us? Med Hypotheses 70: 929–932. Coulter, D. L. 1984. Partial seizures with apnea and bradycardia. Arch Neurol 41: 173–174. Di Maio, T. G., and J. V. M. Di Maio, eds. 2006. Excited Delirium Syndrome. Cause of Death and Prevention. Boca Raton, FL: CRC Press. Dodd-O, J. M., and C. M. Lathers. 1990. A characterization of the lock step phenomenon in Phenobarbital treated cats. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, Chapter 13, 199–220. New York, NY: Marcel Dekker. Domenighetti, A. A., C. Boixel, D. Cefai, H. Abriel, and T. Pedrazzini. 2007. Chronic angiotensin II stimulation in the heart produces an acquired long QT syndrome associated with IK1 potassium current downregulation. J Mol Cell Cardiol 42: 63–70. Druschky, A., M. J. Hiltz, P. Hopp et al. 2001. Interictal cardiac autonomic dysfunction in temporal lobe epilepsy demonstrated by [123I]metaiodobenzylguanidine-SPECT. Brain 124: 372–382. Everengul, H., H. Tanriverdi, D. Dursunoglu et al. 2005. Time frequency domain analyses of heart rate variability in patients with epilepsy. Epilepsy Res 63: 131–139. Faingold, C. L., S. Tupal, Y. Mhaskar, and V. V. Uteshev. 2010. DBA mice as models of sudden unexpected death in epilepsy. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 41. Boca Raton, FL: CRC Press.

28 Sudden Death in Epilepsy: Forensic and Clinical Issues Ferguson, P. L., J. Chiprich, G. Smith et al. 2008. Prevalence of self-reported epilepsy, health care access, and health behaviors among adults in South Carolina. Epilepsy Behav 13: 529–534. Finsterer, J., and C. Stöllberger. 2010. Cardiac and pulmonary risk factors and pathomechanisms of sudden unexplained death in epilepsy patients (SUDEP). In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 42. Boca Raton, FL: CRC Press. Franco, F. A. 2007. Cerebral autoregulation and syncope. Prog Cardiovasc Dis 50: 49–80. Gaita, F., C. Giustetto, F. Bianchi et al. 2003. Short QT syndrome: A familial cause of sudden death. Circulation 108: 965–970. Gaita, F., C. Giustetto, F. Bianchi et al. 2004. Short QT syndrome: Pharmacological treatment. J Am Coll Crdiol 43: 1494–1499. Gary-Bobo, E. G., and M. Bonvalette. 1977. Commissural projection to the amygdala through the ἀmbria fornix system in the cat. Exp Brain Res 27: 61–70. Ghali and Nashef. 2010. Genetics of SUDEP. In Sudden Death in Epilepsy: Forensic and Clinical Issues,€ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 18. Boca Raton, FL: CRC Press. Gleick, J. 1987. Chaos: Making a New Science. New York, NY: Viking Press. Goldberger, A. L., V. Bhargava, and B. J. West. 1985. Nonlinear dynamics of the heart beat. Physica 17D: 207–214. Goodman, J. H., R. W. Homan, and I. L. Crawford. 1989. Kindled seizures activate both branches of the autonomic nervous system. Epilepsy Res 34: 169–176. Goodman, H. J., R. W. Homan, and I. L. Crawford. 1990. Kindled seizures elevate blood pressure and induce cardiac arrhythmias. Epilepsia 31: 489–495. Goodman, J. H., R. E. Berger, and T. K. Tcheng. 2005. Preemptive low-frequency stimulation decreases the incidence of amygdala-kindled seizures. Epilepsia 46: 1–7. Graves, T. D. 2006. Ion channels and epilepsy. Q J Med 99: 201–217. Gray, M. A., P. Taggart, P. M. Sutton et al. 2007. A cortical potential reflecting cardiac function. Proc Natl Acad Sci USA 104: 6818–6823. Grubb, B. P., G. Gerard, D. A. Wolfe et al. 1992. Syncope and seizures of psychogenic origin: identiἀcation with head-upright tilt table testing. Clin Cardiol 15: 839–842. Han, J., and G. K. Moe. 1964. Nonuniform recovery of excitability in ventricular muscle. Circ Res 14: 44–60. Hanna, N. J., and R. J. Panelli. 2010. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, J. E. Leestma, Chapter 57. Boca Raton, FL: CRC Press. Herreros, B. 2010. Cardiac channelopathies and sudden death. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 19. Boca Raton, FL: CRC Press. Hirsch, C. S., and D. L. Martin. 1971. Unexpected death in young epileptics. Neurology 21: 682–690. Hughes, J. R. 2009. A review of sudden unexpected death in epilepsy: Prediction of patients at risk. Epilepsy Behav 14: 280–287. Ieda, M., H. Kanazawa, Y. Ieda et al. 2006. Nerve growth factor is critical for cardiac sensory innervation and rescues neuropathy in diabetic hearts. Circulation 114: 2351–2363. Ieda, M., H. Kanazawa, K. Kimura et al. 2007. Sema3a maintains normal heart rhythm through sympathetic innervation patterning. Nat Med 13: 604–612. Ieda, M., K. Kimura, H. Kanazawa, and K. Fukuda. 2008. Regulation of cardiac nerves: a new paradigm in the management of sudden cardiac death? Curr Med Chem 15: 1731–1736. Jain, V. 2007. Clinical perspective of obstructive sleep apnea-induced cardiovascular complications. Antioxid Redox Signal 9: 701–710. Jehi, L., and I. M. Najm. 2008. Sudden unexpected death in epilepsy: Impact, mechanisms, and prevention. Clevel Clin J Med 75 (Suppl 2): S66–S70. Jim, K. F., C. M. Lathers, V. L. Farris, L. F. Pratt, and W. H. Spivey. 1989. Suppression of pentylenetetrazol elicited seizure activity by intraosseous lorazepam in pigs. Epilepsia 30: 480–486.

Neurocardiologic Mechanistic Risk Factors in SUDEP

29

Jim, K. F., C. M. Lathers, W. H. Spivey, W. D. Matthews, C. Kahn, and K. Dolce. 1988. Suppression of pentylenetetrazolâ•‚elicited seizure activity by intraosseous propranolol in pigs. J Clin Pharmacol 28: 1106–1111. Johnston, A., and P. Smith. 2007. Sudden unexpected death in epilepsy. Expert Rev Neurother 7: 1751–1761. Johnston, S. C., R. Siedenberg, J. K. Min, E. H. Jerome, and K. D. Laxer. 1997. Central apnea and acute cardiac ischemia in a sheep model of epileptic sudden death. Ann Neurol 42: 588–594. Johnston, S. C., J. K. Horn, J. Valente, and R. P. Simon. 1995. The role of hypoventilation in a sheep model of epileptic sudden death. Ann Neurol 37: 531–537. Karch, S. 2010. Drug abuse and SUDEP. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 10. Boca Raton, FL: CRC Press. Kerling, F., M. Dutsch, R. Linke, T. Kuwert, H. Stefan, and M. J. Hilz. 2009. Relation between ictal asystole and cardiac sympathetic dysfunctions shown by MIBG-SPECT. Acta Neurol Scand 120: 123–129. Kiehne, N., and S. Kauferstein. 2007. Mutations in the SCN5A gene: Evidence for a link between long TQ syndrome and sudden death? Forensic Sci Int Genet 1: 170–174. Kimura, K. M. Ieda, H. Kanazawa et al. 2007. Cardiac sympathetic rejuvenation: a link between nerve function and cardiac hypertrophy. Circ Res 100 (12): 1755–1764. Kinney, H., A. Chadwick, L. A. Crandall, M. Grafe, D. L. Armstrong, W. J. Kupsky, F. L. Trachtenberg, and H. F. Krous. 2009. Sudden death, febrile seizures, and hippocampal maldevelopment in toddlers: A new entity. Pediatr Dev Pathol Jul 16: 1. [Epub ahead of print]. Kornick, C. A., M. J. Kilborn, J. Santiago-Palmer et al. 2003. QTc interval prolongation associated with intravenous methadone. Pain 105 (3): 499–506. Lathers, C. M. 1980. Effect of timolol on postganglionic cardiac and preganglionic splanchnic sympathetic neural discharge associated with ouabainâ•‚induced arrhythmia. Eur J Pharmacol 64: 95–106. Lathers, C. M. 1981. Carnivores. Induced disease. Myocardial infarction. In Mammalian Models for Research on Aging 224–228. Washington, DC: National Academy Press. Lathers, C. M. 1982. Lack of effect of methylprednisolone on cardiac neural discharge associated with coronary occlusionâ•‚induced arrhythmia and death. Eur J Pharmacol 85: 233–238. Lathers, C. M. 2009. Epilepsy and sudden death: Personal reflections and call for global action. Epilepsy Behav 15 (3): 269–277. Lathers, C. M. 2010a. Sudden death animal models to study nervous system sites of action for disease and pharmacological intervention. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 25. Boca Raton, FL: CRC Press. Lathers, C. M. 2010b. Arrhythmogenic, respiratory, and psychological risk factors for sudden unexpected death and epilepsy. Case histories. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Chapter 44. Boca Raton, FL: CRC Press. Lathers, C. M., and R. M. Levin. 2010. Animal model for sudden cardiac death. Sympathetic innervation and myocardial beta receptor densities. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, J. E. Leestma. Chapter 33. Boca Raton, FL: CRC Press. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy. Characterization of autonomic cardiac neural discharge associated with pentylenetetrazolâ•‚induced epileptogenic activity. Epilepsia 23: 633–648. Lathers, C. M., and P. L. Schraeder. 1987. Review of autonomic dysfunction, cardiac arrhythmias, and epileptogenic activity. J Clin Pharmacol 27: 346–356. Lathers, C. M., and P. L. Schraeder. 1990a. Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, 485–505. New York, NY: Marcel Dekker.

30 Sudden Death in Epilepsy: Forensic and Clinical Issues Lathers, C. M., and P. L. Schraeder. 1990b. Synchronized cardiac neural discharge and epileptogenic activity, the Lock Step Phenomenon. Lack of correlation with cardiac arrhythmias. In Epilepsy and Sudden Death, ed. C. M Lathers and P. L. Schraeder, Chapter 12, 187–198. New York, NY: Marcel Dekker. Lathers, C. M., and P. L. Schraeder. 2002. Clinical pharmacology: Drugs as a beneἀt and/or risk in sudden unexpected death in epilepsy? J Clin Pharmacol 42: 123–136. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9: 236–242. Lathers, C. M., and P. L. Schraeder. 2009. Verbal autopsies and SUDEP. Epilepsy Behav 14: 573–576. Lathers, C. M., and P. L. Schraeder. 2010. Antiepileptic drugs beneἀt/risk clinical pharmacology: Possible role in cause and/or prevention of SUDEP. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, J. E. Leestma, Chapter 47. Boca Raton, FL: CRC Press. Lathers, C. M., R. F. Flax, and L. J. Lipka. 1986a. The effect of C1 spinal cord transection or bilateral adrenal vein ligation on thioridazineâ•‚induced arrhythmia and death in the cat. J Clin Pharmacol 26: 515–523. Lathers, C. M., R. M. Levin, and W. H. Spivey. 1986b. Regional distribution of myocardial receptors. Eur J Pharmacol 130: 111–117. Lathers, C. M., W. H. Spivey, L. E. Suter, J. P. Lerner, N. Tumer, and R. M. Levin. 1986c. The effect of acute and chronic administration of timolol on cardiac sympathetic neural discharge, arrhythmia, and beta adrenergic receptor density associated with coronary occlusion in the cat. Life Sci 39: 2121–2141. Lathers, C. M., J. L. Gerard-Ciminera, S. I. Baskin, J. C. Krusz, G. J. Kelliher, and J. Roberts. 1981. The action of reserpine, 6-hydroxydopamine, and bretylium on digitalis-induced cardiotoxicity. Eur J Pharmacol 76: 371–379. Lathers, C. M., K. F. Jim, and W. H. Spivey. 1989a. A comparison of intraosseous and intravenous routes of administration for antiseizure agents. Epilepsia 30: 472–479. Lathers, C. M., K. F. Jim, W. High, W. H. Spivey, W. D. Matthews, and T. Ho. 1989b. An investigation of the pathological and physiological effects of intraosseous sodium bicarbonate in pigs. J Clin Pharmacol 29: 354–359. Lathers, C. M., K. F. Jim, W. H. Spivey, C. Kahn, K. Dolce, and W. D. Matthews. 1990. Antiepileptic activity of beta-blocking agents. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, Chapter 24, 485–505. New York, NY: Marcel Dekker. Lathers, C. M., G. J. Kelliher, and J. Roberts. 1974a. Correlation of ouabainâ•‚induced arrhythmia and nonuniformity in cardiac accelerator nerves. Clin Res 22: 682A. Lathers, C. M., G. J. Kelliher, J. Roberts, and A. B. Beasley. 1978. Nonuniform cardiac sympathetic nerve discharge: Mechanism for coronary occlusion and digitalisâ•‚induced arrhythmia. Circulation 57: 1058–1065. Lathers, C. M., G. J. Kelliher, J. Roberts, and A. B. Beasley. 1977a. Role of the adrenergic nervous system in arrhythmia produced by acute coronary artery occlusion. In Pathophysiology and Therapeutics of Myocardial Ischemia, ed. A. Lefer, G. Kelliher, and M. Rovetto, 123–147. Myocardial Ischemia Symposium, Philadelphia Physiological Society, May 6, 1976. New York, NY: Spectrum Publ. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1977b. Correlation of ouabainâ•‚induced arrhythmia and nonuniformity in the histamineâ•‚evoked discharge of cardiac sympathetic nerves. J Pharmacol Exp Ther 203: 467–479. Lathers, C. M., S. A. Koehler, C. H. Wecht, and P. L. Schraeder. 2003a. Forensic antiepileptic drug levels in 2001 autopsy cases of sudden, unexpected deaths in persons with epilepsy in Allegheny County Pennsylvania. FDA Science Forum, Washington, DC. Lathers, C. M., S. A. Koehler, C. H. Wecht, and P. L. Schraeder. 2003b. Forensic antiepileptic drug levels in 2001 autopsy cases of sudden, unexpected deaths in persons with epilepsy in Allegheny County, Pennsylvania. Annual Meeting of the American College Clinical Pharmacology, Orlando, FL.

Neurocardiologic Mechanistic Risk Factors in SUDEP

31

Lathers, C. M., P. L. Schraeder, and H. G. Claycamp. 2003c. Clinical pharmacology of topirimate vs. lamotrigine vs. Phenobarbital: Comparison of efficacy and side effects using odds ratios. J Clin Pharmacol 43: 491–503. Lathers, C. M., and L. J. Lipka. 1986. Chlorpromazine: Cardiac arrhythmogenicity in the cat. Life Sci 38: 521–538. Lathers, C. M., and L. J. Lipka. 1987. Cardiac arrhythmia, sudden death, and psychoactive agents. J€Clin Pharmacol 27: 1–14. Lathers, C. M., L. J. Lipka, and H. A. Klions. 1988. Digitalis glycosides: A discussion of the similarities and differences in actions and existing controversies. Rev Clin Basic Pharmacol 7: 1–108. Lathers, C. M., and J. Roberts. 1980. Digitalis cardiotoxicity revisited. Life Sci 27: 1713–1733. Lathers, C. M., and J. Roberts. 1985. Are the sympathetic neural effects of digoxin and quinidine involved in their action on cardiac rhythm? J Cardiovasc Pharmacol 7: 350–360. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1974b. Relationship between the effect of ouabain on€ arrhythmia and interspike intervals (ISI) of cardiac accelerator nerves. Pharmacologist 16:€201. Lathers, C. M., P. L. Schraeder, and M. Bungo. 2008a. Mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12: 3–24. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008b. Reply. Epilepsy Behav 13: 265–269. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008c. Sudden death: Neurocardiologic mystery, Chapter 13. In: Psychological Factors and Cardiovascular Disorders, ed. L. Sher, Chapter 13. New York, NY: Nova Science Publishers, Inc. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2010a. Sodium channel dysfunction. Common physiopathologic mechanism associated with sudden death ECG abnormalities in Brugada syndrome and some types of epilepsy. Case histories. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, J. E. Leestma, Chapter 20, Boca Raton, FL: CRC Press. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2010b. Syncope, seizures and SUDEP. Case histories. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 22. Boca Raton, FL: CRC Press. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2010c. Unanswered questions: SUDEP Clinical studies needed. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P.€L.€Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 4. Boca Raton, FL: CRC Press. Lathers, C. M., P. L. Schraeder, and S. B. Carnel. 1984. Neural mechanisms in cardiac arrhythmias associated with epileptogenic activity: The effect of phenobarbital in the cat. Life Sci 34: 1919–1936. Lathers, C. M., P. L. Schraeder, and N. Tumer. 1993. The effect of phenobarbital on autonomic function and epileptogenic activity induced by the hippocampal injection of penicillin in cats. J Clin Pharmacol 33: 837–844. Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lock step phenomenon. Electroencephalogr Clin Neurophysiol 67: 247–259. Lathers, C. M., W. H. Spivey, and R. M. Levin. 1988. The effect of chronic timolol in an animal model for myocardial infarction. J Clin Pharmacol 28: 736–745. Lathers, C. M., W. H. Spivey, R. M. Levin, and N. Tumer. 1990. The effect of dilevalol on cardiac autonomic neural discharge, plasma catecholamines, and myocardial beta receptor density associated with coronary occlusion. J Clin Pharmacol 30: 241–253. Lear-Kaul, K. C., L. Coughlin, and M. J. Dobersen. 2005. Sudden unexpected death in epilepsy: a retrospective study. Am J Forensic Med Pathol 26: 11–17. Leestma, J. E., T. Walczak, J. R. Hughes, M. B. Kalelkar, and S. S. Teas. 1989. A prospective study on sudden unexpected death in epilepsy. Ann Neurol 26: 195–203. Lehmann-Horn F., and K. Jurkat-Rott. 1999. Voltage-gated ion channels and hereditary disease. Physiol Rev 79: 1317–1372.

32 Sudden Death in Epilepsy: Forensic and Clinical Issues Leiter, J. C., and I. Böhm. 2007. Mechanisms of pathogenesis in the sudden infant death syndrome. Respir Physiol Neurobiol 159(2): 127–138. Lipka, L. J., and C. M. Lathers. 1987. Psychoactive agents, seizure production, and sudden death in epilepsy. J Clin Pharmacol 27: 169–183. Lipka, L. J., C. M. Lathers, and J. Roberts. 1988. Does chlorpromazine produce cardiac arrhythmia via the central nervous system? J Clin Pharmacol 28: 968–983. Malow, B. A., K. Levy, K. Maturen, and R. Bowes. 2000. Obstructive sleep apnea is common in medically refractory epilepsy patients. Neurology 55: 1002–1007. Martin, P., A. Schreiner, K. Rettig, and B. Schauble. 2009. Topiramate in patients with epilepsy and intellectual disability. Epilepsy Behav 14: 496–502. Matschke, J., and R. Laas. 2007. Central alveolar hypoventilation syndrome causes SUD. Am J Forensic Med Pathol 28: 141–144. Michelsen, J. W., and J. M. Meyer. 2007. Cardiovascular effects of antipsychotics. Expert Rev Neurother 7: 829–839. McGregor, A., and J. Wheless. 2006. Pediatric experience with sudden unexplained death in epilepsy at a tertiary epilepsy center. J Child Neurol 21: 782–787. Monte, C. P., J. B. Arends, I. Y. Tan, A. P. Aldenkamp, M. Limburg, and M. C. de Krom. 2007. Sudden unexpected death in epilepsy patients: Risk factors. A systematic review. Seizure 16: 1–7. Nagele, H., S. Hashagen, M. Azizi, S. Behrens, and M. A. Castel. 2007. Analysis of terminal arrhythmias stored in the memory of pacemakers from patients dying suddenly. Europace 9: 380–384. Nashef, L., S. Garner, J. W. Sander, D. R. Fish, and S. D. Shorvon. 1998. Circumstances of death in sudden death in epilepsy: interviews of bereaved relatives. J Neurol Neurosurg Psychiatry 64: 349–352. Nashef, L., N. Hindocha, and A. Makoff. 2007. Risk factors in sudden death in epilepsy (SUDEP): The quest for mechanisms. Epilepsia 48: 859–871. Nilsson, L., U. Bergman, V. Diwan, B. Y. Farahmand, P. G. Persson, and T. Tomson. 2001. Antiepileptic drug therapy and its management in sudden unexpected death in epilepsy: A case-control study. Epilepsia 42: 667–673. Novak, V., A. L. Reeves, P. Novak, P. A. Low, and F. W. Sharbrough. 1999. Time-frequency mapping of R–R interval during complex partial seizures of temporal lobe origin. J Auton Nerv Syst 24: 195–202. O’Rourke, K., and C. Lathers. 1990. Interspike interval histogram characterization of synchronized cardiac sympathetic neural discharge and epileptogenic activity in the electrocorticogram of the cat. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, Chapter 13, 239– 260. New York, NY: Marcel Dekker. Owada, M., Y. Aizawa, K. Kurihara, N. Tanabe, T. Aizaki, and T. Izumi. 1999. Risk factors and triggers of sudden death in the working generation: An autopsy proven case-control study. Tohoku J Exp Med 189: 245–258. Paterson, D. S. 2010. Medullary serotonergic abnormalities in sudden infant death syndrome. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 5. Boca Raton, FL: CRC Press. Penἀeld, W., and H. Jasper. 1954. Summary of clinical analysis and seizure patterns. In Epilepsy and the Functional Anatomy of Human Brain, 830–831. Boston, MA: Little, Brown. Persson, H., E. Kumlien, M. Ericson, and T. Topmson. 2007. Circadian variation in heart-rate variability in localization-related epilepsy. Epilepsia 48: 917–922. Pezzella, M., P. Striano, C. Ciampa, L Errichiello, P. Penza, and S. Striano. 2009. Severe pulmonary congestion in a near-miss at the ἀrst seizure: further evidence for respiratory dysfunction in sudden unexpected death in epilepsy. Epilepsy Behav (in press). Pickworth, W. B., J. Gerard-Ciminera, and C. M. Lathers. 1990. Stress, arrhythmias and seizures. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, Chapter 22, 417–446. New York, NY: Marcel Dekker.

Neurocardiologic Mechanistic Risk Factors in SUDEP

33

Quint, S. R., J. A. Messenheimer, and M. B. Tennison. 1990. Power spectral analysis: a procedure for assessing autonomic activity related to risk factors for sudden and unexplained death in epilepsy. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, Chapter 16, 261–291. New York, NY: Marcel Dekker. Randall, W. C. 1977. Neural Regulation of the Heart, 1–440. New York, NY: Oxford University Press. Randall, W. C. 1984. Nervous Control of Cardiovascular Function, 1–476. New York, NY: Oxford University Press. Randall, W. C., J. X. Thomas, D. E. Euler, and G. J. Rozanski. 1978. Cardiac dysrhythmias associated with autonomic nervous system imbalance in the conscious dog. In Perspectives in Cardiovascular Research, Vol. 2, Neural Mechanisms in Cardiac Arrhythmias, ed. P. J. Schwartz, A. M. Brown, A. Malliani, and A. Zanchetti, 123–138. New York, NY: Raven Press. Reingardiene, D., and J. Vilcinskaite. 2007. QTc-prolonging drugs and the risk of sudden death. Medicina (Kaunas) 43: 347–353. Rusli, M., W. H. Spivey, H. Bonner, R. M. McNamara, C. K. Aaron, and C. M. Lathers. 1987. Pathological effects of endotracheal diazepam in cats. Ann Emerg Med 16: 314–318. Ryvlin, P., T. Tomson, and A. Montavont. 2009. Excess mortality and sudden unexpected death in epilepsy. Press Med 38 (6): 905–910. Saffitz, J. E. 2008. Sympathetic neural activity and the pathogenesis of sudden cardiac death. Heart Rhythm 5: 140–141. Salama, G., and B. London. 2007. Mouse models of long QT syndrome. J Physiol 578 (Pt 1): 43–53. Sarkozy, A., and P. Brugada. 2005. Sudden cardiac death: What is inside our genes? Can J Cardiol 21: 1099–1110. Schott, G. D., A. A. McLeod, and D. E. Jewitt. 1977. Cardiac arrhythmias that masquerade as epilepsy. Br Med J 1: 1454–1457. Schraeder, P. L. 1987. Adult respiratory distress syndrome (ARDS) associated with nonconvulsive status epilepticus. Epilepsia 28: 605. Schraeder, P. L. 2010. SUDEP case histories: Typical and atypical. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 52. Boca Raton, FL: CRC Press. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32: 1371–1382. Schraeder, P. L., and C. M. Lathers. 1989. Paroxysmal autonomic dysfunction, epileptogenic activity and sudden death. Epilepsy Res 3: 55–62. Schraeder, P. L., C. M. Lathers, and J. B. Charles. 1994. The spectrum of syncope. J Clin Pharmacol 34: 454–459. Schraeder, P. L., R. Pontzer, and T. R. Engel. 1983. A case of being scared to death. Arch Intern Med 143: 1793–1794. Schuele, S. U. 2009. Effects of seizures on cardiac function. J Clin Neurophysiol. Sep 11. [Epub ahead of print]. Schwartz, R. D., X. Yu, M. R. Katzman, D. M. Hayden-Hixson, and J. M. Perry. 1995. Diazepam, given postischemia, protects selectively vulnerable neurons in the rat hippocampus and striatum. J Neurosci 15 (1 Pt 2): 529–539. Scorza, F. A., A. M. Abreu, M. Albuquerque et al. 2007. Quantiἀcation of respiratory parameters in patients with temporal lobe epilepsy. Arq Neuropsiquiatr 65: 450–453. Scorza, F. A., M. Albuquerque, R. M. Arida, V. C. Terra, H. R. Machado, and E. A. Cavalheiro. 2010a. Beneἀts of sunlight: Vitamin D deἀciency might increase the risk of sudden unexpected death in epilepsy. Med Hypotheses 74: 158–161. Scorza, F. A., E. A. Cavalheiro, R. M. Arida et al. 2010b. Omega-3 fatty acids in SUDEP: Guardian of the brain-heart connection. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Chapter 3. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 22. Boca Raton, FL: CRC Press.

34 Sudden Death in Epilepsy: Forensic and Clinical Issues Scorza, F. A., R. M. Arida, and E. A. Cavalheiro. 2008a. Preventive measures for sudden cardiac death in epilepsy beyond therapies. Epilepsy Behav 13: 263–264. Scorza, F. A., R. M. Cysneiros, and R. M. Arida et al. 2008b. Adult hippocampal neurogenesis and sudden unexpected death in epilepsy: Reality or just an attractive history? Med Hypotheses 71: 914–922. Sharma, S., T. Ho, and B. K. Kantharia. 2010. Not seizure but syncope: Causes and investigations to establish the diagnosis. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, J. E. Leestma, Chapter 21. Boca Raton, FL: CRC Press. Simon, R. P., L. L. Bayne, R. F. Tranbaugh, and F. R. Lewis. 1982. Elevated pulmonary flow and protein content during status epilepticus in sheep. J Appl Physiol 52: 91–95. Sheldon, R. S., A. G. Sheldon, A. Serletis et al. 2007. Investigators of the prevention of syncope study. Worsening of symptoms before presentation with vasovagal syncope. J Cardiovasc Electrophysiol 18: 954–959. So, E. L. 2008. What is known about the mechanisms underlying SUDEP? Epilepsia 49 (Suppl 9): 93–98. So, E. L., J. Brainbridge, and J. R. Buchhalter et al. 2008. Report of the American Epilepsy Society and the Epilepsy Foundation Joint Task Force on sudden unexplained death in epilepsy. Epilepsia 50: 917–922. So, E. L., M. C. Sam, and T. L. Lagerlund. 2000. Postictal central apnea as a cause of SUDEP: Evidence from near-SUDEP incident. Epilepsia 41: 1494–1497. Sonoda E. Y., D. B. Colugnati, C. A. Scorza, R. M. Arida, A. P. Pansani, A. C. de Almeida, E. A. Cavalheiro, and F. A. Scorza. 2008. Is cold the new hot in sudden unexpected death in epilepsy? Effect of low temperature on heart rate of rats with epilepsy. Arq Neuropsiquiatr 66: 848–852. Spivey, W. H., and C. M. Lathers. 1985. Effect of timolol on the sympathetic nervous system in coronary occlusion in cats. Ann Emerg Med 134: 939–944. Spivey, W. H., C. M. Lathers et al. 1985. A comparison of intraosseous, central and peripheral routes of sodium bicarbonate administration during CPR in pigs. Ann Emerg Med 14: 1135–1139. Spivey, W. H., H. D. Unger, R. M. McNamara, M. M. LaManna, T. Ho, and C. M. Lathers. 1987a. The effect of intraosseous sodium bicarbonate on bone in swine. Ann Emerg Med 16: 773–776. Spivey, W. H., H. D. Unger, C. M. Lathers, and R. M. McNamara. 1987b. Intraosseous diazepam suppression of pentylentetetrazol-induced epileptogenic activity in pigs. Ann Emerg Med 16: 56–59. Stauffer, A. Z., J. Doddâ•‚O, and C. M. Lathers. 1989. The relationship of the lock step phenomenon and precipitous changes in mean arterial blood pressure. Electroencephalogr Clin Neurophysiol 72: 340–345. Stauffer, A. Z., J. M. Dodd-O, and C. M. Lathers. 1990. Relationship of the lockstep phenomenon and precipitous changes in blood pressure. In Epilepsy and Sudden Death, ed. C. M. Lathers and P.€L. Schraeder, Chapter 14, 199–220. New York, NY: Marcel Dekker. Strzelczyk, A., S. Bauer, S. Knake, W. H. Oertel, H. M. Hamer, and F. Rosenow. 2008. Ictal asystole in temporal lobe epilepsy before and after pacemaker implantation. Epileptic Disord 10: 39–44. Stopper, M., T. Joska, and M. M. Burg et al. 2006. Electrophysiologic characteristics of anger-Â�triggered arrhythmias. Heart Rhythm 4: 268–273. Tavee, J., and H. Morris 3rd. 2008. Severe postictal laryngospasm as a potential mechanism for sudden unexpected death in epilepsy: a near-miss in an EMU. Epilepsia 49: 2113–2117. Terra-Bustamante, V. C., C. A. Scorza, and M. de Albuquerque et al. 2009. Does the lunar phase have an effect on sudden unexpected death in epilepsy? Epilepsy Behav 14: 404–406. Terrence, C. E., G. R. Rao, and J. A. Perper. 1981. Neurogenic pulmonary edema in unexpected, unexplained death in epileptic patients. Ann Neurol 25: 594–595. Tomson, T. 2010. Compliance with antiepileptic drug treatment and the risk of SUDEP. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapters 51 and 22. Boca Raton, FL: CRC Press.

Neurocardiologic Mechanistic Risk Factors in SUDEP

35

Tomson, T., L. Nashef, and P. Ryvlin. 2008. Sudden unexpected death in epilepsy: current knowledge and future directions. Lancet Neurol 7: 1021–1031. Tupal, S., and C. L. Faingold. 2006. Evidence supporting a role of serotonin in modulation of sudden death induced by seizures in DBA/2 mice. Epilepsia 47: 21–26. Tupal, S., and C. L. Faingold. 2006. Respiratory arrest mechanisms, modulated in part by serotonin, may cause SUDEP. Epilepsia 47: 21–26. Tumer, N., P. L. Schraeder, and C. M. Lathers. 1985. The effect of phenobarbital upon autonomic function and epileptogenic activity induced by the hippocampal injection of penicillin in cats. Epilepsia 26: 520. Verrier, R. L., K. Kumar, and B. D. Nearing. 2009. Basis for sudden cardiac death prediction by T-wave alternans from an integrative physiology perspective. Heart Rhythm 6 (3): 416–422. Verrier, R. L., and S. C. Schachter. 2010. Neurocardiac interactions in sudden unexpected death in epilepsy: Can ambulatory electrocardiogram-based assessment of autonomic function and T-wave alternans help to evaluate risk? In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo and J. E. Leestma, Chapters 43 and 22. Boca Raton, FL: CRC Press. Vindrola, O., M. Asai, M. Zubieta, E. Talavera, E. Rodriguez, and G. Linares. 1984. Pentylenetetrazol kindling produces a long-lasting elevation of IR-Met-enkephalin but not IR-Leu-enkephalin in rat brain. Brain Res 297: 121–125. Vohra, J. 2007. The Long QT Syndrome. Heart Lung Circ 16 (Suppl 3): S5–S12. Walczak, T. 2003. Do antiepileptic drugs play a role in sudden unexpected death in epilepsy? Drug Saf 26: 673–683. Walczak, T. 2010. Risk factors for SUDEP. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 12. Boca Raton, FL: CRC Press. Wang, J., Y. Chen, K. Li, and L. Hou. 2006. Blockade of inhibitory neurotransmission evoked seizurelike ἀring of cardiac parasympathetic neurons in brainstem slices of newborn rats: Implications for sudden deaths in patients of epilepsy. Epilepsy Res 70: 172–183. Wannamaker, B. 1985. Autonomic nervous system and epilepsy. Epilepsia 26 (Suppl 1): S31–S39. Wichter, T., T. M. Paul, and L. Eckardt et al. 2005. Arrhythmogenic right ventricular cardiomyopathy. Antiarrhythmic drugs, catheter ablation, or ICD? Herz 30: 91–101. Williams, L., and M. Frenneaux. 2007. Syncope in hypertropic cardiomyopathy mechanism and consequences for treatment. Europace 9 (9): 817–822. Winfree, A. T. 1987. When Time Breaks Down: The Three-Dimensional Dynamics of Electrochemical Waves and Cardiac Arrhythmias. Princeton, NJ: Princeton University Press.

Forensic Considerations and Sudden Unexpected Death in Epilepsy

2

Jan E. Leestma

Contents 2.1 Background of Sudden or Unexpected Deaths 2.2 Forensic Issues Regarding Epilepsy and SUDEP Deaths 2.2.1 Classifying Deaths as SUDEP 2.3 Drowning 2.4 SUDEP Deaths in the Home or Workplace 2.5 SUDEP Deaths Outdoors 2.6 Role of SUDEP in Traffic Deaths 2.7 SUDEP and Deaths in Agitated Delirium or Restraint 2.8 SUDEP in Criminal Cases 2.9 SUDEP and Anticonvulsant Medication References

37 39 39 41 41 42 42 44 48 50 52

2.1â•… Background of Sudden or Unexpected Deaths Sudden and unexpected deaths generally occur outside hospitals, although victims may be brought to an emergency room. In most locales in the United States, and likely in other developed countries, such deaths are usually not attended by a physician and are without detailed historical or medical information. They will usually be brought to the attention of a medical examiner or coroner, who is responsible for determining the cause and manner of death and for generating a death certiἀcate before the remains may be interred or otherwise disposed of. These responsibilities are statutory in most jurisdictions, but how they are discharged varies widely (DiMaio and DiMaio 2001). When a coroner is the responsible party, this individual may be elected or appointed, and may or may not be a medical doctor. A local pathologist (forensically trained or not) will usually be employed to conduct an autopsy or other examination of the body to make the determination of cause and manner of death. When a medical examiner is involved, that individual is almost always a trained forensic pathologist or will employ a staff of forensic pathologists to perform the required examinations to determine the medical cause of the death and its manner, which by convention may be labeled as a homicide (death at the hands of another), suicide (death by one’s own hand), accident, natural (natural disease processes), or “undetermined.” The coroner or medical examiner may direct that an autopsy be performed and that other studies that may include toxicological examinations of tissues or body fluids be undertaken (DiMaio and DiMaio 2001; Knight 1996). 37

38 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 2.1â•… Manner of Death Statistics Manner of Death

Percentage of Death Certiἀcates (%)

Natural diseases Homicide Suicide Accidents Undetermined

64–66 10–11 4–5 20–22 4–5

Sources: Office of the Medical Examiner, Cook County, IL, Annual Report, 1977– 1979, Office of the Medical Examiner, Chicago, IL, 1979; Leestma, J. E., and E. W. Sharp, in Forensic Neuropathology, CRC Press, Boca Raton, FL, 2009. With permission.

So-called sudden deaths may or may not be truly sudden, with many deἀnitions that surround such deaths (Davis and Wright 1980; de la Grandmaison and Durigon 2002; Kuller 1966; Kuller et al. 1966; Luke and Helpern 1968; Moritz 1954). When one refers to sudden deaths, more often than not, one is really referring to the fact that the victim actually died unexpectedly and, perhaps, also suddenly. When such deaths are witnessed, death may be described as having occurred in minutes rather than hours or longer. When a death is not witnessed, but the victim is found in circumstances that indicate death occurred in the course of normal activities and does not suggest a protracted agonal period, the unexpected and sudden quality of the death can be presumed (Haerem 1978). Such circumstances often involve ἀnding the victim dead in bed, in the bathroom (not bathing), or in a den or living room (in a chair or on the floor). When the victim is found in the bath (submerged or not), swimming pool, sauna or Jacuzzi, near electrical equipment or machinery, or in a vehicle (not involved in a crash), interpretations regarding the exitus may become complicated. When death occurs in the context of an apparent accident, with or without trauma, many facts and factors will have to be considered in reaching a proper determination of cause and manner of death (Leestma 1990a) (see Table 2.1). Owing to the variability in coroners’ and medical examiners’ statistics, it is generally recognized that 10–15% of their cases occurred suddenly and/or unexpectedly (see Table 2.2). Not unexpectedly, the majority of these deaths are due to some form of heart attack. Often this cannot be proven anatomically, although it can be inferred (Haerem 1978; Fineschi et al. 2006; Greenberg and Dwyer 1982; Schwartz and Gerrity 1975). Most central nervous system diseases do not kill suddenly, even though they present suddenly

Table 2.2â•… Medical Causes of Death in Sudden and/or Unexpected Deaths Disease Process Heart and great vessels Respiratory system Brain or meninges Digestive or urogenital systems Miscellaneous

Percentage (%) 56.1 (±7.4%) 14.5 (±6.4%) 15.8 (±2.4%) 8 (±1.7%) 9.5 (±8.9%)

Sources: Kuller, L., J Chronic Dis, 19 (11), 1165–1192, 1966; Kuller, L., A. Lilienfeld, and R. Fisher, Circulation, 34 (6), 1056–1068, 1966; Leestma, J. E., in Forensic NeuroÂ� pathology, CRC Press, Boca Raton, FL, 2009. With permission.

Forensic Considerations and Sudden Unexpected Death in Epilepsy

39

and unexpectedly as in strokes (Leestma 2009a). The largest group of CNS-related sudden deaths appear to occur in the context of epilepsy [sudden death in epilepsy (SUDEP)] (Leestma 2009a, 1990b; Lhatoo and Sander 2002). A number of chapters in this volume deal with population statistics for SUDEP and these data will not be repeated here. Suffice it to say that deaths classiἀed as SUDEP—not a cause of death, but a category of death for which no anatomically evident cause is found, much like the label sudden infant death syndrome (SIDS)—forms an important and nonrare problem for forensic pathologists. As noted before in previous publications (Leestma 1990a, 1990b; Leestma et al. 1989, 1997), the apparent incidence of SUDEP cases at a large metropolitan medical examiner’s facility (Chicago-Cook County, IL) is 60–80 cases per year (Leestma et al. 1989). A casual survey of other large metropolitan offices by the author indicates similar numbers of cases in all of them (Los Angeles, Denver, and Miami). Collecting precise and accurate statistics on SUDEP incidence is problematic because there is wide variation in practices among the forensic community on how such cases are labeled on death certiἀcates as Schraeder et al. (2006) have reported. This is a function of the awareness or lack of awareness of the phenomenon and/or philosophical positions on how, presumably, epilepsy-related deaths should be classiἀed or labeled by forensic pathologists. As alluded to earlier, it is the task and responsibility of the forensic pathologist to determine the medical cause and the manner of death if possible. If this is not possible, then, after all the evidence is gathered, the medical cause of death and, sometimes, the manner of death, must remain undeclared (undetermined). This circumstance is certainly intellectually disappointing and frustrating and may lead to unwarranted speculations on the part of family, friends, attorneys, the media, and colleagues. Such cases, whether assigned the SUDEP designation or not, may ἀnd their way into the legal system as civil cases (torts) or even as criminal cases. The following discussion will explore some of these possibilities and, when possible, provide case examples of the issues involved.

2.2â•…Forensic Issues Regarding Epilepsy and SUDEP Deaths 2.2.1â•…Classifying Deaths as SUDEP Which cases are labeled with the designation SUDEP can be difficult and sometimes controversial. An attempt to provide some guidelines was arrived at in conjunction with data collected during the development of lamotrigine (Lamictal) at the behest of the then Burroughs-Wellcome pharmaceutical company, now Glaxo-SmithKline, Inc. In the clinical trial databases that drew upon studies in many countries with 4700 patients (5747 patient years of exposure) over several years, 45 sudden and/or unexpected deaths were discovered in all arms of the study. It was important to evaluate, if possible, if there was a different incidence of SUDEP deaths between the treated and control groups. To this end, a study group of professionals that included epidemiologists, epileptologists, and pathologists was impaneled to review the data and to determine what criteria might be applied to the cases in order to stratify them for analysis (Leestma et al. 1997; Tennis et al. 1995). As might be expected, the quality and quantity of individual case data varied widely. It was not possible to obtain further data beyond that reported. Cases were stratiἀed according to the following classiἀcation (Leestma et al. 1997):

40 Sudden Death in Epilepsy: Forensic and Clinical Issues

• SUDEP (definite or highly probable) The victim suffered from epilepsy as deἀned by Gastaut (1969) and Gastaut and Zifkin (1985) and had been treated with one or more anticonvulsive agents, usually for many years, and died unexpectedly in a reasonable state of health. The fatal attack occurred suddenly, but death might not have occurred for several hours, when it was usually associated with cardiorespiratory arrest, resuscitative efforts, and their complications. The attack occurred during normal activities in benign circumstances. An obvious medical cause of death was not found after autopsy. Cardiac arrhythmia may have been observed after the attack. If the victim was found in the bath but did not show evidence of drowning, the death may be assignable as SUDEP. If status epilepticus or acute neurotrauma occurred during seizure, the case was excluded. • Possible SUDEP These cases met most or all of the SUDEP criteria, but data suggested more than one possible cause of death associated with seizures, such as death while bathing, swimming, or due to aspiration, with or without an observed seizure. • Other non-SUDEP These cases could not be assigned to either of the ἀrst two categories and an obvious or likely cause of death had been established. • Insufficient data These cases could not be interpreted because of lack of information or ambiguous data concerning the circumstances of death or concurrent medical conditions in the victim. There are and were inevitable issues with the above classiἀcations, owing to all the informational problems that are inherent to any patient-based study. A particularly thorny one was the role of possible drowning in some of the victims, which is discussed later. Other confounding variables include the role of concurrent diseases in the victim, especially heart disease in the older victim, as well as conditions related to alcohol and drug abuse. There are a host of issues in infants and children that include possible SIDS-type deaths. Highly important to any death investigation of an individual who apparently dies suddenly and/or unexpectedly is the death scene investigation (DiMaio and DiMaio 2001; Knight 1996; Leestma 2009b). The physical death scene environment must be documented and carefully inspected. It is important to note if the body is in a position or location that would be consistent with normal activity that was interrupted by the fatal attack. The presence of bruises on the face, head, and extremities does not necessarily mean that the victim was assaulted; rather that if a seizure occurred, the victim may have fallen and been injured or become injured during the clonic phase of a seizure. Context and common sense in evaluating the scene is vital. Furthermore, the investigator should search for and collect any pill bottles that are present and any materials that suggest illicit drug use. The pill bottles may indicate the use of anticonvulsant medications and can provide the names of the pharmacies that dispensed the drug and the physician who prescribed it, who may be contacted later for additional information. If the pill bottles have pills in them and the date of prescription is on the label, or if the bottle is empty, one can make a judgment regarding medication compliance by the victim. If other medications are present, this can also give a

Forensic Considerations and Sudden Unexpected Death in Epilepsy

41

clue as to other medical problems the victim may have had and can provide some guidance as to which toxicological examinations may need to be made. If any witnesses to the attack are present, a detailed description of what was seen is vital. If the attack was not witnessed, it may be possible to determine the timeframe of the attack by inference (when the individual was last seen and found). If friends or relatives are available, they may have knowledge about the health status of the decedent and, speciἀcally, if the victim was epileptic. To aid investigators in collecting this vital information, the author has previously described a “check sheet” that can serve as a template for investigators at death scenes where the victim was epileptic and sudden or unexpected death has occurred (Leestma et al. 1989; Leestma 1990b, 2009b).

2.3â•…Drowning Epileptic individuals are frequently found dead in bathtubs, swimming pools, hot tubs, and natural bodies of water in conjunction with swimming. It is a common instruction to epileptic persons not to bathe upon arising or immediately before going to bed, rather to bathe at another time of the day to avoid the increased likelihood of seizures that occur proximate to sleep (Laidlaw and Richens 1982; Engel and Pedley 2008). The postmortem diagnosis of drowning is imprecise and subject to many evidentiary problems that are well-known to any forensic pathologist such as the so-called “wet” and “dry” forms of drowning (DiMaio and DiMaio 2001; Knight 1996); thus, it cannot be known with complete assurance that a victim who was found submerged in the bathtub or swimming pool, and whose lungs did not show evidence of water aspiration, did not drown (laryngospasm), but rather died from the supposed mechanism(s) typical of SUDEP and simply sank into the water after the attack, when the victim was lifeless. If the potential drowning occurred in a natural body of water, such as a stream, lake, or ocean, it may be possible to recover diatoms and other small particles from the water in the lower respiratory passages, which may indicate aspiration. The forensics of this process and its reliability is controversial (Piette and De Letter 2006; Modell et al. 1999; Pachar and Cameron 1993). If the victim is found in the bathtub but the head is not immersed, this does not necessarily mean that drowning did not occur. The onset of rigor mortis may have shifted the position of the body after death, causing the head to rise out of the bath water. A careful examination of the oral cavity at autopsy may reveals bites of the tongue, lips, or buccal mucosa, which is presumptive evidence that a recent seizure occurred but still does not necessarily mean that drowning did not occur (Ulrich and Maxeiner 2003).

2.4â•…SUDEP Deaths in the Home or Workplace The most common location for SUDEP deaths is in the home, in locations that bespeak an attack that took place during normal activity (Leestma et al. 1989; Leestma 1990b). The most common of these is in the bedroom. The victim may be found in bed in a normal position, or one that suggests some movement, as in a seizure with disordered bedding apparent. The victim may also be found on the floor in the bedroom dressed in normal

42 Sudden Death in Epilepsy: Forensic and Clinical Issues

clothing or in bed clothes beside the bed or in another position that suggests that the fatal attack took place proximate to going to bed. The association of SUDEP, and for that matter epileptic seizures in general (Laidlaw and Richens 1982; Engel and Pedley 2008), with sleep is well known and it is also not uncommon to ἀnd SUDEP victims dead in an easy chair in a den or living room, sometimes with the television on, which suggests that the victim may have been sleeping or dozing when the attack occurred. Death in the bathroom, discussed earlier with respect to drowning, may have no connection with bathing when the victim is found dead on the bathroom floor. In such circumstances, various injuries from the impact of the body surfaces with bathroom ἀxtures is common and may confuse the interpretation of the death scene, suggesting a homicidal attack or perhaps a suicide. A careful documentation of the death scene and blood spatter or blood flow (DiMaio and DiMaio 2001; James et al. 2005) as well as the autopsy may clear up suspicion of foul play. The body of a SUDEP victim may be found in other areas of the home, in the garage or workshop area, or in a workplace area that raises the issue of other means of death than a seizure-related cause. A careful scene inspection and investigation will go a long way in ruling in or out electrocution or intoxication, be they accidental or self-inflicted. Deaths may not be discovered immediately in SUDEP victims and the body may have begun to decompose, making interpretations complex and making a proper scene investigation even more germane. The process of decomposition has been the subject of careful study by forensic scientists and there is a large body of literature on the subject (Milroy 1999; Knight 1996). Occasionally, a body that is not immediately found in a home may be subject to predation by insects, rodents, or pets. Most experienced forensic pathologists are familiar with these phenomena and can differentiate marks upon the body from premortem injuries.

2.5╅SUDEP Deaths Outdoors While it is not common, SUDEP victims can experience a fatal attack outdoors with prompt or delayed discovery of the body. Obviously, the longer a body has lain outside, the more decomposition and environmental effects such as insects, vermin, and other animals will complicate examination of determinations on the body. Even in decomposed bodies, it may be possible to glean valuable information that may make a diagnosis of SUDEP practicable (Milroy 1999; Knight 1996). As with an indoor death scene investigation, the outdoor death scene can yield important information about what happened to the victim and rule in or€out foul play, suicide, or accident. The following case example (Case A) is illustrative.

2.6â•…Role of SUDEP in Traffic Deaths A not uncommon type of forensic case is that involving vehicular crashes in which the driver appears to have experienced some sort of attack that rendered that person unable to control the vehicle. These may be single-vehicle or multivehicle incidents. Such cases are often very difficult to evaluate from a pathological point of view, owing to the often overlying trauma to the body and brain caused by the accident, and also the coexistence of other disease processes, the most important and common being cardiovascular disease.

Forensic Considerations and Sudden Unexpected Death in Epilepsy

43

Case A A middle-aged man had been complaining of headaches and “strange” feelings and set out toward his automobile to go to his family physician for an examination. The man’s wife observed the man fall to the ground and apparently jerk beside his car. When she reached her husband, he was apparently dead. Autopsy revealed a 2.5-cm unruptured, encapsulated mass ἀlled with brownish greasy material indenting the right inferior frontal lobe anterior to the amygdala. Microscopic examination revealed the mass to be an epidermoid cyst. The remainder of the autopsy was unremarkable. It appeared that this man died from an attack that represented a generalized convulsion (GTC), apparently his ἀrst and last, from a lesion in a commonly epileptogenic region of the brain. From the standpoint of SUDEP criteria, this man was not known to be epileptic, although it appears that at the time of his death he experienced a GTC event and, thus, it could be argued the death was SUDEP.

A not uncommon cause of vehicular accidents is an epileptic attack that causes the driver-victim to lose control of the vehicle and incur injuries that may or may not prove to be fatal. In the case of fatal accidents, an examination of the vehicle, its contents, and condition may provide important information for the forensic pathologist. As is always the case, witness accounts and a medical history of the victim may point the way for further inquiry, especially if the victim was suffering from epilepsy and did not appear to have died from traumatic injuries. As in other cases of SUDEP, an effort should be made to determine what drugs the victim was taking, who prescribed them, and if the victim was compliant. A thorough autopsy examination will generally reveal if there are signiἀcant traumatic injuries and a similar thorough neuropathological examination will reveal any acute lesions, or chronic lesions in the brain that might have been epileptogenic. Examples of such lesions are old traumatic contusions, vascular anomalies, tumors, and congenital malformations (Leestma et al. 1989; Leestma 2009b). Autopsies of non-SUDEP deaths in epileptic persons generally show a relatively low incidence of structural lesions in the brain (about 10%), whereas SUDEP victims have a much higher incidence, between 50% and 70% (Leestma et al. 1985, 1989; Monte et al. 2007; Shields et al. 2002; Thom et al. 1999). Not all studies have shown signiἀcant neuropathology, however (Morentin and Alcaraz 2002). If there are acute traumatic lesions in the central nervous system, it may still be possible to determine if there are more chronic lesions present. It may, however, be unlikely to determine if the older lesions had anything to do with the vehicle crash, although sometimes collateral information may suggest a causal event. Examination of the oropharynx may reveal tongue, lip, or buccal bites that may indicate a recent seizure (Ulrich and Maxeiner 2003). Toxicological examination will reveal if anticonvulsants were present and if they were within therapeutic range. While many studies of SUDEP victims have shown that they have been or are taking multiple anticonvulsant medications, they are more likely than not to be noncompliant with respect to anticonvulsant medications (Bell and Sander 2006; George and Davis 1998; Langan 2000; Langan et al. 2005; Lear-Kaul et al. 2005; Lund and Gormsen 1985; McKee and Bodἀsh 2000; Monte et al. 2007; Nilsson et al. 1999; Tomson et al. 2005; Vlooswijk et al. 2007; Walczak 2003).

44 Sudden Death in Epilepsy: Forensic and Clinical Issues

Not all studies have shown this (Opeskin et al. 1999, 2000; Schwender and Troncoso 1986; Walczak et al. 2001), and some have suggested that toxicological studies may not truly reflect proper or improper drug levels (McGugan 1999), thus the role of medications on the risk of SUDEP is, at present, suggestive but has not been proven. Some have suggested that carbamazepine may play some role in SUDEP, either by some idiosyncratic reaction or in connection with changes in dosage of this drug (Hitiris et al. 2007; Nilsson et al. 1999, 2001; Timmings 1993; Walczak 2003). A thorough pathological examination of the heart must be performed to deἀne the extent of arteriosclerotic cardiovascular disease, previous areas of myocardial scarring or necrosis, and any other morphological abnormality of the heart or its valves. One inevitably encounters some degree of cardiovascular disease in accident victims and in sudden or unexpected deaths, and it becomes difficult to compare the importance of what is found to the death circumstances (Natelson et al. 1998; Scorza et al. 2007). There is a robust literature on the heart and sudden death (Okada and Kawai 1983; Rossi 1982; Schwartz and Gerrity 1975; Schwartz and Walsh 1971; Fineschi et al. 2006; Greenberg and Dwyer 1982; Bharati and Lev 1990) and on the so-called neurocardiology (Armour and Ardell 1984; Johnson et al. 1984), which cannot be reviewed here but the issue of cardiac dysfunction and pathology is discussed in detail elsewhere in this volume. Since SUDEP deaths are apparently physiological in the sense that the process that leads to SUDEP is electrophysiological and, thus, may or may not have a morphological counterpart such as a conduction system defect in the heart, an ion channelopathy, or some other deἀnable and demonstrable disease process, the science of forensic pathology can only go so far in determining the mechanism of death in some traffic-related deaths.

2.7â•…SUDEP and Deaths in Agitated Delirium or Restraint There is a category of sudden death in which the victim dies suddenly and unexpectedly while in a state of delirium that may involve an arrest or restraint circumstance, during which the victim dies. This problem has been the subject of a robust literature over many years and recent monographs by DiMaio and DiMaio (2006) and by Ross and Chan (2006) review much of it. It is germane to this discussion that some victims are epileptic. A typical restraint death circumstance is one where an individual is involved in an assault, robbery, disturbance, or some other event that may result in law enforcement personnel, emergency personnel, or rescue personnel being summoned to the scene. Once on the scene, there may be an apparent offender who is violent, disoriented, or delusional and may seem to need to be restrained or taken into custody to prevent flight, injury to the offender or to others. In the course of the activity of restraining the individual, many personnel may become involved in trying to subdue the violent person and in so doing may place handcuffs or restraint ties to the person and may place the individual in the prone position, “hog tie” them, and perhaps overlie the individual with several people. In this circumstance, the violent person may suddenly become quiet or complain of not being able to breathe and suffer a cardiorespiratory arrest from which the victim may or may not be able to be resuscitated. Such events almost inevitably result in some form of litigation in which the personnel involved in the restraint are held to account for what actions they may or may not have

Forensic Considerations and Sudden Unexpected Death in Epilepsy

Case B This 29-year-old man had apparently had a seizure disorder since childhood, for which he had been treated with a variety of anticonvulsant medications and a vagal nerve stimulator (about 3 years before this admission) in an effort to suppress his frequent seizures. On a recent occasion, he had been admitted to a hospital after having suffered a seizure. During admission the patient experienced a cardio/pulmonary arrest from which he was successfully resuscitated. The patient had a history of respiratory troubles including asthma. A brain CT on this admission was said to be negative. The shortness of breath the patient had complained of was thought possibly to be due to aspiration. As the patient began to emerge from his postictal state and resuscitation procedures, he was disoriented and agitated, for which he was medicated. He recovered from this episode. Seven months later in the presence of his mother, the patient experienced a seizure event and apparently, while postictal, he became delirious and combative and assaulted his mother, who called for assistance. Police and ἀre department personnel responded and subdued the man by placing him in handcuffs and hobbles in a prone position on a litter. Soon afterward, he became unconscious and pulseless. Immediately after emergency medical personnel arrived, the patient had a heart rate of 140 and a respiratory rate of 28, but then became pulseless. Resuscitative treatment was begun and the man was conveyed to an emergency facility where he was reported to be asystolic on admission. During resuscitation, the patient urinated and vomited. Resuscitation did not succeed and the patient was pronounced dead. The victim was known to be hypertensive and to have had temporal lobe epilepsy for many years. When he was postical, it was not uncommon for him to be delusional and agitated. Autopsy examination reported a body weight of 251 lb and a height of about 72 in. A number of bruises and abrasions to the face, arms, and hands were noted. There were petechial hemorrhages of the lip mucosa and right flank. A subcutaneous hemorrhage was noted near the left occiput. There was evidence of food aspiration. Heart–blood toxicological examination revealed 10 µg/ml barbiturates, 40 mS. (From Tigaran, S., Acta Neurol Scand Suppl, 177, 9–32, 2002. With permission.)

1. The changes are due to subendocardial scarring present in the epilepsy patients’ myocardium. However, because of the discrete nature of ἀbrosis found postmortem in the hearts of the epilepsy patients, we may not be able to detect these abnormalities with sufficient sensitivity by means of the present technique. The author must admit that the results following the investigation of the ἀrst ἀve patients enrolled in the study, all of which exhibited positive late potentials, were surprising. Consequently, it was decided to repeat the investigation after the antiepileptic drugs were withdrawn (the patients did not receive any other drugs but antiepileptics), and simultaneously control the antiepileptic drug plasma concentration when investigating the rest of the patients, both in the medicated and in the unmedicated Positive *Vector Butterworth 40–250 Hz

Negative RMS40 : 10.2 µV RMS50 : 19.2 µV QRS : 140 mS

*Vector Butterworth 40–250 Hz

RMS40 : 25.3 µV RMS50 : 40.1 µV QRS : 101 mS

LPD : 46 mS

LPD : 24 mS

Noise : 0.4 µV

Noise : 0.3 µV

200 mm/s 10 µV/10 mm

200 mm/s 10 µV/10 mm

Figure 7.2╇ Example positive versus negative LP image. (From Tigaran, S., Acta Neurol Scand Suppl, 177, 9–32, 2002. With permission.)

Sudden Unexpected Death in Epilepsy

115

phase. Our ἀrst conclusion, coupled with the delayed depolarization of the left ventricle observed in 47% (positive late potentials) of our patients on drugs when compared with the off-drugs phase, led us to the second working hypothesis. 2. The changes are due to the antiepileptic drugs, thus generating the hypothesis that antiepileptics might also influence the electrical properties of the myocyte due to their electrophysiological effects on the conduction system and the myocardium. As previously shown, many antiepileptic drugs exert their effects by blocking sodium channels, which generally slow conduction. Several cardiac antiarrhythmic drugs also exert their effects by blocking this channel (Anderson et al. 1994). This relationÂ� ship has also been observed in other clinical studies of sodium channel blocker drugs. Antiepileptic drugs may thus, through their as yet unrecognized influence on the depolarization process (Freedman and Steinberg 1991), potentially increase the propensity for malignant arrhythmias. In a larger perspective, this could be a potential so-called proarrhythmic effect, which could precipitate cardiac arrhythmias. To the best of the author’s knowledge, there are no available data describing the relationship between the occurrence of late potentials and antiepileptic drugs. While not claiming to have found the ultimate explanation, the author would like to propose that although the two hypotheses are apparently discordant, both variants may be true. Likewise a combination of both hypotheses would also be a possibility, an idea supported by the ἀndings of the electrocardiographic positive ST segment and the positive late potentials. However, the interpretation of these ἀndings requires much caution, warranting further studies to verify the reproductibility of the described ἀndings (Tigaran et al. 2002). Future studies employing the magnetic resonance imaging (MRI) technique will help in clarifying the questions rained by these two hypotheses.

7.6â•…Evidence of Sinus Tachycardia Associated with Epileptic Seizures More than 25 years ago, Marshall et al. (1983) wrote that, “a seldom-recognized accompaniment of temporal lobe seizures is tachycardia.” They were the very ἀrst authors to document by simultaneous “electroencephalographic, electrocardiographic, and videotape monitoring” the presence of this phenomenon in 12 consecutive patients with spontaneous seizures. Several subsequent articles have described this phenomenon, and as a result of the analysis of more than 1500 seizures from more than 1000 epilepsy patients, all of them established without any doubt that, despite the origin of the epileptic seizure, sinus tachycardia is by far the most common cardiac phenomenon temporally associated with epileptic seizures, as one would expect as part of a normal response to stress. Some of the previous articles in the literature are lacking in details related to the exact brain origin of the seizure or the exact type of seizure, i.e., secondary generalization versus complex versus simple partial seizures. In the older studies, this inconsistency is due to the technical limitations of the imaging methods employed at the time of the study. Unfortunately, at times, the electroencephalographic onset of the seizure can be deceiving due to either rapid propagation of the seizure or inadequacy of the scalp EEG recording(s), as shown by studies using intracranial electrodes or magnetoencephalogram studies (Shibasaki et al. 2007). Overall, the heart rate is thought to be signiἀcantly higher with seizures arising from the temporal than from the frontal lobe (Galimberti et al. 1996; Schernthaner et al.

116 Sudden Death in Epilepsy: Forensic and Clinical Issues

130

1999; Garcia et al. 2001; Opherk et al. 2002; Tigaran et al. 2002; Leutmezer et al. 2003) and to occur earlier in patients whose seizures arise during sleep (Opherk et al. 2002). Of note, Tigaran et al. (2003) showed that epileptic seizures induce a rapid increase in heart rate from resting levels to more than 180 beats/min shortly before the onset of electrocardiographic seizures. This early increase in heart rate indicates that seizures can induce a turning on/off of very high levels of efferent neural cardiac sympathetic activity (see Figure 7.3). Such high levels of sympathetic activity (Tigaran et al. 2001) become toxic principally to the cardiac myocyte and could, in the long run, be a contributing factor to the scarring of the myocardium observed in SUDEP victims (see Figure 7.4) (P-Codrea Tigaran et al. 2005; Natelson et al. 1998; Falconer and Rajs 1976), as well as the formation of cardiac contraction bands (Manno et al. 2005). Moreover, Tigaran et al. (2003) have demonstrated the occurrence of myocardial ischeÂ� mia triggered by high heart rate values, as shown by the presence of electrocardiographic ST-segment depression, which is associated with an epileptic seizure (see Figure 7.5), and could represent an additional contributing factor to the pathologic changes of the myocardium, such as ἀbrotic scarring. Strikingly, the failure of the recorded heart rate to recover to the baseline level for as long as 1 h after seizure cessation can be a possible result of the impaired autonomic function (Ushijima et al. 2009) such as Nei et al. (2000a) have demonstrated to be present in some of their epilepsy patients. It is important to consider that, in epilepsy patients, tachycardia triggering cardiac ischemia may have even more serious consequences, especially in patients who already have cardiac disease. This suggests that methods of addressing it might be important and worthy of investigation, such as, for example, blunting the ischemic and heart rate response with beta-blockers. The observations and present hypothesis represent an exciting challenge with regard to experimental clinical testing. Salutary in this context are studies looking into the association between tachycardia and diverse types of ethnic groups. A pioneer in this ἀeld of research, Wilder-Smith

110 100 90 70

80

Average heart rate

120

No ST-segment change ST-segment change

0

10

20

30

40

50

60

Minutes relative to seizure

Figure 7.3╇ Average heart rate (HR) over time by ST-segment changes. 0 = max HR during seizure. (From Tigaran, S., Acta Neurol Scand Suppl, 177, 9–32, 2002. With permission.)

Sudden Unexpected Death in Epilepsy 9

1

2

117 1 2 3 4 5 6 7 8 9 10 11 12 13 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

8

3

10

4

7

5

6

11

(a)

X

1 2 3 4 5 6 7 8 9 10 11 12 13

8

3

10

4 5

(b)

9

1

2

7 6

11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

X X X

Figure 7.4╇ Comparison between the distribution of the fibrotic changes of the deep and sub-

endocardial myocardium in SUDEP versus control patients. (a) Control patients for fibrotic myocardial areas. (b) Fibrotic areas in the hearts of SUDEP patients. 1–15 (vertical), subject’s identification number; 1–11 (horizontal), sections from the transmural tissue blocks containing the deep and endocardial myocardium; 12, longitudinal tissue block from intermedial septum; 13, longitudinal tissue block from the papillary muscle of the left ventricle; X, presence of fibrotic changes in singular areas which are not present in the stylized figure. Light gray: slight fibrosis: 1 patient with fibrotic changes confined to 1 area. Darker gray, moderate fibrosis: 2 patients with fibrotic changes observed in the same area. Darkest gray, severe fibrosis: 3 patients with fibrotic changes observed in the same area. (From P-Codrea, Tigaran, S. et al., Am J Forensic Med Pathol, 26 (2), 99–105, 2005. With permission.)

(Wilder-Smith and Lim 2001), published a study concerning the changes in heart rate amongst non-Caucasian Singaporean patients. Notably, this study concluded that sinus tachycardia was considerably less frequent in non-Caucasian epilepsy patients. Despite being a very comprehensive review of SUDEP and its mechanism, the Nigerian Kwara State article lacked regional ethnic data that could have addressed this particular topic (Sanya 2005). By the same token, it is quite possible that the American melting pot, more than anywhere else, will claim its own speciἀc studies looking at the issue of racial and ethnic differences in epilepsy-related cardiac arrhythmias.

118 Sudden Death in Epilepsy: Forensic and Clinical Issues

00:49:13 Pre start of Event

01:03:57 Pre start of Event

00:49:47 Max. depress.

01:08:28 Max. depress.

00:51:33 Post end of Event

01:12:11 Post end of Event

Figure 7.5╇ Example from Patient 1 (left) and Patient 9 displaying dynamic ST-segment depression in relation to seizure with secondary generalization. (From Tigaran, S., Acta Neurol Scand Suppl, 177, 9–32, 2002. With permission.)

A conceptually different idea, outside of the direct association between heart rate variations and SUDEP, involves the epidemiology of obesity, which is thought to be a speciἀc important risk factor for occult cardiovascular disease. Thus, complications of obesity in epilepsy patients might influence even more the ability of the heart to adjust to abrupt, sudden variations in its rate and represent a contributory mechanism in SUDEP. Obesity is a chronic metabolic disorder associated with cardiovascular disease and increased morbidity and mortality. It is a widely acknowledged risk factor for developing coronary artery disease (Franzosi 2006). The number of deaths per year attributable to obesity is about 30,000 in the United Kingdom, a country that has produced a signiἀcant amount of SUDEP data. This number is 10 times higher in the United States, roughly about 300,000 deaths per year (Allison et al. 1999), where more than half of the adults are overweight (Flegal et al. 1998) and where obesity is thought to have overtaken smoking in 2005 as the main preventable cause of illness and premature death (Franzosi 2006). If we do death rate extrapolations for the United States for obesity, there will be 300,000 deaths per year, 25,000 deaths per month, 5769 deaths per week, 821 deaths per day, and 34 deaths per hour (WrongDiagnosis.com 2009). Moreover, we also have to keep in mind the speciἀc association between obesity and obstructive sleep apnea, which represents an aggravating and important contributing factor to the genesis of cardiovascular diseases (Hiestand et€al. 2006). Since apnea has already been identiἀed to be associated with epileptic seizures and SUDEP (Nashef et al. 1996; So et al. 2000), this association should prompt even more attention and many more research studies. Another notable association, which might be very signiἀcant for epilepsy and SUDEP, is the signiἀcant increase in the older population and the speciἀc increase of epilepsy prevalence in this age group (Velez and Selwa 2003). Accordingly, the speciἀc concurrent morbidities, such as stroke and cardiac diseases, present in this age category will prompt an even higher awareness of the cardiac complications in epilepsy.

Sudden Unexpected Death in Epilepsy

119

7.7â•…Bradycardia and Asystole Asystole and atrioventricular block are recognized as potentially malignant conduction abnormalities that can lead to ventricular ἀbrillation (Tigaran et al. 2002). Thus, bradyarrhythmia could be particularly risky in patients with coexistent, underlying ischemic heart diseases; this emphasizes the importance of recognizing these cases. Ictal bradycardia, cardiac asystole, and total atrioventricular block associated with epileptic seizures is to date a more seldom but noteworthy complication of epilepsy. This association is probably still underestimated because of the lack of recognition of cardiac complications in epilepsy, as mentioned before. Because of its implication in the pathophysiology of SUDEP, there were two earlier literature reviews on bradyarrhythmias and asystole as cardiac consequences of epileptic seizures (Devinsky et al. 1997; Tinuper et al. 2001). Several other case reports have been published since 2001 (Rocamora et al. 2003; Venugopalan et al. 2001; Mondon et al. 2002; Leutmezer et al. 2003; Toth et al. 2008; Carinci et al. 2007; Strzelczyk et al. 2008; Schuele et al. 2008; Ghearing et al. 2007; Almansori et al. 2006; Mascia et al. 2005; So and Sperling 2007; Rugg-Gunn et al. 2004a; Carvalho et al. 2004), adding valuable information about the occurrence of bradycardia in epilepsy. Of note, in this context, are recent studies from the Mayo Clinic (Britton et al. 2006), which showed no consistent hemispheric lateralization of seizure activity at the onset of bradycardia. Another study (Ghearing et al. 2007) found over a 14-year period that only 29 seizures were associated with ictal bradycardia in 13 patients. Out of the 29 bradycardia episodes, 7 patients had a total of 11 complex partial seizures that were associated with asystole. Despite the apparent multitude of published case reports, this association still seems to be infrequent. A feasible explanation of the paucity of data regarding the association between epilepsy and bradycardia may be an underestimation of this association due to the lack of adequate monitoring of the epilepsy patients, with concurrent electrocardiographic and intracranial electroencephalographic analysis (Leung et al. 2006). Important in this context, Rugg-Gunn et al. (2004b) developed a new monitoring strategy. Epilepsy patients were implanted with subcutaneous electrocardiogram loop recorders for an average of 18 months. Of the 19 patients, 4 (21%) developed bradycardia or asystole, for which a permanent pacemaker was deemed appropriate. Three of these episodes occurred at the time of a clinical seizure and one was not associated with a known clinical event. This study clearly showed how inadequate it is to simply record electroencephalograms and electrocardiograms for a few days or a few seizures. Thus, the ἀndings of Rugg-Gunn and colleagues (2004b) should be applauded. Other investigators should try to replicate these ἀndings in order to acquire more evidence-based data before we apply these ἀndings to the routine care of people with epilepsy (Hirsch and Hauser 2004).

7.8â•…Ictal Cardiac SPECT Imaging Supportive of a Cardiac Mechanism of SUDEP To the best of our knowledge, there is only one study illustrating focal myocardial perfusion defects concurrent with electrocardiographic ST-segment depression at the cessation of an epileptic seizure as an indicator for myocardial ischemia (see Figure 7.6) (Tigaran 2001; Dam et al. 2001). Notably, the criteria and the methods available for the detection of the myocardial ischemia with myocardial perfusion scintigraphy in this study were not

120 Sudden Death in Epilepsy: Forensic and Clinical Issues

Seizure

Rest

Perfusion defect during seizure

Patient 8

Figure 7.6╇ Perfusion defect during epileptic seizure.

used to determine if ischemia could occur globally in the myocardium. For future studies, the limitation of employing this type of imaging technique could be potentially addressed by the use of newer techniques, such as hybrid positron emission tomography/computerized tomography (PET/CT) or PET/magnetic resonance imaging systems, which would be able to detect more accurately any perfusion defects (Slomka et al. 2008). In data from an unpublished study conducted by the author of this chapter, 14 men and 9 women (age range 20–59 years, mean 42) with intractable focal epilepsy for 4 to 55 years (mean age 26 years) participated in this prospective study. All patients had normal baseline results on electrocardiography, Holter monitoring, echocardiography, myocardial scintigraphy at rest, and coronary angiography. Technetium 99m Cardiolite (DuPont Pharma, UK) was injected into 18 unmedicated patients during ictal events in which nine patients experienced secondarily generalized tonic–clonic seizures with a mean of maximum ictal heart rate of 133 beats/min. The other seven patients were injected in relation to complex partial seizures with an average maximal ictal heart rate of 108 beats/min. One patient received the tracer during a simple partial status epilepticus episode, during which the Holter recording was not available. Of the 18 myocardial perfusion studies, there were several ictal studies positive for focal myocardial perfusion defects (n = 3 or 16%). Unfortunately, only one was a positive ictal study in which the tracer was injected in direct relationship to a seizure where a 1-mm ST-segment depression, which is reflective of myocardial ischemia, was detected on the Holter monitor subsequent to a complex partial seizure without any motor phenomenon, recorded at 03:05 p.m. The maximum ictal heart rate during this seizure was 134 beats/min. This patient, a 56-year-old female with MRI-veriἀed right-sided hippocampus gliosis, had both infero- and anteroseptal reversible defects. This particular ἀnding may advocate for SUDEP being more frequent in women than was previously estimated (Walczak et al. 2001). The patient was moderately obese and had a normal 12-lead electrocardiogram, rest and stress test, and myocardial scintigraphy, as well as normal coronary angiography. Her baseline plasma norepinephrine was higher than normal, namely 1033 pg/mL. Her urine norepinephrine was 55,200 pg/mL

Sudden Unexpected Death in Epilepsy

121

or 135 μg per 24 h. Her baseline plasma epinephrine was also elevated above the normal reference interval, up to 87 pg/mL, and her urine norepinephrine was 13,800 pg/mL or 34€μg per 24€h (diuresis was 2450 mL for 24 h when the values were collected). Normal reference ranges are laboratory-speciἀc, vary according to methodology of testing, and differ between blood and urine samples. Supine (lying down): epinephrine less than 50 pg/ mL, norepinephrine less than 410 pg/mL; reference ranges for urine catecholamines are: epinephrine 0–20 μg per 24 h and norepinephrine 15–80 μg per 24 h (Answers.com 2009). The ἀndings in the above case concur with those of Nei et al. (2004), who suggests “. . . that patients with evidence of a great degree of change in autonomic tone during seizures might be at increased risk for SUDEP.” During the epileptic seizure, an almost threefold increase was noted in both plasma and urine catecholamines (plasma norepinephrine, 2891 pg/mL; plasma epinephrine, 296 pg/mL; urine norepinephrine, 121,300 pg/mL; and urine epinephrine, 29,700 pg/mL) for a total diuresis of 2400 mL. During the seizure, a drop in O2 saturation from 100% to 93% was recorded by the pulse oximeter, which did not represent any signiἀcant drop in the O2 saturation. The oxygen saturation dropped, but not below the limit of 90%. Most physicians would not be concerned when the oxygen saturation is above 90% (93% in this study). The imaging results from a small cohort of drug refractory epilepsy patients, whose antiepileptic drugs were withdrawn during the course of the study, and without any evidence of cardiovascular disease, substantiate the hypothesis that cardiac ischemic abnormalities may exist and could potentially provide some of the pathophysiologic explanation for SUDEP. The data also suggest that epileptic seizures of temporal origin may induce myocardial ischemia in the absence of coronary pathology, presumably by autonomicÂ�mediated vasospasm. This ἀnding points again toward an association between cardiac abnormalities in epilepsy and the sudden, large imbalance in cardiac efferent autonomic activity (Druschky et al. 2001) that may result in autonomic-mediated ischemia and, thus, possible fatal arrhythmia. Unfortunately, this study (Tigaran 2001) has several limitations. The tracer was simultaneously injected during the electrocardiographic detected myocardial ischemia as reflected by the ST-segment changes in one patient out of the 23 who were enrolled in the study. Studies related to the anatomical brain localization of the site of the seizure employing the subtraction ictal SPECT coregistered to MRI (SISCOM) method documented the importance of the timing for the tracer injection (O’Brien et al. 1998). Late injection of the radiotracer (>45 s) after the onset of the epileptic seizure was associated with a falsely localizing or nonlocalizing SISCOM study, a ἀnding that could be extrapolated to our epilepsy study. This might represent the explanation for why 15 of the 18 scans were negative myocardial perfusion studies.

7.9â•…Future Research Avenues for an Underlying Cardiac Mechanism of SUDEP: Epilepsy, Depression, and Cardiac Disease For centuries poets and folklore have asserted that there is a relationship between the mind and body in general, and human moods and the heart in particular. Almost 400 years ago Shakespeare wrote, “. . . My life being made of four, with two alone sinks down to death, oppress’d with melancholy . . . .” (Shakespeare). However, only in the past few years has this

122 Sudden Death in Epilepsy: Forensic and Clinical Issues

conviction been scientiἀcally tested (Glassman and Shapiro 1998). Indeed, it has also been proved that depression is the most frequent psychiatric disorder in patients with epilepsy and that it is more common in patients with partial seizure disorders of temporal or frontal lobe origin and among patients with poorly controlled seizures. In three communitybased studies, prevalence rates of depression ranged between 21% and 33% among patients with persistent seizures and between 4% and 6% among seizure-free patients (LaFrance et al. 2008). Ettinger et al. (2005) reported the results of a population-based survey that investigated a lifetime prevalence of depression, epilepsy, diabetes, and asthma in 185,000 households. Among the 2900 patients with epilepsy, 32% reported having experienced at least one episode of depression. This contrasted with an 8.6% prevalence among healthy respondents, 13% among patients with diabetes, and 16% among people with asthma. For a long time, people speculated about whether depression increases the risk of mortality only in individuals with established coronary disease, or if those without a history of heart disease are at increased risk as well. Glassman and Shapiro (1998) concluded that both groups are at risk, but the risk is higher in groups with established coronary artery disease. This conclusion was validated in a study by Penninx et al. (2001), which concluded that, regardless of the preexistence of coronary artery disease, the effects of depression result from the same mechanisms in both groups. Because sudden cardiac death accounts for most of the excess mortality in association with depression in patients with established coronary artery disease, one should search for proarrhythmic mechanisms associated with depression (Carney et al. 2001). Multiple possibilities exist. Heart rate variability is lower in depressed than in nondepressed patients with established coronary artery disease. In addition, plasma catecholamines, known provokers of arrhythmias and sudden cardiac death, are elevated in depressed patients (Carney et al. 2001). This is also the case in epilepsy patients (Tigaran et al. 2001; Lathers et al. 1984, 2008). Especially supportive of the altered autonomic tone in depression is the study of Sheline et al. (1996) who demonstrated that patients with a history of major depression had signiἀcantly smaller hippocampal volumes bilaterally when compared with matched control subjects. The decrement in hippocampal volume correlates with the cumulative lifetime duration of major depression, possibly as result of a progressive process mediated by glucocorticoid neurotoxicity (Carney et al. 2001). This process may be responsible for an increase in corticotropin-releasing factor secretory drive and may thereby contribute to the elevated hypothalamic–pituitary–adrenal axis activity observed in depression. Corticotropin-releasing factor is also a potent stimulus for sympathetic nervous system activation, which may account for the sympathetic hyperactivity observed in major depression (Carney et al. 1999, 2001) and in epilepsy patients (Nei et al. 2000a). In concert with the hippocampal ἀndings in depression, recent imaging studies in epilepsy patients revealed the existence of similar pathological ἀndings, namely those with right temporal lobe epilepsy and depression have hippocampal atrophy that cannot be explained by epilepsy alone (Shamim et al. 2009). In this context and supportive of this theory is the recent study of Bernhardt et al. (2009), which provides a unique quantitative assessment in patients with temporal lobe epilepsy. The study indicates that regions of the brain remote from the lobe from which the seizure originates may be adversely affected with signiἀcant changes in cerebral cortical structures that may be important in the development of comorbidities (Cascino 2009). Unfortunately, the sample size was too small to allow any conclusions regarding the left side hippocampus. Until now, there have been no studies looking into the association between the size of the hippocampal sclerosis, epilepsy, and depression in SUDEP patients. The ἀndings of

Sudden Unexpected Death in Epilepsy

123

this recent imaging study (Shamim et al. 2009) may open new research avenues for a better understanding of the SUDEP phenomenon, and a possible explanation of why not all epilepsy patients have an equal risk of dying of SUDEP. Finally, depression is associated with platelet activation and with inflammatory processes that may increase the risk of developing coronary artery disease or, in patients with established coronary artery disease or myocardial infarction (Carney et al. 2001), which might also be the case in epilepsy.

7.10â•…Evidence of Cardiac Arrhythmogenic Substrate for SUDEP: The Influence of Antiepileptic Drugs on the Heart Blumhardt et al. (1986) suggested that treated patients with epilepsy had lower mean rates of the ictal cardiac acceleration than the untreated patients, whereas Opherk et al. (2002) stated that antiepileptic drug regimes were mostly similar in seizure patients with and without cardiac abnormalities. However, due to the small number of patients studied, further conclusions concerning the possible protective effect of antiepileptic drugs were precluded. Others studies evaluated the role of heart rate, but did not specify the part played by antiepileptic drugs upon the heart rate. In contradistinction to the antiepileptic drug protective role theory, Devinsky et al. (1994) showed that patients with epilepsy have greater blood pressure and heart rate variability and reactivity than control patients, with those ἀndings partly attributable to carbamazepine. Conversely, by employing the technique of the electrocardiographic late potentials (see Figure 7.2) (Tigaran et al. 2002), it has been shown, with no preference for any of the antiepileptic drugs, that all may also influence the electrical properties of the myocytes, mostly through an inactivation of the sodium channels, thus potentially precipitating cardiac arrhythmias. The electrocardiographic late potential technique was shown to reveal the presence of diseased myocardium with delayed depolarization, which may serve as the substrate for reentrant arrhythmias causing ventricular tachycardia and sudden cardiac death (Tigaran et al. 2002). Further work is needed to determine whether techniques such as signal-averaged electrocardiography (SAECG) or other, newer techniques will be useful screening tools to identify whether some persons with epilepsy are at risk for SUDEP.

7.11â•…Summary and Clinical Perspectives The concept that abnormal electrical discharges in the brain trigger cardiac arrhythmias does not seem to be widely recognized by clinicians, although it has long been proposed in the neurological literature (Blumhardt et al. 1986; Mameli et al. 2006; Jallon 1997). Unfortunately, even now, electrocardiographic recordings are not always conducted or reviewed during epilepsy monitoring. Thus, the deἀnite relationship between epileptic seizures and cardiac abnormalities is still the subject of an ongoing debate. The diversity of the methods employed in the assessment of this relationship, in contrast to the use of only a single electrocardiogram rhythm monitoring channel that is usually a supplement to the EEG recordings, along with the infrequent use of simultaneous long-term EEG-Holter recordings, presents another limitation of this type of investigation.

124 Sudden Death in Epilepsy: Forensic and Clinical Issues

Of note is the consistent lack of respiratory recordings, which need to be conducted during assessments related to electroencephalographic investigations before epilepsy surgery or admittance for differential diagnosis purposes, since apnea and hypoxia clearly represent a major contributory mechanism to the occurrence of cardiac events (Schulz et al. 2006). Nonetheless, the variable results from studies seeking a relationship between epilepsy and the risk of potentially fatal cardiac events, as discussed in this chapter, support the need for different animal models (Lathers et al. 2008, 2010, Chapter 28) that would provide comparison with clinical study data from patients with epilepsy. Also, studies addressing the association between epilepsy, depression, and cardiac abnormalities and SUDEP are deemed to be of future research importance. We need to encourage cardiologists to exercise caution when concluding from Holter monitoring alone, especially in the setting of ambulatory recordings that presumed cardiogenic arrhythmias, even when they coincide with symptoms, are the primary cause of any patient’s complaints. Only prolonged monitoring, deἀned as at least 24 h of recordings of simultaneous EEG and ECG records, will improve the diagnostic yield and, consequently, optimize the treatment of the patients with seizure-related arrhythmias and also of persons with cardiogenic arrhythmias that can result in seizure like activity. Moreover, besides the determination of the genetic etiology of epilepsy and association with depression, it would be useful to conduct collaborative studies aimed at evaluating both the epileptogenic and the cardiologic aspect of the underlying substrate that may lead to increasing the chance of potentially fatal events. Only by making both doctors and patients aware of the presence of the potential for adverse symbiosis between neurological and cardiac abnormalities will the quality of preventive services, especially as relates to SUDEP, be improved. Only time, education, and interdisciplinary joint efforts will determine whether well-conducted research will be feasible in large population samples.

Acknowledgment For valuable criticism when reviewing this book chapter, I have to thank Allan S. Jaffe, MD, Mayo Clinic, Rochester, MN.

References Allison, D. B., K. R. Fontaine, J. E. Manson, J. Stevens, and T. B. VanItallie. 1999. Annual deaths attributable to obesity in the United States. JAMA 282 (16): 1530–1538. Almansori, M., M. Ijaz, and S. N. Ahmed. 2006. Cerebral arrhythmia influencing cardiac rhythm: a case of ictal bradycardia. Seizure 15 (6): 459–461. Anderson, J. L., E. V. Platia, A. Hallstrom, R. W. Henthorn, T. A. Buckingham, M. D. Carlson, and P. E. Carson. 1994. Interaction of baseline characteristics with the hazard of encainide, flecainide, and moricizine therapy in patients with myocardial infarction. A possible explanation for increased mortality in the Cardiac Arrhythmia Suppression Trial (CAST). Circulation 90 (6): 2843–2852. Answers.com. 2009. Catecholamines Tests: Normal Results. Available from http://www.answers╉ .com/catecholamines-tests-normal-results (accessed February 2009).

Sudden Unexpected Death in Epilepsy

125

Aung-Din, R., J. H. Mitchell, and J. C. Longhurst. 1981. Reflex alpha-adrenergic coronary vasoconstriction during hindlimb static exercise in dogs. Circ Res 48 (4): 502–509. Begun, J. W., J. Tornabeni, and K. R. White. 2006. Opportunities for improving patient care through lateral integration: the clinical nurse leader. J Healthc Manag 51 (1): 19–25. Beran, R. G. 2006. SUDEP—To discuss or not discuss: That is the question. Lancet Neurol 5 (6): 464–465. Bernhardt, B. C., K. J. Worsley, H. Kim, A. C. Evans, A. Bernasconi, and N. Bernasconi. 2009. Longitudinal and cross-sectional analysis of atrophy in pharmacoresistant temporal lobe epilepsy. Neurology 72 (20): 1747–1754. Blumhardt, L. D., P. E. Smith, and L. Owen. 1986. Electrocardiographic accompaniments of temporal lobe epileptic seizures. Lancet 1 (8489): 1051–1056. Brazier, J. R., and G. D. Buckberg. 1975. Effects of tachycardia on the adequacy of subendocardial oxygen delivery in experimental aortic stenosis. Am Heart J 90 (2): 222–230. Britton, J. W., G. R. Ghearing, E. E. Benarroch, and G. D. Cascino. 2006. The ictal bradycardia syndrome: localization and lateralization. Epilepsia 47 (4): 737–744. Carinci, V., G. Barbato, A. Baldrati, and G. Di Pasquale. 2007. Asystole induced by partial seizures: a rare cause of syncope. Pacing Clin Electrophysiol 30 (11): 1416–1419. Carney, R. M., K. E. Freedland, and A. S. Jaffe. 2001. Depression as a risk factor for coronary heart disease mortality. Arch Gen Psychiatry 58 (3): 229–230. Carney, R. M., K. E. Freedland, R. C. Veith, and A. S. Jaffe. 1999. Can treating depression reduce mortality after an acute myocardial infarction? Psychosom Med 61 (5): 666–675. Carvalho, K. S., V. Salanova, and O. N. Markand. 2004. Cardiac asystole during a temporal lobe seizure. Seizure 13 (8): 595–599. Cascino, G. D. 2009. Temporal lobe epilepsy is a progressive neurologic disorder: Time means neurons! Neurology 72 (20): 1718–1719. Chin, P. S., K. R. Branch, and K. J. Becker. 2004. Myocardial infarction following brief convulsive seizures. Neurology 63 (12): 2453–2454. Christiansen, E. H., L. Frost, H. Mlgaard, P. E. Thomsen, T. T. Nielsen, and A. K. Pedersen. 1995. The signal-averaged ECG becomes late potential-positive at low noise levels in healthy subjects. Eur Heart J 16 (11): 1731–1735. Dam, M., S. Tigaran, K. Rasmussen, M. Rehling, and H. Molgaard. 2001. Sudden unexpected death in epilepsy (SUDEP): Regional myocardial blood flow during seizure. Neurology 56 (S-3): A233. DeGiorgio, C. M., P. Miller, S. Meymandi, and J. A. Gornbein. 2008. n-3 fatty acids (ἀsh oil) for epilepsy, cardiac risk factors, and risk of SUDEP: Clues from a pilot, double-blind, exploratory study. Epilepsy Behav 13 (4): 681–684. Devinsky, O., K. Perrine, and W. H. Theodore. 1994. Interictal autonomic nervous system function in patients with epilepsy. Epilepsia 35 (1): 199–204. Devinsky, O., S. Pacia, and G. Tatambhotla. 1997. Bradycardia and asystole induced by partial seizures: a case report and literature review. Neurology 48 (6): 1712–1714. Druschky, A., M. J. Hilz, P. Hopp, G. Platsch, M. Radespiel-Troger, K. Druschky, T. Kuwert, H. Stefan, and B. Neundorfer. 2001. Interictal cardiac autonomic dysfunction in temporal lobe epilepsy demonstrated by [(123)I]metaiodobenzylguanidine-SPECT. Brain 124 (Pt 12): 2372–2382. Duncan, H. W., R. J. Barnard, G. K. Grimditch, J. Vinten-Johansen, and G. D. Buckberg. 1987. Cardiovascular response to sudden strenuous exercise. Basic Res Cardiol 82 (3): 226–232. El Sherif, N. 1994. Re-entrant mechanism in venticular arrhythmias. In Cardiac Electrophysiology: From Cell to Bedside, ed. D. Zipes and J. Jalife. Philadelphia, PA: W. B. Saunders. Espinosa, P. S., J. W. Lee, U. B. Tedrow, E. B. Bromἀeld, and B. A. Dworetzky. 2009. Sudden unexpected near death in epilepsy: Malignant arrhythmia from a partial seizure. Neurology 72 (19): 1702–1703. Ettinger, A. B., M. L. Reed, J. F. Goldberg, and R. M. Hirschfeld. 2005. Prevalence of bipolar symptoms in epilepsy vs other chronic health disorders. Neurology 65 (4): 535–540.

126 Sudden Death in Epilepsy: Forensic and Clinical Issues Falconer, B., and J. Rajs. 1976. Post-mortem ἀndings of cardiac lesions in epileptics: A preliminary report. Forensic Sci 8 (1): 63–71. Flegal, K. M., M. D. Carroll, R. J. Kuczmarski, and C. L. Johnson. 1998. Overweight and obesity in the United States: Prevalence and trends, 1960–1994. Int J Obes Relat Metab Disord 22 (1): 39–47. Foster, C., J. D. Anholm, C. K. Hellman, J. Carpenter, M. L. Pollock, and D. H. Schmidt. 1981. Left ventricular function during sudden strenuous exercise. Circulation 63 (3): 592–596. Foster, C., D. S. Dymond, J. Carpenter, and D. H. Schmidt. 1982. Effect of warm-up on left ventricular response to sudden strenuous exercise. J Appl Physiol 53 (2): 380–383. Franzosi, M. G. 2006. Should we continue to use BMI as a cardiovascular risk factor? Lancet 368 (9536): 624–625. Freedman, R. A., and J. S. Steinberg. 1991. Selective prolongation of QRS late potentials by sodium channel blocking antiarrhythmic drugs: Relation to slowing of ventricular tachycardia. Electrophysiologic Study Versus Electrocardiographic Monitoring Trial (ESVEM) Investigators. J Am Coll Cardiol 17 (5): 1017–1025. Galimberti, C. A., E. Marchioni, F. Barzizza, R. Manni, I. Sartori, and A. Tartara. 1996. Partial epileptic seizures of different origin variably affect cardiac rhythm. Epilepsia 37 (8): 742–747. Garcia, M., C. D’Giano, S. Estelles, R. Leiguarda, and A. Rabinowicz. 2001. Ictal tachycardia: Its discriminating potential between temporal and extratemporal seizure foci. Seizure 10 (6): 415–419. Ghearing, G. R., T. M. Munger, A. S. Jaffe, E. E. Benarroch, and J. W. Britton. 2007. Clinical cues for detecting ictal asystole. Clin Auton Res 17 (4): 221–226. Glassman, A. H., and P. A. Shapiro. 1998. Depression and the course of coronary artery disease. Am J Psychiatry 155 (1): 4–11. Hanna, J. 2007. Epilepsy and sudden death: A personal view. Epilepsia 38 (s11): S3–S5, doi: 10.1111/ j.1528-1157.1997.tb06121.x. Harris, C. D. 2005. Neurophysiology of sleep and wakefulness. Respir Care Clin N Am 11 (4): 567–586. Heart, Cambridge. 2009. T-Wave Alternans. Available from http://www.cambridgeheart.com. Hiestand, D. M., P. Britz, M. Goldman, and B. Phillips. 2006. Prevalence of symptoms and risk of sleep apnea in the US population: Results from the national sleep foundation sleep in America 2005 poll. Chest 130 (3): 780–786. Hirsch, L. J., and W. A. Hauser. 2004. Can sudden unexplained death in epilepsy be prevented? Lancet 364 (9452): 2157–2158. Ho, R. T., T. Wicks, D. Wyeth, and M. Nei. 2006. Generalized tonic–clonic seizures detected by implantable loop recorder devices: Diagnosing more than cardiac arrhythmias. Heart Rhythm 3 (7): 857–861. Hoffman, J. I., and G. D. Buckberg. 1975. Pathophysiology of subendocardial ischaemia. Br Med J 1€(5949): 76–79. Jallon, P. 1997. Epilepsy and the heart. Rev Neurol (Paris) 153 (3): 173–184. Johnston, S. C., J. K. Horn, J. Valente, and R. P. Simon. 1995. The role of hypoventilation in a sheep model of epileptic sudden death. Ann Neurol 37 (4): 531–537. Katz, L. N., and H. Feinberg. 1958. The relation of cardiac effort to myocardial oxygen consumption and coronary flow. Circ Res 6 (5): 656–669. Khouri, E. M., D. E. Gregg, and C. R. Rayford. 1965. Effect of exercise on cardiac output, left coronary flow and myocardial metabolism in the unanesthetized dog. Circ Res 17 (5): 427–437. LaFrance, W. C., Jr., A. M. Kanner, and B. Hermann. 2008. Psychiatric comorbidities in epilepsy. Int Rev Neurobiol 83: 347–383. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008. The mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12 (1): 3–24. Lathers, C. M., P. L. Schraeder, and S. B. Carnel. 1984. Neural mechanisms in cardiac arrhythmias associated with epileptogenic activity: The effect of phenobarbital in the cat. Life Sci 34 (20): 1919–1936.

Sudden Unexpected Death in Epilepsy

127

Lathers, C. M., and P. L. Schraeder. 2010. SUDEP animal models. Ch. 28 in Sudden Death in Epilepsy: Forensic and Clinical Issues. Boca Raton: CRC Press. Leung, H., P. Kwan, and C. E. Elger. 2006. Finding the missing link between ictal bradyarrhythmia, ictal asystole, and sudden unexpected death in epilepsy. Epilepsy Behav 9 (1): 19–30. Leutmezer, F., C. Schernthaner, S. Lurger, K. Potzelberger, and C. Baumgartner. 2003. ElectroÂ� cardiographic changes at the onset of epileptic seizures. Epilepsia 44 (3): 348–354. Mameli, O., M. A. Caria, A. Pintus, G. Padua, and S. Mameli. 2006. Sudden death in epilepsy: an experimental animal model. Seizure 15 (5): 275–287. Manno, E. M., E. A. Pfeifer, G. D. Cascino, K. H. Noe, and E. F. Wijdicks. 2005. Cardiac pathology in status epilepticus. Ann Neurol 58 (6): 954–957. Marshall, D. W., B. F. Westmoreland, and F. W. Sharbrough. 1983. Ictal tachycardia during temporal lobe seizures. Mayo Clin Proc 58 (7): 443–446. Mascia, A., P. P. Quarato, A. Sparano, V. Esposito, F. Sebastiano, G. Occhiogrosso, and G. Di Gennaro. 2005. Cardiac asystole during right frontal lobe seizures: A case report. Neurol Sci 26 (5): 340–343. Mondon, K., B. Charbonnier, C. Hommet, P. Corcia, A. Autret, and B. de Toffol. 2002. Ictal bradycardia followed by cardiac asystole: A case report. Epileptic Disord 4 (4): 261–264. Murray, P. A., and S. F. Vatner. 1981. Carotid sinus baroreceptor control of right coronary circulation in normal, hypertrophied, and failing right ventricles of conscious dogs. Circ Res 49 (6): 1339–1349. Nashef, L., F. Walker, P. Allen, J. W. Sander, S. D. Shorvon, and D. R. Fish. 1996. Apnoea and bradycardia during epileptic seizures: Relation to sudden death in epilepsy. J Neurol Neurosurg Psychiatry 60 (3): 297–300. Natelson, B. H., R. V. Suarez, C. F. Terrence, and R. Turizo. 1998. Patients with epilepsy who die suddenly have cardiac disease. Arch Neurol 55 (6): 857–860. Nei, M., R. T. Ho, M. R. Sperling. 2000. EKG abnormalities during partial seizures in refractory epilepsy. Epilepsia May;41(5): 542–548. Nei, M., R. T. Ho, B. W. Abou-Khalil, F. W. Drislane, J. Liporace, A. Romeo, and M. R. Sperling. 2004. EEG and ECG in sudden unexplained death in epilepsy. Epilepsia 45 (4): 338–345. O’Brien, T. J., E. L. So, B. P. Mullan, M. F. Hauser, B. H. Brinkmann, N. I. Bohnen, D. Hanson, G. D. Cascino, C. R. Jack Jr., and F. W. Sharbrough. 1998. Subtraction ictal SPECT co-registered to MRI improves clinical usefulness of SPECT in localizing the surgical seizure focus. Neurology 50 (2): 445–454. Opherk, C., J. Coromilas, and L. J. Hirsch. 2002. Heart rate and EKG changes in 102 seizures: Analysis of influencing factors. Epilepsy Res 52 (2): 117–127. P-Codrea Tigaran, S., S. Dalager-Pedersen, U. Baandrup, M. Dam, and A. Vesterby-Charles. 2005. Sudden unexpected death in epilepsy: Is death by seizures a cardiac disease? Am J Forensic Med Pathol 26 (2): 99–105. Pedley, T. A., and W. A. Hauser. 2002. Sudden death in epilepsy: A wake-up call for management. Lancet 359 (9320): 1790–1791. Penninx, B. W., A. T. Beekman, A. Honig, D. J. Deeg, R. A. Schoevers, J. T. van Eijk, and W. van Tilburg. 2001. Depression and cardiac mortality: Results from a community-based longitudinal study. Arch Gen Psychiatry 58 (3): 221–227. Rocamora, R., M. Kurthen, L. Lickfett, J. Von Oertzen, and C. E. Elger. 2003. Cardiac asystole in epilepsy: Clinical and neurophysiologic features. Epilepsia 44 (2): 179–185. Rugg-Gunn, F. J., R. J. Simister, M. Squirrell, D. R. Holdright, and J. S. Duncan. 2004a. Cardiac arrhythmias in focal epilepsy: A prospective long-term study. Lancet 364 (9452): 2212–2219. Rugg-Gunn, F. J., R. J. Simister, M. Squirrell, D. R. Holdright, and J. S. Duncan. 2004b. Cardiac arrhythmias in focal epilepsy: A prospective long-term study. Lancet 364 (9452): 2212–2219. Sanya, E. O. 2005. Increasing awareness about sudden unexplained death in epilepsy—A review. Afr J Med Med Sci 34 (4): 323–327.

128 Sudden Death in Epilepsy: Forensic and Clinical Issues Schernthaner, C., G. Lindinger, K. Potzelberger, K. Zeiler, and C. Baumgartner. 1999. Autonomic epilepsy—The influence of epileptic discharges on heart rate and rhythm. Wien Klin Wochenschr 111 (10): 392–401. Schuele, S. U., A. C. Bermeo, E. Locatelli, R. C. Burgess, and H. O. Luders. 2008. Ictal asystole: A benign condition? Epilepsia 49 (1): 168–171. Schulz, R., M. Grebe, H. J. Eisele, K. Mayer, N. Weissmann, and W. Seeger. 2006. Obstructive sleep apnea-related cardiovascular disease. Med Klin (Munich) 101 (4): 321–327. Scorza, F. A., D. B. Colugnati, A. P. Pansani, E. Y. Sonoda, R. M. Arida, and E. A. Cavalheiro. 2008. Preventing tomorrow’s sudden cardiac death in epilepsy today: what should physicians know about this? Clinics (Sao Paulo) 63 (3): 389–94 LID-S1807-593220. Shakespeare, W. Sonnet 45. Available from http://www.shakespeare-online.com/sonnets/45.html. Shamim, S., G. Hasler, C. Liew, S. Sato, and W. H. Theodore. 2009. Temporal lobe epilepsy, depression, and hippocampal volume. Epilepsia 50 (5): 1067–1071. Sheline, Y. I., P. W. Wang, M. H. Gado, J. G. Csernansky, and M. W. Vannier. 1996. Hippocampal atrophy in recurrent major depression. Proc Natl Acad Sci U S A 93 (9): 3908–3913. Shibasaki, H., A. Ikeda, and T. Nagamine. 2007. Use of magnetoencephalography in the presurgical evaluation of epilepsy patients. Clin Neurophysiol 118 (7): 1438–1448. Slomka, P. J., D. S. Berman, and G. Germano. 2008. Applications and software techniques for integrated cardiac multimodality imaging. Expert Rev Cardiovasc Ther 6 (1): 27–41. So, E. L., M. C. Sam, and T. L. Lagerlund. 2000. Postictal central apnea as a cause of SUDEP: evidence from near-SUDEP incident. Epilepsia 41 (11): 1494–1497. So, N. K., and M. R. Sperling. 2007. Ictal asystole and SUDEP. Neurology 69 (5): 423–424. Spratling, W. P. 1902. The cause and manner of death in epilepsy. Med News 80: 1225–1227. Strzelczyk, A., S. Bauer, S. Knake, W. H. Oertel, H. M. Hamer, and F. Rosenow. 2008. Ictal asystole in temporal lobe epilepsy before and after pacemaker implantation. Epileptic Disord 10 (1): 39–44. Ten Tusscher, K. H., and A. V. Panἀlov. 2007. Influence of diffuse ἀbrosis on wave propagation in human ventricular tissue. Europace 9 (Suppl 6): vi38–vi45. Tigaran, S. 2001. Regional myocardial blood flow during seizures. Paper read at the American Academy of Neurology Annual Meeting in Philadelphia, on May 9, 2001 during session no. 32. Tigaran, S. 2002. Cardiac abnormalities in patients with refractory epilepsy. Acta Neurol Scand Suppl 177: 9–32. Tigaran, S., F. Andreasen, and M. Dam. 2001. Changes in plasma catecholamines after epileptic seizures. Possible explanation of SUDEP. Epilepsia 42 (S-2): 158. Tigaran, S., H. Molgaard, and M. Dam. 1999. Detection of myocardial ἀbrosis by ECG technique in epilepsy patients—Further evidence for cardiac pathogenesis of SUDEP. Epilepsia 40 (S-7): 105. Tigaran, S., H. Molgaard, and M. Dam. 2002. Atrio-ventricular block: A possible explanation of sudden unexpected death in epilepsy. Acta Neurol Scand 106 (4): 229–233. Tigaran, S., H. Molgaard, R. McClelland, M. Dam, and A. S. Jaffe. 2003. Evidence of cardiac ischemia during seizures in drug refractory epilepsy patients. Neurology 60 (3): 492–495. Tinuper, P., F. Bisulli, A. Cerullo, R. Carcangiu, C. Marini, G. Pierangeli, and P. Cortelli. 2001. Ictal bradycardia in partial epileptic seizures: Autonomic investigation in three cases and literature review. Brain 124 (Pt 12): 2361–2371. Toth, V., L. Hejjel, Z. Kalmar, A. Fogarasi, T. Auer, C. Gyimesi, A. Szucs, and J. Janszky. 2008. Effect of epileptic seizures on the heart rate. Ideggyogy Sz 61 (5–6): 155–161. Ushijima, A., N. Fukuma, Y. Kato, N. Aisu, and K. Mizuno. 2009. Sympathetic excitation during exercise as a cause of attenuated heart rate recovery in patients with myocardial infarction. J€Nippon€Med Sch 76 (2): 76–83. Velez, L., and L. M. Selwa. 2003. Seizure disorders in the elderly. Am Fam Physician 67 (2): 325–332. Venugopalan, P., P. M. Nair, and R. L. Koul. 2001. An infant with seizure-related bradycardia and asystole. J Paediatr Child Health 37 (1): 96–97.

Sudden Unexpected Death in Epilepsy

129

Walczak, T. S., I. E. Leppik, M. D’Amelio, J. Rarick, E. So, P. Ahman, K. Ruggles, G. D. Cascino, J. F. Annegers, and W. A. Hauser. 2001. Incidence and risk factors in sudden unexpected death in epilepsy: A prospective cohort study. Neurology 56 (4): 519–525. WHO. 2009. Epilepsy. 2009/01/01. Available from http://www.who.int/mediacentre/factsheets/ fs999/en/index.html (accessed February 2009). Wilder-Smith, E., and S. H. Lim. 2001. Heart rate changes during partial seizures: A study amongst Singaporean patients. BMC Neurol 1: 5. Willams, B., and A. Richens. 1998. Epilepsy and sudden death. UK winter 1998 Epilepsy Toronto: EA updates. London: Epilepsy Research Foundation. WrongDiagnosis.com. 2009. Deaths from obesity. Available from http://www.wrongdiagnosis╉ .com/o/obesity/deaths.htm. Yasue, H. 1983. Role of autonomic nervous system in the pathogenesis of angina pectoris. Arch Mal Coeur Vaiss 76 Spec No: 3–6. Zijlmans, M., D. Flanagan, and J. Gotman. 2002. Heart rate changes and ECG abnormalities during epileptic seizures: Prevalence and deἀnition of an objective clinical sign. Epilepsia 43 (8): 847–854.

Forensic Postmortem Examination of Victims of Sudden Unexpected Death in Epilepsy

8

Claire M. Lathers Paul L. Schraeder Steven A. Koehler Cyril H. Wecht

Contents 8.1 Introduction 8.2 Hypothesis and Future Retrospective Studies 8.3 Speciἀc Aims 8.3.1 Year One 8.3.2 Year Two 8.3.3 Year Three 8.4 Questions for Retroprospective Postmortem Studies References

131 135 135 136 136 136 137 139

8.1â•…Introduction Evaluation of all potential cardiac risks for sudden death needs to start with detailed cardiac postmortem examinations. Detailed postmortem examinations of hearts obtained from patients with epilepsy who experienced sudden and unexpected death (SUDEP) will be helpful in identifying associated cardiac risk factors in persons with epilepsy, and also family members that could be predisposed to sudden death. A proἀle of representative pathology associated with sudden death is found in Table 8.1. Research work in the areas of SUDEP suggests that changes in autonomic peripheral and central mechanisms involved in the control and regulation of blood pressure, heart rate, and rhythm may contribute to sudden death. More speciἀcally, changes in postganglionic cardiac sympathetic discharges, which result in nonuniform neural discharges associated with epileptiform discharges, may be part of the mechanism of risk. Another major risk factor seems to be noncompliance with antiepileptic drug use. A national survey of coroners and medical examiners throughout the United States was conducted to determine their depth of understanding of causes of death in persons with epilepsy and, in particular, the phenomenon of sudden death (Schraeder et al. 2006, 2009). The survey concluded that, while there is a valid classiἀcation of SUDEP, it is not routinely used in autopsy reports of patients with known seizures who die suddenly and unexpectedly with no cause of death found on postmortem examination. Instead, the autopsy reports emphasized the negative ἀndings, i.e., that the death might have been 131

132 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 8.1â•… Pathological Findings in Victims of Sudden Death Victims of Sudden Death Victims of stress-related death (Selye 1958) 51 SUDEP cases non-East Asian subjects including indigenous Saudis, 1/1995 to 6/1995; most subcontinent Indians (43%) (Elfawal 2000) Autopsies of sudden death cases in Japan, 5/1994 to 2/1998 Of 271 cases, 176 patients (20 to 59 years old) were sudden death (Owada et al. 1999) 200 cases of sudden death in persons younger than 35 years in Veneto region, Italy – Unexplained deaths occurred in 12 cases (6%) – More than 1 in 20 cases in sudden death not explained by structural risk factors (Basso et al. 1999)

Cardiac Pathology Microscopic myoἀbrillar degeneration or myocytolysis Identical in hearts of patients dying subarachnoid hemorÂ� rhage and other acute strokes Autopsies on 22 victims 7 mild to moderate cardiac hypertrophy and 2 mild to moderate coronary stenosis 4 similar degrees of coronary narrowing but no myocardial hypertorphy Of the sudden death cases, 29 (31.9%) were due to coronary artery disease 18 (19.8%) acute cardiac dysfunction 6 (6.6%) other cardiac diseases 4 (4.4%) acute aortic dissection 163 cases (81.5%) due to cardiovascular etiologies

Obstructive coronary atherosclerosis, 23%; arrhythmogenic right ventricular cardiomyopathy, 12.5%; mitral valve prolapse 10%; conduction system abnormalities, 10%; congenital coronary artery anomalies, 8.5%; myo�carditis, 7.5%; hyper�trophic cardiomyopathy 5.5%; aortic rupture, 5.5%; dilated cardiomyopathy, 5%; nonatherosclerotic-acquired coronary artery disease, 3.5%; postoperative congenial heart disease, 13%; aortic stenosis, 2%

Pulmonary Pathology

Cerebral Pathology

Other

18 of 22 with severe pulmonary congestion and alveolar hemorrhage

10 cases (5%) due to respiratory causes

4 (4.4%) due to cardiovascular disease

30 (32.9%) due to other diseases

15 cases (7.5%) due to cerebral causes

Other causes (2%)

Pulmonary embolism, 2%

caused by a seizure disorder when there was no pathological ἀnding. Two important questions remain: 1. How do we determine if the death of a person is a sudden unexpected death? 2. How do we establish a true prevalence of the phenomenon of SUDEP?

Forensic Postmortem Examination of Victims of SUDEP

133

In addition to the need for establishing the true prevalence of the phenomenon of SUDEP is the importance of looking for evidence of the roles of the brain and the heart in the cause of death when conducting an autopsy on a person with epilepsy who has died suddenly. SUDEP refers to sudden death of an individual with a clinical history of epilepsy, in whom a postmortem examination fails to uncover a gross anatomic, toxicological, or environmental cause of death. Evidence of terminal seizure activity may not be present. In 2002, Shields et al. reported that 1–2% of natural deaths certified by the medical–legal death investigators in the United States are attributed to epilepsy and that increased microscopic examination of the brain postmortem has allowed identification of structural changes representative of epileptogenic foci. They examined 70 death cases, all with known clinical history of seizures, and classified them as: 1. Individuals who lacked a gross brain lesion 2. Those with a brain lesion demonstrable at autopsy 3. Those who lacked neuropathological evaluation because of decomposition so that only an external examination was done. Microscopic ἀndings include neuronal clusters, increased perivascular oligodendrogÂ� lia, gliosis, cystic gliotic lesions, decreased myelin, cerebellar Bergmann’s gliosis, and folial atrophy were found to be present in a higher percentage of the brains of SUDEP victims, when compared to brains from age- and sex-matched control subjects. Additional autopsy results are needed to clarify the role of changes in the heart in SUDEP. Insight into the mechanism of death in persons with epilepsy who die unexpectedly is not much greater than what it was when the ἀrst book (Lathers and Schraeder 1990b) addressed the topic of epilepsy and sudden death in 1990. This chapter emphasizes the need for clinical studies to focus on the contribution of cardiac autonomic dysfunction and/or antiepileptic drug use, with or without other drugs, to the development of cardiac abnormalities to gain an understanding of the mechanism of death in these persons. While most investigators agree that disturbance in the function of the autonomic nervous system may be a contributory cause (Han and Moe 1964; Randall et al. 1968; Lathers 1975; Lathers 1980a, 1980b; Lathers and Roberts 1985; Lathers and Spivey 1987; Spivey and Lathers 1985; Lathers et al. 1977a, 1977b, 1978; Lathers and Schraeder 1982; Lathers and Roberts 1985; Lathers et al. 1986; Lipka et al. 1988), how disruption of autonomic function contributes to the risk of death is not known. There are some clues from clinical observations. For example, the occurrence of seizures is often associated with measurable changes in cardiac conduction and rhythm. Two recent papers are relevant. The ἀrst, by Chin et al. (2004), reports myocardial infarction following brief convulsive seizures. The second paper, also by Chin et al. (2005), describes the occurrence of “postictal neurogenic stunned myocardium,” resulting from the consequence of seizures. A reversible multifocal ventricular dysfunction, developed in a nonvascular pattern, is thought to be the result of a high sympathetic tone. The initiating factor(s) for a fatal outcome in some at-risk individuals has not been deἀned. Experimental evidence indicates that cortical epileptiform activity can affect cardiac autonomic regulatory function (Lathers and Schraeder 1982; Lathers et al. 1987; Lathers and Schraeder 1990a, 1995, 2002; Lathers et al. 2003a; Lathers and Schraeder 1983, 1989, 1995, 2006, 2009; Lathers et al. 1984, 1988,

134 Sudden Death in Epilepsy: Forensic and Clinical Issues

1989, 1993, 2008a; Scorza et al. 2008; Carnel et al. 1985; Kraras et al. 1987; Stauffer et al. 1989, 1990; Suter and Lathers 1984; Tumer et al. 1985) and the integrity of pulmonary vasculature function (Simon et al. 1982; Johnston et al. 1995, 1997). In young adults with epilepsy, there is an increased risk for sudden death, often, but not always, as a function of the severity of their seizure disorder. Why there is an increased risk of presumed cardiac arrhythmia in this group is not known. Many investigators assume that death in these patients is a function of neurogenically induced cardiac arrhythmias since, in most cases, no gross pathological explanations are found on postmortem examination. However, there is a small body of literature that suggests that microscopic changes in the subendocardial region of the heart may be a contributory factor to increased risk of death in this population. Natelson et al. (1998) found irreversible pathological changes in the form of subendocardial perivascular and interstitial ἀbrosis in four of seven hearts from persons with epilepsy who died suddenly and unexpectedly. These ἀndings support the premise that patients with epilepsy who die suddenly and unexpectedly have subtle microscopic cardiac pathological conditions that may be responsible for increased risk of neurogenically induced arrhythmias. A series of questions must be addressed to focus on the issue of whether drugs are a beneἀt or a risk for SUDEP (Lathers 2002, 2003; Lathers and Schraeder 1995, 2002; Lathers et al. 2003, 2008a, 2008b; Scorza et al. 2008; Leestma et al. 1997; Schraeder and Lathers 1995). The ἀrst question is: “Do cardiac changes induced by antiepileptic drugs contribute to the risk of sudden death in persons with epilepsy?” A second question is: “Are particular combinations of antiepileptic drugs more likely to be associated with an increased risk of sudden death?” A third question to be addressed is: “Does a combination of antiepileptic drugs with other drugs, i.e., psychotropics, increase the risk of dying in a sudden, unexpected manner?” The FDA recognizes the importance of addressing this series of questions related to the use of pharmacological agents in that most of the new antiepileptic drugs have to be evaluated for the risk of SUDEP. As such, the FDA requires a statement in the package insert addressing the relative risk of occurrence of SUDEP with use of the drug. It is important for postmortem protocols to address the details of types of drugs and dosage schedules and possible adverse interactions as contributing to SUDEP. Speculation as to risk factors for SUDEP and/or contributory causes of SUDEP include the following:

1. Autonomic neurohumoral dysfunction 2. Autonomic cardiac neural dysfunction 3. Cardiac changes of subendocardial perivascular and interstitial ἀbrosis 4. Other cardiac changes (e.g., coronary heart disease, cardiomyopathy, aortic valvular stenosis, right ventricular dysplasia, postictal neurogenic stunned myocardium, coronary artery thrombosis, and postmyocardial infarction ἀbrosis 5. Cardiac changes possibly induced by antiepileptic drugs 6. The likelihood of cardiac changes being induced by particular combinations of antiepileptic drugs and associated with an increased risk of SUDEP 7. Whether a combination of antiepileptic drugs and nonantiepileptic drugs are more likely to increase the risk of dying in SUDEP (nonantiepileptic drugs are deἀned as therapeutic drugs and do include drugs of abuse)

Forensic Postmortem Examination of Victims of SUDEP

135

8. Whether an inherited predisposition to increased risk of cardiac arrhythmias (e.g., long QT and Brugada syndromes) could also be an additional risk factor in persons with epilepsy

8.2â•…Hypothesis and Future Retrospective Studies Autonomic neurohumoral and autonomic cardiac neural dysfunction, in combination with susceptibility associated with cardiac pathological changes, are hypothesized to be risk factors for the development of circumstances that predispose a person with epilepsy to sudden death. The potential risk factors of antiepileptic drug use versus nonantiepileptic drug use also must be examined. Study design should evaluate and differentiate the contribution of intrinsic parameters of autonomic dysfunction and/or cardiac pathological changes from the extrinsic parameter of therapeutic pharmacological agents (antiepileptic drugs and/or nonantiepileptic drugs). Microscopic pathological cardiac markers manifest in the heart in association with autonomic catecholamine-mediated arrhythmias and cardiac damage and/or by the administration of drugs (antiepileptic drugs and nonantiepileptic drugs should be examined). It could be assumed that catecholamine toxicity at the local level in the myocardium is a contributory factor to the development of microἀbrotic changes thought to be a predisposing risk factor for cardiac arrhythmias. Another contributory factor to cardiac arrhythmias and/or sudden death is abnormal cardiac postganglionic discharge. It is theorized that the combination of nonuniform postganglionic cardiac discharge due to abnormal anatomy of the postganglionic nerves in combination with catecholamine toxicity at the local level in the myocardium, possibly in combination with a genetically determined cardiac predisposition, are risk factors that combine to contribute to arrhythmias culminating in a fatal sudden death event. It is unclear whether or not the use of certain antiepileptic drugs such as carbamazepine compound diminish the physiologic risk factors for SUDEP (Devinsky 2004). The role of genetics in epilepsy needs to be considered when evaluating risk factors for SUDEP and/or the role of sodium channel dysfunction. The reader is referred to the following references and chapters discussing the genetics of epilepsy in this book: Chioza et al. (2001, 2002a, 2002b, 2002c), Sisodiya et al. (2007), Hindocha et al. (2008), Helbig et al. (2009), Mullen and Scheffer (2009), Herreros (2010), Ghali and Nashef (2010), and Lathers et al. (2010 Chapter 1).

8.3â•…Specific Aims Study designs should examine the potential contributory role of the following parameters to the risk of SUDEP to determine if: 1. Consistent microscopic, cardiac changes of subendocardial, perivascular, and interstitial ἀbrosis are present and may have contributed to the SUDEP. 2. Changes in postganglionic cardiac sympathetic nerves may be present. 3. Cardiac changes are induced by speciἀc antiepileptic drugs that could contribute to the risk of SUDEP.

136 Sudden Death in Epilepsy: Forensic and Clinical Issues

4. Particular combinations of antiepileptic drugs are associated with an increased risk of SUDEP. 5. A combination of antiepileptic and nonantiepileptic drugs is likely to increase the risk of SUDEP. Nonantiepileptic drugs are deἀned as therapeutic agents and will not include drugs of abuse. 6. Controls for the retrospective and prospective arms of the studies will consist of matched persons without a history of epilepsy who have died. See the chapter by Ghali and Nashef (2010) for a discussion of what genetic issues should be considered. To determine the validity of the proposed hypothesis that autonomic neurohumoral and autonomic neural dysfunction, in combination with susceptibility to cardiac pathological changes and/or the use of antiepileptic or nonantiepileptic drugs, are risk factors for the development of circumstances that predispose a person with epilepsy to sudden death, the following timeline and methodology could be used by those designing retrospective postmortem studies. 8.3.1â•…Year One During the ἀrst year, study should be devoted to analyzing retrospectively the frozen samples of heart tissue obtained from victims of SUDEP in order to determine the optimal parameters of analysis of samples obtained from the prospective arm of the study. Collection of prospectively obtained samples of cardiac tissue and preparations for postmortem examination of cardiac postganglionic innervations, using the method of Druschky et al. (2001), in persons who have died suddenly and unexpectedly should be initiated during year one. Likewise, methods should be utilized to determine the presence or absence of genetic defects at the Na+ channel level (Mullen and Scheffer 2009; Herreros 2010; Lathers and Schraeder 2010). 8.3.2â•…Year Two During the second year, the retrospective analysis should be completed and the prospective acquisition of samples should be completed. 8.3.3â•…Year Three During the third year, analysis of the pathological materials should be completed. The following analyses should be done: • Microscopic histopathology of the heart, especially the conduction system and subendocardial tissue. • Transmission electron microscopy, scanning transmission electron microscopy, scanning electron microscopy, and laser scanning fluorescence confocal microscopy. • Use 123I-metaiodobenzylguanidine-SPECT to look for sympathetic changes in the postganglionic cardiac sympathetic innervation in patients with chronic temporal lobe epilepsy, since the study of Druschky et al. (2001) suggested that altered

Forensic Postmortem Examination of Victims of SUDEP

137

postganglionic cardiac sympathetic innervation may increase the risk of cardiac abnormalities and/or SUDEP.

8.4â•…Questions for Retroprospective Postmortem Studies

1. What was the gender of the victim? 2. What was the victim’s age? 3. Did the victim of sudden, unexpected death have a known seizure disorder? 4. Was the victim witnessed exhibiting seizure(s) at the time of death? If so, what type of seizure was observed? 5. What was the actual date of death? How does the date of death for this victim compare with the date of the last known seizure? 6. Was any other cause of death found upon autopsy for any of these patients, such as cardiomyopathy and glial tumor? Did the victim have a terminal illness or any known associated illnesses? 7. What was the extent of postmortems? Were microscopic examinations done of the heart, lungs, brain, and other organs? 8. What was the microscopic pathology of the lungs and the hearts? 9. How did you interpret ἀndings of “heavy lungs” and data from hearts obtained in the postmortem autopsy? What do these ἀndings mean in terms of mechanism and signiἀcance (see Terrence 1990)? 10. Did the autopsy conduct postmortem examination of postganglionic cardiac sympathetic innervation of the heart? 11. Were quantitative antiepileptic drug levels found in the toxicological screen conducted postmortem? What assay method was used? 12. What were the actual antiepileptic drug levels postmortem? Were they within the therapeutic range or were they subtherapeutic? Were any levels below the lower level of quantiἀcation of the assay method? 13. Did the postmortem examine tears and aqueous humor or just aqueous humor for antiepileptic drug levels? 14. Were hair samples obtained to examine antiepileptic drugs? 15. Was the patient considered to be a compliant patient prior to death? 16. Was any genetic testing for entities that predispose an individual to cardiac arrhythmias done on the victim and/or the victim’s family prior to death? 17. How can the diagnosis of a seizure-induced death be differentiated from a diagnosis of a SUDEP death? 18. Was there any prior history of cardiac rhythm irregularities? 19. Was there any family history of cardiac rhythm irregularities or any history of sudden deaths?

Sympathetic changes in the postganglionic cardiac sympathetic innervation in patients with chronic temporal lobe epilepsy suggest that altered postganglionic cardiac sympathetic innervation may increase the risk of cardiac abnormalities and/or SUDEP (Druschky et al. 2001). Developmental and regulatory mechanisms determining density and pattern of cardiac sympathetic innervation are still unclear. Likewise, the exact role of innervation

138 Sudden Death in Epilepsy: Forensic and Clinical Issues

in arrhythmogenesis is unclear. The clinical study of Druschky et al. (2001) conἀrms the animal studies conducted by Lathers and colleagues (Lathers 1975, 1980a, 1980b; Lathers and Roberts 1985; Lathers and Spivey 1987; Spivey and Lathers 1985; Lathers et al. 1977a, 1977b, 1978; Lathers et al. 1986; Lathers and Schraeder 1982, 1983, 1987, 1989, 1990a, 2002; Lathers et al. 1984, 1987, 2003a, 1989, 1993, 2008a; Scorza et al. 2008; Lathers et al 2008b) in which postganglionic cardiac sympathetic neural discharge was monitored before and as arrhythmias developed. When considering likely mechanisms of SUDEP, it is important to consider whether there is any underlying, undeἀned genetic predisposition to arrhythmias and, thus, the mechanism of sudden death. Obviously, genetic predisposition to arrhythmias varies from patient to patient. The status of any subtle symptomatic disease present at the time of death will also vary from patient to patient and emphasize the importance of a genetic component to the autopsy. Animal data (Lathers et al. 1987; Staufferet al. 1989, 1990; O’Rourke and Lathers 1990; Dodd-O and Lathers 1990) demonstrated the lockstep phenomenon, i.e., postganglionic cardiac sympathetic discharge time locked to cortical epileptiform activity. The lockstep phenomenon may explain propagation of electrical impulses to autonomic nervous system regulatory centers, thus initiating arrhythmogenic potentials. During the lockstep phenomenon, cardiac postganglionic sympathetic and vagal discharges were synchronized with both ictal and interictal discharges and premature ventricular contractions, ST/T changes, and conduction blocks occurring concurrent with interictal spikes at a time when nonuniform cardiac sympathetic and vagal discharges were also observed. These experimental observations suggest that altered cardiac sympathetic innervation of hearts in patients who die suddenly and unexpectedly may contribute to nonuniform neural discharge, arrhythmias, and/or death. Detailed microscopic pathologic examination of cardiac autonomic nerves should be part of an autopsy study of SUDEP. A regulatory response, resulting from increased awareness of SUDEP, occurred in 1993 when the FDA focused the attention of practitioners and pharmaceutical manufacturers on the question of whether the use of anticonvulsant drugs contributes to or prevents sudden unexpected death in epileptic persons (Lathers 2002, 2003; Lathers and Schraeder 1995, 2002; Lathers et al. 2008a, 2008b; Scorza et al. 2008; Leestma et al. 1997; Schraeder and Lathers 1995). The FDA-convened panel of scientists considered the prevalence of sudden unexpected death in patients involved in studies associated with developing new anticonvulsant drugs and reviewed data on the risk of sudden unexpected death in patients taking lamotrigine. The risk of SUDEP was no different from that found in the young epilepsy population in general. Estimated SUDEP rates in patients receiving the new anticonvulsant drugs lamotrigine, gabapentin, topiramate, tigabine, and zonisamide were found to be similar to those in patients receiving standard anticonvulsant drugs, suggesting that SUDEP rates reflect population rates and not a speciἀc drug effect. The FDA requires warning labels on the risk of SUDEP in association with the use of each of the above-mentioned drugs. However, little data is available on the relative risk associated with the older, commonly prescribed antiepileptic drugs. The reader is referred to the chapters by Lathers and Schraeder (2010) and by Lathers, Schraeder, and Claycamp (Lathers et al. 2010) in this book for an in-depth discussion of the possible role of antiepileptic drugs in the cause and/or prevention of SUDEP. Widdes-Walsh and Devinsky (2007) discuss drug-resistant epilepsy, stating that it is a prevalent problem in spite of the fact that multiple ADE drug options are available. Mimics of drug-resistant epilepsy exist and cause diagnostic confusion. Fortunately, advances in epilepsy research and pharmacogenomics

Forensic Postmortem Examination of Victims of SUDEP

139

prove that new understandings of the mechanisms of drug resistance and tolerance allow rational antiepileptic drug strategies to prevent drug resistance. A different question regarding the clinical pharmacology of SUDEP has been raised by Tigaran and coauthors (Tigaran et al. 1997; Tigaran 2002; Tigaran et al. 2002; Tigaran et al. 2003). The question to be addressed is whether some persons with severe drug-resistant epilepsy, and without any indication of previous cardiac disease, may experience a beneἀcial effect of prophylactic treatment with cardioactive drugs to reduce the risk of sudden death (Tigaran et al. 1997). These investigators found ECG changes and ST-segment depression—many of which were closely related to the occurrence of epileptic seizures— that were suggestive of myocardial ischemia in the patients with severe drug-resistant epilepsy and no previous indication of cardiac disease. Of twelve patients with medically intractable epilepsy in studies with both ECG and EEG recordings, the ECG recording found in one person with chest pain minor yet morphologically conspicuous changes in the ECG, suggestive of myocardial infarction. This person with epilepsy died in cardiac arrest. In two publications in 2002, these authors reported atrio-ventricular block occurring as a life threatening cardiac arrhythmia complicating epileptic seizures in one person with medically intractable epilepsy (Tigaran et al. 2002) and cardiac abnormalities in patients with refractory epilepsy (Tigaran 2002). In another study, Tigaran et al. (2003b) examined 23 patients with drug-refractory epilepsy and found ST-segment depression in 40% of the patients and this was associated with a higher maximum heart rate during seizures, indicative of myocardial ischemia occurring in these individuals. These studies emphasize the importance of conducting both EEG and ECG in persons with severe drug-resistant epilepsy and the necessity of evaluating whether these patients would beneἀt from prophylactic treatment with cardioactive drugs to reduce the risk of sudden death. This chapter focuses on the need for retrospective forensic postmortem examination of not only the brains but also the hearts, myocardial autonomic innervations, and antiepileptic drug levels obtained from patients with epilepsy who die suddenly. A recent series of publications has addressed the multiple risk factors for sudden death in cardiac patients and in persons with epilepsy who have died suddenly and unexpectedly (Lathers and Schraeder 2010; Lathers et al. 2008a, 2008b; Scorza et al. 2008; (Herreros 2010). An evaluation of all potential risks for sudden death requires that risk assessments be done (Lathers 2002). These risk assessment efforts will aid all working in clinical and research ἀelds as they work to unravel the mystery of SUDEP. We recommend that retrospective postmortem examinations, including genetic determinations, be done using hearts and, of course, brains, obtained from patients who have died suddenly and unexpectedly. In addition, an effort should be made to obtain data on preexisting diatheses for cardiac arrhythmias in SUDEP victims and family members. The ἀndings from these rÂ�etrospective studies will allow prospective identiἀcation of persons in families of patients with epilepsy who have died suddenly. Finally, in the event that a physical autopsy was not conducted, the value of verbal autopsies must be determined as a relative substitute (Aspray 2005; Nashef et al. 1998; Lathers and Schraeder 2009; Nashef and Sahloldt 2010; Schraeder et al. 2010).

References Aspray, T. J. 2005. The use of verbal autopsy in attributing cause of death from epilepsy. Epilepsia 46 (Suppl 11): 15–17.

140 Sudden Death in Epilepsy: Forensic and Clinical Issues Carnel, S. B., P. L. Schraeder, and C. M. Lathers. 1985. Effect of phenobarbital pretreatment on cardiac neural discharge and pentylenetetrazol-induced epileptogenic activity in the cat. Pharmacology 30 (4): 225–240. Chin, P. S., K. R. Branch, and K. J. Becker. 2004. Myocardial infarction following brief convulsive seizures. Neurology 63 (12): 2453–2454. Chin, P. S., K. R. Branch, and K. J. Becker. 2005. Postictal neurogenic stunned myocardium. Neurology 64 (11): 1977–1978. Chioza, B., H. Wilkie, L. Nashef, J. Blower, D. McCormick, P. Sham, P. Asherson, and A. J. Makoff. 2001. Association between the alpha(1a) calcium channel gene CACNA1A and idiopathic generalized epilepsy. Neurology 56 (9): 1245–1246. Chioza, B., A. Osei-Lah, H. Wilkie, L. Nashef, D. McCormick, P. Asherson, and A. J. Makoff. 2002a. Suggestive evidence for association of two potassium channel genes with different idiopathic generalised epilepsy syndromes. Epilepsy Res 52 (2): 107–116. Chioza, B., A. Osei-Lah, L. Nashef, B. Suarez-Merino, H. Wilkie, P. Sham, J. Knight, P. Asherson, and A. J. Makoff. 2002b. Haplotype and linkage disequilibrium analysis to characterise a region in the calcium channel gene CACNA1A associated with idiopathic generalised epilepsy. Eur J Hum Genet 10 (12): 857–864. Chioza, B., L. Nashef, P. Asherson, and A. Makoff. 2002c. CACNA1A and P/Q-type calcium channels in epilepsy. Lancet 359 (9302): 258. Devinsky, O. 2004. Effects of seizures on autonomic and cardiovascular function. Epilepsy Curr 4 (2): 43–46. Dodd-O, J. H., and C. M. Lathers. 1990. A characterization of the lockstep phenomenon in pÂ�henobarbital-pretreated cats. In Epilepsy and Sudden Death, Chapter 13. New York, NY: Marcel Dekker. Druschky, A., M. J. Hilz, P. Hopp, G. Platsch, M. Radespiel-Troger, K. Druschky, T. Kuwert, H. Stefan, and B. Neundorfer. 2001. Interictal cardiac autonomic dysfunction in temporal lobe epilepsy demonstrated by [(123)I]metaiodobenzylguanidine-SPECT. Brain 124 (Pt 12): 2372–2382. Ghali, N., and L. Nashef. 2010. Genetics of sudden death in epilepsy. In Sudden Death in Epilepsy: Forensic and Clinical Issues, Chapter 18. ed. C. M. Lathers, P. Schraeder, M. Bungo, and J. Leestma. Boca Raton: CRC Press. Han, J., and G. K. Moe. 1964. Nonuniform recovery of excitability in ventricular muscle. Circ Res 14: 44–60. Helbig, I., H. C. Mefford, A. J. Sharp, M. Guipponi, M. Fichera, A. Franke, H. Muhle et al. 2009. 15q13.3 microdeletions increase risk of idiopathic generalized epilepsy. Nat Genet 41 (2): 160–162. Herreros, B. 2010. Cardiac channelopathies and sudden death. In Sudden Death in Epilepsy: Forensic and Clinical Issues, Chapter 19. ed. C. M. Lathers, P. Schraeder, M. Bungo, and J. Leestma. Boca Raton: CRC Press. Hindocha, N., L. Nashef, F. Elmslie, R. Birch, S. Zuberi, A. Al-Chalabi, L. Crotti, P. J. Schwartz, and A. Makoff. 2008. Two cases of sudden unexpected death in epilepsy in a GEFS+ family with an SCN1A mutation. Epilepsia 49 (2): 360–365. Johnston, S. C., J. K. Horn, J. Valente, and R. P. Simon. 1995. The role of hypoventilation in a sheep model of epileptic sudden death. Ann Neurol 37 (4): 531–537. Johnston, S. C., R. Siedenberg, J. K. Min, E. H. Jerome, and K. D. Laxer. 1997. Central apnea and acute cardiac ischemia in a sheep model of epileptic sudden death. Ann Neurol 42 (4): 588–594. Kraras, C. M., N. Tumer, and C. M. Lathers. 1987. The role of enkephalins in the production of epileptogenic activity and autonomic dysfunction: Origin of arrhythmia and sudden death in the epileptic patient? Med Hypotheses 23 (1): 19–31. Lathers, C. M. 1975. Effect of practolol and stalol on ouabrain-induced nonuniform adrenergic nerve activity. Pharmacologist 34: 745. Lathers, C. M. 1980a. Effect of timolol on autonomic neural discharge associated with ouabaininduced arrhythmia. Eur J Pharmacol 64 (2–3): 95–106.

Forensic Postmortem Examination of Victims of SUDEP

141

Lathers, C. M. 1980b. The effect of metoprolol on coronary occlusion-induced arrhythmia and autonomic neural discharge. Fed Proc 39: 771. Lathers, C. M. 2002. Risk assessment in regulatory policy making for human and veterinary public health. J Clin Pharmacol 42 (8): 846–866. Lathers, C. M. 2003. Challenges and opportunities in animal drug development: A regulatory perspective. Nat Rev Drug Discov 2 (11): 915–918. Lathers, C. M., and J. Roberts. 1985. Are the sympathetic neural effects of digoxin and quinidine involved in their action on cardiac rhythm? J Cardiovasc Pharmacol 7 (2): 350–360. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23 (6): 633–647. Lathers, C. M., and P. L. Schraeder. 1987. Review of autonomic dysfunction, cardiac arrhythmias, and epileptogenic activity. J Clin Pharmacol 27 (5): 346–356. Lathers, C. M., and P. L. Schraeder. 1990a. Arrhythmias associated with epileptogenic activity elicited by penicillin. In Epilepsy and Sudden Death., ed. C. M. Lathers and P. Schraeder, Chapter€9. New York, NY: Marcel Dekker. Lathers, C. M., and P. L. Schraeder, eds. 1990b. Epilepsy and Sudden Death. New York, NY: Marcel Dekker. Lathers, C. M., and P. L. Schraeder. 1995. Experience-based teaching of therapeutics and clinical pharmacology of antiepileptic drugs. Sudden unexplained death in epilepsy: Do antiepileptic drugs have a role? J Clin Pharmacol 35 (6): 573–586; quiz 586–587. Lathers, C. M., and P. L. Schraeder. 2002. Clinical pharmacology: Drugs as a beneἀt and/or risk in sudden unexpected death in epilepsy? J Clin Pharmacol 42 (2): 123–136. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9 (2): 236–242. Lathers, C. M., and P. L. Schraeder. 2009. Verbal autopsies. Epilepsy Behav 14: 573–576. Lathers, C. M., and P. L. Schraeder. 2010. Antiepileptic drugs beneἀt/risk clinical pharmacology: Possible role in cause and/or prevention of SUDEP. In Sudden Death in Epilepsy: Forensic and Clinical Issues, Chapter 47. ed. C. M. Lathers, S. P. L., M. W. Bungo, and J. Leestma. Boca Raton: CRC Press. Lathers, C. M., and W. H. Spivey. 1987. The effect of beta blockers on cardiac neural discharge associated with coronary occlusion in the cat. J Clin Pharmacol 27 (8): 582–592. Lathers, C. M., K. M. Keller, J. Roberts, and A. B. Beasley. 1977a. Role of the adrenergic nervous system in arrhythmia produced by acute coronary artery occlusion. In Pathophysiology and Therapeutics of Myocardial Ischemia, ed. L. A., G. J. Kelliher, and M. Rovetto, Chapter 5. New York, NY: Spectrum. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1977b. Correlation of ouabain-induced arrhythmia and nonuniformity in the histamine-evoked discharge of cardiac sympathetic nerves. J Pharmacol Exp Ther 203 (2): 467–479. Lathers, C. M., G. J. Kelliher, J. Roberts, and A. B. Beasley. 1978. Nonuniform cardiac sympathetic nerve discharge: Mechanism for coronary occlusion and digitalis-induced arrhythmia. Circulation 57 (6): 1058–1065. Lathers, C. M., P. L. Schraeder, and S. B. Carnel. 1984. Neural mechanisms in cardiac arrhythmias associated with epileptogenic activity: The effect of phenobarbital in the cat. Life Sci 34 (20): 1919–1936. Lathers, C. M., W. H. Spivey, L. E. Suter, J. P. Lerner, N. Tumer, and R. M. Levin. 1986. The effect of acute and chronic administration of timolol on cardiac sympathetic neural discharge, arrhythmia, and beta adrenergic receptor density associated with coronary occlusion in the cat. Life Sci 39 (22): 2121–2141. Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67 (3): 247–259.

142 Sudden Death in Epilepsy: Forensic and Clinical Issues Lathers, C. M., L. J. Lipka, and H. Klions. 1988. Digitalis glycosides: A discussion of the similarities and differences in actions and existing controversies. Rev Clin Basic Pharm 7 (1-4): 1–108. Lathers, C. M., A. Z. Stauffer, N. Tumer, C. M. Kraras, and B. D. Goldman. 1989. Anticonvulsant and antiarrhythmic actions of the beta blocking agent timolol. Epilepsy Res 4 (1): 42–54. Lathers, C. M., P. L. Schraeder, and N. Tumer. 1993. The effect of phenobarbital on autonomic function and epileptogenic activity induced by the hippocampal injection of penicillin in cats. J Clin Pharmacol 33 (9): 837–844. Lathers, C. M., P. L. Schraeder, and H. G. Claycamp. 2003a. Clinical pharmacology of topiramate versus lamotrigine versus phenobarbital: Comparison of efficacy and side effects using odds ratios. J Clin Pharmacol 43 (5): 491–503. Lathers, C. M., S. A. Koehler, C. H. Wecht, and P. L. Schraeder. 2003b. Forensic antiepileptic drug levels in 2001 autopsy cases of sudden, unexpected deaths in persons with epilepsy in Allegheny County Pennsylvania. Paper read at the Annual Meeting of American College of Clinical Pharmacology, September, Orlando, FL. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008a. The mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12 (1): 3–24. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008b. Sudden death: Neurocardiologic mystery. In Psychological Factors and Cardiovascular Disorders, ed. L. Sher, Chapter 13. Hauppauge, NY: Nova Science. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2010. Sodium channel dysfunction: Common pathophysiologic mechanism associated with sudden death ECG abnormalities in Brugada Syndrome and some types of epilepsy. Case histories. In Sudden Death in Epilepsy: Forensic and Clinical Issues, Chapter 20. ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton: CRC Press. Leestma, J. E., J. F. Annegers, M. J. Brodie, S. Brown, P. Schraeder, D. Siscovick, B. B. Wannamaker, P. S. Tennis, M. A. Cierpial, and N. L. Earl. 1997. Sudden unexplained death in epilepsy: Observations from a large clinical development program. Epilepsia 38 (1): 47–55. Lipka, L. J., C. M. Lathers, and J. Roberts. 1988. Does chlorpromazine produce cardiac arrhythmia via the central nervous system? J Clin Pharmacol 28 (11): 968–983. Mullen, S. A., and I. E. Scheffer. 2009. Translational research in epilepsy genetics: Sodium channels in man to interneuronopathy in mouse. Arch Neurol 66 (1): 21–26. Nashef, L., and L. Sahloldt. 2010. Bereavement and sudden unexpected death in epilepsy. In Sudden Death in Epilepsy: Forensic and Clinical Issues, Chapter 58. ed. C. M. Lathers, P.€Schraeder, M. W. Bungo, and J. Leestma. Boca Raton: CRC Press. Nashef, L., S. Garner, J. W. Sander, D. R. Fish, and S. D. Shorvon. 1998. Circumstances of death in sudden death in epilepsy: Interviews of bereaved relatives. J Neurol Neurosurg Psychiatry 64€(3): 349–352. Natelson, B. H., R. V. Suarez, C. F. Terrence, and R. Turizo. 1998. Patients with epilepsy who die suddenly have cardiac disease. Arch Neurol 55 (6): 857–860. O’Rourke, D. K., and C. M. Lathers. 1990. Interspike interval histogram characterization of synchronized cardiac sympathetic neural discharge and epileptogenic activity in the electrocorticogram of the cat. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. Schraeder, Chapter 15. New York, NY: Marcel Dekker. Randall, W. C., M. Szentivanyi, J. B. Pace, J. S. Wechsler, and M. P. Kaye. 1968. Patterns of sympathetic nerve projections onto the canine heart. Circ Res 22 (3): 315–323. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32 (12): 1371–1382. Schraeder, P. L., and C. M. Lathers. 1989. Paroxysmal autonomic dysfunction, epileptogenic activity and sudden death. Epilepsy Res 3 (1): 55–62. Schraeder, P. L., and C. M. Lathers. 1995. Clinical pharmacology of antiepileptic drug use: Clinical pearls about the perils of patty. J Clin Pharmacol 35 (12): 1120–1135.

Forensic Postmortem Examination of Victims of SUDEP

143

Schraeder, P. L., K. Delin, R. L. McClelland, and E. L. So. 2006. Coroner and medical examiner documentation of sudden unexplained deaths in epilepsy. Epilepsy Res 68 (2): 137–143. Schraeder, P. L., K. Delin, R. L. McClelland, and E. L. So. 2009. A nationwide survey of the extent of autopsy in sudden unexplained death in epilepsy. Am J Forensic Med Pathol 30 (2): 123–126. Schraeder, P. L., E. L. So, and C. M. Lathers. 2010. Forensic case identiἀcation. In Sudden Death in Epilepsy: Forensic and Clinical Issues, Chapter 6. ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton: CRC Press. Scorza, F. A., R. M. Arida, and E. A. Cavalheiro. 2008. Preventive measures for sudden cardiac death in epilepsy beyond therapies. Epilepsy Behav 13 (1): 263–264; author reply 265. Shields, L. B., D. M. Hunsaker, 3rd, J. C. Hunsaker, and J. C. Parker Jr. 2002. Sudden unexpected death in epilepsy: Neuropathologic ἀndings. Am J Forensic Med Pathol 23 (4): 307–314. Simon, R. P., L. L. Bayne, R. F. Tranbaugh, and F. R. Lewis. 1982. Elevated pulmonary lymph flow and protein content during status epilepticus in sheep. J Appl Physiol 52 (1): 91–95. Sisodiya, S., J. H. Cross, I. Blumcke, D. Chadwick, J. Craig, P. B. Crino, P. Debenham et al. 2007. Genetics of epilepsy: Epilepsy research foundation workshop report. Epileptic Disord 9 (2): 194–236. Spivey, W. H., and C. M. Lathers. 1985. Effect of timolol on the sympathetic nervous system in coronary occlusion in cats. Ann Emerg Med 14 (10): 939–944. Stauffer, A. Z., J. Dodd-O, and C. M. Lathers. 1989. The relationship of the lock-step phenomenon and precipitous changes in mean arterial blood pressure. Electroencephalogr Clin Neurophysiol 72 (4): 340–345. Stauffer, A. Z., J. Dodd-O, and C. M. Lathers. 1990. Relationship of the lockstep phenomenon and precipitous changes in blood pressure. In Epilepsy and Sudden Death, Chapter 14. New York, NY: Marcel Dekker. Suter, L. E., and C. M. Lathers. 1984. Modulation of presynaptic gamma aminobutyric acid release by prostaglandin E2: Explanation for epileptogenic activity and dysfunction in autonomic cardiac neural discharge leading to arrhythmias? Med Hypotheses 15 (1): 15–30. Terrence, C. F. 1990. Unexpected, unexplained death of epileptic persons: clinical correlation including pulmonary changes. In Epilepsy and Sudden Death, ed. C. Lather and P. Schraeder, Chapter€6. New York, NY: Marcel Dekker. Tigaran, S. 2002. Cardiac abnormalities in patients with refractory epilepsy. Acta Neurol Scand Suppl 177: 9–32. Tigaran, S., V. Rasmussen, M. Dam, S. Pedersen, H. Hogenhaven, and B. Friberg. 1997. ECG changes in epilepsy patients. Acta Neurol Scand 96 (2): 72–75. Tigaran, S., H. Molgaard, and M. Dam. 2002. Atrio-ventricular block: A possible explanation of sudden unexpected death in epilepsy. Acta Neurol Scand 106 (4): 229–233. Tigaran, S., A. Cascino Fuglsang-Frederiksen, and G. D. Cascino. 2003a. Temporal distribution of partial seizures during the sleep–wake cycle: possible signiἀcance for sudden unexpected death. Epilepsia 44: 123. Tigaran, S., H. Molgaard, R. McClelland, M. Dam, and A. S. Jaffe. 2003b. Evidence of cardiac ischeÂ� mia during seizures in drug-refractory epilepsy patients. Neurology 60 (3): 492–495. Tumer, N., P. L. Schraeder, and C. M. Lathers. 1985. The effect of phenobarbital upon autonomic function and epileptogenic activity induced by hippocampal injection of penicillin in cats. Epilepsia 26: 520. Widdess-Walsh, P., and O. Devinsky. 2007. Antiepileptic drug resistance and tolerance in epilepsy. Rev Neurol Dis 4 (4): 194–202.

One-Year Postmortem Forensic Analysis of Deaths in Persons with Epilepsy

9

Steven A. Koehler Paul L. Schraeder Claire M. Lathers Cyril H. Wecht

Contents 9.1 Introduction 9.2 Methods 9.3 Results 9.4 Discussion Acknowledgments References

145 148 148 155 158 158

9.1â•…Introduction Individuals with epilepsy can die from the progression of the primary underlying brain process, status epileptics, trauma, drowning precipitated by a seizure, or causes totally unrelated to the epilepsy (Earnest et al. 1992). The medical literature reports sudden, unexpected and unexplained deaths among individuals with epilepsy (Earnest et al. 1992). This disorder has been termed sudden unexpected death in epilepsy (SUDEP). SUDEP has been deἀned as a sudden, unexpected, witnessed or unwitnessed death, nontraumatic and nondrowning, in patients with epilepsy, with or without evidence of a seizure, excluding documented status epileptics, in whom postmortem examination does not reveal an anatomical or toxicological cause of death (Nashef and Shorvon 1997). SUDEP occurs most frequently among young individuals with a history of generalized tonic-clonic seizures. SUDEP has been reported in the literature to be responsible for 1.7–17% of all deaths among individuals with epilepsy (Ficker et al. 1998; Ficker 2000). However, these prior studies to ascertain the incidence of SUDEP suffered from selection bias and methodological limitations resulting in a tenfold difference between the various studies. Populationbased and forensic (medical examiner or coroner’s office) studies are few. Postmortem examinations of victims of SUDEP fail to establish the cause of death and toxicological analysis reveals subtherapeutic antiepileptic drug blood levels (Earnest et al. 1992). Some of the commonly reported features of SUDEP are highlighted in Table 9.1. Previous postmortem-based studies have reported the following changes. The brains in these series showed increased weight. The cerebral hemispheres were anemic and congested, and there were signs of hypoxia in the hippocampal regions. The hearts showed 145

146 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 9.1â•… Characteristics of SUDEP Characteristics History of epilepsy Age Sex Race Seizure event Death Location found dead Postmortem ἀndings Toxicological ἀndings

Features of SUDEP Victims 8–17% of deaths Average age is 28–35 years, rare in children Number of male victims is twice the number of female victims More frequent among blacks Witnesses to the seizure event are rare Occurred within minutes Bed Autopsies do not reveal a cause of death Heavier than normal weights of the heart, lungs, and liver Postmortem antiepileptic drug (AED) levels were either subtherapeutic or absent

ἀbrosis localized in the conductive system around the atrioventriclular bundle, edema of the conductive tissue, perivascular and interstitial ἀbrosis, and reversible myocyte vacuolization. The lungs showed increased weight, mild to moderate pulmonary edema, and alveolar hemorrhage (Leestma et al. 1989; Terrence et al. 1981; Jallon 1999; So 2008). The livers showed increased weight and venous congestion (Leestma et al. 1989). Leestma stated that the degree of passive congestion in both the liver and lungs, as well as the edema in the lungs, suggests some element of acute backward cardiac failure in SUDEP cases (Leestma 1990). The true incidence of SUDEP is not known and the estimates vary greatly. Possible reasons for this may be linked to the fact that SUDEP is not a diagnosis as such, since it is assigned when all other possible diagnoses have been eliminated, making it a default label. Another may be the general lack of understanding by medical examiners and coroner’s office personnel of what deἀnes a SUDEP death and when they should list it on the death certiἀcate. One method of investigating the protocol used within a medical examiner’s or coroner’s office regarding SUDEP deaths is to examine the death scene investigation, the autopsy and toxicology results, and the resulting death certiἀcate. In addition, by examining all the deaths at the medical examiner’s or coroner’s office and calculating the number of deaths that conform to the SUDEP deἀnition, an approximate incidence rate of SUDEP can be ascertained. All deaths that are sudden, unexpected, and unexplained by past medical history are investigated by the medical examiner or coroner’s offices. They typically conduct a detailed death scene investigation, a forensic autopsy, and a toxicological analysis of the body fluids. The level of investigation and the anatomical and toxicological data collected by the medical examiner or coroner’s office offer investigators studying SUDEP a wealth of information. The Allegheny County Coroner’s Office was selected for several reasons. First, it was the site of a previous examination of SUDEP death and, second, the forensic examination of these types of deaths involves a thorough death scene investigation, complete postmortem examination, and a comprehensive toxicological analysis of body fluids. The ἀrst study to examine SUDEP deaths at the Allegheny County Coroner’s Office was a retrospective study conducted in 1981 (Terrence et al. 1981). The study examined SUDEP cases that occurred between January 1, 1978 and December 31, 1979 and located a total of 8 cases: 4 male, 4 female, 4 white, and 4 black, ranging in age from 9 to 31 years. The pathological ἀndings at autopsy are shown in Table 9.2 (Terrence et al. 1981). While, the

One-Year Postmortem Forensic Analysis of Deaths in Persons with Epilepsy

147

Table 9.2â•… Pathological Findings Heart Weight Normal Weight Range (250–350 g)

Combined Lung Weight Normal Weight Range (650–1140 g)

1 2 3 4 5

275 400 360 370 200

1090 850 800 1250 755

6 7 8

374 300 110

1325 930 350

Patient Number

Neuropathological Findings None Cerebral edema None None Old contusion of L temporal pole, olfactory bulb, and orbitofrontal surfaces None None Cerebral edema

Source: Terrence, C. F., Rao, G. R., and Perper, J. A., Ann Neurol, 9 (5), 458–464, 1981.

normal heart weight by sex could not be calculated, using the overall range of 250–350€g as the normal heart weight, 50% of the cases had an enlarged heart (Cotran et al. 1994). No signiἀcant abnormalities in the brain were noted. Using the range of 650–1140 g as a normal weight range of the combined lung weight, 25% of the cases had heavy lungs. The type, number of prescription medications, and the postmortem blood levels are shown in Table 9.3 (Terrence et al. 1981). Phenytoin and phenobarbital was prescribed to all 8€patients,€primidone prescribed to 4, and carbamazepine, valproic acid, and mephenytoin prescribed to one patient. Toxicological analysis revealed subtherapeutic or absent blood levels of phenytoin in 7 cases and phenobarbital in 4 out of the 8 cases. Levels for carbaÂ� mazepine, valproic acid, and mephenytoin were also measured postmortem. The purpose of the current study was to present an overview of all deaths examined at the Allegheny County Coroner’s Office among individuals with epilepsy in AlÂ�Â� legheny County during 2001, to describe epidemiological, anatomical, pathological, and Table 9.3â•… Prescribed Medications and the Postmortem Drug Levels Patient Number 1 2 3 4 5 6 7 8

Postmortem Drug Levels Drug Prescribed Phenytoin, phenobarbital Phenytoin, phenobarbital, primidone Phenytoin, phenobarbital, primidone Phenytoin, phenobarbital, primidone Phenytoin, phenobarbital, primidone Phenytoin, phenobarbital Phenytoin, phenobarbital Phenytoin, phenobarbital, carbamazepine, valproic acid, mephenytoin

Phenytoin

Phenobarbital

Other

None detected None detected

None detected None detected

None detected None detected

None detected

Therapeutic

None detected

Subtherapeutic

Therapeutic

None detected

Therapeutic

Therapeutic

None detected

None detected Subtherapeutic None detected

None detected Subtherapeutic Therapeutic

None detected None detected None detected

Source: Terrence, C. F., Rao, G. R., and Perper, J. A., Ann Neurol, 9 (5), 458–464, 1981.

148 Sudden Death in Epilepsy: Forensic and Clinical Issues

toxicological features of each SUDEP case, and to determine if SUDEP is present on the death certiἀcates issued by the coroner’s office.

9.2â•…Methods The Allegheny County Coroner’s Office has jurisdiction to investigate all deaths within Allegheny County, which is located in western Pennsylvania and encompasses a population of ~1.2 million individuals. All deaths investigated by the Allegheny County Coroner’s Office were reviewed from January 1, 2001 to December 31, 2001. This office investigates more than 6000 cases and conducts more than 1200 autopsies annually. All cases were identiἀed by conducting a computer and hand search of all the case reports by the Chief Forensic Epidemiologist (SAK) for all deaths with the words “seizure disorders,” “epilepsy,” or “SUDEP” appearing in Part I or Part II of the death certiἀcate. The following epidemiological information was collected: age, sex, race, time and date last seen alive, and the time and date of death. Seizure-related information collected included past medical€ history, prescribed medications, and who witnessed the event. Pathological information collected included the weights of the internal organs obtained from the forensic autopsy report. The normal weight parameters of the internal organs are based on published data and, where possible, separated by age and sex. Toxicological analysis was conducted on the blood, bile, urine, and eye fluid recovered during the autopsy. The blood used for the toxicological analysis was collected from the heart during the autopsy. The number and level of drugs detected in the body fluids were obtained from the toxicological report. The data was entered and analyzed by Statistical Package for the Social Sciences® (11.0 Chicago, IL). The anatomical and toxicological data for all deaths with a diagnosis of epilepsy were summarized. Based upon the summarized data, the cases that meet the classiἀcation criteria for SUDEP were determined.

9.3â•…Results A total of 12 deaths were identiἀed in which seizure disorder was listed as either the immediate cause of death or contributed to the death during the study period. Epidemiological characteristics are shown in Table 9.4. Among the 12 cases, 58.3% were male, 41.7% were female, and all cases were white. All cases were between the ages of 38 and 54 years old with the mean age of the males being slightly higher than that of the females. Overall, more than 90% of the seizures were unwitnessed events. The only witnessed seizure was that of a 42-year-old female. She was in her residence when she had a sudden seizure and became unresponsive. She was pronounced dead at the emergency room one hour after the seizure. Among the males, four were found in their bathroom and three in bed. Among the females, all were found in their residence (one on the kitchen floor, one on the bed, one in a chair, and one outside on the rear deck). Deaths were most frequent in January among the males and in September for the females. According to information contained within the death investigation report, all 12 cases had a past medical history for either a diagnosis of seizures or epilepsy. All cases would be considered as having a seizure disorder. A complete postmortem examination was conducted on all but one case, due to an advanced level of decomposition.

One-Year Postmortem Forensic Analysis of Deaths in Persons with Epilepsy

149

Table 9.4â•… Epidemiological Characteristics: Age, Sex, Race, Seizure Events, Month of Event, Medical History, and Type of Postmortem Examination Epidemiological Characteristics Total number of cases (all were Caucasian) Age range (mean) Seizure event witnessed Month of seizure

Medical history of seizures disorder or epilepsy Manner of death Type of postmortem examination a

Males

Females

7 38–50 (x = 45.4) Yes: 0 No: 7 January: 3 February: 1 March: 1 August: 1 October: 1 Yes: 7

5 39–54 (x = 44.8) Yes: 1 No: 4 June: 1 July: 1 August: 1 September: 2

Natural: 6 Accident: 1 Complete: 6 External only: 1a

Natural: 5

Yes: 5

Complete: 5 External only: 0

Decomposed.

An examination of the information contained on Part I or Part II of the death certiἀcate by sex is shown in Tables 9.5 and 9.6. The immediate cause of death is the event that directly and immediately resulted in the death. The contributory cause of death consists of conditions that play a role in causing the death, but do not cause immediate death. The section criteria for this study included any death where seizure disorder appeared on the death certiἀcate. Seizure disorder was listed as the immediate cause of death in 83.3% of the cases (6 males and 4 females), cardiomyopathy was listed in one male case, and a glial brain tumor was listed in one female case (Tables 9.5 and 9.6). Among males, seizure disorder was the immediate cause of death among 6 of the 7 deaths. In Male Case #7, the seizure disorder was considered by the forensic pathologist conducting the examination to have played a signiἀcant role in contributing to the death. The manner of death was ruled natural in six cases (85.7%) and accidental in one (14.3%). Among the females, seizure disorder Table 9.5â•…Immediate and Contributory Causes of Death for Males Death Certiἀcate Part I Immediate cause of death

Part II Contributory cause of death

Case No. 1

2

3

Seizure Seizure Seizure disorder disorder disorder (clinical)

4 Seizure disorder

5 Seizure disorder

6 Seizure disorder

Chronic Liver Liver Chronic cirrhosis cirrhosis obstructive obstructive disease pulmonary pulmonary disease disease

7 Dilated cardiomyopathy with arteriosclerotic cardiovascular disease, acute pneumonia, emphysema Seizure disorder

150 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 9.6â•…Immediate and Contributory Causes of Death for Females Case No.

Death Certiἀcate Part I Immediate cause of death

1

2

Seizure disorder

Seizure disorder (clinical)

3

4 Seizure disorder with hypertensive and arteriosclerotic cardiovascular disease

Seizure disorder (clinical) with arteriosclerotic cardiovascular disease

5 Convulsive seizure developed from grade II glial tumor

Part II Contributory cause of death

was the immediate cause of death among four of the ἀve deaths. In one case, the immediate cause of death was a convulsive seizure that developed in association with a grade II glial tumor. The manner of death was ruled natural in all ἀve cases. Overall, the incidence of death in persons with a history of seizures in which no cause of death was found on postmortem was 0.83% (10 in 1200 autopsies). The 2001 Allegheny County Coroner’s data found 0.833% (10 of 1200) of the autopsy cases met SUDEP criteria. Among the 12 cases, 85.7% (six of all males) and 80% (four of all females) represented SUDEP deaths. Autopsies were performed on all ἀve females and six of the seven males. The weights of the internal organs were available in 92% of the cases. The weights of the hearts, thicknesses of the left ventricles, lungs, and brains, and brain pathologies by sex are shown Table 9.7â•…Observed and Normal Organ Weights of the Heart, Lungs, and Brain by Sex

Sex

Weight (lb)

Heart Weight (g) (Normal Weight)a

Left Ventricle Thickness (cm)

Combined Lung Weight (g)b

Brain Weight (g)c

Male

118

310 (281)

1.2

1440

1070

121 131 145 240

445 (292) 360 (302) 400 (317) 455 (406)

1.4 1.2 1.2 1.5

1095 1200 1510 955

1240 1275 1760 1570

268 99.5 108 160 226

575 (432) 300 (230) 345 (243) 445 (284) 395 (329)

1.6 1.5 1.2 1.7 1.2

2065 915 550 1055 975

1760 1215 1205 1450 920

300

400 (371)

1.4

970

1365

Female

a b c

Normal heart range (mean expected heart weight based on body weight and sex). Normal combined weights of lungs: males, 720–1140 g; females: 650–960 g. Normal brain weights: males, 1100–1700 g; females: 1050–1550 g.

Brain Pathology Craniotomy, patchy subarachnoid hemorrhage Normal Slight swelling Normal Congestion, periventricular white matter Normal Chronic mild swelling Normal Normal Small crowded gyri, w/o atrophy Congested, mild swelling

One-Year Postmortem Forensic Analysis of Deaths in Persons with Epilepsy

151

700 600 500 400

Actual

300

Normal

200 100 0 1

2

3

4

5

6

Figure 9.1╇ Normal and actual heart weight (g) among males.

in Table 9.7. The normal ranges of the internal organs are also shown in Table 9.7. The observed weight of the heart was above the expected mean weight based on body weight and sex in all cases. Figures 9.1 and 9.2 show the normal and actual heart weights among males and females, respectively. The normal range was based on the weight and sex of the subjects. The weight of the heart was above the expected weight, after adjusting for body weight and sex in all cases. Left-ventricle hypertrophy was seen in two cases. The combined weight of the right and left lungs showed that more than 66% of the male and 60% of the female lungs exceeded normal parameters. The normal range of combined lung weights was based on sex. The mean weight for all the female lungs was normal (449.2 g), while that of the males was elevated (688.75 g). Figures 9.3 and 9.4 show the weight of the normal and actual combined lung weights by sex. Among males, the combined lung weight exceeds the upper limits of normal in four cases. Among females, the combined lung weight exceeds the upper limits of normal in three cases and dropped below the lower limits in one case. Examinations of the brains of the males showed that 33% were above normal limits, 50% were within normal limits, and 16% were below. The normal range of brain weights was based on sex. Examinations of the brains of the females showed that 80% were within

500 450 400 350 300

Actual

250

Normal

200 150 100 50 0 1

2

3

4

Figure 9.2╇ Normal and actual heart weight (g) among females.

5

152 Sudden Death in Epilepsy: Forensic and Clinical Issues 2500 2000 Actual

1500

Lower Limit Upper Limit

1000 500 0 1

2

3

4

5

6

Figure 9.3╇ Actual and normal upper and lower ranges of the combined weight of lungs (g) among males.

normal limits and 16% were below. Figures 9.5 and 9.6 show the normal and actual brain weight by sex. In the male group, the past medical history, prescribed medications, drugs identiἀed by the toxicology screen, level of postmortem compounds, and pathological factors are shown in Table 9.8. All seven had a history of seizures or epilepsy disorder. In ἀve cases (71%), antiepileptic drugs were prescribed (clonazepam, phenytoin). The death scene investigation failed to ascertain a list of prescription medications in two deaths. Toxicological analysis revealed that AED medications were detected in 57% of the cases, with phenytoin the most frequently detected. All the detected AED were at subtherapeutic levels. In only one case was the toxicological analysis negative. The bodies in three of the cases were in varying stages of decomposition. In the female group, the past medical history, prescribed medications, drugs identiἀed by the toxicology screen, level of postmortem compounds, and pathological factors are shown in Table 9.9. All cases had a past medical history of seizures. In two cases (40%), medications for seizures were prescribed (Dilantin, Lamotrigine, Tegretol). The death 1200 1000 800

Actual Lower Limit

600

Upper Limit

400 200 0 1

2

3

4

5

Figure 9.4╇ Actual and normal upper and lower ranges of the combined weight of lungs (g) among females.

One-Year Postmortem Forensic Analysis of Deaths in Persons with Epilepsy

153

2000 1800 1600 Male

1400

Lower Upper

1200 1000 800

118

121

131

145

240

268

Figure 9.5╇ Actual and normal upper and lower ranges of brain weight (g) among males by body weight (lb).

scene investigation failed to ascertain a list of prescription medications in one death. Toxicological analysis revealed that AED medications were detected in 60% of the cases, with phenytoin the most frequently detected. Toxicological analysis showed that the AED medications were at therapeutic levels in three, subtherapeutic in one, and above the therapeutic level in one case. Even though all the female cases had a diagnosis of seizure disorder, no information was collected as to why three of them were not taking any antiepileptic medications. This omission is one of the pitfalls of a retrospective study. This category of information (i.e., why a person with a diagnosis of epilepsy is not on medication) should be a speciἀc issue addressed in future studies.

1600 1500 1400 1300

Female

1200

Lower

1100

Upper

1000 900 800

99.5

108

160

226

300

Figure 9.6╇ Actual and normal upper and lower ranges of brain weight (g) among females by body weight (lb).

154 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 9.8â•… Medical History, Prescribed Medication, Toxicology Screen, and Pathological Features among Males

Medical History

Prescribed Medications

Drugs Identiἀed (with Levels) in Toxicology Screen (AED Drugs in€Bold)

Seizures

Dilantin

Acetone (6 mg%) Alcohol (5 mg%)

Seizures

Prinvil Clonazepam K-Dur Oramorph Depo Medrol Provential inhaler

Doxylamine (0.014 mg%) Benzodiazepines: ╇ Nordiazepam (0.01€mg%) ╇ Chlordiazepoxide (0.013 mg%) ╇ Demoxepam (0.023 mg%) ╇ Oxazepam (0.010 mg%) ╇ Clonazepam (too low to quantify) Morphine (0.087 μg/mL) Dextromethorphanpositive Ibuprofen-positive Diphenhydraminepositive Phenytoin (8.31 μg/mL)

Seizures

Dilantin

DVT, seizures, alcoholism

Dilantin Keflex

None detected

Seizures, alcoholism

Celexa Neurontin Phenytoin Indomethacin Unknown

Ethanol (0.04%) Phenytoin (1.05 μg/mL)

HIV, seizures, chronic lung disease Bipolar depression Epilepsy, depression

Unknown

Citalopram-positive Desmethylcitaloprampositive Ethanol (0.02%) Phenytoin (2.45 μg/mL) Olanzapine (0.041 mg%) Sertraline (0.021 mg%) Valproic acid (TDX) (8.99 μg/mL)

Levels of Postmortem Compounds — —

Above therapeutic levels Therapeutic level Subtherapeutic level — Subtherapeutic level — Subtherapeutic level — — — Subtherapeutic level —

— Subtherapeutic level — — — Subtherapeutic level Above therapeutic level Therapeutic level Subtherapeutic level

Pathological Features Alive: 1/7/01 Found dead: 1/11/01 12:30 p.m. Body moderately decomposed Alive: 1/17/01 4:30€p.m. Dead: 1/18/01 12:30 p.m.

Alive: 1/18/01 7:30€a.m. Found dead: 1/18/01 4:50 p.m. Alive: 2/11/01 1:00€a.m. Found dead: 8:00€p.m. Decomposed

Putriἀcation

Alive: 10/16/01 11:00 p.m. Found dead: 10/16/01 12:00€p.m.

One-Year Postmortem Forensic Analysis of Deaths in Persons with Epilepsy

155

Table 9.9â•… Medical History, Prescribed Medication, Toxicology Screen, and Pathological Features among Females

Medical History

Prescribed Medications

Drugs Identiἀed in Toxicology Screen (AED Drugs in Bold)

Level of Postmortem Compounds Subtherapeutic level Therapeutic level Subtherapeutic level Above therapeutic level

Hysterectomy, seizures

Dilantin Lamotrigine

Phenytoin (5.36 μg/mL) Lamotrigine (0.79 mg%) Ibuprofen (0.81 mg%)

Seizures

Neurontin Tegretol Dilantin

Clonazepram (8.37 μg/ mL)

Amyotrophic lateral sclerosis, depression, seizures

Bupropion Ditropanxl Zanaflex Baclofen Rilutek Unknown

Bupropion (0.006 mg%) Threoaminobupropionpositive

Therapeutic level —

Trazodone (0.296 mg%) Venlfaxine (0.175 mg%) O-Desmethylvenlafaxine (0.025 mg%) Carbamazepine (4.56 μg/mL) Phenytoin (14.68 μg/ mL) Butalbital (0.246 mg%) Acetaminophen (14.05 mg%)

Therapeutic level Therapeutic level —

Diabetes, ╇ seizures, ╇ hypertension, ╇ asthma Hypertension, ╇ seizures, TMJ, ╇ osteoporosis, ╇ cervical fusion ╇ psychiatric history

Prempro Gybutynin Butslbitalapac-caff-tap

Therapeutic level Therapeutic level Subtherapeutic level Nearly toxic

Pathological Features Alive: 6/7/01 evening Found dead: 6/8/01 3:30 p.m. Alive: 7/6/01 12:20€a.m. Found dead: 7/6/01 12:21 a.m. Witnesses to seizure Alive: Unknown Found dead: 8/7/01 5:37 p.m. Moderate putriἀcation Alive: 9/28/01 11:30€p.m. Found dead: 9/29/01 9:31 a.m. Alive: 9/4/01 2:30€p.m. Found dead: 9/5/01 11:32 p.m.

9.4â•…Discussion The sudden and unexpected deaths among individuals with a history of epilepsy are cases that fall under the jurisdiction of the medical examiner or coroner’s office for a forensic investigation. These investigations involve a detailed review of the victims’ past medical history, a list of current medications with dosage levels, and a determination of the events immediately surrounding the time of death. The body, in most cases, also undergoes a complete forensic autopsy and toxicological analysis. All these factors can provide a fairly accurate representation of the events at the time of death, including any anatomical or disease processes that might have contributed to the death and the circulating blood levels of any medications or other compounds at the time of death. Therefore, those studying SUDEP via forensic data will be provided with not only the circumstances surrounding the death, but the weights and state of the internal organs, and a detailed toxicological analysis of the body fluids. The difficulty is locating these types of cases within the medical examiners’ or coroners’ ἀles. When medical examiner’s or coroner’s offices are confronted with a death where the autopsy and toxicological analyses fail to identify a speciἀc cause of death, and that individual has a well-documented history of seizures, the cause of death on the death certiἀcate would be listed as seizure disorder and not as SUDEP. A lack of any other

156 Sudden Death in Epilepsy: Forensic and Clinical Issues

explanation for the cause of death indicates that these victims should be considered more appropriately as having a deἀnite classiἀcation of SUDEP. However, some authors have highlighted concerns with the death certiἀcate generated from the medical examiner and coroner data. If the diagnosis of epilepsy and/or seizure disorder is not stated on death certiἀcates with accuracy, many SUDEP deaths will go undetected. Coyle et al. (1994) examined 40 cases of SUDEP identiἀed in the United Kingdom in 1992. Postmortem reports and witness statements were examined to look at the accuracy of the coroners’ diagnoses. In 70% of those cases, the type of seizure either was not known or was not referred to. The review also found inconsistent reports of the position of the victim’s body, examination of the organs, especially the brain, details of the medication history, and the toxicology examinations conducted. Epilepsy or seizure disorder as an attributed cause of death was used in less than half of the cases even though the victims had a history consistent with this diagnosis. All of these ἀndings raise issues of the quality of data included in postmortem reports. Within the United States, the majority of medical examiners’ and coroners’ offices traditionally list such deaths as seizure disorders rather than epilepsy. This practice of not using the diagnosis of SUDEP on the death certiἀcate is not due to a lack of understanding or acknowledgment of SUDEP as a valid classiἀcation of the cause of death. The lack of utilization of SUDEP as a ἀnal diagnosis in appropriate cases is in agreement with data obtained in a recent nationwide survey (Schraeder et al. 2006) that found most medical examiners and coroners did not use the diagnosis of SUDEP when entering the cause of death on the death certiἀcate in those seizure disorder cases where no cause of death was identiἀed on postmortem. This lack of use of SUDEP on the death certiἀcates results in an underreporting of SUDEP and overreporting of deaths from seizure disorder. This study highlights the increased need to educate medical examiners’ and coroners’ offices regarding when the category of SUDEP should be used. Forensic pathologists face their greatest challenge when attempting to determine causes of death that rely on exclusion criteria, such as in SUDEP. In these types of cases it is important for the examination to attempt to rule out other possible causes before classifying the death as SUDEP on the death certiἀcate. In a case that is likely to be SUDEP, the following should be part of the standard protocol. First, conduct a detailed death scene investigation with an emphasis on collecting prescription medicines at the residence. Second, obtain a detailed past medical history with a thorough review of the medical history. Third, conduct a meticulous forensic autopsy with a comprehensive examination of the heart and its conductive system. Finally, review the results of the toxicological analysis of the body fluids with special attention to the postmortem levels of AED medications. When comparing the results of the study conducted in 1981 (Terrence et al. 1981) with our data, there were some similarities and some differences. Enlarged hearts were reported in 50% of the victims in the earlier study; our study reported that all victims had an enlarged heart. However, a retrospective case control study in a medical examiner’s population by Davis and McGwin (2004) found no difference in mean heart mass between the two groups. With increasing age, they did ἀnd competing causes of death in patients with epilepsy. Their ἀndings support the concept that SUDEP occurs in young adults. Opeskin et al. (2000) conducted cardiac pathology exams in 10 victims of SUDEP and in 10 controls. They found no abnormalities in the conduction system nor in the other cardiac pathological parameters studied. In contrast, the ἀndings in our study showed above-normal heart weights. Other studies found cardiac pathological changes obtained

One-Year Postmortem Forensic Analysis of Deaths in Persons with Epilepsy

157

Table 9.10â•… Questions to Be Asked by Death Investigators Relating to Possible Seizure-â•›Related Deaths Questions 1 2 3 4 5 6 7 8 9 10

Was the deceased diagnosed with a seizure disorder? At what age was the deceased diagnosed with a seizure disorder? Was the deceased taking any antiepileptic drugs? Do you have a list of the deceased’s antiepileptic drugs? Was the deceased compliant with taking his/her medication? What was the date of the deceased’s last seizure? Were the deceased’s seizures well-controlled? Is there a family history of seizures in the deceased’s family? When was the last time the deceased took his/her antiepileptic drugs? Do you have the name of the deceased’s treating physician?

from SUDEP victims. George and Davis (1998) looked at the pathology of arrhythmogenic right ventriclular cardiomyopathy/dysplasia. Inflammatory inἀltrates (i.e., lymphocytes) were observed in 60% of cases while myocyte necrosis was found in only one case. In the current study, 63% of the deaths had above-normal lungs weights while the earlier data reported that only 25% had heavy lungs. Both studies showed that the levels of AED medications detected during toxicological analysis were typically at the subtherapeutic level, or absent entirely. In this study, 7 of the 12 deaths were determined to have occurred in patients who had been prescribed anticonvulsant medications. Based on the postmortem analysis for antiepileptic drugs, ἀve were at subtheraputic levels, one was within the therapeutic range, one was€above therapeutic levels, and ἀve were totally absent of antiepileptic drugs. Based on these results and supported by the Terrence et al. (1981) study, a high percentage were either due to individuals being noncompliant with their medications or the detection methods of the antiepileptic drugs; levels in postmortem samples were below the lower limit of quantiἀcation of the assay method. The issue of using postmortem blood to estimate the circulating blood level in the living due to the phenomena of redistribution is a concern in forensic€toxicology. In this study, all the analyses were conducted on blood collected from the chambers of the heart. This blood typically shows a higher concentration of the drug than does peripheral blood. Therefore, the recoded levels of the antiepileptic drugs were lower in the circulating blood, meaning that the individuals with subtherapeutic heart blood levels have an even lower level of the drug in their circulating blood. The current observation of increased lung weights has also been reported to be present in SUDEP by other investigators. Noncompliance of anticonvulsants has been suggested as a risk factor for neurogenic pulmonary edema and the ἀnding of pulmonary edema is thought to be associated with an adrenergic component in the death event (Lathers and Schraeder 2002; Tomson et al. 2005; Hughes 2009). For future studies, there is a need to have more detailed data collection, for example, about seizure types, the premorbid antiepileptic drug levels, the general level of compliance, and any stressful circumstances preceding death. In essence, the need for detailed information could be obtained with oral autopsy data (Lathers and Schraeder 2009). Finally, there is a need to emphasize the importance of prospective data collection. Those investigating a possible death related to a seizure-related medical history should obtain a detailed description of the events leading up to the death from the next of kin and obtain all medical records and a complete list of prescription

158 Sudden Death in Epilepsy: Forensic and Clinical Issues

medications. To ensure that the important information is ascertained, Table 9.10 lists the key questions.

Acknowledgments The authors thank Shaum Ladham, MD, Leon Rozin, MD, Abdulrezak Shakir, MD, and Joseph Dominick, RN, LFD, for technical support in the data collection process.

References Cotran, R. S., V. Kumar, S. L. Robbins, and F. J. Schoen. 1994. Robbins Pathological Basis of Disease, 5th ed. Philadelphia, PA: W. B. Saunders. Coyle, H. P., N. Baker-Brian, and S. W. Brown. 1994. Coroners’ autopsy reporting of sudden unexplained death in epilepsy (SUDEP) in the UK. Seizure 3 (4): 247–254. Davis, G. G., and G. McGwin Jr. 2004. Comparison of heart mass in seizure patients dying of sudden unexplained death in epilepsy to sudden death due to some other cause. Am J Forensic Med Pathol 25 (1): 23–28. Earnest, M. P., G. E. Thomas, R. A. Eden, and K. F. Hossack. 1992. The sudden unexplained death syndrome in epilepsy: Demographic, clinical, and postmortem features. Epilepsia 33 (2): 310–316. Ficker, D. M. 2000. Sudden unexplained death and injury in epilepsy. Epilepsia 41 (Suppl 2): S7–S12. Ficker, D. M., E. L. So, W. K. Shen, J. F. Annegers, P. C. O’Brien, G. D. Cascino, and P. G. Belau. 1998. Population-based study of the incidence of sudden unexplained death in epilepsy. Neurology 51 (5): 1270–1274. George, J. R., and G. G. Davis. 1998. Comparison of anti-epileptic drug levels in different cases of sudden death. J Forensic Sci 43 (3): 598–603. Hughes, J. R. 2009. A review of sudden unexpected death in epilepsy: Prediction of patients at risk. Epilepsy Behav 14 (2): 280–287. Jallon, P. 1999. Sudden death of epileptic patients. Presse Med 28 (11): 605–611. Lathers, C. M., and P. L. Schraeder. 2002. Clinical pharmacology: Drugs as a beneἀt and/or risk in sudden unexpected death in epilepsy? J Clin Pharmacol 42 (2): 123–136. Lathers, C. M., and P. L. Schraeder. 2009. Verbal autopsies. Epilepsy Behav 14: 573–576. Leestma, J. E. 1990. Sudden unexpected death associated with seizures: A pathological review. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Leestma, J. E., T. Walczak, J. R. Hughes, M. B. Kalelkar, and S. S. Teas. 1989. A prospective study on sudden unexpected death in epilepsy. Ann Neurol 26 (2): 195–203. Nashef, L., and S. D. Shorvon. 1997. Mortality in epilepsy. Epilepsia 38 (10):1059–1061. Opeskin, K., A. Thomas, and S. F. Berkovic. 2000. Does cardiac conduction pathology contribute to sudden unexpected death in epilepsy? Epilepsy Res 40 (1): 17–24. Schraeder, P. L., K. Delin, R. L. McClelland, and E. L. So. 2006. Coroner and medical examiner documentation of sudden unexplained deaths in epilepsy. Epilepsy Res 68 (2): 137–143. So, E. L. 2008. What is known about the mechanisms underlying SUDEP? Epilepsia 49 (Suppl 9): 93–98. Terrence, C. F., G. R. Rao, and J. A. Perper. 1981. Neurogenic pulmonary edema in unexpected, unexplained death of epileptic patients. Ann Neurol 9 (5): 458–464. Tomson, T., T. Walczak, M. Sillanpaa, and J. W. Sander. 2005. Sudden unexpected death in epilepsy: A review of incidence and risk factors. Epilepsia 46 (Suppl 11): 54–61.

Drug Abuse and SUDEP Steven B. Karch

10

Contents 10.1 Introduction 10.2 Channelopathies and Abused Drugs 10.2.1 QT Interval Prolongation 10.2.2 QT Shortening 10.3 QT Dispersion 10.4 Abnormal Catecholamine Metabolism References

159 161 161 164 164 165 165

10.1â•…Introduction Nearly a quarter century ago, Leestma et al. (1985) described the ἀndings seen at autopsies of 66 epileptics who had died unexpectedly during or just after experiencing a seizure. The autopsy ἀndings were insufficient to explain the cause of death in any of the cases. The syndrome of sudden death during or immediately after a seizure has come to be called sudden unexplained death in epilepsy (SUDEP). Over a 10-year period, the mean age of epileptics with SUDEP in Leestma’s studies was 31.4 years. Of these, 37% were found dead in bed, 49% were black males, 25% were white males, 11% were black females, and 15% were white females ranging in age from 10 months to 60 years (mean age, 28 years); nearly half of the decedents were found “dead in bed,” a description that applies equally well to deaths from long QT syndrome (LQTS), an entity due to heritable channelopathies (Ackerman et al. 2001), or to diabetes with hypoglycemia (Rothenbuhler et al. 2008; Tu et al. 2008). By convention, the abbreviation sudden unexplained death syndrome (SUDS) is used to describe the death of anyone older than 2 years who dies in this fashion, whereas for children 2 years and younger, the diagnosis that is used is sudden infant death syndrome (SIDS). More recently, a similar syndrome has been recognized in young diabetics (Gill et al. 2009). As with SUDEP, the cause of death is not apparent in any of these disorders, although it seems probable that an explanation is to be found only at the molecular level. One popular hypothesis holds that SUDEP victims may have died because of QT interval prolongation (QTd) and that SUDEP is just another variety of LQTS. The presence of prolonged QT intervals reflects delayed cardiac repolarization. It can be caused by a variety of different abnormalities, including acute hypoglycemia, superimposed upon the presence of cardiac autonomic neuropathy (Tu et al. 2008). This process may account for the death of young diabetics as noted above, although why the QT interval in young diabetics should be prolonged has never really been explained. Other factors, such as channelopathies and cardiomyocyte membrane abnormalities, may also be involved. The notion that the same abnormality might account for both SUDEP and LQTS (Aurlien et al. 2009) has the potential to open up new areas of research. 159

160 Sudden Death in Epilepsy: Forensic and Clinical Issues

Numerous risk factors for SUDEP have been proposed and a few have been consistently identiἀed, including young age, early seizure onset, refractory seizures, generalized tonic clonic seizures, and male gender. Studies have found that many of these individuals are in bed, presumably asleep, at the time of death, that anticonvulsant blood levels are subtherapeutic, and that a structural brain lesion often can be identiἀed (Leestma et al. 1985). The current consensus seems to be that SUDEP is primarily a seizure-related cause of death, but the mechanisms underlying SUDEP are unknown (Jehi and Najm 2008; Hughes 2009). Most SUDEP victims are found alone in a room at home and only rarely have they been vigorously exercising. A recent meta-analysis, which included a review of the Cochrane database (Monte et al. 2007), found that patients were more likely to be discovered asleep or, at least, in their beds. Subtherapeutic concentrations of antiseizure medications were found in nearly 70% of Leestma’s cases, many of the decedents having no medications detectable at all. Signiἀcant but nonprogressive brain abnormalities were present in more €than half the decedents. Leestma et al. (1985) estimated the prevalence of seizureassociated SUDEP was between 1:525 and 1:2100 epileptics (Tellez-Zenteno et al. 2005). Other retrospective studies have reported both slightly higher and slightly lower incidence rates (Tellez-Zenteno et al. 2005; Hughes 2009), but the fact remains that SUDEP is the most common cause of death in persons with epilepsy. SUDEP is rare in those who have only recently been diagnosed with epilepsy, and it is equally uncommon in those who are in remission. Many theories accounting for the etiology of SUDEP have been proposed (Kloster and Engelskjon 1999; Lathers and Schraeder 2006; So and Sperling 2007; Aurlien et al. 2009). Most involve cardiac arrhythmias, mediated by sympathetic autonomic events, which are thought to occur during the seizures. The possibility of heritable channelopathy is receiving much closer scrutiny than in the past. Some candidate genes make the QT interval longer and some make it shorter, but both abnormalities are associated with a sudden arrhythmia-related death. These changes are important not only because they have been identiἀed in epileptics (Akalin et al. 2003), but also because they are associated with the use of some abused drugs, in particular cocaine and alcohol (Gamouras et al. 2000; Karle and Kiehn 2002; Uyarel et al. 2005; Yap et al. 2009). The possibility that genetic polymorphisms may be responsible for SIDS, SUDS, and even SUDEP cannot be ignored, but neither can the process known as myocardial remodeling, which causes the QT interval to be longer in some parts of the heart than in others. This phenomenon is referred to as QT interval dispersion; if the difference between the longest and shortest QT interval measured in a 12-lead electrocardiogram exceeds 80 ms, a state of QT dispersion (QTd) is said to exist (Anderson 2003). QTd has the same strong association with sudden cardiac death as interstitial ἀbrosis and channelopathy (Cuddy et al. 2009). The process of myocardial remodeling has received relatively little attention from neurologists, but is increasingly recognized as a cause of sudden death by cardiologists and electrophysiologists (Fischer et al. 2007). Stimulant drugs initiate the same remodeling process as seen in states of chronic catecholamine excess, and such a state can be said to exist during recurrent seizure activity (Meierkord et al. 1994; Henning and Cuevas 2006). Hypertension (Haider et al. 1998), catecholamine excess (Rona 1985), and stimulant drug abuse (Karch et al. 1998 1999) always lead to myocardial ἀbrosis and left ventricular hypertrophy (i.e., myocardial remodeling). Either or both of these processes can provide

Drug Abuse and SUDEP

161

the substrate for sudden arrhythmia death (John et al. 2004; Haider et al. 1998). Ventricular hypertrophy and ἀbrosis are both common ἀndings in SUDEP victims (P-Codrea Tigaran et al. 2005) and it may well be that QTd is responsible for some cases of SUDEP.

10.2â•…Channelopathies and Abused Drugs 10.2.1â•…QT Interval Prolongation Ackerman and others have proposed that perturbations of the hERG channels (“rapid delayed repolarizing channel”) can confer a susceptibility for seizures themselves or, alternatively, cause QT prolongation, resulting in syncopal episodes that trigger seizures (Johnson et al. 2009; Nemec et al. 2009). This is not an unreasonable theory given that many patients who have seizures and QT prolongation are often confused with patients suffering from primary seizure disorders, and are treated with antiepileptic drugs instead of the internal deἀbrillators they really need (Lathers et al. 2010). In a recently published controlled case study, seizure phenotype was recorded in 98 of 343 (29%) patients who were randomly checked for the most common channelopathies. A seizure phenotype was more common in individuals with LQT2 (36 of 77 or 47%) than those with LQT1 (16 of 72 or 22%, p < 0.002) and LQT3 (7 of 28 or 25%, p < 0.05, not signiἀcant). LQT1 and LQT3 combined cohorts did not differ signiἀcantly from the expected background rates of a seizure phenotype, but a personal history of seizures was much more commonly found in those individuals with LQT2 (30 of 77 or 39%) than all other subtypes of LQTS (11 of 106 or 10%, p < 0.001) (Johnson et al. 2009). If this connection can be proven in other studies, the observation is potentially very signiἀcant. The gene, KCNH2, is responsible for LQT2. The KCNH2 channel was cloned originally from the hippocampus; it encodes a potassium channel active in hippocampal astrocytes and in the heart (Zehelein et al. 2004). Not only can this channel interact with some antiepileptic medications, e.g., phentoyin, phenobarbital, toprimate, and flunariÂ� zine (Danielsson et al. 2003; Trepakova et al. 2006; Yang et al. 2008), but it also interacts with abused drugs such as cocaine, methamphetamine, methadone, anabolic steroids, and even alcohol. This observation suggests that, in some cases of SUDEP, drug abuse may be the cause of death. That this observation has not yet been recorded in the literature only reflects the fact that full toxicology screening for abused drugs is not part of the routine investigation of SUDEP. Indeed, the autopsy is often performed even before a history of epilepsy is established (Kloster and Engelskjon 1999). Lathers et al. (1990) suggest, “… since both cocaine and epilepsy alone are associated with sudden unexpected death and since both are capable of modifying cardiac sympathetic neural discharge to produce changes in heart rate and rhythm, the question of whether the use of cocaine in the epileptic person places this individual at risk for sudden death must be raised.” It is also known that alcohol use is associated with an increased risk of autonomic dysfunction, seizure, and sudden death (Chan et al. 1990). It is clear that no single neurochemical system can adequately explain the complex nature of epileptic seizures. The same can be said for the neurochemical mechanisms that underlie alcohol withdrawal reactions in that there are complex, dynamic interactions among the neurotransmitters and neuromodulator systems in the brain. Another complicating factor is that

162 Sudden Death in Epilepsy: Forensic and Clinical Issues

occurrence of withdrawal seizures is only one component of the withdrawal syndrome. The challenge is to separate the neurochemical changes that may trigger seizures from those that are actually the result of the seizures themselves, or some other complication of the withdrawal reaction. Some of these factors include the individual’s genetic makeup, seizure threshold, health status, malnutrition, polydrug abuse, a history of epilepsy, prior alcohol withdrawal seizures, and trauma. Despite the difficulties associated with interpreting the neurochemical changes associated with ethanol withdrawal, there do appear to be similarities between the abnormalities observed in alcoholics and those postulated to be involved in the mechanism of epileptic seizures (Lathers et al. 1990). Twenty years have lapsed since Lathers et al. (1990) and Chan et al. (1990) discussed the possible interactions of cocaine and alcohol and SUDEP. The time is long overdue for us to answer these questions. The results of very recent discoveries make the effort to answer these questions even more worthwhile. When genome-wide data from ἀve different population-based cohorts, composed of 15,842 individuals of European ancestry, were analyzed, a total of 10 loci associated with the occurrence of LQTS were identiἀed. Four of these loci map near the monogenic LQTS genes: KCNQ1, KCNH2, SCN5A, and KCNJ2. Two other loci, ATP1B1 and PLN, have already been shown to be genes with established electrophysiological functions, whereas three of the newly discovered genes map to RNF207, near LITAF and within the NDRG4–GINS3–SETD6–CNOT1 complex, respectively. Until this study was undertaken, not a single one of these genes was thought to have anything to do with cardiac function (Newton-Cheh et al. 2009; Pfeufer et al. 2009). Any one of these genes could be responsible for LQTS, torsades de pointes, and sudden cardiac death (Pfeufer et al. 2009). The most striking thing about this new discovery is that it was replicated within the same week by another group of scientists who came up with exactly the same results (Newton-Cheh et al. 2009). If this sequence of events were to occur in a person with epilepsy, it would be called SUDEP, since the genetic component would not be detected by the medical examiner.€Even though these new discoveries very strongly suggest a nexus between SUDEP, channelopathies, and drug abuse, further understanding of the pathophysiology is required. Associations between LQT2 and epilepsy that had not previously been suspected are now known to exist. This raises the possibility that LQT2 perturbations in the KCNH2-encoded potassium channel may confer susceptibility for recurrent seizure activity (Johnson et al. 2009). When a cardiomyocyte depolarizes, the rapid component of the delayed rectiἀer K+ current, abbreviated as IKr, plays a key role in cell repolarization. During the plateau phase of the depolarization and repolarization cycle, the current generated through the IKr channel is small. As repolarization proceeds, a transient increase in the IKr outward current occurs due to fast recovery from inactivation and slow deactivation, ultimately leading to repolarization of the cardiac cell. The LQTII or hERG gene controls the IKr channel (Thomas et al. 2006). In congenital forms of LQTII syndrome, flow through the channel is slowed because the structure of the potassium pore itself is abnormal. As a consequence, the action potential is prolonged. This leads to the occurrence of early after depolarization currents that, in turn, can lead to lethal arrhythmias (torsades de pointes, literally “twisting of the points,” a form of ventricular tachycardia). However, LQTS can occur even in someone with a perfectly normal hERG gene because so many drugs, both licit and illicit, such as ἀrst and second generation antidepressants, interact with the channel (Sala et al. 2006). Since the

Drug Abuse and SUDEP

163

majority of antiepileptic drugs block voltage-dependent sodium and calcium channels, enhance GABAergic transmission, and/or antagonize glutamate receptors, it only seems reasonable to assume that an epileptic who abuses drugs is at far greater risk for arrhythmia, even if the effects of their antiseizure medications predominate. Although there is nothing in the literature to suggest that any of the commonly used anti-seizure medications cause QT prolongation, there is ample evidence that some abused drugs do. Cocaine, alcohol, and cocaethylene (a cocaine metabolite that is formed only in the presence of signiἀcant amounts of alcohol) all block the hERG potassium channel, effectively producing an acquired form of LQTII (Karle and Kiehn 2002). Whether they interact with any of the other recently discovered channels is not known. When hERG currents are measured in vitro using the patch-clamp technique, cocaethylene (the only metabolite of cocaine that is psychoactive) increases the QT interval and causes torsades de pointes (Ferreira et al. 2001; O’Leary 2002). Other studies have shown cocaethylene accelerates the inactivation of hERG current without affecting recovery from inactivation. Depending on the molecular conἀguration of an ion pore, there are several different ways to disrupt normal IKr function. Cocaine and its metabolite, cocaethylene, produce what is called open channel block, which prevents the channel closing by putting a “foot in the door” (Wang et al. 2000). Much the same mechanism has been proposed for quaternary ammonium compounds. Study of cocaethylene is particularly important because the IKr current is found to be altered, even at realistic plasma drug concentrations (O’Leary 2002). hERG K+ channels have also been found in neurons, putatively contributing to the neuropsychological behavior of individuals who have been drinking alcohol and/or using cocaine. Methamphetamine abuse also causes QT prolongation, even though it does not interact with the hERG channel (or any other known cardiac channel) (Haning and Goebert 2007). This somewhat perplexing result may be a consequence of catecholamine toxicity, as it is well established that epinephrine induces LQTS (Urao et al. 2004), and methamphetamine usage causes catecholamine release from synaptic vesicles, thereby increasing circulating levels of catecholamines (Stuerenburg et al. 2002). Central sympathetic stimulation, such as occurs during seizures, also causes release of catecholamines and could provide a plausible explanation for this phenomena (Eckard et al. 1999). Clinical studies show that cerebral infarction causes various cardiovascular and electrocardiographic abnormalities, depending on the location and the size of the infarct. The two most frequently encountered abnormalities in patients with stroke are QT interval prolongation and widening of the QRS complex. Disease of the insular cortex seems to be very important for activation of the sympathetic nervous system. Patients with brain stem infarction have substantially higher mean plasma norepinephrine levels than patients with hemispheric infarction; on the other hand, hemispheric lesions are associated with a signiἀcantly higher incidence of cardiac arrhythmias when compared to patients with brain stem infarction (Klingelhofer and Sander 1997). Whether any of these changes can be related to the mechanism that causes arrhythmias in epileptics is not known. Methadone has recently been added to the list of drugs shown to produce QT prolongation, arrhythmias, and sudden death. It has been known for several years that highdose methadone can induce QT prolongation (Kornick et al. 2003) by hERG inhibition. However, new evidence shows that QT prolongation can occur at much lower doses of methadone, even when the drug is not given intravenously. Methadone is a chiral€drug, but only (R)-methadone provides any pain relief. Laboratory experiments have shown that

164 Sudden Death in Epilepsy: Forensic and Clinical Issues

(S)-methadone, which does not bind the mu receptor or provide pain relief, blocks the hERG current three-and-a-half times more potently than (R)-methadone. If an individual is a CYP2B6 slow metabolizer (SM), they can still metabolize (R)-methadone normally, but they cannot metabolize the (S) form. Concentrations of the (S) form of methadone will continue to rise, eventually leading to hERG blockade and, potentially, torsades de pointes (Eap et al. 2007). 10.2.2â•…QT Shortening Pathologic reduction in the QT interval is much less common than prolongation but it is also associated with sudden death. Brugada et al. (2004) linked SQTS to a KCNH2 gene mutation, the same gene responsible for QT prolongation. This disorder is characterized by a corrected QT (QTc) interval that is shorter than normal (QTc ≤€320 ms), and it is often associated with atrial ἀbrillation, syncopal episodes, and/or sudden cardiac death in patients who are said to have no anatomic evidence of heart disease (Zareba and Cygankiewicz 2008). While there is general agreement about what constitutes a prolonged QT interval, the deἀnition of short QT (SQTS) remains somewhat controversial, though most accept that the lower boundary of normal is on the order of 320 ms. Unlike the mutations in LQTS, which result in loss of function, mutations in SQTS cause an increase in function with rapid repolarization; cardiac arrest may be the ἀrst symptom (Giustetto et al. 2006). The recent discovery that anabolic steroid abuse is associated with a reduction in the QT interval (Bigi et al. 2009) is most provocative, given the repeated observation that abnormalities of reproductive endocrine hormones are more often found in men with epilepsy than in the general population (Roste et al. 2005). There is an ongoing debate whether this increased risk in males can be attributed to the use of antiepileptic drugs or the epilepsy itself. The corrected QT interval in proven anabolic steroid abusers is signiἀcantly shorter than the QT interval of drug-free bodybuilders and that of sedentary men. In this recently study, sedentary men were found to have a QTc interval of 418 ± 23.6 ms, drug-free bodybuilders had a QTc interval of 422 ± 24.5 ms, and steroid abusing bodybuilders had a QTc interval of 367 ± 17.1 ms (p < 0.01). In fact, the correlation between steroid abuse and QT interval is so strong that some have recommended EKG screening as a method of detecting steroid abusers (Bigi et al. 2009). Similar ἀndings have been produced by the administration of anabolic steroids to experimental animals (Fulop et al. 2006; Liu et al. 2003).

10.3╅QT Dispersion Animal studies suggest that hippocampal norepinephrine transporters are downregulated when chronically exposed to cocaine (Kitayama et al. 2006), and human studies have shown a decrease in dopamine and serotonin transporters in the same areas (Mash et al. 2000). It should not be forgotten that all local anesthetics are Na channel blockers, both in the brain and the heart, and cocaine is a local anesthetic. Thus, it might be reasonable to postulate that some drug abusers who also have epilepsy, of which there appear to be more than a few (Opeskin et al. 2000),€die as a consequence of the drug abuse acting in synergy with their primary disease to cause seizure-related death.

Drug Abuse and SUDEP

165

10.4â•…Abnormal Catecholamine Metabolism Disruption of postganglionic reuptake is thought to be the mechanism that leads to bradyarrhythmia or even asystole during epileptic seizures. In the heart, as opposed to the brain, the action of norepinephrine is terminated primarily by reuptake (Kloner et al. 1992; Schafers et al. 1998) by the actions of catechol-O-methyl transferase (COMT). Recent studies of epileptics who have experienced bradyarrhythmia or asystole have shown that these individuals have dramatically abnormal postganglionic cardiac norepinephrine uptake (Kerling et al. 2009), suggesting impaired sympathetic cardiac innervation, resulting in a limited ability to adjust and modulate heart rate, or even cause asystole. A large number of commonly abused drugs (cocaine, methamphetamine, and 3,4-methylenedioxymethamphetamine) profoundly disrupt catecholamine metabolism, and their use may well be responsible for the death of an occasional drug abuser who also has epilepsy. However, the most current studies suggest variations in COMT activity are more likely due to genetic polymorphisms and, no doubt, more work will be done in this area in the near future (Haile et al. 2009).

References Ackerman, M. J., D. J. Tester, and D. J. Driscoll. 2001. Molecular autopsy of sudden unexplained death in the young. Am J Forensic Med Pathol 22 (2): 105–111. Akalin, F., A. Tirtir, and Y. Yilmaz. 2003. Increased QT dispersion in epileptic children. Acta Paediatr 92 (8): 916–920. Anderson, K. P. 2003. Sympathetic nervous system activity and ventricular tachyarrhythmias: Recent advances. Ann Noninvasive Electrocardiol 8 (1): 75–89. Aurlien, D., T. P. Leren, E. Tauboll, and L. Gjerstad. 2009. New SCN5A mutation in a SUDEP victim with idiopathic epilepsy. Seizure 18 (2): 158–160. Bigi, M. A., A. Aslani, and A. Aslani. 2009. Short QT interval: A novel predictor of androgen abuse in strength trained athletes. Ann Noninvasive Electrocardiol 14 (1): 35–39. Brugada, R., K. Hong, R. Dumaine, J. Cordeiro, F. Gaita, M. Borggrefe, T. M. Menendez et al. 2004. Sudden death associated with short-QT syndrome linked to mutations in HERG. Circulation 109 (1): 30–35. Chan, A. W. K., C. M. Lathers, and J. E. Leestma. 1990. Chapter 19. Alcohol, arrhythmias, seizures, and sudden death. In Epilepsy and Sudden Death, ed. C. Lathers and P. Schraeder. New York, NY: Marcel Dekker. Cuddy, T. E., P. S. Halli, and R. B. Tate. 2009. QT dispersion and heart rate predict the risk of sudden unexpected cardiac death in men: The Manitoba Follow-Up Study. Prev Cardiol 12 (1): 27–33. Danielsson, B. R., K. Lansdell, L. Patmore, and T. Tomson. 2003. Phenytoin and phenobarbital inhibit human HERG potassium channels. Epilepsy Res 55 (1–2): 147–157. Eap, C. B., S. Crettol, J. S. Rougier, J. Schlapfer, L. Sintra Grilo, J. J. Deglon, J. Besson, M. CroquetteKrokar, P. A. Carrupt, and H. Abriel. 2007. Stereoselective block of hERG channel by (S)-methadone and QT interval prolongation in CYP2B6 slow metabolizers. Clin Pharmacol Ther 81 (5): 719–728. Eckardt, M., L. Gerlach, and F. L. Welter. 1999. Prolongation of the frequency-corrected QT dispersion following cerebral strokes with involvement of the insula of Reil. Eur Neurol 42 (4): 190–193. Ferreira, S., W. J. Crumb Jr., C. G. Carlton, and C. W. Clarkson. 2001. Effects of cocaine and its major metabolites on the HERG-encoded potassium channel. J Pharmacol Exp Ther 299 (1): 220–226.

166 Sudden Death in Epilepsy: Forensic and Clinical Issues Fischer, R., R. Dechend, A. Gapelyuk, E. Shagdarsuren, K. Gruner, A. Gruner, P. Gratze et al. 2007. Angiotensin II-induced sudden arrhythmic death and electrical remodeling. Am J Physiol Heart Circ Physiol 293 (2): H1242–H1253. Fulop, L., T. Banyasz, G. Szabo, I. B. Toth, T. Biro, I. Lorincz, A. Balogh, K. Peto, I. Miko, and P. P. Nanasi. 2006. Effects of sex hormones on ECG parameters and expression of cardiac ion channels in dogs. Acta Physiol (Oxf) 188 (3–4): 163–171. Gamouras, G. A., G. Monir, K. Plunkitt, S. Gursoy, and L. S. Dreifus. 2000. Cocaine abuse: Repolarization abnormalities and ventricular arrhythmias. Am J Med Sci 320 (1): 9–12. Gill, G. V., A. Woodward, I. F. Casson, and P. J. Weston. 2009. Cardiac arrhythmia and nocturnal hypoglycaemia in type 1 diabetes—The ‘dead in bed’ syndrome revisited. Diabetologia 52 (1): 42–45. Giustetto, C., F. Di Monte, C. Wolpert, M. Borggrefe, R. Schimpf, P. Sbragia, G. Leone et al. 2006. Short QT syndrome: Clinical ἀndings and diagnostic–therapeutic implications. Eur Heart J 27€(20): 2440–2447. Haider, A. W., M. G. Larson, E. J. Benjamin, and D. Levy. 1998. Increased left ventricular mass and hypertrophy are associated with increased risk for sudden death. J Am Coll Cardiol 32 (5): 1454–1459. Haile, C. N., T. R. Kosten, and T. A. Kosten. 2009. Pharmacogenetic treatments for drug addiction: Cocaine, amphetamine and methamphetamine. Am J Drug Alcohol Abuse 35 (3): 161–177. Haning, W., and D. Goebert. 2007. Electrocardiographic abnormalities in methamphetamine abusers. Addiction 102 (Suppl 1): 70–75. Henning, R. J., and J. Cuevas. 2006. Cocaine activates calcium/calmodulin kinase II and causes cardiomyocyte hypertrophy. J Cardiovasc Pharmacol 48 (1): 802–813. Hughes, J. R. 2009. A review of sudden unexpected death in epilepsy: Prediction of patients at risk. Epilepsy Behav 14 (2): 280–287. Jehi, L., and I. M. Najm. 2008. Sudden unexpected death in epilepsy: Impact, mechanisms, and prevention. Cleve Clin J Med 75 (Suppl 2): S66–S70. John, B. T., B. K. Tamarappoo, J. L. Titus, W. D. Edwards, W. K. Shen, and S. S. Chugh. 2004. Global remodeling of the ventricular interstitium in idiopathic myocardial ἀbrosis and sudden cardiac death. Heart Rhythm 1 (2): 141–149. Johnson, J. N., N. Hofman, C. M. Haglund, G. D. Cascino, A. A. Wilde, and M. J. Ackerman. 2009. Identiἀcation of a possible pathogenic link between congenital long QT syndrome and epilepsy. Neurology 72 (3): 224–231. Karch, S. B., B. G. Stephens, and C. H. Ho. 1999. Methamphetamine-related deaths in San Francisco: Demographic, pathologic, and toxicologic proἀles. J Forensic Sci 44 (2): 359–368. Karch, S. B., B. Stephens, and C. H. Ho. 1998. Relating cocaine blood concentrations to toxicity—An autopsy study of 99 cases. J Forensic Sci 43 (1): 41–45. Karle, C. A., and J. Kiehn. 2002. An ion channel ‘addicted’ to ether, alcohol and cocaine: The HERG potassium channel. Cardiovasc Res 53 (1): 6–8. Kerling, F., M. Dutsch, R. Linke, T. Kuwert, H. Stefan, and M. J. Hilz. 2009. Relation between ictal asystole and cardiac sympathetic dysfunction shown by MIBG-SPECT. Acta Neurol Scand 120€(2): 123–129. Kitayama, T., L. Song, K. Morita, N. Morioka, and T. Dohi. 2006. Down-regulation of norepinephrine transporter function induced by chronic administration of desipramine linking to the alteration of sensitivity of local-anesthetics–induced convulsions and the counteraction by coadministration with local anesthetic. Brain Res 1096 (1): 97–103. Klingelhofer, J., and D. Sander. 1997. Cardiovascular consequences of clinical stroke. Baillieres Clin Neurol 6 (2): 309–335. Kloner, R. A., S. Hale, K. Alker, and S. Rezkalla. 1992. The effects of acute and chronic cocaine use on the heart. Circulation 85 (2): 407–419. Kloster, R., and T. Engelskjon. 1999. Sudden unexpected death in epilepsy (SUDEP): A clinical perspective and a search for risk factors. J Neurol Neurosurg Psychiatry 67 (4): 439–444.

Drug Abuse and SUDEP

167

Kornick, C. A., M. J. Kilborn, J. Santiago-Palma, G. Schulman, H. T. Thaler, D. L. Keefe, A. N. Katchman et al. 2003. QTc interval prolongation associated with intravenous methadone. Pain 105 (3): 499–506. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9 (2): 236–242. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2010. Neurocardiologic mechanistic risk factors in sudden unexpected death in epilepsy. Ch. 1 in Sudden Death in Epilepsy: Forensic and Clinical Issues, eds. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton: CRC Press. Lathers, C. M., M. M. Spino, I. Agarwal, L. S. Y. Tyau, and W. B. Pickworth. 1990. Chapter 27. Cocaine-induced seizures, arrhythmias, and sudden death. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. Schraeder. New York, NY: Marcel Dekker. Leestma, J. E., J. R. Hughes, S. S. Teas, and M. B. Kalelkar. 1985. Sudden epilepsy deaths and the forensic pathologist. Am J Forensic Med Pathol 6 (3): 215–218. Liu, X. K., A. Katchman, B. H. Whitἀeld, G. Wan, E. M. Janowski, R. L. Woosley, and S. N. Ebert. 2003. In€vivo androgen treatment shortens the QT interval and increases the densities of inward and delayed rectiἀer potassium currents in orchiectomized male rabbits. Cardiovasc Res 57 (1): 28–36. Mash, D. C., J. K. Staley, S. Izenwasser, M. Basile, and A. J. Ruttenber. 2000. Serotonin transporters upregulate with chronic cocaine use. J Chem Neuroanat 20 (3–4): 271–280. Meierkord, H., S. Shorvon, and S. L. Lightman. 1994. Plasma concentrations of prolactin, noradrenaline, vasopressin and oxytocin during and after a prolonged epileptic seizure. Acta Neurol Scand 90 (2): 73–77. Monte, C. P., J. B. Arends, I. Y. Tan, A. P. Aldenkamp, M. Limburg, and M. C. de Krom. 2007. Sudden unexpected death in epilepsy patients: Risk factors. A systematic review. Seizure 16 (1): 1–7. Nemec, J., M. Buncova, V. Shusterman, B. Winter, W. K. Shen, and M. J. Ackerman. 2009. QT interval variability and adaptation to heart rate changes in patients with long QT syndrome. Pacing Clin Electrophysiol 32 (1): 72–81. Newton-Cheh, C., M. Eijgelsheim, K. M. Rice, P. I. de Bakker, X. Yin, K. Estrada, J. C. Bis et al. 2009. Common variants at ten loci influence QT interval duration in the QTGEN Study. Nat Genet 41 (4): 399–406. O’Leary, M. E. 2002. Inhibition of HERG potassium channels by cocaethylene: A metabolite of cocaine and ethanol. Cardiovasc Res 53 (1): 59–67. Opeskin, K., A. S. Harvey, S. M. Cordner, and S. F. Berkovic. 2000. Sudden unexpected death in epilepsy in Victoria. J Clin Neurosci 7 (1): 34–37. P-Codrea Tigaran, S., S. Dalager-Pedersen, U. Baandrup, M. Dam, and A. Vesterby-Charles. 2005. Sudden unexpected death in epilepsy: Is death by seizures a cardiac disease? Am J Forensic Med Pathol 26 (2): 99–105. Pfeufer, A., S. Sanna, D. E. Arking, M. Muller, V. Gateva, C. Fuchsberger, G. B. Ehret et al. 2009. Common variants at ten loci modulate the QT interval duration in the QTSCD Study. Nat Genet 41 (4): 407–414. Rona, G. 1985. Catecholamine cardiotoxicity. J Mol Cell Cardiol 17 (4): 291–306. Roste, L. S., E. Tauboll, L. Morkrid, T. Bjornenak, E. R. Saetre, T. Morland, and L. Gjerstad. 2005. Antiepileptic drugs alter reproductive endocrine hormones in men with epilepsy. Eur J Neurol 12 (2): 118–124. Rothenbuhler, A., C. P. Bibal, S. Le Fur, and P. Bougneres. 2008. Effects of a controlled hypoglycaemia test on QTc in adolescents with Type 1 diabetes. Diabet Med 25 (12): 1483–1485. Sala, M., F. Coppa, C. Cappucciati, P. Brambilla, G. d’Allio, E. Caverzasi, F. Barale, and G. M. De Ferrari. 2006. Antidepressants: Their effects on cardiac channels, QT prolongation and Torsade de Pointes. Curr Opin Investig Drugs 7 (3): 256–263. Schafers, M., D. Dutka, C. G. Rhodes, A. A. Lammertsma, F. Hermansen, O. Schober, and P. G. Camici. 1998. Myocardial presynaptic and postsynaptic autonomic dysfunction in hypertrophic cardiomyopathy. Circ Res 82 (1): 57–62. So, N. K., and M. R. Sperling. 2007. Ictal asystole and SUDEP. Neurology 69 (5): 423–424.

168 Sudden Death in Epilepsy: Forensic and Clinical Issues Stuerenburg, H. J., K. Petersen, T. Baumer, M. Rosenkranz, C. Buhmann, and R. Thomasius. 2002. Plasma concentrations of 5-HT, 5-HIAA, norepinephrine, epinephrine and dopamine in ecstasy users. Neuro Endocrinol Lett 23 (3): 259–261. Tellez-Zenteno, J. F., L. H. Ronquillo, and S. Wiebe. 2005. Sudden unexpected death in epilepsy: Evidence-based analysis of incidence and risk factors. Epilepsy Res 65 (1–2): 101–115. Thomas, D., C. A. Karle, and J. Kiehn. 2006. The cardiac hERG/IKr potassium channel as pharmaÂ� cological target: Structure, function, regulation, and clinical applications. Curr Pharm Des 12€(18): 2271–2283. Trepakova, E. S., S. J. Dech, and J. J. Salata. 2006. Flunarizine is a highly potent inhibitor of cardiac hERG potassium current. J Cardiovasc Pharmacol 47 (2): 211–220. Tu, E., S. M. Twigg, and C. Semsarian. 2008. Sudden death in type 1 diabetes: The mystery of the ‘dead in bed’ syndrome. Int J Cardiol. Urao, N., H. Shiraishi, K. Ishibashi, M. Hyogo, M. Tsukamoto, N. Keira, S. Hirasaki, T. Shirayama, and M. Nakagawa. 2004. Idiopathic long QT syndrome with early after depolarization induced by epinephrine. A case report. Circ J 68 (6): 587–591. Uyarel, H., C. Ozdol, A. M. Gencer, E. Okmen, and N. Cam. 2005. Acute alcohol intake and QT dispersion in healthy subjects. J Stud Alcohol 66 (4): 555–558. Wang, J., C. D. Myers, and G. A. Robertson. 2000. Dynamic control of deactivation gating by a soluble amino-terminal domain in HERG K(+) channels. J Gen Physiol 115 (6): 749–758. Yang, Z. Q., J. C. Barrow, W. D. Shipe, K. A. Schlegel, Y. Shu, F. V. Yang, C. W. Lindsley et al. 2008. Discovery of 1,4-substituted piperidines as potent and selective inhibitors of T-type calcium channels. J Med Chem 51 (20): 6471–6477. Yap, Y. G., E. R. Behr, and A. J. Camm. 2009. Drug-induced Brugada syndrome. Europace 11 (8): 989–994. Zareba, W., and I. Cygankiewicz. 2008. Long QT syndrome and short QT syndrome. Prog Cardiovasc Dis 51 (3): 264–278. Zehelein, J., D. Thomas, M. Khalil, A. B. Wimmer, M. Koenen, M. Licka, K. Wu et al. 2004. Identiἀcation and characterisation of a novel KCNQ1 mutation in a family with Romano–Ward syndrome. Biochim Biophys Acta 1690 (3): 185–192.

Cocaine-Induced Seizures, Arrhythmias, and Sudden Death

11

Claire M. Lathers Michelle M. Spino Isha Agarwal Laurie S. Y. Tyau Wallace B. Pickworth

Contents 11.1 Introduction 11.2 Mechanisms of Action of Cocaine 11.3 Cocaine-Induced Sudden Death 11.3.1 Cocaine-Induced Changes in Mean Arterial Blood Pressure and Heart Rate 11.3.2 Cocaine-Induced Myocardial Ischemia, Infarction, Arrhythmia, and Cardiomyopathies 11.3.3 Cocaine-Induced Changes in Postganglionic Cardiac Sympathetic Neural Function 11.3.4 Central Actions of Cocaine 11.3.5 Cocaine-Induced Seizures 11.4 Treatment of Cocaine-Induced Arrhythmias and Seizures 11.5 Use of Cocaine in Persons with Epilepsy 11.6 Summary References

169 170 172 172 174 176 176 179 180 181 181 182

11.1â•…Introduction The presence of coca leaves in the tombs of South American Indian mummies suggests that cocaine was used as early as a .d. 600. The use of cocaine is prevalent in modern society. Cregler and Mark (1986a) reviewed the demographics of current cocaine users and found that approximately 1 out of every 10 Americans, have used cocaine at least once (Cregler and Mark 1987). The fallacy that cocaine is a benign, nonaddicting substance may be part of the reason for the alarming rise in abuse (Cregler and Mark 1986b). Although cocaine has been found to be a cardiotoxin, the pathogenesis of this toxicity is not well deἀned (Cregler and Mark 1987). Cocaine use has also been linked to the occurrence of subarachnoid hemorrhage, hypertension, ventricular arrhythmia, tachycardia, acute myocardial infarction, seizure, and sudden death (Lichtenfeld et al. 1984; Nahas et al. 1985; Tazelaar et al. 1987; Young and Glauber 1947). Persons with epilepsy have been shown to manifest autonomic dysfunctions similar to those manifested by cocaine users, 169

170 Sudden Death in Epilepsy: Forensic and Clinical Issues

including changes in blood pressure and heart rate and rhythm, phenomena that may be contributory to sudden unexpected death (Leestma et al. 1984; Penἀeld and Erickson 1941; Phizackerly et al. 1954; Walsh et al. 1968). Thus, one must ask whether the use of cocaine in individuals with epilepsy places these individuals at risk of dying in a sudden unexplained manner.

11.2â•…Mechanisms of Action of Cocaine Cocaine (Figure 11.1), extracted from the leaves of Erythroxylon coca, is a potent local anesthetic agent (Cregler and Mark 1986a; Gould et al. 1985) possessing membrane-stabilizing effects at low plasma levels (Tazelaar et al. 1987). It is also a sympathomimetic agent at higher plasma concentrations (Benchimol et al. 1978; Duke 1986). Cocaine ampliἀes the effect of catecholamines by blocking the reuptake at the synaptic junctions, causing a local excess of norepinephrine at the synaptic cleft. As a result of the excess of norepinephrine at the nerve terminal, there is a prolongation and potentiation of the activity of norepinephrine (Weiss 1986). Norepinephrine is the primary neurotransmitter of the sympathetic nervous system. Excitation of the sympathetic nervous system produces physiological characteristics, such as mobilization of adrenal catecholamines, causing an increase in blood pressure and the heart rate, dilatation of the pupils, a rise in blood sugar levels, vasoconstriction of vessels in the brain and muscles, tightening of the sphincters, and an elevation of body temperature. The intense peripheral vasoconstriction retards reabsorption. Drug effects include intense euphoria and elation, garrulousness, excitability, and irritability; with repeated administration, paranoid ideation, delirium, and assaultiveness occur. Table 11.1 summarizes the actions of cocaine on the cardiovascular, respiratory, and central nervous system (Gay 1982). Cocaine as a hydrochloride salt is brought into the United States with purity ranging up to 95% (Gay 1982). The purity is decreased to 25–90% of its original state through the addition of diluents and adulterants such as procaine, lidocaine (Cregler and Mark 1986b), caffeine, benzocaine, amphetamines, heroin, quinine, talc, and phencyclidine. All adulterants contribute to the toxicity of cocaine. Finally, it is combined with sugars such as mannitol, lactose, and glucose to attain a ἀnal volume and weight (Gay 1982). The resulting cocaine street product can be administered by various routes, including intravenous and subcutaneous injections, intranasal inhalation (snorting), and the current vogue of smoking a “freebase” form of cocaine (crack). Freebase smoking or intravenous injections of

CH3 N

COOCH3

OOCC6H5 H

Figure 11.1╇ Structure of cocaine (C17 H 21NO4).

Cocaine-Induced Seizures, Arrhythmias, and Sudden Death

171

Table 11.1â•… The Cocaine Reaction Phase I: Early stimulation

II: Advanced stimulation

III: Depressive

Central Nervous System

Cardiovascular System

Euphoria: stated feelings of “soaring,” well-being Elation, expansive good humor, laughing, mydriasis Talkative, garrulous Excited, flighty, emotionally unstable Restless, irritable, apprehensive, unable to sit still Stereotyped movements (such as “picking” or “stroking”), bruxism Nausea, vomiting, vertigo Sudden headache Cold sweats Tremor (nonintentional) Twitching of small muscles, especially of the face, ἀngers, and feet Tics, generalized Preconvulsive, tonic and clonic jerks Possible psychosis, hallucinations Core body temperature rises Verbalization of impending doom (precedes imminent total collapse) Unresponsive to voice; decreased responsiveness to all stimuli Increased deep tendon reflexes Generalized hyperreflexia Convulsions: tonic and clonic Status epilepticus Incontinence Malignant encephalopathy possible

Pulse vanes at ἀrst, may immediately slow because of reflex vagal effect; will increase 30% to 50% above normal with system absorption of 25 mg of cocaine Blood pressure usually elevates 15% to 20% above normal with similar dosages as noted above Skin pallor caused by vasoconstriction Premature ventricular contractions

Increased respiratory rate and depth Dyspnea

Increased pulse and blood pressure: high output failure possible Blood pressure falls as ventricular dysrhythmias supervene and inefficient cardiac output results Pulse becomes rapid, weak, and irregular Peripheral, then central cyanosis Ventricular ἀbrillation Circulation failure Ashen gray cyanosis No palpable pulse Cardiac arrest Paralysis of medullary brain center Exitus

Gasping, rapid, or irregular respiration (Chevne–Stokes)

Flaccid paralysis of muscles Coma Pupils ἀxed and dilated Loss of reflexes Loss of vital support functions Paralysis of medullary brain center Exitus

Respiratory System

Agonal gasps Respiratory failure Gross pulmonary edema Paralysis of medullary brain center Exitus

cocaine cause a euphoric or “rush” experience, which occurs within 45 s and is associated with a rapid increase in plasma cocaine concentrations. The effect lasts for approximately 20 min. In contrast, intranasal administration results in euphoria occurring within 3 to 5 min of administration and lasts for 1 to 1.5 h (Van Dyke and Byck 1983). Regardless of the route of administration, accounts of cocaine-induced sudden death have become common.

172 Sudden Death in Epilepsy: Forensic and Clinical Issues

11.3â•…Cocaine-Induced Sudden Death Sudden death has been shown to be induced by cocaine (Amon et al. 1986; Estroff and Gold 1986; Mittleman and Welti 1987; Welti and Fishbain 1985). Reports indicate 1.2 g to be a lethal dose; however, severe toxic effects have been reported with doses as low as 20 mg (Estroff and Gold 1986). Because it is so sudden, medical personnel do not ordinarily witness cocaineinduced death; victims usually collapse and die before resuscitation efforts can begin. Confusion or convulsions precede death induced by cocaine. Estroff and Gold (1986) reported seven cases of sudden death associated with the use of cocaine, in whom a state of excited delirium was the fatal symptom. The initial symptom was intense paranoia, followed by bizarre and violent behavior necessitating the use of force to restrain the patient. The unexpected outbursts of strength were associated with hyperthermia, which was thought to be due to a direct effect of cocaine on the central nervous system center for temperature regulation, and due to peripheral vasoconstriction, with resultant reduction in heat (Ritchie and Greene 1980). Status epilepticus, respiratory paralysis, or cardiac arrhythmias genrally precede sudden death induced by cocaine. Abramowicz (1986) suggested that most sudden deaths associated with cocaine use are caused by seizures leading to anoxia. Recent clinical data have been correlated with pathological ἀndings, generating several hypotheses that attempt to deἀne forensically the pathological mechanisms of cocaine-induced sudden death. 11.3.1â•…Cocaine-Induced Changes in Mean Arterial Blood Pressure and Heart Rate The circulatory effects of cocaine are believed to be of both central and peripherally induced vasoconstriction and cardioacceleration (Young and Glauber 1947). Change in heart rate is a sensitive measure of cocaine-induced cardiovascular effect (Fischman et al. 1976). Cocaine results in dose-related changes in heart rate (Javiad et al. 1978), with small doses decreasing heart rate via central vagal action and moderate doses increasing heart rate via atrial and peripheral sympathetic stimulation (Benchimol et al. 1978). Extremely high intravenous doses have direct toxic effects on the heart and cause immediate death (Nanji and Filipenko 1984; Young and Glauber 1947). The duration of the cardiovascular action is dependent on the dose of cocaine. Fischman et al. (1976) showed that an increase in heart rate was evident after intravenous injections of varying doses of cocaine; the increase began 2–5 min after infusion, peaked at 10 min, and rapidly returned to baseline (Figure 11.2). Cocaine also increased blood pressure in a dose-related manner, but more variability is seen in this measure. In one study (Pitts et al. 1987), cocaine was administered intravenously and evoked a rapid, transient, dose-dependent rise in mean arterial pressures (Figure 11.3). Jain et al. (1987) reported that the administration of cocaine (0.25 mg/kg, i.v.) to anesthetized cats increased systolic and diastolic blood pressure by 33 ± 11 and 31 ± 7 mm Hg, respectively. The dose also enhanced the pressor responses to intravenous norepinephrine and to bilateral carotid occlusion. Doses of 0.5 and 1.0 mg/kg (i.v.) also caused an increase in blood pressure and responses to intravenous norepinephrine but did not increase the blood pressure response to bilateral carotid occlusion. Higher doses had no additive effect on the blood pressure, but rather slowed the heart rate, attenuated blood pressure responses to norepinephrine, prolonged the QRS duration, and decreased tidal volumes. All effects

Mean heart rate (beats/min)

Cocaine-Induced Seizures, Arrhythmias, and Sudden Death 120 115 110 105 100 95 90 85 80 75 70 65 60 55 50

173 4 mg 8 mg 16 mg 32 mg Saline

Pre-drug

0

8

16

24

32

40

48

56

Time since drug injection (min)

Figure 11.2╇ Mean heart rate as a function over time after cocaine is injected. 40 20 (a)

0 –20 –40 –60

Mean percent change

40 20 (b)

0 –20 –40 –60 40 20

(c)

0 –20 –40 –60

0

M 1

2

3

5 10 15 30

(min)

Figure 11.3╇ Time course for the cardiovascular and respiratory effects of three different doses of

cocaine: 0.312, 1.25, and 5 mg/kg (i.v.) depicted in (a) (N = 6), (b) (N = 5), and (c) (N = 4), respectively. The ordinate scale is percentage of change and the abscissa scale is time in minutes. M€= period during maximal pressor response. Squares represent mean values for arterial pressure. Vertical lines represent standard error of the mean. All animals were anesthetized with pentobarbital (6.5 mg/kg, i.p.). (From Pitts, D.K., Udom, C.E., and Marwah, J., Life Sci., 40, 1099–1111, 1987.)

174 Sudden Death in Epilepsy: Forensic and Clinical Issues

were increased after a dose of 4 mg/kg or greater of cocaine i.v., with arrhythmias occurring with 4 and 8 mg/kg. Doses as low as 0.25 mg/kg (i.v.) evoked substantial cardiovascular responses and lethal responses of apnea. In another study, the administration of cocaine intravenously to conscious rats increased arterial blood pressure (Rockhold et al. 1987). The heart rate was elevated initially but subsequently was decreased. With the onset of cocaine-induced seizures, a further elevation in heart rate and blood pressure occurred, ultimately progressing to cardiovascular collapse and death. Preliminary studies utilizing intravenous administration of cocaine to anesthetized dogs elicited a dose-dependent increase in blood pressure and heart rate and alterations in the ST segment (Tackett and Jones 1987). These changes were associated with elevated cerebrospinal fluid levels of norepinephrine and dopamine, ἀndings that suggest a role for central catecholaminergic mechanisms in the cardiovascular actions of cocaine. Therefore, as cocaine raises the blood pressure and heart rate to excessively high levels, there is an increased risk of aneurism, arteriovenous malformation, and stroke or hemorrhage from ruptures of cerebral arteries weakened by drug-related arteritis. 11.3.2â•…Cocaine-Induced Myocardial Ischemia, Infarction, Arrhythmia, and Cardiomyopathies There has been a recent and dramatic increase in cardiac abnormalities among cocaine users (Duke 1986; Wiener and Putnam 1987; Wiener et al. 1986) that has raised questions concerning the effect of cocaine on the cardiovascular system. Indeed, cocaine is clearly cardiotoxic, being temporally linked to myocardial ischemia, arrhythmias, and many cardiomyopathies. Cocaine use in the presence of preexisting coronary artery disease may predispose the individual to the development of angina, arrhythmias, or myocardial infarction (Coleman et al. 1982; Young and Glauber 1947). It is possible that a patient with hypercholesterolemia who is using cocaine may be further increasing the likelihood of coronary artery spasm (Rosendorff et al. 1981), leading to myocardial ischemia and necrosis. Numerous cases of suspected cocaine-induced myocardial ischemias and infarctions have been reported (Isner et al. 1985, 1986; Kassowsky and Lyon 1984; Mathias 1986; Rod and Zucker 1987; Rollingher et al. 1986; Schachne et al. 1984; Simpson and Edwards 1986). Simpson and Edwards (1986) reported a case of a 21-year-old man with a history of recreational intravenous cocaine abuse who developed chest pain within 1 min and cardiopulmonary collapse within 1 h after injection of cocaine. Postmortem ἀndings revealed severe coronary obstructive lesions and acute platelet thrombosis, with secondary chronic and acute myocardial ischemic lesions, focal endothelial injury, and platelet aggregations being observed. The author proposed that coronary artery spasm induced by cocaine caused the endothelial lesions and favored platelet adherence and aggregation. The chronic obstructions that were also found may have resulted from a similar mechanism. According to Weiss (1986), ἀxed coronary atherosclerotic lesions play a permissive role in the induction of coronary vasospasm. It has been proposed that the ability of both intrinsic atherosclerotic plaques and cocaine-induced norepinephrine uptake blockade increases local levels of catecholamine, producing coronary vasospasms. Furthermore, preexisting coronary artery disease sensitizes the vascular smooth muscle to norepinephrine-induced vasoconstriction, predisposing the cocaine user to life-threatening ischemia (Gould et al. 1985; Weiss 1986). Also, with chronic cocaine abuse, the excessive accumulation of norepinephrine may prime the myocardium for a fatal arrhythmia (Tazelaar et al. 1987).

Cocaine-Induced Seizures, Arrhythmias, and Sudden Death

175

Cocaine increases the local concentrations of catecholamines from blocked adrenergic nerve endings to other cell receptors by inhibiting the neuronal uptake of norepinephrine. Thus the adrenergic response in susceptible organs is increased, leading to the development of catecholamine supersensitivity (Benchimol et al. 1978; Trendelenburg 1968). Therefore, cocaine is capable of eliciting both an inhibitory and an excitatory response of sympathetically innervated structures to endogenous and exogenous catecholamines (Pitts et al. 1987). It has also been suggested that cocaine accentuates the action of norepinephrine on beta receptors in the heart (Nanji and Filipenko 1984) by increasing the concentration of norepinephrine at the synaptic cleft. Beta stimulation increases automaticity, heart rate, and the conduction velocity of the His–Purkinje system and decreases atrioventricular nodal refractoriness (Tazelaar et al. 1987). Tazelaar et al. (1987) characterized cocaine-induced pathophysiology in a postmortem study of 30 cocaine-related deaths. Morphologic characteristics of acute ischemia were observed in 93% of the cases. These involved the formation of myocardial contraction bands in association with polymorphonucleocytes in the initial 12–24 h; they were replaced with lymphocytes by 24–48 h. The formation of contraction bands and myocardial interstitial ἀbrosis may be one pathogenic mechanism for fatal arrhythmias. These contraction bands may also represent an anatomical route for reentrant mechanism, thus priming the heart for fatal arrhythmias. It is quite possible that arrhythmias may be generated by cocaine through a decrease in the refractoriness of the myocardial ἀbers, accumulation of excess norepinephrine, formation of contraction bands, and production of interstitial ἀbrosis. Cocaine causes myocardial ischemia by direct and indirect actions (Vitullo et al. 1987). Direct effects are the stimulation of the sinoatrial node, with an increase in heart rate, contractility, and wall tension. Indirectly, the effects can be due to the sympathetic vasoconstriction of the peripheral smooth muscle vasculature. Arterial vasoconstriction leads to an increase in afterload and blood pressure, which, in turn, increases the work that the heart must pump against. Both the direct and the indirect actions increase the oxygen consumption of the myocardium, with an ischemic event occurring when the demand for oxygen supersedes the supply (Tyau and Lathers 1988). Pasternack et al. (1985) reported three male patients in their middle to late thirties who were referred for coronary angiography after having angina pectoris and/or an acute myocardial infarction, coincident with an increase in the frequency of cocaine abuse. The onset of angina and acute myocardial infarctions may have been caused by a cocaineinduced potentiation of the activities of the central nervous system resulting in systemic hypertension and tachycardia. Isner et al. (1986) reported a temporal relationship between cocaine and cardiac sequelae in seven nonintravenous cocaine abusers. It was concluded that cocaine may precipitate fatal arrhythmias, myocarditis, acute infarctions, and possible sudden death in patients with either anatomically normal or abnormal coronary arteries. In these cases there was a temporal relationship between the administration of cocaine and the onset of a myocardial ischemia and/or infarction. Because of many other medical factors involved, it is difficult to discern the actual etiological mechanisms of the infarctions. However, in a case report by Howard et al. (1985), a young woman with normal coronary arteries, blood glucose, and lipid levels and no history of cardiovascular disease or smoking was admitted to the hospital for loss of consciousness and epigastric pain; 5€h earlier, she had inhaled 1.5 g of cocaine. On admission, an ECG showed precordial ST segment elevation and a loss of R waves. On the day following admission, an echocardiography revealed akinesis and dyskinesis of the left ventricle apex and septum. In this

176 Sudden Death in Epilepsy: Forensic and Clinical Issues

healthy individual with no coronary risk factors or demonstrable coronary artery disease, infarction occurred. This ἀnding suggests that cocaine-induced myocardial infarctions should be considered when examining individuals who may not appear to be vulnerable. The cardiovascular events produced through cocaine abuse can be seen to involve a range of pathological responses, including coronary vasospasm, arrhythmia, myocardial ischeÂ� mia, infarction, and cardiomyopathies. Therapeutic use of cocaine is not without risk. For example, Chiu et al. (1986) reported a patient who was anesthesized for a closed reduction of a nasal fracture by spraying 2 ml of 1% cocaine solution into the nasal airways. The patient complained of an acute onset of chest pain and shortness of breath. An electrocardiogram indicated ST-T wave changes in the precordial leads, suggestive of an acute coronary ischemic event. The rise in the MB creatine kinase fraction and reversed LDH isoenzyme fractional values were consistent with a small nontransmural myocardial infarction. The published accounts of cardiovascular events, myocardial infarction, and mortality related to cocaine use as described here and consist mostly of case reports (Loveys 1987). Experimental research looking into the cardiovascular effects of cocaine is warranted. 11.3.3â•…Cocaine-Induced Changes in Postganglionic Cardiac Sympathetic Neural Function Cocaine potentiates the ganglionic blocking action of norepinephrine (Christ et al. 1982). In the isolated hamster stellate ganglia preparation, cocaine exaggerated the inhibitory action of exogenously applied norepinephrine. In pithed rats, cocaine potentiated the pressor effect of norepinephrine more than it potentiated the pressor effect of sympathetic stimulation (Bayorh et al. 1983). Cocaine increased plasma norepinephrine levels and extended the inotropic and chronotropic responses to sympathetic neural stimulation in anesthetized dogs (Matsuda et al. 1980). These actions were attributed to the inhibition of the neuronal catecholamine uptake by cocaine (Matsuda et al. 1980). It is possible that cocaine-induced exaggeration of sympathetic discharge may enhance the arrhythmias experimentally caused by ouabain (Lathers et al. 1977), coronary occlusion (Lathers et al. 1978), and seizures (Lathers and Schraeder 1982; Schraeder and Lathers 1983). Experimental arrhythmias are hypothesized to be a useful model to study possible mechanisms of sudden death associated with myocardial infarctions (Lathers et al. 1986) and epilepsy (Lathers and Schraeder 1987; Schraeder and Lathers 1989). Consequently, any changes induced by cocaine in cardiac sympathetic neural discharge may well augment the development of arrhythmias and/or sudden death. On the other hand, Dart et al. (1983) demonstrated that stimulation of postganglionic cardiac sympathetic nerves in a Langendorff rat–isolated heart preparation produced a stimulation frequency-dependent overflow of endogenous norepinephrine into the venous effluent with an increase in the heart rate. Cocaine signiἀcantly reduced the norepinephrine outflow while the heart rate continued to increase. 11.3.4â•…Central Actions of Cocaine Many of the effects of cocaine result from actions in the central nervous system. Cocaineinduced euphoria, for example, was among the ἀrst effects described (Freud 1884) and is the most well-known central effect. Generalized convulsions, which often precede

Cocaine-Induced Seizures, Arrhythmias, and Sudden Death

177

cocaine-induced death, unfortunately are less well-known outside the medical community. It has been shown that seizures are a major determinant of cocaine-induced death (Catravas and Waters 1981; Catravas et al. 1978). The common concomitants of generalized seizures, including hyperthermia, acidosis, increased blood pressure, cardiac arrhythmia, and hypoventilation, may be responsible for the lethality. In animal studies, many of the effects of cocaine have been localized to limbic structures. For example, Castellani et al. (1983) found that cocaine initiated high-voltage spindles that began in the amygdala 5–25 s after the injection of cocaine (Figure 11.4) and spread within seconds to other olfactory sites in synchronous bursts. Cocaine induced an increase in spindle frequency in the olfactory bulb and amygdala that was inhibited by atropine administration. Yasuda et al. (1984) reported that low concentrations of cocaine potentiated a norepinephrine-induced increase in spike amplitude of hippocampal splices. The authors concluded that the action was due to the inhibition of catecholamine uptake based on the observation that other inhibitors had the same effect. Lesse and Collins (1979) found that cocaine increased the speed at which epileptiform discharges spread to the amygdala and hippocampus. They postulated that subconvulsive doses of cocaine have an excitatory effect on limbic structures, which increases their sensitivity to repetitive discharges from distant foci. Matsuzaki (1978) reported that chronic high doses of cocaine in the rhesus monkey engendered persistent behavioral depression, with cortical and limbic slowing of EEG. They concluded that it was the action of cocaine on limbic structures that played an important role in the persistence of these effects. Overall, the evidence indicates that cocaine enhances the propagation of limbic seizures. Since such activity has been associated with cardiac arrhythmia (Lathers and Schraeder 1982; Schraeder and Lathers 1983), it is quite possible that cocaine-induced seizures could be a factor in the deaths of persons using the drug. The cortical EEG effects of cocaine in humans were among the earliest documented effects of the drug (Berger 1937). Cocaine increases power in the fast frequency (beta bands) of the resting EEG after subcutaneous, intravenous, or oral administration (Berger 1937; Herning et al. 1985). Four-hour intravenous infusions of high doses of cocaine in humans sustains the increase in EEG beta power (Pickworth et al. 1986). The increase of power in

1 min after injection

7th week after cocaine initiation

µV/mm 15

L OLF BULB

7.5

R OLF BULB

7.5

L AMYG

15

R AMYG

10

L OLF TUB

7.5

L ACCUMB 1s

10 mm

Figure 11.4╇ Electrographic amygdala–olfactory spindling and spike response to 5 mg/kg (i.v.) cocaine during preseizure behaviors.

178 Sudden Death in Epilepsy: Forensic and Clinical Issues

the fast EEG frequency bands is ordinarily associated with increased attention, vigilance, or arousal. Pickworth et al. (1986) measured subjective, cardiovascular, and EEG effects of large (60 mg) intravenous doses of cocaine in human volunteers. Although the subjective report of “rush” lasted for only a few moments, the pressor effect and tachycardia persisted for up to 60 min (Figure 11.5). The rush, or intense cocaine-induced euphoria, is the effect for which the drug is self-administered. It is quite probable that inadvertent overdosage may occur when subjects readminister cocaine at a time when the cardiovascular and central nervous systems are at jeopardy. In reviewing the effect of cocaine on the electrophysiology of the central monoaminergic neurons, Pitts and Marwah (1986) found that intravenous cocaine activated cerebellar Purkinje neurons and inhibited serotonergic dorsal raphe and noradrenergic locus coeruÂ� leus neurons. The authors concluded that cocaine-induced increases in the mean arterial blood pressure were correlated with changes in the discharge of the central neurons. Pitts and Marwah (1987) also found that reserpine pretreatment diminished the inhibitory effects of intravenous cocaine on neuronal discharges in the locus coeruleus and dorsal raphe as well as the excitatory action of cocaine on the cerebellar Purkinje neurons. Thus, although stimulation of the inhibitory locus coeruleus afferent input to the cerebellar Purkinje neurons can reduce the activity of the Purkinje neurons via a betaadrenoceptor mechanism, intravenous cocaine (1 mg/kg) did not precipitate the inhibitory actions of locus coeruleus stimulation on cerebellar Purkinje neurons. This dose of cocaine also did not potentiate the inhibitory effects of iontophoretically applied norepinephrine or GABA on cerebellar Purkinje neurons. Locus coeruleous neurons were inhibited by intravenous cocaine (1 mg/kg) in conscious animals paralyzed with gallamine. It was proposed that intravenous cocaine (1 mg/kg) reduced impulse flow in locus coeruleus neurons, possibly through an alpha2-autoreceptor mechanism, without augmenting the effect

Systolic pressure (mm Hg)

4

120

160

105

150

90

140

2

75

130

1

60

120

0

Heart rate (beats/min) “Rush”

–30

–15

0 Cocaine

15

30

45

60

75

3

90

105

120

135

150

Time (min)

Figure 11.5╇ Effects of intravenous cocaine (60 mg) in seven drug-experienced volunteers. The high dose caused a transient “rush” (drug-induced euphoria) but prolonged increases in blood pressure and heart rate.

Cocaine-Induced Seizures, Arrhythmias, and Sudden Death

179

of norepinephrine at the level of the noradrenergic terminals impinging on postsynaptic cerebellar Purkinje neurons. The action of cocaine on norepinephrine-containing locus coeruleus neurons was also evaluated in freely moving, unanesthetized cats (Trulson and Trulson 1987). Cocaine eÂ�licited a dose-dependent reduction in the activity of the neurons, which was suppressed by a prior administration of an alpha2 antagonist, piperoxane. Also, the activity of the locus coeruleus remained unchanged by the administration of the structurally related local anesthetic agent procaine. It was concluded that the local anesthetic actions of cocaine were not the inhibitory factor in its effect on the activity of norepinephrine-containing nÂ�eurons. Nevertheless, as Yasuda et al. (1984) stated, “It is remarkable that its effects upon the electrophysiological activity of the brain remain virtually unknown.” Whether cocaineiÂ�nduced changes in the activity of central neurons contributes to sudden death remains to be determined. 11.3.5â•…Cocaine-Induced Seizures Seizures have been shown to play an essential role in the pathophysiology of cocaine toxicity. A major determinant of lethality in cocaine-treated dogs was the presence of seizures (Catravas and Waters 1981; Catravas et al. 1978). Cocaine infusions produced prolonged seizures that led to lactic acidosis and hyperthermia prior to death while cardiac output, systemic vascular resistance, respiration, and oxygenation were stable just prior to death. Seizures and death could have been prevented with pretreatment using diazepam, highdose chlorpromazine, or neuromuscular blockade with pancuronium (Antelman et al. 1981; Fekete and Borsy 1971). The use of diazepam is particularly important since it also counteracts the sympathomimetic action of cocaine on the heart. Treatment of acidosis alone did not prevent death, unless the animals were maintained in a hypothermic state. Jonsson et al. (1983) reported one patient who, as a result of cocaine intoxication, showed combined metabolic and respiratory acidosis consequent to seizures and hypoventilation. Improved ventilation and the administration of bicarbonate reversed the hypotension and accelerated idioventricular rhythm to sinus rhythm. Blood pH increased from 6.33 to normal and the pCO2 decreased from 70 to 46 mm Hg. Jonsson et al. concluded that respiratory arrest compounded the acidosis in patients intoxicated with cocaine and may have contributed signiἀcantly to their deaths. Acidosis has a particularly negative effect on myocardial contractility (Fabiato and Fabiato 1978; Spivey et al. 1985) and acidosis can heighten the effects of catecholamines on the heart (Ford et al. 1968; Lathers et al. 1988; Spivey et al. 1985), and thereby contribute to the initiation of arrhythmias by cocaine. Carbamazepine is an antiepileptic drug that seems to be particularly effective in treating limbic system seizures. In experimental studies, repeated high doses of cocaine produce a convulsive response classiἀed as pharmacologic kindling. Weiss et al. (1987) reported that chronic carbamazepine administration inhibited the development of lidocaine- or cocaine-kindled seizures and lethality. Chronic, but not acute, pretreatment with carbaÂ� mazepine inhibited the high-dose cocaine seizures. It was suggested that carbamazepine may interact with local anesthetic mechanisms mediating the progressive development of seizures and that the effects of this antiepileptic drug at the level of the sodium channels should be further explored since both carbamazepine and the local anesthetics are believed to interact at this site. Investigation of the mechanism responsible for this effect should be undertaken, as it may prove clinically useful in preventing cocaine toxicity.

180 Sudden Death in Epilepsy: Forensic and Clinical Issues

11.4â•…Treatment of Cocaine-Induced Arrhythmias and Seizures Treatment of cocaine toxicity must ultimately involve deconditioning therapy to reduce drug craving and drug-seeking behavior (Kumor et al. 1988). Tricyclic antidepressants, bromocriptine, amantadine, methylphenidate, and lithium may decrease cocaine selfmedication. The hypertension and tachycardia that follow administration of cocaine are mediated by both alpha-adrenergic and beta-adrenergic receptors to induce vasoconstriction and an increase in heart rate and cardiac output, respectively (Olsen et al. 1983). One clinical management procedure of the adrenergic cocaine crisis involves the judicious use of intravenous propranolol, given in doses of 1 mg at 1-min intervals to a total of up to 6€mg (Gay 1982). Although intervention calms the excitable patient and decreases tachyÂ� arrhythmias, the efficacy of propranolol is limited by its receptor sensitivity. It has been argued that although propranolol effectively blocks beta receptors to decrease heart rate, it leaves the alpha-adrenergic receptors unopposed (Olsen et al. 1983). Thus stimulation of the alpha1-adrenergic receptors in the smooth muscle vasculature results in a worsening of vasoconstriction with resultant dangerous hypertension. Olsen et al. (1983) propose the use of phentolamine or nitroprusside to effect rapid vasodilatation. However, Gay (1983) argues that the nonselective alpha-adrenergic blockade properties of phentolamine may, in fact, further aggravate matters. Indeed, phentolamine will block alpha1 postsynaptic receptors to decrease vasoconstriction, but it will also block alpha2 presynaptic receptors. This blocks the normal regulatory control of the catecholamines, resulting in an increase in synthesis and output of norepinephrine, and may even spur a reflex sympathetic response to the heart. One agent recently used to treat cocaine toxicity is labetalol, which possesses both alpha- and beta-blocking capabilities. Thus the establishment of the alpha blockade counters the cocaine-induced vasoconstriction and hypertension while the beta blockade decreases the tachyarrhythmias (Gay and Loper 1988). The use of chlorpromazine and haloperidol to calm the hyperkinetic state is contraindicated in the cocaine user as they can lower seizure threshold activity and cause cardiac arrhythmias and/or sudden death (Lathers and Lipka 1986, 1987; Lipka and Lathers 1987). Instead, an effective means of quieting the stimulatory phase of cocaine intoxication is the use of diazepam, 15–20 mg orally every 8 h. Antelman et al. (1981) serendipitously found that amytriptiline, a tricyclic antidepressant, protected against sudden cardiac death due to cocaine intoxication in animals. Amytriptiline, 10 mg/kg, administered experimentally in animals 1 h before intraperitoneal injection of cocaine (35 mg/kg) resulted in no protection against sudden death. However, 24-h pretreatment with a single injection of amytriptiline markedly increased survival, while 10-day pretreatment conferred complete protection against sudden death. The mechanism of action, to date, has not been deἀned. However, pretreatment is not a useful tool in the management of clinical toxicity associated with cocaine use. It has recently been suggested that Ca2+ channel blockers may be a useful antidote for cocaine toxicity (Duke 1986; Mittleman and Welti 1987; Trouve and Nahas 1986). Ca2+ channel blockers inhibit the vasoconstrictive effects of norepinephrine by blocking the release of Ca2+ into the smooth muscle of the vasculature. Trouve and Nahas (1986) studied the cocaine antagonistic effects of nitrendepine, a Ca2+ channel blocker, in animals. Nitrendipine was selected for its lack of myocardial depressant activity and its ability to cause coronary vasodilatation. Nitrendipine (1.46 × 10−3 mg/kg/min) was concomitantly administered with 2 mg/ kg/min of cocaine and caused an inhibition of cocaine-induced tachycardia, pressor, and

Cocaine-Induced Seizures, Arrhythmias, and Sudden Death

181

vasoconstriction. Nitrendipine also suppressed the cocaine-induced arrhythmias observed in control animals. In a comparative study, nitrendipine and propranolol were able to slow cocaine-induced tachycardia while nitrendepine alone increased coronary flow and pulse pressure. Nitrendipine, alone and in combination with propranolol, decreased coronary flow and performance (Trouve and Nahas 1986). In addition to its cardioprotective properties, nitrendipine appears to possess central activity. Nitrendipine prevented motor tremors, convulsions, and seizures (Trouve and Nahas 1986). It was concluded that the sympatho� mimetic properties of cocaine can be antagonized by Ca2+ channel blockers. Ca2+ channel blockers may become the drugs of choice for the treatment of cocaine intoxication.

11.5â•…Use of Cocaine in Persons with Epilepsy Since cocaine use has been reported to produce hypertension, ventricular arrhythmias, tachycardia, myocardial infarction, seizures, and sudden death (Nahas et al. 1985; Tazelaar et al. 1987), and since persons with epilepsy have been shown to manifest autonomic dysfunction, including changes in blood pressure, heart rate, and rhythm (Leestma et al. 1984; Penἀeld and Erickson 1941; Phizackerly et al. 1954; Walsh et al. 1968), one must raise the question of whether the use of cocaine in individuals with epilepsy places them at risk of dying in a sudden, unexplained manner. Furthermore, dysfunction in the activity of peripheral cardiac autonomic neural discharge contributes to the production of cardiac arrhythmias (Gillis 1969; Gillis et al. 1972; Lathers et al. 1974, 1977, 1978; Verrier and Lown 1978; Weaver et al. 1976) and to sudden death (Lown and Verrier 1978). Lathers et al. (1977, 1978) and Lathers (1980) reported that nonuniform discharge in the cardiac postganglionic nerves, i.e., simultaneous increases and decreases in the various sympathetic branches innervating the myocardium, contributes to the production of arrhythmias by altering ventricular automaticity and excitability in the manner reported by Han and Moe (1964). Similar autonomic cardiac neural dysfunction was reported in association with arrhythmias and interictal and ictal discharges (Carnel et al. 1985; Lathers and Schraeder 1982, 1987; Lathers et al. 1984, 1987; Schraeder and Lathers 1983, 1988). Cocaine also modiἀes postganglionic cardiac sympathetic neural function, increasing the inotropic and chronotropic responses to sympathetic neural stimulation (Matsuda et al. 1980). Thus it is possible that the actions of cocaine and the autonomic cardiac neural dysfunction associated with epileptogenic activity may combine to produce cardiac arrhythmias and, at worst, sudden unexpected death. The question of whether cocaine use in the individual with epilepsy places the individual at risk for sudden death should be examined.

11.6â•…Summary This chapter has reviewed the incidence, characteristics, risk factors, and clinical management of cocaine-induced toxicity. Cocaine causes death by actions on the cardiovascular system, including cardiomyopathy, arrhythmia production, accelerated heart rate, and increased blood pressure. Seizures often accompany cocaine toxicity, leading to death. Cocaine is known to activate the EEG in humans, cause seizures in animals, and lower the seizure threshold. Patients with preexisting risk factors for cardiovascular pathology (high cholesterol, high blood pressure, cardiac arrhythmia, etc.) and those with epilepsy may

182 Sudden Death in Epilepsy: Forensic and Clinical Issues

be especially sensitive to cocaine-induced toxicity. Most research in animals suggests that cocaine-induced cardiovascular responses are due to enhanced noradrenergic response on the heart and arteriolar smooth muscles. While there is controversy surrounding the management of cocaine-induced toxicity, a symptomatic approach involves controlling the seizures with diazepam, the cardiovascular response with beta-adrenergic blockers or labetolol, a combined alpha- and beta-blocking agent, while correcting the systemic acidosis and hyperthermia. Use of the Ca2+ channel blockers may represent a new, more effective treatment. Finally, since both cocaine and epilepsy alone are associated with sudden unexpected death and since both are capable of modifying cardiac sympathetic neural discharge to produce changes in heart rate and rhythm, the question of whether the use of cocaine in the epileptic person places this individual at risk for sudden death must be raised.

References Abramowicz, M. 1986. Crack. Med Lett Drug Ther 28: 69–70. Amon, C. A., L. G. Tate, R. K. Wright, and W. Matusiak. 1986. Sudden death due to ingestion of cocaine. J Ann Toxicol 10: 217–218. Antelman, M., D. Kocan, N. Rowland, L. De Giovanni, and L. A. Chiodo. 1981. Amitriptyline provides long-lasting immunization against sudden cardiac death from cocaine. Eur J Pharmacol 69: 119–120. Bayorh, M. A., Z. Zukowska-Grojec, and I. J. Kopin. 1983. Effect of desipramine and cocaine on plasma norepinephrine and pressor responses to adrenergic stimulation in pithed rats. J Clin Pharmacol 23: 24–31. Benchimol, A., H. Bartall, and K. Desser. 1978. Accelerated ventricular rhythm and cocaine abuse. Ann Intern Med 88: 519–520. Berger, H. 1937. Hans Berger on the electroencephalogram of man. Arch Psychiatr Nervenk 106: 577–584. Translated in P. Gloor (Ed.). EEG Clin Neurophysiol Suppl 28: 291–297. Carnel, S. B., P. L. Schraeder, and C. M. Lathers. 1985. Effect of phenobarbital pretreatment on cardiac neural discharge and pentylenetetrazol-induced epileptogenic activity in the cat. Pharmacology 30: 225–240. Castellani, S., E. H. Ellinwood Jr., M. M. Kilbey et al. 1983. Cholinergic effects on arousal and cocaineinduced olfactory-amygdala spindling and seizures in cats. Sociol Behav 31: 461–466. Catravas, J. D., and I. W. Waters. 1981. Acute cocaine intoxication in the conscious dog: Studies on the mechanism of lethality. J Pharmacol Exp Ther 217: 350–356. Catravas, J. D., I. W. Waters, M. A. Walz et al. 1978. Acute cocaine intoxication in the conscious dog: Pathophysiologic proἀle of acute lethality. Arch Int Pharmacodyn 235: 328–340. Chiu, Y. C., K. Brecht, D. S. DasGupta, and E. Mhoon. 1986. Myocardial infarction with topical cocaine anesthesia for nasal surgery. Arch Otolaryngol Head Neck Surg 112: 988–990. Christ, D., J. Curry, and T. Zitaglio. 1982. Potentiation of the ganglionic blocking action of norepinephrine by cocaine. J Pharmacol Exp Ther 220: 97–101. Coleman, D. L., T. F. Ross, and J. L. Naughton. 1982. Myocardial ischemia and infarction related to recreational cocaine use. West J Med 136: 444–446. Cregler, L. L., and H. Mark. 1986a. Medical complications of cocaine abuse. New Engl J Med 315: 1495–1500. Cregler, L. L., and H. Mark. 1986b. Cardiovascular dangers of cocaine abuse. Am J Cardiol 57: 1185–1186. Cregler, L. L., and H. Mark. 1987. Cocaine-induced myocardial ischemia. Am J Med 82: 388. Dart, A. M., R. Dietz, W. Kubler, A. Schömig, and R. Strasser. 1983. Effects of cocaine and desipramine on the neurally evoked overflow of endogenous noradrenaline from the rat heart. Br J Pharmacol 79: 71–74.

Cocaine-Induced Seizures, Arrhythmias, and Sudden Death

183

Duke, M. 1986. Cocaine, myocardial infarction and arrhythmias—A review. Conn Med 50: 440–442. Estroff, T. W., and M. S. Gold. 1986. Medical and psychiatric complications of cocaine abuse with possible points of pharmacological treatment. In Advances in Alcohol Substance Abuse, Vol. 5, 61–68. New York, NY: Haworth Press. Fabiato, A., and F. Fabiato. 1978. Effects of pH on the myoἀlaments and the sarcoplasmic reticulum of skinned cells from cardiac and skeletal muscle. J Physiol 276: 233–255. Fekete, M., and J. Borsy. 1971. Chlorpromazine–cocaine antagonism: Its relation to changes of dopaÂ� mine metabolism in the brain. Eur J Pharmacol 16: 171–175. Fischman, M. W., C. R. Schuster, L. Resnekov et al. 1976. Cardiovascular and subjective effects of intravenous cocaine administration in humans. Arch Gen Psych 33: 983–989. Freud, S. 1884. Oncoca. Centralblatt für die Gesammte Theupie (Wien). Vol. 2, 289–314. Reprinted in the Cocaine Papers (S. Edminster, trans.). Vienna: Dunquin Press, 1963. Ford, G. D., W. H. Cline Jr., and W. W. Flemin. 1968. Influence of lactic acidosis on cardiovascular response to sympathomimetic amines. Am J Physiol 215: 1123–1129. Gay, G. R. 1982. Clinical management of acute and chronic cocaine poisoning. Ann Emerg Med 11: 562–572. Gay, G. R. 1983. Letter to the editor: In response to management of cocaine poisoning. Ann Emerg Med 10: 656–657. Gay, G. R., and K. A. Loper. 1988. Control of cocaine-induced hypertension with labetalol. Anesth Analg 67: 92. Gillis, R. A. 1969. Cardiac sympathetic nerve activity: Changes induced by ouabain and propranolol. Science 166: 508–510. Gillis, R. A., A. Raines, Y. L. Sohn, B. Levitt, and F. G. Standaert. 1972. Neuroexcitatory effects of digitalis and their role in the development of cardiac arrhythmias. J Pharmacol Exp Ther 183: 154–168. Gould, L., C. Gopalaswamy, C. Patel, and R. Betzu. 1985. Cocaine-induced myocardial infarction. N€Y State J Med, November, 660–661. Han, J., and G. K. Moe. 1964. Nonuniform recovery of excitability in ventricular muscle. Circ Res 14: 44–60. Herning, R. I., R. T. Jones, Hooker et al. 1985. Cocaine increases EEG beta: A replication and extension of Hans Berger’s historic experiments. EEG Clin Neurophysiol 60: 470–477. Howard, R. E., D. C. Hueter, and G. J. Davis. 1985. Acute myocardial infarction following cocaine abuse in a young woman with normal coronary arteries. J Am Med Assoc 254: 95–96. Isner, J. M., N. A. M. Estes, P. D. Thompson, Mark et al. 1985. Cardiac consequences of cocaine: Premature myocardial infarction, ventricular tachyarrhythmia, myocarditis, and sudden death. Circulation 72 (Suppl 3): 414. Isner, J. M., N. A. M. Estes, P. D. Thompson et al. 1986. Acute cardial events temporally related to cocaine abuse. New Engl J Med 315: 1438–1443. Jain, R., P. J. Gatti, M. Visner, K. G. Albrecht, M. G. Moront, C. E. Rackley, and R. A. Gillis. 1987. Effects of cocaine on cardiorespiratory function and on cardiovascular responses produced by bilateral carotid occlusion and IV norepinephrine. Fed Proc 46: 402. Javiad, J. I., M. W. Fishman, C. R. Schuster et al. 1978. Cocaine plasma concentration: Relation to physiological and subjective effects in humans. Science 202: 227–228. Jonsson, S., M. O’Meara, and J. B. Young. 1983. Acute cocaine poisoning. Am J Med 75: 1061–1064. Kassowsky, W. A., and A. F. Lyon. 1984. Cocaine and acute myocardial infarction. A probable connection. Chest 5: 729–731. Lathers, C. M. 1980. Effect of timolol on postganglionic cardiac and preganglionic splanchnic sympathetic neural discharge associated with ouabain-induced arrhythmia. Eur J Pharmacol 64: 95–106. Lathers, C. M., and L. J. Lipka. 1986. Chlorpromazine: Cardiac arrhythmogenicity in the cat. Life Sci 38: 521–538.

184 Sudden Death in Epilepsy: Forensic and Clinical Issues Lathers, C. M., and L. J. Lipka. 1987. Cardiac arrhythmia, sudden death and psychoactive agents. J Clin Pharmacol 27: 1–14. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy. Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23: 633–648. Lathers, C. M., and P. L. Schraeder. 1987. Review of autonomic dysfunction, cardiac arrhythmias, and epileptogenic activity. J Clin Pharmacol 27: 346–356. Lathers, C. M., N. Turner, and J. M. Schoffstall. 1989. The effect of different routes of sodium bicarbonate administration on plasma catecholamines, pH, and blood pressure during cardiac arrest in pigs. Resuscitation 18: 59–74. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1974. Relationship between the effect of ouabain on arrhythmia and interspike intervals (ISI) of cardiac accelerator nerves. Pharmacologist 16: 201. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1977. Correlation of ouabain-induced arrhythmia and nonuniformity in the histamine-evoked discharge of cardiac sympathetic nerves. J Pharmacol Exp Ther 203: 467–479. Lathers, C. M., G. J. Kelliher, J. Roberts, and A. B. Beasley. 1978. Nonuniform cardiac sympathetic nerve discharge: Mechanism for coronary occlusion and digitalis-induced arrhythmia. Circulation 57: 1058–1065. Lathers, C. M., W. H. Spivey, L. E. Suter, J. P. Lerner, N. Turner, and R. M. Levin. 1986. The effect of acute and chronic administration of timolol on cardiac sympathetic neural discharge, arrhythmias, and beta receptor density associated with coronary occlusion in the cat. Life Sci 39: 2121–2141. Lathers, C. M., P. L. Schraeder, and S. B. Cornel. 1984. Neural mechanisms in cardiac arrhythmias associated with epileptogenic activity: The effects of phenobarbital. Life Sci 34: 1919–1936. Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67: 247–259. Lathers, C. M., W. H. Spivey, and N. Turner. 1988. The effect of timolol given ἀve minutes post coronary occlusion on plasma catecholamines. J Clin Pharmacol 28: 289–299. Leestma, J. E., M. G. Kalelkar, S. S. Teas, G. W. Jay, and J. R. Hughes. 1984. Sudden unexpected death associated with seizures: Analysis of 66 cases. Epilepsia 25: 84–88. Lesse, H., and J. P. Collins. 1979. Effects of cocaine on propagation of limbic seizure activity. Pharmacol Biochem Behav 11: 689–694. Lichtenfeld, P. J., D. B. Rubin, and R. S. Feldman. 1984. Subarachnoid hemorrhage precipitated by cocaine snorting. Arch Neurol 41: 223–224. Lipka, L. J., and C. M. Lathers. 1987. Psychoactive agents, seizure production, and sudden death in epilepsy. J Clin Pharmacol 27: 169–183. Loveys, B. J. 1987. Physiologic effects of cocaine with particular reference to the cardiovascular system. Heart Lung 16: 175–181. Lown, B., and R. L. Verrier. 1978. Neural factors and sudden death. In Perspectives in Cardiovascular Research. Vol. 2. Neural Mechanisms in Cardiac Arrhythmias, ed. P. J. Schwartz, A. M. Brown, A. Malliani, and A. Zanchetti, 87–98. New York, NY: Raven Press. Mathias, D. W. 1986. Cocaine associated myocardial ischemia. Am J Med 81: 675–678. Matsuda, Y., Y. Masuda, B. Blattberg, and M. N. Levy. 1980. The effects of cocaine, chlorpheniramine and tripelennamine on the cardiac responses to sympathetic nerve stimulation. Eur J Pharmacol 63: 25–33. Matsuzaki, M. 1978. Alteration in pattern of EEG activities and convulsant effect of cocaine following chronic administration in the rhesus monkey. EEG Clin Neurophysiol 45: 1–15. Mittleman, R. E., and C. V. Welti. 1987. Cocaine and sudden “natural” death. J Forensic Sci 32: 11–19. Nahas, G., R. Trouve, J. F. Demus, and M. von Sitbon. 1985. A calcium-channel blocker as antidote to the cardiac effects of cocaine intoxication. New Engl J Med 313: 519–520.

Cocaine-Induced Seizures, Arrhythmias, and Sudden Death

185

Nanji, A. A., and J. D. Filipenko. 1984. Asystole and ventricular ἀbrillation associated with cocaine intoxication. Chest 85: 132–133. Olsen, K., N. Benowitz, and P. Pentel. 1983. Management of cocaine poisoning. Ann Emerg Med 10: 655–656. Pasternack, P. F., S. B. Colvin, and F. G. Baumann. 1985. Cocaine-induced angina pectoris and acute myocardial infarction in patients younger than 40 years. Am J Cardiol 55: 847. Penἀeld, W., and T. C. Erickson. 1941. Epilepsy and Cerebral Localizations, 320–362. Springἀeld, IL: Charles C. Thomas. Phizackerly, P. J. R., E. W. Poole, and C. W. M. Whitty. 1954. Sinoauricular heart block as an epileptic manifestation: A case report. Epilepsia 3: 89–91. Pickworth, W. B., R. I. Herning, K. Kumor, and M. Sherer. 1986. Spontaneous EEG during chronic cocaine infusion. Pharmacologist 28: 236. Pitts, D. K., and J. Marwah. 1986. Electro physiological effects of cocaine on central monoaminergic neurons. Eur J Pharmacol 131: 95–98. Pitts, D. K., and J. Marwah. 1987. Cocaine inhibits central monoaminergic neurons and activates cerebellar Purkinje neurons. Fed Proc 46: 400. Pitts, D. K., C. E. Udom, and J. Marwah. 1987. Cardiovascular effects of cocaine in anesthetized and conscious rats. Life Sci 40: 1099–1111. Ritchie, J. M., and N. M. Greene. 1980. Local anesthetics. In The Pharmacological Basis of Therapeutics, ed. L. S. Goodman and A. Gilman, 302–321. New York, NY: Macmillan. Roberts, D. C. S., M. E. Corcoran, and H. C. Fibiger. 1977. On the role of ascending catecholaminergic systems in intravenous self-administration of cocaine. Pharmacol Biochem Behav 6: 615–620. Rockhold, R. W., B. Hoskins, and I. K. Ho. 1987. Spontaneously hypertensive rats are resistant to convulsive and lethal actions of cocaine. Fed Proc 46: 402. Rod, J. L., and R. P. Zucker. 1987. Acute myocardial infarction shortly after cocaine inhalation. Am J Cardiol 59: 161. Rollingher, I. M., A. S. Belzberg, and I. L. MacDonald. 1986. Cocaine-induced myocardial infarction. Can Med Assoc 135: 45–46. Rosendorff, C., J. T. E. Hoffman, E. D. Verrier, J. Rouleau, and L. E. Boerboom. 1981. Cholesterol potentiates the coronary artery response to norepinephrine in anesthetized and conscious dogs. Circ Res 45: 320–329. Schachne, J. S., B. H. Roberts, and P. D. Thompson. 1984. Coronary-artery spasm and myocardial infarction associated with cocaine use. New Engl J Med 310: 1665–1666. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained sudden death. Life Set 32: 1371–1382. Schraeder, P. L., and C. M. Lathers. 1989. Paroxysmal cardiovascular dysfunction and epileptogenic activity. Epilepsy Res 3: 55–62. Simpson, R. W., and W. D. Edwards. 1986. Pathogenesis of cocaine-induced ischemic heart disease. Arch Pathol Lab Med 110: 479–484. Spivey, W. H., C. M. Lathers, D. R. Malone, H. D. Unger, S. Blat, R. M. McNamara, J. Schroffstall, and N. Turner. 1985. A comparison of intraosseous, central and peripheral routes of sodium bicarbonate administration during CPR in pigs. Arch Pathol Lab Med 14: 1135–1139. Tackett, R. L., and L. F. Jones. 1987. Central catecholaminergic changes and cardiovascular responses following acute administration of cocaine. Pharmacologist 29: 159. Tazelaar, H. D., S. B. Karch, B. G. Stephens, and M. E. Billingham. 1987. Cocaine and the heart. Hum Pathol 18: 195–199. Trendelenburg, U. 1968. The effect of cocaine on the pacemaker of isolated guinea-pig atria. J Pharmacol Exp Ther 161: 222–231. Trouve, R., and G. Nahas. 1986. Nitrendipine: An antidote to cardiac and lethal toxicity of cocaine. Proc Soc Exp Biol Med 183: 392–397. Trulson, T. J., and M. E. Trulson. 1987. Cocaine suppresses the activity of noradrenergic locus coeruÂ� leus neurons in freely moving cats. Pharmacologist 29: 159.

186 Sudden Death in Epilepsy: Forensic and Clinical Issues Tyau, L. S. Y., and C. M. Lathers. 1988. Cocaine-induced myocardial ischemia, infarction, and arrhythmias: A review of possible mechanisms causing sudden death. FASEB J 2: A1518. Van Dyke, C., and R. Byck. 1983. Cocaine use in man. Adv Subst Abuse 3: 1–24. Verrier, R. L., and B. Lown. 1978. Sympathetic–parasympathetic interactions and ventricular electrical stability. In Perspectives in Cardiovascular Research. Vol. 2. Neural Mechanisms in Cardiac Arrhythmias, ed. P. J. Schwartz, A. M. Brown, A. Malliani, and A. Zanchetti, 75–85. New York, NY: Raven Press. Vitullo, J. C., C. Lakios-Cherpas, and P. A. Khairallah. 1987. Effects of intraaortic injection of cocaine hydrochloride on rat electrocardiograms. Fed Proc 46: 1143. Walsh, G., W. Masland, and E. Goldenshon. 1968. Paroxysmal cerebral discharge associated with paroxysmal atrial tachycardia. Electroencephalogr Clin Neurophysiol 24: 187. Weaver, L. C., T. Akera, and T. M. Brody. 1976. Digoxin toxicity: Primary sites of drug action on the sympathetic nervous system. J Pharmacol Exp Ther 197: 1–9. Weiss, R. J. 1986. Recurrent myocardial infarction caused by cocaine abuse. Am Heart J 111: 793. Weiss, S. R. B., M. Costello, R. Woodward et al. 1987. Chronic carbamazepine inhibits the development of cocaine-kindled seizures. Abstr Soc Neurosci 13: 950. Welti, C. V., and D. A. Fishbain. 1985. Cocaine-induced psychosis and sudden death in recreational cocaine users. J Forensic Sci 30: 873–880. Wiener, M. D., and C. Putnam. 1987. Pain in the chest in a user of cocaine. J Am Med Assoc 258: 2087–2088. Wiener, R. S., J. T. Lockhart, and R. G. Schwartz. 1986. Dilated cardiomyopathy and cocaine abuse: Report of two cases. Am J Med 81: 699–701. Wilkerson, R. D. 1987. Yohimbine pretreatment enhances the cardiovascular actions of cocaine in anesthetized dogs. Fed Proc 46: 1143. Yasuda, R. P., N. R. Zahniser, and T. V. Dunwiddie. 1984. Electrophysiological effects of cocaine in the rat hippocampus in vitro. Neurosci Lett 45: 199–204. Young, D., and J. J. Glauber. 1947. Electrocardiographs changes resulting from acute cocaine intoxication. Am Heart J 34: 272–279.

Risk Factors for Sudden Death in Epilepsy Thaddeus s. Walczak

12

Contents 12.1 12.2 12.3 12.4 12.5

Introduction Epidemiologic Sources and Biases Traditional Risk Factors for SUDEP: Epilepsy Severity and Seizure Type SUDEP Is Not Conἀned to Severe Epilepsy Other Traditional Risk Factors 12.5.1 Age and Gender 12.5.2 Symptomatic Causes of Epilepsy 12.5.3 Sleep, Sleep Position, and Supervision during Sleep 12.5.4 Psychotropic Drugs 12.6 AED Use and SUDEP 12.6.1 AED Compliance and SUDEP 12.6.2 AED Polytherapy and SUDEP 12.6.3 Individual AEDs and SUDEP 12.7 Novel SUDEP Risk Factors: Recent Work 12.7.1 SCN1A Mutations 12.7.2 Heart Rate Variability and Cardiac Autonomic Instability 12.7.3 Anatomic and Electrophysiologic Substrates of SUDEP 12.7.3.1 Anatomic Substrates of Postictal Apnea 12.7.3.2 Anatomic Substrates of Cardiac Arrhythmia in SUDEP 12.7.3.3 Electrophysiologic Substrates of Cardiac Arrhythmia in€SUDEP 12.8 The Next Steps in the Study of SUDEP Risk Factors References

187 188 189 192 192 192 193 193 193 194 194 194 194 195 196 196 196 196 196 197 197 198

12.1â•…Introduction Understanding risk factors for sudden unexpected death in epilepsy (SUDEP) is important for both research-oriented and practical reasons. Current thinking regarding the pathophysiology and prevention of SUDEP remains largely speculative. Deἀning circumstances surrounding SUDEP and the patient and epilepsy characteristics associated with SUDEP could help direct investigations of pathophysiology. Risk factors can sometimes be modiἀed and could offer an approach to prevention, though we must remember that additional prospective studies with active intervention are necessary to demonstrate that modifying risk factors is effective. More practically, a consistent set of risk factors could deἀne a population that may be especially prone to SUDEP. This group can then be targeted for more detailed discussion of SUDEP and for any potential interventions. 187

188 Sudden Death in Epilepsy: Forensic and Clinical Issues

Growing interest in SUDEP over the past 25 years has prompted a series of investigations into the risk factors associated with this condition. This chapter attempts to review and synthesize this information. Such a synthesis is challenging because of differences in deἀnitions, study populations, and study design. However, the picture that emerges is remarkably consistent and corresponds well to what we are beginning to learn about pathophysiology.

12.2â•… Epidemiologic Sources and Biases Deἀning risk factors is initially a task for epidemiology. Selection and other biases associated with various epidemiologic approaches can signiἀcantly affect results. This is especially true when investigating an uncommon condition such as SUDEP. Different results in various studies may be due to the differing methodologies used. Before considering results, it is important to consider the biases associated with the various epidemiologic approaches used to determine SUDEP risk factors. Initial reports of SUDEP were uncontrolled case series collected at medical examiners’ offices (Leestma et al. 1989; Terrence et al. 1975; Freytag and Lindenberg 1964). These solidiἀed belief in SUDEP as an entity and provided an initial guess at risk factors. However, these populations are not representative of the great majority of people with epilepsy. Risk factor assessments based on this material are likely to be affected by selection bias, especially given the absence of control populations. Subsequent studies have evaluated SUDEP risk factors in more representative populations, including large prescription databases, cohorts of developmentally delayed persons with epilepsy, larger cohorts of persons with epilepsy at epilepsy centers, and drug and device development programs. These also do not represent the general population with epilepsy, though risk factors in SUDEP subjects are usually compared to nested controls in these studies, which reduces bias due to selection. A SUDEP population derived from a cohort of incident epilepsy would provide the best information on risk factors. However, such information is very difficult to obtain because SUDEP is rare in a population-based cohort with prevalent epilepsy (Ficker et al. 1998) and even rarer in a group with incident epilepsy (Lhatoo et al. 2001). Even if a reasonable number of cases were to be found in a large community-based cohort, it would be extremely difficult to assess circumstances of death and risk factors in retrospect. Less information is typically available about people with epilepsy drawn from a community; often the diagnosis of epilepsy is not secure. More information is usually available in groups followed in epilepsy clinics, where the diagnosis of epilepsy is probably more reliable. Most information regarding SUDEP risk factors is derived from retrospective SUDEP ascertainment. These studies often do not fully deἀne the cohort from which the deaths are drawn, so the population in which SUDEP is being described is often unclear. Further potential sources of bias include: 1. The possibility that not all deaths are ascertained 2. Difficulty in elucidating the circumstances of death several years later 3. Uncertainty about when risk factors for SUDEP and controls should be ascertained in the course of clinical follow-up 4. Difficulty in ascertaining risk factors in retrospect

Risk Factors for Sudden Death in Epilepsy

189

The general availability of mortality indices in developed countries allows reliable determination of whether a given individual is alive or not. However, determination of whether SUDEP occurred or not is very much dependent on a thorough understanding of the circumstances of death and this can be very difficult to reconstruct several years later. Deἀning a cohort and potential risk factors prospectively, assessing for death at regular intervals, and assessing for SUDEP promptly after death reduces these biases; however, such studies are more costly and challenging to perform as they need to utilize case investigators using a standardized set of questions to be asked of family and healthcare providers, as well as performance of autopsies for most, if not all, persons with a history of epilepsy who died. Choice of controls can also influence results. Some studies use people with epilepsy dying from causes other than SUDEP as controls. This is thought to provide information regarding circumstances of death (Tellez-Zenteno et al. 2005). However, mortality in the epilepsy population is mostly related to underlying causes of epilepsy. It is not clear that comparison of SUDEP deaths to epilepsy deaths (which are mostly related to the causes of epilepsy) adds much to understanding why SUDEP occurs. If the goal is determining which people with epilepsy are at risk for SUDEP, comparing risk factors in live people with epilepsy and SUDEP victims is more appropriate (Tomson et al. 2008). Finally, individual risk factors potentially associated with SUDEP may be related to other potential risk factors. An important example is seizure frequency, seizure severity, and anticonvulsant drug polytherapy, all of which may influence one another (Nilsson et al. 1999; Walczak et al. 2001; Langan et al. 2005). Establishing that a risk factor is independently responsible requires multivariate analysis, which, in turn, requires a reasonable number of cases and controls.

12.3â•…Traditional Risk Factors for SUDEP: Epilepsy Severity and Seizure Type Many potential risk factors have been examined (Table 12.1). Epilepsy severity and correlates such as epilepsy duration, seizure type, and seizure frequency have perhaps been most intensely studied. SUDEP risk has been consistently associated with more severe epilepsy, longer duration of epilepsy, and more frequent seizures. Table 12.2 summarizes SUDEP incidence in several populations with differing epilepsy severity. SUDEP incidence is very low in new onset (incident) epilepsy cohorts and somewhat increased in community-based prevalence cohorts. SUDEP incidence is higher in cohorts of persons with epilepsy at referral centers where more severe epilepsy cases may be expected to congregate. Incidence rates are still higher in drug and device development programs, which are usually limited to patients who have failed treatments with several antiepileptic drugs (AEDs). SUDEP incidence is highest in persons with epilepsy undergoing epilepsy surgery, where epilepsy is especially refractory. Furthermore, the percentage of all reported deaths due to SUDEP also increases with increasing epilepsy severity (Table 12.2). These two ἀndings establish a clear gradient of risk that strongly supports the idea that SUDEP risk increases with epilepsy severity. Most studies with live patient controls report that SUDEP is associated with higher seizure frequency (Table 12.1). The larger controlled studies (Nilsson et al. 1999; Langan et al. 2005; Walczak et al. 2001) all demonstrate progressively increased relative risk with

0

ns

20/80

154/616

62/124

+

0

18

0

0

0

ns

+

0

Male Sex

11/?

ns

0

14/1806

57/171

ns

11/20

Young Age

0

+

ns

+

0

ns

Epilepsy Duration

+

+

+

ns

+

+

ns

0

Frequent Seizures

ns

+

+

ns

+

+

0

ns

0

+

+

nsa

ns

ns

+

Mental Retardation

0

0

+

+

+

+

0

ns

AED Polytherapy at Time of Death

0

0

ns

ns

ns

ns

0

Lack of€super� vision at night, treatment with CBZ

Rx with€anti� psychotic drugs

Treatment with CBZ, lack of supervision at night Rx with€anti� psychotic drugs Nonambulatory status

Noncompliance with AED Treatment Other Risk Factors

Note: SUDEP, sudden unexpected death in epilepsy; AED, antiepileptic drug; CBZ, carbamazepine. 0, item was not a risk factor for SUDEP; +, item was a risk factor for SUDEP; ns, item was not studied. a All SUDEP cases had mental retardation in this study.

Hiltris et al. (2007)

Nilsson et al. (1999) McKee and Bodἀsh (2000) Tennis et al. (1995) Walczak et al. (2001) Langan et al. (2005)

Jick et al. (1992) Timmings (1993)

Cases/ Controls

Frequent Tonic– Clonic Seizures

Table 12.1â•…Risk Factors for SUDEP in Some Studies with Living Persons with Epilepsy as Controls

190 Sudden Death in Epilepsy: Forensic and Clinical Issues

Risk Factors for Sudden Death in Epilepsy

191

Table 12.2â•… SUDEP Incidence with Increasing Epilepsy Severity

Study Lhatoo et al. (2001) Ficker et al. (1998) Jick et al. (1992) Tennis et al. (1995) Timmings (1993) Lip and Brodie (1992) Leppik (1995) Leestma et al. (1997) Annegers et al. (2000) Sperling et al. (1999) Dashieff (1991)

Population

Incidence (per 1000 patient years)

Percentage of Deaths That Are SUDEP (%)a

Population-based cohort of incident epilepsy Population-based cohort of prevalent epilepsy Large prescription database Large prescription database Epilepsy clinic Epilepsy clinic

0.09

0.5

0.35

8.6

1.3 1.35 2.0 4.9

26 11 nr

Tiagabine clinical trial Lamotrigine clinical trial Vagus nerve stimulator clinical trial Patients who underwent epilepsy surgery Candidates for epilepsy surgery

3.9 3.5 4.1 4.0

29 40 52 54

10.0

Source: Leestma et al., Epilepsia, 38, 47–55, 1997. Note: nr, not reported. a SUDEP included deἀnite and probable SUDEP cases.

higher seizure frequencies. When multivariate models include generalized tonic–clonic seizures, other seizure types, and AEDs, SUDEP risk appears associated with generalized tonic–clonic seizures rather than partial seizures (Walczak et al. 2001; Langan et al. 2005; Tomson et al. 2005). As few as three tonic–clonic seizures per year signiἀcantly increase SUDEP risk, compared to no tonic–clonic seizures; risk increases further with more frequent tonic–clonic seizures. Other evidence strongly supports the position that tonic–clonic seizures are an important risk factor for epilepsy. All studies assessing seizure type report a history of tonic–clonic seizures in at least 90% of SUDEP cases. These ἀndings are consistent no matter what the study design or source population (Hirsch and Martin 1971; Terrence et al. 1975; Earnest et al. 1992; Leestma et al. 1989; Kloster and Torstein 1999; Walczak et al. 2001; Timmings 1993; Langan et al. 2005). Furthermore, studies addressing circumstances of death report tonic–clonic seizure prior to death in most cases (Terrence et al. 1975; Leestma et al. 1989; Earnest et al. 1992; Nillson 1999; Langan et al. 2000; Opeskin and Berkovic 2003). The consistency of these ἀndings indicates that tonic–clonic seizures are an important proximate cause of SUDEP. The relationship between epilepsy duration and SUDEP risk is less clear. SUDEP did not occur in studies assessing outcome in the several years after an initial seizure (Beghi et al. 2005). When risk is stratiἀed by duration of epilepsy, a clinically and statistically signiἀcant risk is noted after 10 to 30 years of epilepsy (Walczak et al. 2001; Leestma et al. 1989). One case control study found increased SUDEP risk with longer epilepsy duration (Walczak et al. 2001) while another did not (Langan et al. 2005). Age of seizure onset is generally lower in SUDEP cases than in controls (Nillson 1999; Jick et al. 1992; Kloster and Torstein 1999). This also supports the idea that duration of epilepsy is an important risk factor.

192 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 12.3â•… Expected Yearly Number of SUDEP Cases in a Hypothetical City of 6,000,000 Inhabitantsa People with Epilepsy

SUDEP Rates (patient years)

Number of SUDEP Cases

Percentage of All SUDEP Cases (%)

24,000

0.5/1000 patient years

12

50

6,000 30,000

2.0/1000 patient years

12 24

50

Well-controlled epilepsy Intractable epilepsy Total a

See Section 12.4 for details.

12.4╇SUDEP Is Not Confined to Severe Epilepsy This information relating higher SUDEP risk to increased seizure severity is leading to the impression that SUDEP is conἀned to people with severe epilepsy. Population-based medical examiner series and clinical experience clearly indicate that this is not the case. SUDEP also occurs in people with what are generally considered to be well-controlled seizures (Opeskin and Berkovic 2003). In fact, a SUDEP presenting to a random physician is probably equally likely to occur in a person with well-controlled seizures as in a person with poorly controlled seizures. This is because, on a population basis, a large majority of people with epilepsy have reasonable seizure control. A thought experiment illustrates why this is the case (Table 12.3). Consider a large city with 6 million inhabitants. Assuming an epilepsy prevalence of 0.5%, we would expect the city to contain 30,000 individuals with epilepsy. Let us further assume that 80% of the individuals with epilepsy have well-controlled seizures and a SUDEP incidence of 0.5/1000 patient years. Of the individuals with epilepsy, 20% have poorly controlled seizures and a SUDEP risk of 2.0/1000 patient years. With these assumptions (Hauser and Hesdorffer 1990, Table 2), we would expect 12 SUDEP cases per year in the 24,000 individuals with well-controlled seizures and 12 SUDEP cases in the 6000 individuals with poorly controlled seizures. Thus, a SUDEP case presenting to a random health care worker in this city is as likely to have had well-controlled seizures as poorly controlled seizures. General practitioners would be more likely to encounter the SUDEP cases whose seizures had been well-controlled because most people with well-controlled epilepsy are followed by community physicians.

12.5â•…Other Traditional Risk Factors 12.5.1╇Age and Gender Initial uncontrolled case series described SUDEP as a phenomenon found in young men with excessive alcohol use (Leestma et al. 1989; Terrence et al. 1975). This constellation of risk factors appears to have reflected the cases typically referred to medical examiners’ offices and has not been found in controlled studies. Mean age at death in most studies is between 25 and 39 years (see Tomson et al. 2005 for review). Somewhat higher ages were found in the large cohort-based studies (Walczak et al. 2001; Langan et al. 2005; Nillson 1999). Neither population-based studies (Ficker et al. 1998) nor the large case control studies found a male predominance. In fact, two controlled studies (Walczak et al. 2001; Opeskin

Risk Factors for Sudden Death in Epilepsy

193

and Berkovic 2003) found increased incidence of SUDEP among women. Controlled series have not found evidence for increased alcohol abuse among victims of SUDEP (Nilsson et al. 1999; Langan et al. 2005; Opeskin and Berkovic 2003; Kloster and Torstein 1999). However, SUDEP risk appears lower among children than adults. SUDEP incidence in four pediatric studies ranged between 0.11 and 0.43/1000 patient years (Harvey et al. 1993; Donner et al. 2001; Weber et al. 2005; Camἀeld et al. 2002). A prospective cohort study supports this: SUDEP incidence was 0/3625 patient years (0/1000) in those aged 0–19 years, 13/7792 (1.67/1000) in those aged 20–39 years, and 7/3958 patient years (1.77/1000) in those aged 40–59 years (Walczak et al. 2001). Children with the rare and severe myoclonic epilepsy of infancy are an exception; unusually high rates of SUDEP have been reported in this condition (Dravet et al. 2005). 12.5.2â•… Symptomatic Causes of Epilepsy Early series also found an apparent excess of epileptogenic lesions among SUDEP victims (Leestma et al. 1989; Terrence et al. 1975). This ἀnding also appears to be related to selection bias. Controlled clinical (Walczak et al. 2001) and medical examiner’s series (Kloster and Torstein 1999; Opeskin and Berkovic 2003) have not found this association. Controlled studies have not found either symptomatic epilepsy (Nillson 1999; Kloster and Torstein 1999) or any particular epilepsy syndrome (Walczak et al. 2001; Opeskin and Berkovic 2003) to be more common among SUDEP victims either. However, there is some evidence that mental retardation is more common among SUDEP victims (Jick et al. 1992; Walczak et al. 2001) and the inability to ambulate may confer further risk in mentally retarded people with epilepsy (McKee and Bodἀsh 2000). Nonetheless, most SUDEP cases are not afflicted with mental retardation or cerebral palsy. 12.5.3â•… Sleep, Sleep Position, and Supervision during Sleep SUDEP victims are found dead in bed in the majority of cases (Leestma et al. 1989; Terrence et al. 1975; Opeskin and Berkovic 2003; Kloster and Torstein 1999; Nashef et al. 1995), suggesting that sleep increases SUDEP risk. Most patients found dead in bed are found prone in both controlled and uncontrolled studies (Leestma et al. 1989; Kloster and Torstein 1999). Supervision at night (deἀned here as the presence in the bedroom of a responsible individual of normal intelligence), the use of a listening device, and regular checks throughout the night were associated with lower SUDEP rates in a large, retrospective, controlled study (Langan et al. 2005). These ἀndings are consistent with the idea that SUDEP related to postictal apnea is more likely to be lethal when appropriate positioning and ἀrst aid does not occur following a nocturnal seizure. Please see the chapter by Drs. Sato and Hughes discussing SUDEP and sleep in this book. 12.5.4â•… Psychotropic Drugs Antipsychotic drugs are associated with increased risk of sudden death (Ray et al. 2009) and treatment with selective serotonin reuptake inhibitors decreases SUDEP risk in an animal model of SUDEP (Tupal and Faingold 2006). Use of psychotropic medications is probably more common in persons with epilepsy than in the general population. This raises the question of whether use of psychotropic drugs affects SUDEP risk. Several controlled

194 Sudden Death in Epilepsy: Forensic and Clinical Issues

studies have found that the use of psychotropic drugs in general is not more common in SUDEP victims than in control persons with epilepsy (Tennis et al. 1995; Nilsson et al. 1999; Walczak et al. 2001; Langan et al. 2005; Opeskin and Berkovic 2003). Anxiolytic drugs speciἀcally were found to be more common in SUDEP in one study (Nilsson et al. 1999) and less common in SUDEP in another (Opeskin and Berkovic 2003). Antipsychotic use was speciἀcally examined in one study (Nillson 1999) and was not associated with SUDEP. The association between clinical use of selective serotonin reuptake inhibitors and SUDEP has not been examined.

12.6â•…A ED Use and SUDEP 12.6.1â•…A ED Compliance and SUDEP Initial medical examiner series found that most SUDEP victims had subtherapeutic AED levels (Leestma et al. 1989; Terrence et al. 1975; Kloster and Torstein 1999). This led to the conclusion that noncompliance was associated with SUDEP. Controlled medical examiner series have not consistently found subtherapeutic AED levels in SUDEP cases (George and Davis 1998; Opeskin et al. 1999). Furthermore, antiepileptic drug metabolism continues after death so postmortem AED levels may not accurately reflect compliance (see Walczak 2003 for review). Comparison of AED levels at antemortem visits in SUDEP and control patients has not found evidence of decreased noncompliance in SUDEP patients (Walczak et al. 2001; Nilsson et al. 2001; Langan et al. 2005). Overall, noncompliance with antepileptic drug treatment does not appear to be an important risk factor for SUDEP. 12.6.2â•…A ED Polytherapy and SUDEP AED polytherapy at death was noted to be common in initial studies (Leestma et al. 1989; Terrence et al. 1975; Tennis et al. 1995). This was initially thought to be a surrogate marker for severe epilepsy rather than an independent risk factor. However, two controlled studies found that treatment with more than two AEDs was associated with SUDEP, even after adjusting for the number of seizures and seizure type (Walczak et al. 2001; Nilsson et al. 1999). A third controlled study adjusting for seizure frequency and type did not ἀnd SUDEP to be associated with the number of AEDs at the time of death, but did ἀnd SUDEP to be associated with the number of AEDs ever used (Langan et al. 2005). Thus, it appears that AED polytherapy is associated with SUDEP, even after adjusting for seizure severity and frequency. The pathophysiologic implications of these ἀndings are unclear. AED polytherapy could have additive adverse effects on cardiac function (see Section 12.6.3). Alternatively, increased sedation associated with polytherapy could prolong recovery from the postictal state and cause greater susceptibility to the postictal central apnea and positional asphyxia that may play a role in SUDEP. 12.6.3â•…Individual AEDs and SUDEP This raises the question as to whether any individual AED or combination is associated with SUDEP. Carbamazepine use appears particularly likely to be associated with cardiac arrhythmia and alteration of cardiac autonomic function (see Walczak et al. 2003 for

Risk Factors for Sudden Death in Epilepsy

195

review). Consequently several reports have asked whether carbamazepine use is associated with SUDEP with mixed ἀndings. An uncontrolled series found what appeared to be a high rate of carbamazepine use in people succumbing to SUDEP (Timmings 1998). Four larger controlled series found that carbamazepine use was equally likely in SUDEP cases and in control subjects (Walczak et al. 2001; Nilsson et al. 1999; Kloster and Torstein 1999; Opeskin et al. 1999). A cohort based controlled study (Jick et al. 1992) found that carbaÂ� mazepine use was less likely in SUDEP cases. The largest SUDEP series published (Langan et al. 2005) found that SUDEP was associated with current carbamazepine use (odds ratio 2.0, 95% conἀdence interval 1.1–3.8), even after adjustment for potential confounders. This series of 154 SUDEP cases had the highest power to detect association between individual AED use and SUDEP. However, this was not a cohort-based study; cases and controls were drawn from differing sources, allowing an introduction of bias. Further analyses have asked whether carbamazepine toxicity is associated with SUDEP. Two studies have not found association between carbamazepine toxicity and SUDEP (Walczak et al. 2001; Opeskin et al. 1999). However, a more detailed analysis (Nilsson et al. 2001) found that SUDEP risk was increased more than nine-fold with toxic carbamazepine concentrations at the time of last visit, even after adjusting for confounders. Risk was further increased after adjusting for number of AED dose changes in the last year. SUDEP risk was also increased nine-fold with low carbamazepine concentrations at last visit, but only when more than one AED dose change had been made in the last year. In contrast, SUDEP risk was not increased with therapeutic carbamazepine concentrations, irrespective of how many AED dose changes had been made. The authors concluded that frequent changes of carbamazepine dose with concentrations outside therapeutic range were an independent risk factor for SUDEP after adjustment for seizure severity. This would go along with the idea that abrupt large fluctuations in carbamazepine levels exacerbate cardiac autonomic instability in people with epilepsy and increase SUDEP risk. The occurrence of SUDEP with other epilepsy treatments has not been thoroughly examined. SUDEP may be less frequent with phenytoin use than with carbamazepine use (Nilsson et al. 2001); it is not clear whether this reflects a deleterious effect of carbaÂ� mazepine or a protective effect of phenytoin. Single retrospective studies have reported that SUDEP is less common when lamotrigine (Leestma et al. 1997) or the vagal nerve stimulator (Annegers et al. 2000) is used.

12.7â•…Novel SUDEP Risk Factors: Recent Work Several risk factors have been proposed based on preliminary observations. We discuss them because they illustrate the study of SUDEP risk factors related to pathophysiologic theories, rather than standard demographic variables. In general, two major theories of SUDEP pathogenesis have been proposed (Tomson et al. 2008). One holds that severe postÂ� ictal cerebral inhibition leads to postictal central apnea, which, together with obstructive apnea, leads to arrhythmia and death. Another theory holds that the signiἀcant adrenergic stimulation associated with frequent tonic–clonic seizures results in microscopic cardiac lesions such as subendocardial ἀbrosis or contraction band necrosis. These then act as potential foci for arrhythmia, perhaps triggered by adrenergic stimulation associated with further tonic–clonic seizures. Arrhythmia risk may already be increased in this population because of the cardiac autonomic abnormalities thought to be more common in people

196 Sudden Death in Epilepsy: Forensic and Clinical Issues

with epilepsy, as previously discussed. Some literature has begun to address risk factors based on these pathophysiologic theories. 12.7.1â•…SCN1A Mutations Two cases of SUDEP have been reported in a family with generalized epilepsy with febrile seizures and with a mutation in the sodium channel gene SCN1A (Hindocha et al. 2008). Mortality and SUDEP rates are increased in severe myoclonic epilepsy of infancy—an epilepsy syndrome also caused by an SCN1A mutation (Dravet et al. 2005). SCN1A is expressed in the heart. This has led to the idea that SCN1A mutations may predispose people with epilepsy to arrhythmia and SUDEP (Nashef et al. 2007). In principle, comparing the prevalence of SCN1A mutations in SUDEP victims and persons with epilepsy controls dying of other causes could answer this question. This approach could be extended to other genes predisposing to cardiac arrhythmias (Nashef et al. 2007). 12.7.2â•…Heart Rate Variability and Cardiac Autonomic Instability Heart rate variability is an indicator of cardiac autonomic function and can be easily assessed by a standardized analysis of R–R intervals during electrocardiography. Heart rate variability changes are known to be associated with sudden death in the general population. Heart rate variability changes are more common in people with epilepsy, though it is not clear whether the epilepsy itself, use of AEDs, or conditions comorbid with epilepsy are responsible (Tomson et al. 1998; Walczak 2003). This has led to the idea that altered heart rate variability may be a marker of SUDEP risk (Yuen and Sander 2004; DeGiorgio et al. 2008). Observers have further noted that omega-3 fatty acids reduce sudden cardiac deaths in healthy subjects and may therefore help prevent SUDEP (Yuen and Sander 2004). Preliminary studies suggest that omega-3 fatty acids normalize HRV in people with epilepsy (DeGiorgio et al. 2008). A large study examining whether omega-3 fatty acids can prevent SUDEP is being planned. 12.7.3â•…A natomic and Electrophysiologic Substrates of SUDEP 12.7.3.1â•…A natomic Substrates of Postictal Apnea If postictal central and obstructive apnea are responsible for SUDEP, anatomic features associated with obstructive sleep apnea (increased body mass index, increased neck circumference, nasal obstruction, decreased pharyngeal diameter, etc.) should be more common in people with SUDEP than in persons with epilepsy controls. This information does not appear in the literature, though such data should be easy to obtain from material in medical examiner series. Similarly, a person with epilepsy will adjust spontaneously in the postictal state to maintain an open airway. One would expect such spontaneous adjustments to be less common in people with cerebral palsy or other conditions limiting movement. This idea is supported by reports that SUDEP is more common in people with developmental delay (Walczak et al. 2001) and inability to ambulate (McKee and Bodἀsh 2000). 12.7.3.2â•…A natomic Substrates of Cardiac Arrhythmia in SUDEP If cardiac arrhythmia due to microscopic cardiac lesions and the adrenergic surge associated with a tonic–clonic seizure is responsible for SUDEP, microscopic subendocardial

Risk Factors for Sudden Death in Epilepsy

197

abnormalities or conduction abnormalities should be more common in SUDEP victims than in control persons with epilepsy. Information from small, controlled series indicates that subendocardial abnormalities may be more common in persons with epilepsy than in control subjects without epilepsy (P-Codrea Tigaran et al. 2005; Natelson 1998). However, prevalence of subendocardial or conduction abnormalities does not differ between people with epilepsy dying from SUDEP and those dying of other causes (Opeskin et al. 2000). Nonetheless, these are small studies with inadequate power to exclude a potential contribution from subtle cardiac lesions, so further study is warranted. 12.7.3.3â•…Electrophysiologic Substrates of Cardiac Arrhythmia in SUDEP If the hearts of persons with epilepsy are, in fact, more susceptible to arrhythmia, one would expect nonfatal arrhythmia to be more common in persons with epilepsy than in control populations. A controlled study of 24 to 48 hours of cardiac monitoring has not found this to be the case (Blumhardt et al. 1986). Long-term electrocardiographic recordings of small numbers of people with severe epilepsy, usually lasting for many months, have found periods of asystole generally thought to require intervention in 15% (Rugg-Gunn et al. 2004). However, these studies are not controlled; the incidence of asystole during prolonged recordings in people with chronic disease (or, for that matter, in healthy young men) is not known. This amalgam of information raises the question whether people with severe epilepsy should undergo long term electrocardiographic monitoring and whether asymptomatic arrhythmias found by such monitoring should be treated. Long-term cardiographic monitoring of a larger group of persons with epilepsy is currently underway (Cooper 2008). However, this appears to be an uncontrolled study so it may be difficult to determine the clinical relevance of the information obtained.

12.8â•…The Next Steps in the Study of SUDEP Risk Factors The studies reviewed here have established a reasonably consistent risk proἀle for SUDEP. Persons with epilepsy succumbing to SUDEP suffer from generalized tonic–clonic seizures, have longer durations of epilepsy, and are often treated with multiple AEDs. Other potential risk factors such as treatment with speciἀc classes of psychotropic drugs, or treatment with speciἀc AEDs deserve further exploration. Examining these potential risk factors will be challenging for two reasons. First, SUDEP is uncommon in the general population and persistent surveillance of large groups is required for detection. Second, any risks associated with putative risk factors will require adjustment for risks known to be associated with the traditional risk factors described above. It is clear that future research should move beyond the old approach of retrospective case accumulation with convenience controls in convenience populations. Analysis of what we have called traditional risk factors together with animal studies have led to reasonably supported pathophysiologic theories (Lathers 2010, Chapter 25; Lathers and Schraeder 2010, Chapter 28; Lathers and Levin 2010, Chapter 33; Alkadhi and Alzoubi 2010, Chapter 26; Bealer et al. 2010, Chapter 38; Faingold et al. 2010, Chapter 41; Stewart 2010, Chapter 39; Goodman et al. 2010, Chapter 40). We are now in a position to study novel risk factors that support or detract from those theories. The state of knowledge is such that prospective studies validating traditional risk factors can now be undertaken. This would require periodic standardized longitudinal

198 Sudden Death in Epilepsy: Forensic and Clinical Issues

surveillance for SUDEP in multiple centers but should be feasible with a well-organized approach. With a little more information, hypothesis-based intervention studies in populations at high risk should soon be feasible as well. Much work will need to be done to increase recognition of SUDEP and set up the core infrastructure for such studies (So et al. 2009). Given what we have learned already, such an investment is quite likely to lead to more complete understanding and, ultimately, prevention of this tragic condition.

References Annegers, J. F., S. P. Coan, W. A. Hauser, and J. Leestma. 2000. Epilepsy, vagal nerve stimulation by the NCP system, all-cause mortality, and sudden, unexpected, unexplained death. Epilepsia 41: 549–553. Beghi, E., M. Leone, and A. Solari. 2005. Mortality in patients with a ἀrst unprovoked seizure. Epilepsia 46 (11): 40–42. Blumhardt, L. D., P. E. Smith, and L. Owen. 1986. Electrocardiographic accompaniments of temporal lobe epileptic seizures. Lancet 1 (8489): 1051–1056. Camἀeld, C. S., P. R. Camἀeld, and P. J. Veugelers. 2002. Death in children with epilepsy: A populationbased study. Lancet 39: 1891–1895. Cooper, P. N. 2008. Cardiac arrhythmias and SUDEP—Commentary on Hindocha et al. Epilepsia 49: 366–367. Dashieff, R. M. 1991. Sudden unexpected death in epilepsy: A series from an epilepsy surgery program and speculation on the relationship to sudden cardiac death. J Clin Neurosci 8: 216–222. DeGiorgio, C. M., P. Miller, S. Meymandi, and J. A. Gombein. 2008. n-3 fatty acids (eicosapentanoic acid and docosahexanoic acids) in epilepsy and for the prevention of sudden death in epilepsy. Epilepsy Behav 13: 681–684. Donner, E. J., C. R. Smith, and O. C. Snead. 2001. Sudden unexplained death in children with epilepsy. Neurology 57: 430–434. Dravet, C., M. Bureau, H. Oguni, Y. Fukayama, and O. Cokar. 2005. Severe myoclonic epilepsy in infancy (Dravet syndrome). In Epileptic Syndromes in Infancy, Childhood and Adolescence, 4th ed, ed. J. Roger, M. Bureau, C. Dravet, P. Genton, C. A. Tassinari, and P. Wolf, 89–114. London: John Libbey. Earnest, M. P., G. E. Thomas, R. A. Eden, and K. F. Hossak. 1992. The sudden unexplained death syndrome in epilepsy: Demographic, clinical and postmortem features. Epilepsia 33: 310–316. Ficker, D. M., E. L. So, and W. K. Shen et al. 1998. Population based study of the incidence of sudden unexplained death in epilepsy. Neurology 51: 1270–1274. Freytag, E., and R. Lindenberg. 1964. 294 medicolegal autopsies in epileptics. Arch Pathol 78: 274–286. George, J. R., and G. G. Davis. 1998. Comparison of anti-epileptic drug levels in different cases of sudden death. J Forensic Sci 43: 598–603. Harvey, A. S., T. Nolan, and J. B. Carlin. 1993. Community based study of mortality in children with epilepsy. Epilepsia 34: 597–603. Hauser, W. A., and D. C. Hesdorffer. 1990. Incidence and prevalence. In Epilepsy: Frequency, Causes, and Consequences, 1–51. New York, NY: Demos Publications. Hiltris, N., S. Suratman, K. Kelly, L. J. Stephen, G. J. Sills, and M. J. Brodie. 2007. Sudden unexpected death in epilepsy: A search for risk factors. Epilepsy Behav 10: 138–141. Hindocha, N., L. Nashef, and F. Elmslie et al. 2008. Two cases of sudden unexpected death in epilepsy in a GEFS+ family with an SCN1A mutation. Epilepsia 49: 360–365. Hirsch, C. S., and D. L. Martin. 1971. Unexpected death in young epileptics. Neurology 21: 682–689. Jick, S. S., T. B. Cole, R. A. Mesher, P. Tennis, and H. S. Jick. 1992. Sudden unexpected death in young persons with primary epilepsy. Pharmacoepidemiol Drug Safety 1: 59–64.

Risk Factors for Sudden Death in Epilepsy

199

Kloster, R., and E. Torstein. 1999. Sudden unexpected death in epilepsy (SUDEP): A clinical perspective and a search for risk factors. J Neurol Neurosurg Psychiatry 67: 439–444. Langan, Y., L. Nashef, and J. W. A. S. Sander. 2000. Sudden unexpected death in epilepsy: A series of witnessed deaths. J Neurol Neurosurg Psychiatry 68: 211–213. Langan, Y., L. Nashef, and J. W. Sander. 2005. Case-control study of SUDEP. Neurology 64: 1131–1133. Leestma, J. E., T. W. Walczak, J. R. Hughes, M. B. Kalelkar, and S. S. Teas. 1989. A prospective study on sudden unexpected death in epilepsy. Ann Neurol 26: 195–203. Leestma, J. E., J. F. Annegers, and M. J. Brodie et al. 1997. Sudden unexplained death in epilepsy: Observations from a large clinical development program. Epilepsia 38: 47–55. Leppik, I. E. 1995. Tiagabine: The safety landscape. Epilepsia 36: (Sl6): S10–S13. Lhatoo, S. D., A. L. Johnson, D. M. Goodrich, B. K. MacDonald, J. W. Snyder, and S. D. Shorvon. 2001. Mortality in epilepsy in the ἀrst 11 to 14 years after diagnosis: Multivariate analysis of a long-term prospective, population-based cohort. Ann Neurol 49: 336–344. Lip, G. W., and M. J. Brodie. 1992. Sudden death in epilepsy: An avoidable outcome? J R Soc Med 85: 609–611. McKee, J. R., and J. W. Bodἀsh. 2000. Sudden unexpected death in epilepsy in adults with mental retardation. Am J Ment Retard 105: 229–235. Nashef, L., D. R. Fish, J. W. A. S. Sander, and S. D. Shorvon. 1995. Incidence of sudden unexpected death in an adult outpatient cohort with epilepsy at a tertiary referral center. J Neurol Neurosurg, Psychiatry 58: 462–464. Nashef, L., N. Hindocha, and A. Makoff. 2007. Risk factors in sudden death in epilepsy (SUDEP): The quest for mechanisms. Epilepsia 48: 859–871. Natelson, B. H., R. V. Suzrez, C. F. Terrence, and R. Turzio. 1998. Patients with epilepsy who die suddenly have cardiac disease. Arch Neurol 55: 857–60. Nilsson, L., U. Bergman, and V. Diwan. 2001. Antiepileptic drug therapy and its management in sudden unexpected death in epilepsy: A case-control study. Epilepsia 42: 667–673. Nilsson, L., B. Y. Farahmand, P. G. Persson, I. Thiblin, and T. Tomson. 1999. Risk factors for sudden unexpected death in epilepsy: A case control study. Lancet 35: 888–893. Opeskin, K., and S. F. Berkovic. 2003. Risk factors for sudden unexpected death in epilepsy: A controlled prospective study based on coroners cases. Seizure 12: 456–464. Opeskin, K., M. P. Burke, and S. M. Cordner. 1999. Comparison of antiepileptic drug levels in sudden unexpected deaths in epilepsy with deaths from other causes. Epilepsia 40: 1795–1798. Opeskin, K., A. Thomas, and S. F. Berkovic. 2000. Does cardiac conduction pathology contribute to sudden unexpected death in epilepsy? Epilepsy Research 40: 17–24. P-Codrea Tigaran, S., S. Dalanger-Pedersen, U. Baandrup, M. Dam, and A. Vesterby-Charles. 2005. Sudden unexpected death in epilepsy: Is death by seizures a cardiac disease? Am J Forensic Med Pathol 26: 99–105. Ray, W. A., C. P. Chung, K. Murray, K. Hall, and C. M. Stein. 2009. Atypical antipsychotic drugs and the risk of sudden cardiac death. N Engl J Med 360: 225–235. Rugg-Gunn, F. J., R. J. Simister, M. Squirrell, D. R. Holdright, and J. S. Duncan. 2004. Cardiac arrhythmias in focal epilepsy: A prospective long-term study. Lancet 364: 2212–2219. Sperling, M. R., H. Fedman, J. Kinman, J. D. Liporace, and M. J. O’Connor. 1999. Seizure control and mortality in epilepsy. Ann Neurol 46: 45–50. So, E. L., J. Bainbridge, J. R. Buchhalter, and J. Donalty et al. 2009. Report of the American Epilepsy Society and the Epilepsy Foundation Joint Task Force on Sudden Unexplained Death in Epilepsy. Epilepsia 50: 917–922. Tellez-Zenteno, J. F., L. H. Ronquillo, and S. Wiebe. 2005. Sudden unexpected death in epilepsy: Evidence-based analysis of incidence and risk factors. Epilepsy Res 65: 101–115. Tennis, P., T. B. Cole, J. F. Annegers, J. E. Leestma, M. McNutt, and A. Rajput. 1995. Cohort study of incidence of sudden unexplained death in persons with seizure disorder treated with antiepileptic drugs in Saskatchewan, Canada. Epilepsia 36: 29–36.

200 Sudden Death in Epilepsy: Forensic and Clinical Issues Terrence, C. F., H. M. Wiszotzkey, and J. A. Perper. 1975. Unexpected, unexplained death in epileptic patients. Neurology 25: 594–598. Timmings, P. L. 1993. Sudden unexpected death in epilepsy: A local audit. Seizure 2: 287–290. Timmings, P. L. 1998. Sudden unexpected death in epilepsy: Is carbamazepine implicated? Seizure 7: 289–291. Tomson, T., M. Ericson, C. Ihrman, and L. E. Lindblad. 1998. Heart rate variability in patients with epilepsy. Epilepsy Res 30: 77–83. Tomson, T., L. Nashef, and P. Ryvlin. 2008. Sudden unexpected death in epilepsy: Current knowledge and future directions. Lancet Neurol 7 (11): 1021–1031. Tomson, T., T. Walczak, M. Silanpaa, and J. W. A. S. Sander. 2005. Sudden unexpected death in epilepsy: A review of incidence and risk factors. Epilepsia 46: (S11): 54–61. Tupal, N. and C. L. Faingold. 2006. Evidence supporting a role of serotonin in modulation of sudden death induced by seizures in DBA/2 mice. Epilepsia 47: 21–26. Walczak, T. 2003. Do antiepileptic drugs play a role in sudden unexpected death in epilepsy? Drug Saf 26: 673–683. Walczak, T. S., I. E. Leppik, M. D’Amelio et al. 2001. Incidence and risk factors in sudden unexpected death in epilepsy: A prospective cohort study. Neurology 6: 51–55. Weber, P., R. Bubl, U. Blauenstein, B. U. Tillman, and J. Lutschg. 2005. Sudden unexplained death in children with epilepsy: A cohort study with an eighteen-year follow up. Acta Paediatr 94: 564–567. Yuen, A. W., and J. W. Sander. 2004. Is omega-3 fatty acid deἀciency a factor contributing to refractory seizures and SUDEP? A hypothesis. Seizure 13: 104–107.

EEG Findings in SUDEP Maromi Nei Nicole Simpkins

13

Contents 13.1 Introduction 13.2 EEG Data in Patients Who Subsequently Died due to SUDEP 13.3 EEG in SUDEP and Near-SUDEP 13.4 Conclusions References

201 201 202 205 206

13.1â•…Introduction Most of the data regarding the EEG during SUDEP are obtained from case reports from patients undergoing ambulatory EEG or video-EEG monitoring. These data are limited, but the EEG during SUDEP has generally shown a generalized tonic–clonic seizure with subsequent suppression. Ictal and interictal EEG recordings in patients with SUDEP have revealed varied ἀndings.

13.2â•…EEG Data in Patients Who Subsequently Died due to SUDEP In the majority of cases of witnessed SUDEP, a seizure is reported to precede death. In 12 of 15 witnessed cases of SUDEP, a generalized tonic–clonic seizure preceded death (Langan et al. 2000). In the other three cases in this series, either an aura occurred or the patient was thought to be in a postictal state. Since seizures usually precede death in SUDEP, this suggests that seizures are often responsible for triggering the physiologic changes that ultimately lead to SUDEP. Thus, the evaluation of the EEG and other physiologic data during seizures and interictally could yield clues regarding the pathophysiology of SUDEP. The video-EEG monitoring data in individuals who subsequently died due to SUDEP have revealed seizures of various localization and lateralization (Nei et al. 2004). In this study, video-EEG data from 21 patients who subsequently died due to SUDEP were reviewed and compared with video-EEG data from a control refractory focal epilepsy population. The majority (86%) of patients had ictal EEG recorded, in addition to interictal EEG data. Localization of seizure onset was conἀdently identiἀed as arising from the left hemisphere in 29% and from the right hemisphere in 14%. The remainder had seizures that were either nonlateralized (24%), multifocal (5%), or generalized (14%) in onset. Of these patients, 43% were thought to have temporal lobe onset for their seizures and 19% had a probable frontal lobe onset; however, these data are likely biased since the majority of patients were admitted as part of their evaluations for potential epilepsy surgery. No speciἀc lobe of onset or lateralization for seizures was more common in SUDEP than in control patients. 201

202 Sudden Death in Epilepsy: Forensic and Clinical Issues

There are also data to suggest that prolonged seizures, generalized tonic–clonic seizures, as well as seizure clusters, might increase the risk for increased autonomic stimulation during seizures and might increase the risk for cardiac arrhythmias (Nei et al. 2000, 2004). Unfortunately, there were no speciἀc EEG ἀndings that were predictive of SUDEP. Most patients with SUDEP die in their sleep. EEG data also suggest that patients with SUDEP are more likely to have a history of and/or documentation of seizures arising from sleep, as compared with control patients (Nei et al. 2004). Opherk et al. also found that in patients with refractory epilepsy (in a non-SUDEP population), there was a trend toward increased ictal-related EKG abnormalities during seizures associated with sleep, suggesting a potential speciἀc sleep-related risk on autonomic status during seizures arising at this time, which might increase risk for SUDEP. The reader is referred to the chapter on sleep and SUDEP by Drs. Sato and Hughes in this book. While it is clear that seizures can cause ictal and postictal EKG rate and repolarization abnormalities, the potential interictal effects of epileptiform abnormalities on cardiac and pulmonary function are not as clear. One study evaluated the effect of interictal epileptiform EEG discharges on the QTc interval of the EKG in patients who subsequently died due to SUDEP (Tavernor et al. 1996). In this study, the EEGs influence on EKG data from eleven patients with SUDEP were compared with data from 11 age and sex matched control patients, also with uncontrolled tonic–clonic seizures who were alive at the time of the investigation. They found that only for those with SUDEP, the QTc interval was signiἀcantly prolonged during epileptiform EEG discharges. This led to the speculation that prolonged QTc intervals might increase the likelihood for potentially lethal ventricular arrhythmias and sudden cardiac death. However, additional data are needed to conἀrm these ἀndings. No speciἀc information regarding the type or localization of the EEG epileptiform abnormalities is available from this study. Most epidemiologic studies on SUDEP have focused on seizure type, rather than speciἀc EEG ἀndings. Generalized tonic–clonic seizures increase the risk for SUDEP and these may be either primarily or secondarily generalized seizures (Nei et al. 2004).

13.3â•…EEG in SUDEP and Near-SUDEP There are few case reports of SUDEP or near SUDEP captured during EEG recording available in the literature (see Table 13.1). One case of SUDEP captured during video-EEG monitoring includes a 41-year-old woman with refractory focal epilepsy since infancy (Lee 1998). Interictally, she had independent bitemporal sharp waves, with left greater than right. During sleep, she had an unwitnessed secondarily generalized tonic–clonic seizure lasting 70 seconds, which was followed by diffuse slowing on the EEG and left temporal sharp waves for 40 seconds, then marked suppression, with overlying EKG artifact seen that failed to recover. The EKG initially showed bradycardia to 30 beats per minute (bpm), which slowly increased to 70 bpm ἀve minutes after the seizure. However, the heart rate then began to slow and stopped 18 minutes after the seizure. There was no evidence of cardiovascular nor pulmonary abnormalities at autopsy, and no evidence of asphyxia. The cause of SUDEP in this case was postulated to be cessation of brain function. McLean and Wimalaratna (2007) reported a case of a woman in her ἀfties who died following a seizure while undergoing ambulatory EEG monitoring. Interictally, she had slow and sharp wave discharges that increased in frequency during sleep. During sleep,

EEG Findings in SUDEP

203

Table 13.1â•… EEG and Seizure Data: Cases of SUDEP and Near-SUDEP Interictal EEG

Report

Death

Dasheiff and Dickinson (1986)

Yes

N/A

Bird et al. (1997)

Yes

Lee et al. (1998)

Yes

L Temp, R temp, R occipital Bitemporal sharp waves, L€>€R

McLean and Wimalaratna (2007) So et al. (2000)

Yes

Tavee and Morris (2008) Espinosa et al. (2009)

No

Slow and sharp discharge Bifrontal sharp waves

No

L temp

No

Bitemporal theta/delta

Postictal EEG

Sz Type Prior to Event

N/A

N/A

CPS

R medial temporal

R suppressed, GTC then bilateral suppression GTC Diffusely Slow; L temp sharp waves × 40 secs Diffuse EEG c/w No suppression GTC

EKG: pulse artifact at 48 bpm EKG:€brady� cardia at 30 bpm€post� ictally

Bifrontal

Diffusely slow, then suppressed

GTC

Yes: 4 GTC in 6 h

L hem, maximal temp R temp

Continuous slowing

GTC

Yes: 1 aura, 4 partial and 2 GTC No

Apnea postictally, then€brady� cardia and asystole Stridor/ resp distress€af�� ter sz ended EKG: VT, then VF

Ictal EEG

N/A

Spike-wave discharge

Diffusely slow GTC

Sz Cluster Prior to Event Yes: cluster of 2 GTC and 1 CPS within 3 h Yes: cluster of 5 GTCs in 24 h No

Notes EKG: VF



Note: sz, seizure; L, left; R, right; temp, temporal; N/A, not available; hem, hemisphere; secs, seconds; GTC, generalized tonic–clonic seizures (either primary or secondarily generalized); c/w, consistent with; CPS, complex partial seizure; bpm, beats per minute; VF, ventricular ἀbrillation; resp, respiratory; VT, ventricular tachycardia.

a 52-second seizure beginning with spike and wave discharges was captured, followed abruptly by marked suppression that failed to recover. Further details regarding localization of ictal and interictal discharges are not available. Rhythmic movement artifact was seen at the T3 electrode, with associated muscle activity that became less frequent and disappeared completely 3 minutes after seizure termination, leaving a suppressed EEG. No EKG or respiratory data were available. One case of SUDEP during intracranial EEG monitoring was reported by Bird et al. (1997). A 47-year-old man with refractory seizures since age 19 was admitted for videoEEG monitoring. He had undergone ambulatory scalp monitoring and had three complex partial seizures with secondary generalization that began with right hemispheric slow waves, but were otherwise nonlocalized. Interictally, there were complex left temporal discharges, right temporal spikes, and occasional right occipital spikes. He then underwent implantation of bilateral temporal depth electrodes and subdural electrodes of the anterior

204 Sudden Death in Epilepsy: Forensic and Clinical Issues

and posterior temporal regions for further evaluation of his seizures. The operation and immediate postoperative period were uneventful. He subsequently had four secondarily generalized seizures and died following the ἀfth seizure at 3 a .m. Electrographic onset in all ἀve seizures was in the right medial temporal lobe. The ἀfth seizure occurred during sleep and began with head version to the left followed by turning of his body, then a generalized convulsion lasting for 2.5 minutes. At the end of the convulsion, he no longer moved. The EEG showed right mesial temporal onset, with the electrographic discharge spreading to the left hemisphere after 15 seconds, and subsequent generalized discharge lasting 2.5 minutes. The right-sided ictal discharge then briefly flattened, alternating with spindling spike discharges for 16 seconds, before stopping completely, leaving a suppressed background on the right. The left hemisphere continued to show spikes for 8 more seconds, then stopped. Pulse artifact, at 46 beats per minute, was seen for another 2 minutes and then gradually decreased in amplitude (continuing at the same rate) until the heart beat stopped. There were no respiratory or EKG recordings. Postmortem examination revealed mild congestion in the lungs and normal cardiac examination. The neuropathological examination showed an acute infarct in the right temporal lobe, attributed to insertion of the depth electrodes, acute hypoxic changes in the right hippocampus, evidence of old frontal contusions bilaterally, and bilateral occipital ulegyria. Dasheiff and Dickinson (1986) reported a case of sudden unexpected death in a patient undergoing video-EEG monitoring with intracranial electrodes. The patient was a 48-yearold man with a history of prior myocardial infarction and refractory focal epilepsy who had been implanted with depth electrodes for seizure focus localization. He had two secondarily generalized tonic–clonic seizures within 1 hour of each other. After the second seizure, the patient complained of chest and left arm pain. An EKG revealed ST segment elevation and inverted T waves. One hour later, the patient complained of chest pain, then was witnessed to have a complex partial seizure, becoming cyanotic and apneic. The ἀrst available EKG revealed coarse ventricular ἀbrillation, followed by asystole. Cardiopulmonary resuscitation was ineffective. Postmortem examination revealed no acute coronary artery thromboses or pulmonary emboli but did reveal an old myocardial infarction. The EEG ἀndings were not reported. It is likely that the acute sympathetic discharge due to the cluster of seizures precipitated both the angina and ventricular arrhythmia, to which he was already predisposed, due to his underlying previous myocardial infarction. A similar mechanism has been proposed as resulting in myocardial infarction in the setting of seizure (Chin et al. 2004). It is of interest to speculate that a vasospasm mechanism such as that observed in Prinzmetal Angina could account for the ST elevation and cause infarction in the absence of atherosclerosis. A case of near SUDEP due to postictal severe laryngospasm was reported by Tavee and Morris (2008). A 42-year-old man with refractory epilepsy since age 6 was admitted for video-EEG monitoring. Interictally, he had frequent left sphenoidal electrode sharp waves and less frequent posterior temporal sharp waves. He had one simple partial seizure and 5 complex partial seizures, two of which secondarily generalized. Ictal EEG showed left hemisphere, maximal in the temporal region, ictal fast activity. During seizure six, the patient awakened from sleep with right arm and face twitching, followed by right arm extension, left arm flexion, head version to the right and a generalized convulsion. The entire seizure lasted 82 seconds. Following the seizure, he developed inspiratory stridor and marked cyanosis, eventually requiring intubation for respiratory support. The anesthesiologist noted laryngospasm at the time of intubation. It was postulated that aspiraÂ� tion€may have triggered the laryngospasm. The postictal EEG showed diffuse slowing until

EEG Findings in SUDEP

205

6 hours after the seizure, when he developed nonconvulsive status epilepticus, which eventually resolved. He recovered to baseline and was ultimately discharged home. So et al. (2000) reported a case of near-SUDEP due to postictal central apnea. A 20-yearold woman with refractory epilepsy since age 1, was admitted for video EEG monitoring. She had frequent generalized convulsions and complex partial seizures and also had a history of convulsive status epilepticus at age 9 when she developed mononucleosis, which culminated in cardiorespiratory arrest. There were no permanent neurologic deἀcits after this event. She had no known cardiovascular or pulmonary disorders; however, following both clusters of seizures, as well as isolated, self-limited seizures, she had postictal respiratory arrest and had required cardiopulmonary resuscitation since age 10. It had been observed that following a seizure, her pulse was initially regular and strong, but this disappeared€as the apnea continued. Outpatient interictal EEG showed bisynchronous frontal sharp waves. During the monitoring, she had a cluster of four generalized convulsions within 6€hours, lasting 55–85 seconds each. Ictal EEG showed bifrontal slow waves at onset. She had fully recovered from the ἀrst three seizures when she had a fourth seizure, lasting 56 seconds. She was noted to have apnea immediately following the seizure. The EKG remained unchanged for 10 seconds, then gradually slowed until the heart stopped 57 seconds later. Postictally, the EEG showed diffuse slowing for 20 seconds, which was followed by marked suppression. She was successfully resuscitated, and follow-up cardiac evaluation showed no evidence of cardiac or pulmonary disease. A demand cardiac pacemaker was placed. Following implantation of the pacemaker, 10–15 second periods of apnea were noted after seizures; however, she no longer required cardiopulmonary resuscitation. Recently, a case of near-SUDEP revealed a right temporal complex partial seizure with secondary generalization associated with ventricular tachycardia (Espinosa et al. 2009). A 51-year-old woman underwent video-EEG monitoring for refractory focal epilepsy. Her interictal EEG revealed bitemporal independent theta and delta activity, and a baseline EKG revealed a ἀrst-degree atrioventricular block, with a normal QTc interval. She had a typical complex partial seizure with secondary generalization, which was associated with ventricular tachycardia, then ventricular ἀbrillation, toward the end of the seizure. The patient underwent successful cardiopulmonary resuscitation and subsequent deἀbrillator implantation.

13.4â•…Conclusions The EEG data from patients with SUDEP, near-SUDEP, or subsequent SUDEP reveal a variety of interictal ἀndings, with both focal and generalized interictal epileptiform abnormalities. While the case numbers are limited, the ictal EEG recordings from patients with SUDEP or near-SUDEP have uniformly recorded a terminal generalized tonic–clonic seizure, except in one case, which reported a cluster of two secondarily generalized tonic–clonic seizures and then a complex partial seizure just prior to death. This ἀnding is consistent with epidemiologic data that generalized tonic–clonic seizures increase risk for SUDEP. The data thus far do not implicate a speciἀc seizure focus lateralization or lobe of the brain being associated with a higher risk for SUDEP. These data, while limited, also suggest that seizure clusters might also increase risk for SUDEP. The EEGs from cases of SUDEP also reveal diffuse suppression after the seizure ends. Based on this ἀnding, the possibility of primary cerebral shutdown has been proposed

206 Sudden Death in Epilepsy: Forensic and Clinical Issues

(Bird et al. 1997). The sudden cessation of cerebral activity has been suggested to be due to primary irreversible brain failure, with cardiorespiratory failure occurring as a secondary consequence. However, it is important to note that there are limited cardiorespiratory data available in these cases. Alternatively, it is possible that seizures could result in concomitant cardiopulmonary abnormalities during the ictal or postictal phase of the seizure, resulting in either anoxia or decreased cardiac output. While there can be diffuse suppression of the EEG after an uncomplicated generalized tonic–clonic seizure, anoxia or decreased cardiac output, such as related to a seizure-related arrhythmia or pulseless electrical activity, may explain the persistence of this suppression and lack of recovery of the EEG in these SUDEP cases. While a primary cerebral etiology is possible, the data thus far suggest that the seizure itself is an important trigger for a cascade of respiratory and/or cardiac abnormalities that ultimately cause death. The near-SUDEP case of Espinosa et al. (2009) and the SUDEP case of Dasheiff and Dickinson (1986) were associated with ventricular tachyarrhythmias occurring during or toward the end of a seizure. In the two SUDEP cases with EKG or pulse artifact, bradycardia was recorded, suggesting that cardiac function was affected. The respiratory status in these cases is unknown, but it is possible that respiratory compromise could have occurred during the seizure, resulting in persistent postictal suppression of the EEG and reflex bradycardia. In the So et al. (2000) case, it appears that apnea triggered the bradycardia and treatment with a cardiac pacemaker insertion has been helpful. Even though the apnea was the initial event, the more concerning life-threatening cardiorespiratory response may have been the secondary cardiac effect of bradycardia, which had been eliminated by the pacemaker. More detailed analysis of EEG data and close correlation with cardiac and pulmonary function is needed to fully interpret the EEG ἀndings obtained thus far during SUDEP. Perhaps the use of more routine pulmonary and cardiac monitoring, along with EEG recordings, may yield further insights into the pathophysiology of SUDEP. Hopefully, specialized investigation, ideally evaluating the combined neurologic, cardiac, and pulmonary data in people with epilepsy during both the ictal as well as interictal states, will identify risk factors for SUDEP and provide targets for preventative therapy.

References Bird, J. M., K. A. T. Dembny, D. Sandeman, and S. Butler. 1997. Sudden unexplained death in epilepsy: An intracranially monitored case. Epilepsia 38 (S11): S52–S56. Chin, P. S., K. R. Branch, and K. J. Becker. 2004. Myocardial infarction following brief convulsive seizures. Neurology 63 (12): 2453. Dasheiff, R. M., and L. J. Dickinson. 1986. Sudden unexpected death of epileptic patient due to cardiac arrhythmia after seizure. Arch Neurol (43): 194–196. Espinosa, P. S., J. W. Lee, U. B. Tedrow, E. Bromἀeld, and B. A. Dworetzky. 2009. Sudden unexpected near death in epilepsy (SUNDEP): Malignant ventricular arrhythmia from a partial seizure. Neurology 72: 1702–1703. Langan, Y., L. Nashef, and J. W. A. S. Sander. 2000. Sudden unexpected death in epilepsy: A series of witnessed deaths. J Neurol Neurosurg Psychiatry 68: 211–213. Lee, M. A. 1998. EEG video recording of sudden unexpected death in epilepsy. Epilepsia 39 (S6): 120–121. McClean, B. N., and S. Wimalaratna. 2007. Sudden death in epilepsy recorded in ambulatory EEG. J Neurol Neurosurg Psych 78: 1395–1397.

EEG Findings in SUDEP

207

Nei, M., R. T. Ho, and B. W. Abou-Khalil et al. 2004. EEG and ECG in sudden unexplained death in epilepsy. Epilepsia 45 (4): 338–345. Nei, M., R. T. Ho, and M. R. Sperling. 2000. EKG abnormalities during partial seizures in refractory epilepsy. Epilepsia 41 (5): 542–548. Opherk, C., J. Coromilas, and L. J. Hirsch. 2002. Heart rate and EKG changes in 102 seizures: Analysis of influencing factors. Epilepsy Res 52: 117–127. So, E., M. Sam, and T. Lagerlund. 2000. Postictal central apnea as a cause of SUDEP: Evidence from near-SUDEP incident. Epilepsia 41 (11): 1494–1497. Tavee, J., and H. Morris. 2008. Severe postictal laryngospasm as a potential mechanism for sudden unexpected death in epilepsy: A near-miss in an EMU. Epilepsia 49 (12): 2113–2117. Tavernor, S. J., S. W. Brown, R. M. E. Tavernor, and C. Gifford. 1996. Electrocardiograph QT lengthening associated with epileptiform EEG discharges—A role in sudden unexplained death in epilepsy? Seizure 5: 79–83.

Severity of Seizures as a Forensic Risk and Case Reports Edward H. Maa Michael P. Earnest Mark C. Spitz Jacquelyn Bainbridge

14

Contents 14.1 What Is a Severe Seizure? 14.2 Deἀnition 14.3 Epidemiology 14.4 SUDEP Risk Factors 14.5 Witnessed Cases 14.6 University of Colorado Epilepsy Monitoring Unit Case 14.7 Conclusion References

209 210 211 211 211 215 218 219

14.1â•…What Is a Severe Seizure? Seizures are a debilitating neurological condition characterized by the sudden evolution of synchronous electrical activity in the brain that can lead to loss of awareness, confusion, sudden falls, odd motor behaviors or sensations, and convulsions (Spitz 1998). Medically inexperienced, ἀrst-time witnesses of seizures are frightened or confused and often misinterpret the event as the impending death of the patient. What makes a seizure severe? The lay media implies the severity of seizures by the intense shaking, drooling, and eye rolling associated with a convulsion. The severity of a seizure in a typical medical encounter is judged by its duration and the presence of witnessed convulsion, tongue biting, and incontinence. To the patient stricken with seizures, however, each event is severe not only because of the clinical manifestation of the seizure, but also because of the psychosocial impact of the unpredictability of the seizure. The epileptologist’s approach to seizure severity is in alignment with the patient perspective but also emphasizes the likelihood of injury. This likelihood of injury is the foundation of seizure precautions and is intended as a safety recommendation because of the unpredictability of epileptic seizures. The sudden change or loss of awareness associated with seizures would not in and of itself be harmful if it only occurred in sleep. In the settings of operating heavy machinery, swimming or bathing alone, work or play at heights, tight spaces, or with open flames including cooking stoves, seizures can have devastating consequences. Even a single seizure in these settings increases the risk of death or serious injury. 209

210 Sudden Death in Epilepsy: Forensic and Clinical Issues

Unpredictability inevitably leads to anxiety. Any new neurology resident taking stroke calls can attest to many sleepless nights waiting for their ἀrst tissue plasminogen activator (tPA) candidate. New parents experience anxious and uneasy sleep, listening to the irregular breaths and sounds of a sleeping newborn infant. In much the same way, it is the overwhelming anticipation, rather than the seizure itself, that leads to chronic anxiety. In fact, of 1023 epilepsy patients who responded to an Epilepsy Foundation questionnaire, uncertainty and fear of the next seizure was rated as the worst thing about having epilepsy (Fisher et al. 2000). Fear eventually leads to restricted movements outside of the home, decreased productivity, and impaired livelihood. Psychiatric disturbances including depression and agoraphobia can be found in as many as 70%, and the risk of suicide in uncontrolled epilepsy patients is 13%, nine times that of the general population (Nowack 2006). Seizures themselves have graded severity; intuitively, the more violent the motor involvement, the more severe the seizure. This concept is manifest in the old terminology of “grand mal” and “petit mal” seizures. The imprecision of these terms from the perspective of therapeutics and modern epilepsy care ignores the fact that to the lay person, blinking and staring cannot possibly be as serious as convulsing and becoming cyanotic. The higher the frequency and longer the duration of an individual seizure, as well as the overall duration of epilepsy, add to the concept of severity, as do a number of surrogate markers including increased numbers and doses of medications, specialists, and surgeries. The ultimate marker of seizure and epilepsy severity is any seizure event that results in death. Recent sensationalized deaths in persons having a seizure emphasize the fact that uncontrolled epilepsy should not be thought of as a chronic condition that merely necessitates annual reἀlling of medications (Epilepsy Foundation 2009; Phillips 2008). Obviously, patients with persistent seizures are at increased risk of fractures, burns, and drowning, but they also exhibit increased rates of depression, anxiety, and suicide (Sperling 2004; Spitz et al. 1994). Less well known in the general medical community is that sudden unexplained death in epilepsy (SUDEP) is the most common cause of seizure-related death (Langan et al. 2000; Langan and Nashef 2003), accounting for as many as 50% of early deaths in refractory epilepsy patients (Sperling 2001).

14.2â•…Definition Sudden unexplained death in epilepsy can be established by applying the criteria developed by an expert panel (Leestma et al. 1997):

1. Diagnosis of epilepsy 2. Death occurring unexpectedly while in a reasonable state of health 3. Death occurring suddenly 4. Death occurring during normal activities and benign circumstances 5. No obvious medical cause of death determined during postmortem examination 6. Death is not the result of trauma, asphyxia from aspiration, or status epilepticus

Death from SUDEP is “deἀnite” if all conditions are satisἀed and “probable” if no postmortem data is available.

Severity of Seizures as a Forensic Risk and Case Reports

211

14.3â•…Epidemiology Fortunately, SUDEP remains relatively rare with annual incidence ἀgures ranging from 0 to 10 per 1000 patients, depending on the studied population (Tellez-Zenteno et al. 2005). Reflected in these numbers is the suggestion that seizure severity accounts for a higher incidence of SUDEP, with epilepsy surgical candidates representing the higher end of the spectrum and general population coroner’s cases representing the lower end of the spectrum. SUDEP appears to be mainly a problem in patients with refractory epilepsy. In the National General Practice Study of Epilepsy in the United Kingdom (NGPSE), there was only one conἀrmed case of SUDEP in 7147 person years. In this prospective cohort of 564 patients, 70% became seizure-free for at least 5 years of the mean 15 years of follow-up (Sander et al. 1990).

14.4â•…SUDEP Risk Factors Accumulated risk factors from descriptive cohort studies over the years include youth (Leestma et al. 1997; Opeskin and Berkovic 2003), male gender (Tennis et al. 1995), early onset of epilepsy (Kloster and Engelskjon 1999; Nilsson et al. 1999), duration of epilepsy and seizure frequency (Walczak et al. 2001; Leestma et al. 1997), poor control of seizures (Sperling et al. 1999), convulsive seizure type (Kloster and Engelskjon 1999; Birnbach et al. 1991), high antiepileptic drug number (Nilsson et al. 1999; Tennis et al. 1995; Walczak et al. 2001; Racoosin et al. 2001), frequency of antiepileptic drug changes (Nilsson et al. 1999), subtherapeutic antiepileptic drug levels (Kloster and Engelskjon 1999; Earnest et al. 1992), mental retardation (Walczaket al. 2001), concomitant use of psychotropic medications (Nilsson et al. 1999; Tennis et al. 1995), prone position (Kloster and Engelskjon 1999), and being found in bed or home (Kloster and Engelskjon 1999; Opeskin and Berkovic 2003). Despite inconsistent methodologies, patient populations, and autopsy availability, Stollberger and Finsterer (2004) and Tellez-Zenteno et al. (2005) summarized these ἀndings in their works, providing a picture of a mid to late 30s male living alone with poorly€controlled, long-standing symptomatic convulsive epilepsy, on multiple medications, found dead in bed in the prone position, often with evidence of a recent seizure.

14.5â•…Witnessed Cases The elusiveness of SUDEP’s etiology likely remains because of its predilection for unwitnessed sleep, relative rarity, and its multifactorial nature. Witnessed case reports over the years have shed some light on this devastating condition. Dasheiff and Dickinson (1986) described a 48-year-old male from the Wisconsin Epilepsy Center with a witnessed complex partial seizure who recovered, but then suffered what appeared to be a second seizure accompanied by ventricular ἀbrillation on EKG. Even though witnessed in the hospital, he was not successfully resuscitated and went on to receive an unrevealing autopsy. The authors suspected a cardiac origin of SUDEP. Later, Dasheiff (1991) described a much higher incidence of SUDEP, almost 1:100, than previously reported from their experience at the Pittsburgh Epilepsy Center. One of the seven patients he described was in the midst of transfer from the intensive care unit to

212 Sudden Death in Epilepsy: Forensic and Clinical Issues

the hospital floor after receiving a temporal lobectomy for refractory epilepsy when she suddenly expired while sitting unmonitored in bed. Efforts to resuscitate her were also fruitless and, despite another negative autopsy, the author suspected a cardiac cause as the leading etiologic hypothesis. Purves et al. (1992) reported a case of a 27-year-old woman from the British Columbia epilepsy program. She had complex partial seizures during her monitoring stay, but her last event was a secondarily generalized tonic–clonic seizure that resulted in her lying in the prone position. She was found 24 min later, cyanotic and unable to be resuscitated. Authors attributed this death to asphyxia following severe postictal depression. Bird et al. (1997) reported the ἀrst intracranially monitored case, from their experience in Bristol, England. The patient had bilateral, anterior and posterior temporal depth electrodes placed. Two days later, all medications were withdrawn and he had four typical seizures with right mesial temporal onset of electrical activity followed by blank staring, left head turn, then 2−4 min of convulsion. His ἀfth seizure occurred in sleep during which his left head turn was followed by his whole body turning to the prone position. He convulsed for 2.5 min then stopped moving and never recovered. Evidence of labored breathing or asphyxia was not seen by the authors and, unfortunately, there was no cardiac rhythm strip associated with this recording. The electroencephalogram (EEG) revealed an unusual right mesial spindling spike-discharge burst suppression activity for 16 s, followed by electrical silence. The left hemisphere showed rhythmic spike discharges for an

7 µV/mm

Fp1 - F7

7 µV/mm

F7 - T3 T3 - T5

7 µV/mm

T5 - O1

7 µV/mm

Fp2 - F8

7 µV/mm

F8 - T4

7 µV/mm

T4 - T6

7 µV/mm

T6 - O2

7 µV/mm

Fp1 - F3

7 µV/mm 7 µV/mm

F3 - C3

7 µV/mm

C3 - P3 P3 - O1

7 µV/mm

Fp2 - F4

7 µV/mm 7 µV/mm

F4 - C4

7 µV/mm

C4 - P4 P4 - O2

7 µV/mm

Fz - Cz

7 µV/mm

Cz - Pz

7 µV/mm

ECG

50 µV/mm

00:17:34

10.5 s/page, 1 s/div, 21.6 mm/s Gain: 50 µV/mm, LFF: 1 Hz, HFF: 70 Hz, Notch: Off

xltek

Figure 14.1╇ The time stamps on each figure show the progression of time. EKG rhythm strips illustrate the lengthening between complexes and the peaking of T waves.

Severity of Seizures as a Forensic Risk and Case Reports

213

additional 8 s when they also disappeared permanently. Pulse artifact was seen for an additional 2 min, prompting the authors to downplay the cardiac arrhythmia hypothesis of SUDEP. The lack of slow wave changes on EEG was also not suggestive of cerebral anoxia. Based on the pattern of spindling spike discharges seen as the terminal event, the authors suggested that dysregulation of thalamocortical regulatory systems may not only shut off cortical activity, but may be responsible for cessation of all life functions. Lee (1998) reported a case from Calgary during which a woman’s convulsion was followed by bradycardia to a rate of 30 beats/min. A minute after seizure cessation, the EEG became silent and never recovered. Interestingly, the bradycardia resolved after 4 additional minutes, returning to a rate of 70 beats/min, until the patient ἀnally expired 18 min later. A cardiac cause was also not suspected as an explanation for the death. Finally, McLean and Wimalaratna (2007) described an ambulatory EEG case from the United Kingdom of a woman with poorly controlled epilepsy who underwent ambulatory monitoring. Her ambulatory recorder documented increasing spike frequency once she had fallen asleep, eventually coalescing into an ictal event at 08:27:18 the next morning. Reminiscent of the intracranial case, the seizure exhibited polyspike (up to 6 spikes) activity for 52 s, abruptly terminated at 08:28:14, leaving an isoelectric EEG. No cardiac rhythm information was recorded, but rhythmic movement artifact was seen in the left temporal leads associated with muscle activity. It slowed to complete cessation of activity over the next 3 min. The patient was found the next morning in her night clothes, prone on

7 µV/mm

Fp1 - F7

7 µV/mm

F7 - T3

7 µV/mm

T3 - T5 T5 - O1

7 µV/mm

Fp2 - F8

7 µV/mm 7 µV/mm

F8 - T4

7 µV/mm

T4 - T6 T6 - O2

7 µV/mm

Fp1 - F3

7 µV/mm 7 µV/mm

F3 - C3

7 µV/mm

C3 - P3

7 µV/mm

P3 - O1

7 µV/mm

Fp2 - F4

7 µV/mm

F4 - C4

7 µV/mm

C4 - P4 P4 - O2

7 µV/mm

Fz - Cz

7 µV/mm 7 µV/mm

Cz - Pz

50 µV/mm

ECG

00:18:54

10.5 s/page, 1 s/div, 21.6 mm/s Gain: 50 µV/mm, LFF: 1 Hz, HFF: 70 Hz, Notch: Off

xltek

Figure 14.2╇ The time stamps on each figure show the progression of time. EKG rhythm strips illustrate the lengthening between complexes and the peaking of T waves.

214 Sudden Death in Epilepsy: Forensic and Clinical Issues

the floor with arm outstretched toward the telephone. The authors also could not attribute the death to hypotension or cerebral anoxia due to the lack of slow wave activity on the EEG (Neidemeyer and Da Silva 1987). They coined the phrase abrupt irreversible “cerebral electrical shutdown” to explain the primary mechanism of SUDEP. The largest case-control study of SUDEP is from the United Kingdom, where autopsies are routinely performed on suspected seizure deaths. Langan et al. (2005) published their results of 154 cases of conἀrmed SUDEP. Of the conἀrmed SUPDEP cases, 15% (23 cases) were witnessed, with the majority following a convulsive seizure and associated with breathing difficulties. In addition to previous accounts of cardiac arrhythmias, they suggest central and obstructive apneas are likely contributors to SUDEP mechanisms, based on the high frequency of labored breathing reported. In the controlled environment of epilepsy monitoring and telemetry units, cardiac cases may be enriched because nurse interventions, such as rolling patients into the recovery position, nasal cannula oxygen, as well as engaging the postictal patient in the neurological exam, may be sufficient to prevent deaths from apneic mechanisms. This argument is supported by their ἀnding that supervision was a protective factor in their study, and further supported by a study of SUDEP incidence at a residential school for children with epilepsy who were supervised at night and carefully monitored after a seizure. Of the 310 students enrolled between 1970 and 1993, there were no SUDEP deaths during term, but 14 sudden deaths while at home on vacation. Most were unwitnessed (Nashef et al. 1995).

Fp1 - F7

15 µV/mm

F7 - T3

15 µV/mm

T3 - T5

15 µV/mm

T5 - O1

15 µV/mm

Fp2 - F8

15 µV/mm

F8 - T4

15 µV/mm

T4 - T6

15 µV/mm

T6 - O2

15 µV/mm

Fp1 - F3

15 µV/mm

F3 - C3

15 µV/mm

C3 - P3

15 µV/mm

P3 - O1

15 µV/mm

Fp2 - F4

15 µV/mm

F4 - C4

15 µV/mm

C4 - P4

15 µV/mm

P4 - O2

15 µV/mm

Fz - Cz

15 µV/mm

Cz - Pz

15 µV/mm

ECG

50 µV/mm

00:19:54

10.5 s/page, 1 s/div, 21.6 mm/s Gain: 15 µV/mm, LFF: 1 Hz, HFF: 15 Hz, Notch: Off

xltek

Figure 14.3╇ The time stamps on each figure show the progression of time. EKG rhythm strips illustrate the lengthening between complexes and the peaking of T waves.

Severity of Seizures as a Forensic Risk and Case Reports

215

14.6â•…University of Colorado Epilepsy Monitoring Unit Case The patient was a 63-year-old, right-handed male with seizures beginning at age 6 or 7. His seizures began with a vague, “strange feeling,” followed quickly by a sensation of nervousness and feeling of “wanting to get away.” This lasted 30 s and was followed by lethargy and trouble with expressive speech for several minutes. These events occurred multiple times a day despite maximal medical management, and very rarely would be followed by a secondarily generalized tonic–clonic seizure lasting 1 to 2 min. His last convulsive seizure occurred more than 5 years ago, and he had no history of status epilepticus. He had failed multiple antiepileptic medications including phenobarbital, valproic acid, gabaÂ� pentin, lamotrigine, topiramate, zonisamide, and levetiracetam, and was currently being managed by phenytoin, 300 mg twice daily, and carbamazepine, 600 mg twice daily, by a community neurologist. His past medical history was signiἀcant for complex partial seizures, depression (fluoxetine, 20 mg nightly), hypertension (atenolol, 50 mg nightly), and a history of head trauma before 2 years of age during which he was unconscious for more than a week. He was the product of an uncomplicated pregnancy and birth and there was no family history of epilepsy. Physical examination was remarkable for blood pressure of 147/77 and pulse 62, II/VI systolic crescendo/decrescendo cardiac murmur at the right upper sternal border, and brisk but symmetric lower extremity reflexes with flexor plantar responses. Routine EEG revealed left temporal sharp waves. MRI scans were performed at an outside hospital and not available for review. Fp1 - F7

7 µV/mm

F7 - T3

7 µV/mm

T3 - T5

7 µV/mm

T5 - O1

7 µV/mm

Fp2 - F8

7 µV/mm

F8 - T4

7 µV/mm

T4 - T6

7 µV/mm

T6 - O2

7 µV/mm

Fp1 - F3

7 µV/mm

F3 - C3

7 µV/mm

C3 - P3

7 µV/mm

P3 - O1

7 µV/mm

Fp2 - F4

7 µV/mm

F4 - C4

7 µV/mm

C4 - P4

7 µV/mm

P4 - O2

7 µV/mm

Fz - Cz

7 µV/mm

Cz - Pz

7 µV/mm

ECG

50 µV/mm

00:20:44

10.5 s/page, 1 s/div, 21.6 mm/s Gain: 7 µV/mm, LFF: 1 Hz, HFF: 15 Hz, Notch: Off

xltek

Figure 14.4╇ The time stamps on each figure show the progression of time. EKG rhythm strips illustrate the lengthening between complexes and the peaking of T waves.

216 Sudden Death in Epilepsy: Forensic and Clinical Issues

He was admitted for epilepsy monitoring for potential ablative epilepsy surgery. Phenytoin and carbamazepine doses were halved on day 2. Phenytoin was discontinued on day 3, and carbamazepine was discontinued on day 4. During the latter half of day 4, the patient had two of his typical complex partial events with the second one generalizing into a mild convulsion. These events evolved out of the left temporal lobe and, of note, the patient’s postictal pulse following his convulsion was 140 beats/min. In the early morning of day 5, he had a third and ἀnal event. The seizure began typically, evolving from the left temporal region, but clinically the patient was sleeping in a prone position. His seizure generalized after 45 s, but the patient remained prone, his face hidden in his pillow. Audio and visual conἀrmation of progressively labored postictal breathing with good chest expansion was evident after the gentle convulsion ended 1 min later. One-and-a-half minutes after this, his EEG attenuated to essentially a flat baseline with only pulse artifact appreciated. Less than 30 s later, audio and visual evidence of breathing stopped as the EKG rhythm strip also terminated in electrical silence (Figures 14.1 through 14.8). The cardiac rhythm strip is particularly insightful in this tragic case, and suggests that previous cases may beneἀt from reevaluation by a multidisciplinary team. The EKG following the cessation of the convulsive seizure begins to suggest a peaked T wave (Figure 14.4 compared with Figure 14.1). In Figure 14.5, junctional escape beats are appreciated as the T wave continues to exhibit a more peaked appearance. By Figure 14.7, signiἀcant ST elevation is appreciated as the cardiac rhythm begins to slow signiἀcantly and, by Figure 14.8, cardiac electrical activity is silent. In a review of these series of rhythm strips with Fp1 - F7

7 µV/mm

F7 - T3

7 µV/mm

T3 - T5

7 µV/mm

T5 - O1

7 µV/mm

Fp2 - F8

7 µV/mm

F8 - T4

7 µV/mm

T4 - T6

7 µV/mm

T6 - O2

7 µV/mm

Fp1 - F3

7 µV/mm

F3 - C3

7 µV/mm

C3 - P3

7 µV/mm

P3 - O1

7 µV/mm

Fp2 - F4

7 µV/mm

F4 - C4

7 µV/mm

C4 - P4

7 µV/mm

P4 - O2

7 µV/mm

Fz - Cz

7 µV/mm

Cz - Pz

7 µV/mm

ECG

50 µV/mm 00:21:26

10.5 s/page, 1 s/div, 21.6 mm/s Gain: 7 µV/mm, LFF: 1 Hz, HFF: 15 Hz, Notch: Off

xltek

Figure 14.5╇ The time stamps on each figure show the progression of time. EKG rhythm strips illustrate the lengthening between complexes and the peaking of T waves.

Severity of Seizures as a Forensic Risk and Case Reports

217

a cardiologist, the peaking of T waves followed by junctional escape beats suggests acidemic hyperkalemia (C. Long, personal communication, April 21, 2009). Not having the beneἀt of the corresponding video that clearly demonstrates the patient as being postictal with labored, facedown breathing, Long suggested the EKG showed the patient becoming acidemic, possibly by a rebreathing mechanism. Acute potassium shifts associated with respiratory acidosis are predicted (Perez et al. 1981), but whether levels associated with rebreathing are sufficient to produce cardiac disturbance are less certain (S. Linus, personal communication, May 4, 2009). Linus reviewed the postictal EKG as well, but as a nephrologist, he was much less impressed with the EKG being explained by hyperkalemia, claiming that despite the minimal peaking of the T wave, the QRS complex remained narrow throughout. Montague et al. (2008) retrospectively reviewed the frequency of EKG changes in hyperkalemia. Despite subjective corroboration of peaking of the T wave with quantitative amplitude measurements, no diagnostic threshold could be established. In fact, the EKG had such poor sensitivity and speciἀcity for hyperkalemia, they recommended against EKGs in guiding treatment of hyperkalemia in stable patients. The cardiologic and nephrologic interpretation of this rhythm strip remains unresolved, but suggests a different and broader approach to the problem of SUDEP. Could a combination of convulsion-related lactic acidosis plus respiratory acidosis from prone rebreathing be sufficient to cause a potassium-related fatal arrhythmia? Much like the ambulatory and intracranial EEG cases, there was no evidence of slow waves to suggest that hypotension or cerebral anoxia were the cause of death. Additionally, the EEG Fp1 - F7

7 µV/mm

F7 - T3

7 µV/mm

T3 - T5

7 µV/mm

T5 - O1

7 µV/mm

Fp2 - F8

7 µV/mm

F8 - T4

7 µV/mm

T4 - T6

7 µV/mm

T6 - O2

7 µV/mm

Fp1 - F3

7 µV/mm

F3 - C3

7 µV/mm

C3 - P3

7 µV/mm

P3 - O1

7 µV/mm

Fp2 - F4

7 µV/mm

F4 - C4

7 µV/mm

C4 - P4

7 µV/mm

P4 - O2

7 µV/mm

Fz - Cz

7 µV/mm

Cz - Pz

7 µV/mm

ECG

50 µV/mm 00:22:04

10.5 s/page, 1 s/div, 21.6 mm/s Gain: 7 µV/mm, LFF: 1 Hz, HFF: 15 Hz, Notch: Off

xltek

Figure 14.6╇ The time stamps on each figure show the progression of time. EKG rhythm strips illustrate the lengthening between complexes and the peaking of T waves.

218 Sudden Death in Epilepsy: Forensic and Clinical Issues

Fp1 - F7

7 µV/mm

F7 - T3

7 µV/mm

T3 - T5

7 µV/mm

T5 - O1

7 µV/mm

Fp2 - F8

7 µV/mm

F8 - T4

7 µV/mm

T4 - T6

7 µV/mm

T6 - O2

7 µV/mm

Fp1 - F3

7 µV/mm

F3 - C3

7 µV/mm

C3 - P3

7 µV/mm

P3 - O1

7 µV/mm

Fp2 - F4

7 µV/mm

F4 - C4

7 µV/mm

C4 - P4

7 µV/mm

P4 - O2

7 µV/mm

Fz - Cz

7 µV/mm

Cz - Pz

7 µV/mm

ECG

50 µV/mm 00:24:49

10.5 s/page, 1 s/div, 21.6 mm/s Gain: 7 µV/mm, LFF: 1 Hz, HFF: 15 Hz, Notch: Off

xltek

Figure 14.7╇ The time stamps on each figure show the progression of time. EKG rhythm strips illustrate the lengthening between complexes and the peaking of T waves.

had become silent a full 4 min before the last EKG activity, possibly supporting McLean’s concept of “cerebral electrical shutdown.” Could hyperkalemia or respiratory acidosis contribute to this phenomenon? Or are these patient’s EKG changes wholly unrelated to the underlying mechanism of SUDEP? More questions than answers remain.

14.7â•…Conclusion Seizure severity appears to be a forensic risk for the development of SUDEP. Seizures that are convulsive, frequent, and of long duration appear to increase the risk of SUDEP to almost 1:100 per year (Tellez-Zenteno et al. 2005), but not all SUDEP deaths are preceded by convulsions. Instead complex partial seizures and recovery from complex partial seizures with death a short time later have been reported (Langan et al. 2005). The severity of a speciἀc convulsion is also not particularly helpful, as witnessed in the University of Colorado case. The patient began convulsing in a prone position and essentially never moved from the spot. The convulsion was not of an unusual duration or violent motor behavior, but appeared to be position dependent. The nexus of coroners’ (Leestma 1990) and witnessed cases of SUDEP seems to suggest that a signiἀcant number of deaths are associated with prone positioning, which really has nothing at all to do with seizure severity. As increased interest and research in SUDEP further clariἀes the pathophysiologic explanations of sudden death in epilepsy, our epidemiological concept of seizure severity

Severity of Seizures as a Forensic Risk and Case Reports

219

7 µV/mm

Fp1 - F7

7 µV/mm

F7 - T3

7 µV/mm

T3 - T5 T5 - O1

7 µV/mm

Fp2 - F8

7 µV/mm

F8 - T4

7 µV/mm

T4 - T6

7 µV/mm

T6 - O2

7 µV/mm

Fp1 - F3

7 µV/mm

F3 - C3

7 µV/mm

C3 - P3

7 µV/mm

P3 - O1

7 µV/mm

Fp2 - F4

7 µV/mm

F4 - C4

7 µV/mm

C4 - P4

7 µV/mm

P4 - O2

7 µV/mm

Fz - Cz

7 µV/mm

Cz - Pz

7 µV/mm

ECG

50 µV/mm

00:28:45

10.5 s/page, 1 s/div, 21.6 mm/s Gain: 7 µV/mm, LFF: 1 Hz, HFF: 15 Hz, Notch: Off

xltek

Figure 14.8╇ The time stamps on each figure show the progression of time. EKG rhythm strips illustrate the lengthening between complexes and the peaking of T waves.

may likely play a less signiἀcant role in our approach to SUDEP understanding, education, and avoidance.

References Bird, J. M., K. A. T. Dembny, D. Sandeman, and S. Butler. 1997. Sudden unexplained death in epilepsy: An intracranially monitored case. Epilepsia 38 (s11): 52–56. Birnbach, C. D., A. J. Wilensky, and C. B. Dodril. 1991. Predictors of early mortality and sudden death in epilepsy: A multidisciplinary approach. J Epilepsy 4: 11–17. Dasheiff, R. M. 1991. Sudden unexpected death in epilepsy and its relationship to sudden cardiac death. J Clin Neurophysiol 8: 216–222. Dasheiff, R. M., and L. J. Dickinson. 1986. Sudden unexpected death following a seizure in an epileptic patient: A case report. Arch Neurol 43: 194–196. Earnest, M. P., G. E. Thomas, R. A. Eden, and K. F. Hossack. 1992. The sudden unexplained death syndrome in epilepsy: Demographic, clinical, and postmortem features. Epilepsia 33: 310–316. Epilepsy Foundation. 2009. http://www.epilepsyfoundation.org/epilepsyusa/news/Travolta.cfm (acÂ�Â� cessed February 8, 2009). Fisher, R. S., B. G. Vickrey, P. Gibson et al. 2000. The impact of epilepsy from the patient’s perspective I. Descriptions and subjective perceptions. Epilepsy Res 41 (1): 39–51. Kloster, R., and T. Engelskjon. 1999. Sudden unexpected death in epilepsy (SUDEP): A clinical perspective and a search for risk factors. J Neurol Neurosurg Psychiatry 67: 439–444. Langan, Y., and L. Nashef. 2003. Sudden unexpected death in epilepsy (SUDEP). ACNR 2 (6): 6–8.

220 Sudden Death in Epilepsy: Forensic and Clinical Issues Langan, Y., L. Nashef, and J. W. Sander. 2000. Sudden unexpected death in epilepsy: A series of witnessed deaths. J Neurol Neurosurg Psychiatry 68: 211–213. Langan, Y., L. Nashef, and J. W. Sander. 2005. Case-control study of SUDEP. Neurology 64: 1131–1133. Lee, M. A. 1998. EEG Video recording of sudden unexpected death in epilepsy (SUDEP). Epilepsia 39 (s6): 123–124. Leestma, J. E. 1990. Sudden unexpected death associated with seizures: A pathological review. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, 61–88. New York, NY: Marcel Dekker. Leestma, J. E., J. F. Annegers, M. J. Brodie et al. 1997. Sudden unexplained death in epilepsy: Observations from a large clinical development program. Epilepsia 38: 47–55. McLean, B. N., and S. Wimalaratna. 2007. Sudden death in epilepsy recorded in ambulatory EEG. J Neurol Neurosurg Psychiatry 78: 1395–1397. Montague, B. T., J. R. Ouellette, and G. K. Buller. 2008. Retrospective review of the frequency of ECG changes in hyperkalemia. Clin J Am Soc Nephrol 3(2): 324–330. Nashef, L., D. R. Fish, S. Garner, J. W. Sander, and S. D. Shorvon. 1995. Sudden death in epilepsy: A study of incidence in a young cohort with epilepsy and learning difficulty. Epilepsia 36: 1187–1194. Neidemeyer, E., and F. L. Da Silva. 1987. Electroencephalography, Basic Principles, Clinical Applications and Related Fields, 2nd ed., 385. Baltimore, MD: Urban and Schanarzenberg. Nilsson, L., B. Y. Farahmand, P. G. Persson et al. 1999. Risk factors for sudden unexpected death in epilepsy: A case-controlled study. Lancet 13: 888–893. Nowack, W. J. 2006. Psychiatric disorders associated with epilepsy. http://www.emedicine.medscape╉ .com/article/1186336 (accessed May 12, 2009). Opeskin, K., and S. Berkovic. 2003. Risk factors for sudden unexpected death in epilepsy: A controlled prospective study based on coroners cases. Seizure 12: 456–464. Perez, G. O., J. R. Oster, and C. A. Vaamonde. 1981. Serum potassium concentration in acidemic states. Nephron 27 (4–5): 233–243. Phillips, L. A. 2008. Death in epilepsy monitoring unit raises questions about safety policies and practice standards. Neurology Today 8 (16): 1–15. Purves, S. J., M. Wilson-Young, and V. P. Sweeney. 1992. Sudden death in epilepsy: Single case report with video-EEG documentation. Epilepsia 33 (Sl3): 123. Racoosin, J. A., J. Feeney, G. Burkhart et al. 2001. Mortality in antiepileptic drug development programs. Neurology 56: 514–519. Sander, J. W., Y. M. Hart, A. L. Johnson, and S. D. Shorvon. 1990. National General Practice Study of Epilepsy: Newly diagnosed epileptic seizures in a general population. Lancet 336: 1267–1271. Sperling, M. R. 2001. Sudden unexplained death in epilepsy. Epilepsy Curr 1 (1): 21–23. Sperling, M. R. 2004. The consequences of uncontrolled epilepsy. CNS Spectr 9: 98–101, 106–109. Sperling, M. R., H. Feldman, J. Kinman et al. 1999. Seizure control and mortality in epilepsy. Ann Neurol 46: 45–50. Spitz, M. C. 1998. Injuries and death as a consequence of seizures in people with epilepsy. Epilepsia 39: 904–907. Spitz, M. C., J. A. Towbin, D. Shantz, and L. E. Adler. 1994. Risk factors for burns as a consequence of seizures in people with epilepsy. Epilepsia 35: 764–767. Stollberger, C., and J. Finsterer. 2004. Cardiorespiratory ἀndings in sudden unexplained/unexpected death in epilepsy (SUDEP). Epilepsy Res 59: 51–60. Tellez-Zenteno. J. F., L. H. Ronquillo, and S. Weibe. 2005. Sudden unexpected death in epilepsy: Evidence-based analysis of incidence and risk factors. Epilepsy Res 65: 101–115. Tennis, P., T. B. Cole, J. F. Annegers et al. 1995. Cohort study of incidence of sudden unexplained death in persons with seizure disorder treated with antiepileptic drugs in Saskatchewan, Canada. Epilepsia 36: 29–36. Walczak, T. S., I. E. Leppik, M. D’Amelio et al. 2001. Incidence and risk factors in sudden unexpected death in epilepsy: A prospective cohort study. Neurology 56: 519–525.

Intractable Epilepsy in the Setting of Malformations of Cortical Development as a Mechanism for SUDEP

15

Lara Jehi Imad Najm

Contents 15.1 Introduction 15.2 SUDEP: Epidemiology and Risk Factors 15.3 Proposed Mechanisms of SUDEP 15.3.1 Pulmonary Pathophysiology 15.3.1.1 Clinical Evidence 15.3.1.2 Experimental Evidence 15.3.2 Cardiac Pathophysiology 15.3.2.1 Clinical Evidence 15.3.2.2 Experimental Evidence 15.4 Central Autonomic and Respiratory Control 15.5 Malformations of Cortical Development and SUDEP 15.5.1 Classiἀcation of MCD 15.5.2 Neuroimaging of MCD 15.5.3 Relevance of MCD Classiἀcation and Imaging to SUDEP 15.6 Localization of MCD and SUDEP 15.6.1 Case Reports 15.6.2 Cleveland Clinic Epilepsy Center Experience 15.7 Mechanisms of Epileptogenicity in MCD and SUDEP 15.7.1 Localized Disruption of Excitatory and Inhibitory Neurotransmission 15.7.1.1 Experimental Evidence 15.7.1.2 Clinical Evidence 15.7.1.3 Relevance to SUDEP 15.7.2 Diffuse Disruption of Normal Neural Circuitry 15.7.2.1 Experimental Evidence 15.7.2.2 Clinical Evidence 15.7.2.3 Relevance to SUDEP 15.8 Conclusion References

221

222 222 222 222 223 223 223 223 224 225 226 226 227 227 228 228 229 229 229 229 230 230 230 230 231 231 231 232

222 Sudden Death in Epilepsy: Forensic and Clinical Issues

15.1â•…Introduction Sudden unexpected death in epilepsy (SUDEP) is currently accepted as the most important epilepsy-related mode of death, and is the leading cause of death in chronic uncontrolled epilepsy (Jehi and Najm 2008; Tomson 2000). Despite signiἀcantly increased interest in SUDEP over the past few decades, the exact mechanisms leading to its occurrence remain unknown (Nashef et al. 2007; Pedley and Hauser 2002; Schraeder et al. 2009; Tomson et al. 2008a). Furthermore, despite higher rates of SUDEP observed in patients with structural brain abnormalities (Monte et al. 2007), little is known about how, or even if, speciἀc epilepsy etiologies and brain pathologies interact with other potential triggers leading up to sudden death in a given epilepsy patient. Speciἀcally, the role played by malformations of cortical development (MCD), a major cause of intractable epilepsy, remains unknown. This chapter will ἀrst briefly review general concepts related to SUDEP and its proposed mechanisms, outline basic concepts pertaining to intractable epilepsy in MCD, and then focus on how those two topics—intractable epilepsy in MCD and SUDEP—may be related. The discussion will be based on a review of SUDEP occurrences among a cohort of patients evaluated at Cleveland Clinic Epilepsy Center over a 15-year period, and on a review of the currently available literature.

15.2â•… SUDEP: Epidemiology and Risk Factors SUDEP is most often deἀned as the sudden, unexpected, witnessed or unwitnessed, nontraumatic, and nondrowning death of patients with epilepsy with or without evidence of a seizure, excluding documented status epilepticus, and in whom postmortem examination does not reveal a structural or toxicological cause for death (Nashef et al. 2007). Estimates of its incidence range from 0.7 to 1.3 cases per 1000 patient years in large cohorts of patients with epilepsy (Nilsson et al. 1997; Tennis et al. 1995), and from 3.5 to 9.3 cases per 1000 patient years in anticonvulsant drug registries, medical device registries, and epilepsy surgery programs (Leestma et al. 1997; Nashef et al. 1995; Tomson et al. 2008b). Several potential risk factors for SUDEP have been investigated with conflicting ἀndings (Jehi and Najm 2008). Consistently identiἀed risk factors include young age, early onset of seizures, refractoriness of epilepsy, the presence of generalized tonic–clonic seizures, male sex, and being in bed at the time of death (Langan et al. 2005; Monte et al. 2007; Nashef et al. 2007; Tomson et al. 2008b). Weaker risk factors include being in the prone position at the time of death, having one or more subtherapeutic blood levels of anticonvulsant medication, having a structural brain lesion, and being asleep (Monte et al. 2007; Tomson et al. 2008b). At any rate, the current consensus is that SUDEP is primarily a seizure-related occurrence, with patients having poorly controlled epilepsy and frequent generalized tonic–clonic seizures being particularly vulnerable (Jehi and Najm 2008; Tomson et al. 2008b) (see Table 15.1).

15.3â•… Proposed Mechanisms of SUDEP 15.3.1â•… Pulmonary Pathophysiology The two major proposed respiratory mechanisms of SUDEP are central apnea and acute neurogenic pulmonary edema.

Intractable Epilepsy in the Setting of MCD as a Mechanism for SUDEP

223

Table 15.1â•… Summary of the General Mechanisms Thought to Contribute to SUDEP Respiratory mechanisms â•… Central apnea â•… Pulmonary edema Cardiac mechanisms â•… Arrhythmia â•… Asystole Effects of long-standing seizure disorder â•… Altered autonomic function â•… Structural heart change

15.3.1.1â•…Clinical Evidence In a prospective study of epilepsy patients undergoing a video-EEG evaluation, central apnea lasting at least 10 s was observed postictally in 40% of the recorded seizures (Nashef et al. 1996); otherwise healthy young epilepsy patients have been reported to develop central apnea immediately following complex partial seizures (Blum et al. 2000; Jehi and Najm 2008). Apnea might also represent the only ictal symptom of temporal lobe seizures, especially in children (Lee et al. 1999; Singh et al. 1993). Pulmonary edema is frequently found in SUDEP patients at autopsy (Terrence et al. 1981) and is known to occur in other neurological disorders affecting the central neurorespiratory control centers such as head trauma and subarachnoid hemorrhage. 15.3.1.2â•…Experimental Evidence In a sheep animal model of SUDEP, one third of animals died from hypoventilation and had associated pulmonary edema at autopsy (Johnston et al. 1995). DBA/2 mice are another proposed SUDEP model because they exhibit respiratory arrest after audiogenic seizures (Tupal and Faingold 2006), and where sudden death was preventable by oxygenation without any change in seizure severity (Venit et al. 2004). 15.3.2â•…Cardiac Pathophysiology The most signiἀcant and widely discussed cardiac mechanism of SUDEP is cardiac arrhythmia precipitated by seizure discharges acting via the autonomic nervous system (Jehi and Najm 2008; Nei et al. 2004; Tomson et al. 2008b). 15.3.2.1â•…Clinical Evidence A wide spectrum of cardiac arrhythmias, such as ictal asystole, atrial ἀbrillation, repolarization abnormalities, and bundle branch blocks, has been reported during seizures (Blumhardt et al. 1986; Galimberti et al. 1996; Leung et al. 2006; Nei et al. 2004; Opherk et al. 2002). Ictal cardiac arrhythmias occurred in 42% of hospitalized epilepsy patients in one study, with the most common being an irregular series of abrupt rate changes toward the end of the electroencephalographic (EEG) seizure discharge (Blumhardt et al. 1986). In another study, analysis of R–R intervals during the ἀrst 10-s period of EEG discharge showed a signiἀcant early heart rate increase in 49% of seizures and an early heart rate

224 Sudden Death in Epilepsy: Forensic and Clinical Issues

reduction in 25.5% (Galimberti et al. 1996) (Figures 15.1 and 15.2). Certain clinical seizure characteristics have been correlated with the occurrence of ictal electrocardiographic (ECG) abnormalities. While one study found that mean seizure duration was longer in patients with ECG abnormalities than in those without such changes (Nei et al. 2000), others observed that ictal ECG abnormalities occurred more often and were more severe in generalized tonic–clonic seizures relative to complex partial seizures (Nei et al. 2000, 2004; Opherk et al. 2002). Those same clinical seizure characteristics were correlated with a higher risk of SUDEP (Langan et al. 2005), suggesting an interrelation between seizure semiology, ECG abnormalities, and SUDEP. 15.3.2.2â•…Experimental Evidence Electrical brain stimulation of the limbic system and insular cortex has repeatedly been shown to provoke heart rate changes, including bradycardia, tachycardia, and asystole (Leung et al. 2006). Some studies have even suggested a lateralized influence of the insulae on cardiovascular autonomic control with intraoperative stimulation of the left posterior insula eliciting a cardioinhibitory response and hypotension, and stimulation of the right anterior insula eliciting tachycardia and hypertension (Jehi and Najm 2008). Other studies have suggested a localization-related influence of the limbic system on cardiovascular responses with stimulation of the amygdala alone being insufficient to produce the ictal

Fp1–F7

50 µV

F7–SP1 SP1–T7 T7–P7 P7–O1 Fp2–F8 F8–SP2 SP2–T8 T8–P8 P8–O2 SP1–SP2 TP9–TP10 Fp1–F3 F3–C3 C3–P3 P3–O1 Fp2–F4 F4–C4 C4–P4 P4–O2 EKG1–EKG2 50 µV

Figure 15.1╇ Asystole documented during a left hemispheric seizure (10-s epoch) in a patient with a normal MRI, and subsequently pathologically proven malformation of cortical development. The arrow identifies movement artifact secondary to patient fall.

Intractable Epilepsy in the Setting of MCD as a Mechanism for SUDEP

225

EEG EMG (masseter) EMG (biceps) Pupils Electrodermogram C. H. R. B.

R. H. B. C. Time (min)

Interictal

Preictal

Ictal

Enuresis Extinction Postictal

Aura

Ictal

Postictal

Recuperation

Interictal

Figure 15.2╇ Autonomic changes during a motor seizure. Clear increases in the heart rate

(H) and blood pressure (B) correspond to a suppression in the respiratory rate (R) during the ictal phase. Autonomic changes continue well after the ictal event is finished. C, cystogram. (Reprinted From Gastaut, H., and Broughton, R., Epileptic Seizures: Electrographic Features, Diagnosis, and Treatment, Springfield, IL: Charles C Thomas Publisher, 1972: 28. With permission.)

tachycardia so commonly seen in epileptic seizures, suggesting that cortical involvement is essential for the increase in HR (Keilson et al. 1987). Such cortical stimulation-induced HR changes may explain how massive seizure-related discharges can affect the cardiac rhythm during the seizure itself. There is, however, also evidence of a baseline epilepsy-related autonomic dysfunction. A recent study showed a pronounced reduction in cardiac metaiodobenzylguanidine (MIBG) uptake in patients who had ictal asystole compared to other epileptic and nonepileptic controls, suggesting a postganglionic cardiac catecholamine disturbance. The authors then propose that epilepsy-related impaired sympathetic cardiac innervation limits adjustment and heart rate modulation and may thus increase the risk of asystole and, ultimately, SUDEP (Kerling et al. 2008).

15.4â•…Central Autonomic and Respiratory Control A tight, interconnected network exists throughout the neuraxis to control various elements of the cardiovascular autonomic system. A solid understanding of this network provides

226 Sudden Death in Epilepsy: Forensic and Clinical Issues

useful insights for consideration of a cardiac pathophysiology of SUDEP (Jehi and Najm 2008). Key components of the central cortical control of autonomic functions include the insula, the anterior cingulate gyrus, and the ventromedial prefrontal cortex. The insula represents the primary viscerosensory cortex, while the cingulate gyrus and prefrontal cortices form the premotor autonomic region. At the subcortical level, the hypothalamus provides the interface with the endocrine stimuli and triggers corresponding autonomic responses to maintain homeostasis. The amygdala, an integral component of the limbic system linking the cortical and subcortical centers, mediates the autonomic response to emotions. In addition to playing a key role in autonomic control, the insula, amygdala, cingulate gyrus, and prefrontal cortex also represent the most common foci of partial epilepsy, a concept that we will elaborate on further later and that may explain the frequent observation of autonomic changes in relation to epileptic seizures (Leung et al. 2006). Although central apnea has been observed with focal epileptiform activity alone (Lee et al. 1999), a more accepted hypothesis is that neurotransmitters mediating the brain’s own seizure-terminating mechanism could also be inhibiting respiratory centers in the brainstem and causing postictal apnea (Jehi and Najm 2008). Understanding the concepts of central autonomic pathways and respiratory control will facilitate the discussion of possible mechanisms of SUDEP in MCD. In the subsequent sections of this chapter, we will discuss how the general concepts of SUDEP discussed so far apply speciἀcally to intractable epilepsy in MCD patients.

15.5â•…Malformations of Cortical Development and SUDEP Alzheimer and Rahcke recognized the presence of aberrant cortical lamination in autopsies of patients with a history of chronic epilepsy almost a century ago. It was, however, the detailed report published by Taylor et al. (1971) that has since raised awareness of the role of misshapen dysmorphic (dysplastic) neurons in the setting of cortical architectural disorganization (both columnar and laminar) as the pathological substrate in some patients with drug-resistant epilepsy. The report identiἀed the possible role of disorientated and giant neurons (and balloon cells with eccentric nuclei) in temporal cortex resected from patients with temporal lobe epilepsy. In the past 20 years, studies showed that MCDs include a broad spectrum of architectural anomalies, such as cortical laminar disorganization, neuronal heterotopia in the subcortical white matter, the persistence of neurons in the superἀcial cortical layer (layer I of the neocortex), clustering of neurons in the gray matter, nodular heterotopia, and the presence of aberrant neurons such as giant neurons and balloon cells (Fauser et al. 2006; Lawson et al. 2005; Palmini et al. 2004; Tassi et al. 2002). The relationship between epilepsy and MCD in general, and focal cortical dysplasia (FCD) in particular, has been well established. In fact, 8–12% of cases with intractable epilepsy are attributed to MCD, whereas 14–26% of surgically treated cases have MCD (Tassi et al. 2002). 15.5.1â•…Classification of MCD MCDs are due to abnormalities in neuronal migration, proliferation, and/or differentiation that result in four distinct pathological subtypes: IA, IB, IIA, and IIB (Widdess-Walsh et al. 2005). Those various subtypes have different microscopic and imaging characteristics, as well as distinct outcomes with epilepsy surgery. Type I is characterized by a lack of

Intractable Epilepsy in the Setting of MCD as a Mechanism for SUDEP

227

dysmorphic neurons or balloon cells (abnormal cellular elements with a thin membrane; pale, glassy, and eosinophilic cytoplasm, eccentric nucleus, usually of increased size compared with gemistocytic astrocytes).. Type IA is characterized by patchy and isolated architectural abnormalities, mainly consisting of dyslamination and columnar disorganization, accompanied, or not, by other abnormalities of mild MCD. Type IB is characterized by architectural disorganization intermixed with giant or immature, but not dysmorphic, neurons. Most of these mild abnormalities defy in vivo imaging recognition. The only available evidence is from patients undergoing epilepsy surgery in whom such mild abnormalities were the only histopathological ἀnding. This suggests that at least some patients with type IA/B FCD can have medically refractory epilepsy (Palmini et al. 2004). Type II or Taylor-type FCD show dysmorphic neurons without (IIA) or with balloon cells (IIB) in the setting of diffuse cortical architectural disorganization. These are the focal lesions most commonly identiἀed on MRI. However, studies showed various degrees of imaging abnormalities in patients with Taylor-type FCD ranging from thickening of the cortical mantle to focal and severe fluid-attenuated inversion recovery (FLAIR) signal abnormalities (Palmini et al. 2004). Figure 15.3 illustrates the various histopathological ἀndings of those four MCD subtypes. 15.5.2â•…Neuroimaging of MCD MRI can be either normal (most often in type I), despite the use of high-resolution techniques, or may demonstrate one or more of the following characteristics: (1) focal areas of increased cortical thickness; (2) blurring of the cortex (gray)/white matter junction; (3)€ increased signal on T2-weighted proton density or FLAIR sequences (more likely to occur in balloon cell-containing lesions); and (4) extension of cortical tissue with increased FLAIR signal from the surface to the periventricular region (transmantle dysplasia) (Palmini et al. 2004; Ruggieri et al. 2004; Tassi et al. 2002; Widdess-Walsh et al. 2005). 15.5.3â•… Relevance of MCD Classification and Imaging to SUDEP Several studies have shown that MRI and histopathological ἀndings in MCD correlate with seizure outcome following resective surgery for intractable epilepsy, with better outcomes

Normal

Type I

Type IIA

Type IIB

Figure 15.3╇ Cresyl echt violet staining of sections representing normal neocortex and type I,

type IIA, and type IIB malformations of cortical development; scale bar, 100 µm. (Palmini, A. et al., Neurology, 62 (6 Suppl 3), S2–S8, 2004. With permission.)

228 Sudden Death in Epilepsy: Forensic and Clinical Issues

observed in patients with clear MRI lesions, and more severe histopathological changes (mainly type IIB), as opposed to those with normal MRI (types IA and IB) (Jeha et al. 2007; Tassi et al. 2002; Widdess-Walsh et al. 2005). The poorer outcome following surgery in nonlesional MCD cases is mainly attributed to difficulties localizing the epileptogenic focus and/or its incomplete resection due to more diffuse cytoarchitectural changes that may be invisible on MRI (Jeha et al. 2007). This is relevant in a discussion about SUDEP because persistent seizures are the main risk factor for SUDEP, as extensively discussed earlier in this chapter (Jehi and Najm 2008; Tomson 2000; Tomson et al. 2008b). In fact, it has been shown in a recent prospective study that persistent seizures speciἀcally following failed epilepsy surgery carry a signiἀcantly high risk of sudden death, estimated at 6.3 cases per 1000 patient years (Sperling et al. 2005). As such, patients with intractable epilepsy, MCD, and normal imaging may be particularly more vulnerable to SUDEP.

15.6â•…Localization of MCD and SUDEP Several studies showed a predisposition of speciἀc MCD pathological subtypes to localize to certain brain regions. In a review of 145 cases of MCD operated on for intractable epilepsy at Cleveland Clinic Epilepsy Center between 1990 and 2002, we found that pathological subtypes IIA and IIB were predominantly frontal in location and had a more severe epilepsy syndrome than patients with subtypes IA and IB. Patients with subtype IA FCD had less severe, later onset epilepsy that was predominantly located in the temporal lobe (Widdess-Walsh et al. 2005). This is similar to ἀndings of another review of 52 MCD surgical cases of intractable epilepsy where patients with architectural dysplasia alone (type IA) had lower seizure frequency than those with Taylor-type dysplasia, and the epileptogenic zone was mainly in the temporal lobe, while in patients with Taylor-type dysplasia, the epileptogenic zone was mainly extratemporal, predominantly frontal (Tassi et al. 2002). Beyond these anatomical distributions of MCD in relation to histopathology, it remains€to be shown whether MCDs have a speciἀc affinity to localize in brain regions particularly relevant in the proposed mechanisms of SUDEP, such as the insula, cingulate gyrus, orbitoÂ�frontal regions, or amygdalae. We will review now the limited available data concerning this point. 15.6.1â•…Case Reports A few case reports have indeed documented ictal cardiac or respiratory changes in patients with MCD involving brain regions thought to be part of the central autonomic control centers. Known examples include one case report of a three-and-a-half-year-old child with episodic sinus bradycardia during habitual seizures and prolonged interictal discharges due to FCD in the anterior two-thirds of the insula and the inferior frontal cortex (Seeck et al. 2003). In another case report of a 30-year-old man who was found dead on arrival to the hospital following an hour-and-a-half of complex partial seizure, post mortem examination showed bilateral occipital frontal polymicrogyria in the brain and chronic interstitial and perivascular ἀbrosis in the heart without previous vascular risk factors, a ἀnding which the authors attributed to possibly chronic repetition of seizures (Ribacoba Montero et al. 2002).

Intractable Epilepsy in the Setting of MCD as a Mechanism for SUDEP

229

15.6.2â•…Cleveland Clinic Epilepsy Center Experience We recently reviewed 3841 patients evaluated with prolonged video-EEG monitoring in our epilepsy monitoring unit between 1990 and 2005 and identiἀed 301 mortalities (Jehi et al., in preparation). Of 223 mortalities evaluated so far, 29 died of SUDEP, including six who had MCD. Of these six cases, four had dual pathology with associated hippocampal sclerosis affecting their mesial temporal structures (three with type IA MCD, including two in their temporal lobe, and another in his lateral temporal and inferior frontal lobes; and one with MRI-visible dysplasia affecting his posterior temporal and angular gyrus), one had posterior left temporal lobe closed lip schizencephaly, and one had a temporoparietal ictal onset zone on invasive EEG with normal imaging. Five of the above patients had epilepsy surgery (some even up to three times), but continued to have seizures postoperatively. The epileptogenic zone thus localized to the peri-insular region in all of our MCD cases with SUDEP suggests a direct anatomical disruption of baseline physiologic autonomic centers in the insula as a possible mechanism of SUDEP. Although the observation is clear, a direct relationship between the suspected anatomical location of the FCD and epileptogenicity in the insular region and SUDEP is difficult to prove. An alternative hypothesis in this group may be that the increased risk of SUDEP is simply due to continuous seizures following a failed epilepsy surgery.

15.7â•…Mechanisms of Epileptogenicity in MCD and SUDEP Multiple mechanisms of epileptogenicity in MCD have been proposed, mainly revolving around two central concepts: (1) localized disruption of excitatory and inhibitory neurotransmission and (2) diffuse disruption of normal neural circuitry. We will now elaborate on those two general principles and discuss how they might be related to SUDEP. 15.7.1â•…Localized Disruption of Excitatory and Inhibitory Neurotransmission Dysplastic lesions have a high degree of intrinsic epileptogenicity: In one study, up to 67% of patients with MCD manifested continuous or frequent rhythmic epileptogenic discharges recorded directly from their cortical dysplastic lesions during intraoperative electrocorticography, as opposed to only 2.5% of patients with intractable partial epilepsy associated with other types of structural lesions (Palmini et al. 1995). Data collected through immunocytochemical and clinical studies support an increase in excitatory amino acid neurotransmission and an overall decrease in intralesional and perilesional inhibition as a signiἀcant contributor to this high degree of intrinsic epileptogenicity (Avoli et al. 1999; Najm et al. 2004; Palmini et al. 1995, 2004). 15.7.1.1â•…Experimental Evidence Ferrer et al. (1992) ἀrst showed abnormalities in the morphology and distribution of localcircuit (inhibitory) neurons in FCD, and hypothesized they may have a pivotal role in the appearance and prolongation of electrical discharges. Since then, Najm et al. (2000) correlated signiἀcantly higher immunoreactivity of the excitatory N-methyl-d-aspartate (NMDA) receptor (NR) 2A/B in both the dysplastic somata and all their dendritic processes

230 Sudden Death in Epilepsy: Forensic and Clinical Issues

with in vivo epileptic activity recorded through subdural EEG, whereas White et al. (2001) observed reduced levels of the inhibitory gamma aminobutyric acid A (GABA-A) receptors alpha1 and alpha2 mRNA in both dysplastic neurons and giant cells compared to control neurons. Neurotransmission changes in the dysplastic cortex extend, however, beyond the classical excitatory and inhibitory NMDA and GABA systems. Trottier et al. (1996) found evidence of serotonergic hyperinnervation and altered patterns of the catecholaminergic innervation in dysplastic cortex of epilepsy patients with MCD, as opposed to tissue from patients with cryptogenic neocortical epilepsy and normal controls (Trottier et al. 1996). 15.7.1.2â•…Clinical Evidence Most patients diagnosed by imaging studies as having lesions identiἀed as type IIA/B FCD have medically intractable partial epilepsy, with frequently disabling motor and secondary generalized seizures (Fauser et al. 2006; Lawson et al. 2005; Palmini et al. 2004; Tassi et al. 2002; Widdess-Walsh et al. 2005). Many patients have a history of status epilepticus, including epilepsia partialis continua, and scalp EEG and acute electrocorticography often show continuous spiking or other highly epileptogenic patterns, attesting to some type of re-entrant excitatory circuitry unopposed by faulty inhibition (Avoli et al. 1999; Ferrer et al. 1992; Palmini et al. 1995). 15.7.1.3â•…Relevance to SUDEP It is reasonable to hypothesize that the previously discussed disturbances in excitatory versus inhibitory balance leading to epileptogenicity of dysplastic tissue may also translate to disturbances in the normal sympathetic versus parasympathetic balance if the MCD happens to involve cortical regions crucial for central autonomic control. There is however no information available currently on the status of either NMDA or GABA receptors in SUDEP victims. Data presented previously supporting a role for serotonin in the modulation of sudden death induced by audiogenic seizures in DBA/2 mice via causing respiratory arrest (Tupal and Faingold 2006) may be related to the altered serotonergic pathways described speciἀcally in MCD (Trottier et al. 1996), but further investigation is needed. 15.7.2â•…Diffuse Disruption of Normal Neural Circuitry This hypothesis is based on the concept that rather than intrinsic epileptogenicity of the dysplastic lesion itself, the main mechanism of epilepsy, in the context of MCD, is disruption of neuronal circuitry extending far beyond the lesion itself. Following this idea, focal epileptogenesis associated with MCD, as opposed with postnatally acquired lesions such as those due to tumors or trauma, is best conceptualized as a disorder of widespread and patchy disturbance of cortical networks. This developmental perspective implies that the epileptogenic region in MCD is rarely discrete, even in patients with focal dysplasia, and may include remote cortical or subcortical areas (Duchowny et al. 2000). 15.7.2.1â•…Experimental Evidence Subtle structural abnormalities are seen beyond the clearly visible dysplastic lesion in many patients with MCD. Measurements of cerebral surface area and volume reveal widespread and unusually extensive anatomic changes throughout extralesional gray and subcortical white matter in a high proportion of patients with FCD (Sisodiya et al. 1997). Similarly, prenatal damage during critical maturational stages of primates resulted in anomalous

Intractable Epilepsy in the Setting of MCD as a Mechanism for SUDEP

231

sulcation and reorganization at remote cortical sites in both cerebral hemispheres (Goldman 1978). Functional neuroimaging abnormalities on proton MR spectroscopy and flumazenil PET studies of benzodiazepine receptors in FCD reveal abnormal activation patterns extending far beyond the lesional boundary (Richardson et al. 1996). Ictal SPECT in tuberous sclerosis complex also shows increased blood flow in the perituberal penumbra, suggesting dysfunction beyond the tuber (Koh et al. 1999). 15.7.2.2â•…Clinical Evidence Neurological deἀcits in patients with MCD are often more extensive than expected based on the extent of the lesion visible on MRI. Besides epilepsy, patients with MCD often suffer from cognitive impairment, and are prone to attention and behavioral problems and autism. Verbal IQ scores are decreased even in patients with very focal lesions of the dominant cerebral hemisphere (Duchowny et al. 2000). These ἀndings may be attributed to dysfunction beyond the lesion itself. However, there are other factors to consider. Younger age of seizure onset and larger lesions are associated with diminished cognitive outcome. Epileptogenic activity and medication also make it difficult to separate cause from effect. However, not all patients with frequent seizures and chronic medications exhibit cognitive disturbance (Duchowny et al. 2000; Fauser et al. 2006; Lawson et al. 2005) and the extent of the MRI lesion itself does not always correlate with cognitive outcome suggesting an additional functional mechanism to the disturbances seen with MCD besides the structural disturbance visible on routine imaging. 15.7.2.3â•…Relevance to SUDEP Parallels can be drawn between the dysfunctional connectivity alluded to previously in MCD, and the disruptions in the sympathetic and parasympathetic networks felt to be contributing to SUDEP. This issue needs, however, to be further investigated.

15.8â•…Conclusion SUDEP is the leading cause of death in chronic uncontrolled epilepsy and is a devastating complication, usually occurring in young, otherwise healthy individuals with persistent seizures or poor compliance to antiepileptic medications. Despite signiἀcantly increased interest in SUDEP over the past few decades, the exact mechanisms leading to its occurrence remain unknown. On the other hand, MCD have been increasingly recognized as a common and very signiἀcant cause of epilepsy that still remains difficult to control with epilepsy surgery. As such, MCDs may provide a unique substrate for further study of SUDEP mechanisms and possible preventative interventions. A lot remains unknown though, and areas of further research may include: 1. Determination of the incidence of SUDEP in MCD speciἀcally, as compared to other epilepsy etiologies, to evaluate whether this patient population is particularly more vulnerable to sudden death. 2. Investigation of the autonomic function in patients with MCD as compared to those with other structural abnormalities and epilepsy to clarify whether strategically located MCD can lead to distant autonomic dysfunction, possibly contributing to cardiac arrhythmias hypothesized in SUDEP.

232 Sudden Death in Epilepsy: Forensic and Clinical Issues

3. Characterization of the immunocytochemical characteristics of brain tissue from patients with SUDEP and MCD, especially excitatory and inhibitory pathways, and serotonergic pathways to investigate further the potential for common mechanisms between intractable epilepsy in MCD and SUDEP.

References Avoli, M., A. Bernasconi, D. Mattia, A. Olivier, and G. G. Hwa. 1999. Epileptiform discharges in the human dysplastic neocortex: In vitro physiology and pharmacology. Ann Neurol 46 (6): 816–826. Blum, A. S., J. R. Ives, A. L. Goldberger, I. C. Al-Aweel, K. B. Krishnamurthy, F. W. Drislane, and D.€L.€Schomer. 2000. Oxygen desaturations triggered by partial seizures: Implications for cardiopulmonary instability in epilepsy. Epilepsia 41 (5): 536–541. Blumhardt, L. D., P. E. Smith, and L. Owen. 1986. Electrocardiographic accompaniments of temporal lobe epileptic seizures. Lancet 1 (8489): 1051–1056. Duchowny, M., P. Jayakar, and B. Levin. 2000. Aberrant neural circuits in malformations of cortical development and focal epilepsy. Neurology 55 (3): 423–428. Fauser, S., H. J. Huppertz, T. Bast, K. Strobl, G. Pantazis, D. M. Altenmueller, B. Feil et al. 2006. Clinical characteristics in focal cortical dysplasia: A retrospective evaluation in a series of 120 patients. Brain 129 (Pt 7): 1907–1916. Ferrer, I., M. Pineda, M. Tallada, B. Oliver, A. Russi, L. Oller, R. Noboa, M. J. Zujar, and S. Alcantara. 1992. Abnormal local-circuit neurons in epilepsia partialis continua associated with focal cortical dysplasia. Acta Neuropathol 83 (6): 647–652. Galimberti, C. A., E. Marchioni, F. Barzizza, R. Manni, I. Sartori, and A. Tartara. 1996. Partial epileptic seizures of different origin variably affect cardiac rhythm. Epilepsia 37 (8): 742–747. Goldman, P. S. 1978. Neuronal plasticity in primate telencephalon: Anomalous projections induced by prenatal removal of frontal cortex. Science 202 (4369): 768–770. Jeha, L. E., I. Najm, W. Bingaman, D. Dinner, P. Widdess-Walsh, and H. Luders. 2007. Surgical outcome and prognostic factors of frontal lobe epilepsy surgery. Brain 130 (Pt 2): 574–584. Jehi, L., and I. M. Najm. 2008. Sudden unexpected death in epilepsy: Impact, mechanisms, and prevention. Cleve Clin J Med 75 (Suppl 2): S66–S70. Johnston S. C., J. K. Horn, J. Valente, and R. P. Simon. 1995. The role of hypoventilation in a sheep model of epileptic sudden death. Ann Neurol 37 (4): 531–537. Keilson, M. J., W. A. Hauser, J. P. Magrill, and M. Goldman. 1987. ECG abnormalities in patients with epilepsy. Neurology 37 (10): 1624–1626. Kerling, F., I. Blumcke, and H. Stefan. 2008. Pitfalls in diagnosing limbic encephalitis—A case report. Acta Neurol Scand 118 (5): 339–342. Koh, S., P. Jayakar, T. Resnick, L. Alvarez, R. E. Liit, and M. Duchowny. 1999. The localizing value of ictal SPECT in children with tuberous sclerosis complex and refractory partial epilepsy. Epileptic Disord 1 (1): 41–46. Langan, Y., L. Nashef, and J. W. Sander. 2005. Case-control study of SUDEP. Neurology 64 (7): 1131–1133. Lawson, J. A., S. Birchansky, E. Pacheco, P. Jayakar, T. J. Resnick, P. Dean, and M. S. Duchowny. 2005. Distinct clinicopathologic subtypes of cortical dysplasia of Taylor. Neurology 64 (1): 55–61. Lee, H. W., S. B. Hong, W. S. Tae, D. W. Seo, and S. E. Kim. 1999. Partial seizures manifesting as apnea only in an adult. Epilepsia 40 (12): 1828–1831. Leestma, J. E., J. F. Annegers, M. J. Brodie, S. Brown, P. Schraeder, D. Siscovick, B. B. Wannamaker, P. S. Tennis, M. A. Cierpial, and N. L. Earl. 1997. Sudden unexplained death in epilepsy: Observations from a large clinical development program. Epilepsia 38 (1): 47–55. Leung, H., P. Kwan, and C. E. Elger. 2006. Finding the missing link between ictal bradyarrhythmia, ictal asystole, and sudden unexpected death in epilepsy. Epilepsy Behav 9 (1): 19–30.

Intractable Epilepsy in the Setting of MCD as a Mechanism for SUDEP

233

Monte, C. P., J. B. Arends, I. Y. Tan, A. P. Aldenkamp, M. Limburg, and M. C. de Krom. 2007. Sudden unexpected death in epilepsy patients: Risk factors. A systematic review. Seizure 16 (1): 1–7. Najm, I., Z. Ying, T. Babb, P. B. Crino, R. Macdonald, G. W. Mathern, and R. Spreaἀco. 2004. Mechanisms of epileptogenicity in cortical dysplasias. Neurology 62 (6 Suppl 3): S9–S13. Najm, I. M., Z. Ying, T. Babb, A. Mohamed, J. Hadam, E. LaPresto, E. Wyllie et al. 2000. Epileptogenicity correlated with increased N-methyl-d-aspartate receptor subunit NR2A/B in human focal cortical dysplasia. Epilepsia 41 (8): 971–976. Nashef, L., D. R. Fish, J. W. Sander, and S. D. Shorvon. 1995. Incidence of sudden unexpected death in an adult outpatient cohort with epilepsy at a tertiary referral centre. J Neurol Neurosurg Psychiatry 58 (4): 462–464. Nashef, L., F. Walker, P. Allen, J. W. Sander, S. D. Shorvon, and D. R. Fish. 1996. Apnoea and bradycardia during epileptic seizures: Relation to sudden death in epilepsy. J Neurol Neurosurg Psychiatry 60 (3): 297–300. Nashef, L., N. Hindocha, and A. Makoff. 2007. Risk factors in sudden death in epilepsy (SUDEP): The quest for mechanisms. Epilepsia 48 (5): 859–871. Nei, M., R. T. Ho, B. W. Abou-Khalil, F. W. Drislane, J. Liporace, A. Romeo, and M. R. Sperling. 2004. EEG and ECG in sudden unexplained death in epilepsy. Epilepsia 45 (4): 338–345. Nei, M., R. T. Ho, and M. R. Sperling. 2000. EKG abnormalities during partial seizures in refractory epilepsy. Epilepsia 41 (5): 542–548. Nilsson, L., T. Tomson, B. Y. Farahmand, V. Diwan, and P. G. Persson. 1997. Cause-speciἀc mortality in epilepsy: A cohort study of more than 9,000 patients once hospitalized for epilepsy. Epilepsia 38 (10): 1062–1068. Opherk, C., J. Coromilas, and L. J. Hirsch. 2002. Heart rate and EKG changes in 102 seizures: Analysis of influencing factors. Epilepsy Res 52 (2): 117–127. Palmini, A., A. Gambardella, F. Andermann, F. Dubeau, J. C. da Costa, A. Olivier, D. Tampieri et al. 1995. Intrinsic epileptogenicity of human dysplastic cortex as suggested by corticography and surgical results. Ann Neurol 37 (4): 476–487. Palmini, A., I. Najm, G. Avanzini, T. Babb, R. Guerrini, N. Foldvary-Schaefer, G. Jackson et al. 2004. Terminology and classiἀcation of the cortical dysplasias. Neurology 62 (6 Suppl 3): S2–S8. Pedley, T. A., and W. A. Hauser. 2002. Sudden death in epilepsy: A wake-up call for management. Lancet 359 (9320): 1790–1791. Ribacoba Montero, R., C. Garcia Pravia, A. Astudillo, and J. Salas Puig. 2002. Bilateral fronto-occipital polymicrogyria and epilepsy. Seizure 11 (Suppl A): 298–302. Richardson, M. P., M. J. Koepp, D. J. Brooks, D. R. Fish, and J. S. Duncan. 1996. Benzodiazepine receptors in focal epilepsy with cortical dysgenesis: An 11C-flumazenil PET study. Ann Neurol 40 (2): 188–198. Ruggieri, P. M., I. Najm, R. Bronen, M. Campos, F. Cendes, J. S. Duncan, H. G. Weiser, and W. H. Theodore. 2004. Neuroimaging of the cortical dysplasias. Neurology 62 (6 Suppl 3): S27–S29. Schraeder, P. L., K. Delin, R. L. McClelland, and E. L. So. 2009. A nationwide survey of the extent of autopsy in sudden unexplained death in epilepsy. Am J Forensic Med Pathol 30 (2): 123–126. Seeck, M., S. Zaim, V. Chaves-Vischer, O. Blanke, M. Maeder-Ingvar, M. Weissert, and E. Roulet. 2003. Ictal bradycardia in a young child with focal cortical dysplasia in the right insular cortex. Eur J Paediatr Neurol 7 (4): 177–181. Singh, B., A. al Shahwan, and S. M. al Deeb. 1993. Partial seizures presenting as life-threatening apnea. Epilepsia 34 (5): 901–903. Sisodiya, S. M., N. Moran, S. L. Free, N. D. Kitchen, J. M. Stevens, W. F. Harkness, D. R. Fish, and S. D. Shorvon. 1997. Correlation of widespread preoperative magnetic resonance imaging changes with unsuccessful surgery for hippocampal sclerosis. Ann Neurol 41 (4): 490–496. Sperling, M. R., A. Harris, M. Nei, J. D. Liporace, and M. J. O’Connor. 2005. Mortality after epilepsy surgery. Epilepsia 46 (Suppl 11): 49–53.

234 Sudden Death in Epilepsy: Forensic and Clinical Issues Tassi, L., N. Colombo, R. Garbelli, S. Francione, G. Lo Russo, R. Mai, F. Cardinale et al. 2002. Focal cortical dysplasia: Neuropathological subtypes, EEG, neuroimaging, and surgical outcome. Brain 125 (Pt 8): 1719–1732. Taylor, D. C., M. A. Falconer, C. J. Bruton, and J. A. Corsellis. 1971. Focal dysplasia of the cerebral cortex in epilepsy. J Neurol Neurosurg Psychiatry 34 (4): 369–387. Tennis, P., T. B. Cole, J. F. Annegers, J. E. Leestma, M. McNutt, and A. Rajput. 1995. Cohort study of incidence of sudden unexplained death in persons with seizure disorder treated with antiepileptic drugs in Saskatchewan, Canada. Epilepsia 36 (1): 29–36. Terrence, C. F., G. R. Rao, and J. A. Perper. 1981. Neurogenic pulmonary edema in unexpected, unexplained death of epileptic patients. Ann Neurol 9 (5): 458–464. Tomson, T. 2000. Mortality in epilepsy. J Neurol 247 (1): 15–21. Tomson, T., L. Nashef, and P. Ryvlin. 2008a. Sudden unexpected death in epilepsy: Current knowledge and future directions. Lancet Neurol 7 (11): 1021–1031. Tomson, T., L. Nashef, and P. Ryvlin. 2008b. Sudden unexpected death in epilepsy: Current knowledge and future directions. Lancet Neurol 7 (11): 1021–1031. Trottier, S., B. Evrard, J. P. Vignal, J. M. Scarabin, and P. Chauvel. 1996. The serotonergic innervation of the cerebral cortex in man and its changes in focal cortical dysplasia. Epilepsy Res 25 (2): 79–106. Tupal, S., and C. L. Faingold. 2006. Evidence supporting a role of serotonin in modulation of sudden death induced by seizures in DBA/2 mice. Epilepsia 47 (1): 21–26. Venit, E. L., B. D. Shepard, and T. N. Seyfried. 2004. Oxygenation prevents sudden death in seizureprone mice. Epilepsia 45 (8): 993–996. White, R., Y. Hua, B. Scheithauer, D. R. Lynch, E. P. Henske, and P. B. Crino. 2001. Selective alterations in glutamate and GABA receptor subunit mRNA expression in dysplastic neurons and giant cells of cortical tubers. Ann Neurol 49 (1): 67–78. Widdess-Walsh, P., C. Kellinghaus, L. Jeha, P. Kotagal, R. Prayson, W. Bingaman, and I. M. Najm. 2005. Electro-clinical and imaging characteristics of focal cortical dysplasia: Correlation with pathological subtypes. Epilepsy Res 67 (1–2): 25–33.

16

Neurogenic Cardiac Arrhythmias Howan Leung Anne Y. Y. Chan

Contents 16.1 Introduction 16.2 What Can We Learn from Electrical Brain Stimulation Studies about the Cortical Control of the Autonomic System? 16.3 What Can We Learn from Functional Magnetic Resonance Imaging Studies Demonstrating Cortical Control of the Autonomic System? 16.4 Illustrating Ictal Bradyarrhythmia and Asystole with Scalp EEG Data 16.5 Illustrating Ictal Bradyarrhythmia and Asystole with Intracranial EEG Data (Diagram 2) 16.6 Can Ictal Bradyarrhythmia Enlighten Us about the Various Mechanisms of Neurogenic Cardiac Arrhythmia? 16.7 Could There Be a Link between Ictal Bradyarrhythmia and SUDEP? 16.8 What Lies in the Future for Researchers? Acknowledgments References

235 236 238 239 240 241 244 247 247 248

16.1╅Introduction The analysis of abnormal changes in heart rate is not only a matter of concern for cardi� ologists, but also a topic of importance among neurologists. Through the study of patients with epilepsy, which is characterized by recurrent abnormal and excessive discharge of a set of neurons in the brain (Commission on Epidemiology and Prognosis 1993), cardiac arrhythmias of neurogenic origin may be examined. In fact, it is a recognized clinical observation that during seizures there might be an increase or decrease in the rate of cardiac rhythm. Through the functional study of the brain during investigations and work-up of epilepsy patients, the cortical control of cardiac rhythmic activities may be uncovered. Blumhardt et al. (1986) analyzed the electrocardiograms (ECGs) of 26 patients with temporal lobe epilepsy and showed that ictal cardiac arrhythmias occurred in 42% of patients. Ictal bradyarrhythmia and ictal asystole, in particular, have received much attention recently because of the postulation that autonomic manifestations of seizures could be one of the mechanisms underlying sudden unexpected death in epilepsy (SUDEP), although the evidence supporting this remains fragmented. This short chapter aims to review the potential role of cardiac arrhythmia in the pathogenesis of SUDEP by examining the data pertaining to the cortical control of the autonomic system, the clinical observation of potentially life-threatening ictal bradyarrhythmia, and various proposed mechanisms for SUDEP. 235

236 Sudden Death in Epilepsy: Forensic and Clinical Issues

16.2â•…What Can We Learn from Electrical Brain Stimulation Studies about the Cortical Control of the Autonomic System? The central representation of autonomic states has long suspected, as stipulated by the James–Lange theory, that arousal and emotion were closely related (James 1894). More information about the representation of the autonomic system above the level of the brain stem was reported more than 50 years ago beginning with studies in primates and other animals (Kaada 1951). The limbic structures are thought to be the principal mediators of this function and the candidate areas include cingulate gyrus, amygdala, and insular and orbitofrontal cortex. In Kaada’s experiments, stimulation of the anterior portion of the cingulate gyrus in monkeys showed inhibition and arrest of respiration. Penἀeld and Jasper (1954) carried out experiments in humans in the 1950s and showed that stimulation of the cingulate gyrus in its anterior and inferior portion produced apnea, which could not be overcome even with a conscious effort. In addition, stimulation of the uncus and of the right anterior margin of insular cortex also produced a similar effect on respiration. The cortical influence on cardiovascular responses has been likewise demonstrated in various animal studies (Kaada 1951; Ward 1948; Burns and Wyss 1985; Chefer et al. 1997; Healy and Peck 1997) and in humans. This was demonstrated by delivering electrical stimulation to psychotic patients before bilateral fractional ablation of the cortex (Pool and Ransohoff 1949). A total of 12 patients were tested with bilateral cingulate stimulation, of which three cases had heart rate elevation, seven had heart rate reduction, and two cases showed no change. The authors did not explain what factors determined the direction of change nor whether the change in heart rate was unidirectional in any given patient. Van Buren (1961) performed electric stimulation to right orbitofrontal and right mesial temporal areas in 11 patients before pituitary surgery. The stimulation of the right orbitofrontal area was associated with a slight drop in heart rate, whereas stimulation of the right mesial temporal region produced a mixture of increase and decrease in heart rate. In an often-quoted study by Oppenheimer et al. (1992), electrical stimulation of the insular regions of ἀve epileptic patients (three with origin on the right side and two on the left side) was associated with changes in both heart rate and blood pressure. Electrical stimulation with 5 to 10 V was carried out with a pulse duration of 2 ms at 40 Hz. They found that bradycardia and depressor responses were more frequently produced than tachycardia when the left insular cortex was stimulated. The converse was noted when the right insular cortex was stimulated. Of interest, the amplitude of the bradycardia was greater on stimulation of the left posterior insular cortex compared with the left anterior insular cortex. This study was able to demonstrate that there is a left–right difference in the cortical control of heart rate (the implications of this ἀnding for SUDEP will be discussed in Section 16.7). Electrical stimulations are sometimes performed as part of the presurgical evaluation of medically refractory epilepsy, particularly in determining cortical eloquent areas and sensorimotor areas, or for the reproduction of habitual seizures. The amount of electrical stimulation given in those situations may vary from 1 to 10 mA depending on the apparatus used and the interpatient variation of parenchymal impedance. The frequency of stimulation may be higher for stimulation of eloquent areas but lower for motor areas to minimize discomfort. The duration of electrical stimulation is usually short (e.g., 40 s into the typical semiology of a complex partial seizure. However, EEG changes were only typical of cerebral hypoperfusion and no clinical predisposing factors can be identiἀed (Schuele et al. 2007).

16.5â•…Illustrating Ictal Bradyarrhythmia and Asystole with Intracranial EEG Data (Diagram 2) Ictal bradyarrhythmia captured on invasive EEG monitoring can give invaluable information, although the event is actually one that occurs rarely. Among the few reported cases (Altenmuller et al. 2004; Broglin and Bancaud 1991; Munari et al. 1995; Devinsky et al. 1997; Manitius-Robeck et al. 1998; Kahane et al. 1999; Rossetti et al. 2005), only four provided sufficient clinical information enabling correlation between ictal EEG and onset of bradyarrhythmia. In one of these reports, a patient with right hippocampal atrophy was described who underwent presurgical evaluation with placement of left temporal subdural grid and strip electrodes, right temporal subdural electrodes, and right orbitofrontal and

Neurogenic Cardiac Arrhythmias

241

frontal electrodes. Subsequent evaluations led to left temporal resective surgery rather than the right. An 8-s period of asystole took place after the onset of seizure and the intracranial recording registered a left temporal onset, although the asystole did not start until the seizure had spread to the right temporal lobe (Devinsky et al. 1997). In another case, a patient with hypothalamic hamartoma was implanted with 10 electrodes exploring the right frontal, central, and temporal cortices, two electrodes within the left frontal lobe, and one implanted within the lesion. The patient had dacrystic (i.e., crying seizures with intracranial EEG monitoring showing epileptiform activities beginning from and remaining localized in the hamartoma). An episode of ictal asystole, however, was registered with the onset of epileptiform discharges in the right frontocentral and temporal neocortical areas rather than the hypothalamic mass itself (Kahane et al. 1999). It was noteworthy that this patient was left-handed. In another report, a patient had foci of encephalomalacia in the right inferior frontal gyrus, temporal pole, and anterior parietal lobe. Invasive monitoring was carried out with bilateral temporal depth electrodes and right dorsolateral frontoparietal subdural strips. Asystole occurred in a seizure episode with left temporal seizure onset, and sharing similarity with the ἀrst case report, the actual phenomenon of asystole only occurred after an interhemispheric spread to opposite mesial structures. The proposed route of spread to the opposite side was thought to be via the anterior commissure or the corpus callosum. Although no intracranial electrode was placed in the neocortical portions of the left temporal lobe, the surface EEG showed rhythmic activity over the left lateral temporal region before involvement of the opposite mesial structure so that involvement of the left insular region before interhemispheric spread remained possible. Intracranial electrodes conἀrmed that the right frontal lobe was spared (Rossetti et al. 2005). In the last of the four reports, a patient with left temporal lobe epilepsy was implanted with left lateral and basal temporal subdural grid (x1) and strips (x2) with an additional left temporal depth electrode. Further evaluation eventually led to left temporal lobe surgery, the pathology of which was shown to be glioneuronal hamartoma. High-grade AV blocks were observed with left temporal seizure onset that was recorded before intracranial electrode placement. A left basal and anterior lateral temporal lobe seizure onset was thought to be present during the ictal cardiac event as this was subsequently inferred from the intracranial recording. Epileptic discharges were thought to be localized during the event of AV block on the basis of scalp EEG ἀndings, although the time lag from seizure onset and the placement of intracranial recording at a separate time may still raise doubts about a possible spread. Electrical stimulation to this region also reproduced AV block once, as described previously in the section about human brain stimulation studies (Altenmuller et al. 2004).

16.6â•…Can Ictal Bradyarrhythmia Enlighten Us about the Various Mechanisms of Neurogenic Cardiac Arrhythmia? Factors predisposing an epilepsy patient to ictal bradyarrhythmia remain unknown. It is tempting to suspect that certain seizure mechanisms may predispose to ictal bradyarrhythmia and there might be additional pathological factors contributing to this, such as an aberrant condition in the mediators of cortical autonomic control (e.g., the vagus nerve), concomitant antiepileptic drugs thought to be arrhythmogenic, imbalance between the

242 Sudden Death in Epilepsy: Forensic and Clinical Issues

sympathetic and parasympathetic divisions of the autonomic nervous system, and/or a cardiac substrate (i.e., underlying disease process affecting the cardiac functions). The observation of bradyarrhythmia associated with temporal and frontal lobe onset seizures may lend support to the hypothesis that activation of these regions is part of the mechanism, and in particular, the operculo–insulo–mesiotemporal–orbital complex (Mufson and Mesulam 1982). However, none of the intracranial reports of ictal bradyarrhythmia gave information about involvement of the left insular cortex as the electrodes are seldom placed in this area in clinical practice. Selection bias is also possible, given the fact that there is a preponderance of temporal and frontal epilepsies in specialized centers. The concept of recognizing a pattern of seizure spread during ictal bradyarrhythmia is important, as our analysis of intracranial EEG data showed that seizures beginning from the central autonomic network on the dominant hemisphere were present in the episodes with ictal bradycardia. In addition, a spread to the contralateral central autonomic network was frequently witnessed. The timing of onset of bradycardia in relation to seizure onset was supporting evidence for this. We suspect that the cerebral dominance may alter the “internal wiring” of the central autonomic network, hence a patient with left cerebral dominance may be observed with a left-to-right spread, and vice versa for a patient with right cerebral dominance, although in the intracranial reports there was no patient with bilateral cerebral dominance. This idea of cerebral dominance may be mirrored by the fact that language representation can be lateralized to either hemisphere or bilaterally during the patient’s developmental process. However, we did not know exactly what effect the lesions had on the mechanisms producing bradycardia, as the spreading of epileptiform discharges to the opposite hemisphere cannot be easily correlated with whether the lesion was ipsilateral or contralateral to the seizure onset using the intracranial data. In the only case with hypothalamic hamartoma, there was no seizure spread required for the process and the bradycardia was registered at the onset of seizure. In the case where the lesion was on the same side as seizure onset, no spreading was noted (Altenmuller et al. 2004). In the two cases where the lesion was contralateral to the seizure onset, spreading can be witnessed (Devinsky et al. 1997; Rossetti et al. 2005). We also suspect that the presence of a lesion may disrupt the original central autonomic network, keeping the speciἀc network required for the production of bradycardia more localized than it otherwise would be (e.g., only localized to one part of the central autonomic network, or simply one cerebral hemisphere). The inception of bradycardia may also be viewed as the indirect result of epileptiform discharges, and hence a less localization-speciἀc mechanism. In this respect, bradycardia may be viewed as a release phenomenon comparable to the release mechanisms observed in many mesial temporal seizures, such as chewing and epigastric sensation, which are also vegetative in nature. Under this precinct, the generation of the release mechanism did not have to have a one-to-one transmission relationship with the ictal epileptiform discharges. Regarding the issue of lateralization of seizures in ictal bradyarrhythmia, the basic science data as outlined above may lend support for this to occur at ἀrst glance, despite the existence of some inconsistencies. There was also an additional study investigating heart rate during a Wada test (Zamrini et al. 1990) with a total of 25 patients, using a repeatedmeasures analysis of variance design. (A Wada test is a commonly performed procedure in the presurgical evaluation of epilepsy in which intracarotid injections of amobarbital are given and the subjects tested for language and memory. The test indicates which of

Neurogenic Cardiac Arrhythmias

243

the two cerebral hemispheres is the dominant hemisphere and it may give an approximation to the effect of surgery after lobectomy.) The authors suggested that left carotid artery infusion was associated with tachycardia whereas right carotid artery infusion was noted with bradycardia. However, these subjects were not patients reported to have any ictal bradycardia. In practice, as we have seen with the observations in ictal bradycardia recorded with intracranial EEGs, the pattern of seizure spread, hand dominance, and the presence of lesion contribute to the lateralization issue, making the simplistic left/right dichotomy less likely for one to predict. Mediators of the cortical signals for the autonomic system should also be considered. The parasympathetic system and the vagus nerves are essential in this respect. The functional distribution of the efferent ἀbers of the vagi also demonstrates laterality, although it is not entirely clear how this may relate to the laterality observed in the cortical representation of cardiac autonomic control. The left vagus nerve carries fewer efferent ἀbers to the ventricle, which is why traditionally vagus nerve stimulation is performed on the left side. (Fewer efferent ἀbers may mean that electrical stimulation of the vagus nerve produces less vagal-mediated parasympathetic activation and/or bradycardia.) Using information about patients in whom vagus nerve stimulators were implanted, it was observed that neither left-sided nor right-sided vagal stimulation gave rise to frequent bradycardia or asystole (McGregor et al. 2005; Handforth et al. 1998). However, one case report during intraoperative implantation of left vagus stimulator did show persistent bradycardia and the author (Asconape et al. 1999) proposed, among other speculations, that the intensity of the vagus stimulation, or some idiosyncratic mechanism at work, may contribute to this. It was not known whether during a seizure such intensity or amplitude of vagus nerve activation can be achieved or if an ictal bradyarrhythmia occurs at all. Some authorities may view that both sympathetic and parasympathetic systems are at work during ictal bradyarrhythmia, owing to the synchronization of the cardiac autonomic neural discharge with epileptogenic activity, the so-called “lockstep” phenomenon (Lathers et al. 1983). In this theory, the effect may vary from animal to animal or patient to patient, so sometimes there might be excessive sympathetic stimulation and at other times only parasympathetic nervous system dominance, or an imbalance between the two divisions. Previous studies (Stauffer et al. 1989, 1990) and O’Rourke and Lathers (2010) reported precipitous mean arterial blood pressure changes correlating with unstable lockstep phenomenon. Four possible mechanisms through which lockstep phenomenon may be related to arrhythmia and sudden death in persons with epilepsy were postulated: (1) excessive sympathetic stimulation of a heart that is already electrically unstable due to prior damage (Jay and Leestma 1981); (2) a nonuniform discharge in the postganglionic cardiac sympathetic nerve branches (Lathers et al. 1977, 1978); (3) the parasympathetic nervous system causing sinus arrest and bradycardia during seizures (Kiok et al. 1986; Lathers and Schraeder 1982); and (4) the associated precipitous changes in blood pressure per se. A coexisting pathological cardiac substrate is, and has always been, a potential confounding factor in determining the mechanism underlying ictal bradyarrhythmia. Studies on refractory epilepsy patients showed that echocardiographic abnormalities can be found in up to 9% of patients (Tigaran 2002). Among the ἀve patients observed to have ictal asystole in one study mentioned earlier, two had underlying cardiac disease that was reported by the authors to be a history of myocardial infarction in one and a history of an aberrant complex in the other (Rocamora et al. 2003). Another school of thought proposes that long-standing epilepsy may alter the neuronal network system on the cardiac tissues. One

244 Sudden Death in Epilepsy: Forensic and Clinical Issues

study found altered postganglionic cardiac sympathetic innervation with a predominantly parasympathetic cardiac activity in patients with temporal lobe epilepsy by using MIBGSPECT (123-I metaiodobenzylguanidine single photon emission computed tomography) (Druschky et al. 2001). Others argue that there may be a common pathology that leads to both seizure and arrhythmia, such as a channelopathy (e.g., mutations of genes encoding nicotinic acetylcholine receptors) that may underlie many forms of inherited idiopathic generalized epilepsy (Hiroso et al. 2005). The potential arrhythmogenic effect of antiepileptic drugs should also be considered. It is well documented that carbamazepine can slow the AV conduction and increase the sympathetic tone in the autonomic nervous system (Isojarvi et al. 1998) and decrease heart rate variability (Persson et al. 2003). One study demonstrated that abrupt withdrawal of carbamazepine can lead to enhanced sympathetic activity in sleep (Hennessy et al. 2001). Other studies have reported conflicting results (Kenneback et al. 1992, 1997). In the clinical cases discussed above, not all patients were receiving carbamazepine when ictal bradyarrhythmia occurred. However, the contributory ἀndings for other antiepileptic drugs are not easy to decipher because most, if not all, of those patients in whom ictal bradyarrhythmia occurred had been put on multiple drugs rather than a single one. Another caveat in the discussion of ictal bradyarrhythmia is the careful ascertainment of the time sequence between the seizure event and the cardiac arrhythmia. A neurocardiogenic syncope in which the bradyarrhythmic event precedes the clinical attack will be a differential diagnosis for all patients who have ictal bradyarrhythmia.

16.7â•…Could There Be a Link between Ictal Bradyarrhythmia and SUDEP? The incidence of SUDEP varies between 1 in 200 and 1 in 1000 (Lhatoo and Sander 2002). A widely accepted deἀnition of SUDEP is “a sudden unexpected nonaccidental death in an individual with epilepsy, with or without evidence of a seizure having occurred, excluding status epilepticus, where autopsy does not reveal an anatomical or toxicological cause of death” (Nashef 1997). Another deἀnition consisting of six criteria was put forward by an expert panel in 1997 (Leestma et al. 1997). In all these deἀnitions, a “deἀnite” case can only be reached if all criteria are satisἀed, but where postmortem data are lacking, the term “probable” should be used. Many risk factors have been stratiἀed for SUDEP and on face value they may not always be consistent (Langan and Sander 1999; Kloster and Engelskjon 1999; Birnbach et al. 1991; George and Davis 1998; Opeskin and Berkovic 2003; Schnabel et al. 2000; Shields et al. 2002; Opeskin et al. 2000; Jick et al. 1992; McKee and Bodἀs 2000; Nilsson et al. 1999; Timmings 1993; Walczak et al. 2001). But on the other hand, these risk factors for SUDEP may depend on the type of controls used, and therefore may be regarded as complementary (Tellez-Zenteno et al. 2005). For instance, in the studies in which non-SUDEP deaths were used as controls, seizure preceding death, subtherapeutic drug levels, and patient-found-in-bed were considered the most consistent factors (Kloster and Engelskjon 1999; Birnbach et al. 1991; George and Davis 1998; Opeskin and Berkovic 2003; Schnabel et al. 2000; Shields et al. 2002; Opeskin et al. 2000). These studies may be clinically useful in terms of ascertaining the peri-SUDEP circumstances of the patients. In studies that used persons living with epilepsy as controls, the main risk factors were high seizure frequency, high number of antiepileptic drugs, youth (but not children), and

Neurogenic Cardiac Arrhythmias

245

long duration of epilepsy (Jick et al. 1992; McKee and Bodἀsh 2000; Nilsson et al. 1999; Timmings 1993; Walczak et al. 2001). These studies may be better in terms of determining enduring, long-term factors affecting SUDEP. Although risk factors located in epidemiological studies may not offer detailed mechanistic explanations for the mechanism of SUDEP, they may nonetheless provide lines of thinking to support or refute proposed mechanisms of SUDEP. Central to this issue is that seizures do have an important role to play in SUDEP. Research into the pathogenic mechanism of SUDEP has proposed several paradigms: (1) cortical control of autonomic system, in terms of ictal bradyarrhythmia; (2) primary cardiac disease; (3) arrhythmogenic side effects of antiepileptic drugs; and (4) respiratory€mechanisms (e.g., the contribution of peri-ictal apnea and hypoxia in generalized tonic–clonic seizures is well recognized and pulmonary edema to varying extent has been reported in patients with SUDEP), mainly with retrospective ἀndings, particularly autopsy cases (Stöllberger and Finsterer 2004, Lathers and Schraeder 2010). It is not difficult for one to decipher the overlap between the mechanisms underlying ictal bradyarrhythmia and SUDEP in this regard owing to the suggestion that the cortical control of the autonomic system is a prevailing underlying explanation. The discussions about the pathogenic mechanisms of ictal bradyarrhythmia may very well apply to SUDEP, but this is made on the assumption that ictal bradyarrhythmia actually occurs during a SUDEP episode. Epidemiological data suggesting that a seizure may occur shortly before SUDEP bear witness to the possibility of ictal bradyarrhythmia having taken place. In fact, in the study in which the highest ἀgure was quoted, signs of seizures starting the ἀnal event were observed in 67% of SUDEP patients, all due to generalized motor seizures (Kloster and Engelskjon 1999). If the patient was found to have SUDEP in bed or during sleep, then it may well be possible that during such times the patient was in a physiological state that predisposed the patient to SUDEP, such as reduced sympathetic tone to counteract ictal bradyarrhythmia or increased autonomic instability leading to extreme fluctuations in heart rate. More discussion regarding sleep and SUDEP will be given in another chapter in this book (Chapter 23; Hughes and Sato 2010). However, in our analysis of patients with ictal bradyarrhythmia (both scalp and intracranial EEG data), there was insufficient information to indicate whether ictal bradyarrhythmia may occur more often during sleep. Another useful clinical method would be to look for the characteristics in a cohort with both SUDEP and ictal arrhythmia. In one recent study (Nei et al. 2004) involving 21 patients with deἀnite and probable SUDEP, ictal cardiac repolarization and rhythm abnormalities were found to occur in 56% of cases, although only 16 out of the 21 patients had continuous ECG data recorded with video EEG and the analysis was understandably performed in retrospect. The rhythm abnormalities may range from atrial ἀbrillation, through ventricular premature depolarizations, to junctional escape. No overt ictal asystole and no association with laterality were found. Authors from this study stipulated that seizures from sleep can cause sudden and extreme fluctuations in autonomic tone, which can trigger lethal cardiac arrhythmia, including bradycardia. In one recent SUDEP review, the association between cardiac dysfunction and SUDEP was not substantiated (Tellez-Zenteno et al. 2005). Thus, we may see that the clinical data were not in perfect agreement with the mechanistic explanation of SUDEP using ictal bradyarrhythmia alone, and there is a possibility that individual variability can be important and different patients may have different mechanisms (Lathers 1982, 2008). Intrinsic cardiac diseases among SUDEP cases were suspected often by virtue of autopsy analyses. These were made on histological grounds such as focal myocarditis or

246 Sudden Death in Epilepsy: Forensic and Clinical Issues

hypertrophic or dilative cardiomyopathy or other problems in the conduction system (Cohle et al. 2002) although the quality of specimen analysis might be less than desirable due to the advanced stage of death. Speciἀc ion channel and preexcitation syndromes were also suspected by researchers, and, for instance, the long QT syndromes may be due to mutations of sodium and potassium cardiac ion channels and the short QT syndrome may be due to missense mutation in the potassium ion channel IKrHERG (de la Grandmaison 2006). Such pathology may, however, not be readily demonstrable at autopsy, making retrospective analysis of this theory rather difficult (Davies 1999). There was speculation that such genetic mutation may underlie both an epilepsy syndrome and a cardiac syndrome although the exact disease entity remains unknown (Hiroso et al. 2005). Moreover, these entities may comprise theoretically both tachycardia and bradycardia syndrome. Epidemiological studies did not identify as yet intrinsic cardiac disease as a risk factor (Tellez-Zenteno et al. 2005) and cardiac enzymes (such as troponin T) during seizure were previously shown not to be elevated ictally (Woodruff et al. 2003). Therefore, the contribution of an intrinsic cardiac disease to SUDEP, with or without direct reference to ictal bradyarrhythmia, remains unknown from a clinical point of view. The speciἀc association of carbamazepine with SUDEP has been previously investigated. An audit of an epilepsy clinic in Wales showed that carbamazepine was used in higher proportion among the 14 SUDEP patients than among the general epilepsy patients attending the clinic (Timmings 1998). However, confounding with other variables had not been removed. In a Norwegian study, carbamazepine was also shown to be in small excess among the SUDEP cases but polytherapy precluded worthwhile analysis (Kloster and Engelskjon 1999). The arrhythmogenic properties of carbamazepine had been already discussed, but once again, the contribution of carbamazepine to SUDEP with or without a linkage to ictal bradyarrhythmia is still open to debate. Please see a separate chapter in this book regarding the risk of SUDEP with pharmaceutical drugs (Chapter 51; Tomson 2010). Respiratory mechanisms have been reported to be compromised during an ictal event using polysomnography in 20 of 47 clinical seizures according to one study (Nashef et al. 1996). Central apnea was most often observed but obstructive apnea was also present. What is more, bradycardia was shown to be associated with apnea during the same time. In two reports featuring witnessed SUDEP cases, seizure associated respiratory embarrassment was a prominent observation (Nashef et al. 1998; Langan et al. 2000). The relationship between ictal bradyarrhythmia and ictal apnea lies at the proposition that either may be mediated by a similar set of central autonomic networks as suggested by previous electrical stimulation studies (Penἀeld and Jasper 1954). However, apnea due to noncentral influence may not be explained by this. In studies examining the risk factors for SUDEP, the proportion of patients found dead in a prone position ranged from 42% to 71% (Kloster and Engelskjon 1999; Nashef et al. 1996; Earnest et al. 1992). It was postulated that in such a position ventilation may be easily compromised, such as bringing about collapse of upper airways, particularly if seizure had happened. Pathological examination of SUDEP cases also revealed the common occurrence of pulmonary edema in the autopsy that pointed toward the involvement of apnea in peri-SUDEP circumstances (Black and Graham 2002). In the only case of SUDEP occurring in an intracranially monitored patient, the death occurred during a seizure with right mesial temporal onset with the EEG becoming flat after 16 s in the right hemisphere. The left hemisphere showed spike discharges for a further 8 s before ceasing suddenly as well. However, there was no respiratory or ECG recording

Neurogenic Cardiac Arrhythmias

247

although pulse artifact was found for a further 120 s at a rate of 46 s. During the event the patient was lying face down although no apparent evidence of asphyxia was noted. Bird et al. (1997) proposed that the mechanism of SUDEP in this case was entirely cerebrogenic. Therefore, the mechanism underlying SUDEP may be a multifactorial one, with contribution of seizure playing a major role, and with the central influence of cardiac and respiratory functions giving rise to potentially life-threatening events, with or without extra predisposition from intrinsically altered physiological parameters such as a cardiac substrate (Amlie 1997), or drugs, together with certain peri-SUDEP events, such as a prone position or during sleep, which ultimately brings about the demise of the patient.

16.8â•…What Lies in the Future for Researchers? We have reexamined the biological plausibility of the cortical control of the autonomic system in the explanation of ictal bradyarrhythmia. Although the evidence from scientiἀc studies is in keeping with this general notion, further analysis to allow for a clear dissection of the mechanism is not readily available, such as the left-right paradigm. Further extrapolation of ictal bradyarrhythmia to a mechanistic explanation for SUDEP has remained elusive. The missing links are (1) clinical evidence of common factors shared by ictal bradyarrhythmic patients and SUDEP patients, (2) evidence of arrhythmia from epidemiological studies as a risk factor for SUDEP, and (3) ascertaining the importance of ictal bradyarrhythmia in SUDEP with regard to other proposed mechanisms including apnea and intrinsic cardiac abnormalities. It may well be possible that SUDEP has an underlying mechanism attributable to multiple causes rather than a single, unifying factor. From the seizure mechanistic perspective, and also based on data from electrical stimulation and functional imaging studies, it might be logical to speciἀcally examine cases in which intracranial EEG monitoring of the left insular region took place. However, there is ethical concern in putting an intracranial electrode near the insular region simply to look for seizure spread rather than origin. In addition, in individual patients with intracranial recording showing seizure onset from the insular region, the presence of lesion may alter the normal physiological location of the autonomic cortical pathways. From the clinical point of view, it would be logical to look again at the underlying seizure mechanism and location of seizure onset and spread in each and every case of SUDEP. Given the increasing utilization of presurgical work-up for refractory epilepsy patients, the number of SUDEP cases with previous intracranial recording may be increasingly found. Comparison of SUDEP cohorts with ictal bradyarrhythmia patients and exploration of other alternative mechanisms for SUDEP may be potential areas for further research.

Acknowledgments We thank the research team at the Division of Neurology, Department of Medicine and Therapeutics, Chinese University of Hong Kong. We are grateful to Professor Lawrence K. S. Wong, the chair professor in Neurology at the Chinese University of Hong Kong, for guidance over our research interests over the years. We are also most indebted to our collaborators at the Department of Epileptology, University of Bonn, Germany, and to Professor Christian Elger for his most kind and generous offers of training opportunities

248 Sudden Death in Epilepsy: Forensic and Clinical Issues

during our visits to the University of Bonn. We are grateful to Drs. Vincent Ip and Lisa Au for proofreading the manuscript.

References Allman, J. M., A. Hakeem, J. M. Erwin, and E. Nimchinsky. 2001. The anterior cingulate cortex: The evolution of an interface between emotion and cognition. Ann N Y Acad Sci 935: 107–117. Altenmuller, D. M., M. Zehender, and A. Schulze-Bonhage. 2004. High grade atrioventricular block triggered by spontaneous and stimulation induced epileptic activity in the left temporal lobe. Epilepsia 45 (12): 1640–1644. Amlie, J. P. 1997. QT dispersion and sudden cardiac death. Eur Heart J 18: 189–190. Asconape, J. J., D. D. Moore, D. P. Zipes, L. M. Hartman, W. H. Duffell. 1999. Bradycardia and asystole with the use of vagus nerve stimulation for the treatment of epilepsy: A rare complication of intraoperative device testing. Epilepsia 40 (10): 1452–1454. Bird, J. M., K. A. T. Dembny, D. Sandeman, and S. Butler. 1997. Sudden unexplained death in epilepsy: An intracranially monitored case. Epilepsia 38 (suppl. 11): S52–S56. Birnbach, C. D., A. J. Wilensky, and C. B. Dodril. 1991. Predictors of early mortality and sudden death in epilepsy: A multidisciplinary approach. J Epilepsy 4: 11–17. Black, M., and D. I. Graham. 2002. Sudden unexplained death in adults caused by intracranial pathology. J Clin Pathol 55: 44–50. Blumhardt, L. D., P. E. Smith, and L. Owen. 1986. Electrocardiographic accompaniments of temporal lobe epileptic seizures. Lancet 1 (8489): 1051–1056. Britton, J. W., G. R. Ghearing, E. E. Benarroch, and G. D. Gascino. 2006. The ictal bradycardia syndrome: Localization and lateralization. Epilepsia 47 (4): 737–744. Broglin, D., and J. Bancaud. 1991. Manifestations neurovegetatives au cours des crises partiÂ� elles du lobe temporal. In: Crises epileptiques et epilepsies du lobe temporal, 69–96. Gentilly: Documentation Medicale Labaz. Burns, S. M., and J. M. Wyss. 1985. The involvement of the anterior cingulate cortex in blood pressure control. Brain Res 340: 71–77. Chefer, S. I., M. I. Talan, and B. T. Engel. 1997. Central neural correlates of learned heart rate control during exercise: Central command demystiἀed. J Appl Physiol 83 (5): 1448–1453. Cohle, S. D., M. P. Suarez-Mier, and B. Aguilera. 2002. Sudden death resulting from lesions of the cardiac conduction system. Am J Forensic Med Pathol 23: 83–89. Commission on Epidemiology and Prognosis, International League Against Epilepsy. 1993. Guidelines for epidemiologic studies on epilepsy. Epilepsia 34 (4): 592–596. Critchley, H. D., C. J. Mathias, O. Josephs, J. O’Doherty, J. Zanini, B. K. Dewar, L. Cipolotti, T. Shallice, and R. J. Dolan. 2003. Human cingulate cortex and autonomic control: Converging neuroimaging and clinical evidence. Brain 126: 2139–2152. Davies, M. J. 1999. The investigation of sudden cardiac death. Histopathology 34: 93–98. de la Grandmaison, G. L. 2006. Is there progress in the autopsy diagnosis of sudden unexpected death in adults? Forensic Sci Int 156: 138–144. Devinsky, O., S. Pacia, and G. Tatambhotla. 1997. Bradycardia and asystole induced by partial seizures: A case report and literature review. Neurology 48: 1712–1714. Druschky, A., M. J. Hiz, P. Hopp, G. Platsch, M. Radespiel-Troger, K. Druschky, T. Kuwert, H. Stefan, B. Neundorfer, and S. Oppenheimer. 2001. Interictal cardiac autonomic dysfunction in temporal lobe epilepsy demonstrated by (123I)metaiodobenzylguanidine-SPECT. Brain 124: 2372–2382. Earnest, M. P., G. E. Thomas, R. A. Eden, and K. F. Hossack. 1992. The sudden unexplained death syndrome in epilepsy: Demographic, clinical, and postmortem features. Epilepsia 33: 310–316. George, J. R., and G. G. Davis. 1998. Comparison of antiepileptic drug levels in different cases of sudden death. J Forensic Sci 43: 598–603.

Neurogenic Cardiac Arrhythmias

249

Handforth, A., C. M. DeGiorgio, S. C. Schachter, B. M. Uthman, D. K. Naritoku, E. S. Tecoma, T. R. Henry, et al. 1998. Vagus nerve stimulation therapy for partial onset seizures: a randomized, active control trial. Neurology 51: 48–55. Healy, B., and J. Peck. 1997. Bradycardia induced from stimulation of the left versus right central nucleus of the amygdala. Epilepsy Res 28: 101–104. Hennessy, M. J., M. G. Tighe, C. D. Binnie, and L. Nashef. 2001. Sudden withdrawal of carbamaÂ� zepine increases cardiac sympathetic activity in sleep. Neurology 57 (9): 1650–1654. Hiroso, S., A. Mitsudome, M. Okada, and S. Kaneko. 2005. Genetics of idiopathic epilepsies. Epilepsia 46 (suppl. 1): 38–43. Hughes, J. D. and S. Sato. 2010. Sudden death in epilepsy: Relationship to the sleep–wake circadian cycle and fractal physiology. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 23. Boca Raton: CRC€Press. Isojarvi, J. I., H. Ansakorpi, K. Suominen, U. Tolonen, M. Repo, and V. V. Myllyla. 1998. Interictal cardiovascular autonomic responses in patients with epilepsy. Epilepsia 39: 420–426. James, W. 1894. Physical basis of emotion. Psychol Rev I: 516–529. Jay, G. W., and J. E. Leestma. 1981. Sudden death in epilepsy. A comprehensive review of the literature and proposed mechanisms. Acta Neurol Scand Suppl 82: 1–66. Jick, S. S., T. B. Cole, A. Mesher, and P. J. H. Tennis. 1992. Sudden unexplained death in young persons with primary epilepsy. Pharmacoepidemiol Drug Saf 1: 59–64. Kaada, B. R. 1951. Somatomotor, autonomic and electrocorticographic responses to electrical stimulation of rhinencephalic and other structures in primates, cat and dog: A study of responses from the limbic, subcallosal, orbitoinsular, piriform and temporal cortex, hippocampus fornix and amygdala. Acta Physiol Scandinav Suppl 24 (83): 1–262. Kahane, P., M. Di Leo, D. Hoffmann, and C. Munari. 1999. Ictal bradycardia in a patient with a hypothalamic hamartoma: A stereo-EEG study. Epilepsia 40: 522–527. Kenneback, G., L. Bergfeldt, T. Tomson, E. Spina, and O. Edhag. 1992. Carbamazepine induced bradycardia—A problem in general or only in susceptible patients? A 24 h long term electrocardiogram study. Epilepsy Res 13 (2): 141–145. Kenneback, G., M. Ericson, T. Tomson, and L. Bergfeldt. 1997. Changes in arrhythmia proἀle and heart rate variability during abrupt withdrawal of antiepileptic drugs. Implications for sudden death. Seizure 6 (5): 369–375. Kiok, M. C., C. F. Terrence, G. H. Fromm, and S. Lavine. 1986. Sinus arrest in epilepsy. Neurology 36 (1): 115–116 Kloster, R., and T. Engelskjon. 1999. Sudden unexpected death in epilepsy (SUDEP): A clinical perspective and a search for risk factors. J Neurol Neurourg Psychiatry 67: 439–444. Langan, Y., L. Nashef, and J. W. A. S. Sander. 2000. Sudden unexpected death in epilepsy: A series of witnessed deaths. J Neurol Neurosurg Psychiatry 68: 211–213. Langan, Y., and J. W. A. S. Sander. 1999. Patients with epilepsy who die suddenly do not always have cardiac disease. Arch Neurol 56 (2): 249. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1977. Correlation of ouabain-induced arrhythmia and non-uniformity in the histamine-evoked discharge of cardiac sympathetic nerves. J Pharmacol Exp Ther 203 (2): 467–479. Lathers, C. M., G. J. Kelliher, J. Roberts, and A. B. Beasley. 1978. Non-uniform cardiac sympathetic nerve discharge: Mechanism for coronary occlusion and digitalis-induced arrhythmia. Circulation 57 (6): 1058–1065. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23: 633–647. Lathers, C. M., and P. L. Schraeder. 2010. Antiepileptic drugs beneἀt/risk clinical pharmacology: Possible role in cause and/or prevention of SUDEP. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton: CRC Press.

250 Sudden Death in Epilepsy: Forensic and Clinical Issues Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008. The mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12 (1): 3–24. Lathers, C. M., F. L. Weiner, and P. L. Schraeder. 1983. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lock-step phenomenon. Circ Res 31: 630A. Leestma, J. E., J. F. Annegers, M. J. Brodie, S. Brown, P. Schraeder, D. Siscovick, B. B. Wannamaker, P. S. Tennis, M. A. Cierpial, and N. L. Earn. 1997. Sudden unexplained death in epilepsy: Observations from a large clinical development program. Epilepsia 38: 47–55. Leung, H., K. Schindler, P. Kwan, and C. Elger. 2007. Asystole induced by electrical stimulation of the left cingulated gyrus. Epileptic Disord 9 (1): 77–81. Lhatoo, S. D., and J. W. Sander. 2002. Sudden unexpected death in epilepsy. Hong Kong Med J 8 (5): 354–358. Manitius-Robeck, S., P. Schuler, H. Feistel, G. Platsch, and H. Stefan. 1998. Ictal syncopes. Cardiac sympathetic innervation disorder as the etiology? (German) Der Nervenarzt 69: 712–716. Matthews, S. C., M. P. Paulus, A. N. Simmons, R. A. Nelesen, and J. E. Dimsdale. 2004. Functional subdivisions within anterior cingulate cortex and their relationship to autonomic nervous system function. NeuroImage 22: 1151–1156. McGregor, A., J. Wheless, and J. Baumgartner. 2005. Right sided vagus nerve stimulation as a treatment for refractory epilepsy in humans. Epilepsia 46 (1): 91–96. McKee, J. R., and J. W. Bodἀsh. 2000. Sudden unexpected death in epilepsy in adults with mental retardation. Am J Ment Retard 105: 235. Mufson, E. J., and M. M. Mesulam. 1982. Insula of the old world monkey: Part II. Afferent cortical input and comments on function. J Comp Neurol 212: 23–37. Munari, C., L. Tassi, and M. Di Leo, P. Kahane, D. Hoffman, S. Francione, and P. Quarato. 1995. Video-stereo-electroencephalographic investigation of orbitofrontal cortex: Ictal electroclinical patterns. In Epilepsy and the Functional Anatomy of the Frontal Lobe, ed. H. H. Jasper, S. Riggio, and P. S. Goldman-Rakic, 273–795. New York, NY: Raven Press. Nashef, L. 1997. Sudden unexpected death in epilepsy: Terminology and deἀnitions. Epilepsia 38: S6–S8. Nashef, L., F. Walker, P. Allen, J. W. Sander, S. D. Shorvon, and D. R. Fish. 1996. Apnoea and bradycardia during epileptic seizures: Relation to sudden death in epilepsy. J Neurol Neurosurg Psychiatry 60 (3): 297–300. Nashef, L., S. Garner, J. Sander, D. Fish, and S. Shorvon. 1998. Circumstances of death in sudden death in epilepsy: Interviews of bereaved relatives. J Neurol Neurosurg Psychiatry 64 (3): 349–352. Nilsson, L., B. Y. Farahmand, P. G. Persson, I. Thiblin, and T. Tomson. 1999. Risk factors for sudden unexpected death in epilepsy: A case-control study. Lancet 13: 888–893. Nei, M., R. T. Ho, B. W. Abdou-Khalil, F. W. Drislane, J. Liporace, A. Romeo, and M. R. Sperling. 2004. EEG and ECG in sudden unexplained death in epilepsy. Epilepsia 45(4): 338–345. Ongur, D., X. An, and J. L. Price. 1998. Prefrontal cortical projections to the hypothalamus in macaque monkeys. J Comp Neurol 401: 480–505. Opeskin, K., and S. Berkovic. 2003. Risk factors for sudden unexpected death in epilepsy: A controlled prospective study based on coroners cases. Seizure 12: 456–464. Opeskin, K., A. S. Harvey, S. M. Cordner, and S. F. Berkovic. 2000. Sudden unexpected death in epilepsy in Victoria. J Clin Neurosci 7: 34–37. Oppenheimer, M. B., A. Gelb, J. P. Girvin, and V. C. Hachinski. 1992. Cardiovascular effects of human insular cortex stimulation. Neurology 42: 1727–1732. O’Rourke, D., and C. M. Lathers. 2010. Interspike interval histogram characterization of synchronized cardiac sympathetic neural discharge and epileptogenic activity in the electrocorticogram of the cat. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 31. Boca Raton: CRC Press. Penἀeld, W., and H. Jasper. 1954. Epilepsy and the Functional Anatomy of the Human Brain. London: J. and A. Churchill Ltd (whole chapter). Persson, H., M. Ericson, and T. Tomson. 2003. Carbamazepine affects autonomic cardiac control in patients with newly diagnosed epilepsy. Epilepsy Res 57 (1): 69–75.

Neurogenic Cardiac Arrhythmias

251

Pool, J. L., and J. Ransohoff. 1949. Autonomic effects on stimulating the rostral portion of the cingulate gyri in man. J Neurophysiol 12: 385–392. Reeves, A. L., K. E. Nollet, D. W. Klass. F. W. Sharbrough, and E. L. So. 1996. The ictal bradycardia syndrome. Epilepsia 37: 983–987. Rocamora, R., M. Kurthen, L. Lickfett, J. von Oertzen, and C. E. Elger. 2003. Cardiac asystole in epilepsy: Clinical and neurophysiologic features. Epilepsia 44 (2): 179–185. Rossetti, A. O., B. A. Dworetzky, J. R. Madsen, O. Golub, J. A. Beckman, and B. E. Broomἀeld. 2005. Ictal asystole with convulsive syncope mimicking secondary generalization: A depth electrode study. J Neurol Neurosurg Psychiatry 76: 885–887. Rugg-Gunn, F. J., R. J. Simister, M. Squirrell, D. R. Holdright, and J. S. Duncan. 2004. Cardiac arrhythmias in focal epilepsy: A prospective long-term study. Lancet 364: 2212–2219. Russell, A. E. 1906. Cessation of the pulse during the onset of epileptic ἀts. Lancet 2: 152–154. Schnabel, R., M. Beblo, and T. W. May. 2000. Is geomagnetic activity a risk factor for sudden unexplained death in epilepsies? Neurology 54: 903–908. Schuele, S. U., A. C. Bermeo, A. V. Alexopoulos, E. R. Locatelli, R. C. Burgess, and N. FoldvarySchaefer. 2007. Video-electrographic and clinical features in patients with ictal asystole. Neurology 69 (5): 434–441. Shields, L. B., D. M. Hunsaker, J. C. Hunsaker 3rd, and J. C. Parker Jr. 2002. Sudden unexpected death in epilepsy: Neuropathologic ἀndings. Am J Forensic Med Pathol 23: 307–310. Stauffer, A. Z., J. Dodd-O, and C. M. Lathers. 1989. The relationship of the lock-step phenomenon and precipitous changes in mean arterial blood pressure. Electroencephalogr Clin Neurophysiol 72: 340–345. Stauffer, A. Z., J. Dodd-O, and C. M. Lathers. 1990. Relationship of the lock-step phenomenon and precipitous changes in blood pressure. In Epilepsy and Sudden Death, ed. C. M. Lathers and P.€L. Schraeder, 221–238. New York, NY: Marcel Dekker. Stöllberger, C., and J. Finsterer. 2004. Cardiorespiratory ἀndings in sudden unexplained/unexpected death in epilepsy (SUDEP). Epilepsy Res 59: 51–60. Stöllberger, C., and J. Finsterer. 2010. Cardiac and pulmonary risk factors and pathomechanisms of sudden unexplained death in epilepsy patients (SUDEP). In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 42. New York, NY: Taylor & Francis. Tellez-Zenteno, J. F., L. H. Hernandez Ronquillo, and S. Wiebe. 2005. Sudden unexpected death in€epilepsy: Evidence based analysis of incidence and risk factors. Epilepsy Res 65: 101–115. Tigaran, S. 2002. Cardiac abnormalities in patients with refractory epilepsy. Acta Neurol Scand Suppl 177: 9–32. Timmings, P. L. 1993. Sudden unexpected death in epilepsy: A local audit. Seizure 2: 287–290. Timmings, P. L. 1998. Sudden unexpected death in epilepsy: Is carbamazepine implicated? Seizure 7: 289–291. Tinuper, P., F. Bisulli, R. Cerullu, R. Carcangiu, C. Marini, G. Pierangeli, and P. Cortelli. 2001. Ictal bradycardia in partial epileptic seizures. Autonomic investigation in three cases and literature review. Brain 124: 2361–2371. Tomson, T. 2010. Compliance with antiepileptic drug treatment and the risk of sudden unexpected death in epilepsy. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C.€M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 51. Boca Raton: CRC Press. Van Buren, J. M. 1961. Sensory, motor and autonomic effects of mesial temporal stimulation in man. J Neurosurg 18: 273–288. Vilensky, J. A., and G. W. van Hoesen. 1981. Corticopontine projections from the cingulate cortex in the rhesus monkey. Brain Res 205: 391–395. Walczak, T. S., I. E. Leppik, M. D’Amelio, J. Rarick, E. So, P. Ahman, K. Ruggles, G. C. Gascino, J. F. Anngers, and W. A. Hauser. 2001. Incidence and risk factors in sudden unexpected death in epilepsy: A prospective cohort study. Neurology 56: 519–525. Ward, A. A. 1948. The cingular gyrus: Area 24. J Neurophysiol 11: 13–34.

252 Sudden Death in Epilepsy: Forensic and Clinical Issues Woodruff, B. K., J. W. Britton, S. Tigaran, G. D. Gascino, M. F. Burritt, J. P. McConnell, J. Ravkilde et al. 2003. Cardiac troponin levels following monitored epileptic seizures. Neurology 60: 1690–1692. Zamrini, E. Y., K. J. Meador, D. W. Loring, F. T. Nichols, G. P. Lee, W. O. Tomson. 1990. Unilateral cerebral inactivation produces differential left/right heart rate responses. Neurology 40: 1408–1411.

17

Stress and SUDEP Claire M. Lathers Paul L. Schraeder

Contents 17.1 Introduction 17.2 Stress-Related Risk Factors 17.3 Positive Life Events as a Stress: A Case Report of SUDEP References

253 254 260 262

17.1â•…Introduction Heart–brain interactions function during normal daily routine actions and during times of stress. The occurrence of stress itself is a powerful change initiator and may trigger transient ischemia and acute coronary syndrome in some persons (Pickworth et al. 1990; Lathers and Schraeder 2006; Lathers et al. 2008; Soufer and Burg 2007). These individuals are at increased risk for recurrent cardiac events and early death. Psychosocial stress can become an acute trigger of myocardial infarction in patients with preexisting coronary artery disease. Stress, via actions on the central and autonomic nervous systems, may produce a cascade of physiologic responses in individuals at risk that may lead to myocardial ischemia, ventricular ἀbrillation, plaque rupture, or coronary thrombosis (Krantz et al. 1996). Use of simultaneous single-photon emission computed tomography imaging with technetium99m tetrofosmin myocardial perfusion imaging and transthoracic echocardiography was done at rest and during mental stress induced in patients with stable coronary artery disease (Shah et al. 2006). It was concluded that C-reactive protein levels may be a risk marker for mental stress-induced myocardial ischemia. The role of C-reactive protein levels in stress associated with interictal and ictal seizure activity and the development of cardiac arrhythmias and/or death is unknown and should be examined. Nevertheless, today we still do not understand the pathophysiology of mental stress-induced ischemia, what diagnostic tests are needed to identify susceptible persons, nor how to develop risk stratiἀcation algorithms to be applied in the clinical workplace. Research is needed to understand the brain–heart relationship during the occurrence of mental stress that underlies the cognitive and emotional aspects of mental stress as the distinct patterns of brain activity occurring during mental stress trigger silent myocardial ischemia (Soufer and Burg 2007). Cardiac function itself may be adversely changed during an episode of acute emotional stress (Ziegelstein 2007). Left ventricular contractile dysfunction, myocardial ischemia, and/or cardiac arrhythmias have been demonstrated to be triggered by acute emotional stress. These events may be transient but have damaging and/or fatal consequences. The understanding of the anatomical substrate and physiological pathways involved in this heart–brain interaction are not clear, but new data obtained with functional neuroimaging suggest asymmetric brain activity are important in making the heart more susceptible 253

254 Sudden Death in Epilepsy: Forensic and Clinical Issues

to ventricular arrhythmias. Lateralization of cerebral activity during emotional stress may stimulate the heart asymmetrically and produce areas of inhomogeneous repolarization that create electrical instability and facilitate development of cardiac arrhythmias. Ziegelstein (2007) states that patients with ischemic heart disease, who do survive an episode of sudden cardiac death in the setting of acute emotional stress, should be treated with a beta blocker. Nonpharmacologic methods to manage the effects of stress in persons with or without coronary artery disease include social support, relaxation therapy, yoga, meditation, controlled slow breathing, and biofeedback. Clues for effective treatment of mental and emotional stress associated with heart– brain interactions exist. As early as 1993, the pathophysiologic effects of mental stress appear to involve alterations in both myocardial oxygen demand and supply (Merz et al. 1993). Intense negative emotion, such as hostility, and heightened cardiovascular reactivity are associated with occurrence of this type of ischemia. Thus, if persons at risk are taught to recognize these factors, interventional training may help protect them from unwanted consequences. Using traditional anti-ischemic therapy, such as beta blockers and vasodilators, has been shown to reduce mental stress–triggered ischemia in coronary artery disease. Both behavioral and psychosocial interventions, such as decreasing environmental stress via use of social support, alteration of stress perception by behavioral training, and altered physiologic reaction to stress through physical training were discussed as therapeutic options. Posttraumatic stress disorder, a psychiatric disorder that develops after a psychological trauma generally triggered by a situation perceived by the person experiencing the event as one that deeply threatens his/her life or integrity, is thought to be triggered by complex neurobiological changes. Kozaric-Kovacic (2008) reports that selective serotonin-reuptake inhibitors are the ἀrst line of treatment for posttraumatic stress disorders. These agents are more effective than noradrenalin-reuptake inhibitors or tricyclic antidepressants. Antipsychotic drugs, especially the atypical ones, are effective in posttraumatic stress disorder patients with psychotic characteristics or refractoriness to other drugs, reducing the overall overreaction to stress. Monoamine oxidase inhibitors have not been clearly identiἀed as beneἀcial. Serotonin agonists and antagonists, new antidepressants that are dual inhibitors of serotonin and noradrenaline reuptake, anticonvulsants, and opiate antagonists may be used. Additional rigorous clinical trials are needed to establish use, efficacy, tolerability, and safety in treating this pharmacotherapeutic disorder. The occurrence of acute postictal psychiatric symptoms are well recognized, but, fortunately, relatively uncommon, and most likely manifest after partial complex seizures. These symptoms may mimic anxiety, depression, or an acute psychotic disorder (Kanner et al. 1996). Violent behavior can also be an uncommon postictal manifestation. When such episodes of€ violence occur, they are not well organized, planned, or speciἀcally directed. Directed aggression is rare, but when postictal aggressive behavior—whether undirected or seemingly directed—does occur, the patient can be at risk for an aggressive police response including total body restraints. Mendez (1998) described a patient with directed postictal aggression.

17.2â•…Stress-Related Risk Factors The mechanisms involved in emotion-associated seizure activation are not known but multiple factors may explain the role of emotion. These factors include activation of neural networks, sleep deprivation, noncompliance, alcohol use, and hyperventilation. Increased

Stress and SUDEP

255

risk of seizure recurrence is also associated with use of antidepressants and/or neuroleptics, drugs that may lower the seizure threshold. Poor seizure control is a major risk factor for SUDEP (Dominian et al. 1963; Mattson 1991; Nashef et al. 1995; Sperling et al. 1999). Acute stress caused by emotions such as fear may cause sudden death in persons with coronary heart disease. Chronic stress also contributes to development of long-term coronary disease (Ackerman et al. 2001). Stress management in standard cardiac rehabilitation programs is important. The distinction between type A versus type B personalities as risk correlates is overly simplistic. Psychosocial factors of depression, hostility, social isolation, anxiety, anger, and other stresses are related to increased cardiac death and illness in all groups with coronary heart disease (Buselli and Stuart 1999) and these psychosocial risk factors may beneἀt from a biopsychosocial model of intervention. Death can result from acute stress related to disasters. Sudden deaths related to atherosclerotic coronary artery disease increased ἀvefold on the day of the Los Angeles earthquake in 1994 (Leor et al. 1996). Acute mental stress induced experimentally may be associated with ST-segment deviation and wall-

Heart –Brain Int erplay : A 53-Year-Old Wo man Reco v ering fro m Mit ral Valv e Repair Anxiety, panic, fatigue, and depression can have a profound adverse effect on physical well-being. This case illustrates how these adverse psychological symptoms interfered with recovery from mitral valve surgery. A 53-year-old woman had an uncomplicated mitral valve repair, after which she refused to use the incentive spirometer, ambulate, or even sit in a chair. She experienced unexplained episodes of shortness of breath and tachycardia. At 4 weeks after operation, the ejection fraction, which was 50% pre-op, remained at 40%. She was complaining of more post-op pain and fatigue than was usual for this type of successful surgery. The patient experienced tearful episodes, and reluctantly was interviewed by a psychiatrist. While the initial diagnosis was adjustment disorder with anxious features, subsequent evaluation established a diagnosis of depression. Ultimately, the patient agreed to take an SSRI, citalopram. After several weeks of treatment, she had a profound improvement in her mental and physical states, with she and her family concurring that she had returned to her “old self” (Callahan et al. 2008). Discussio n This case illustrates how postoperative cardiac rehabilitation can be signiἀcantly affected by an adverse psychological state. Although the pathophysiological mechanism of sudden death in epilepsy (SUDEP) is not established, acute cardiac dysfunction is thought to be a major factor. Cardiologists have long known about the physical risks of adverse mental states in cardiac patients including unexpected sudden cardiac death. The role of stress and other psychological disturbances require consideration as possible risk factors associated with SUDEP. However, in contrast to the interest cardiologists have manifested in psychogenic risks associated with cardiac sudden death, little research has been undertaken on the subject of psychological issues relative to the risk of SUDEP.

256 Sudden Death in Epilepsy: Forensic and Clinical Issues

motion abnormalities (Rozanski et al. 1988). Mental stress may be a greater risk factor than eÂ�xercise-induced ischemia in increasing the rate of fatal and nonfatal cardiac events (Jiang et al. 1996). Psychosocial stress management treatment in a cardiac rehabilitation program does reduce cardiac related mortality and morbidity (Linden et al. 1996). Distress at the family level and/or stress at work are strong predictors for developing stress-related disorders and need intervention (Anderberg 2001). Management of stress factors in coronary artery disease may also help to decrease the risk of sudden death. While depression and stress are major risk factors in sudden cardiac death, it is not known if the same risks apply to SUDEP. Having epilepsy in and of itself is stress-producing and stress does increase the frequency of seizures. The uncertainty of when a seizure can occur, the consequences of having a seizure on employment status and driving privileges are stress-producing circumstances. Both depression and anxiety are symptoms associated with epilepsy (Trimble and Perez 1980; Blumer 1992). Earnest et al. (1992) suggested there was a role for acutely stressful circumstances as a possible contributor near the time of death in a case control study of the metropolitan Denver area. The strong interest of cardiologists in adverse emotional states as a risk of sudden cardiac death implies that there is a need to investigate this issue more thoroughly in persons with epilepsy. Activation of stress-responsive systems during depressive episodes may contribute to metabolic risk factors and imbalance of the autonomic heart regulation (Blumer 1992; Linden et al. 1996; Deuschle and Lederbogen 2002). Clearly, cerebral activity can have a profound effect upon the autonomic regulation of cardiac function. Lathers et al. (1977, 1978) found that nonuniform postganglionic cardiac sympathetic neural discharge is capable of triggering cardiac arrhythmias in the manner described by Han and Moe (1964) (i.e., nonuniform cardiac repolarization). These authors noted that the aberrant nonuniform neural discharge was associated with cardiac arrhythmias triggered by abrupt coronary occlusion of the left anterior descending coronary artery to produce cardiac ischemia mimicking events occurring in the sudden death heart attack victim. The nonuniform postganglionic cardiac sympathetic neural activity also occurs with cardiac changes associated with ouabain-induced toxicity characterized by cardiac arrhythmias and death. In both animal models the potentially damaging role of adrenal catecholamines in the production of arrhythmias and death was discussed. Synchronization of brain electrical activity with both cardiac sympathetic and vagal neural discharge was also identiἀed by those working in Dr. Lathers’ laboratory using the cat model (Lathers et al. 1987; O’Rourke and Lathers 1990; Dodd-O and Lathers 1990; Stauffer et al. 1989; Lathers et al. 2010). This autonomic neural discharge synchronized with cerebral ictal and interictal discharges was termed the lockstep phenomenon and was hypothesized to be one mechanism contributing to the development of cardiac arrhythmias and/or sudden death associated with both interictal and ictal discharges in persons with epilepsy found dead in a sudden, unexpected manner. Subsequently, Davis and Natelson (1993) focused on brain–heart interactions and the neurocardiology of arrhythmia and sudden cardiac death, with an emphasis on the nervous system direction of the events leading to cardiac damage associated with raising catecholamine levels in experimental and clinical entities of stroke, epilepsy, and environmental stress. Autonomic sympathetic and parasympathetic cardiac neuronal dysfunctions are associated with interictal as well as ictal epileptiform discharges and with cardiac arrhythmias (Lathers and Schraeder 1982). Both types of epileptogenic activity are associated with temporal lobe epilepsy, autonomic dysregulation, and predominant sympathetic overactivity

Stress and SUDEP

257

Case Repo rt o f Po st ict al Ag g ressi v e Behav io r A 37-year-old man had complex partial seizures, consisting of an olfactory aura followed by alteration of consciousness, since age ἀve, with occasional tonic–clonic seizures. He also had a history of depression and bipolar affective disorder. During the postictal period, at times he would feel a sense of being threatened and having been harmed, and would focus his aggressive response by attacking any individual in his immediate environment, often resulting in physical injury to that person. The patient’s postictal confusion would remit usually after one hour, but the fear of harm and sense of being threatened would last for 24 hours, after which he felt great remorse. He had been charged with aggravated assault several times. His MRI was unremarkable, but sleep-deprived EEG manifested left anterior temporal spikes and sharp waves. The aggressive postictal episodes ended with control of his seizures resulting from the addition of carbamazepine to his original regimen of valproic acid and sertraline. Discussio n by M endez The author describes aggressive acts as direct consequence of seizures, occurring directly during seizures or postictally (Mendez 1998). He also emphasizes that postÂ� ictal violence is most commonly resistive behavior during the postictal delirium and is associated with attempts at restraint. He also discussed the observation that seemingly violent automatisms such as flailing or spitting can occur during complex partial seizures, and that secondary violent automatisms can be behavioral responses to ictal fear, hallucinations, or other disagreeable seizure related experiences. Discussio n by L at hers and Schraeder Another issue to consider in relationship to the occurrence of putative ictal or postictal violent behavior is that of the possible induction of an excited delirium syndrome when attempts are made to restrain the behavior of an individual who is manifesting actual or seemingly violent behavior. As deἀned by DiMaio and DiMaio (2006), excited delirium syndrome “involves the sudden death of an individual, during or following an episode of excited delirium, in which an autopsy fails to reveal evidence of sufficient trauma or natural disease to explain the death. In virtually all cases, the episode of excited delirium is terminated by a struggle with police or medical personnel, and the use of physical restraint. Typically, within a few to several minutes following cessation of the struggle, the individual is noted to be in cardiopulmonary arrest. Attempts at resuscitation are usually unsuccessful.” During the state of delirium, there are varying transient disturbances of consciousness and cognition with disorientation, disorganized and inconsistent thought processes, inability to distinguish reality from hallucinations, speech disturbance, and disorientation to time, place, and person. Deaths occur most commonly in individuals who have abused stimulants such as cocaine and methamphetamine, but also in persons with endogenous mental disease who have not used these drugs. The majority of deaths occur between the ages of 17 and 35. Although the mechanism of death in these individuals is not deἀned, DiMaio and DiMaio (2006) concluded that stimulation of the sympathetic nervous system causes release of norepinephrine at the synapses and in combination

258 Sudden Death in Epilepsy: Forensic and Clinical Issues

with epinephrine into the bloodstream from the adrenals. This response then results in subsequent increase in myocyte activity and oxygen demand in combination with decreased myocardial blood flow secondary to coronary artery constriction. We need to be aware of the potential for induction of this potentially fatal state of agitation when attempts are made to restrain persons with epilepsy who manifest ictal or postictal agitation or seemingly violent behavior. This is a highly stressful state and in combination with the history of epilepsy could be contributory to the occurrence of SUDEP in these individuals. Since both ictal and postictal states are self limited, it is imperative for family members, police, and emergency care personnel to understand that watchful observation to keep the affected individual out of harm’s way, rather than the high-risk intervention of physical restraint, is the most appropriate intervention.

(Hilz et al. 2002). Surgical treatment for temporal lobe epilepsy reduced sympathetic cardiomodulation and decreased baroreflex sensitivity (i.e., decreased the influence of sympathetically mediated tachyarrhythmias and excessive bradycardiac counterregulation). These factors are thought to contribute to the risk of SUDEP and thus the temporal lobe surgery itself appears to be one method to reduce and/or eliminate some risk factors associated with SUDEP (Hilz et al. 2002; Burgerman et al. 1995). Parasympathetic nervous system activity also regulates cardiac rhythm (Richter 1957; Talman 1985). Stimulation of the vagus nerve or application of acetylcholine to the SA node slows or abolishes depolarization of sinus node ἀbers (Schwartz et al. 1976; West et al. 1956). Decreased depolarization and shifts of membrane threshold potentials change the SA node rate, causing sinus bradycardia. A similar mechanism occurs in the AV node (Kralios and Millar 1981). ECG changes and cardiac muscle necrosis result from stimulation of the efferent limb of the sympathetic nervous system and by stimulation of the aortic arch and carotid baroreceptors regulated by the autonomic nervous system reflex activity (Pavlov 1951; Samuels 1997; Zavodskaya et al. 1980; Natelson et al. 1998). Sympathetic stimulation causes a sudden release of norepinephrine from cardiac nerve endings into the heart muscle. This leads to microscopic changes in the form of cardiac myocyte necrosis, cardiac dysfunction, and arrhythmias (Schwartz et al. 1976; West et al. 1956). Stimulation of cardiac sympathetic nerves accelerates sinoatrial depolarizations and shortens the cycle length of ἀring of the sinus node. Natelson et al. (1998) found pathologic changes in the form of irreversible perivascular and interstitial ἀbrosis and myocyte vacuolization in hearts of persons with epilepsy who died suddenly. These lesions occurred mostly in the subendocardium. Thus, it is possible they also resulted from sympathetic nerve catecholamine release with consequent cardiac arrhythmias and repolarization changes that predispose a patient to a form of cardiac damage known as myoἀbrillar degeneration or contraction band necrosis and possible sudden death. This lesion is associated with four types of etiologies: stress plus or minus steroids, catecholamine infusion, nervous system stimulation, and reperfusion. Sympathetic overactivity, with secondary catecholamine toxicity, is the common factor in all four etiologies. Samuels (1997) concludes that all forms of sudden death are based on the anatomic connection between the nervous system and the heart and lungs.

Stress and SUDEP

259

Exposure of rats to stress initially resulted in structural changes of stereotypical cardiac contraction bands, regardless of the type of stress factor, and differed only in severity. At a later stage, contractures were gradually replaced by cytolytic injuries and also did not depend on the type of stress. In the case of early predominance of myocytolysis in combination with excessive contracture injuries that led to rapid death, a genetically determined predisposition was proposed as an explanation (Zavodskaya et al. 1980). Since emotional stressors impact on the autonomic nervous system, this ensures that psychogenic factors are important in leading to cardiac dysrhythmias and to coronary and noncoronary sudden death syndromes in humans and other species (Bohus and Korte 2000). There is a causal relationship among depression, stress, and increased cardiac mortality. Depression may be a state of prolonged negative arousal or mental stress associated with a measurably higher risk of fatal cardiac events. Stress may also result in histopathological changes in a previously normal heart while genetically determined and/or acquired dysfunction of the opioidegic, GABAegic, cholinergic, adenosinergic, and other transmitter/modulator systems may interact to predispose to arrhythmias and sudden death (Tulner and den Boer 2000). The influence of psychological factors on autonomic, neuroendocrine, and immune systems is complex and exerts adverse effects on cardiac function. The amygdala is involved in integration of autonomic responses to emotional stimuli (Cechetto 2000). There is a common pattern of sympathetic representation in the medial prefrontal cortex, insular cortex, ventromedial temporal lobe, and ventral hippocampal region (Westerhaus and Loewy 2001). The ventromedial temporal lobe regions studied included the central, basomedial posterior, and lateral amygdaloid transition area and posterior medial cortical amygdaloid nucleus. This latter anatomical substrate for sympathetic control is of considerable theoretical importance. The amygdala has an important role in expression of emotional behaviors while it integrates autonomic responses to emotional stimuli under conditions of fear and anxiety. The amygdala is important in cardiovascular control within the limbic system, having reciprocal connections with the insular cortex and direct projections to other autonomic control centers in the hypothalamus, pons, and medulla. Other important brain–heart connections exist. Limbic cortex activity associated with an emotionally charged stimulus occurs with cardiac neural changes, resulting in intense autonomic stimulation of both sympathetic and parasympathetic neurons. This, in turn, may result in sudden stress-related death. It is assumed that the hypothalamic and brainstem structures are involved secondarily and that stress activates the hypothalamic–pituitary– adrenal axis. Prolactin levels are elevated after seizures and are used as an indicator that a seizure has actually occurred. The prolactin level is most consistently elevated after generalized tonic–clonic and complex partial seizures, and less predictably after a simple partial seizure. Frontal lobe, absence, myoclonic, and akinetic seizures are not generally associated with elevation of the prolactin level (Pritchard 1997). This observation raised the possibility of the frequent occurrence of a stress response during and after a seizure as a risk factor for SUDEP. An alternative explanation is that the hypothalamic-pituitary axis is directly simulated by the epileptiform activity. In either case, the occurrence of acute autonomic and/or neuroendocrine dysfunction appears to put the patient at risk. Elevated prolactin levels at necropsy were examined as a marker of antemortem stress (Jones and Hallworth 1999), but postmortem prolactin values differed according to the cause of death, with higher values in postoperative deaths and in the chronically ill.

260 Sudden Death in Epilepsy: Forensic and Clinical Issues

In the study of Wannamaker and Booker (1998), patients with epilepsy have identiἀed the common stressors of fear, worry, frustration, and anger as trigger factors. The seizure event usually does not occur immediately in association with the stressor. Animal models suggest that injured populations of neurons surround and interact with an epileptogenic focus. This may cause the focus to function independently when heightened brain excitability triggers neuronal activity in this network. As discussed by Homan (1998), patients often associate increased stress with increased seizure activity. Interventions to reduce stress include the use of behavioral techniques such as biofeedback, relaxation, and desensitization, and pharmacological agents such as psychotherapeutic or benzodiazepine drugs. Physical stress associated with elevated body temperatures induced by infection also triggers seizures and antipyretic drugs are recommended for prophylactic use. Adjunctive management of the patient with seizure and stress results in improvement in the control of seizures and overall quality of life (Moffett and Scott 1984; Fenwick 1995). Education of patients about the importance of drug compliance and stress management techniques ultimately will decrease the need for high therapeutic antiepileptic drug levels while decreasing the occurrence of dose-related side effects. This will also result in an improved lifestyle. Physicians must individualize recommendations for intervention and use referrals to a clinical psychologist with expertise in stress reduction. Both internal and external stressors must be decreased. Treatment with tranquilizers, antidepressants, or neuroleptics is used with counseling. The long-acting anxiolytics clorazepate and clonazepam are relatively effective as antiepileptic drugs in selected patients. Paranoia, thought disorder, hallucinations, and extreme agitation require concurrent psychiatric consultation in addition to treatment with neuroleptics. Care must be exerted when managing the patient since a rapid change in the levels of neuoroleptics, high doses, or induction of drowsiness may worsen the occurrence of seizures. Reduction in agitation, thought disorder, and hallucinations usually exerts a calming effect and contributes to lowering the stress level and to restoring the patient’s sense of well-being (Lathers and Schraeder 2006). As can be concluded from the above discussion, most life events that may be implicated as risk factors for sudden death result in psychological responses that are regarded as negative. The following brief case history raises the possibility that, at least in some persons with epilepsy, an intensely positive event may also be a risk factor for SUDEP.

17.3â•…Positive Life Events as a Stress: A Case Report of SUDEP Many diseases that can affect autonomic balance exhibit patterns of temporal variation during circadian, seasonal, reproductive, and life span cycles that remain unexplained. Termination of organisms during senescence, achieved by emergence of autonomic imbalance and other systemic dysfunction has been examined from a Darwinian perspective. This variation in autonomic balance and disease symptoms of epilepsy has yet to be carefully studied. Cardiac neural control may be estimated by frequency domain characterization of R–R interval variations and this technique is a clinical tool to examine the role of autonomic dysfunction in the pathophysiology of sudden cardiac death (Molgaard et al. 1994). Simultaneous examination of the quality of life and changes in heart rate variability of patients immediately after acute myocardial infarction showed that survivors exhibited heart rate variability within the ἀrst 3 days that was signiἀcantly higher than nonsurvivors and had developed a clear circadian pattern after 3 weeks (Kummell et al. 1993). The authors

Stress and SUDEP

261

SUDEP AFT ER RECEIVING V ERY GOO D NEWS An 18-year-old black female high school student had a history of infrequent (less than once yearly) generalized tonic–clonic seizures that were controlled with moderate doses of carbamazepine with therapeutic blood levels. She was an outstanding student, and during her senior year was offered admission to an Ivy League school with a full scholarship. Her social circumstances were modest in that her father worked as a municipal trash collector and she was the ἀrst family member to go to college. Shortly thereafter, her distraught parents notiἀed her neurologist (PLS) that their daughter was found dead in bed. As no postmortem was performed, and no other cause of death was evident, the diagnosis of probable SUDEP was applied. This case may demonstrate that intensely positive surprise life events can produce as much of a stress as negative events. concluded circadian patterns of heart rate and heart rate variability may be assessed meaningfully immediately post acute myocardial infarction and may ἀnd common expression in changes in sympathovagal balance. Future studies of persons with epilepsy who may be at risk for SUDEP could examine sympathovagal balance using 24-hour ambulatory BP monitoring to examine circadian blood pressure variation, power spectral analysis of R–R interval oscillation to measure autonomic function, and measurement of QTc-d and QTc intervals to monitor cardiac depolarization time. These techniques were employed in a 1-year study designed to examine sympathovagal balance, nighttime blood pressure and QT intervals (Esposito et al. 2003). Sustained weight loss eliminated the diastolic night time drop in blood pressure and sympathetic overactivity detected in normotensive obese women. This weight loss may reduce the cardiovascular risk in obese women. One could speculate that as the blood pressure falls during sleep, there may be, in some persons with epilepsy and autonomic dysfunction, an increased sympathetic discharge that results in the triggering of cardiac arrhythmias and/or sudden death. A study of the role of sympathovagal interaction in diurnal variations of QT interval suggested that although change in sympathovagal balance was responsible for diurnal variation in QT interval, the enhanced sympathetic activity in the day was a major determinate of the phenomenon (Murakawa et al. 1992). Examination of heart rate variability circadian patterns and effect on the QT interval dispersion in healthy subjects (Bilan et al. 2005) was studied. Multiple regression analysis revealed relations between mean QTd and R–R as well as mean QTd and high frequency after adjustment for periods, correlations were only observed during morning hours. The authors concluded that sympathovagal balance, as reflected in heart rate variability, and not the tone of both autonomic components that affects QTd variability. These data suggest that in persons with epilepsy, the sympathovagal balance, as reflected in heart rate variability, should be examined for changes in QTd variability as a risk factor for sudden death during both awake and sleep cycles. In Brugada syndrome, ventricular ἀbrillation occurs mainly during sleep, and Brugada ECG signs are intensiἀed by parasympathomimetic drugs (Mizumaki et al. 2004). Spontaneous augmentation of ST elevation in daily life was demonstrated along with an increase in vagal activity. The ST elevation was increased more in those patients with Brugada syndrome related ventricular ἀbrillation than in those without ventricular ἀbrillation under similar vagal tone. It may be that some patients with epilepsy at certain

262 Sudden Death in Epilepsy: Forensic and Clinical Issues

times exhibit a sympathovagal balance that is dominated by the parasympathetic nervous system and these patients may die in asystole. Most of the time, we consider stress to be associated only with adverse or negative circumstances. However, unexpected good news can also be a stress in a positive sense, especially if it is the fulἀllment of a heretofore seemingly out-of-reach quest. An explanation of why such an association could occur is speculative. However, while it is common knowledge that negative live events such as death, divorce, loss of a job, and so forth, can result in an increase in sympathetic parameters such as heart rate and blood pressure, relief from an adverse circumstance or the occurrence of a positive life event can result in more prominence of parasympathetic parameters such as decreased heart rate and blood pressure (see discussion above). One could speculate that in individuals who have the potential for an excessive parasympathetic response, the intervention of epileptiform discharges could augment this tendency to the point of asystole. Thus, when information about a SUDEP victim’s circumstances prior to the demise is solicited, one should consider eliciting the possibility of an intensely positive event as a risk factor.

References Ackerman, M. J., D. J. Tester, and D. J. Driscoll. 2001. Molecular autopsy of sudden unexplained death in the young. Am J Forensic Med Pathol 22 (2): 105–111. Anderberg, U. M. 2001. Stress-related syndromes—Contemporary illnesses. Lakartidningen 98 (51–52): 5860–5863. Basso, C., D. Corrado, and G. Thiene. 1999. Cardiovascular causes of sudden death in young individuals including athletes. Cardiol Rev 7 (3): 127–135. Bilan, A., A. Witczak, R. Palusinski, A. Ignatowicz, and J. Hanzlik. 2005. Circadian rhythm of the QT interval dispersion in healthy subjects. Correlation with heart rate variability circadian pattern. J Electrocardiol 38 (1): 36–42. Blumer, D. 1992. Postictal depression: signiἀcance for treatment of the neurobehavioral disorder of epilepsy. J Epilepsy 5: 214–219. Bohus, B., and S. M. Korte. 2000. Stress, the hypothalamo–pituitary–adrenal system and the heart. In The Nervous System and the Heart, ed. G. Ter Horst, chap. 6. Totowa, NJ: Humana Press. Burgerman, R. S., M. R. Sperling, J. A. French, A. J. Saykin, and M. J. O’Connor. 1995. Comparison of Mesial Versus Neocortical Onset Temporal Lobe Seizures: Neurodiagnostic Findings and Surgical Outcome. New York, NY: Raven Press. Buselli, E. F., and E. M. Stuart. 1999. Influence of psychosocial factors and biopsychosocial interventions on outcomes after myocardial infarction. J Cardiovasc Nurs 13 (3): 60–72. Callahan, T. D., U. Khokhar, L. Pozuelo, and J. B. Young. 2008. Case study in heart–brain interplay: A 53-year-old woman recovering from mitral valve repair. Cleve Clin J Med 75 (Suppl 2): S10–S14. Cechetto, D. 2000. Neuropathology and cardiovascular regulation. In The Nervous System and the Heart, ed. G. Ter Horst, chap. 4. Totowa, NJ: Humana Press. Davis, A. M., and B. H. Natelson. 1993. Brain–heart interactions. The neurocardiology of arrhythmia and sudden cardiac death. Tex Heart Inst J 20 (3): 158–169. Deuschle, M., and F. Lederbogen. 2002. Depression and cardiovascular disease: pathogenetic factors of the stress concept. Fortschr Neurol Psychiatr 70 (5): 268–275. DiMaio, T. G., and V. J. M. DiMaio. 2006. Excited Delirium Syndrome. Boca Raton: CRC Press. Dodd-O, J. H., and C. M. Lathers. 1990. A characterization of the lockstep phenomenon in phenobarbital-pretreated cats. In Epilepsy and Sudden Death, chap. 13. New York, NY: Marcel Dekker. Dominian, J., E. A. Serafetinides, and M. Dewhurst. 1963. A follow-up study of late-onset epilepsy: II. Psychiatric and social ἀndings. Br Med J 1 (5328): 431–435.

Stress and SUDEP

263

Earnest, M. P., G. E. Thomas, R. A. Eden, and K. F. Hossack. 1992. The sudden unexplained death syndrome in epilepsy: Demographic, clinical, and postmortem features. Epilepsia 33 (2): 310–316. Elfawal, M. A. 2000. Sudden unexplained death syndrome. Med Sci Law 40 (1): 45–51. Esposito, K., R. Marfella, P. Gualdiero, C. Carusone, A. Pontillo, G. Giugliano, G. Nicoletti, and D. Giugliano. 2003. Sympathovagal balance, nighttime blood pressure, and QT intervals in normotensive obese women. Obes Res 11 (5): 653–659. Fenwick, P. 1995. The basis of behavioral treatment in seizure control. Epilepsia 36: 46–50. Han, J., and G. K. Moe. 1964. Nonuniform recovery of excitability in ventricular muscle. Circ Res 14: 44–60. Hilz, M. J., O. Devinsky, W. Doyle, A. Mauerer, and M. Dutsch. 2002. Decrease of sympathetic cardiovascular modulation after temporal lobe epilepsy surgery. Brain 125 (Pt 5): 985–995. Homan, R. W. 1998. Adjunctive and combination therapy. Chapter 114. In Epilepsy. A Comprehensive Textbook, ed. J. J. Engel, T. A. Pedley, J. Aicardi, M. Dichter, U. Heinemann, W. Moshe, R. Porer, and D. Taylor. Philadelphia, PA: Lippincott-Raven. Jiang, W., M. Babyak, D. S. Krantz, R. A. Waugh, R. E. Coleman, M. M. Hanson, D. J. Frid et al. 1996. Mental stress–induced myocardial ischemia and cardiac events. JAMA 275 (21): 1651–1656. Jones, T. J., and M. J. Hallworth. 1999. Postmortem prolactin as a marker of antemortem stress. J Clin Pathol 52 (10): 749–751. Kanner, A. M., S. Stagno, P. Kotagal, and H. H. Morris. 1996. Postictal psychiatric events during prolonged video-electroencephalographic monitoring studies. Arch Neurol 53 (3): 258–263. Kozaric-Kovacic, D. 2008. Psychopharmacotherapy of posttraumatic stress disorder. Croat Med J 49 (4): 459–475. Kralios, F. A., and C. K. Millar. 1981. Sympathetic neural effects on regional atrial recovery properties and cardiac rhythm. Am J Physiol 240 (4): H590–H596. Krantz, D. S., W. J. Kop, H. T. Santiago, and J. S. Gottdiener. 1996. Mental stress as a trigger of myocardial ischemia and infarction. Cardiol Clin 14 (2): 271–287. Kummell, H. C., P. Van Leeuwen, C. Heckmann, P. Engelke, G. Kesting, H. Kremer, and A. Becher. 1993. Quality of life and circadian variation of heart rate and heart rate variability in short-term survivors and nonsurvivors after acute myocardial infarction. Clin Cardiol 16 (11): 776–782. Lathers, C. M., G. J. Kelliher, J. Roberts, and A. B. Beasley. 1978. Nonuniform cardiac sympathetic nerve discharge: mechanism for coronary occlusion and digitalis-induced arrhythmia. Circulation 57 (6): 1058–1065. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1977. Correlation of ouabain-induced arrhythmia and nonuniformity in the histamine-evoked discharge of cardiac sympathetic nerves. J Pharmacol Exp Ther 203 (2): 467–479. Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: the lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67 (3): 247–259. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008. The mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12 (1): 3–24. Lathers, C. M., and P. L. Schraeder. 2010. Animal model for sudden unexpected death in persons with epilepsy. Chapter 28 in Sudden Death in Epilepsy: Forensic and Clinical Issues, eds., C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton: CRC Press. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23 (6): 633–647. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9 (2): 236–242. Leor, J., W. K. Poole, and R. A. Kloner. 1996. Sudden cardiac death triggered by an earthquake. N Engl J Med 334 (7): 413–419. Linden, W., C. Stossel, and J. Maurice. 1996. Psychosocial interventions for patients with coronary artery disease: A meta-analysis. Arch Intern Med 156 (7): 745–752. Mattson, R. H. 1991. Emotional effects on seizure occurrence. Adv Neurol 55: 453–460.

264 Sudden Death in Epilepsy: Forensic and Clinical Issues Mendez, M. F. 1998. Postictal violence and epilepsy. Psychosomatics 39 (5): 478–480. Merz, C. N., D. S. Krantz, and A. Rozanski. 1993. Mental stress and myocardial ischemia. Correlates and potential interventions. Tex Heart Inst J 20 (3): 152–157. Mizumaki, K., A. Fujiki, T. Tsuneda, M. Sakabe, K. Nishida, M. Sugao, and H. Inoue. 2004. Vagal activity modulates spontaneous augmentation of ST elevation in the daily life of patients with Brugada syndrome. J Cardiovasc Electrophysiol 15 (6): 667–673. Moffett, A., and D. F. Scott. 1984. Stress and epilepsy: The value of a benzodiazepine–lorazepam. J Neurol Neurosurg Psychiatry 47 (2): 165–167. Molgaard, H., K. Hermansen, and P. Bjerregaard. 1994. Spectral components of short-term RR interval variability in healthy subjects and effects of risk factors. Eur Heart J 15 (9): 1174–1183. Murakawa, Y., H. Inoue, A. Nozaki, and T. Sugimoto. 1992. Role of sympathovagal interaction in diurnal variation of QT interval. Am J Cardiol 69 (4): 339–343. Nashef, L., D. R. Fish, J. W. Sander, and S. D. Shorvon. 1995. Incidence of sudden unexpected death in an adult outpatient cohort with epilepsy at a tertiary referral centre. J Neurol Neurosurg Psychiatry 58 (4): 462–464. Natelson, B. H., R. V. Suarez, C. F. Terrence, and R. Turizo. 1998. Patients with epilepsy who die suddenly have cardiac disease. Arch Neurol 55 (6): 857–860. O’Rourke, D. K., and C. M. Lathers. 1990. Interspike interval histogram characterization of synchronized cardiac sympathetic neural discharge and epileptogenic activity in the electrocorticogram of the cat. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. Schraeder, chap. 15. New York, NY: Marcel Dekker. Owada, M., Y. Aizawa, K. Kurihara, N. Tanabe, T. Aizaki, and T. Izumi. 1999. Risk factors and triggers of sudden death in the working generation: An autopsy proven case-control study. Tohoku J Exp Med 189 (4): 245–258. Pavlov, I. 1951. The Complete Collection of Works, Vol. 1. Moscow: ANSSR. Pritchard, P. B. 1997. Hormone changes in epilepsy. In Epilepsy. A Comprehensive Textbook, ed. J. J. Engel and T. A. Pedley, chap. 188. Philadelphia, PA: Lippincott-Raven. Pickworth, W. B., J. Gerard-Ciminara, and C. M. Lathers. 1990. Stress, arrhythmias, and seizures. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. Schraeder, chap. 22. New York, NY: Marcel Dekker. Richter, C. P. 1957. On the phenomenon of sudden death in animals and man. Psychosom Med 19 (3): 191–198. Rozanski, A., C. N. Bairey, D. S. Krantz, J. Friedman, K. J. Resser, M. Morell, S. Hilton-Chalfen, L. Hestrin, J. Bietendorf, and D. S. Berman. 1988. Mental stress and the induction of silent myocardial ischemia in patients with coronary artery disease. N Engl J Med 318 (16): 1005–1012. Samuels, M. 1997. Voodoo death revisited: The modern lessons of neurocardiology. The Neurologist 3: 293–304. Schwartz, P. J., H. L. Stone, and A. M. Brown. 1976. Effects of unilateral stellate ganglion blockade on the arrhythmias associated with coronary occlusion. Am Heart J 92 (5): 589–599. Selye, H. 1958. The Chemcal Prevention of Cardiac Necrosis. New York, NY: Ronald Press. Shah, R., M. M. Burg, A. Vashist, D. Collins, J. Liu, F. Jadbabaie, B. Graeber, C. Earley, R. Lampert, and R. Soufer. 2006. C-reactive protein and vulnerability to mental stress-induced myocardial ischemia. Mol Med 12 (11–12): 269–274. Soufer, R., and M. M. Burg. 2007. The heart–brain interaction during emotionally provoked myocardial ischemia: Implications of cortical hyperactivation in CAD and gender interactions. Cleve Clin J Med 74 Suppl 1: S59–S62. Sperling, M. R., H. Feldman, J. Kinman, J. D. Liporace, and M. J. O’Connor. 1999. Seizure control and mortality in epilepsy. Ann Neurol 46 (1): 45–50. Stauffer, A. Z., J. Dodd-O, and C. M. Lathers. 1989. The relationship of the lock-step phenomenon and precipitous changes in mean arterial blood pressure. Electroencephalogr Clin Neurophysiol 72 (4): 340–345.

Stress and SUDEP

265

Talman, W. T. 1985. Cardiovascular regulation and lesions of the central nervous system. Ann Neurol 18 (1): 1–13. Trimble, M. R., and M. M. Perez. 1980. Quantiἀcation of psychopathology in adult patients with epilepsy. In Epilepsy and Behavior, ed. B. Kulig, H. Meinardi, and G. Stores. Lisse: Swets & Zeitlinger. Tulner, D. M., and H. A. den Boer. 2000. Heart disease and cognitive/neuropsychiatric disorders. In The Nervous System and the Heart, ed. G. Ter Horst, chap. 15. Totowa, NJ: Humana Press. Wannamaker, B. B., and H. E. Booker. 1998. Treatment of provoked seizures. In Epilepsy. A Comprehensive Textbook, ed. Engel, T. A. Pedley, J. Aicardi, M. Dichter, U. Heinemann, S. Moshe, R. Porer, and D. Taylor, chap. 119. Philadelphia, PA: Lippincott-Raven. West, T. C., G. Falk, and P. Cervoni. 1956. Drug alteration of transmembrane potentials in atrial pacemaker cells. J Pharmacol Exp Ther 117 (2): 245–252. Westerhaus, M. J., and A. D. Loewy. 2001. Central representation of the sympathetic nervous system in the cerebral cortex. Brain Res 903 (1–2): 117–127. Zavodskaya, I. S., E. V. Moreua, and N. A. Novikova. 1980. Neurogenic Heart Lesions. Oxford: Pergamon. Ziegelstein, R. C. 2007. Acute emotional stress and cardiac arrhythmias. JAMA 298 (3): 324–329.

18

Genetics of Sudden Death in Epilepsy Neeti Ghali Lina Nashef

Contents 18.1 The Genetics of Epilepsy 18.2 Mortality in Speciἀc Syndromes 18.2.1 15q Inversion/Duplicaton [inv dup(15)] 18.2.2 Rett Syndrome 18.2.3 Other Examples of Syndromes with Multisystem Involvement 18.2.4 SCN1A-Related Disorders 18.3 SUDEP and Idiopathic Epilepsy 18.4 SUDEP in Genetic Epilepsies: How Does It Occur? 18.5 Genetic Susceptibility to SCD and SIDS 18.6 Conclusion References

267 269 269 270 270 271 272 275 276 278 278

18.1â•… The Genetics of Epilepsy Idiopathic epilepsies account for a substantial proportion of all epilepsies and are considered to be largely genetically determined (Steinlein 2008). The inheritance of these epilepsies may be Mendelian, whereby a single identiἀed mutation results in epilepsy and/or febrile seizures. However, there is often interfamilial and intrafamilial variability in the clinical phenotype, suggesting an effect of other genetic variants. The inheritance is more often non-Mendelian or complex, whereby the phenotype is thought to be determined by several more minor genetic defects as well as environmental effects. Epilepsies inherited in a complex or multifactorial manner will arise when a chance combination of certain susceptibility alleles come together with sufficient effect in the individual to push neuronal hyperexcitability over the seizure threshold (Mulley et al. 2005), but where each susceptibility allele alone is insufficient to cause seizures. Susceptibility genes with minor effect have hitherto been much harder to identify, and most genes conἀrmed to be implicated in idiopathic epilepsy have thus far been Mendelian. Almost all mutations identiἀed in idiopathic Mendelian epilepsies, mostly in large pedigrees, are known to be in ion channel genes or genes interacting with ion channel genes (see Table 18.1). While the broad phenotype is largely determined by a mutation in a major gene, other genes with minor effects as well as environmental factors may modulate its expression. The consequences of these modiἀer effects are twofold: incomplete penetrance, whereby not all carriers of the mutation are clinically affected, and variable phenotypic expression, whereby within a family, factors such as age of onset, type, severity and frequency of seizures, response to antiepileptic drug (AED) treatment, and duration of 267

268 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 18.1â•… Genes Found to Be Factors in Epilepsy Monogenic Epilepsies

Gene

Reference

Benign familial neonatal–infantile seizures

KCNQ2 KCNQ3 SCN2A

GEFS+, SMEI, FS

SCN1A SCN1B GABRG2 GABRG2 CLCN2 GABRA1 LGI1

Singh et al. 1998 Charlier et al. 1998 Heron et al. 2002; Striano et al. 2006 Claes et al. 2001; Escayg et al. 2000 Wallace et al. 1998 Harkin et al. 2002 Baulac et al. 2001 Haug et al. 2003 Cossette et al. 2002 Kalachikov et al. 2002

CHRNA4 CHRNB2 CHRNA2

Steinlein et al. 1995 De Fusco et al. 2000 Aridon et al. 2006

Benign familial neonatal seizures

GEFS+ GEFS+, SMEI, FS Absence epilepsy and FS IGE JME Autosomal dominant lateral temporal lobe epilepsy Autosomal dominant nocturnal frontal lobe epilepsy

Note: GEFS+, generalized epilepsy with febrile seizures plus; SMEI, severe myoclonic epilepsy of infancy; FS, febrile seizures; IGE, idiopathic generalized epilepsy; JME, juvenile myoclonic epilepsy.

the epileptic disorder may be different. Mechanisms for incomplete penetrance are largely unknown. Some believe that if there are mutations in more than one ion channel gene, the individual will have a more severe phenotype (Kearney et al. 2006). However, of interest is an animal study whereby mutations in two different ion channel genes that would individually result in opposing excitability defects resulted in nullifying the epilepsy phenotype in mice (Glasscock et al. 2007). In addition, there may be some genotype–phenotype correlation with interfamilial variability, whereby the severity of the disease relates to the type of mutation in a speciἀc gene; a nonsense mutation resulting in a stop codon, for example, may lead to a more severe phenotype. For all these reasons, it can sometimes be difficult to distinguish between monogenic epilepsies and those inherited in a complex manner. Genetic mutations may also result in symptomatic epilepsies as a result of a cortical malformation (e.g., a mutation in the GPR56 gene results in bilateral frontoparietal polymicrogyria that often presents with seizures). Mutations in the MECP2 gene result in Rett syndrome, an X-linked syndrome presenting with seizures, ataxia, and severe learning difficulties. Epilepsy in the context of learning difficulties may be a result of singlegene disorders such as Rett syndrome or due to a chromosomal abnormality such as 1p36 microdeletion syndrome, ring 20 and ring 14 syndromes, or other chromosomal aberrations such as inversion-duplication 15. The genes responsible for the epilepsy in these chromosomal syndromes are at present largely unknown. The development of higher resolution chromosomal studies such as comparative genomic hybridization using arrays may help delineate this and identify further epilepsy genes. With the advent of array technology, submicroscopic deletions and duplications are now being identiἀed. Microdeletions of 15q13.3 have been reported to be associated with epilepsy and learning difficulties (Sharp et al. 2008). Furthermore, microdeletions of 15q13.3 have been associated with idiopathic generalized epilepsy (IGE) (Helbig et al. 2009). Genomic imbalances in these regions in the form of microduplications have been found to be associated with autistic spectrum

Genetics of Sudden Death in Epilepsy

269

disorder (Miller et al. 2009) and schizophrenia (International Schizophrenia Consortium 2008). In both single-gene disorders and chromosomal disorders, the epilepsy may not always be a consistent feature of a syndrome and the phenotype and severity often varies. In some genetic syndromes, there may be an underlying brain abnormality, whereas in others, the brain appears developmentally normal. A classiἀcation of diseases frequently associated with epileptic seizures or syndromes is available in Engel’s (2001) proposed diagnostic scheme. A United Kingdom Workshop in 2006 concluded that although epilepsy genetics has advanced signiἀcantly over the past decade, this is merely the beginning. Truly characterizing the full extent of genomic variation related to multifactorial or complex epilepsy remains very much in the future (Sisodiya et al. 2007). It was noted, however, that an impact on clinical practice has already been observed. For example, as a result of the identiἀcation of SCN1A mutations in Mendelian families with a phenotype of generalized epilepsy with€ febrile seizures plus (GEFS+), sporadic cases of Dravet syndrome or severe€ myoÂ� clonic€epilepsy of infancy (SMEI) were also found to carry SCN1A mutations. These cases have been shown to respond well to stiripentol, which is considered to enhance central GABA transmission by increasing duration of channel opening (Quilichini et al. 2006) and a meta-analysis has shown signiἀcant improvement in seizure control (Kassai et al. 2008). It is hoped that this beneἀcial effect will translate into improved developmental outcome in what is currently a disorder with an extremely poor prognosis and demonstrates the key role of research into basic science for these disorders (Mullen and Scheffer 2009).

18.2â•… Mortality in Specific Syndromes A family history of an increased risk of sudden death, where looked for, has generally not been reported in sudden unexpected death in epilepsy (SUDEP) cohorts or in a case control study of SUDEP (Nashef et al. 1998; Langan et al. 2005). However, this should not be taken to exclude the possibility of a familial tendency in a subgroup. Most epidemiological studies include all epilepsies and not only those with a likely genetic basis. Studies may not be designed to identify increased risk of syncope, sudden infant death syndrome (SIDS), or sudden cardiac death (SCD), particularly if inheritance is complex or if penetrance is reduced. Data are also very limited when it comes to mortality studies in epilepsy in speciἀc syndromes, and much of what is discussed below does not have a ἀrm basis. Nevertheless, there is a suggestion that mortality rates may differ between epileptic syndromes over and above that expected from the severity of the epilepsy, as in the examples below. 18.2.1â•… 15q Inversion/Duplicaton [inv dup(15)] This chromosomal anomaly is caused by the presence of a supernumerary chromosome 15 and results in tetrasomy of the 15q11-q13 region. This region is also involved in both Angelman and Prader–Willi syndromes and overlaps with recently reported microdeletions and microduplications referred to above. Inv dup(15) syndrome is characterized by hypotonia, minor dysmorphic features, moderate to severe learning difficulties, autistic spectrum disorder, and seizures. Seizure types include spasms, atypical absences, and tonic and atonic seizures. The electroencephalograph (EEG) shows atypical hypsarhythmia, with large amplitude diffuse slow spike waves and/or multifocal abnormalities. Case

270 Sudden Death in Epilepsy: Forensic and Clinical Issues

reports indicate a variable phenotype. Mosaic inv dup (15) has been identiἀed in a healthy child (Loitzsch and Bartsch 2006), while a child with mild generalized epilepsy and a developmental disorder was found to have a large inv dup (15) (Chifari et al. 2002). Most case reports describe a phenotype where seizures are a signiἀcant and consistent problem. The IDEAS support group for inv dup (15) (Isodicentric 15 Exchange, Advocacy & Support) released a physician advisory update alerting of the risk of sudden, unexpected death in this group of patients, suggesting that the risk is in the order of 1% per year (IsoDicentric 15 Exchange, Advocacy & Support 2009). Six cases are reported on the Web site and the€mechÂ� anism is unknown in all cases with each individual dying in bed during the night, presumably while asleep. According to the Web site, “Five of the six young people had recognized seizure disorders. One had no recent seizures, and the remaining three had seizures that were described as well controlled at the time of death, and one had not had a seizure for more than a month.” While caution is required in assuming an increased risk without the data being published in a peer-reviewed publication, it is clear that the advisory clinicians felt there was cause for concern. At the time of writing, two sudden unexpected deaths in individuals with Inv dup(15) syndrome have been published (Hogart et al. 2009). Further studies on this rare chromosomal anomaly need to be carried out to conἀrm if there is a higher-than-expected rate of sudden unexpected death in these patients. 18.2.2â•…Rett Syndrome Epilepsy is a manifestation of Rett syndrome with partial and generalized seizures being reported in 50% to 90% of cases. Sudden death with no preceding symptoms is a recognized problem associated with Rett syndrome and seizures may be a partial explanation although the precise etiology is not always understood (Byard 2006). Sudden death has been reported in 22% to 26% of cases compared to 2.3% in the general population of the same age (Byard 2006). Several studies have been carried out demonstrating a prolongation of QTc interval in patients with Rett syndrome, the pathogenesis of which is unknown (Acampa and Guideri 2006). Brainstem dysfunction resulting in cardiac autonomic dysregulation is also described as being characterized by disturbed breathing and heart rate during sleep. Some studies have demonstrated labile breathing patterns and a reduction in cardiac vagal tone, indicating brainstem immaturity (Julu et al. 2001). Other support for autonomic dysregulation stems from the observation of decreased heart rate variability and sinus bradycardia (Axelrod et al. 2006). Therefore, the increase in sudden death may partly be related to the autonomic disturbance as well as perhaps to the seizures (Byard 2006). Defective autonomic nervous system control and cardiac arrhythmias relate more to functional problems than any defects demonstrated at postmortem (Byard, 2006). The pathophysiology of sudden death in Rett syndrome is an example where a single gene mutation may result in increased susceptibility to sudden death by a number of different mechanisms. 18.2.3â•…Other Examples of Syndromes with Multisystem Involvement There are other examples of syndromes with multisystem involvement, including epilepsy and increased mortality. Lafora disease, an inherited progressive myoclonic epilepsy characterized by intractable epilepsy in association with progressive neurological and cognitive deἀcit due to a mutation in the EPM2A gene on chromosome 6q24 is also associated with

Genetics of Sudden Death in Epilepsy

271

premature death from a variety of causes, including sudden death (Wick and Byard 2006). The same applies to the mitochondrial mutation 3243A>G resulting in mitochondrial encephalomyopathy, lactic acidosis, and strokelike episodes (MELAS), which increases the risk of death among carriers and their ἀrst-degree maternal relatives. Death was due to a variety of causes including metabolic disturbances, cardiac involvement, status epilepticus, and sudden death associated with epilepsy or diabetes (Majamaa-Voltti et al. 2008). These examples emphasize the need for mortality studies in speciἀc syndromes and the limitations of “all inclusive” mortality studies in epilepsy. 18.2.4â•… SCN1A-Related Disorders The vast majority of mutations identiἀed in the SCN1A gene have been in sporadic cases of SMEI. SMEI is characterized by normal early development followed by the onset of feversensitive and refractory epilepsy with GTCS or unilateral seizures beginning within the ἀrst year of life. Most mutations identiἀed are nonsense mutations although some are missense mutations. Approximately 5% to 10% of families with GEFS+, an idiopathic familial epilepsy syndrome characterized by febrile seizures extending beyond 6 years with or without afebrile seizures, have also been shown to have SCN1A missense mutations (Marini et al. 2007), although other genes (GABRG2, SCN1B) are also implicated (Helbig et al. 2008). In addition, SCN1A mutations have been identiἀed in familial hemiplegic migraine. Mortality rates in SMEI are observed to be higher than in some other epilepsies, somewhere in the order of 15% versus 5% (Kassai et al. 2008), although controlled prospective studies have not been carried out and would require multicenter collaboration due to the rarity of SMEI. In Dravet’s series of 63 patients with a clinical diagnosis of SMEI and a mean age at follow-up of 11 years 4 months, 10 died, 2 of whom were sudden unexplained deaths. One death was unclassiἀed, while other causes of death included drowning, accidents, infection, and status epilepticus. In a Japanese series of clinically diagnosed SMEI or borderline SMEI, 12 of 85 patients died. Seven patients died from status epilepticus, one from severe infection, and one by accidental drowning. Three died suddenly of unknown causes (Oguni et al. 2001). In a further study of clinical cases of SMEI, two patients died; one from an unknown cause (Caraballo and Fejerman 2006). Accurate person years of follow-up are generally not provided to allow for incidence of sudden death, standardized mortality rates, and conἀdence intervals to be calculated. Mortality rates in GEFS+ families have not been reported to be higher than expected from patients with epilepsy, although this has not been systematically examined. We described a family with GEFS+ and a novel SCN1A mutation with two SUDEP cases (Hindocha et al. 2008) in individuals with more severe epilepsy. Although DNA samples were unavailable from the SUDEP cases to conἀrm their carrier status of the SCN1A mutation, their epilepsy phenotype was consistent with the familial diagnosis of GEFS+. To exclude as far as possible an alternative cause for the sudden death in these individuals, a ἀrst-degree relative of each SUDEP case was screened for the most common cardiac channelopathy mutations with negative results. SUDEP incidence in this family was found to be 7/1000 (95% conἀdence interval, 1–25), compared with an incidence of less than 1/1000, in population-based epilepsy cohorts. The wide conἀdence intervals suggest that the two deaths could still have occurred by chance. Nevertheless, the two SUDEP cases occurred in individuals with uncontrolled epilepsy and there was no other family history of sudden premature death. These two cases raise the possibility of an increased genetic predisposition to sudden death in people with

272 Sudden Death in Epilepsy: Forensic and Clinical Issues

SCN1A mutations in the setting of uncontrolled seizures. Possible mechanisms are discussed below. While initial characterization of ion channels presumed that these proteins were more localized to a speciἀc tissue, it is being increasingly recognized that this is not the case. Tissues are now thought to be mosaic for these proteins, composed of many isoforms, each expressed in different proportions (Haufe et al. 2007). Although SCN1A may primarily be expressed in the brain, several studies have shown that Nav1.1 (SCN1A gene product) is present in various regions of the heart in rat and mouse (Rogart et al. 1989; Dhar et al. 2001; Marionneau et al. 2005), in rabbit neonate (Baruscotti et al. 1997), and in dog (Haufe et al. 2005). There is good evidence for a role for Nav1.1 in pacemaker function of the sinoatrial (SA) node, but a lack of expression in atrial muscle (Tellez et al. 2006). In mice, Nav1.1 (but not Nav1.5, a SCN5A gene product) was detected in the SA node, and moreover, when brain-type Na+ channels were selectively blocked, signiἀcantly reduced spontaneous heart rate and greater heart rate variability were observed (Maier et al. 2003). A role for Nav1.1 in pacemaker activity in the mouse SA node (Lei et al. 2004) and rat SA node (Du et al. 2007) was conἀrmed in other studies. In contrast, evidence for a role for Nav1.1 in ventricular function is contradictory. Nav1.1 has been detected in mouse ventricular myocytes (Maier et al. 2002) and when brain-type Na+ channels are blocked, ventricular function is reduced, suggesting a role in excitation-contraction coupling (Maier et al. 2002). However, another similar study in rat ventricular myocytes demonstrated no reduction in ventricular function (Brette and Orchard 2006). SCN1A mutations may also result in dysfunction in the brainstem, resulting in alteration in autonomic function and thereby theoretically predisposing to sudden death. SCN1A mutations have been identiἀed in familial hemiplegic migraine. The pathophysiology of migraine implicates the brainstem, whereby brain imaging studies have established reproducible changes in the brain (Goadsby 2007; Goadsby and Hargreaves 2008). In addition, a study describing a missense mutation in a case of SMEI has suggested possible dysfunction of the brainstem in this disorder (Kimura et al. 2005). Two brothers with SMEI were found to have a missense mutation and their father, also a carrier of the mutation, had experienced two simple FSs before the age of 4. Both siblings had a deranged sleep-wake cycle after late infancy, postulated to be due to the dysfunction of aminergic neurons in the brainstem. Studies in rat models have also shown good expression of SCN1A in the brainstem (Gong et al. 1999).

18.3â•… SUDEP and Idiopathic Epilepsy Although most SUDEP cases are associated with more intractable and usually focal or symptomatic epilepsies, cases with IGEs with a history of generalized tonic seizures are nevertheless well represented in SUDEP cohorts. This was evident in an early study by one of the authors (L.N.) where 9 of 26 SUDEP cases were classiἀed as having idiopathic primary generalized epilepsy (see Chapter 58, this book). This cohort was largely identiἀed through the self-help group Epilepsy Bereaved and not through specialized services. Of these IGE cases, one had reportedly never been treated, and one had reportedly discontinued medication independently. Another, with juvenile myoclonic epilepsy in remission€on valproic acid, had been independently considering medication reduction, but it is not known if this had taken place. Two others had only ever been treated with carbamazepine or phenytoin.

Genetics of Sudden Death in Epilepsy

273

Two, both with a positive photoparoxysmal response, were found dead near visual display units, in one case a computer game was on and the person was found dead with a bitten tongue; the other was watching a video at the time of death. One had discontinued medication, while the other had only had a small number of convulsions in similar settings previously, and as already referred to, had never been treated (Nashef et al. 1998). In Langan et al.’s (2005) case control study, no association was found with syndromic diagnosis, but information on this variable was incomplete. In Nilsson et al.’s (1999) study, no increase in risk was associated with any particular type of epilepsy, but a lowered risk was associated with localization-related symptomatic epilepsy compared with generalized idiopathic epilepsy, especially among men, but numbers were small and conἀdence intervals wide. Genton and Gelisse (2001) reported three cases of premature death in juvenile myoÂ� clonic epilepsy. The report was useful in both highlighting this as a potential problem€and in providing person years of follow-up. Among 170 consecutive juvenile myoclonic epilepsy cases, three female patients died prematurely without autopsy being performed. One case had anorexia nervosa and died from severe aspiration pneumonia after provoked vomiting. The second had a history of psychosis and a case of IGE in her family. She had uncontrolled epilepsy and was found, aged 16, cyanotic and unconscious one morning “in the toilets of her institution” and died before resuscitation. The third had uncontrolled epilepsy, was also on neuroleptics, had a borderline personality disorder, a history of alcoholism, and low compliance. She was found dead at home at the age of 42 years. The authors concluded that severe mental disorders were risk factors for unexpected death in juvenile myoclonic epilepsy with a death ratio, if only two of these three cases are considered, of 0.9/1000. Aurlien et al. (2007) reported on four consecutive female SUDEP cases aged 25, 16, 37, and 24 years with idiopathic epilepsy treated with lamotrigine monotherapy. One case was unclassiἀed (auras suggested focal epilepsy but MRI was normal and EEG showed bilateral synchronous epileptogenic activity) and the other three were idiopathic generalized (one juvenile myoclonic epilepsy with photosensitivity, another with concomitant diabetes). The authors considered four possible explanations for this observation: an insufficient effect of lamotrigine leading to fatal seizures, a direct effect of lamotrigine on vital functions, such as cardiac rhythmicity (given that lamotrigine inhibits the cardiac rapid delayed rectiἀer potassium ion current (Ikr)), a combination of drug-induced effects and seizures, or coincidence. It is interesting that although they had not systematically identiἀed all SUDEP cases in their area during this period, these four cases were the only ones they were aware of among their outpatients and represented all SUDEP patients reported by the department pathologist. Also of interest was the control of the epilepsy. Seizure frequency on last outpatient hospital visit in cases 1 through 4, respectively, was: 1.5 simple partial seizures/ month; seizure free for 6 months; seizure free for 7 months; two seizures during the last week (previously seizure-free for 3 months). At autopsy, case 1 did not have AED levels performed, case 2 did not have detectable blood concentration of lamotrigine (previously 7 µmol/l), case 3 had a level of 15 µmol/l, previously 24.4, and one case had a lower postmortem lamotrigine (3.2 µmol/l) compared with the last antemortem level (27.5 µmol/l) when on combination therapy with carbamazepine. The authors considered the possibility that “there may be subgroups of patients with idiopathic epilepsy and generalized tonic– clonic seizures treated with lamotrigine that are at an increased risk of SUDEP.” They cited reports of a greater risk of drug-induced torsade in females, who also have an increased prevalence of symptoms of congenital long QT syndrome (LQTS) and an increase in episodes of supraventricular tachycardia in the perimenstrual period in susceptible patients.

274 Sudden Death in Epilepsy: Forensic and Clinical Issues

These careful observations yet again highlight the importance of mortality studies in speciἀc syndromes and suggest a greater risk of SUDEP in those with idiopathic epilepsy, even when the epilepsy is not severe. Which of the possible explanations suggested by the authors are true remain unknown. Note that a limited number of other studies (Timmings 1998; Langan et al. 2005; for review, see Rugg-Gunn and Nashef 2009) suggested that carbamazepine might also increase risk to a small extent again through an effect on cardiac function. In Aurlien et al.’s (2007) series, there is also possible selection bias in terms of preferential prescribing of lamotrigine to females with idiopathic epilepsy of childbearing age, whereas valproate, although particularly effective, may have been avoided because of potential teratogenicity. Theoretically, a particular AED or AED combination could have a detrimental effect by either increasing risk of sudden death through a variety of mechanisms or by giving insufficient protection for the seizure disorder. One angle that has not been explored but perhaps suggested by absent or lower levels in some of these cases (not withstanding reservations about postmortem blood levels), is a possible differential SUDEP risk associated with AED withdrawal, through an effect on seizure severity or autonomic function. This potential AED factor may have relevance if there is nonadherence to treatment or abrupt withdrawal. It is generally considered, for example, that the full effect of starting and stopping valproate in IGE cases is not all immediate with a delayed effect often observed. Thus, it is possible that effective control may be lost less quickly with valproate than when some other drugs are omitted. The same group (Aurlien et al. 2009) published a later follow up report on case 1 when autopsy DNA sequencing of LQTS-associated genes revealed a novel missense mutation in the SCN5A gene coding for the cardiac sodium channel, voltage-gated, type V alpha subunit. They discussed whether the mutation may explain both the epilepsy and the sudden death and the possible effect of lamotrigine on cardiac ion channel function. This is a particularly interesting report as it provides evidence in support of the same genetic mutation giving rise to both epilepsy and susceptibility to cardiac death. We have studied a small pedigree of an otherwise well young woman with juvenile myoclonic epilepsy who died suddenly while on lamotrigine, having been previously fully controlled on valproate, but who changed medication because of weight gain and potential teratogenicity, and whose control was not as good on lamotrigine as it had been on valproate. As can be seen in Figure 18.1, her mother had undiagnosed blackouts and her brother died of SIDS. LQT channel mutation screen in this pedigree, however, was negative. The pedigree raises the possibility of wider overlap presentations, as yet unexplored. Two other individuals with idiopathic epilepsy and cardiac arrhythmias have also so far been negative on screening for the most common LQTS and Brugada syndrome (BS) gene mutations (70% and 20%, respectively). Recently, it has been demonstrated that individuals with mutations in KCNH2 seen in LQTS are more likely to have a personal history of seizures than other subtypes of LQTS, raising the possibility that the KCNH2-encoded potassium channel could confer susceptibility to seizures (Johnson et al. 2009). Also of particular interest is the SCN1B gene previously implicated in GEFS+, absence epilepsy and temporal lobe epilepsy, and now in BS (Watanabe et al. 2008). This is discussed in more detail below. In our view, the available, though limited, evidence raises the possibility that in idiopathic epilepsy, with a history of generalized tonic clonic seizures, SUDEP may occur more often than in other syndromes with comparable epilepsy severity. At present, this is only a

Genetics of Sudden Death in Epilepsy

275

• Age of seizure onset—15 years 1.1

1.2

1.3 1.4 Late-onset epilepsy

II.I

III.1

II.2 Syncopal episodes

III.2

Seizures/epilepsy

III.3 III.4 SUDEP SIDS

• Myoclonic jerks and rare generalized tonic–clonic seizure • Idiopathic uncontrolled epilepsy at time of death • Found dead in bed • Clear bite mark on tip and left side of tongue found at postmortem supporting probability of epileptic seizure

DNA screened for the most common LQTS and BS gene mutations (account for 75% of all LQTS and BS). Mutations in these genes have been excluded as far as possible, as a cause of the sudden death.

Figure 18.1╇ SUDEP case study.

suggestion, but it needs to be explored further. AED choices may be particularly important in this subgroup as suggested by Aurlien et al. (2007). It is important in this context, however, to stress the potential for misdiagnosis of LQT as epilepsy, and to emphasize the need for clinicians to be constantly vigilant in ensuring diagnostic accuracy and in screening for alternative or concomitant diagnosis in those diagnosed with epilepsy.

18.4â•… SUDEP in Genetic Epilepsies: How Does It Occur? In SUDEP, different risk factors and mechanisms may operate with a ἀnal common pathway of cardiorespiratory compromise. Respiratory compromise and hypoxia occur frequently in seizures, particularly if convulsive, and the severity may be influenced by position and airway during postictal coma, particularly in the absence of someone capable of giving assistance. Cardiac changes occur during seizures and although sinus tachycardia is most commonly observed during seizures, ictal sinus arrest also occurs as do rare malignant dysrhythmias. Functional cardiac changes such as apical ballooning may also occur (Stöllberger and Finsterer 2004). In discussing genetic predisposition, an increased susceptibility to sudden death may be due to cardiac mechanisms, reflecting underlying processes common to neurological and cardiac functions, autonomic function, or brainstem control of respiration. Etiological factors underlying SUDEP are likely to be heterogeneous, in much the same way as in sudden unexpected death (SUD) and SIDS (see below). Given the association between intractable epilepsy and SUDEP, one can postulate that in some cases of idiopathic epilepsy a primarily “neuronal” mutation, if it can also cause a predisposition to cardiac arrhythmias, may manifest as sudden death in persons in whom the epilepsy is uncontrolled. In others there may be an unrelated coexisting “mild” susceptibility to SCD that would manifest itself in the presence of uncontrolled seizures (Nashef et al. 2007). In our 2007 review, we discussed the possibility of overlap between susceptibility to SCD and idiopathic epilepsy and that further genetic and epidemiological studies are needed (Nashef et al. 2007). However, at the time there was no bridging evidence to

276 Sudden Death in Epilepsy: Forensic and Clinical Issues

support this hypothesis. Since then there have been reports as above, which suggest that ion channel gene mutations already known to have more than a single pathological role, as demonstrated, for example, by LQTS with deafness, can cause both epilepsy and LQTS or other inherited susceptibility to cardiac dysrhythmias, although such reports are currently uncommon. Animal models provide evidence in support of this hypothesis as in the case of the Ca2+ release channel ryanodine receptor 2 (RyR2) required for excitation-Â�contraction coupling in the heart- and expressed in the brain. Mutations in RyR2, which result in “leaky” RyR2 channels, have been linked to exercise-induced SCD and catecholaminergic polymorphic ventricular tachycardia. Mice heterozygous for the R2474S mutation in Ryr2 exhibited spontaneous generalized tonic–clonic seizures (without cardiac arrhythmias), exercise-induced ventricular arrhythmias, and SCD (Lehnart et al. 2008). Treatment with a compound inhibiting the channel leak prevented cardiac arrhythmias and raised the seizure threshold. The authors proposed that this was a combined neurocardiac disorder.

18.5â•… Genetic Susceptibility to SCD and SIDS In considering possible genetic predisposition to SUDEP, it may useful to briefly review the evidence of genetic susceptibility to SCD and SIDS. At least 4% of sudden deaths are unexplained at autopsy and on average a quarter of these may be due to inherited cardiac disease (Behr et al. 2008). These ἀgures will vary according to age group (Rodriguez-Calvo et al. 2008). Diagnosis is crucial as close relatives may be at potential risk of also having a fatal cardiac event. While the genetic causes of SCD also include structural abnormalities (e.g., inherited cardiomyopathies), a proportion (Saenen and Vrints 2008) are a result of primary arrhythmogenic disorders, also known as cardiac channelopathies, such as LQTS, short QT syndrome, BS, and catecholaminergic polymorphic ventricular tachycardia. Most cases of LQTS and BS are inherited from a parent who may or may not show clinical symptoms (Morales et al. 2008). In a study of autopsies in cases of patients aged 5 to 35 years who died suddenly from cardiac causes, 29% were arrhythmia-related SCD (Puranik et al. 2005). Of the cardiac cases, SCD in a ἀrst-degree relative was reported in only 4.5% of cases with a low yield of signiἀcant positive family history from these cases. The diagnostic yield of investigating relatives of individuals who had died suddenly with no explanation has also been examined. In one study (Tan et al. 2005), 43 families with one sudden unexpected death victim who had died under the age of 40 were investigated. Seven of the 43 families studied (16%) were found clinically to have LQTS or BS, diagnoses that were conἀrmed with molecular techniques in four of these seven families (9%). A slightly larger study showed that 21 of 57 families (37%) were clinically identiἀed to have either deἀnite or probable LQTS or BS. Molecular conἀrmation was made in six of the 57 families (11%) (Behr et al. 2008). Once molecular conἀrmation is achieved, testing may be offered to appropriate relatives. One study suggests that clinical evaluation alone is no longer appropriate to exclude a diagnosis of LQTS, as penetrance may be very low (around 25%) (Priori et al. 1998). Congenital LQTS is mostly inherited in an autosomal dominant fashion (RomanoWard syndrome) and up until now, mutations in 12 genes have been identiἀed to result in LQTS (KCNQ1, KCNH2, SCN5A, ANK2, KCNE1, KCNE2, KCNJ2, CACNA1C, CAV3, SCN4B, AKAP9, and α-1-syntrophin) (Van Norstrand and Ackerman 2009). All of these genes are expressed primarily in cardiac tissue and most of these genes encode an ion

Genetics of Sudden Death in Epilepsy

277

channel protein. Those that do not have been found to affect the sodium or potassium channel (Rodriguez-Calvo et al. 2008). BS is characterized by episodes of VF and in approximately 20% of cases a positive family history of unexplained sudden death is present (Miura et al. 2008). BS is genetically heterogeneous and mutations in the sodium channel gene SCN5A account for 20% to 30% of cases (Van Norstrand and Ackerman 2009). More recently, a mutation in SCN1B, primarily a neuronal voltage-gated sodium channel and implicated in GEFS+ (Wallace et al. 1998; Singh et al. 1999) but also expressed in the heart, has been implicated in BS (Watanabe et al. 2008). This is an exciting discovery, as it is an example of a gene having a pathological effect on the heart and the brain. Mutations in SCN1B have been previously reported in a family with early onset absence epilepsy without the occurrence of febrile seizures (Audenaert et al. 2003) and temporal lobe epilepsy even without febrile seizures (Scheffer et al. 2007). Cardiac abnormalities have not been commented on in these families. β1-null mice exhibit a severe seizure disorder and premature death (Chen et al. 2004). In addition, these mice exhibit bradycardia and prolonged rate-corrected QT intervals (Lopez-Santiago et al. 2007). These changes suggest that this protein has a possible role in the heart. Other genes such as CACNA1C, CACNB2b, and GPD1-L, known to modulate sodium channel function, are also implicated in BS. Since the advent of genome-wide association studies, research examining genetic susceptibility to various disorders has developed signiἀcantly. Single nucleotide polymorphisms (SNPs) in the LQT genes have been identiἀed as demonstrating a genetic propensity to developing increased QT interval/cardiac arrhythmias/SCD, not related to a single gene defect but because of multiple low-risk alleles (Pfeufer et al. 2005; Gouas et al. 2005). However, replication data at present is lacking (Schulze-Bahr 2006). Of interest in this context is the variant allele (S1103Y) of the cardiac sodium channel gene SCN5A, which (1) has a subtle effect on risk in African Americans (13% of whom are carriers), manifesting if there are additional acquired risk factors, with most carriers never having an arrhythmia (Splawski et al. 2002), (2) has a strong effect on risk in Caucasians where it is very rare (Chen et al. 2002), and (3) is associated, if homozygous, with SIDS in AfricanAmericans (Plant et al. 2006). Using similar techniques, genetic susceptibility to SIDS has been examined (WeeseMayer et al. 2007). Around 5% to 10% of SIDS cases (9.5% by Arnestad et al. 2007) are linked to ion channelopathies (Tester and Ackerman 2005). From an 18-year study (Schwartz et al. 1998), an increase in the QT interval in the ἀrst week of life was established in individuals who went on to die from SIDS. SNPs in common cardiac channel genes may confer risk to sudden death at any age including infancy, even prenatally. Several studies have identiἀed differences between infants with SIDS and control infants with gene polymorphisms in SCN5A (Plant et al. 2006). Further research has identiἀed other genetic factors predisposing to SIDS. For example, decreased serotonin (5HT) receptor binding in the 5HT pathways of the medulla has been identiἀed in a small number of SIDS cases (Paterson et al. 2006; see Paterson’s chapter 5). In addition, association between the serotonin transporter gene (5-HTT) and SIDS has been demonstrated (Opdal and Rognum, 2004), suggesting that serotonin may play a regulatory role in SIDS. Of interest in this context is the protective effect of fluoxetine, a selective serotonin reuptake inhibitor, in reducing ictal respiratory arrest in DBA/2 mice with audiogenic seizures at doses that did not reduce seizure severity (Tupal and Faingold 2006). Polymorphisms in genes involved in inflammatory and infectious processes such as interleukin-10 (IL-10) gene have been shown to be associated with both SIDS and infectious

278 Sudden Death in Epilepsy: Forensic and Clinical Issues

death (Korachi et al. 2004) such as pneumonia and Epstein-Barr virus (Weese-Mayer et al. 2007). In addition, genes pertinent to the development of the autonomic nervous system are also being studied because of reports of autonomic dysregulation in SIDS victims and polymorphisms have been found to be associated in several genes (PHOX2a, RET, ECE1, TLX3, EN1) (Weese-Mayer et al. 2004).

18.6â•… Conclusion While genetic predisposition may be less prominent than the severity of the epilepsy, treatment-related factors, and supervision in many cases of SUDEP, it is still a potentially very important area requiring further research, particularly in idiopathic epilepsy. Future research in this area should include epidemiological studies looking at mortality in speciἀc syndromes, as well as studies of possible overlap of epilepsy and syncope in idiopathic epilepsy and in LQT cohorts. Clinicians need to be more aware, not only of the potential for misdiagnosis but also for potential clinical overlap. There is also an urgent need to collect DNA and clinical details in SUDEP cases, initially to carry out screening for genetic mutations in selected cases, but also to carry out sufficiently powered association studies. This is a difficult but not insurmountable task; one possible source is DNA from dried blood spots from Guthrie cards taken from newborns in the United Kingdom. These are stored for many years and can be retrospectively retrieved to carry out molecular studies (Skinner et al. 2004).

References Acampa, M., F. Guideri. 2006. Cardiac disease and Rett syndrome. Arch Dis Child 91: 440–443. Aridon, P., C. Marini, C. Di Resta, E. Brilli, M. De Fusco, F. Politi, E. Parrini et al. 2006. Increased sensitivity of the neuronal nicotinic receptor alpha 2 subunit causes familial epilepsy with nocturnal wandering and ictal fear. Am J Hum Genet 79: 342–350. Arnestad, M., L. Crotti, T. O. Rognum, R. Insolia, M. Pedrazzini, C. Ferrandi, A. Vege et al. 2007. Prevalence of long-QT syndrome gene variants in sudden infant death syndrome. Circulation 115: 361–367. Audenaert, D., L. Claes, B. Ceulemans, A. Lofgren, C. Van Broeckhoven, and P. De Jonghe. 2003. A deletion in SCN1B is associated with febrile seizures and early-onset absence epilepsy. Neurology 61: 854–856. Aurlien, D., E. Tauboll, and L. Gjerstad. 2007. Lamotrigine in idiopathic epilepsy—Increased risk of cardiac death? Acta Neurol Scand 115: 199–203. Aurlien, D., T. P. Leren, E. Tauboll, and L. Gjerstad. 2009. New SCN5A mutation in a SUDEP victim with idiopathic epilepsy. Seizure 18: 158–160. Axelrod, F. B., G. G. Chelimsky, and D. E. Weese-Mayer. 2006. Pediatric autonomic disorders. Pediatrics 118: 309–321. Baruscotti, M., R. Westenbroek, W. A. Catterall, D. DiFrancesco, and R. B. Robinson. 1997. The newborn rabbit sino-atrial node expresses a neuronal type I-like Na+ channel. J Physiol 498 (Pt 3): 641–648. Baulac, S., G. Huberfeld, and I. Gourἀnkel-An et al. 2001. First genetic evidence of GABA(A) receptor dysfunction in epilepsy: A mutation in the gamma2-subunit gene. Nat Genet 28: 46–48. Behr, E. R., C. Dalageorgou, M. Christiansen, P. Syrris, S. Huges, M. T. Tome Esteban, E. Rowland, S. Jeffery, and W. J. McKenna. 2008. Sudden arrhythmic death syndrome: Familial evaluation identiἀes inheritable heart disease in the majority of families. Eur Heart J 29: 1670–1680.

Genetics of Sudden Death in Epilepsy

279

Brette, F., and C. H. Orchard. 2006. No apparent requirement for neuronal sodium channels in excitation-Â�contraction coupling in rat ventricular myocytes. Circ Res 98: 667–674. Byard, R. W. 2006. Forensic issues and possible mechanisms of sudden death in Rett syndrome. J Clin Forensic Med 13: 96–99. Caraballo, R. H., and N. Fejerman. 2006. Dravet syndrome: A study of 53 patients. Epilepsy Res 70 (S1): S231–S238. Charlier, C., N. A. Singh, S. G. Ryan, T. B. Lewis, B. E. Reus, R. J. Leach, and M. Leppert. 1998. A pore mutation in a novel KQT-like potassium channel gene in an idiopathic epilepsy family. Nat Genet 18: 53–55. Chen, C., R. E. Westenbroek, X. Xu, C. A. Edwards, D. R. Sorenson, Y. Chen, D. P. McEwen et al. 2004. Mice lacking sodium channel beta1 subunits display defects in neuronal excitability, sodium channel expression, and nodal architecture. J Neurosci 24: 4030–4042. Chen, S., M. K. Chung, D. Martin, R. Rozich, P. J. Tchou, and Q. Wang. 2002. SNP S1103Y in the cardiac sodium channel gene SCN5A is associated with cardiac arrhythmias and sudden death in a white family. J Med Genet 39: 913–915. Chifari, R., R. Guerrini, M. Pierluigi, S. Cavani, V. Sgro, M. Elia, R. Canger et al. 2002. Mild generalized epilepsy and developmental disorder associated with large inv dup(15). Epilepsia 43: 1096–1100. Claes, L., J. Del Favero, B. Ceulemans, L. Lagae, C. Van Broeckhoven, and P. De Jonghe. 2001. De novo mutations in the sodium-channel gene SCN1A cause severe myoclonic epilepsy of infancy. Am J Hum Genet 68: 1327–1332. Cossette, P., L. Liu, K. Brisebois, H. Dong, A. Lortie, M. Vanasse, J. M. Saint-Hilaire et al. 2002. Mutation of GABRA1 in an autosomal dominant form of juvenile myoclonic epilepsy. Nat Genet 31: 184–189. De Fusco, M., A. Becchetti, A. Patrignani, G. Annesi, A. Gambardella, A. Quattrone, A. Ballabio, E. Wanke, and G. Casari. 2000. The nicotinic receptor beta 2 subunit is mutant in nocturnal frontal lobe epilepsy. Nat Genet 26: 275–276. Dhar, M. J., C. Chen, I. Rivolta, H. Abriel, R. Malhotra, L. N. Mattei, F. C. Brosius,€R.€S.€Kass,€and L.€ L.€ Isom. 2001. Characterization of sodium channel alpha- and beta-subunits in rat and mouse cardiac myocytes. Circulation 103: 1303–1310. Du, Y., X. Huang, T. Wang et al. 2007. Downregulation of neuronal sodium channel subunits Nav1.1 and Nav1.6 in the sinoatrial node from volume-overloaded heart failure rat. Pflugers Arch 454€(3): 451–459. Engel, Jr., J. 2001. A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: Report of the ILAE Task Force on Classiἀcation and Terminology. Epilepsia 42: 796–803. Escayg, A., B. T. MacDonald, M. H. Meisler et al. 2000. Mutations of SCN1A, encoding a neuronal sodium channel, in two families with GEFS+2. Nat Genet 24: 343–345. Genton, P., and P. Gelisse. 2001. Premature death in juvenile myoclonic epilepsy. Acta Neurol Scand 104: 125–129. Glasscock, E., J. Qian, J. W. Yoo, and J. L. Noebels. 2007. Masking epilepsy by combining two epilepsy genes. Nat Neurosci 10: 1554–1558. Goadsby, P. J. 2007. Recent advances in understanding migraine mechanisms, molecules and therapeutics. Trends Mol Med 13: 39–44. Goadsby, P. J., and R. Hargreaves. 2008. Refractory migraine and chronic migraine: Pathophysiological mechanisms. Headache 48: 1399–1405. Gong, B., K. J. Rhodes, Z. Bekele-Arcuri, and J. S. Trimmer. 1999. Type I and type II Na(+) channel alpha-subunit polypeptides exhibit distinct spatial and temporal patterning, and association with auxiliary subunits in rat brain. J Comp Neurol 412: 342–352. Gouas, L., V. Nicaud, M. Berthet et al. 2005. Association of KCNQ1, KCNE1, KCNH2 and SCN5A polymorphisms with QTc interval length in a healthy population. Eur J Hum Genet 13: 1213–1322.

280 Sudden Death in Epilepsy: Forensic and Clinical Issues Harkin, L. A., D. N. Bowser, L. M. Dibbens et al. 2002. Truncation of the GABA(A)-receptor gamma2 subunit in a family with generalized epilepsy with febrile seizures plus. Am J Hum Genet 70: 530–536. Haufe, V., J. M. Cordeiro, T. Zimmer et al. 2005. Contribution of neuronal sodium channels to the cardiac fast sodium current INa is greater in dog heart Purkinje ἀbers than in ventricles. Cardiovasc Res 65: 117–127. Haufe, V., C. Chamberland, and R. Dumaine. 2007. The promiscuous nature of the cardiac sodium current. J Mol Cell Cardiol 42: 469–477. Haug, K., M. Warnstedt, A. K. Alekov et al. 2003. Mutations in CLCN2 encoding a voltage-gated chloride channel are associated with idiopathic generalized epilepsies. Nat Genet 33: 527–532. Helbig, I., I. E. Scheffer, J. C. Mulley, and S. F. Berkovic. 2008. Navigating the channels and beyond: Unravelling the genetics of the epilepsies. Lancet Neurol 7: 231–245. Helbig, I., H. C. Mefford, A. J. Sharp et al. 2009. 15q13.3 microdeletions increase risk of idiopathic generalized epilepsy. Nat Genet 41: 160–162. Heron, S. E., K. M. Crossland, E. Andermann et al. 2002. Sodium-channel defects in benign familial neonatal–infantile seizures. Lancet 360: 851–852. Hindocha, N., L. Nashef, F. Elmslie et al. 2008. Two cases of sudden unexpected death in epilepsy in a GEFS+ family with an SCN1A mutation. Epilepsia 49: 360–365. Hogart, A., K. N. Leung, N. J. Wang et al. 2009. Chromosome 15q11-13 duplication syndrome brain reveals epigenetic alterations in gene expression not predicted from copy number. J Med Genet 46: 86–93. International Schizophrenia Consortium. 2008. Rare chromosomal deletions and duplications increase risk of schizophrenia. Nature 455: 237–241. IsoDicentric 15 Exchange, Advocacy & Support 2009. Johnson, J. N., N. Hofman, C. M. Haglund, G. D. Cascino, A. A. M. Wilde, and M. J. Ackerman. 2009. Identiἀcation of a possible pathogenic link between congenital long QT syndrome and epilepsy. Neurology 72: 224–231. Julu, P. O., A. M. Kerr, F. Apartopoulos et al. 2001. Characterisation of breathing and associated central autonomic dysfunction in the Rett disorder. Arch Dis Child 85: 29–37. Kalachikov, S., O. Evgrafov, B. Ross et al. 2002. Mutations in LGI1 cause autosomal-dominant partial epilepsy with auditory features. Nat Genet 30: 335–341. Kassai, B., C. Chiron, S. Augier et al. 2008. Severe myoclonic epilepsy in infancy: A systematic review and a meta-analysis of individual patient data. Epilepsia 49: 343–348. Kearney, J. A., Y. Yang, B. Beyer et al. 2006. Severe epilepsy resulting from genetic interaction between Scn2a and Kcnq2. Hum Mol Genet 15: 1043–1048. Kimura, K., T. Sugawara, E. Mazaki-Miyazaki et al. 2005. A missense mutation in SCN1A in brothers with severe myoclonic epilepsy in infancy (SMEI) inherited from a father with febrile seizures. Brain Dev 27: 424–430. Korachi, M., V. Pravica, A. J. Barson, I. V. Hutchinson, and D. B. Drucker. 2004. Interleukin 10 genotype as a risk factor for sudden infant death syndrome: determination of IL-10 genotype from wax-embedded postmortem samples. FEMS Immunol Med Microbiol 42: 125–129. Langan, Y., L. Nashef, and J. W. Sander. 2005. Case-control study of SUDEP. Neurology 64: 1131–1133. Lehnart, S. E., M. Mongillo, A. Bellinger et al. 2008. Leaky Ca2+ release channel/ryanodine receptor 2 causes seizures and sudden cardiac death in mice. J Clin Invest 118: 2230–2245. Lei, M., S. A. Jones, J. Liu et al. 2004. Requirement of neuronal- and cardiac-type sodium channels for murine sinoatrial node pacemaking. J Physiol 559: 835–848. Loitzsch, A., and O. Bartsch. 2006. Healthy 12-year-old boy with mosaic inv dup(15)(q13). Am J Med Genet A 140: 640–643. Lopez-Santiago, L. F., L. S. Meadows, S. J. Ernst et al. 2007. Sodium channel Scn1b null mice exhibit prolonged QT and RR intervals. J Mol Cell Cardiol 43: 636–647.

Genetics of Sudden Death in Epilepsy

281

Maier, S. K., R. E. Westenbroek, K. A. Schenkman, E. O. Feigl, T. Scheuer, and W. A. Catterall. 2002. An unexpected role for brain-type sodium channels in coupling of cell surface depolarization to contraction in the heart. Proc Natl Acad Sci U S A 99: 4073–4078. Maier, S. K., R. E. Westenbroek, T. T. Yamanushi et al. 2003. An unexpected requirement for braintype sodium channels for control of heart rate in the mouse sinoatrial node. Proc Natl Acad Sci U S A 100: 3507–3512. Majamaa-Voltti, K., J. Turkka, M. L. Kortelainen, H. Huikuri, and K. Majamaa. 2008. Causes of death in pedigrees with the 3243A>G mutation in mitochondrial DNA. J Neurol Neurosurg Psychiatry 79: 209–211. Marini, C., D. Mei, T. Temudo et al. 2007. Idiopathic epilepsies with seizures precipitated by fever and SCN1A abnormalities. Epilepsia 48: 1678–1685. Marionneau, C., B. Couette, J. Liu et al. 2005. Speciἀc pattern of ionic channel gene expression associated with pacemaker activity in the mouse heart. J Physiol 562: 223–234. Miller, D. T., Y. Shen, L. A. Weiss et al. 2009. Microdeletion/duplication at 15q13.2q13.3 among individuals with features of autism and other neuropsychiatirc disorders. J Med Genet 46€(4): 242–248. Miura, D., K. Nakamura, and T. Ohe. 2008. Update on genetic analysis in Brugada syndrome. Heart Rhythm 5: 1495–1496. Morales, A., J. Cowan, J. Dagua, and R. E. Hershberger. 2008. Family history: An essential tool for cardiovascular genetic medicine. Congest Heart Fail 14: 37–45. Mullen, S. A., and I. E. Scheffer. 2009. Translational research in epilepsy genetics: Sodium channels in man to interneuronopathy in mouse. Arch Neurol 66: 21–26. Mulley, J. C., I. E. Scheffer, S. Petrou, L. M. Dibbens, S. F. Berkovic, and L. A. Harkin. 2005. SCN1A mutations and epilepsy. Hum Mutat 25: 535–542. Nashef, L., S. Garner, J. W. Sander, D. R. Fish, and S. D. Shorvon. 1998. Circumstances of death in sudden death in epilepsy: Interviews of bereaved relatives. J Neurol Neurosurg Psychiatry 64: 349–352. Nashef, L., N. Hindocha, and A. Makoff. 2007. Risk factors in sudden death in epilepsy (SUDEP): The quest for mechanisms. Epilepsia 48: 859–871. Nashef, L., and L. Sahlholdt. 2010. Bereavement and sudden unexpected death in epilepsy. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, chap. 58. Boca Raton, FL: CRC Press. Nilsson, L., B. Y. Farahmand, P. G. Persson, I. Thiblin, and T. Tomson. 1999. Risk factors for sudden unexpected death in epilepsy: A case-control study. Lancet 353: 888–893. Oguni, H., K. Hayashi, Y. Awaya, Y. Fukuyama, and M. Osawa. 2001. Severe myoclonic epilepsy in infants—A review based on the Tokyo Women’s Medical University series of 84 cases. Brain Dev 23: 736–748. Opdal, S. H., and T. O. Rognum. 2004. The sudden infant death syndrome gene: Does it exist? Pediatrics 114: e506–e512. Paterson, D. S. 2010. Medullary serotonergic abnormalities in sudden infant death syndrome: Implications in sudden unexpected death in epilepsy. In Sudden Death in Epilepsy: Forensic and Clinical Issues. Chapter 5, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. New York, NY: Taylor & Francis. Paterson, D. S., F. L. Trachtenberg, E. G. Thompson et al. 2006. Multiple serotonergic brainstem abnormalities in sudden infant death syndrome. JAMA 296: 2124–2132. Pfeufer, A., S. Jalilzadeh, S. Perz et al. 2005. Common variants in myocardial ion channel genes modify the QT interval in the general population: Results from the KORA study. Circ Res 96: 693–701. Plant, L. D., P. N. Bowers, Q. Liu et al. 2006. A common cardiac sodium channel variant associated with sudden infant death in African Americans, SCN5A S1103Y. J Clin Invest 116: 430–435. Priori, S. G., and C. Napolitano. 1998. Impact of molecular biology on diagnosis and management of arrhythmogenic disorders. Rev Port Cardiol 17: 323–327.

282 Sudden Death in Epilepsy: Forensic and Clinical Issues Puranik, R., C. K. Chow, J. A. Duflou, M. J. Kilborn, and M. A. McGuire. 2005. Sudden death in the young. Heart Rhythm 2: 1277–1282. Quilichini, P. P., C. Chiron, Y. Ben Ari, and H. Gozlan. 2006. Stiripentol, a putative antiepileptic drug, enhances the duration of opening of GABA-A receptor channels. Epilepsia 47 (4): 704–716. Rodriguez-Calvo, M. S., M. Brion, C. Allegue, L. Concheiro, and A. Carracedo. 2008. Molecular genetics of sudden cardiac death. Forensic Sci Int 182: 1–12. Rogart, R. B., L. L. Cribbs, L. K. Muglia, D. D. Kephart, and M. W. Kaiser. 1989. Molecular cloning of a putative tetrodotoxin-resistant rat heart Na+ channel isoform. Proc Natl Acad Sci U S A 86: 8170–8174. Rugg-Gunn, F. J., and L. Nashef. 2009. Sudden unexpected death in epilepsy, 211–240. In Blue Book of Neurology: The Epilepsies, ed. S. D. Shorvon and T. Pedley. Philadelphia, PA: Saunders. Saenen, J. B., and C. J. Vrints. 2008. Molecular aspects of the congenital and acquired long QT syndrome: Clinical implications. J Mol Cell Cardiol 44: 633–646. Scheffer, I. E., L. A. Harkin, B. E. Grinton et al. 2007. Temporal lobe epilepsy and GEFS+ phenotypes associated with SCN1B mutations. Brain 130: 100–109. Schulze-Bahr, E. 2006. Mutation detection beyond: an example from congenital long QT syndrome. Heart Rhythm 3: 822–823. Schwartz, P. J., M. Stramba-Badiale, A. Segantini et al. 1998. Prolongation of the QT interval and the sudden infant death syndrome. N Engl J Med 338: 1709–1714. Sharp, A. J., H. C. Mefford, K. Li et al. 2008. A recurrent 15q13.3 microdeletion syndrome associated with mental retardation and seizures. Nat Genet 40: 322–328. Singh, N. A., C. Charlier, D. Stauffer et al. 1998. A novel potassium channel gene, KCNQ2, is mutated in an inherited epilepsy of newborns. Nat Genet 18: 25–29. Singh, R., I. E. Scheffer, K. Crossland, and S. F. Berkovic. 1999. Generalized epilepsy with febrile seizures plus: A common childhood-onset genetic epilepsy syndrome. Ann Neurol 45: 75–81. Sisodiya, S., J. H. Cross, I. Blumcke et al. 2007. Genetics of epilepsy: Epilepsy Research Foundation Workshop Report. Epileptic Disord 9: 194–236. Skinner, J. R., B. Chong, M. Fawkner, D. R. Webster, and M. Hegde. 2004. Use of the newborn screening card to deἀne cause of death in a 12-year-old diagnosed with epilepsy. J Paediatr Child Health 40: 651–653. Splawski, I., K. W. Timothy, M. Tateyama et al. 2002. Variant of SCN5A sodium channel implicated in risk of cardiac arrhythmia. Science 297: 1333–1336. Steinlein, O. K., J. C. Mulley, P. Propping et al. 1995. A missense mutation in the neuronal nicotinic acetylcholine receptor alpha 4 subunit is associated with autosomal dominant nocturnal frontal lobe epilepsy. Nat Genet 11: 201–203. Steinlein, O. K. 2008. Genetics and epilepsy. Dialogues Clin Neurosci 10: 29–38. Stöllberger, C., and J. Finsterer. 2004. Cardiorespiratory ἀndings in sudden unexplained/unexpected death in epilepsy (SUDEP). Epilepsy Res 59 (1): 51–60. Striano, P., L. Bordo, M. L. Lispi et al. 2006. A novel SCN2A mutation in family with benign familial infantile seizures. Epilepsia 47: 218–220. Sudden Death in Chromosome 15q11-q13 Duplication Syndrome. 2009. http://www.idic15.org/ dr_physicianadvisory.php (accessed April 3, 2009). Tan, H. L., N. Hofman, I. M. van Langen, A. C. van der Wal, and A. A. Wilde. 2005. Sudden unexplained death: Heritability and diagnostic yield of cardiological and genetic examination in surviving relatives. Circulation 112: 207–213. Tellez, J. O., H. Dobrzynski, I. D. Greener et al. 2006. Differential expression of ion channel transcripts in atrial muscle and sinoatrial node in rabbit. Circ Res 99: 1384–1393. Tester, D. J., and M. J. Ackerman. 2005. Sudden infant death syndrome: How signiἀcant are the cardiac channelopathies? Cardiovasc Res 67: 388–396. Timmings, P. L. 1998. Sudden unexpected death in epilepsy: Is carbamazepine implicated? Seizure 7: 289.

Genetics of Sudden Death in Epilepsy

283

Tupal, S., and C. L. Faingold. 2006. Evidence supporting a role of serotonin in modulation of sudden death induced by seizures in DBA/2 mice. Epilepsia 47: 21–26. Van Norstrand, D. W., and M. J. Ackerman. 2009. Sudden infant death syndrome: Do ion channels play a role? Heart Rhythm 6: 272–278. Wallace, R. H., D. W. Wang, R. Singh et al. 1998. Febrile seizures and generalized epilepsy associated with a mutation in the Na+-channel beta1 subunit gene SCN1B. Nat Genet 19: 366–370. Watanabe, H., T. T. Koopmann, S. Le Scouarnec et al. 2008. Sodium channel beta1 subunit mutations associated with Brugada syndrome and cardiac conduction disease in humans. J Clin Invest 118: 2260–2268. Weese-Mayer, D. E., E. M. Berry-Kravis, L. Zhou et al. 2004. Sudden infant death syndrome: Casecontrol frequency differences at genes pertinent to early autonomic nervous system embryologic development. Pediatr Res 56: 391–395. Wick, R., and R. W. Byard. 2006. Mechanisms of unexpected and/or sudden death in Lafora disease. Forensic Sci Int 163: 144–147.

Cardiac Channelopathies and Sudden Death Benito Herreros

19

Contents 19.1 Introduction 19.1.1 Sudden Cardiac Death in Subjects with a Structurally Normal Heart 19.1.2 Ion Channels and Channelopathies 19.1.3 The Cardiac Action Potential 19.1.4 Inherited Primary Arrhythmia Syndromes 19.2 Long QT Syndrome 19.2.1 Clinical Manifestations 19.2.2 Genetic Background and Pathophysiology 19.2.3 Risk Stratiἀcation and Management 19.3 Brugada Syndrome 19.3.1 Clinical Manifestations 19.3.2 Genetic Background and Pathophysiology 19.3.3 Risk Stratiἀcation and Management 19.4 Catecholaminergic Polymorphic Ventricular Tachycardia 19.4.1 Clinical Manifestations 19.4.2 Genetic Background and Pathophysiology 19.4.3 Management 19.5 Short QT Syndrome 19.5.1 Clinical Manifestations 19.5.2 Genetic Background and Pathophysiology 19.5.3 Management 19.6 Considerations about Cardiac Channelopathies, Epilepsy, and SUDEP 19.6.1 Brugada ECG and Epilepsy 19.6.2 LQT and Epilepsy 19.6.3 The Hypothesis of a Common Channelopathy 19.6.4 SUDEP and Channelopathies References

285 286 286 287 287 289 289 290 291 291 292 293 294 295 295 295 296 296 296 297 297 297 298 298 299 299 299

19.1â•…Introduction Cardiac channelopathies are genetically determined diseases that may cause sudden arrhythmic death in young subjects with a structurally normal heart. Abnormal cellular electrophysiology related to dysfunctional ion channels is the underlying substrate. Some types of idiopathic epilepsy are also caused by ion channel dysfunction in the neuronal cell membrane. This shared pathophysiologic mechanism lets us hypothesize about an arrhythmic cause for some cases of sudden death in epilepsy. In this chapter, cardiac 285

286 Sudden Death in Epilepsy: Forensic and Clinical Issues

channelopathies related to sudden death are reviewed and data supporting a potential relationship with epilepsy are exposed in the last section. 19.1.1â•… Sudden Cardiac Death in Subjects with a Structurally Normal Heart Sudden cardiac death (SCD) is deἀned as a natural death from cardiac causes, heralded by abrupt loss of consciousness within 1 hour of the onset of an acute change in cardiovascular status. Preexisting heart disease may or may not be present, but the time and mode of death are unexpected. It is likely that ventricular ἀbrillation (VF), or ventricular tachycardia (VT) deteriorating to VF, is the initiating event in most cardiac arrests (Myerburg and Castellanos 2008). These arrhythmias most often occur in the presence of a cardiac structural abnormality that is responsible for disturbances in impulse initiation and conduction. Ischemic heart disease is the most common cause, followed by other structural heart diseases such as hypertrophic, dilated, or arrhythmogenic right ventricular cardiomyopathy, but malignant arrhythmias leading to SCD can also happen in some patients with a structurally normal heart. In several postmortem series of cardiac arrest victims, no structural abnormality was found in 5% to 8% of cases (Priori et al. 2003). This occurrence used to be designated as “idiopathic VF.” The etiology of SCD in these cases may be a primary electrical disorder of a genetic basis (and thus can be inherited) that predisposes to rhythm alteration. In fact, SCD without structural heart disease is proportionally more common at younger ages, in contrast with the adult age, in which the main cause of SCD, by far, is coronary artery disease (Myerburg 2001) (Table 19.1). During the past 15 years, with the help of molecular biology, the substrate of idiopathic VF has been better deἀned and genetically determined abnormalities of proteins that control the electrical activity of the heart have been demonstrated to cause cardiac arrest in the structurally intact heart (Priori et al. 2008). 19.1.2â•…Ion Channels and Channelopathies Each normal heartbeat is initiated by a pulse of electrical excitation that begins in a group of specialized pacemaker cells and subsequently spreads throughout the heart. This electrical impulse is made possible by the electrochemical gradient that exists across the cell membrane of the cardiomyocyte. At rest, the electrochemical potential inside the cell is negative with respect to the outside due to energy-consuming processes, such as the Na+/K+ ATPase, which maintain the ionic gradients. The resting cell membrane is relatively permeable to K+, but much less so to Na+ and Ca2+. During electrical excitation, the membrane becomes permeable to Na+ ions and the electrochemical gradient reverses (depolarization). Next, Ca2+ ions move into the cell to activate the contractile process. Finally, the negative

Table 19.1â•… Etiology of Sudden Cardiac Death Coronary artery disease Cardiomyopathies Valvular, inflammatory, and inἀltrative cardiac diseases Normal hearts/idiopathic VF

80% 10%–15% ±5% ?%

Source: Myerburg, R.J., J Cardiovasc Electrophysiol, 12, 369–381, 2001. With permission.

Cardiac Channelopathies and Sudden Death

287

membrane potential is restored, mainly due to the efflux of K+, and thus the wave of depolarization is self-limiting (Marbán 2002). All these changes in permeability are accomplished by the opening and closing of ion channels that are speciἀc for the individual ions. Ion channels are transmembrane proteins that allow the passage of ions from the interior of the cell to outside and vice versa (Figure 19.1). The movements of ions are passive down their respective concentration gradients. Some channels are tissue-speciἀc (e.g., Na+ channels have several isoforms with different expression in the heart, skeletal muscle, central, or peripheral nervous system), whereas others are widely distributed throughout the body. Ion channels are basic to the processes of electrical signaling and excitation essential to the functioning of the heart. Furthermore, for the normal function of any given ion channel, not only its protein subunits (alpha, beta), but also multiple other gene products with various functions (e.g., phosphorylation, assembly, posttranslational modiἀcation, anchoring units) are necessary. Mutation in any of these genes can cause ion channel dysfunction, or channelopathy, which may provoke cardiac arrhythmias and SCD as the most severe clinical expression (Roden et al. 2002). This chapter reviews the most common inherited arrhythmogenic diseases that may cause sudden death in subjects with a structurally normal heart. As most of them are due to dysfunction of ion channels participating in the cardiomyocyte excitation, it is important to review the cardiac action potential and its involved ionic currents. These are summarized in Figure 19.1. 19.1.3â•… The Cardiac Action Potential The cardiac action potential is a graphic representation of the stereotypical voltage changes against time, which follow one another when an appropriate voltage stimulus reaches the cardiomyocyte (Figure 19.1). Thus, the action potential is a reflection of the electrical activity of a single cardiac cell. A sharp depolarizing upstroke (phase 0) is the result of rapid inflow of Na+ ions through voltage-dependent Na+ channels. Phase 1 reflects the activation of a transient outward current (Ito), mostly carried by K+, which produces a notch or early repolarization soon after the initial depolarizing upstroke. During the plateau (phase 2), the net influx of Ca2+ is balanced by the efflux of K+ through several K+ channels (IKr, IKs, IK1, Ito). Phase 3 is the downslope after the plateau, representing the late depolarization, due to the chemical forces, that favors the efflux of K+ predominating over the electrostatic forces that would favor its influx through the same channels. Phase 4 represents the resting potential, during which there tends to be a net diffusion (efflux) of K+ in the direction of its concentration gradient through IK1 channels (Berne and Levy 1992). 19.1.4â•…Inherited Primary Arrhythmia Syndromes A number of genes are associated with inherited arrhythmia syndromes that predispose to sudden death in individuals with a structurally normal heart (Table 19.2). Studies in large cohorts suggest that cardiac channelopathies could be responsible for 35% of sudden deaths in the young and in 9% of cases of sudden infant death syndrome (SIDS) (Schwartz and Crotti 2007). Recently, some authors have proposed a classiἀcation of inherited primary arrhythmia syndromes according to their pathophysiologic basis; that is, into Na+ channelopathies, K+ channelopathies, and so forth (Lehnart et al. 2007; Wilde 2008). This chapter uses the traditional clinical nomenclature to review the four main inherited arrhythmia

288 Sudden Death in Epilepsy: Forensic and Clinical Issues

(a)

Na+

K+ (b)

K+

Ca2+

Na+

Ca2+

Probable gene SCN5A

INa ICa, L

CACNA1c

INa/Ca

NCX1

0

IK1

KCNJ2

It01

KCND

It02

?

IKr

KCNH2/KCNE2

IKs

KCNQ1/KCNE1

IKp

KCNK?

Figure 19.1╇ Ion channels underlie cardiac excitability. (a) The key ion channels (and an elec-

trogenic transporter) in cardiac cells. K+ channels mediate K+ efflux from the cell; Na+ channels and Ca 2+ channels mediate Na+ and Ca 2+ influx, respectively. The Na+/Ca 2+ exchanger transports three Na+ ions for each Ca 2+ ion across the surface membrane. (b) Ionic currents and genes underlying the cardiac action potential. Top, depolarizing currents as functions of time, and their corresponding genes; center, a ventricular action potential; bottom, repolarizing currents and their corresponding genes. (From Marbán, E., Nature, 415, 213–218, 2002. With permission.)

Cardiac Channelopathies and Sudden Death

289

Table 19.2â•… Genes Involved in Arrhythmogenic Cardiac Channelopathies Related to SCD Syndrome

Subtype

Long QT syndrome

LQT1 LQT2 LQT3 LQT4

KCNQ1 KCNH2 SCN5A ANK2

IKs K+ channel α subunit IKr K+ channel α subunit Na+ channel α subunit Anchoring protein ankyrin B

LQT5 LQT6 LQT7 LQT8 LQT9 LQT10 LQT11

KCNE1 KCNE2 KCNJ2 CACNA1c CAV3 SCN4B AKAP9

Brugada syndrome

BrS1 BrS2

SCN5A GPD1L

Catecholaminergic VT

CPVT1 CPVT2 SQT1 SQT2 SQT3

RyR2 CASQ2 KCNH2 KCNQ1 KCNJ2

IKs K+ channel β subunit IKr K+ channel β subunit IK1 K+ channel α subunit ICa,L Ca2+ channel α subunit Caveolin 3 Na+ channel β4 subunit Yotiao (A-kinase anchoring protein) Na+ channel α subunit Glycerol-3-P dehydrogenase 1 like protein Cardiac ryanodine receptor Cardiac calsequestrin IKr K+ channel α subunit IKs K+ channel α subunit IK1 K+ channel α subunit

Short QT syndrome

Gene

Protein

Effect IKs loss of function IKr loss of function INa gain of function Reduction of several ionic currents IKs loss of function IKr loss of function IK1 loss of function ICa,L gain of function INa gain of function INa gain of function IKs loss of function INa loss of function INa loss of function Citoplasmic Ca2+ overload Citoplasmic Ca2+ overload IKr gain of function IKs gain of function IK1 gain of function

syndromes which, to the moment, have been shown to be responsible for sudden death: long QT syndrome (LQTS), Brugada syndrome (BrS), short QT syndrome (SQTS), and catecholaminergic polymorphic ventricular tachycardia (CPVT). There are other inherited arrhythmic syndromes not necessarily linked to SCD that are responsible for atrial arrhythmias, sick sinus syndrome, and progressive cardiac conduction disease (PCCD).

19.2â•… Long QT Syndrome The congenital form of the LQTS is characterized by prolongation of the QT interval on the ECG and susceptibility to malignant ventricular arrhythmias. Two major forms have been described: one inherited as an autosomal dominant trait (Romano–Ward syndrome), and the other (more rare) transmitted as an autosomal recessive trait (Jervell and Lange– Nielsen syndrome) with associated neurosensorial deafness in affected individuals. 19.2.1â•… Clinical Manifestations The characteristic ECG in LQTS shows a prolonged QT interval, often with T-wave morphological abnormalities, and occasionally polymorphic VT of the torsades de pointes type. Clinical manifestations are recurrent syncope or fainting and sudden death, often precipitated by stress (fear, loud noise, sudden awakening), but they can also occur at rest (Schwartz et al. 2001). Typical onset of symptoms occurs in the ἀrst two decades of life. In the neonatal period, LQTS can be responsible for a certain number of SIDS (Arnestad et al.

290 Sudden Death in Epilepsy: Forensic and Clinical Issues

2007). Since the clinical presentation varies greatly, ranging from asymptomatic mutation carriers with normal QT interval to typical cases with marked QT prolongation and recurrent syncope, diagnostic criteria have been developed with a scoring system that allows classiἀcation of patients into categories of low, intermediate, or high probability of having LQTS (Schwartz et al. 1993). With the advent of the DNA diagnosis, these scoring systems have been shown to have less sensitivity than measurement of QTc alone (Napolitano et al. 2003; Hofman et al. 2007) for the detection of mutation carriers. Furthermore, an important proportion of mutation carriers have a normal QT interval (Priori et al. 2003), so molecular diagnosis has become part of the routine clinical management of LQTS. 19.2.2â•… Genetic Background and Pathophysiology At the time of writing, 11 genes have been proven to be or are thought to be associated with LQTS (Lehnart et al. 2007) (Table 19.2). However, LQT1, LQT2, and LQT3 variants account for more than 90% of all genotyped LQTS patients, whereas the remaining genes are responsible for a minority of cases (Priori et al. 2008). Most of these genes encode cardiac ion channels subunits. If not, they modulate ionic currents. In patients with LQT1, there is loss of function of the IKs potassium channel due to a mutation in the gene KCNQ1, which encodes the alpha subunit of IKs channel. The result is a reduced K+ outward current in phases 2 and 3 of the action potential, leading to an abnormal repolarization and reduced rate-dependent shortening of the action potential (Roden et al. 2002). Mutations in KCNQ1 resulting in IKs loss of function accounts for 40% to 50% of all genotyped LQT mutations (Splawski et al. 2000; Tester et al. 2005). The same channel IKs exhibits loss of function in patients with LQT5, but is due to mutation in gene KCNE1, which encodes its beta subunit. This variant is much less frequent than LQT1, but it is noteworthy that both genes KCNQ1 and KCNE1 are also expressed in the inner ear, where K+ channel function contributes to production of endolymph. That is the reason why mutations in both genes may cause the Jervell and Lange–Nielsen syndrome (LQT and bilateral deafness) in homozygous subjects (Neyroud et al. 1997; Schulze-Bahr et al. 1997). Patients with LQT2 exhibit loss of function of the IKr potassium channel due to a mutation in gene KCNH2, which encodes the alpha subunit of IKr channel. These mutations represent 35% to 45% of the genotyped LQTS mutations. The beta subunit of IKr channel is encoded by KCNE2, whose mutations are the substrate for the rarer LQT6 (Abbot et al. 1999). Whereas K+ channels mutations involved in LQTS produce a loss of function effect, a gain of function mutation in SCN5A encoding the alpha subunit of the cardiac Na+ channel is the substrate for the variant LQT3. As a consequence, persistent inward Na+ current during phase 2 (plateau) prolongs the action potential. LQT3 accounts for 2% to 8% of all LQTS patients (Splawski et al. 2000; Tester et al. 2005). Overall, penetrance of LQTS is 60%, but it is not the same for speciἀc variants. In LQT1 the proportion of mutation carriers without QTc prolongation is as high as 36%, in LQT2 it is 19%, and in LQT3 it is 10% (Priori et al. 2003). There are some speciἀc phenotypic features in the different variants. Patients with LQT1 have increased risk of developing VT during exercise (75%, especially swimming) or emotional stress (15%). IKs current is activated by fast heart rates and by catecholamines, so under these circumstances, LQT1 patients cannot shorten their QT interval appropriately. Accordingly, beta blocker therapy should be most effective in LQT1 patients (Schwartz et al. 2001). The ECG in LQT1 is characterized by long T wave duration (Figure 19.2). In LQT2 patients, cardiac events

Cardiac Channelopathies and Sudden Death LQT3

LQT2

291 LQT1

II aVF

V5

Figure 19.2╇ Electrocardiographic distinctive features in long QT syndrome variants. Left

panel: note the long ST segment with late onset of T wave in an LQT3 patient. Center panel: low amplitude T waves in an LQT2 patient. Right panel: early onset of broad-based T waves in an LQT1 patient. (Modified from Moss, A. J. et al., Circulation, 92 (10), 2929–2934, 1995. With permission.)

are often triggered by emotions, sudden loud noises, and acute arousal (37%). They are also at€risk during sleep or at rest (63%) but not during exercise. Their ECGs show small (low-amplitude), notched, or biphasic T waves (Figure 19.2). Finally, LQT3 patients suffer cardiac events usually at night, sleep, or rest (80%), and occasionally with emotions (15%) or during exercise (5%) (Schwartz et al. 2001). Typical ECG in LQT3 shows a flat, long ST segment with late onset of a narrow, peaked T wave (Figure 19.2). 19.2.3â•…Risk Stratification and Management Highest risk patients are those who have QTc ≥ 500 ms. Other ECG markers of bad prognosis are T wave alternans and torsades de pointes VT. Those who have survived after a cardiac arrest, and those who have recurrent syncope despite beta blockers, usually receive an implantable cardioverter deἀbrillator (ICD) (Zareba et al. 2003). On the other hand, the increasing knowledge of clinical evolution in genotyped patients has demonstrated that the different LQTS variants have distinct prognoses. LQT2 and LQT3 patients have a lower event-free survival than LQT1. Gender has influence in LQT2 (higher risk in females) and LQT3 patients (higher risk in males), but not in LQT1 (Priori et al. 2003). Once a diagnosis of LQTS has been made, lifestyle modiἀcation is recommended Â�(e.€g., avoidance of swimming among LQT1 patients, no use of loud alarm clocks in LQT2, prohibition of drugs with QT prolonging effects). Experts recommend beta blocker therapy in all symptomatic patients and in asymptomatic patients younger than 40 years with a clearly prolonged QTc (in which risk for SCD if untreated is around 13%) (Priori et al. 2003). Implantation of a deἀbrillator is recommended after a resuscitated cardiac arrest and in those patients who experience syncope or VT under beta blocker therapy (Zipes et al. 2006). Also, the recognition of the speciἀc arrhythmic risk of the different LQTS variants will probably have an increasing impact on clinical management.

19.3â•… Brugada Syndrome BrS is characterized by an ST segment elevation in the right precordial ECG leads and a high incidence of SCD in patients with structurally normal hearts. Sudden death is

292 Sudden Death in Epilepsy: Forensic and Clinical Issues

provoked by ventricular arrhythmias (polymorphic VT leading to ventricular ἀbrillation) that can also produce syncope when they are self-limited. Inheritance of BrS occurs via an autosomal dominant mode of transmission. 19.3.1â•… Clinical Manifestations Characteristic Brugada ECG shows a coved ST segment elevation ≥2 mm followed by a negative T wave in more than one right precordial lead (V1–V3, Figure 19.3). This is the only accepted diagnostic pattern (the so-called type 1), although two more ECG types have been described: type 2 has a saddleback ST elevation with a J point ≥ 2 mm, a trough ≥ 1 mm, and a positive/biphasic T wave; and in type 3, ST elevation is 10% from baseline (cardioinhibitory response), or both (mixed response). Most patients with vasovagal syncope have a mixed response. In general, tilt table testing is reserved for patients with recurrent syncope or a single high-risk syncope episode when there is no evidence of structural heart disease or when other causes of syncope have been excluded. In patients with a single episode of uncomplicated syncope where the clinical picture is typical for neurally mediated reflex syncope, tilt table testing is unnecessary. Furthermore, tilt table testing is not useful in establishing the diagnosis of neurally mediated reflex syncope with a speciἀc trigger (e.g., micturation) (Brignole et al. 2001; Schnipper and Kapoor 2001; Grubb 2005).

Figure 21.2╇ Retrieved episode data of syncopal event in a patient with unexplained syncope in

whom an implantable loop recorder was inserted for a prolonged monitoring of cardiac rhythm. The tracing shows an initial three beats of sinus rhythm followed by a premature ventricular complex and a sinus beat before the onset of a rapid ventricular tachycardia at 176 beats/min (mean cycle length of 340 ms).

320 Sudden Death in Epilepsy: Forensic and Clinical Issues

21.3.6â•… Electrophysiologic Testing The electrophysiology study is an invasive study in which electrode-tipped catheters are placed in the heart in speciἀc locations and stimulation protocols are performed to assess the cardiac electrical system. In general, electrophysiology study is indicated when syncope is associated with structural heart disease. The role of electrophysiology study in recurrent unexplained syncope with negative tilt table test and no structural heart disease is not ἀrmly established. In patients with unexplained syncope and structural heart disease (e.g., prior myocardial infarction) induction of ventricular tachycardia indicates a poor prognosis (up to 30% mortality in 3 years) (Gouello et al. 1992). Ventricular tachycardia is the most common abnormality revealed by electrophysiology study. Since recent guidelines recommend implantable cardioverter deἀbrillator implantation for primary prevention of ventricular arrhythmia in patients with severe left ventricular systolic dysfunction (ejection fraction 100 ms), and/or (2) block below the His conduction system (Gouello et al. 1992). 21.3.7â•… Miscellaneous Tests Transient ischemic attacks from carotid atherosclerosis do not cause loss of consciousness. Therefore, carotid ultrasonography is not indicated in the evaluation of syncope. Blood tests rarely assist in the diagnosis of syncope, unless a metabolic etiology is strongly suspected (e.g., hypoglycemia). Brain imaging with computerized tomography (CT) and magnetic resonance imaging (MRI) are usually unnecessary and only yields a diagnosis when there is a focal neurologic deἀcit or a witnessed seizure. If there are no signs and symptoms of seizure, electroencephalography is not useful in the diagnosis of syncope (Britton 2004). Simultaneous electroencephalography and electrocardiography (especially with video monitoring) may help to diagnose frequent episodes that cannot be distinguished as syncope or seizure (Britton 2004). Myocardial ischemia is an unlikely cause of syncope, especially in the absence of angina or exertional symptoms. Therefore, stress testing and cardiac catheterization should be reserved for patients with syncope associated with exertion when the suspicion for coronary artery disease is high.

21.4â•… Syncope and Epilepsy There are reports of patients with recurrent syncope who develop cardiac asystole, transient AV block, and severe sinus bradycardia (heart rate 55% of time in seconds) while the cat was under the influence of PTZ, this phenomenon was not found in any cat during the control period. Epileptogenic activity induced by PTZ is an experimental model of primary generalized epilepsy. The method of action of this drug, however, is uncertain (Stone 1972). There are three proposed methods of action of PTZ, which would allow the LSP to express itself: (1) spatial and temporal summation of neuronal discharges in a subcortical center producing a stimulus strong enough to overcome the cortical and ganglionic threshold (Hahn 1960); (2) increased synaptic recruitment, resulting in the ampliἀcation of subcortical stimuli along their path so that, upon reaching the cortex and sympathetic ganglion, they are capable of causing these neurons to discharge; and (3) increased irritability of all neurons so that subcortical impulses could stimulate cortical and ganglionic neurons (Hahn 1960). In each case, PTZ effectively creates a hyperirritable state of epileptogenic electrical activity present in the central and autonomic nervous systems. Although phenobarbital can act to minimize this irritability, the effect of this pharmacological agent in this study eventually was overcome by increased epileptogenic activity. The 2.8-s repeated ECoG interval, a type of latency period between the end of one spike and the beginning of another associated spike, appeared to convey a stabilizing effect on the presence of LSP. When this repeated ECoG interval was present, LSP was present a signiἀcantly greater percentage of time than when LSP was absent. When the repeated ECoG interval was not present, LSP was much less common and the distinct ECoG spikes often degenerated into prolonged ictal activity. Our analysis minimized the degree of association between the repeated ECoG interval and the presence of LSP. Our deἀnition of a repeated ECoG interval excludes periods when two or more additional ECoG spikes are contained within the 2.8-s interval, even if these additional ECoG spikes are time locked to sympathetic spikes. A more liberal deἀnition would result in a more frequent association between LSP and repeated ECoG interval. The mean proportion of each minute during which LSP was present was directly proportional to the time interval elapsed following the administration of PTZ. The direct relationship reflects the fact that the episodes of epileptogenic activity, and particularly prolonged ictal activity, are most frequently observed shortly after PTZ is administered. Similarly, the incidence of precipitous blood pressure change was signiἀcantly greater (p€10-s duration. For each cat, the mean numbers of episodes for the two durations were calculated for the control and for each dose of PTZ and timolol. Means were calculated by combining the ECoG data from all cats. To determine if there were any statistically signiἀcant differences among the means, data were analyzed by a Friedman rank ANOVA followed by a Bonferroni-corrected Wilcoxon post hoc tests. The following time points were analyzed: control; PTZ 10 and 20 mg; and timolol 10 μg, 100 μg, 500 μg, and 1 mg/kg. 34.2.2â•… Seizure Studies in Anesthetized Pigs Domestic swine weighing 13–20 kg were anesthetized with ketamine (20 mg/kg i.v.) and alpha-chloralose (80 mg/kg i.v.). Animals were prepared as described by Spivey et al. (1987). After a 10-min equilibration period, seizure activity was induced by PTZ (100 mg/ kg i.v.). Sixty seconds after the onset of epileptogenic activity, the animals were treated with no drug (control group) or propranolol 2.5 mg/kg i.v. Seizure activity was monitored for 20€min. Plasma levels of propranolol were determined by drawing blood samples (8 mL) at 1, 2, 5, 10, 15, and 20 min after propranolol. The samples were centrifuged at 1000 g for min and the plasma (3 mL) was stored frozen at −20°C. The concentrations of propranolol in plasma were determined by a modiἀcation of the method of Albani et al. (1982). The procedure involves reverse-phase high-pressure liquid chromatographic resolution with fluorometric detection of propranolol and an internal standard, Carvedilol (1-(4-carbazolyloxy)-3-[2(2-methoxyphenoxy)ethylamino]-2-propranol).

34.3â•…Results 34.3.1â•…Data Obtained in Anesthetized Cats A one-factor repeated-measures ANOVA revealed that the increase in the control blood pressure of 103 ± 13 to 149 ± 13 mm Hg minutes after intracerebroventricular PTZ was signiἀcant, whereas the mean heart rate increase from 159 ± 17 to 162 ± 10 bpm 3 min after

554 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 34.1â•… Sequence in Which Incidence of the Epileptiform Activity and an Increase in Mean Arterial Blood Pressure and Heart Rate Occurred after Intracerebroventricular PTZa Precededb

Same time

Proceededc

No change

0 0 1 0

2 1 1 1

0 1 0 0

BP 5 HRd 5 BPe 11 HRe 12 d

a b

c

d e

Incidence indicates the number of cats of the 13 studied. Preceded indicates that the increase in blood pressure or heart rate occurred before the onset of epileptiform activity elicited by PTZ. Proceeded indicates that the increase in blood pressure or heart rate occurred after the onset of epileptiform activity elicited by PTZ. Seven cats receiving only PTZ; the ECoG activity was not recorded in the eighth cat. Thirteen cats receiving PTZ and timolol.

PTZ and to 175 ± 14 bpm 15 min after PTZ was not signiἀcant. The PTZ-induced increase in blood pressure and heart rate preceded the initiation of epileptiform activity in most of these cats (Table 34.1). This trend was also observed in the 13 cats receiving PTZ and then timolol (Table 34.2). During the occurrence of PTZ-induced epileptiform activity, premature ventricular contractions of 3–8 min were observed in three of the eight cats. Data obtained from one cat are illustrated in Figure 34.1. In panel a (control), the cerebral activity present was that associated with the induction of anesthesia. With the intracerebroventricular administration of 10 mg PTZ (panel b), there was induction of epileptiform activity, a slight increase in blood pressure, and a slight decrease in heart rate. When timolol (10 μg/kg i.c.v.) was administered (panel c), the epileptiform activity was diminished, although the blood pressure and heart rate values were still elevated above control. The intracerebroventricular administration of PTZ increased both the mean arterial blood pressure and the heart rate in six cats receiving only PTZ and timolol; no other pharmacological agents were administered (data not shown). When timolol was given to the six Table 34.2â•… Incidence of the Sequence in Which Suppression of Intracerebroventricular PTZ-Induced Epileptiform Activity and a Decrease in Mean Arterial Blood Pressure and Heart Rate Occurred after Timolola Partial Suppression Compared to Control

Total Suppression Compared to Control

Total Suppression Compared to Mean of Previous 5-min Interval

No No Precededb Same Proceededc Preceded Same Proceeded Change Preceded Same Proceeded Change BP 1 HR 0 a b

c

d

0 0

12 13

11 12

0 0

1 0

1 1

7 2

0 1

4 6d

2 5d

Incidence indicates the number of cats of the 13 studied. Preceded indicates that the decrease in blood pressure or heart rate occurred before the partial or total suppression of epileptiform activity. Proceeded indicates that the decrease in blood pressure or heart rate occurred after the partial or total suppression of epileptiform activity. Includes same cat, hippocampus no change; both left and right cortex proceeded total suppression of epileptogenic activity.

Electroencephalogram (µV)

Antiepileptic Activity of Beta-Blocking Agents

Mean arterial blood 200 pressure 1000 (mm Hg) Electrocardiogram (Lead II)

(a)

(b)

555

(c)

L. motor cortex R. motor cortex R. hippocampus 108

123

143

122

113

130

1s Control

1s PTZ 10 mg i.c.v.

Timolol 10 µg/kg i.c.v.

Figure 34.1╇ Effect of timolol on the ECoG and cardiovascular parameters associated with

intracerebroventricular PTZ-induced epileptiform activity in one cat. Represented in all three panels are ECoGs from the left and right motor cortex and right hippocampus, as well as mean arterial blood pressure, and electrocardiogram from top to bottom, respectively. (Reproduced from Lathers, C. M., et al., Epilepsy Res, 4, 42–54, 1989b. With permission.)

cats after the administration of PTZ, the mean arterial blood pressure and heart rate values began to decrease. Similar data were obtained when PTZ and timolol were administered to cats that had previously received other agents (n = 7 cats). In the cats receiving other pharmacological agents before the administration of PTZ, the blood pressure, heart rate, and ECoG activity had returned to baseline values before the administration to PTZ. No signiἀcant effect of previous drugs or interaction between previous drug and time was found. The lack of a signiἀcant effect indicated that the two groups of animals responded the same ways to the drug injections over time and permitted combination of the two groups for purposes of analysis (n = 13 cats; Figure 34.2). Thus, the administration of PTZ increased both mean arterial blood pressure and heart rate. When timolol was given intracerebroventricularly, the mean arterial blood pressure and heart rate values began to decrease. In the two-way ANOVA, both main effects and the interaction were signiἀcant for the heart rate. A comparison of the two experimental groups (PTZ i.c.v. and PTZ–timolol, n = 13 cats) at each experimental time showed that the timolol group was lower in the control period, sharply rising to a nonsigniἀcant difference after the second dose of PTZ and after the ἀrst and second doses of timolol. At higher doses of timolol, the timolol group was consistently signiἀcantly lower in heart rate than the group receiving only PTZ. Experimental times within groups, done as separate one-way repeated-measures ANOVAs, due to the different variances, showed no signiἀcant difference across time in the heart rate for the cats receiving no timolol and a sharp rise followed by a fall (Tukey HSD) in the cats receiving PTZ and increasing doses of timolol. The two-way ANOVA revealed that the group main effects on blood pressure were not signiἀcant. The time effect and the interactions were signiἀcant. When the experimental groups were compared at each experimental time, no differences were found until the two highest doses of timolol. The data in the control group were steady at the time equivalent to that when timolol would have been administered to the animals in the experimental group, whereas the timolol group exhibited a rapid fall in blood pressure. Consequently,

556 Sudden Death in Epilepsy: Forensic and Clinical Issues 180

11 11

160

11

140

Mean arterial blood pressure (mean ± SE mm Hg)

10

11

N = 13 cats

11 12 12 12

120

12 12

100 12

80

11 12 12 12 12

60

12

12 12

40

12

8

8

20 –10

6

2

Control

14

18

22

26

30

34

38

42

20 mg PTZ, i.c.v.

100 µg/kg 1 mg/kg 10 mg/kg Timolol, i.c.v. Timolol, i.c.v. Timolol, i.v. 10 µg/kg 500 µg/kg 5 mg/kg 20 mg/kg Timolol, i.c.v. Timolol, i.c.v. Timolol, i.v. Timolol, i.v.

10 mg PTZ, i.c.v.

(a)

10

Time (min)

180

Heart rate (beats/min) mean ± SE

N = 13 cats 160

11

11

11 1110

11

140

12

12 12

120

12 11

100

12 12 12 11 12 12 11 1212 12 8 8

80 –10 Control

2

6

10

20 mg PTZ, i.c.v.

10 mg PTZ, i.c.v.

14

18

22

26

30

34

38

42

100 µg/kg 1 mg/kg 10 mg/kg Timolol, i.c.v. Timolol, i.c.v. Timolol, i.v. 10 µg/kg 500 µg/kg 5 mg/kg 20 mg/kg Timolol, i.c.v. Timolol, i.c.v. Timolol, i.v. Timolol, i.v.

Time (min)

(b)

Figure 34.2╇ The effect of timolol on mean arterial blood pressure and heart rate changes

induced by PTZ administered intracerebroventricularly in 13 cats. In the upper graph, the mean arterial blood pressure is graphed as a function of time. The lower graph depicts the mean heart rate. The arrows along the abscissa indicate the administration of PTZ intracerebroventricularly and timolol intracerebroventricularly or intravenously. (Reproduced from Lathers, C.€M., et€al., Epilepsy Res, 4, 42–54, 1989b. With permission.)

Antiepileptic Activity of Beta-Blocking Agents

557

the latter group was signiἀcantly lower. Experimental times within each group showed no signiἀcant effect of time course on blood pressure in those cats receiving only PTZ, whereas in the cats receiving both PTZ and timolol, the PTZ increased the blood pressure and timolol reversed the effect of PTZ. In the two cats receiving the same doses of timolol at time intervals other than 5 min (i.e., 10 or 15 min), timolol suppressed the epileptogenic discharges and decreased the blood pressure and heart rate. In the one cat in which all doses of timolol were given every 15 min but no epileptogenic activity occurred since PTZ was not given, both the heart rate and blood pressure values were decreased, but the magnitude of the decrease was much less than that observed in the cats given PTZ. Although PTZ induced both interictal and ictal epileptiform discharges in all cats, most epileptiform activity exhibited durations of ≤10 s, that is, interictal and brief ictal activity (Table 34.3 and Figure 34.3). The dashed curve in Figure 34.3 depicts the epileptiform activity with a duration of ≤10 s obtained in the left cortex of the cats receiving only PTZ and no timolol. The mean number of episodes of epileptiform activity remained approximately the same for the 45-min period, the time equivalent to the entire experimental duration in the cats receiving both PTZ and increasing doses of timolol. The mean number of episodes of epileptiform activity lasting 10 s (seconds/minute; mean ± SE) Experimental Groups

Left Cortex

Right Cortex

Right Hippocampus

Control PTZ, 10 mg i.c.v. PTZ, 20 mg i.c.v. Timolol, 10 μg/kg i.c.v. Timolol, 100 μg/kg i.c.v. Timolol, 500 μg/kg i.c.v. Timolol, 1 mg/kg i.c.v. Timolol, 5 mg/kg i.v. Timolol, 10 mg/kg i.v.

0 0 0.31 ± 0.31b 0.50 ± 0.50b 0 0 0 0 0

0 0 0.54 ± 0.54b 0.45 ± 0.45b 0 0 0 0 0

0 1.21 ± 1.21b 1.00 ± 0.65 5.51 ± 4.63 4.22 ± 4.22b 0 4.62 ± 4.62b 0.89 ± 0.89b 0

a

Source: Lathers, C. M., et al., Epilepsy Res, 4, 42–54, 1989b. With permission. Deἀned as the mean ± SE for the 13 cats in which a mean 10-min control period was obtained. b Mean ± SE was calculated by averaging 12 zeros and 1 number, resulting in an SE value that is the same as the mean. a

558 Sudden Death in Epilepsy: Forensic and Clinical Issues 22

}

Left cortex Right cortex Cats received PTZ and timolol Hippocampus Left cortex in cats receiving PTZ only

Mean number of episodes epileptiform activity lasting ≤10 s (episodes/min; mean ± SE)

20 18 16 14 12 10 8 6 4 2 0

Control

5

10

15

20

25

30

35

30

45

PTZ PTZ Timolol Timolol Timolol Timolol Timolol Timolol Timolol 10 100 500 1 5 10 25 10 20 mg/kg mg/kg µg/kg µg/kg µg/kg mg/kg mg/kg mg/kg mg/kg

50

55

60

Figure 34.3╇ Epileptiform activity with a duration of ≤10 s obtained in cats treated with PTZ and timolol. For purposes of comparison, data obtained in cats receiving only PTZ are shown for the comparable experimental duration. The control values obtained in the cats receiving both PTZ and timolol are means obtained in the 10-min period before the first dose of PTZ. (Reproduced from Lathers, C. M., et al., Epilepsy Res, 4, 42–54, 1989b. With permission.)

mean times to total suppression of epileptiform activity in the left and right cortices and the hippocampus were 26 ± 2, 26 ± 2, and 27 ± 2 min, respectively. In six cats, the time to total suppression of epileptiform activity was the same in only three brain regions. In six of the other seven cats, the dose of timolol producing total suppression of epileptiform activity was the same for all three brain areas; the times to suppression varied by only a few seconds. In one cat the epileptiform activity was suppressed in the left and right cortices at a dose of 5 mg timolol, whereas a dose of 10 mg was required to elicit total suppression in the hippocampus. The administration of timolol decreased the duration of all types of epileptiform activity, that is, prolonged ictal (>10 s), brief ictal, and interictal (Table 34.3). Table 34.4 Number of Cats and Dose of Timolol Inducing Total Suppression of Epileptiform Activitya Number of Cats Doses of Timolol (mg)

1

5

10

20

Left cortex Right cortex Hippocampus

2 2 2

3 3 2

1 1 2

6 5 7

a

Epileptiform activity was not induced by PTZ in the left and right cortex of one cat and in the right cortex of a second animal.

Antiepileptic Activity of Beta-Blocking Agents

559

The Friedman ANOVAs were signiἀcant (p ≤ 0.0001) for all three areas of the brain. The Wilcoxon post hoc tests revealed the expected PTZ-induced increase in epileptiform activity from control. In the 10 cats receiving 20 mg PTZ, this dose was not different from control (for any of the three areas), although the means were higher than for 10 mg PTZ. In fact, none of the Bonferroni-corrected Wilcoxon tests were signiἀcant between 20 mg PTZ and any other dose, for any area. The ἀrst dose of timolol elicited slightly (not signiἀcant) higher rates of prolonged and brief ictal and interictal activity than 20 mg PTZ, after which a steady decrease in epileptiform activity occurred, becoming signiἀcant (one-tailed) by the dose of 1 mg/kg i.c.v. timolol. This trend occurred in all three areas of the brain. None of the Friedman ANOVAs or duration more than 10 s approached signiἀcance; this was anticipated because many subjects had no prolonged ictal episodes at any dose. Table 34.2 shows the sequence in which the mean arterial blood pressure and heart rate were depressed by timolol in relation to the time when timolol partially or totally suppressed the epileptiform activity. In 12 of 13 cats, the epileptiform activity was partially suppressed before the fall in mean arterial blood pressure, that is, the fall in this parameter followed partial suppression induced by 10 μg timolol. The heart rate decreased in all 13 cats after the epileptiform activity was partially suppressed by this dose of timolol. In almost all of the cats, the decrease in the mean arterial blood pressure and heart rate preceded total suppression of epileptogenic activity when the cardiovascular parameters were compared to control values. When mean arterial blood pressure and heart rate values at the time of total suppression of epileptogenic activity were compared to blood pressure and heart rate values in the preceding 5-min interval, in most cats (7/13) the blood pressure decreased before total suppression of the epileptiform activity by timolol. However, the decrease in the mean heart rate followed total suppression of epileptogenic activity in six cats. In ἀve cats, the heart rate did not change immediately before or after the occurrence of total suppression of epileptiform activity. 34.3.2â•…Data Obtained in Anesthetized Pigs A transient increase (16.3–50.0%) in the mean arterial blood pressure occurred after the PTZ administration. The elevated blood pressure gradually declined to the basal level within 10 min after PTZ. Intravenously administered propranolol signiἀcantly reduced this transient pressor response and returned the elevated blood pressure to the basal level at 2 min after drug infusion (Figure 34.4). There was no signiἀcant change in the basal heart rate after PTZ administration. However, a marked bradycardia was observed at 1 min after intravenous propranolol administration. Propranolol produced a maximal decrease of 32–38% in the basal heart rate; the bradycardia persisted throughout the experiment (Figure 34.5). Epileptogenic activity induced by PTZ was associated with the occurrence of premature ventricular contractions in some pigs, similar to those observed in anesthetized cats when administered PTZ. Figure 34.4 illustrates the duration of seizure activity elicited by PTZ over an experimental period of 20 min. The seizure activity was continuous from t = 0 to 1 min for both groups. Intravenous propranolol produced a signiἀcant reduction in the duration of seizure activity 1 min after drug infusion; the seizure durations (second per minute interval) were 36.3 ± 4.8 and 12.3 ± 5.1 for the control and intravenous groups, respectively. Animals treated with intravenous propranolol had reduced duration of seizure activity throughout the entire experiment when compared to the control animals.

560 Sudden Death in Epilepsy: Forensic and Clinical Issues

Duration of seizure activity (s/min interval)

60 50 40 30 20 10 0

0

5 Propranolol (2.5 mg/kg)

10 Time (min)

15

20

PTZ (100 mg/kg, i.v.)

Figure 34.4╇ The effect of intravenous propranolol (2.5 mg/kg) on seizure activity induced by PTZ (100 mg/kg i.v.) in pigs (n = 5–6). PTZ was given to induce seizure activity. Propranolol was administered 60 s after the onset of seizure activity. The seizure activity at time zero was determined from the seizure duration of 0- to 1-min interval. A significant suppression of seizure activity was observed at 1 min after propranolol administration. (Modified and reproduced from Lathers, C. M., et al., Epilepsia, 30, 473–479, 1989a. With permission.)

Propranolol µg/ml plasma

10.0

1.0

0.1

0

2

4

6

8

10

12

14

16

18

20

Minutes after the dose

Figure 34.5╇ Propranolol plasma concentrations versus time in pigs administered propranolol 2.5 mg/kg i.v. Values are means ± SD (n = 6).

Antiepileptic Activity of Beta-Blocking Agents

561

Plasma propranolol concentrations were determined from 1 to 20 min after the intravenous administration of 2.5 mg/kg. After a rapid fall from 6.04 ± 1.43 μg/mL at minute 1 to 1.69 ± 0.31 μg/mL at minute 5, a steady decline in plasma concentration was observed up to 20 min, with a mean half-life of 23.3 ± 4.8 min (Figure 34.5). The brevity of the experiment precluded the determination of further kinetic parameters.

34.4â•…Discussion Previous studies have shown that PTZ administered intravenously induced epileptiform activity associated with cardiac arrhythmias and changes in the mean arterial blood pressure and heart rate (Carnel et al. 1985; Lathers and Schraeder 1982; Lathers et al. 1984; Schraeder and Lathers 1983). The present study demonstrated that the central administration of PTZ produced similar effects within seconds of its intracerebroventricular injection. These data support the conclusion that the intravenous administration of PTZ has little direct effect on the heart in eliciting cardiac arrhythmias and that such arrhythmias are associated with the epileptiform discharges induced by PTZ. The central administration of timolol decreased mean arterial blood pressure and heart rate. In another study, timolol administered intravenously to anesthetized cats also decreased the mean arterial blood pressure and heart rate (Lathers 1980) and exhibited an antiarrhythmic action against ouabain-induced arrhythmias. Lathers et al. (1986) demonstrated that chronic oral dosing with timolol for 1 or 2 weeks increased the time, although not signiἀcantly, to arrhythmia induced by acute permanent occlusion of the left anterior descending coronary artery. In the present study, increasing doses of timolol administered intracerebroventricularly and intravenously not only signiἀcantly decreased the elevation of mean arterial blood pressure and heart rate but also decreased and subsequently abolished the incidence of cardiac arrhythmias associated with the epileptiform activity. Epileptiform activity elicited in the present study with a duration less than or equal to 10 s includes both interictal and brief ictal discharges. PTZ-induced bilateral interictal spike activity is indicative of increased cortical excitability often evident in the EEG records of epileptic individuals. PTZ-induced generalized asynchronous clonic movements followed by a tonic convulsion in which limb movements are flexion followed by extension are analogous to the brief ictal discharges. Motor activity characterized by forelimb clonus is analogous to the prolonged ictal activity, that is, duration greater than 10€s. In the animal model using the cat and intracerebroventricular injection of PTZ, most of the epileptiform activity elicited was interictal and/or brief ictal activity. It has been hypothesized that the cardiac arrhythmias associated with interictal activity could be one potential mechanism for sudden unexplained death in epileptic persons (Carnel et al. 1985; Lathers and Schraeder 1982; Lathers et al. 1984; Schraeder and Lathers 1983). That timolol either partially or totally abolishes the epileptiform activity in the shorter duration category suggests that it may be a useful therapeutic agent to suppress the interictal discharges associated with cardiac arrhythmias. The administration of PTZ elicited epileptiform activity that was followed by increases in blood pressure, heart rate, and cardiac arrhythmias. Exactly how these changes develop is unknown, but this laboratory (Kraras et al. 1987; Suter and Lathers 1984) has proposed a possible mechanism to explain how epileptogenic activity and autonomic dysfunction may occur in epileptic patients, resulting in fatal cardiovascular changes. PTZ, trauma,

562 Sudden Death in Epilepsy: Forensic and Clinical Issues

inhibition of prostaglandin transport across the blood–brain barrier, or altered synthesis or metabolism of central enkephalins may lead to increased central levels of PGE2 and/or enkephalins. The consequence of this is thought to be inhibition of central GABA release, epileptogenic activity, increased blood pressure and heart rate, increased sympathetic and parasympathetic central neural outflow, impaired or imbalanced cardiac sympathetic and parasympathetic discharge, and a resultant arrhythmia and/or death. In the present study, the central intracerebroventricular administration of timolol partially suppressed the epileptiform activity and subsequently decreased the blood pressure and heart rate values elevated by PTZ. It may be that timolol is interfering with the central actions of PGE2 or enkephalins to reverse their known capabilities to induce epileptiform activity (see Chapter 18). Additional experimental studies are required to verify this suggestion. Additional mechanisms to explain the anticonvulsant and antiarrhythmic actions of timolol are discussed later. It has been theorized that pharmacological agents capable of suppressing epileptiform activity and the sympathetic component of cardiac arrhythmias may be the best regimens to prevent interictal activity and the associated cardiac arrhythmias that may contribute to the production of sudden unexplained death in the epileptic person (Carnel et al. 1985; Lathers and Schraeder 1982; Lathers et al. 1984; Schraeder and Lathers 1983). The data obtained in the present study indicate that in the experimental setting the pharmacologic agent timolol possesses components of both of these capabilities. Blockade of cardiac beta receptors, a cardiac neurodepressant effect, and/or membrane depressant actions of betablocking agents are thought to contribute to the antiarrhythmic action of beta-blocking agents (Lathers and Spivey 1987). PTZ has been used to induce seizure activity in humans (Franz 1980; Van Buren 1958), study seizure mechanisms (Faingold and Berry 1973; Krall et al. 1978; Langeluddeke 1936; Swinyard 1972), examine autonomic dysfunction associated with epileptogenic activity (Lathers and Schraeder 1982; Onuma 1957; Orihara 1952; Schraeder and Lathers 1983; Van Buren 1958; Van Buren and Ajmone-Marsan 1960), and screen anticonvulsant agents (Carnel et al. 1985; Faingold and Berry 1973; Lathers et al. 1984). Because it is accepted that PTZ is a convulsive model and that many drugs capable of suppressing the PTZ-induced epileptiform activity are anticonvulsant agents, the results of this study suggest that timolol exhibited an anticonvulsant action. Although the data indicate that timolol can reverse the effects of PTZ on the brain, this does not necessarily mean that timolol has intrinsic “anticonvulsant” properties separate from an ability to reverse the effects of PTZ. To answer this question, additional studies must be done to determine whether timolol will protect against seizures induced in other experimental models of epilepsy. In particular, it would be important to evaluate the capability of timolol to suppress interictal discharges and cardiac arrhythmias elicited in other in vivo experimental models not involving PTZ. If timolol also suppresses both the interictal discharges and the arrhythmias in these experimental models, this would provide additional evidence to support the possibility that timolol may be an effective agent to use in epileptic patients to prevent sudden unexplained death. The concept that beta-blocking agents may possess anticonvulsant action is not new (Bose et al. 1963; Conway et al. 1978; Papanicolaou et al. 1982). The studies of Dashputra et al. (1985), Jaeger et al. (1979), Murmann et al. (1966), and Tocco et al. (1980) demonstrated that propranolol possesses anticonvulsant actions. Mueller and Dunwiddie (1983) showed that timolol selectively blocked the proconvulsant activity of 2-fluoro-norepinephrine and 1-isoproterenol in in vitro hippocampal slice preparations superfused with penicillin and

Antiepileptic Activity of Beta-Blocking Agents

563

elevated levels of potassium. Louis et al. (1982) reported that propranolol or timolol (0.25 μg/kg i.c.v.) produced an anticonvulsant action when PTZ was used to induce convulsions in rats. The anticonvulsant action of timolol reported here for the data obtained in swine is similar to the anticonvulsant action of diazepam when used in the same experimental model (Lathers et al. 1987; Spivey et al. 1987). The anticonvulsant action of beta-blocking agents is commonly ascribed to a membranestabilizing effect, although exceptions have been reported (Lints and Nyquist-Battie 1985). Other proposed anticonvulsant mechanisms include decreased central serotonergic (Conway et al. 1978) and monoamine oxidase activity (Bose et al. 1963). An additional possible antiepileptic mechanism of the beta-blocking agents may include beta-Â�adrenoceptor blockade, especially beta2 receptors in the central nervous system (Papanicolaou et al. 1982). Although norepinephrine is generally believed to be anticonvulsant, studies suggest that norepinephrine may exacerbate seizure activity via activation of beta receptors. The state of abnormal seizure susceptibility, but not severity, in genetically epilepsy-prone rats may be determined by norepinephrine deἀcits in the hypothalamus/thalamus (Dailey and Jobe 1986). Both severity and susceptibility can be determined by norepinephrine deἀcits in the telencephalon, midbrain, and pons medulla, whereas seizure severity but not susceptibility may be determined by norepinephrine abnormalities in the cerebellum. Noradrenergic effects may not be uniform throughout the hippocampus; thus, selective activation of alpha or beta receptors by norepinephrine in the brain areas such as the hippocampus might produce either anticonvulsant or proconvulsant effects, respectively (Mueller and Dunwiddie 1983). Beta-blocking agents can increase norepinephrine concentration in cerebral spinal fluid (Tackett et al. 1981) and potentiate the effects of exogenously administered norepinephrine on vas deferens contraction (Patil et al. 1968). If a similar action occurred in this study, the establishment of beta blockade with timolol would increase the central norepinephrine concentration. The increased norepinephrine activity at the central postsynaptic alpha1 receptor sites may account for the anticonvulsant effects of beta-blocking agents (Goldman et al. 1987). Thus, the protective mechanism for timolol against seizures induced by PTZ may be due to a selective blockade of seizure-inducing beta receptors, allowing available norepinephrine to stimulate the central alpha1 receptors that exert an anticonvulsant action. In addition to the possibility that the central alpha1 receptors may be involved in the anticonvulsant action of beta-blocking agents, the role of central postsynaptic alpha2 receptors must be evaluated. Activation of alpha2 receptors decreases the excitability of CA1 pyramidal neurons (Mueller et al. 1982). Clonidine and 1-m-norepinephrine are more selective for alpha2 than for alpha1 receptors and inhibit epileptiform activity at low concentrations; the alpha1 agonist 1-phenylephrine was ineffective at much higher concentrations. These data suggest that central postsynaptic alpha 2 receptors may play a greater role than the alpha1 receptors in the anticonvulsant action of timolol observed in the present study. Deἀnitive experiments will have to be done to conἀrm this possibility.

34.5â•… Summary The experiments in this study were designed to explore the ability of beta-blocking agents to suppress seizures induced by PTZ in two species: the cat and the pig. Cats were anesthetized with alpha-chloralose and PTZ (10–20 mg i.c.v.) was administered to elicit epileptiform

564 Sudden Death in Epilepsy: Forensic and Clinical Issues

activity, including both interictal and ictal discharges. Various doses of timolol (10, 100, 500 μg/kg i.c.v. and 1, 5, 10, and/or 20 mg/kg i.v.) were then administered at 5-min intervals to determine whether it suppressed the epileptiform activity. Mean arterial blood pressure increased after the administration of PTZ and was associated with the development of epileptiform activity. Heart rate also was increased after PTZ. All doses of timolol caused a decrease in the blood pressure and heart rate elevated by PTZ. The administration of timolol also suppressed the epileptiform activity. Similar ἀndings were obtained in cats that received the same doses of timolol administered at different time intervals. The data indicate that the central administration of timolol reverses the epileptiform activity of PTZ on the brain and suppresses the associated increases in blood pressure and heart rate. Domestic swine (13–20 kg) were prepared for recordings of arterial blood pressure, ECG, and electrocortical activity. Seizure activity was induced by PTZ (100 mg/kg i.v.). Sixty seconds after the onset of seizure activity, the animals received either no drug (control) or propranolol (2.5 mg/kg i.v.). A transient increase in the mean arterial blood pressure was observed after PTZ administration. Intravenous propranolol signiἀcantly suppressed the seizure duration (second per minute interval) at 1 min after drug administration; seizure duration control, 36.3 ± 4.8; i.v. propranolol, 12.3 ± 5.1. Intravenous propranolol also produced a maximal decrease of 32–38% in the basal heart rate and reduced the transient increase in mean arterial blood pressure elicited by PTZ, with no signiἀcant effect on the basal mean arterial blood pressure. Plasma propranolol levels were found to be 6.07 ± 1.43 μg/mL at 1 min after administration, falling to 1.10 ± 0.27 μg/mg over the following 19 min of the experiment. The data demonstrate that propranolol possesses anticonvulsant activity against PTZ-induced seizures in both the pig and in the cat.

Acknowledgments The study was funded by a grant from the Epilepsy Foundation of America and from the Ben Franklin Partnership Fund, a program of the Commonwealth of Pennsylvania. The authors would like to thank Valerie Farris, Larry Pratt, and Michele Spino for technical help and also for typing the manuscript, and Dr. Edward Gracely for statistical analyses.

References Albani, F., R. Riva, and A. Baruzzi. 1982. Simple and rapid determination of propranolol and its active metabolite, 4-hydroxypropranolol, in human plasma by liquid chromatography with fluorescence detection. J Chromatogr 228: 362–365. Bose, B. C., A. Q. Saiἀ, and S. Sharma. 1963. Studies on anticonvulsant and antiἀbrillatory drugs. Arch Int Pharmacodyn 146: 106–113. Carnel, S. B., P. L. Schraeder, and C. M. Lathers. 1985. Effect of phenobarbital pretreatment on cardiac neural discharge and pentylenetetrazol-induced epileptogenic activity in the cat. Pharmacology 20: 225–240. Conway, J., D. T. Greenwood, and D. N. Middlemiss. 1978. Central nervous actions of β-adrenoceptor antagonists. Clin Sci Mol Med 54: 119–124. Dailey, J. W., and P. C. Jobe. 1986. Indices of noradrenergic function in the central nervous system of seizure-naive genetically epilepsy-prone rats. Epilepsy 27: 665–670. Dashputra, P. G., V. P. Patki, and T. J. Hemnani. 1985. Antiepileptic action of beta-adrenergic blocking drugs: Pronethal and propranolol. Mater Med Pol 2: 88–92.

Antiepileptic Activity of Beta-Blocking Agents

565

Faingold, C. L., and C. A. Berry. 1973. Quantitative evaluation of the pentyl-enetetrazol-anticonvulsant interaction on the EEG of the cat. Eur J Pharmacol 24: 381–388. Franz, D. N. 1980. The Pharmacological Basis of Therapeutics. New York, NY: Macmillan. Gillis, R. A. 1969. Cardiac sympathetic nerve activity: Changes induced by ouabain and propranolol. Science 166: 508–510. Goldman, B. D., A. Z. Stauffer, and C. M. Lathers. 1987. Beta blocking agents and the prevention of sudden unexpected death in the epileptic person: Possible mechanisms. Fed Proc 46: 705. Gremels, H. 1931. Uber die Einwirkung einiger zentral-erregender Mittel auf Atmung und Kreislauf. Arch Exp Pathol Pharmacol 162: 29–45. Hahn, F. 1960. Analeptics. Pharmacol Rev 12: 447–530. Han, J., and G. K. Moe. 1964. Nonuniform recovery of excitability in ventricular muscle. Circ Res 14: 44–60. Hildebrandt, E. 1937. Pentamethylenetetrazol (Cardiazol). Handb Exp Pharmacol 5: 151–183. Hirsch, C. S., and D. L. Martin. 1971. Unexpected death in young epileptics. Neurology 21: 682–690. Jaeger, V., B. Esplin, and R. Capek. 1979. The anticonvulsant effects of propranolol and beta-Â�adrenergic blockade. Experientia 35: 8081. Krall, R. L., J. K. Perry, B. G. White, H. J. Kupferberg, and E. A. Swinyard. 1978. Antiepileptic drug development: II. Anticonvulsant drug screening. Epilepsia 19: 409–428. Kraras, C. M., N. Turner, and C. M. Lathers. 1987. The role of enkephalins in the production of epileptogenic activity and autonomic dysfunction: Origin of arrhythmia and sudden death in the epileptic patient? Med Hypotheses 23: 19–31. Langeluddeke, A. 1936. Die diagnostische Bedeutung experimentell erzeugter Krampfe. Dtsch Med Wehnschr 62: 1588–1590. Lathers, C. M. 1980. Effect of timolol on autonomic neural discharge associated with ouabainÂ�induced arrhythmia. Eur J Pharmol 64: 95–106. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23: 633–647. Lathers, C. M., and W. H. Spivey. 1987. The effect of timolol, metoprolol, and practolol on postganglionic cardiac neural discharge associated with acute coronary occlusion-induced arrhythmia. J Clin Pharmacol 27: 582–592. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1977. Correlation of ouabain-induced arrhythmia and nonuniformity in the histamine-evoked discharge of cardiac sympathetic nerves. J Pharmacol Exp Ther 203: 467–479. Lathers, C. M., G. J. Kelliher, J. Roberts, and A. B. Beasley. 1978. Nonuniform cardiac sympathetic nerve discharge. Circulation 57: 1058–1064. Lathers, C. M., P. L. Schraeder, and S. B. Carnel. 1984. Neural mechanisms in cardiac arrhythmias associated with epileptogenic activity: The effect of phenobarbital in the cat. Life Sci 34: 1919–1936. Lathers, C. M., N. Turner, and C. M. Kraras. 1985. Cardiovascular and epileptogenic effects of pentÂ� ylenetetrazol administered intracerebroventricularly in cats. Epilepsia 26: 520. Lathers, C. M., W. H. Spivey, L. E. Suter, J. P. Lerner, N. Turner, and R. M. Levin. 1986. The effect of acute and chronic administration of timolol on cardiac sympathetic neural discharge, arrhythmia, and beta adrenergic receptor density associated with coronary occlusion in the cat. Life Sci 39: 2121–2141. Lathers, C. M., K. Jim, and W. H. Spivey. 1989a. A comparison of intraosseous and intravenous routes of administration for antiseizure agents. Epilepsia 30: 472–479. Lathers, C. M., A. Z. Stauffer, N. Turner, C. M. Kraras, and B. D. Goldman. 1989b. Anticonvulsant and antiarrhythmic actions of the beta blocking agent timolol. Epilepsy Res 4: 42–54. Leestma, J. E., M. B. Kalelkar, S. S. Teas, G. W. Jay, and J. R. Hughes. 1984. Sudden unexpected death associated with seizures: Analysis of 66 cases. Epilepsia 1: 84–88. Lints, C. E., and C. Nyquist-Battie. 1985. A possible role for beta-adrenergic receptors in the expression of audiogenic seizures. Pharmacol Biochem Behav 22: 711–716.

566 Sudden Death in Epilepsy: Forensic and Clinical Issues Louis, W. J., J. Papanicolaou, R. J. Summers, and F. J. E. Vajda. 1982. Role of central beta adrenoceptors in the control of pentylenetetrazol-induced convulsions in rats. Br J Pharmacol 75: 441–446. Lown, B., and R. L. Verrier. 1978. Neural factors and sudden death. In Perspectives in Cardiovascular Research, vol. 2, Neural Mechanisms in Cardiac Arrhythmias, ed. P. J. Schwartz, A. M. Brown, A. Malliani, and A. Zanchetti, 87–88. New York, NY: Raven Press. Mueller, A. L., and T. V. Dunwiddie. 1983. Anticonvulsant and proconvulsant actions of alpha- and beta-noradrenergic agonists on epileptiform activity in rat hippocampus in vitro. Epilepsia 24: 51–64. Mueller, A. L., B. J. Hoffer, and T. V. Dunwiddie. 1981. Noradrenergic responses in rat hippocampus: Evidence for mediation by alpha and beta receptors in the in vitro slice. Brain Res 214: 113–126. Murmann, W., L. Almirante, and M. Saccani-Gueli. 1966. Central nervous system effects of four beta-adrenergic receptor blocking agents. J Pharm Pharmacol 18: 317–318. Onuma, T. 1957. Relationships of the predisposition to convulsions with the action potentials of the autonomic nerves and the brain: II. Changes in action potential of the autonomic nerves and the brain under conditions for increasing the predisposition to convulsions. Tohoku J Exp Med 65: 121–129. Orihara, O. 1952. Comparative observations of the action potential of autonomic nerve with EEC. Tohoku J Exp Med 57: 43–54. Papanicolaou, J., F. J. Vajda, R. J. Summers, and W. J. Louis. 1982. Role of beta-adrenoreceptors in the anticonvulsant effect of propranolol on leptazol-induced convulsions in rats. J Pharm Pharmacol 34: 124–125. Patil, P. N., A. Tye, C. May, S. Hetey, and S. Miyagi. 1968. Steric aspects of adrenergic drugs: XI. Interactions of dibenamine and beta adrenergic blockers. J Pharmacol Exp Ther 163: 309–319. Randall, W. C., J. X. Thomas, D. E. Euler, and G. J. Rozanski. 1978. Cardiac dysrhythmias associated with autonomic nervous system imbalance in the conscious dog. In Perspectives in Cardiovascular Research, vol. 2, Neural Mechanisms in Cardiac Arrhythmias, ed. P. J. Schwartz, A. M. Brown, A. Malliani, and A. Zanchetti, 123–138. New York, NY: Raven Press. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32: 1371–1382. Snider, R. S., and W. T. Neimer. 1970. A Stereotaxic Atlas of the Brain. Chicago, IL: University of Chicago Press. Spivey, W. H., H. D. Unger, C. M. Lathers, and R. M. McNamara. 1987. Intraosseous diazepam suppression of pentylenetetrazol-induced epileptogenic activity in pigs. Ann. Emergency Med. 16: 156–159. Suter, L. E., and C. M. Lathers. 1984. Modulation of presynaptic gamma aminobutyric acid release by prostaglandin E2: Explanation for epileptogenic activity and dysfunction in autonomic cardiac neural discharge leading to arrhythmias? Med Hypotheses 15: 15–30. Swinyard, E. A. 1972. Assay of antiepileptic drug activity in experimental animals: Standard tests. In Anticonvulsant Drugs, International Encyclopedia of Pharmacology and Therapeutics, Section 19.1, 47–65. Oxford: Pergamon Press. Tackett, R. L., J. G. Webb, and P. J. Privitera. 1981. Cerebroventricular propranolol elevates cerebrospinal fluid norepinephrine and lowers blood pressure. Science 213: 911–913. Tocco, D. J., B. V. Clineschmidt, A. E. W. Duncan, F. A. Deluna, and J. R. Baer. 1980. Uptake of the beta-adrenergic blocking agents propranolol and timolol by rodent brains: Relationship to central pharmacological action. J Cardiovasc Res 2: 133–143. Toman, J. E. P., and J. P. Davis. 1949. The effects of drugs upon the electrical activity of the brain. Pharmacol Rev 1: 425–492. Van Buren, J. M. 1958. Some autonomic concomitants of ictal autonomism. Brain 81: 505–528. Van Buren, J. M., and C. Ajmone-Marsan. 1960. Correlations of autonomic and EEG components in temporal lobe epilepsy. Arch Neurol 3: 683–703.

Arrhythmias Associated with Epileptogenic Activity Elicited by Penicillin

35

Claire M. Lathers Paul L. Schraeder

Contents 35.1 Introduction 35.2 Method 35.3 Results 35.4 Discussion 35.5 Summary Acknowledgments References

567 567 568 571 574 574 574

35.1â•…Introduction The association of autonomic dysfunction, clinical epilepsy, and sudden unexplained death has been the subject of many studies (Leestma et al. 1984; Terrence et al. 1975). Lathers and Schraeder (1982) and Schraeder and Lathers (1983) observed autonomic dysfunction in cats after epileptogenic activity induced by pentylenetetrazol. A marked increase in variability in mean autonomic cardiac sympathetic and parasympathetic neural discharge was associated with the epileptogenic activity. It was hypothesized that if altered cardiac neural discharge also occurs in the patient with epilepsy, cardiac arrhythmias and sudden unexplained death may occur. The ideal agent to prevent these events should possess anticonvulsant, antiarrhythmic, and cardiac neural depressant properties. This study developed a new small-animal model to study autonomic dysfunction in association with epileptogenic activity produced by injecting penicillin into the hippocampus of the cat. Epileptogenic activity was monitored as it spread to the left and right hippocampi and cerebral cortices. Data were analyzed to determine whether changes in the autonomic parameters of mean arterial blood pressure and heart rate were associated with both the interictal and the ictal epileptogenic activity. Phenobarbital was administered to determine whether it suppressed the epileptogenic and arrhythmic activities.

35.2â•…Method The stereotaxic hippocampal injection of aqueous penicillin solution in 11 cats anesthetized with general anesthesia elicited both interictal and ictal activities. Cats were anesthetized intravenously (i.v.) with alpha-chloralose (80 mg/kg) and surgically prepared for 567

568 Sudden Death in Epilepsy: Forensic and Clinical Issues

monitoring the mean arterial blood pressure, lead II ECG, and for drug administration as described by Lathers and Schraeder (1982). A burr hole was made in the region of the posterior sylvian and posterior ectosylvian gyri bilaterally after the animal was placed into a stereotaxic head holder (David Kopf Instruments). A microcannula and a concentric bipolar recording electrode were inserted into the hippocampus using coordinates obtained from a stereotaxic atlas of the cat brain (Snider and Niemer 1961). Electrocorticographic recording electrodes were placed on the left and right (motor) cortices and the hippocampi. Penicillin was injected into the right hippocampus (coordinates A +7, HD −6.0, and RL +11.8). Motor cortex activity was recorded because of evidence that the frontal cortex is involved in cardiovascular regulation (Yingling and Skinner 1976). Epileptogenic activity, interictal and ictal spikes, was elicited by the right hippocampal injection of penicillin as an aqueous solution of 400,000 U/mL colored with methylene blue to verify postmortem the injection recording site. A microsyringe in stereotaxic carrier was used to inject 0.0025 mL penicillin (1000 U). The epileptogenic activity, quantiἀed in spikes per minute, was correlated with changes in mean arterial blood pressure, heart rate, and ECG. Either 20 or 40 mg/kg sodium phenobarbital (Elkins Sinn, Inc.), dissolved in 5 mL physiologic (0.9%) saline, was infused into the femoral vein at a rate of 0.5 mL/min, followed by a 2-min wash at the same rate with saline. Phenobarbital was administered after the injection of penicillin into the hippocampus. One-factor repeated-measures analysis of variance were run where the independent variable was time (every 4 min) and the dependent measure was either heart rate or blood pressure. This was repeated for both penicillin and phenobarbital, creating four separate analyses. When a signiἀcant F ratio (using the Huynh–Feldt correction to degrees of freedom) was obtained, the Newman–Keuls post hoc procedure for determining which pairwise comparisons were signiἀcant was run at α = 0.05 (Winer 1962). Analyses of variance were done using biomedical programs, subprogram P2V. Post hoc tests were accomplished using the Statistical Package for the Social Sciences, subprogram one way.

35.3â•…Results Changes in the electrocardiogram that were observed after the administration of 1000 U penicillin in 11 cats included T-wave inversion in ἀve, changes in the ST and P–R intervals in four and three, respectively, alterations in the QRS complex in four, and ST depression in two. The administration of phenobarbital 20 mg/kg (i.v.) to four additional cats or 40 mg/kg (i.v.) to nine additional cats abolished the penicillin-induced changes in the ECG. The ECG changes included alterations in the P and QRS waves, the appearance of a U wave, and premature ventricular contractions. At some experimental times, the premature ventricular contractions occurred before the appearance of the epileptogenic activity; at other times in the same cat, the premature ventricular contractions appeared just after the initiation of the epileptogenic activity. The changes in the ECG were not caused by anesthesia or by surgical stress because they were not observed in the control period. In the rare event that anesthesia initiated arrhythmias, the cat was not included in the study. Data from one cat are depicted in Figure 35.1. Penicillin (1000 U/μL) induced interictal activity in the left motor cortex and in the left and right hippocampi 1 min after administration (not shown). Ictal activity developed in the right hippocampus 5 min after the injection of penicillin and was associated with a 5-mm Hg increase in the mean arterial

Electroencephalogram (µV)

Mean arterial blood pressure (mm Hg)

0

100

200

6

5

1s Control

176

78

R. hippocampus

4

R. motor cortex

3

L. hippocampus

2

L. motor cortex

1

1s 5 min after penicillin injection

176

80

(b)

30 min after penicillin injection

246

133

(c)

115

1 h after penicillin injection

6 242

5

R. hippocampus

R. motor cortex

3

L. hippocampus

L. motor cortex 2

1

(d)

10 min after phenobarbital infusion

138

50

(e)

Figure 35.1╇ Changes in the electroencephalogram and cardiovascular function observed in one cat after the administration of penicillin (100 U/ μL) into the right hippocampus. Phenobarbital (40 mg/kg, i.v.) was administered 2 h after the administration of penicillin. (a) Left motor cortex; (b) left hippocampus; (c) right motor cortex; (d) right hippocampus; (e) blood pressure; and row 6, electrocardiogram. The numbers above the mean arterial blood pressure and electrocardiogram tracings indicate the blood pressure and heart rate values in millimeters of mercury and beats per minute, respectively. Labels on the panels indicate data obtained in the control period and the times at which the data were obtained after the administration of penicillin and phenobarbital 40 mg/kg (i.v.).

Electrocardiogram (lead II)

(a)

Arrhythmias Associated with Epileptogenic Activity Elicited by Penicillin 569

570 Sudden Death in Epilepsy: Forensic and Clinical Issues

blood pressure (column B). Interictal activity was evident in all four electrocorticograms at minute 12 (not shown). Thirty minutes after the administration of penicillin (column C), the mean arterial blood pressure increased to 133 mm Hg and the heart rate to 246 bpm. Interictal spikes were observed in both motor cortices and ictal discharge in both hippocampi 1 h (column D) and 2 h (not shown) after penicillin administration. Mean arterial blood pressure and heart rate values were still increased from control. Penicillin-induced interictal and ictal activities were suppressed at minute 5 of the infusion of phenobarbital (not shown) and were abolished at minute 10 after 40 mg/kg (i.v.) phenobarbital (column E). Penicillin-induced increases in blood pressure and heart rate were also reversed to values lower than the control. Blood pressure and heart rate values were slightly higher than the control 30 min after phenobarbital; no epileptogenic activity was apparent (not shown). The hippocampal injection of penicillin (1000 U/μL) in six cats increased the mean arterial blood pressure from a control of 92 ± 12 to 106 ± 10 mm Hg at 30 min after penicillin. Heart rate was increased from the control of 179 ± 15 to 199 ± 13 bpm at this time. The change in heart rate was signiἀcant (p < 0.05), but the change in blood pressure was not. The administration of phenobarbital (20 mg/kg i.v.) decreased mean arterial blood pressure from the pre-phenobarbital control of 115 ± 14 to 79 ± 14 mm Hg (p < 0.05) 22 min after phenobarbital. Heart rate was decreased from 204 ± 19 to 162 ± 12 bpm (p < 0.05) at this time. To determine whether a higher dose of phenobarbital (40 mg/kg, i.v.) would completely abolish the penicillin-induced epileptogenic activity, ἀve additional cats

Mean arterial blood pressure (mean ± SE; mm Hg)

160

N = 5 cats Cats

140 120 100 80 60

Heart rate (mean ± SE; beats/min)

40 0 250 200 150 100 0

0

10

20

30

Penicillin (1000 U, hippocampal)

40

50

60

70

80

90

100 110 120 Time (min)

Figure 35.2╇ Effect of hippocampal injection of penicillin (1000 U/μL) on mean arterial blood pressure (mm Hg) and heart rate (bpm). Data are graphed as a function of time in minutes and are expressed as the mean ± SE for another group of five cats. Asterisks indicate values that are significantly different from control.

Arrhythmias Associated with Epileptogenic Activity Elicited by Penicillin

571

received hippocampal injections of penicillin (1000 U/μL). The effect of penicillin on the mean arterial blood pressure and heart rate in these ἀve additional cats is depicted in Figure 35.2. Mean arterial blood pressure increased from the control of 105 ± 5 to 143 ± 13 mm Hg 52 min after the administration of penicillin; heart rate increased from 170 ± 22 to 218 ± 16 bpm (p < 0.05) 34 min after penicillin. When phenobarbital (40 mg/kg, i.v.) was administered to these ἀve cats, it decreased mean arterial blood pressure from the prephenobarbital control of 128 ± 14 to 56 ± 28 mm Hg and heart rate from 201 ± 15 to 117 ± 14 bpm (p < 0.05) at 12 and 22 min, respectively (Figure 35.3). Comparison of the effect of 20 and 40 mg/kg (i.v.) phenobarbital revealed that the magnitude of the decrease in blood pressure and heart rate after the larger dose was twice that of the lower dose.

35.4â•…Discussion This study showed that penicillin-induced hippocampal epileptogenic discharges were associated with increases in mean arterial blood pressure and heart rate and changes in the ECG. The dose of 20 mg/kg (i.v.) phenobarbital reversed the associated increases in blood 160 N = 5 cats

Mean arterial blood pressure (Mean ± SE; mm Hg)

140 120 100 80 60 40

Heart rate (mean ± SE; beats/min)

0 250 200 150 100 0

0

10

Phenobarbital (40 mg/kg; i.v.)

20 30 Time (min)

Figure 35.3╇ Effect of phenobarbital (40 mg/kg, i.v.) on penicillin-induced changes in blood

pressure (mm Hg) and heart rate (bpm). Data are graphed as a function of time in minutes and are expressed as the mean ± SE and were obtained in the same five cats depicted in Figure 35.2. Phenobarbital was given 2 h after the administration of penicillin. Asterisks indicate values that are significantly different from control.

572 Sudden Death in Epilepsy: Forensic and Clinical Issues

pressure and heart rate and attenuated the epileptogenic activity; 40 mg/kg (i.v.) phenobarbital completely abolished epileptogenic activity and reversed the associated changes in the cardiovascular parameters. This study demonstrated that the use of general anesthesia with hippocampal injections of aqueous penicillin elicited both interictal and ictal spike activities, in contrast to intracortical penicillin into the cerebral convexity, which does not usually progress to ictal activity in cats under general anesthesia (Prince 1972). Several of our preliminary experiments (unpublished observations), using alpha-chloralose anesthesia and cortical injection of penicillin, elicited only interictal spike activity in the primary focus and in the minor focus, that is, the homologous contralateral cortical area, within 3 to 10 min. Afterdischarges did not develop for as long as 7 h after the intracortical injection of penicillin. In contrast, when ether was used for induction of anesthesia, followed by local anesthesia, afterdischarges occurred within a mean time of 105 min after penicillin injection (Schraeder and Celesia 1977). Thus, the use of alpha-chloralose and cortical convexity injection of penicillin reliably elicits only interictal spikes and therefore is useful only in the examination of the effect of interictal discharge on autonomic cardiovascular function. In contrast, this study found that injection of penicillin into the hippocampus in animals anesthetized only with alpha-chloralose resulted in the progressive development of interictal to ictal discharges. Thus, the use of general anesthesia in the model using hippocampal injection of penicillin does not preclude investigation into the effects of ictal discharge. The signiἀcance of the data reported in this study is that interneuronal pathways connect the hippocampal area to the autonomic cardiovascular areas within the hypothalamus. The hippocampal formation, the amygdaloid complex, the septal region, the gyrus fornicatus, the piriform lobe, and the caudal orbital frontal cortex constitute the limbic forebrain structures (Nauta and Haymaker 1969). The fornix system forms the main efferent pathways from the hippocampus. Fiber systems originating in the limbic forebrain are among the most conspicuous afferents of the hypothalamus. Hippocampal afferents come from the medial septal nucleus and cingulate and the parahippocampal regions of the gyrus fornicatus. The amygdaloid complex is connected with the hypothalamus by the stria terminalis and the ventral amygdalofugal pathway. The septoamygdalar complex projects directly to the hippocampus (Swanson and Cowan 1979). The hypothalamus, at least in part, is then under cerebral cortical control in its influence on the maintenance of homeostasis by virtue of its neural relationships with both divisions of the autonomic nervous system and with both lobes of the pituitary gland. “When the connections of the septohippocampal complex are considered as a whole, the conclusion emerges that it essentially forms the gateway between the hypothalamus and the limbic cortical regions” (Swanson 1983). It is quite possible that with spread of interictal activity in the hippocampus to the hypothalamus, the subclinical epileptogenic activity alters the function of other areas of the brain, with resultant simultaneous changes in the autonomic control of mean arterial blood pressure, heart rate and rhythm, and cardiac neural discharge in the periphery. Furthermore, cardiovascular regulation is a function of neuronal activity in the cerebral cortex, the amygdala, and the medullary reticular formations. Cardioacceleratory and cardioinhibitory centers exist at these levels of the nervous system, with selective activation producing either increased or decreased heart rate. Vasopressor and vasodepressor centers also exist at these central sites and produce their effects through reticulospinal connections to the preganglionic sympathetic neurons of the intermediolateral cell column and through connections to preganglionic parasympathetic neurons. In addition to

Arrhythmias Associated with Epileptogenic Activity Elicited by Penicillin

573

descending input from higher centers, the cardiovascular centers also receive input from peripheral receptors, the most important of which are the baroreceptors of the carotid sinus and the aortic arch. Although peripheral autonomic dysfunction in cardiac autonomic nerves can precede the changes in the ECG associated with subconvulsant interictal discharge, interictal activity alone can also be associated with premature ventricular contractions (Lathers and Schraeder 1982). Thus, minimal epileptogenic activity (single spikes) can be associated with altered cardiac neural discharge and arrhythmias. Schraeder and Celesia (1977) reported that even minimal epileptogenic activity has a wide-ranging effect on cerebral functions monitored at the auditory cortex of the cat. If, then, subclinical epileptogenic activity were to likewise have an effect on other functions of the brain, that is, autonomic regulation, producing autonomic imbalance in cardiac autonomic neural discharge with subsequent arrhythmias, this type of activity could be a contributing factor to the mechanism of unexpected death in epilepsy. That this sequence could occur in the person with epilepsy is supported by the anatomical relationships among the cerebral cortex, the hippocampi, and the hypothalamus. The anatomical and physiological relationship of the frontal cortex with the hypothalamus is complex; studies stimulating the dorsolateral surface of both hemispheres have reported both a rise and a drop in blood pressure and increases or decreases in heart rate (Hoff et al. 1963). Cortical stimulation evoked dilatation of pupils, retraction of the nictitating membrane, piloerection, salivation, sweating, and gastric motility and secretion. The autonomic localization in the motor and premotor cortex corresponds closely with the somatotopic representation (Brooks and Koizumi 1974). The focal model of epilepsy used in this study produced cardiovascular changes that were similar to those produced by pentylenetetrazol-induced epileptogenic activity (Lathers and Schraeder 1982; Schraeder and Lathers 1983). These data support the hypothesis that focal epileptogenic activity can produce alterations in cardiac electrical activity making the heart susceptible to arrhythmias that may cause sudden unexplained death in epileptic persons. The clinical signiἀcance of the data obtained in this type of epileptogenic model is emphasized by Blumhardt et al. (1986). They reported that in some patients with temporal lobe epilepsy, cardiac acceleration preceded the onset of recognizable rhythmic surface EEG seizure activity; this may reflect the onset of electrical discharge in deep limbic circuits and the connections of these structures with the autonomic nervous system. Arrhythmias were observed at times when there were no seizure discharges on the EEG in some patients. Blumhardt and others also noted that the autonomic effects of temporal lobe epilepsy on heart rate and rhythm may be more severe in untreated younger patients. Their suggestion agrees with the observation that young epileptic patients are at high risk of sudden unexplained death. The present study used an animal model that allowed testing of a pharmacologic agent that possesses antiepileptic, antiarrhythmic, and neural depressant activity. Future use of this model should help in the development of better therapeutic regimens designed to eliminate the autonomic dysfunction and arrhythmias associated with epileptogenic activity and ultimately contribute to our understanding of the risk factors for sudden unexplained death in epilepsy. If the data indicate that the autonomic changes are secondary to seizures, the primary clinical therapeutic goal would be to use a pharmacologic agent with maximum anticonvulsant potency. However, if interictal activity is associated with a risk of autonomic dysfunction, questions must be raised about the current therapeutic goal for

574 Sudden Death in Epilepsy: Forensic and Clinical Issues

epilepsy, which is to suppress seizures but not interictal discharges. In the latter case, a new type of drug may be required.

35.5â•…Summary Penicillin-induced epileptogenic activity (1000 U/mL) was recorded bilaterally from the hippocampi and the motor cortices of 11 anesthetized cats. The onset of epileptogenic activity ranged from 1 s to 16 min. Epileptiform activity, consisting of interictal discharges (n = 3) or ictal discharges (n = 3), ἀrst occurred at the injection site, the right hippocampus. Blood pressure increased from 92 ± 12 (control) to 106 ± 10 at 30 min and 115 ± 10 mm Hg at 60 min after penicillin (p > 0.05). Heart rate increased from 179 ± 15 (control) to 194 ± 13 at 30 min and 216 ± 13 bpm 60 min after penicillin (p > 0.05). Maximum increases in blood pressure and heart rate were 55 ± 15 mm Hg and 59 ± 15 bpm, respectively (p < 0.05). ECG alterations included P–R interval changes, increased P-wave amplitude, QRS complex changes, T-wave inversion, ST elevation, and the appearance of premature ventricular contractions. Phenobarbital (20, mg/kg i.v.) diminished the epileptogenic activity and depressed the blood pressure to 79 ± 14 mm Hg at 23 min from 115 ± 14 mm Hg (10 min before phenobarbital; p < 0.05). Heart rate was decreased to 162 ± 12 from the pre-phenobarbital control of 204 ± 19 bpm (p > 0.05). To determine whether a higher dose of phenobarbital (40 mg/kg, i.v.) would completely abolish the penicillin-induced epileptogenic activity, ἀve additional cats received 1000 U/μL penicillin G sodium into the right hippocampus. In these cats the penicillin also produced epileptogenic activity and increased the blood pressure from 105 ± 5 to 143 ± 13 and the heart rate from the control 170 ± 22 to 218 ± 16 (p < 0.05). Phenobarbital (40 mg/kg, i.v.) signiἀcantly reversed the effect of penicillin on the blood pressure and heart rate. Blood pressure dropped from the pre-phenobarbital control of 128 ± 14 to 56 ± 18 mm Hg and heart rate dropped from 201 ± 15 to 117 ± 14 bpm (p < 0.05). This dose of phenobarbital also prevented the penicillininduced epileptogenic activity. Thus, phenobarbital diminished the epileptogenic activity and autonomic dysfunction induced by penicillin. The autonomic dysfunction and epileptogenic activity induced by the peripheral intravenous administration of pentylenetetrazol (Lathers and Schraeder 1982) are similar to those induced by the hippocampal injection of penicillin.

Acknowledgments This study was funded by the Epilepsy Foundation of America. The authors are indebted to Dr. Nihal Tumer, Valerie Farris, and Larry Pratt for technical help, Dr. Edward Gracely for statistical analyses, and Michele Spino for typing the manuscript.

References Blumhardt, L. D., P. E. M. Smith, and L. Owen. 1986. Electrocardiographic accompaniments of temporal lobe epileptic seizures. Lancet 1: 1051–1056. Brooks, C. M., and K. Koizumi. 1974. The hypothalamus and control of integrative processes. In Medical Physiology, 13th ed., ed. V. B. Mountcastle, 813–836. St. Louis, MO: C. V. Mosby.

Arrhythmias Associated with Epileptogenic Activity Elicited by Penicillin

575

Hoff, E. C., J. F. Kell Jr., and M. N. Carrol Jr. 1963. Effects of cortical stimulation and lesions on cardiovascular function. Physiol Rev 43: 68–114. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23: 633–647. Leestma, J. E., M. B. Kalelkar, S. S. Teas, G. W. Jay, and J. R. Hughes. 1984. Sudden unexpected death associated with seizures: Analysis of 66 cases. Epilepsia 25: 84–88. Nauta, W. J. H., and W. Haymaker. 1969. Hypothalamic nuclei and ἀber connections. In The Hypothalamus, ed. W. Haymaker, E. Anderson, and W. J. H. Nauta, 136–209. Springἀeld, IL: Thomas. Prince, D. A. (1972). Topical convulsion drugs and metabolic antagonists. In Experimental Models of Epilepsy. A Manual for the Laboratory Worker, ed. D. P. Purpura, J. K. Peney, D. Tower, D. M. Woodbury, and R. Walter, 51–84. New York, NY: Raven Press. Schraeder, P. L., and G. G. Celesia. 1977. The effects of epileptogenic activity on auditory evoked potentials in cats. Arch Neurol 34: 677–682. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32: 1371–1382. Snider, R. S., and W. T. Niemer. 1961. A Stereotaxic Atlas of the Cat Brain. Chicago, IL: University of Chicago Press. Swanson, L. W. 1983. The hippocampus and the concept of the limbic system. In Neurobiology of the Hippocampus, ed. W. Seifert, 3–20. London: Academic Press. Swanson, L. W., and W. M. Cowan. 1979. The connections of the septal region in the rat. J Comp Neurol 186: 621–656. Terrence, C. F., H. M. Wisotzkey, and J. A. Perper. 1975. Neurogenic pulmonary edema in unexpected, unexplained death in epileptic patients. Neurology 25: 594–598. Winer, B. 1962. Statistical Principles in Experimental Design, 2nd ed. New York, NY: McGraw-Hill. Yingling, C. D., and J. E. Skinner. 1976. Selective regulation of thalamic sensory relay nuclei by nucleus reticularis thalami. Electroencephalogr Clin Neurophysiol 41: 476–482.

Role of Neuropeptides in the Production of Epileptogenic Activity and Arrhythmias

36

Claire M. Lathers

Contents 36.1 Introduction 36.1.1 PTZ-Induced Increased Concentrations of Central Enkephalins and Epileptogenic Activity 36.1.2 Central Neuropeptide-Induced Epileptogenic Activity 36.1.3 Enkephalin Modulation of GABA Release 36.1.4 Met-Enkephalin Modulation of Acetylcholine Release 36.2 Central Cardiovascular and Neurological Effects of Enkephalins 36.2.1 Action on Mean Arterial Blood Pressure, Heart Rate, and Brain Electrical Activity in Conscious Cats 36.2.2 Action on Mean Arterial Blood Pressure, Heart Rate, and Brain Electrical Activity in Anesthetized Animals 36.3 Discussion 36.4 Summary References

577 578 578 578 578 580 580 583 585 588 589

36.1â•…Introduction Pentylenetetrazol (PTZ)-induced interictal and ictal epileptogenic activity associated with autonomic dysfunction—that is, changes in autonomic cardiac neural discharge, mean arterial blood pressure, and heart rate and rhythm—has been reported in the cat (Schraeder and Lathers 1989; Lathers and Schraeder 1982). If the autonomic dysfunction, including the development of arrhythmias, also occurs in humans, it may be a contributory factor to sudden unexplained death in a person with epilepsy. Elevation of immunoreactive (IR) met-enkephalin content in the septum, hypothalamus, amygdala, and hippocampus of rats occurs after PTZ-induced convulsions (Vindrola et al. 1984). This elevation may ultimately change central sympathetic neural discharge to the heart, resulting in the development of arrhythmias. Indeed, numerous reports indicate that neuropeptides may produce epileptic seizures (Elazor et al. 1979; Frenk et al. 1978). Resolution of the question of whether enkephalins elicit epileptogenic activity and autonomic dysfunction via an action to inhibit the release of GABA (Brennan et al. 1980; Snead and Bearden 1980) is important because an understanding of this mechanism should eventually allow the design of pharmacologic agents to prevent the epileptogenic activity and autonomic dysfunction.

577

578 Sudden Death in Epilepsy: Forensic and Clinical Issues

36.1.1â•… P  TZ-Induced Increased Concentrations of Central Enkephalins and Epileptogenic Activity PTZ-induced kindling was associated with a long-lasting elevation in brain content of IR met-enkephalin in the septum, hypothalamus, amygdala, and hippocampus of rats (Vindrola et al. 1984). Kindling was produced by the administration of intraperitoneal injections of 40 mg/kg PTZ every 24 h for 10 days. The control group received an equivalent volume of saline on the same schedule. Every animal was observed for 1 h after each injection for the appearance of convulsions. IR met-enkephalin was quantiἀed in several brain areas 16 days after the last injection of PTZ in both the control and experimental groups. Additional rats received a PTZ dose on day 16 and were sacriἀced 1 and 24 h later. Brain tissue was prepared and enkephalin content was assayed by radioimmunoassay. A long-lasting elevation in amygdala, septum, hypothalamus, and hippocampus€ IR€ metÂ�enkephalin content occurred in animals subjected to kindling and sacriἀced 16 days after the last dose of PTZ. A decrease in IR met-enkephalin occurred 1 h after the PTZ-induced seizure but increased to newly elevated levels 24 h later. Thus, PTZ-induced kindling increased levels of enkephalins that were temporally related to the appearance of seizures. 36.1.2â•… Central Neuropeptide-Induced Epileptogenic Activity In addition to the ἀndings of increased brain concentrations of enkephalins after induction of seizures by PTZ, injection of these agents has been shown to elicit seizure activity. Speciἀcally, the intracerebroventricular injection (Urca and Frenk 1983) and hippocampal injection (Elazor et al. 1979) of leu-enkephalin induced seizures in rats and cats. Snead and Bearden (1980) also found that neuropeptides administered into the central nervous system induced epileptogenic activity in rats. Intraventricular leu-enkephalin produced a consistent dramatic paroxysmal electrical response within the ἀrst 60 s of administration (Figure 36.1), which persisted for up to 6 min. The enkephalin-induced paroxysms increased the 3- to 6-Hz band of the EEG spectrum. This indicated that enkephalin is directly involved in the production of epileptogenic activity (Snead and Bearden 1980). 36.1.3â•… Enkephalin Modulation of GABA Release Met-enkephalin inhibited K+ depolarization-induced release of 3H-GABA from rat synaptosomes in a dose-dependent fashion (Brennan et al. 1980). The concentration of metenkephalin that inhibited 50% of the K+-stimulated release was approximately 5 × 10−10 M (Figure 36.2). In every instance, the reduction of GABA release was prevented by naloxone, suggesting that met-enkephalin may interact with opiate receptors to modulate the release of GABA. 36.1.4â•… Met-Enkephalin Modulation of Acetylcholine Release Met-enkephalin reversibly and speciἀcally reduces the quantal content of acetylcholine release from peripheral nerve terminals in the frog cutaneous pectoris muscle by blocking voltage-dependent Ca+ channels (Bixby and Spitzer 1983). It is likely that met-enkephalin also blocks the release of Ca2+-dependent neurotransmitters, such as GABA, from central synapses. Bixby and Spitzer applied met-enkephalin (10–30 μM) by pressure ejection

Role of Neuropeptides in the Production of Epileptogenic Activity and Arrhythmias 579

RF-RP LF-LP

50 µV

1s

RF-RP LF-LP 50 µV

1s RF-RP LF-LP

50 µV

50 µV

1s

1s

Figure 36.1╇ ECoG changes produced by 100 μg leu-enkephalin injected intracerebroventricularly. The insets show a faster time trace. RF and RP, right frontal and parietal leads; LF and LP, left frontal and parietal leads. The rat was immobile during this paroxysmal electrical activity. The first change in the ECoG occurred within 1 min of administration and was a paroxysm of spikes at a frequency of 7–9 Hz lasting 30–40 s. This tapered off to 2–3 Hz slow-wave activity, which was followed by 20–25 s of low-voltage fast activity. A few seconds of 1-Hz high-voltage slow-wave activity then built up to 25–30 s of low-voltage activity. Finally, a prolonged period of high-voltage single spikes occurred at a rate of 1 paroxysm/5 s. (Reproduced from Snead, O.C., and Bearden, L.J., Science, 210, 1031–1033, 1980. With permission.)

% Inhibition K+-stimulated GABA release

through a “puffer” pipette to the presynaptic terminals of frog neuromuscular junctions before compound nerve stimulation with a suction electrode. Met-enkephalin was also applied in the presence of 15 μM naloxone. Application of saline to the presynaptic terminals before stimulation served as a control. Finally, puffer application of enkephalin was used on the postsynaptic membrane, followed by both iontophoretic application of acetylcholine and nerve stimulation. Application of met-enkephalin to the presynaptic membrane led to a consistent decrease in the size of the evoked response (Figure 36.3a).

100 80 60 40 20 10–13 10–12 10–11 10–10 10–9 10–8 10–7 10–6 10–5 10–4 [Met-enkephalin] M

Figure 36.2╇ A dose–response curve for the inhibition of K+ (55 mM)-induced GABA release

by met-enkephalin. Each point is the mean percentage of inhibition ± SD (n = 5). Basal release of GABA was 7651 ± 980 dpm (n = 46). (From Brennan, M., et al., Life Sci, 27, 1097–1101, 1980. With permission.)

580 Sudden Death in Epilepsy: Forensic and Clinical Issues (a)

Normal

Enkephalin

Normal

4 ms (b)

Normal

10 mV

Normal

Enkephalin

10 mV 200 ms

Figure 36.3╇ (a) Compound nerve stimulation with a suction electrode elicits an endplate

potential that is reduced by focal puffer application of 20 μM met-enkephalin. Saline contained 0.8 mM Ca2+ and 8.0 mM Mg 2+. (b) Responses to iontophoretically applied acetylcholine; the amplitude of the response is increased by focal application of 20 μM met-enkephalin. Note that the response with enkephalin is longer and larger. (Reproduced from Bixby, L., and Spitzer, C., Nature, 301, 431–432, 1983. With permission.)

The decrease, which averaged 40%, was not seen when met-enkephalin was applied in the presence of 15 μM naloxone or when normal saline substituted for enkephalin. In addition, met-enkephalin not only did not reduce but also slightly increased the size of response to iontophoretically applied acetylcholine on the postsynaptic membrane (Figure 36.3b). Thus, the opiates appear to be exerting its effect presynaptically, possibly by blocking the Ca2+ channels. When these channels are blocked, there is decreased release of acetylcholine from nerve terminals in the frog cutaneous pectoris muscle (Bixby and Spitzer 1983). GABA release in the central nervous system has been reported to be Ca2+-dependent (DeBelleroche and Bradford 1972). Therefore, met-enkephalin may be able to inhibit GABA release in the central nervous system by preventing Ca2+ influx into the presynaptic nerve terminal in a manner similar to that by which met-enkephalin reduces the presynaptic release of acetylcholine.

36.2â•… Central Cardiovascular and Neurological Effects of Enkephalins 36.2.1â•…Action on Mean Arterial Blood Pressure, Heart Rate, and Brain Electrical Activity in Conscious Cats The administration of neuropeptides elicits cardiovascular changes as well as epileptogenic activity in conscious male cats (Schaz et al. 1980). After intracerebroventricular application of [(d-Ala 2) methionine-enkephalinamide (DAME)] at a dose of 425 nmol in the cat, arterial systolic and diastolic blood pressures increased, suggesting that (d-Ala2)met-enk may produce a centrally mediated vasopressor response (Figure 36.4). A small increase in heart rate occurred. The maximal cardiovascular response was seen 16 min after the

Role of Neuropeptides in the Production of Epileptogenic Activity and Arrhythmias 581 (-Ala2)-Met-Enkephalin 425 nMol i.c.v. N = 4, n = 9 BP sys BP dio HR

16

32

64

178

40

40

30

30

20

20

10

10

∆ HR%

∆ Bp%

8

Time after injection (min)

Figure 36.4╇ Changes in blood pressure and heart rate (given as percent change of the corresponding control values) elicited by intracerebroventricular injection of DAME, 425 nmol, in freely moving cats. Data are expressed as means ± SEM and have been obtained at different time intervals after the injections of (d-Ala 2)met-enk. Nine experiments were performed in four cats. In control experiments, using intracerebroventricular injections of the same volumes of 0.9% NaCl, changes in arterial blood pressure did not exceed 5–10 mm Hg. Pretreatment values of blood pressure and heart rate were 92 ± 3 mm Hg and 145 ± 3 bpm, respectively. (Reproduced from Schaz, K., et al., Hypertension, 2 (4), 395–407, 1980. With permission.)

intracerebroventricular injection and was attenuated after 64 min. A 170-nmol dose of DAME had no effect on blood pressure and heart rate. A dose of 850 nmol increased arterial pressure and produced catatonia-like behavior during the ἀrst 30 min; this was followed by an excitatory behavior that lasted up to 2 h. Spike-wave complexes occurred within the amygdala and hippocampus after 850 nmol. The time course of electrical activity changes did not exactly follow the hemodynamic changes (Figure 36.5). Spike-wave complexes appeared 5 min after the changes in hemodynamic parameters and lasted only 40 min (Schaz et al. 1980). Yukimura et al. (1981) studied the intracerebroventricular injection of 5, 10, 25, 50, and 100 nmol DAME in cats. In additional experiments, 500 nmol naloxone was injected intracerebroventricularly 3 min before DAME to test the effects of the opioid antagonist on enkephalin action. DAME induced dose-dependent increases in systolic blood pressure and heart rate. Doses of 50 nmol or less produced a maximal increase in blood pressure within 15 min; heart rate did not reach maximum until 30 min. For more than 2 h after 50 and 100 nmol of DAME, sharp waves were seen in the hippocampus recording; theta activity was attenuated. Seizures were not observed. Naloxone given before the injection

582 Sudden Death in Epilepsy: Forensic and Clinical Issues

Am cent

100 µV

Hyphotal

100 µV

Hipp.

500 µV

Sups Gyr lat 200 µV HR beats/min BP mm Hg

240 180 150 50

Am cent

100 µV

Hyphotal

100 µV

Hipp.

500 µV

C

i.c.v. injection (-Ala2)-Met-Enkepahlin 850 nMol

Sups Gyr lat 200 µV

BP mm Hg

240 180 120 150 50

Am cent

100 µV

Hyphotal

100 µV

Hipp.

500 µV

HR beats/min

5’

10’

30’

60’

Sups Gyr lat 200 µV HR beats/min BP mm Hg

240 180 120 150 50

Figure 36.5╇ Recordings of the electrical activity of the central amygdala (Am. cent.), hypo-

thalamus (Hypothal.), hippocampus (Hipp.), and lateral suprasylvian gyrus (Subs. Gyr. lat.); of heart rate (HR), instantaneously recorded as intervals between two heartbeats; and of arterial blood pressure (BP) before, immediately after, and 5, 10, 30, and 60 min after intracerebroventricular injection of (d-Ala 2)met-enk 850 nmol in a freely moving cat. The paper speed can be seen by the continuous marks on the top of each panel: each point represents 1 s. Five minutes (5′) after application of the peptide, there is an increase in arterial pressure by approximately 20 mm Hg. At 10′, arterial pressure is markedly increased by 30 mm Hg, and in the subcortical recordings, there are hypersynchronous waves and spike-wave complexes. At 60 min after intracerebroventricular application of (d-Ala 2)met-enk, electrical recordings are not different from the control period; however, arterial blood pressure is still elevated by 35 mm Hg. (Reproduced from Schaz,€K., et al., Hypertension, 2 (4), 395–407, 1980. With permission.)

Role of Neuropeptides in the Production of Epileptogenic Activity and Arrhythmias 583 Table 36.1â•… Effects of DAME on Blood Pressure, Heart Rate, and Baroreceptor Reflex Sensitivity in Conscious Catsa Treatment

Control

15 min

30 min

60 min

25 nmol DAME Naloxone + 25 nmol DAME 50 nmol DAME Naloxone + 50 nmol DAME

133 ± 8 133 ± 11 143 ± 9 136 ± 9

Systolic Blood Pressure (mm Hg) 154 ± 14* 149 ± 13 144 ± 11 124 ± 17 133 ± 18 118 ± 14 168 ± 15* 163 ± 12 148 ± 11 150 ± 10 153 ± 9 148 ± 21

134 ± 13 125 ± 18 129 ± 7 146 ± 25

25 nmol DAME Naloxone + 25 nmol DAME 50 nmol DAME Naloxone + 50 nmol DAME

154 ± 12 149 ± 24 147 ± 6 144 ± 5

Heart rate (bpm) 188 ± 13* 182 ± 12* 164 ± 21 158 ± 20 166 ± 8* 193 ± 9* 180 ± 20* 193 ± 29*

146 ± 12 172 ± 23 142 ± 6 149 ± 10

164 ± 10 158 ± 27 160 ± 9 146 ± 9

120 min

Source: Yukimura, J., et al., Hypertension, 3 (5), 528–533, 1981. Data are expressed as means ± SEM. *p < 0.05. a

of 25 nmol of DAME blocked all cardiovascular responses (Table 36.1). A 50-nmol dose blocked only the blood pressure responses; the heart rate increases and baroreceptor reflex attenuations were unaltered. The baroreceptor reflex was attenuated for 15–60 min after DAME; higher doses were effective for a longer time (Table 36.1). Naloxone administered before enkephalin injection produced no changes in central electrical activity. It was concluded that enkephalins may play a role in central mechanisms of cardiovascular control by interacting with opiate receptors in the brain (Yukimura et al. 1981). 36.2.2â•…Action on Mean Arterial Blood Pressure, Heart Rate, and Brain Electrical Activity in Anesthetized Animals The injection of DAME into the cisterna magna of anesthetized dogs induced a short period of moderate hypertension followed by a marked and prolonged decrease in blood pressure, heart rate, and splanchnic nerve discharge (Laubie et al. 1977). Intravenous (i.v.) naloxone produced a transient increase in all three parameters and antagonism of DAME, which was extended by a subsequent injection of naloxone (Figure 36.6). It was concluded that the opioid peptides may be involved in central cardiovascular control. The intracerebroventricular injection of DAME (500 μg/kg) has been shown to produce hypotension, bradycardia, and seizure activity when administered to anesthetized cats (Kraras et al. 1987; Lathers et al. 1988). In general, intravenous naloxone (100 μg/kg) reversed the effects of DAME on blood pressure and heart rate while eliminating seizure activity. Figure 36.7 illustrates the effect of DAME in one cat. Although the mean arterial blood pressure dropped from the control value of 108 to 93 mm Hg, the heart rate increased to 132 bpm 18 min after the administration of DAME compared to the control interval rate of 108 bpm (Figure 36.7b). The subsequent administration of 100 μg/kg (i.v.) naloxone to this cat then decreased the heart rate elevated by the administration of DAME; a further decrease in blood pressure occurred after naloxone. The epileptogenic activity induced by DAME was decreased but not abolished by naloxone (Figure 36.7c). DAME also produced brief ictal activity beginning several minutes postadministration in most cats. Brief ictal activity consists of repetitive bilateral bursts of polyspike activity,

584 Sudden Death in Epilepsy: Forensic and Clinical Issues

200 BP mm Hg

mean

0 200 0

20 s 220

72

180

HR beats/min splanchnic discharges 50 µV

500 ms

[-Ala2] met-enkephalin

control

500 µg/kg ic

20 min

naloxone 100 µg/kg, i.v.

Figure 36.6╇ The inhibitory effect of (d-Ala 2)met-enk (500 μg/kg) injected into the cisterna

108

200

93

86

150 100

Electrocorticogram (µV)

Mean arterial blood pressure (mm Hg)

magna on blood pressure, heart rate, and splanchnic neural discharges on a dog anesthetized with alpha-chloralose and the reversal produced by naloxone (100 μg/kg, i.v.). (Reproduced from Laubie, M., et al., Eur J Pharmacol, 46, 67–71, 1977. With permission.)

50 0

132

Heart rate (beats/min)

106

114

2 sec

0 Control

(a) T = 0 min DAME (500 µg/kg) i.c.v.

T = 18 min (b)

T = 4 min (c) post naxalone

T = 0 min naxalone (100 µg/kg) i.v.

Figure 36.7╇ Brief ictal activity produced by DAME and eliminated by naloxone in cats anesthetized with alpha-chloralose. Heart rate and blood pressure changes are also shown. (From Kraras, C. M., et al., Med Hypothesis, 23, 19–31, 1987. With permission.)

Role of Neuropeptides in the Production of Epileptogenic Activity and Arrhythmias 585

each lasting less than 10 s, interspersed with brief periods of depression of cerebral activity. Administration of naloxone (100 μg/kg, i.v.) eliminated brief ictal activity in some cats within 4 min of its administration, whereas in other cats the seizure activity was somewhat depressed although still present. DAME-induced depression of heart rate and blood pressure was generally reversed by naloxone.

36.3â•…Discussion In the studies of Lathers and Schraeder (1982) and Schraeder and Lathers (1983), in the control period, the mean heart rate increased with a decrease in the mean arterial blood pressure in anesthetized cats. This relationship did not always occur with the development of epileptogenic activity induced by the intravenous administration of PTZ. The autonomic cardiac nerves did not always respond in a predictable manner to changes in blood pressure after the development of epileptogenic activity. In contrast, during the control period, all postganglionic cardiac sympathetic nerves exhibited an increased discharge as blood pressure dropped after the administration of a vasodilating test dose of histamine. Discharge in the parasympathetic cardiac nerves followed the changes in the mean arterial blood pressure. These relationships of cardiac neural discharges (sympathetic and parasympathetic) to blood pressure changes represent the normal physiological function (Bronk et al. 1936). With the development of interictal activity, the variability of mean neural discharge for the parasympathetic nerves began to increase, as demonstrated by an increase in the standard deviation. With greater degrees of epileptogenic activity, the standard deviation continued to increase; that is, the variability in the discharge among the parasympathetic nerves monitored became larger. A neural variability, again evidenced by a large standard deviation, also occurred in the mean postganglionic cardiac sympathetic discharge. The variability observed for the mean sympathetic discharge developed subsequent to that occurring in the parasympathetic discharge. Thus, autonomic cardiac neural dysfunction was observed within both divisions of the autonomic cardiac nervous systems and between the two divisions. The altered cardiac neural discharge was associated with minimal epileptogenic activity (i.e., interictal discharges) and the development of cardiac arrhythmias. The proposed mechanisms involved in the development of these arrhythmias and the possible role of enkephalins in the induction of sudden unexplained death in some epileptic patients are summarized in Figure 36.8 and are discussed below. The injection of enkephalins into the central nervous system elicits seizure activity (Frenk et al. 1978; Snead 1983). Intraperitoneal administration of PTZ produced increases in enkephalin content of the amygdala, striatum, and septum (Vindrola et al. 1983). An increase in the level of enkephalin in the amygdala may have initiated seizure activity because the amygdala is extensively interconnected with the hypothalamus; indeed, it is considered to have a higher-order modulating influence on the hypothalamus. Furthermore, almost any visceral or somatic activity, including cardiovascular and respiratory changes, elicited by stimulating the amygdala can also be elicited by stimulating various areas within the hypothalamus (Nolte 1981). Seizure activity originating in the amygdala may have induced changes in the discharge to the hypothalamus; disturbances in hypothalamic function may result in autonomic dysfunction. PTZ (intraperitoneal) has also been shown to induce an increase in enkephalin levels within the hypothalamus (Vindrola et al. 1984). Because the hypothalamus contains autonomic centers, it may be that the PTZ-induced increases

586 Sudden Death in Epilepsy: Forensic and Clinical Issues PTZ

Concentration of central enkephalin acting on oplate receptors K+-evoked release of GABA and/or, Ca+2 entry into presynaptic GABA nerve terminals and/or, K+ conductance in the GABA nerve terminal indirect Ca+2 entry in the same nerve terminal Anesthetized Animals

Conscious Animals

Sympathetic and parasympathetic central neuron outflow and enhancement of central reflex-induced vagal bradycardia and blood pressure

Inhibition of central GABA release epileptogenic activity

Blood pressure and heart rate due to attenuation of the vagal component of the baroreceptor reflex

Imbalance in peripheral sympathetic and parasympathetic neural discharge Arrhythmia

Sudden unexplained death in the epileptic person

Figure 36.8╇ Postulated mechanism by which central enkephalins could antagonize GABA, resulting in autonomic dysfunction, epileptogenic activity, and sudden death. (From Kraras, C.€M., et al., Med Hypothesis, 23, 19–31, 1987. With permission.)

in central enkephalin levels led to the production of epileptogenic activity and autonomic dysfunction in the experiments of Lathers and Schraeder (1982). A central mechanism by which increased concentrations of enkephalins inhibit K+-dependent GABA release may exist because met-enkephalin has been shown to inhibit the release of GABA from rat brain synaptosomes (Brennan et al. 1980). There is also evidence suggesting that increased concentrations of enkephalins directly decrease the entry of Ca2+ into the presynaptic GABA nerve terminals (Bixby and Spitzer 1983). Metenkephalin reduced the amount of acetylcholine released at the frog neuromuscular junction, most likely by blocking voltage-dependent Ca2+ channels in the presynaptic terminal. However, it is also possible that enkephalin reduced Ca2+ entry indirectly, by increasing K+ conductance in the terminal. Nevertheless, it may be that met-enkephalin acts within the central nervous system by interfering with the K+- and/or Ca2+-dependent mechanism of GABA release (Figure 36.8). Decreased GABA levels are thought to initiate epileptogenic activity (Krnjevic 1980; Ribak et al. 1979). Enkephalins may inhibit the release of GABA by acting on central opiate receptors. Inhibition of GABA release by met-enkephalin was prevented by administration of naloxone (Brennan et al. 1980). Pretreatment with intracerebroventricular naloxone also prevented DAME from inducing changes in blood pressure and heart rate as well as producing seizure activity in conscious cats (Yukimura et al. 1981). Administration of intravenous naloxone after DAME reversed the effects of DAME on heart rate and blood pressure (Laubie et al. 1977). Naloxone (i.v.) had the same action on heart rate and blood pressure and either eliminated or depressed DAME-induced seizure activity in anesthetized cats in the experiments of Kraras et al. (1987) and Lathers et al. (1988). Thus, it may be that

Role of Neuropeptides in the Production of Epileptogenic Activity and Arrhythmias 587

enkephalins act on central opiate receptors to inhibit GABA release because the actions of these agents are blocked by opioid antagonists such as naloxone. Inhibition of GABA release in anesthetized cats produces increased sympathetic and parasympathetic neural outflow and the enhancement of central reflex–induced vagal bradycardia (DiMicco et al. 1979; Gillis et al. 1980). The resultant increased parasympathetic central outflow and enhancement of central reflex–induced vagal bradycardia via the enkephalin-induced inhibition of GABA release, as depicted in Figure 36.8, may explain the decrease in heart rate observed after the intracerebroventricular injection of DAME in the experiments of Kraras et al. (1987) and Lathers et al. (1988). The data of Gillis et al. (1980) suggested that an increase in blood pressure should occur in anesthetized cats after removal of the tonically active GABAergic system present in the brain. However, the intracerebroventricular injection of DAME in our experiments and the injection of DAME into the cisterna magna in anesthetized dogs (Laubie et al. 1977) to inhibit the release of GABA produced hypotension. We hypothesize that the unanticipated drop in blood pressure is due to the ability of the epileptogenic activity to produce autonomic dysfunction, as suggested by the studies of Lathers and Schraeder (1982) and Schraeder and Lathers (1983, 1989) and as indicated by the broken arrow in Figure 36.8. Altered central parasympathetic and sympathetic neural outflow may induce an imbalance within each division as well as an imbalance between both peripheral autonomic divisions that innervate the heart. This imbalance may result in the production of arrhythmia (Lathers et al. 1977, 1978) and/or sudden unexplained death (Carnel et al. 1985; Lathers and Schraeder 1982, 1987; Lathers et al. 1984, 1988; Suter and Lathers 1984). The cardiovascular effects of enkephalins in the central nervous system vary in a dose-dependent manner and according to the state of the animal, that is, conscious versus anesthetized. Met-enkephalin had no effect after intracerebroventricular injection in dogs anesthetized with alpha-chloralose; met-enkephalin has a half-life of several seconds and the lack of a visible effect may be due to its rapid inactivation. However, DAME, a synthetic analogue of met-enkephalin that is metabolically more stable than the natural peptide, produced prolonged hypotension and bradycardia in alpha-chloralose-Â�anesthetized dogs€when injected into the cisterna magna (Laubie et al. 1977). In alpha-Â�chloralose-Â�â•›anesthetized cats, the intracerebroventricular administration of DAME also produced a€ marked€ decrease in€blood pressure with a decrease in heart rate occurring in most animals (Kraras et al. 1987; Lathers et al. 1988). In contrast, a dose-dependent increase in blood pressure and heart rate was induced by administration of (d-Ala 2)met-enk in conscious cats (Schaz et al. 1980). Similar changes were reported by Yukimura et al. (1981) when they injected DAME intracerebroventricularly into conscious cats. Maximal increases in blood pressure occurred approximately 15 min before the greatest increase in heart rate. Blood pressure increased 20 min before the increase in heart rate occurred (Kraras et al. 1987; Lathers et al. 1988). The observation of differences in the physiological effects of DAME, depending on whether a conscious or anesthetized preparation is used, raises the question of whether the presence of the anesthetic agent alpha-chloralose explains the differences found in the two experimental models. Alpha-chloralose has been shown to be a good anesthetic agent for neurological studies because many reflexes, including the baroreceptors, are present and, in fact, enhanced (Clifford and Soma 1969). This anesthetic also causes minimal change in the amount of epinephrine present in the adrenal glands of cats and does not depress cardiac renal discharge (Clifford and Soma 1969; Cox et al. 1936). These data indicate

588 Sudden Death in Epilepsy: Forensic and Clinical Issues

that the baroreceptor mechanism was intact in the anesthetized animals receiving alphaÂ�chloralose and centrally administered DAME in the studies of Kraras et al. (1987), Lathers et al. (1988), and Laubie et al. (1977). The occurrence of epileptogenic activity in conscious cats began 5 min after the administration of (d-Ala 2)met-enk and ended before the changes in cardiovascular parameters (Schaz et al. 1980). Epileptogenic activity was also observed in anesthetized cats to which DAME was administered (Kraras et al. 1987; Lathers et al. 1988), although it began during the ἀrst several minutes after the administration of DAME and continued throughout the duration of the experiments. Because the epileptogenic activity began after the cardiovascular changes in the experiments of Schaz et al. (1980), they concluded that the epileptogenic activity was independent of the autonomic changes. However, there are studies that support the concept that autonomic cardiovascular changes may occur initially and be followed by the development of seizure activity. Indeed, seizure activity in patients was abolished when cardiac arrhythmias were eliminated with the insertion of pacemakers or with the initiation of antiarrhythmic agents (Schott et al. 1977). The possibility that cardiac arrhythmias may result in the development of seizure activity via impaired peripheral cardiac neural discharge going back to the central nervous system is depicted in Figure 36.8 by the heavy arrows beginning with arrhythmia. In addition, the studies of Lathers and Schraeder (1982) and Schraeder and Lathers (1983) found that epileptogenic activity may lead to cardiovascular dysfunction in animals and substantiate earlier observations made in humans. As early as 1941, Penἀeld and Erickson reported a patient with temporal lobe seizures and episodes of tachycardia. Additional studies by Mulder et al. (1954), Phizackerly et al. (1954), White et al. (1961), and Walsh et al. (1968) reported changes in the electrocardiogram in humans that were associated with epileptogenic activity. Thus, Figure 36.8 illustrates how epileptogenic activity may initiate an enhanced autonomic central neural outflow that impairs peripheral cardiac neural discharge and results in the production of arrhythmia. Further experiments are needed to determine whether enkephalins elicit epileptogenic activity and autonomic dysfunction in both conscious and anesthetized animal preparations. It will be important to determine whether enkephalins impair autonomic dysfunction via an inhibition of the release of GABA from the nerve terminal because delineation of this mechanism will allow the design of experiments to evaluate the ability of pharmacologic agents to prevent the enkephalin-induced epileptogenic activity and autonomic dysfunction. Suppression of progression to cardiac arrhythmias induced by the autonomic dysfunction should ultimately decrease the incidence of sudden unexplained death in the epileptic patient.

36.4â•… Summary Autonomic dysfunction, including arrhythmias, is often associated with epileptogenic activity. This study examines the potential role for enkephalins in this process. Brennan et al. (1980) reported a greater percentage of inhibition of K+-stimulated GABA release with increasing concentrations of met-enkephalin. Snead and Bearden (1980) found that leuenkephalin in the central nervous system may induce epileptogenic activity. In addition, DAME has been shown to produce a centrally mediated vasopressor response as well as attenuation of the baroreceptor reflex in conscious cats (Schaz et al. 1980) possibly leading

Role of Neuropeptides in the Production of Epileptogenic Activity and Arrhythmias 589

to autonomic imbalance. The latter may precipitate arrhythmias and be a contributor to sudden unexplained death in the epilepsy. Resolution of the question of whether enkephalins elicit epileptogenic activity and autonomic dysfunction via inhibition of GABA release is important because an understanding of this mechanism should eventually allow the design of pharmacologic agents to prevent the epileptogenic activity, autonomic dysfunction, and associated sudden death.

References Bixby, L., and C. Spitzer. 1983. Enkephalin reduces quantal content at the frog neuromuscular junction. Nature 301: 431–432. Brennan, M., R. C. Cantrill, and B. A. Wylie. 1980. Modulation of synaptosomal GABA release by enkephalin. Life Sci 27: 1097–1101. Bronk, D. W., R. Ferguson, and R. Margaria. 1936. The activity of the cardiac sympathetic center. Am J Physiol 117: 237–249. Carnel, S. B., P. L. Schraeder, and C. M. Lathers. 1985. The effect of phenobarbital pretreatment on cardiac neural discharge and pentylenetetrazol-induced epileptogenic activity. Pharmacology 30: 225–240. Clifford, D. H., and L. R. Soma. 1969. Feline anesthesia. Fed Proc 28: 1479–1499. Cox, W. V., M. E. Lewiston, and H. F. Robertson. 1936. The effect of stellate ganglionectomy on the cardiac function of intact dogs (and its effect on the extent of myocardial infarction and on cardiac function following coronary artery occlusion). Am Heart J 12: 285–300. DeBelleroche, J., and J. Bradford. 1972. Metabolism of beds of mammalian cortical synaptosomes: Response to depolarizing influences. J Neurochem 19: 585–602. DiMicco, J., K. Gale, B. L. Hamilton, and R. A. Gillis. 1979. GABA receptor control of parasympathetic outflow to heart: Characterization and brainstem localization. Science 204: 1106–1109. Elazor, F., E. Motles, Y. Elv, and R. Simantov. 1979. Acute tolerance to the excitatory effect of enkephalin microinjections into hippocampus. Life Sci 24: 541–548. Frenk, H., G. Urca, and J. C. Leibeskind. 1978. Epileptic properties of leucine- and methionineÂ�enkephalin: Comparison with morphine and reversibility by naloxone. Brain Res 147: 327–337. Gillis, R. A., J. DiMicco, D. Williford, B. L. Hamilton, and K. Gale. 1980. Importance of the CNS GABAergic mechanisms in the regulation of cardiovascular function. Brain Res Bull 5 (Suppl. 2): 303–315. Kraras, C. M., N. Tumer, and C. M. Lathers. 1987. The role of neuropeptides in the production of epileptogenic activity and autonomic dysfunction: Origin of arrhythmias and sudden death in the epileptic patient? Med Hypothesis 23: 19–31. Krnjevic, K. 1980. Principles of synaptic transmission. In Advances in Neurology, vol. 27, Antiepileptic Drugs: Mechanisms of Action, ed. G. Glaser, J. Penry, and D. Woodbury, 127. New York, NY: Raven Press. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 27: 633–647. Lathers, C. M., and P. L. Schraeder. 1987. Review of autonomic dysfunction, cardiac arrhythmias, and epileptogenic activity. J Clin Pharmacol 27: 346–356. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1977. Correlation of ouabain-induced arrhythmia and nonuniformity in the histamine-evoked discharge of cardiac sympathetic nerves. J Pharmacol Exp Ther 203: 461–479. Lathers, C. M., G. J. Kelliher, J. Roberts, and A. B. Beasley. 1978. Nonuniform cardiac sympathetic nerve discharge: Mechanism for coronary occlusion and digitalis-induced arrhythmias. CirÂ� culation 57: 1058–1065.

590 Sudden Death in Epilepsy: Forensic and Clinical Issues Lathers, C. M., P. L. Schraeder, and S. B. Carnel. 1984. Neural mechanisms in cardiac arrhythmias associated with epileptogenic activity: The effect of phenobarbital. Life Sci 34: 1919–1936. Lathers, C. M., N. Tumer, and C. M. Kraras. 1988. The effect of intracerebroventricular d-Ala2 methiÂ� onine enkephalinamide and naloxone on cardiovascular parameters in the cat. Life Sci 43: 2287–2298. Laubie, M., H. Schmitt, M. Vincent, and G. Remond. 1977. Central cardiovascular effects of morphinomimetic peptides in dogs. Eur J Pharmacol 46: 67–71. Mulder, D. W., D. Daly, and A. A. Bailey. 1954. Visceral epilepsy. Arch Intern Med 93: 481–493. Nolte, J. 1981. Olfactory and limbic systems. In The Human Brain: An Introduction to Its Functional Anatomy, ed. J. Lotz, 304. St. Louis, MO: C. V. Mosby. Penἀeld, W., and I. C. Erickson. 1941. Epilepsy and Cerebral Localization, 320–362. Springἀeld, IL: Thomas. Phizackerly, P. J. R., E. W. Poole, and C. W. M. Whitty. 1954. Sinoauricular heart block as an epileptic manifestation: A case report. Epilepsia 3: 89–91. Ribak, C. E., A. B. Harris, J. E. Vaughn, and E. Roberts. 1979. Inhibitory GABAergic nerve terminals decrease at sites of focal epilepsy. Science 205: 211–214. Schaz, K., G. Stock, W. Simon, K. Schlor, T. Unger, R. Rockhold, and D. Ganten. 1980. Enkephalin effects on blood pressure, heart rate and baroreceptor reflex. Hypertension 2 (4): 395–407. Schott, G. D., A. A. McLeod, and D. E. Jewitt. 1977. Cardiac arrhythmias that masquerade as epilepsy. Br Med J 1: 1454–1457. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32: 1371–1382. Schraeder, P. L., and C. M. Lathers. 1989. Paroxysmal cardiovascular dysfunction and epileptogenic activity. Epilepsy Res 3: 55–62. Snead, O. C. 1983. Seizures induced by carbachol, morphine, leucine-enkephalin: A comparison. Ann Neurol 13: 445–451. Snead, O. C., and L. J. Bearden. 1980. Anticonvulsants speciἀc for petit mal antagonist epileptogenic effect of leucine enkephalin. Science 210: 1031–1033. Suter, L. E., and C. M. Lathers. 1984. Modulation of presynaptic gamma aminobutyric acid release by prostaglandin E2: Explanation for epileptogenic activity and dysfunction in autonomic cardiac neural discharge leading to arrhythmias? Med Hypothesis 15: 15–30. Urca, G., and H. Frenk. 1983. Intracerebral opiates block the epileptic effect of intracerebroventricular (I.C.V.) leucine-enkephalin. Brain Res 259: 103–110. Vindrola, O., M. Asai, M. Zubieta, and G. Linares. 1983. Brain content of immunoreactive (leu5) enkephalin and (met5)enkephalin after pentylenetetrazol-induced convulsions. Eur J Pharmacol 90: 85–89. Vindrola, O., M. Asai, M. Zubieta, E. Talavera, R. Rodriquez, and G. Linares. 1984. Pentylenetetrazol kindling produces a long lasting elevation of IR-met-enkephalin but not IR-leu-enkephalin in rat brain. Brain Res 297: 121–125. Walsh, G., W. Masland, and E. Goldensohn. 1968. Paroxysmal cerebral discharge associated with paroxysmal atrial tachycardia. Electroencephalogr Clin Neurophysiol 24: 187. White, P. T., P. Grant, J. Mosier, and A. Craig. 1961. Changes in cerebral dynamics associated with seizures. Neurology 11: 354–361. Yukimura, J., G. Stock, H. Stumpf, T. Unger, and D. Ganten. 1981. Effects of (d-Ala2)-methionineenkephalin in blood pressure, heart rate, and baro-receptor reflex sensitivity in conscious cats. Hypertension 3 (5): 528–533.

Sudden Epileptic Death in Experimental Animal Models Ombretta Mameli Marcello Alessandro Caria

37

Epilepsy is one of the most serious and most common neurological diseases. Although epidemiological studies provide evidence that 70–80% of patients who develop epilepsy go into remission after treatment, the remaining are often resistant to the common therapeutic treatments and continue to have seizures (Kwan and Sander 2004). In these patients, sudden unexplained epileptic death (SUDEP) is the most common cause of mortality related to seizures (Hauser et al. 1980; Neuspiel and Kuller 1985; Tennis et al. 1995; Nashef 1999; Nashef and Brown 1996; Leestma et al. 1997; Ficker et al. 1998; Annegers and Coan 1999; Sperling et al. 1999). The risk of SUDEP has been reported to be up to 24-fold greater than in the general population (Leestma et al. 1989; Nashef and Shorvon 1997; Schraeder et al. 2006), the overall incidence being 1:680, that is, 1:100 per year. Recent studies aimed at identifying the risk factors for SUDEP (Nilsson et al. 2001; Walczak et al. 2001; Opeskin and Berkovic 2003; Monte et al. 2007; Nashef et al. 2007) concluded that the most important risk factors are related to being a young adult male, having generalized tonic–clonic seizures, and lying in bed. Of great interest would be analysis of the results of autopsies that should be performed in all patients with a history of epilepsy according to a common standardized protocol for detailed macro- and microscopic analysis of the brain, the autonomic nervous system, the lungs, and the heart. Comparison of the incidence estimates for SUDEP is difficult. In fact, evidence for epileptic seizures immediately before death is reported in 24–80% of patients (Leestma et al. 1997; Langan et al. 2000). Furthermore, not all patients have postmortem examinations. Case ascertainment methods and source populations have varied, and different deἀnitions of SUDEP have been used (Tomson et al. 2005). To clarify this matter and drawing on the opinions of several specialists, SUDEP has been deἀned as a “sudden, unexpected, witnessed or unwitnessed, nontraumatic, and non drowning death in patients with epilepsy, with or without evidence of a seizure, and excluding documented status epilepticus, in which postmortem examination does not reveal a toxicological or anatomical cause of death” (Nashef 1997). Different physiopathological events may contribute to SUDEP, and its genesis is probably multifactorial and includes cardiac arrhythmias induced by epileptic seizures (Nei et al. 2000), neurogenic pulmonary edema (Smith and Matthay 1997), respiratory failure (O’Regan and Brown 2005), and asphyxia (Johnston et al. 1995). In a systematic report of patients undergoing simultaneous EEG and ECG monitoring for temporal lobe epilepsy, the most common ἀnding was sinus tachycardia, occurring in 92% of recordings (Blumhardt et al. 1986). Bradycardia was seen in only 4% of patients. Whether these arrhythmias are primary or secondary to other phenomena, including respiratory changes, is unknown. A recent review by Lathers et al. (2008) extensively analyzed the overlapping mechanisms that may enhance the risk of SUDEP in epilepsy and in cardiac disease. 591

592 Sudden Death in Epilepsy: Forensic and Clinical Issues

Cardiac arrhythmias during both seizures and interictal activity may result in heart failure caused by complete atria-ventricular conduction block (Wilder-Smith 1992). In a reported case of a patient with epilepsy who died unexpectedly while undergoing cardiac monitoring, a nonreversible malignant ventricular arrhythmia occurred, indicating that it was the likely arrhythmogenic cause of SUDEP (Dasheiff and Dickinson 1986). Respiratory complications such as obstruction of air pathways, central apnea, neurogenic pulmonary edema, and metabolic impairments are probably concurrent ἀnal events (Tomson 2000). However, the precise role of these factors in the pathogenesis of SUDEP has yet to be clariἀed despite a number of possibilities examined by epidemiological studies. A signiἀcant contribution to better understanding of the physiopathological mechanisms involved in SUDEP and experimental conἀrmation of clinical observations in human beings has been inferred from basic research studies in animals. Among the hypotheses explored is the involvement of the autonomic nervous system, as seizures may be preceded by autonomic symptoms that are also evident during seizure evolution (Venit et al. 2004; Johnson and Davidoff 1964; Lathers 1990; Lathers and Schraeder 1987; Lathers et al. 1987; Schraeder and Lathers 1983, 1989; Kalviainen et al. 1990; Toichi et al. 1998; Freeman 2006; Sathyaprabha et al. 2006). This fact may be dependent on the propagation of electric activity from the epileptic focus to the autonomic centres (Van Buren 1958). From the analysis of heart rate variability in humans, about 30 s before seizures begin, a signiἀcant reduction of parasympathetic tone along with a signiἀcant increase of sympathetic activity occurs (Novak et al. 1999). On the other hand, during epileptic convulsions in both animals (Doba et al. 1975; Benowitz et al. 1986) and humans (Smith and Matthay 1997), elevations of epinephrine and norepinephrine to potentially arrhythmogenic levels have been documented. However, although few cases of potentially fatal arrhythmias have been recorded in humans with epilepsy (Phizackerley et al. 1954; Liedholm and Gudjonsson 1992), the bulk of electrocardiographic recordings during seizures shows nothing more malignant than sinus tachycardia. In fact, some clinical studies conclude that ventricular arrhythmias are no more common in epileptic than in nonepileptic patients (Keilson et al. 1987). Early studies in a variety of species demonstrated that cardiac arrhythmias could be induced by stimulation of a number of areas in the diencephalon, mesencephalon, and medulla (Allen 1931; Dikshit 1934; Van Bogaert 1936; Boeles et al. 1957; Purpura et al. 1958; Fuster and Weinberg 1960; Weinberg and Fuster 1960; Manning and Peiss 1960; Parker et al. 1962; Ueda 1962; Attar et al. 1963; Melville et al. 1963; Hockman et al. 1966; Gunn et al. 1968; Hall et al. 1974; Lisander et al. 1975; Evans and Gillis 1974, 1978; De Riu 1983; McCown et al. 1984) and may be prevented by cooling the vagus nerve and even by ablation of the stellate ganglia (Manning and Peiss 1960). These experiments helped to clarify the autonomic and reflex mechanisms mediating the post-stimulation-induced arrhythmias elicited by brain stimulation and may explain the autonomic disturbances described in epileptic patients. Pathological studies have raised questions about arrhythmia as the cause for epileptic sudden death. Lathers and Schraeder (1982) and Schraeder and Lathers (1983) developed an experimental model of generalized epilepsy in the cat that used pentylenetetrazol to explore the hypothesis that an altered autonomic function may be one cause for unexplained sudden epileptic death. Their studies showed for the ἀrst time that altered sympathetic and parasympathetic cardiac neural discharges preceded the cardiac arrhythmias that occurred with the development of interictal activity and worsened with increasing degrees of ictal activity. An imbalance within and between sympathetic and parasympathetic cardiac

Sudden Epileptic Death in Experimental Animal Models

593

neural discharges was found to be associated with a signiἀcant disruption of the physiological relationships between heart rate and blood pressure. All animals died after 2000 mg/kg pentylenetetrazol of cardiovascular failure, asystole, or ventricular ἀbrillation. The authors concluded that these cardiac arrhythmias could be a major contributing factor in SUDEP even in those patients who showed only interictal spikes at the time of death and thus did not have overt seizure activity. Their hypothesis was consistent with that of Han and Moe (1964) who demonstrated that cardiac sympathetic disturbances secondary to direct sympathetic nerve stimulation increased temporal dispersion of recovery of ventricular excitability and led to an underlying electrical instability that predisposes the ventricular myocardium to arrhythmia. However, other studies showed that pretreatment with phenobarbital induced a delay of onset of paroxysmal activity, without, however, showing protective effects on the associated autonomic neural changes once the epileptiform discharges were established (Lathers et al. 1984; Carnel et al. 1985). Similar results have been also obtained in rats with epilepsy induced by pilocarpine (Colugnati et al. 2005). The cardiac sympathetic nerves are more often activated by epileptogenic events than are the cardiac parasympathetic nerves, with tachyarrhythmias being far more common clinically than epileptogenic bradycardia (Schernthaner et al. 1999; Leutmezer et al. 2003). Despite these observations, some evidence suggests that epileptogenic activation of the cardiac parasympathetic nerves, which is revealed by ictal bradyarrhythmias or cardiac asystole, might be involved in causing sudden death of epileptic patients (Nashef et al. 1996; Fuhr and Leppert 2000; Weinstein and Albertario 2000; Kelly et al. 2001; Seeck et al. 2001; Tinuper et al. 2001; Mondon et al. 2002). Preganglionic cardiac parasympathetic neurons are primarily located in the nucleus ambiguous and in the dorsal motor nucleus of the vagus; these neurons participate in the control of heart rate and other cardiac functions (Izzo et al. 1993; Standish et al. 1994; Standish et al. 1995; Taylor et al. 1999). Little is known about the ἀring pattern of preganglionic cardiac parasympathetic neurons during an epileptic attack. In a recent study (Wang et al. 2006), a fluorescent tracer was injected into the cardiac sac of newborn rats for retrograde labeling of the parasympathetic neurons in the nucleus ambiguous. Fluorescence-labeled NA neurons were further examined using a whole-cell patch-clamp method in medulla slices with a respiratory-like rhythm, and neurons with an inspiratory-related increase of the mixed inhibitory synaptic activity were identiἀed as preganglionic cardiac parasympathetic neurons. The authors demonstrated that blockade of the GABAergic and the glycinergic receptors in medulla slices evoked intermittent seizure-like ἀring (synchronized intensive firing of a mass of neurons) in the preganglionic cardiac parasympathetic neurons under a current-clamp conἀguration and evoked intermittent excitatory inward currents under a voltage-clamp conἀguration. Results of this study, therefore, provided new evidence that preganglionic parasympathetic neurons might ἀre in a seizure-like pattern of activity (Wang et al. 2006). Therefore, during an epileptic attack, these neurons may be responsible for neurogenic ictal bradyarrhythmias, cardiac asystole, or even sudden death in epileptic patients. Convulsive seizures triggered by maximal electroshock also induced a severe disruption of cardiac rhythm in rats. In particular, in the immediate postictal state, marked cardiac arrhythmia often appears, the duration of which relates to the seizure activity (Darbin et al. 2003). These data seem to support the hypothesis that cardiac arrhythmias may be an important risk factor for SUDEP in epileptic patients (Oppenheimer et al. 1990), although in the opinion of several clinicians, the role in human SUDEP remains uncertain (Keilson et al. 1987).

594 Sudden Death in Epilepsy: Forensic and Clinical Issues

Animal models of generalized epilepsy do not help, however, to localize the cortical structures that spread paroxysmal stimulation to forebrain areas involved with cardiovascular regulation that are able to desynchronize sympathovagal cardiac neural ἀring and possibly induce SUDEP. Furthermore, a few studies during human brain surgery have investigated the cardiovascular effect of cortical stimulation. In this regard, stimulation of the cingulate gyrus, tips of the temporal lobes, and orbitofrontal cortex induced both pressor and depressor responses (Chapman et al. 1949; Pool and Ransohoff 1949; Chapman et al. 1950; Delgado 1960), and more recently, superἀcial insular stimulation in patients undergoing surgery for intractable epilepsy-induced cardiovascular changes (Oppenheimer et al. 1991). Emotional stress has been shown to increase the frequency and the severity of ventricular ectopic beats in patients with or without ischemic cardiac disease, suggesting a role for cortical regions connected with the limbic structures (Lown et al. 1976; Lown and DeSilva 1978; Taggart et al. 1973; Lathers and Schraeder 2006). Experimental results in animals showed an area of cardiac representation within the posterior conἀnes of the rat€insular cortex (Oppenheimer and Cechetto 1990), in an area known to have profuse reciprocal connections with the limbic system (Zhang and Oppenheimer 2000). Using a novel technique of phasic microstimulation linked to the R wave of the ECG, Oppenheimer and Cechetto (1990) ἀrst€demonstrated a cardiac chronotropic map of the insula. They identiἀed sites generating pure tachycardia independent from other autonomic or respiratory effects within the rostral posterior insula. More caudal sites within this region generated bradycardia after phasic microstimulation. Moreover, prolonged phasic microstimulation within the rat insular cortex resulted in bradyarrhythmia, complete heart block, QT interval prolongation, ventricular ectopy, and asystolic death (Oppenheimer et al. 1991). These parasympathetic arrhythmias were accompanied by elevated plasma norepinephrine levels, increased sympathetic tone and myocytolisis, a form of cardiac damage of sympathetic neural origin, and subendocardial hemorrhages. On the other hand, left anterior insular damage by stroke is associated, in some patients, with signiἀcant tachyarrhythmias, and in the rat with a reduction in baroreflex sensitivity and resulting parasympathetic tone (Oppenheimer et al. 1996; Zhang et al. 1998). The rat insula receives taste information and gastrointestinal stimuli, respiratory afferents, and chemoreceptor and cardiovascular inputs organized in a viscerotopic fashion. This area has reciprocal connections with the parabrachial nucleus, the contralateral insula, adjacent cortical regions, the infralimbic cortex, the thalamus, the lateral hypothalamic area, and the amygdala. Evidence exists for lateralized effects, although nonmyelinated transcallosal pathways (both inhibitory and excitatory) linking the cardiovascular regions of the two insulae have been recently described (Zhang and Oppenheimer 2000). The left insular cortex is involved in the regulation of the vagal cardiac parasympathetic neuronal pool, and the right insular cortex regulates sympathetic neurones involved in the regulation of cardiac function and in the control of vascular resistance and blood pressure. This structure may therefore play a crucial role in brain–heart interaction and perhaps even additionally by its direct and reciprocal ipsilateral connections with the amygdala. The amygdala represents an important central cardiovascular control structure within the limbic system because it seems to provide the neural basis for cardiovascular responses to stressful stimuli (Cheung et al. 1997). In agreement with this latter consideration, preemptive low-frequency sine wave stimulation of amygdala-kindled animals induced a dramatic decrease in the incidence of stage 5 seizures in fully kindled

Sudden Epileptic Death in Experimental Animal Models

595

animals, demonstrating the possibility that low-frequency sine wave stimulation may be an effective therapy for prevention of seizures in patients with epilepsy (Goodman et al. 2005). The amygdala is a nucleus of the temporal lobe, showing complex interconnections with multiple cortical and brainstem regions involved in the production of autonomic temporal lobe epilepsy. An altered postganglionic cardiac sympathetic innervation that may increase the risk of cardiac abnormalities and/or SUDEP has been described in patients with chronic temporal epilepsy (Druschky et al. 2001). In fact, after temporal lobe epilepsy surgery, a reduction of sympathetic cardiovascular modulation has been demonstrated (Hilz et al. 2002). In a rat kindled seizure model, in which paroxysmal activity is similar to that described during secondary spontaneous seizures, amygdala involvement has been demonstrated in seizure-related autonomic disturbances that involved both branches of the autonomic nervous system (Goodman et al. 1990, 1999). Changes in heart rate and blood pressure were observed during amygdala-kindled seizures, but they were apparently independent of the kindling stimulus because they did not appear at the beginning of the kindling process. On the other hand, microdialysis experiments in conscious kindled rats showed decreases in extracellular concentrations of noradrenaline and dopamine levels in the dorsomedial nucleus of the hypothalamus (Goren et al. 2003), suggesting that autonomic changes in kindling require further studies to explain their relationship to SUDEP. Instead, after insular involvement by ischemic stroke or excitatory injury, increased immunostaining for neuropeptide Y, leucine-enkephalin, dynorphin, and neurotensin was seen in the amygdala of the damaged side (Allen et al. 1995; Cheung and Cechetto 1995; Cheung et al. 1995). Because these neurochemical changes in the amygdala were greatest 3 days after ischemic stroke and subsided by 10 days (Cheung et al. 1995), it can be hypothesized that they participate in mediating the cerebrogenic cardiovascular disturbances originating from the insula. The role of neuropeptides in the origin of arrhythmias and SUDEP in epileptic patients has been analyzed in some studies, and the results suggested that prostaglandin E2 and enkephalins may play a signiἀcant role in the genesis of autonomic dysfunction associated with seizure activity and cardiac arrhythmias (Suter and Lathers 1984; Kraras et al. 1987; Lathers et al. 1985, 1988; Lathers 1990; Schwartz and Lathers 1990). In rats, pentylenetetrazol kindling produces a long-lasting elevation of IR-Met-enkephalin in the septum, hypothalamus, amygdala, and the hippocampus after convulsions (Walczak et al. 2001) and is associated with a signiἀcant inhibition of potassium-stimulated GABA release (Brennan et al. 1980). The enkephalins may elicit epileptogenic activity and autonomic dysfunction by inhibiting GABA release. This possibility has been recently conἀrmed by a study in mice showing that a selective cyclooxygenase-2 inhibitor potentiates the anticonvulsant activity of tiagabine against pentylenetetrazol-induced convulsions (Dhir and Kulkarni 2006). Pathological studies in cases of SUDEP have raised questions about pulmonary complications, as autopsies in patients who died from epileptic sudden death revealed pulmonary edema (Terrence et al. 1981; Leestma et al. 1989; Earnest et al. 1992). In anesthetized and ventilated animals, Johnston et al. (1996) showed elevated left atrial and pulmonary vascular pressures that represent a possible mechanism for the pulmonary edema found in patients who have died from SUDEP. However, the pulmonary edema alone does not appear to be fatal because lung congestion requires some time to develop. In an interesting model of SUDEP in which pulmonary edema and sudden death were generated (Johnston et al. 1995), the authors suggested that arrhythmia was not an important

596 Sudden Death in Epilepsy: Forensic and Clinical Issues

factor in SUDEP as they demonstrated a primary role for hypoventilation in the etiology of epileptic sudden death. Their model of SUDEP used chronically monitored unanesthetized sheep with generalized tonic–clonic status epilepticus induced by bicuculline. Some animals died within 5 min after the induction of seizures, although no signiἀcant differences were found in the epileptic activity, in arrhythmias, and in plasma epinephrine and norepinephrine concentrations compared to the group living the longest. Furthermore, in no instance could the death of an animal be ascribed to a malignant rhythm. On the contrary, marked differences in ventilation were found between those convulsing animals that died suddenly and those that survived. A sudden drop in pO2, a marked elevation in pCO2, and a decline in arterial blood pH were observed in animals that died, whereas animals that survived maintained ventilation and oxygenation at levels close to baseline. Hypoventilation in these experiments, probably centrally induced and indistinguishable from apnea, was independent of preterminal arrhythmias or hypotension and was proposed as the cause of death. In agreement with this experimental ἀnding, hypoventilation and central apnea have been documented in several patients with seizures (Nashef and Shorvon 1997) or after stimulation involving various elements of the limbic system (Nelson and Ray 1968). However, in these experimental animals, respiratory function was not directly measured so that central and obstructive hypoventilation could not be distinguished. Extending their studies, Johnston et al. (1997) monitored awake epileptic sheep with tracheostomies that allowed measurements of airway flow and prevented upper airway obstruction. These experiments proved that central apnea and hypoventilation were causes of death in the sheep. Endogenous opioids that may be released during seizures have been implicated in the pathogenesis of this central hypoventilation (Ramabadran and Bansinath 1990) and could well account for the suppression of respiratory drive, although why this should happen is unclear (Darnell and Jay 1982). Changes in respiratory-modulated neural activities, consistent with obstructive and central apnea, have been conἀrmed during seizures in an in situ anesthetized rat preparation (St. John et al. 2006). This preparation has an advantage over in vivo preparations in that delivery of oxygen to the brain is not dependent on the lungs or the cardiovascular system. The EEG activity was recorded, as was activity of the hypoglossal, vagus, and phrenic nerves. The hypoglossal and vagus nerves innervate muscles of the upper airway and larynx, and the phrenic nerve innervates the diaphragm. Seizures were elicited by injections of penicillin into the parietal cortex or the carotid artery. Results showed that after elicitation of the seizures, activity of the hypoglossal and vagal nerves declined greatly, whereas phrenic activity was slightly affected. The authors concluded that such a differential depression of activity of nerves of the upper airway and larynx compared to that of the diaphragm would predispose to obstructive apnea in intact preparations. Moreover, given more time, even the activity of the phrenic nerve declined or ceased, resulting in changes that are known to characterize central apnea. The major conclusion of this study is that seizures may result in recurrent periods of obstructive and central apnea that may account for SUDEP. Recently, severe postictal laryngospasm, as observed in a patient with refractory epilepsy during monitoring, has been described as a potential mechanism for SUDEP (Tavee and Morris 2008). Partial seizures may be associated with prominent oxygen desaturation and ictal ventilatory dysfunction, which could play a role in certain cases of SUDEP in adult patients (Blum et al. 2000). Experimental studies conducted in DBA/2 mice that exhibit sudden death due to respiratory arrest (Collins 1972; Willott and Henry 1976) in the period

Sudden Epileptic Death in Experimental Animal Models

597

immediately after audiogenic seizure have shown that oxygenation prevents sudden death (Venit et al. 2004). The function of respiratory neurons is largely controlled by speciἀc neurotransmitters, among which serotonin has been implicated as modulating respiratory responses to hypoxia (Feldman et al. 2003; Mitchell et al. 2001). However, serotonin receptors have also been implicated in the modulation of seizures (Browning et al. 1997; Pericic et al. 2005) and serotonin agents (fluoxetine) reduced respiratory arrest in DBA/2 mice at doses that did not reduce seizure severity (Tupal and Faingold 2006). The physiopathological mechanisms of SUDEP have also been analyzed in a series of studies on focal epilepsy induced in hemispherectomized rats. In this experimental animal model, independent of cerebral cortex influence, cardioarrhythmogenic triggers may be activated by paroxysmal activity focally induced at hypothalamic, midbrain, and hindbrain levels by topical application of penicillin G (Mameli et al. 1988, 1990). The cardioarrhythmogenic potential of these epileptic foci decreased in the rostrocaudal direction; being more relevant the cardiovascular impairments observed as a consequence of the hypothalamic epileptic focus ἀring, particularly during simultaneous coactivation of the mesencephalic focus (Mameli et al. 1993). Cardiovascular impairments were characterized by sinus bradyarrhythmias, alterations of the repolarization and atriaventricular blocks of different degrees, and signiἀcant decreases in systolic blood pressure (Mameli et al. 1988, 1989, 1990, 1993) temporally correlated to a signiἀcant increase in spontaneous vagal nerve activity. The cardiac arrhythmias became life-threatening when blood gases and electrolytic parameters were simultaneously impaired (Mameli et al. 2001). However, when paroxysmal activity ended, the cardiovascular alterations always disappeared, and vagal nerve ἀring returned to basal values. This demonstrated that in supported ventilation, and in the absence of metabolic derangements, the cardioarrhythmogenic trigger activation was not sufficient to explain SUDEP. These experiments suggested to the authors that fatal evolution consequent to heart impairment was probably related not only to cardiac dysfunction of autonomic origin but also to concomitant metabolic derangement that most likely shared the same genesis. Further experiments, performed using the same experimental animal model, tested the existence of pulmonary complications during the activation of the central cardioarrhythmogenic triggers (Mameli et al. 2006). In the hemispherectomized and artiἀcially ventilated animals, the following parameters were simultaneously analyzed before, during, and after epileptic foci induced both at hypothalamic and mesencephalic levels: spontaneous electrical activity of the hypothalamic neurons, electrothalamogram, spontaneous multiunit vagal nerve ἀber activity, systemic artery blood pressure, pulmonary artery blood pressure, dynamic ECG, and blood gas analysis (pO2, pCO2, bicarbonate, sodium, potassium, Hgb concentrations, pH value, O2% saturation, and bases excess). Metabolic derangements developing throughout the experiment were not purposely adjusted, and only body temperature was kept constant. This study showed that after hypothalamic epileptic focus and mesencephalic epileptic foci, the paroxysmal activity induced, within a short latency, a signiἀcant increase of spontaneous vagal nerve ἀring that was strictly correlated to ECG impairments such as wandering pacemaker, biphasic or negative P waves, extrasystolia, A-V blocks, derangement of A-V conduction and recovery phase, bundle branch blocks, bradyarrhythmias, flattened T waves, and hypotension. Together with a parasympathetic hypertonicity, a concomitant functional imbalance of the orthosympathetic division also developed. When paroxysmal activity began, despite that vagal activity signiἀcantly increased and corresponding bradycardia was expected, heart rate remained around its basal value, sometimes even

598 Sudden Death in Epilepsy: Forensic and Clinical Issues

showing a tendency to increase (Figure 37.1c). Under the same circumstances, supraventricular and ventricular extrasystoles and ventricular complexes of high voltage appeared concomitantly, which also strongly suggested a possible increase in cardiac inotropism that was dependent on orthosympathetic involvement (Figure 37.2d). In the surviving animals (approximately 75%), when paroxysmal activity ended, vagal nerve activity and cardiovascular parameters returned to basal conditions. Macro- and microscopic examination of their lungs never showed any alterations of pulmonary parenchyma, pulmonary vessels, or the bronchial tree. In the deceased animals (approximately 25%) that had manifested interictal and ictal activity, spontaneous vagal nerve ἀring showed similar electrophysiological features, as well as a time course that overlapped the one observed in animals that survived. However, as previously observed (Mameli et al. 2001), the occurrence of cardiac arrhythmias was always accompanied by hyperkalemia and other metabolic derangements, followed by death, which occurred after 3 to 4 h of paroxysmal activity (Figures 37.1 and 37.2). Systemic hypotension was accompanied by signiἀcant pulmonary hypertension, similar to what was described by Johnston and coworkers (1996) in sheep, that further worsened during ictal activity (50% increase). Postmortem macroscopic examination of the lungs demonstrated no relevant alterations in any case. However, histological preparations demonstrated a slight alveolar and perivascular edema in the subinterstitial spaces of the arterial vessels and a concomitant edematous inἀltration in the alveolar and bronchial spaces. Finally, the bronchial tree was ἀlled with considerable intraluminal mucous secretions. Comparison between surviving and deceased animals during ictal activity showed signiἀcant differences in the deceased animals in heart rate, systolic and diastolic pressures, pulmonary artery pressure, potassium and bicarbonate concentrations, pH value, pO2, and pCO2. Regarding the pattern of ECG impairment, during interictal activity, no signiἀcant differences were observed, whereas during ictal activity, signiἀcant differences were observed in the incidence of sinoatrial blocks, bundle branch blocks, and T-wave impairment. In addition, signiἀcant differences were found in the histological patterns of the lungs, which were most likely caused by a state of pulmonary hypertension concomitant with paroxysmal activity and parasympathetic overflow (Figure 37.3). In the opinion of Mameli et al. (2001), a possible explanation of this observation was that at the level of the respiratory apparatus, the paroxysmal activity triggered an increase of the parasympathetic tone, inducing bronchoconstriction and increased mucous secretion. Bronchoconstriction determined, in turn, a reduction of alveolar ventilation with a consequent fall in pO2 as conἀrmed by blood gas data. Furthermore, the decrease in pO2 could have triggered a reflex constriction of pulmonary vessels. Although the mechanism of this phenomenon is still unknown, it is possible that it is induced by a local reflex whose functional signiἀcance is to shunt blood flow from hypoventilated to normally ventilated zones. In the deceased animals, the parasympathetic hypertonicity prejudiced the normal ventilation of the pulmonary parenchyma, although the animals were artiἀcially ventilated. The parasympathetic hypertone caused, therefore, a hypoxic condition that probably extended to the entire lung and in turn caused a reflex vessel constriction and pulmonary artery blood pressure increase. Pulmonary hypertension then resulted in perivascular edema and the edematous inἀltration in the alveolar and bronchial spaces observed in histological preparations (Mameli et al. 2006). Disruption of pulmonary function has been repeatedly reported both in animal and in human epilepsy (Nelson and Ray 1968; Terrence et al. 1975; Bayne and Simon 1981; Harper et

Sudden Epileptic Death in Experimental Animal Models

599

B

A 1

28583 counts

300

2

19724 counts

300

(Scale = 46.22 s/div)

(Scale = 46.22 s/div)

3

C

D

1

56372 counts

38374 counts 300

300

2

3

(Scale = 46.22 s/div)

(Scale = 46.22 s/div) a b c d

Figure 37.1╇ Simultaneous recordings of spontaneous vagal nerve activity and ECG in a deceased animal. Trace 1: spontaneous electrical activity of multiunit vagal nerve fibers recorded using tungsten in glass electrodes. Trace 2: frequency distribution histograms of the same activity constructed during 231.1 s analysis. Trace 3: electrocardiogram. All the events were simultaneously recorded in basal conditions (a) and after the activation of hypothalamic epileptic focus and mesencephalic epileptic foci at 60 (b), 120 (c), and 130 min (d), respectively. Traces a–d: impairment of the ECG, 130 min after hypothalamic epileptic focus and mesencephalic focus induction. The concomitant imbalance of the orthosympathetic function can be indirectly inferred by analyzing the parasympathetic activity. In fact, despite a reduction in vagal nerve firing (b) compared to basal values (a), heart rate did not increase and ventricular extrasystoles appeared. On the other hand, when the vagal nerve firing markedly increased (c) and a corresponding bradycardia was expected, the heart rate also remained unaffected. Bradyarrhythmias appeared only after 130 min of epileptic foci activity. (From Mameli, O., et al., Seizure, 10 (4), 269–278, 2001. With permission.)

600 Sudden Death in Epilepsy: Forensic and Clinical Issues A

1

B

2

3 2547 counts

100

5816 counts

100 50 50

50 4

(Scale = 45.15 s/div)

(Scale = 45.15 s/div)

5

C

D

1 2

3 100

9225 counts

11322 counts

50

50 4

100

(Scale = 45.15 s/div)

(Scale = 45.15 s/div)

5 (Scale = 500.9 ms/div)

Figure 37.2╇ Simultaneous recordings of electrothalamogram, vagal nerve firing, and ECG in

a deceased animal. Trace 1: electrothalamographic signals were recorded using a pair of silver ball electrodes positioned on the thalamic surface. Recordings were performed with a Grass 7P5 and 7DA polygraph. Trace 2: the same event simultaneously analyzed by a computer (Tecfen Computer Scope Analysis ISC-16 software). Trace 3: spontaneous electrical activity of multiunit vagal nerve fibers. Trace 4: frequency distribution histograms of the same activity during 231.1 s analysis. Trace 5: Electrocardiogram. All the events were simultaneously recorded in basal conditions (a) and after the activation of both hypothalamic epileptic focus and mesencephalic epileptic foci at 40 (b), 60 (c), and 120 min (d), respectively. The concomitant imbalance of the orthosympathetic division is shown by the appearance of ventricular complexes of high voltage (c–d). (From Mameli, O., et al., Seizure, 10 (4), 269–278, 2001. With permission.)

Sudden Epileptic Death in Experimental Animal Models

601

al. 1984; James et al. 1991; Graham 1992; Hanning and Alexander-Williams 1995) to worsen the neurogenic arrhythmias, together with metabolic impairment. In agreement with the observations in sheep (Johnston et al. 1996, 1995, 1997), the ἀndings obtained in hemispherectomized rats (Mameli et al. 1988, 1989, 1990, 1993, 2001, 2006) showed that when considered alone, neurogenic arrhythmias are insufficient to cause the animals’ death. However, it is not clear why, given that the same experimental protocol was applied to all animals, only some of these developed clinical symptoms so severe that survival was unfailingly compromised. Because paroxysmal activity involved the same cerebral structures, it seems difficult to hypothesize that activation of a fatal trigger responsible for animal death acts only in some animals. A more convincing hypothesis to explain these ἀndings could be the existence of animals characterized by a low threshold of excitability, whose central structures are unable during epileptic seizure to maintain a balanced ratio between sympathetic and parasympathetic activities. In particular, orthosympathetic activation could paradoxically synergize the vagal influence on heart activity and transform a serious, but reversible, cardiac bradyarrhythmia into a fatal event. This hypothesis coÂ�Â� incides with the ἀnding of increased concentrations of circulating catecholamines during epileptic seizures to potentially arrhythmogenic levels in both animals (Johnston et al. 1995; Doba et al. 1975; Benowitz et al. 1986) and human beings (Simon et al. 1984). At a peripheral level, in the animals with a low threshold of excitability, these high epinephrine levels could activate the alpha-membrane receptors that are responsible for cellular potassium depletion (Williams et al. 1984) at muscular cell levels (Mauger et al. 1982), with a consequent increase in their plasma concentration. Stimulation of alpha-receptors impairs extrarenal potassium disposition, a speciἀc effect that can be reversed by alpha-blockade (Williams et al. 1984). Although alpha-receptor stimulation is known to account for a transient release of hepatic potassium reserves in response to an epinephrine infusion, an effect that occurs within minutes and lasts briefly (Craig and Mendell 1959; Ellis 1956; Vick et al. 1972; Sterns et al. 1981; Giugliano et al. 1979; Guerra and Kitabchi 1976; DeFronzo et al. 1980; Minaker and Rowe 1982), a sustained effect may be observed in a continuous stimulation, as occurs during seizures. Therefore, these effects on potassium availability probably result from a decreased net cellular uptake of potassium, which depends on an enhanced alpha-adrenergic activity in those animals with a low threshold of excitability. This condition would have worsened the vagal-mediated cardiac arrhythmias to a point capable of inducing their fatal evolution into SUDEP. The short latency of hyperkalemia observed in the deceased animals (Mameli et al. 2001, 2006) seems to support this hypothesis and provides evidence for a likely neurogenic extrarenal origin of the phenomenon. Further evidence is suggested in these experiments by the appearance of sharpened T waves that were always related to the increased potassium plasma concentration. These results agree with the severe changes in potassium plasma concentrations detected during generalized seizures in adolescent baboons (Mello et al. 1993). Finally, extending the above considerations to human epilepsy, it can be postulated that patients with a low threshold of excitability may also exist. In these subjects, the cardioarrhythmogenic triggers, activated by hypothalamic and mesencephalic epileptic foci (hypothalamic epileptic focus and mesencephalic focus), may represent the ἀnal common pathway of a storm of signals originating at the level of cortical epileptic foci. In fact, the extensive neural connections, originating in the cerebral cortex and particularly in the frontal and temporal lobes, converge on hypothalamic and brain stem structures (Papez

602 Sudden Death in Epilepsy: Forensic and Clinical Issues 350 300

(a)

250

Basal conditions

200

p = ns

150 100

Surviving Deceased

Deceased

ASP

ADP

PAP

MUAV

HEF -MEF

121±9.8

80±4.5

13.3±1.4

22.88±4.9

0.0

335.1±10.9

121.34±8.4

82±10

14.58±1.8

19.4±6.5

0.0

Na

+

HCO -

pH

pO2

pCO2

Hb

136.4±4.5

3.45±0.44

27.36±3.3

7.38 ±0.16

93.1±10.3

40.6±3.1

14.9±3.4

136 ±6

3.6±0.4

29.2±2.8

7.39±0.6

95 ±4

40.5 ±2

14.6 ±3.7

K

3

Hb

pCO2

pO2

pH

3

K+ HCO-

Na+

MUAV

HEF-MEF

HR

336.2±17.5

+

Surviving

ADP

t test

PAP

HR

0

ASP

50

350 300

(b) Interictal activity

p < 0.005

250

*

200 150

*p = ns

*

100

*

Surviving Deceased

Surviving Deceased

HR 325±4.3 238.3±43.7

ASP 114.5±5.8 109.3±6.9

ADP 76.6±4.6 64.4±5.3

MUAV 156.7±7 159.9±50.4

PAP 11.2±4.4 20.7±3.6

Hb

pCO2

pH

pO2

K+ HCO-3

Na+

MUAV

*

HEF-MEF

ADP

PAP

HR

0

ASP

50

HEF -MEF 0.097±0.02 0.102± 0.17

Na+

K+

pH 7.40±0.09

pCO2

3.59±0.24

HCO3-26.09±3.4

pO2

137.3±2.4

92.1±7.7

41.7±1.5

Hb 15.1±1.3

135 ±0.6

7.2±1.8

25.8 ±1.9

7.20±0.17

66.7±9.1

44.8 ±3.6

15.5 ±0.9

450 * 400 Figure 37.3╇ Main metabolic, cardiovascular, and respiratory parameters determined in (a) 350 = ns basal conditions and during (b) interictal and (c) ictal activities in surviving and*pdeceased ani300 + (c) p < 0.005 mals. All values are expressed as means ± standard deviation. Sodium (Na ), potassium (K+), 250 − bicarboÂ�nateIctal (HCO as milliequivalent per liter. Hemoglobin (Hb) activity 3 ), and pH values are expressed 200 values are expressed as grams per milliliter. arterial systolic (ASP), 150 pO2 and pCO2 pressures, * arterial dystolic pressures (PAP) are expressed as millimeters of * t test (ADP), and pulmonary artery 100 mercury. Heart rate (HR) values are expressed 50 as beats per minute. Multiunit electrical activity 0 of vagal nerve fibers (MUAV) as well as the electrical activity of hypothalamic epileptic focus

Deceased

325±4.3 238.3±43.7 +

114.5±5.8 109.3±6.9 +

156.7±7 159.9±50.4

0.097±0.02 0.102±0.17

Surviving

Na

K

pCO2

3.76±0.45

24.9±2.8

pH 7.36±0.2

pO2

136.4±6.2

88 ±9.3

42.4±2.7

Hb 15.8±0.7

Deceased

136.5±2

8.6 ±4

13.6 ±4

7.18 ±0.1

47.8 ±17

49.9 ±5.3

16.07 ±2.4

Hb

pCO2

pH

pO2

3

K+ HCO-

Na+

HEF-MEF

PAP

MUAV

11.2±4.4 20.7±3.6

76.6±4.6 64.4±5.3

HCO3-

ADP

HR

Surviving

ASP

* and mesencephalic focus cardioarrhythmogenic trigger values are expressed as the number of spikes per second. Statistical significance of differences between surviving and deceased animals was analyzed by paired t tests. (From Mameli, O., et al., Seizure, 15 (5), 275–287, 2006. With permission.) HR ASP ADP PAP MUAV HEF -MEF

ADP 76.6±4.6 64.4±5.3

MUAV 156.7±7 159.9±50.4

PAP 11.2±4.4 20.7±3.6

HCO -pO pH Sudden Epileptic Na Death in KExperimental Animal Models

Surviving Deceased

HR 325±4.3 238.3±43.7 +

ASP 114.5±5.8 109.3±6.9 +

44.8 ±3.6

15.5 ±0.9

*

* *

MUAV 156.7±7 159.9±50.4

PAP 11.2±4.4

20.7±3.6

HEF -MEF 0.097±0.02 0.102±0.17

Surviving

Na

K

pCO2

3.76±0.45

24.9±2.8

pH 7.36±0.2

pO2

136.4±6.2

88 ±9.3

42.4±2.7

Hb 15.8±0.7

Deceased

136.5±2

8.6 ±4

13.6 ±4

7.18 ±0.1

47.8 ±17

49.9 ±5.3

16.07 ±2.4

3

Hb

*

ADP 76.6±4.6 64.4±5.3

HCO -

*p = ns p < 0.005

pO2

t test

66.7±9.1

pCO2

(c) Ictal activity

7.20±0.17

603

3

450 400 350 300 250 200 150 100 50 0

HEF-MEF

25.8 ±1.9

Hb 15.1±1.3

PAP

7.2±1.8

41.7±1.5

MUAV

135 ±0.6

pCO2

92.1±7.7

ADP

26.09±3.4

7.40±0.09

2

3

3.59±0.24

HR

Deceased

+

137.3±2.4

ASP

+

Surviving

HEF -MEF 0.097±0.02 0.102± 0.17

pH

ASP 114.5±5.8 109.3±6.9

K+ HCO-

Deceased

HR 325±4.3 238.3±43.7

Na+

Surviving

Figure 37.3╇ (Continued)

1937; Wall and Davis 1951; Langan et al. 2000; Landau 1953; Walker 1966; Gray 1973; Saper et al. 1976; Korner 1979; Willis and Grossman 1981; Breusch 1984; Natelson 1985). Therefore, the possibility cannot be excluded that under certain circumstances the cortical signals activate the arrhythmogenic triggers simultaneously, thus inducing cardiopulmonary and metabolic impairments that in these patients with a low threshold of excitability may result in sudden death. As for the existence of individuals with a low threshold of excitability, it may be of some interest to consider studies that investigated genetic conditions related to SUDEP. In double-mutant mice that express GM3 as their major ganglioside, a “sudden death phenotype” has been described that was extremely susceptible to induction of lethal seizures by a sound stimulus (Kawai et al. 2001). The gangliosides are a family of glycosphingolipids that contain sialic acid and, although they are abundant on neuronal cell membranes, their speciἀc functions in the CNS remain largely undeἀned. Results of this study showed that these compounds play essential roles in the proper functioning of the CNS and that their absence may contribute to SUDEP susceptibility. In other transgenic mice, the Sema TG, characterized by cardiac-speciἀc overexpression of Sema3a, reduced sympathetic innervation and attenuation of epicardial-to-endocardial innervation gradient have been described. These mice show susceptibility to ventricular tachycardia due to catecholamine supersensitivity and prolongation of the action potential duration, which can terminate in SUDEP. Results of this study showed that appropriate cardiac Sema3a expression is needed for proper sympathetic heart rate control (Ieda et al. 2007). Another interesting hypothesis for the risk of SUDEP considers the neurogenesis process in the CNS. It has long been believed that in mammals, the origin of new neurons in most CNS regions was a process limited to embryogenesis (Hilz et al. 2002) because once development is complete, the progenitor cells that mature into neurons go through a differentiation process and become incapable of division. In contrast, a neurogenesis process in the CNS of adults has been described in several species such as crustaceans (Hilz et al. 2002), reptiles (Lopez-Garcia et al. 1988), amphibians (Polenov and Chetverukhin 1993),

604 Sudden Death in Epilepsy: Forensic and Clinical Issues

birds (Nottebohm 1989), rodents (Altman and Das 1965), primates (Eckenhoff and Rakic 1988), and human beings (Eriksson et al. 1998). In all the mammalian species already studied, including humans, the mitotically active progenitor cells, capable of generating new neurons in the adult phase, are located in speciἀc regions (Eriksson et al. 1998; Gould et al. 1998) such as the subventricular zone of the lateral ventricles and the dentate gyrus of the hippocampal formation. In adult mice, these cells migrate to the olfactory bulb, where they differentiate into a large variety of cell types, such as periglomerular neurons and interneurons, as well as astrocytes and oligodendrocytes (Lois and Alvarez-Buylla 1994). In the rat, these new cells proliferate and migrate continuously into the granular cell layer (Kuhn et al. 1996) where they develop a morphology typical of granule cells (Cameron et al. 1993), express neuronal differentiation markers (Kuhn et al. 1996), and extend their axons to the pathway of mossy ἀbers that project into the CA3 region of the hippocampus (Stanἀeld and Trice 1988). Several brain insults, including epileptogenic activity associated with epilepsy, are able to stimulate progenitor cell proliferation in the dentate gyrus (Parent 2007; Zhao et al. 2008). Using the pilocarpine model of temporal lobe epilepsy, for the ἀrst time in rats, this status epilepticus has been shown to cause a dramatic and prolonged increase in cell proliferation in the dentate subgranular proliferative zone (Parent et al. 1997). This observation has been conἀrmed in humans and adult rodent models of temporal lobe epilepsy (Parent et al. 2006; Mello et al. 1993), showing the possibility that these cells aberrantly migrated after the status epilepticus and were abnormally integrated. Resultant hyperexcitability may contribute to seizure generation and/or propagation (Parent et al. 1997, 2006; Dashtipour et al. 2001; Scharfman et al. 2000). In a recent study, aberrant neurogenesis was hypothesized to negatively influence the cardiovascular system of patients with epilepsy, leading to cardiac abnormalities and hence SUDEP (Scorza et al. 2008). The analysis of the literature reviewed in the present report shows some inconsistencies concerning risk factors for SUDEP. In our opinion, differences in data collected using experimental animal models could derive from the variability in study design. Moreover, the different behaviors displayed by the same brain structures could be related not only to differences between focal and generalized models but also to the intrinsic characteristics of the same structures when considered in different species, as well as to speciἀc epileptogenic characteristic of the chemicals used to induce the epileptic foci (Prince 1969, 1972; Mameli et al. 1999, 1991). With regard to penicillin-G epileptogenic activity, for instance, the sensitivity of nervous structures has been shown to signiἀcantly decrease in the rostrocaudal direction (Mameli et al. 2006). Moreover, at the bulbar level, where cardiorespiratory neurons are localized, penicillin G–induced discharges are not as severe as those induced at the mesencephalic level and disappear after midcollicular transection (Mameli et al. 1991), showing their dependence on rostral nervous structures (De Riu et al. 1994). Instead, in drug-induced generalized epilepsy, the simultaneous general involvement of all cortical and subcortical neuronal networks must be considered. These can be impaired directly and/or indirectly by the epileptogenic drugs as well as by endogenous opioids and neurotransmitters released during generalized seizures. In conclusion, the different models of experimental epilepsy used in animals to analyze the physiopathological mechanisms of SUDEP conἀrm the clinical events detected in patients as well as the overlapping mechanisms proposed to explain human SUDEP. They€show that in fatal conditions, the cortical signals simultaneously activate the central autonomic regulatory triggers, inducing cardiopulmonary and metabolic impairments€that in some subjects with a low threshold of excitability might result in a sudden death. This

Sudden Epileptic Death in Experimental Animal Models

605

is particularly true when the parasympathetic division is involved. In fact, its activation might be responsible for neurogenic ictal bradyarrhythmias, respiratory and metabolic impairments, cardiac asystole, or even SUD in epileptic patients.

References Allen, W. F. 1931. An experimentally produced premature systolic arrhythmia (pulsus bigeminus) in rabbits. Am J Physiol 98: 344–335. Allen, G. V., R. T. Cheung, and D. F. Cechetto. 1995. Neurochemical changes following occlusion of the middle cerebral artery in rats. Neuroscience 68 (4): 1037–1050. Altman, J., and G. D. Das. 1965. Autoradiographic and histological evidence of postnatal hippocampal neurogenesis in rats. J Comp Neurol 124 (3): 319–335. Annegers, J. F., and S. P. Coan. 1999. SUDEP: Overview of deἀnitions and review of incidence data. Seizure 8 (6): 347–352. Attar, H. J., M. T. Gutierrez, S. Bellet, and J. R. Ravens. 1963. Effect of stimulation of hypothalamus and reticular activating system on production of cardiac arrhythmia. Circ Res 12: 14–21. Bayne, L. L., and R. P. Simon. 1981. Systemic and pulmonary vascular pressures during generalized seizures in sheep. Ann Neurol 10 (6): 566–569. Benowitz, N. L., R. P. Simon, and J. R. Copeland. 1986. Status epilepticus: Divergence of sympathetic activity and cardiovascular response. Ann Neurol 19 (2): 197–199. Blum, A. S., J. R. Ives, A. L. Goldberger, I. C. Al-Aweel, K. B. Krishnamurthy, F. W. Drislane, and D.€L.€Schomer. 2000. Oxygen desaturations triggered by partial seizures: Implications for cardiopulmonary instability in epilepsy. Epilepsia 41 (5): 536–541. Blumhardt, L. D., P. E. Smith, and L. Owen. 1986. Electrocardiographic accompaniments of temporal lobe epileptic seizures. Lancet 1 (8489): 1051–1056. Boeles, J. T., G. C. Korteweg, and J. Ten Cate. 1957. Influence of stimulation of some subcortical areas on electrocardiogram. J Neurophysiol 20 (1): 100–107. Breusch, S. R. 1984. Anatomy of the human hypothalamus. In The Hypothalamus, ed. J. R. Givens. Chicago, IL: Year Book Medical Publishers. Browning, R. A., A. V. Wood, M. A. Merrill, J. W. Dailey, and P. C. Jobe. 1997. Enhancement of the anticonvulsant effect of fluoxetine following blockade of 5-HT1A receptors. Eur J Pharmacol 336 (1): 1–6. Cameron, H. A., C. S. Woolley, B. S. McEwen, and E. Gould. 1993. Differentiation of newly born neurons and glia in the dentate gyrus of the adult rat. Neuroscience 56 (2): 337–344. Carnel, S. B., P. L. Schraeder, and C. M. Lathers. 1985. Effect of phenobarbital pretreatment on cardiac neural discharge and pentylenetetrazol-induced epileptogenic activity in the cat. Pharmacology 30 (4): 225–240. Chapman, W. P., R. B. Livingston, and K. E. Livingston. 1949. Frontal lobotomy and electrical stimulation of orbital surface of frontal lobes; effect on respiration and on blood pressure in man. Arch Neurol Psychiatry 62 (6): 701–716. Chapman, W. P., K. E. Livingston, and J. L. Poppen. 1950. Effect upon blood pressure of electrical stimulation of tips of temporal lobes in man. J Neurophysiol 13 (1): 65–71, illust. Cheung, R. T., and D. F. Cechetto. 1995. Neuropeptide changes following excitotoxic lesion of the insular cortex in rats. J Comp Neurol 362 (4): 535–550. Cheung, R. T., T. Diab, and D. F. Cechetto. 1995. Time-course of neuropeptide changes in periÂ�ischemic zone and amygdala following focal ischemia in rats. J Comp Neurol 360 (1): 101–120. Cheung, R. T., V. C. Hachinski, and D. F. Cechetto. 1997. Cardiovascular response to stress after middle cerebral artery occlusion in rats. Brain Res 747 (2): 181–188. Collins, R. L. 1972. Audiogenic seizures. In Experimental Models of Epilepsy: A Manual for the Laboratory Worker, ed. D. P. Purpura, J. K. Penry, D. B. Tower, D. M. Woodbury, and R. D. Walter. New York, NY: Raven Press.

606 Sudden Death in Epilepsy: Forensic and Clinical Issues Colugnati, D. B., P. A. Gomes, R. M. Arida, M. de Albuquerque, R. M. Cysneiros, E. A. Cavalheiro, and F. A. Scorza. 2005. Analysis of cardiac parameters in animals with epilepsy: Possible cause of sudden death? Arq Neuropsiquiatr 63 (4): 1035–1041. Craig, Jr., A. B., and P. L. Mendell. 1959. Blockade of hyperkalemia and hyperglycemia induced by epinephrine in frog liver and in cats. Am J Physiol 197 (1): 52–54. Darbin, O., D. Casebeer, and D. K. Naritoku. 2003. Effects of seizure severity and seizure repetition on postictal cardiac arrhythmia following maximal electroshock. Exp Neurol 181 (2): 327–331. Darnell, J. C., and S. J. Jay. 1982. Recurrent postictal pulmonary edema: A case report and review of the literature. Epilepsia 23 (1): 71–83. Dasheiff, R. M., and L. J. Dickinson. 1986. Sudden unexpected death of epileptic patient due to cardiac arrhythmia after seizure. Arch Neurol 43 (2): 194–196. Dashtipour, K., P. H. Tran, M. M. Okazaki, J. V. Nadler, and C. E. Ribak. 2001. Ultrastructural features and synaptic connections of hilar ectopic granule cells in the rat dentate gyrus are different from those of granule cells in the granule cell layer. Brain Res 890 (2): 261–271. De Riu, P. L. 1983. Cardiovascular effects induced by lateral hypothalamic stimulation in anesthetized cats. IRCS Med Sci 11: 32–33. De Riu, P. L., F. Melis, A. Becciu, and O. Mameli. 1994. Cerebellar and mesencephalic influence on bulbar penicillin-G epileptogenesis in rats. Physiol Behav 56 (4): 715–721. DeFronzo, R. A., P. Felig, E. Ferrannini, and J. Wahren. 1980. Effect of graded doses of insulin on splanchnic and peripheral potassium metabolism in man. Am J Physiol 238 (5): E421–E427. Delgado, J. M. 1960. Circulatory effects of cortical stimulation. Physiol Rev Suppl 4: 146–178. Dhir, A., and S. K. Kulkarni. 2006. Rofecoxib, a selective cyclooxygenase-2 (COX-2) inhibitor potentiates the anticonvulsant activity of tiagabine against pentylenetetrazol-induced convulsions in mice. Inflammopharmacology 14 (5–6): 222–225. Dikshit, B. B. 1934. The production of cardiac irregularities by excitation of the hypothalamic center. J Physiol 81: 382–394. Doba, N., H. R. Beresford, and D. J. Reis. 1975. Changes in regional blood flow and cardiodynamics associated with electrically and chemically induced epilepsy in cat. Brain Res 90 (1): 115–132. Druschky, A., M. J. Hilz, P. Hopp, G. Platsch, M. Radespiel-Troger, K. Druschky, T. Kuwert, H. Stefan, and B. Neundorfer. 2001. Interictal cardiac autonomic dysfunction in temporal lobe epilepsy demonstrated by [(123)I]metaiodobenzylguanidine-SPECT. Brain 124 (Pt 12): 2372–2382. Earnest, M. P., G. E. Thomas, R. A. Eden, and K. F. Hossack. 1992. The sudden unexplained death syndrome in epilepsy: Demographic, clinical, and postmortem features. Epilepsia 33 (2): 310–316. Eckenhoff, M. F., and P. Rakic. 1988. Nature and fate of proliferative cells in the hippocampal dentate gyrus during the life span of the rhesus monkey. J Neurosci 8 (8): 2729–2747. Ellis, S. 1956. The metabolic effects of epinephrine and related amines. Pharmacol. Rev. 8 (4): 485–565. Eriksson, P. S., E. Perἀlieva, T. Bjork-Eriksson, A. M. Alborn, C. Nordborg, D. A. Peterson, and F. H. Gage. 1998. Neurogenesis in the adult human hippocampus. Nat Med 4 (11): 1313–1317. Evans, D. E., and R. A. Gillis. 1974. Effect of diphenylhydantoin and lidocaine on cardiac arrhythmias induced by hypothalamic stimulation. J Pharmacol Exp Ther 191 (3): 506–517. Evans, D. E., and R. A. Gillis. 1978. Reflux mechanisms involved in cardiac arrhythmias induced by hypothalamic stimulation. Am J Physiol 234 (2): H199–H209. Feldman, J. L., G. S. Mitchell, and E. E. Nattie. 2003. Breathing: Rhythmicity, plasticity, chemosensitivity. Annu Rev Neurosci 26: 239–266. Ficker, D. M., E. L. So, W. K. Shen, J. F. Annegers, P. C. O’Brien, G. D. Cascino, and P. G. Belau. 1998. Population-based study of the incidence of sudden unexplained death in epilepsy. Neurology 51 (5): 1270–1274. Freeman, R. 2006. Cardiovascular manifestations of autonomic epilepsy. Clin Auton Res 16 (1): 12–17.

Sudden Epileptic Death in Experimental Animal Models

607

Fuhr, P., and D. Leppert. 2000. Neuroimages. Cardiac arrest during partial seizure. Neurology 54 (10): 2026. Fuster, J. M., and S. J. Weinberg. 1960. Bioelectrical changes of the heart cycle induced by stimulation of diencephalic regions. Exp Neurol 2: 26–39. Graham, D. I. 1992. Hypoxia and vascular disorders. In Greenfield’s Neuropathology, ed. J. H. Adams and L. W. Duchen. London: Arnold. Giugliano, D., R. Torella, S. Sgambato, and F. D’Onofrio. 1979. Effects of alpha- and beta-adrenergic inhibition and somatostatin on plasma glucose, free fatty acids, insulin glucagon and growth hormone responses to prostaglandin E1 in man. J Clin Endocrinol Metab 48 (2): 302–308. Goodman, J. H., R. E. Berger, and T. K. Tcheng. 2005. Preemptive low-frequency stimulation decreases the incidence of amygdala-kindled seizures. Epilepsia 46 (1): 1–7. Goodman, J. H., R. W. Homan, and I. L. Crawford. 1990. Kindled seizures elevate blood pressure and induce cardiac arrhythmias. Epilepsia 31 (5): 489–495. Goodman, J. H., R. W. Homan, and I. L. Crawford. 1999. Kindled seizures activate both branches of the autonomic nervous system. Epilepsy Res 34 (2–3): 169–176. Goren, M. Z., R. Aker, H. R. Yananli, and F. Y. Onat. 2003. Extracellular concentrations of catecholamines and amino acids in the dorsomedial hypothalamus of kindled rats. A microdialysis study. Pharmacology 68 (4): 190–197. Gould, E., P. Tanapat, B. S. McEwen, G. Flugge, and E. Fuchs. 1998. Proliferation of granule cell precursors in the dentate gyrus of adult monkeys is diminished by stress. Proc Natl Acad Sci U S€A 95 (6): 3168–3171. Gray, H. 1973. In Gray’s Anatomy, Descriptive and Applied, ed. D. V. Davies and R. E. Coupland. London: Longman Greens. Guerra, S. M., and A. E. Kitabchi. 1976. Comparison of the effectiveness of various routes of insulin injection: Insulin levels and glucose response in normal subjects. J Clin Endocrinol Metab 42 (5): 869–874. Gunn, C. G., G. Sevelius, J. Puiggari, and F. K. Myers. 1968. Vagal cardiomotor mechanisms in the hindbrain of the dog and cat. Am J Physiol 214 (2): 258–262. Hall, R. E., H. D. Sybers, J. H. Greenhoot, and C. M. Bloor. 1974. Myocardial alterations following hypothalamic stimulation in the intact conscious dog. Am Heart J 88 (6): 770–776. Han, J., and G. K. Moe. 1964. Nonuniform recovery of excitability in ventricular muscle. Circ Res 14: 44–60. Hanning, C. D., and J. M. Alexander-Williams. 1995. Pulse oximetry: A practical review. BMJ 311 (7001): 367–370. Harper, R. M., R. C. Frysinger, R. B. Trelease, and J. D. Marks. 1984. State-dependent alteration of respiratory cycle timing by stimulation of the central nucleus of the amygdala. Brain Res 306 (1–2): 1–8. Hauser, W. A., J. F. Annegers, and L. R. Elveback. 1980. Mortality in patients with epilepsy. Epilepsia 21 (4): 399–412. Hilz, M. J., O. Devinsky, W. Doyle, A. Mauerer, and M. Dutsch. 2002. Decrease of sympathetic cardiovascular modulation after temporal lobe epilepsy surgery. Brain 125 (Pt 5): 985–995. Hockman, C. H., H. P. Mauck Jr., and E. C. Hoff. 1966. ECG changes resulting from cerebral stimulation: II. A spectrum of ventricular arrhythmias of sympathetic origin. Am Heart J 71 (5): 695–700. Ieda, M., H. Kanazawa, K. Kimura, F. Hattori, Y. Ieda, M. Taniguchi, J. K. Lee, K. Matsumura, Y. Tomita, S. Miyoshi, K. Shimoda, S. Makino, M. Sano, I. Kodama, S. Ogawa, and K. Fukuda. 2007. Sema3a maintains normal heart rhythm through sympathetic innervation patterning. Nat Med 13 (5): 604–612. Izzo, P. N., J. Deuchars, and K. M. Spyer. 1993. Localization of cardiac vagal preganglionic motoneurones in the rat: Immunocytochemical evidence of synaptic inputs containing 5-Â�hydroxytryptamine. J Comp Neurol 327 (4): 572–583.

608 Sudden Death in Epilepsy: Forensic and Clinical Issues James, M. R., H. Marshall, and M. Carew-McColl. 1991. Pulse oximetry during apparent tonic– clonic seizures. Lancet 337 (8738): 394–395. Johnson, L. C., and R. A. Davidoff. 1964. Autonomic changes during paroxysmal EEG activity. Electroencephalogr Clin Neurophysiol 17: 25–35. Johnston, S. C., J. K. Horn, J. Valente, and R. P. Simon. 1995. The role of hypoventilation in a sheep model of epileptic sudden death. Ann Neurol 37 (4): 531–537. Johnston, S. C., T. M. Darragh, and R. P. Simon. 1996. Postictal pulmonary edema requires pulmonary vascular pressure increases. Epilepsia 37 (5): 428–432. Johnston, S. C., R. Siedenberg, J. K. Min, E. H. Jerome, and K. D. Laxer. 1997. Central apnea and acute cardiac ischemia in a sheep model of epileptic sudden death. Ann Neurol 42 (4): 588–594. Kalviainen, R., T. Keranen, J. Mustonen, E. Lansimies, and P. J. Riekkinen. 1990. Autonomic nervous system function in Baltic myoclonus epilepsy. Epilepsy Res 5 (3): 251–254. Kawai, H., M. L. Allende, R. Wada, M. Kono, K. Sango, C. Deng, T. Miyakawa, J. N. Crawley, N. Werth, U. Bierfreund, K. Sandhoff, and R. L. Proia. 2001. Mice expressing only monosialoganglioside GM3 exhibit lethal audiogenic seizures. J Biol Chem 276 (10): 6885–6888. Keilson, M. J., W. A. Hauser, J. P. Magrill, and M. Goldman. 1987. ECG abnormalities in patients with epilepsy. Neurology 37 (10): 1624–1626. Kelly, A. M., C. J. Porter, M. D. McGoon, R. E. Espinosa, M. J. Osborn, and D. L. Hayes. 2001. Breathholding spells associated with signiἀcant bradycardia: Successful treatment with permanent pacemaker implantation. Pediatrics 108 (3): 698–702. Korner, P. I. 1979. Central nervous control of autonomic cardiovascular function. In Handbook of Physiology: The Cardiovascular System, ed. R. M. Berne, N. Sperelaskis, and S. R. Geiger. Bethesda, MD: Am Physiol Soc. Kraras, C. M., N. Tumer, and C. M. Lathers. 1987. The role of enkephalins in the production of epileptogenic activity and autonomic dysfunction: Origin of arrhythmia and sudden death in the epileptic patient? Med Hypotheses 23 (1): 19–31. Kuhn, H. G., H. Dickinson-Anson, and F. H. Gage. 1996. Neurogenesis in the dentate gyrus of the adult rat: Age-related decrease of neuronal progenitor proliferation. J Neurosci 16 (6): 2027–2033. Kwan, P., and J. W. Sander. 2004. The natural history of epilepsy: An epidemiological view. J Neurol Neurosurg Psychiatry 75 (10): 1376–1381. Landau, M. W. 1953. Autonomic responses mediated via the corticospinal tract. J Neurophysiol 16 (3): 299–311. Langan, Y., L. Nashef, and J. W. Sander. 2000. Sudden unexpected death in epilepsy: A series of witnessed deaths. J Neurol Neurosurg Psychiatry 68 (2): 211–213. Lathers, C. M. 1990. Role of neuropeptides in the production of epileptogenic activity and arrhythmias. In Epilepsy and Sudden Death, Chap. 18, ed. C. M. Lathers and P. Schraeder. New York, NY: Marcel Dekker. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23 (6): 633–647. Lathers, C. M., and P. L. Schraeder. 1987. Review of autonomic dysfunction, cardiac arrhythmias, and epileptogenic activity. J Clin Pharmacol 27 (5): 346–356. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9 (2): 236–242. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008. The mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12 (1): 3–24. Lathers, C. M., P. L. Schraeder, and S. B. Carnel. 1984. Neural mechanisms in cardiac arrhythmias associated with epileptogenic activity: The effect of phenobarbital in the cat. Life Sci 34 (20): 1919–1936. Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67 (3): 247–259.

Sudden Epileptic Death in Experimental Animal Models

609

Lathers, C. M., N. Tumer, and C. M. Kraras. 1985. Cardiovascular and epileptogenic effects of pentylÂ� enetetrazol administered intracerebroventricularly in cats. Epilepsia 26: 520. Lathers, C. M., N. Tumer, and C. M. Kraras. 1988. The effect of intracerebroventricular D-ALA2 methionine enkephalinamide and naloxone on cardiovascular parameters in the cat. Life Sci 43 (26): 2287–2298. Leestma, J. E., J. F. Annegers, M. J. Brodie, S. Brown, P. Schraeder, D. Siscovick, B. B. Wannamaker, P. S. Tennis, M. A. Cierpial, and N. L. Earl. 1997. Sudden unexplained death in epilepsy: Observations from a large clinical development program. Epilepsia 38 (1): 47–55. Leestma, J. E., T. Walczak, J. R. Hughes, M. B. Kalelkar, and S. S. Teas. 1989. A prospective study on sudden unexpected death in epilepsy. Ann Neurol 26 (2): 195–203. Leutmezer, F., C. Schernthaner, S. Lurger, K. Potzelberger, and C. Baumgartner. 2003. ElectroÂ� cardiographic changes at the onset of epileptic seizures. Epilepsia 44 (3): 348–354. Liedholm, L. J., and O. Gudjonsson. 1992. Cardiac arrest due to partial epileptic seizures. Neurology 42 (4): 824–829. Lisander, B., B. Jaju, and S. C. Wang. 1975. CNS site of antiarrhythmic action of diphenylhydantoin (DPH) in the cat. Eur J Pharmacol 31 (1): 53–62. Lois, C., and A. Alvarez-Buylla. 1994. Long-distance neuronal migration in the adult mammalian brain. Science 264 (5162): 1145–1148. Lopez-Garcia, C., A. Molowny, J. M. Garcia-Verdugo, and I. Ferrer. 1988. Delayed postnatal neurogenesis in the cerebral cortex of lizards. Brain Res 471 (2): 167–174. Lown, B., and R. A. DeSilva. 1978. Roles of psychologic stress and autonomic nervous system changes in provocation of ventricular premature complexes. Am J Cardiol 41 (6): 979–985. Lown, B., J. V. Temte, P. Reich, C. Gaughan, Q. Regestein, and H. Hal. 1976. Basis for recurring ventricular ἀbrillation in the absence of coronary heart disease and its management. N Engl J Med 294 (12): 623–629. Mameli, O., M. A. Caria, F. Melis, C. Severino, C. Tavera, P. Mameli, and S. Mameli. 2001. Autonomic nervous system activity and life threatening arrhythmias in experimental epilepsy. Seizure 10 (4): 269–278. Mameli, O., M. A. Caria, A. Pintus, G. Padua, and S. Mameli. 2006. Sudden death in epilepsy: An experimental animal model. Seizure 15 (5): 275–287. Mameli, O., F. Melis, M. A. Caria, A. Solinas, S. Mameli, and P. L. De Riu. 1999. Epileptic discharge of cortical, subcortical and spinal neurons in penicillin induced experimental epilepsy. Arch Ital Biol 137 (1): 29–46. Mameli, O., F. Melis, D. Giraudi, M. Cualbu, S. Mameli, P. L. De Riu, and P. Mameli. 1993. The brainstem cardioarrhythmogenic triggers and their possible role in sudden epileptic death. Epilepsy Res 15 (3): 171–178. Mameli, O., F. Melis, E. Tolu, M. A. Caria, P. L. De Riu, and G. B. Azzena. 1991. Penicillin-induced paroxysmal activity in brainstem neurons. Physiol Behav 49 (3): 631–634. Mameli, P., O. Mameli, E. Tolu, D. Giraudi, G. Padua, F. Melis, and M. A. Caria. 1989. Investigation into whether neurogenic cardiovascular alterations are involved in sudden epileptic death. In Advances in Epileptology, ed. J. Manelis, E. Bental, J. N. Loeber, and F. E. Dreifuss. New York, NY: Raven Press. Mameli, P., O. Mameli, E. Tolu, G. Padua, D. Giraudi, M. A. Caria, and F. Melis. 1988. Neurogenic myocardial arrhythmias in experimental focal epilepsy. Epilepsia 29 (1): 74–82. Mameli, O., P. Mameli, E. Tolu, G. Padua, F. Melis, M. A. Caria, and A. Virdis. 1990. Analysis of central cardioarrhythmogenic triggers in experimental epilepsy. Epilepsy Res 7 (3): 210–218. Manning, J. W., and C. N. Peiss. 1960. Cardiovascular responses to electrical stimulation in the diencephalon. Am J Physiol 198: 366–370. Mauger, J. P., F. Sladeczek, and J. Bockaert. 1982. Characteristics and metabolism of alpha 1 adrenergic receptors in a nonfusing muscle cell line. J Biol Chem 257 (2): 875–879. McCown, T. J., R. S. Greenwood, G. D. Frye, and G. R. Breese. 1984. Electrically elicited seizures from the inferior colliculus: A potential site for the genesis of epilepsy? Exp Neurol 86 (3): 527–542.

610 Sudden Death in Epilepsy: Forensic and Clinical Issues Mello, L. E., E. A. Cavalheiro, A. M. Tan, W. R. Kupfer, J. K. Pretorius, T. L. Babb, and D. M. Finch. 1993. Circuit mechanisms of seizures in the pilocarpine model of chronic epilepsy: Cell loss and mossy ἀber sprouting. Epilepsia 34 (6): 985–995. Melville, K. I., B. Blum, H. E. Shister, and M. D. Silver. 1963. Cardiac ischemic changes and arrhythmias induced by hypothalamic stimulation. Am J Cardiol 12: 781–791. Minaker, K. L., and J. W. Rowe. 1982. Potassium homeostasis during hyperinsulinemia: Effect of insulin level, beta-blockade, and age. Am J Physiol 242 (6): E373–E377. Mitchell, G. S., T. L. Baker, S. A. Nanda, D. D. Fuller, A. G. Zabka, B. A. Hodgeman, R. W. Bavis, K. J. Mack, and E. B. Olson Jr. 2001. Invited review: Intermittent hypoxia and respiratory plasticity. J Appl Physiol 90 (6): 2466–2475. Mondon, K., B. Charbonnier, C. Hommet, P. Corcia, A. Autret, and B. de Toffol. 2002. Ictal bradycardia followed by cardiac asystole: A case report. Epileptic Disord 4 (4): 261–264. Monte, C. P., J. B. Arends, I. Y. Tan, A. P. Aldenkamp, M. Limburg, and M. C. de Krom. 2007. Sudden unexpected death in epilepsy patients: Risk factors. A systematic review. Seizure 16 (1): 1–7. Nashef, L. 1997. Sudden unexpected death in epilepsy: Terminology and deἀnitions. Epilepsia 38: S6–S8. Nashef, L. 1999. From mystery to prevention: Sudden unexpected death in epilepsy, time to move on. J Neurol Neurosurg Psychiatry 67 (4): 427. Nashef, L., and S. Brown. 1996. Epilepsy and sudden death. Lancet 348 (9038): 1324–1325. Nashef, L., N. Hindocha, and A. Makoff. 2007. Risk factors in sudden death in epilepsy (SUDEP): The quest for mechanisms. Epilepsia 48 (5): 859–871. Nashef, L., and S. D. Shorvon. 1997. Mortality in epilepsy. Epilepsia 38 (10): 1059–1061. Nashef, L., F. Walker, P. Allen, J. W. Sander, S. D. Shorvon, and D. R. Fish. 1996. Apnoea and bradycardia during epileptic seizures: Relation to sudden death in epilepsy. J Neurol Neurosurg Psychiatry 60 (3): 297–300. Natelson, B. H. 1985. Neurocardiology. An interdisciplinary area for the 80s. Arch Neurol 42 (2): 178–184. Nei, M., R. T. Ho, and M. R. Sperling. 2000. EKG abnormalities during partial seizures in refractory epilepsy. Epilepsia 41 (5): 542–548. Nelson, D. A., and C. D. Ray. 1968. Respiratory arrest from seizure discharges in limbic system. Report of cases. Arch Neurol 19 (2): 199–207. Neuspiel, D. R., and L. H. Kuller. 1985. Sudden and unexpected natural death in childhood and adolescence. JAMA 254 (10): 1321–1325. Nilsson, L., U. Bergman, V. Diwan, B. Y. Farahmand, P. G. Persson, and T. Tomson. 2001. Antiepileptic drug therapy and its management in sudden unexpected death in epilepsy: A case-control study. Epilepsia 42 (5): 667–673. Nottebohm, F. 1989. From bird song to neurogenesis. Sci Am 260 (2): 74–79. Novak V, V., L. A. Reeves, P. Novak, A. P. Low, and W. F. Sharbrough. 1999. Time-frequency mapping of R–R interval during complex partial seizures of temporal lobe origin. J Auton Nerv Syst 77 (2–3): 195–202. Opeskin, K., and S. F. Berkovic. 2003. Risk factors for sudden unexpected death in epilepsy: A controlled prospective study based on coroners cases. Seizure 12 (7): 456–464. Oppenheimer, S. M., and D. F. Cechetto. 1990. Cardiac chronotropic organization of the rat insular cortex. Brain Res 533 (1): 66–72. Oppenheimer, S. M., D. F. Cechetto, and V. C. Hachinski. 1990. Cerebrogenic cardiac arrhythmias. Cerebral electrocardiographic influences and their role in sudden death. Arch Neurol 47 (5): 513–519. Oppenheimer, S. M., G. Kedem, and W. M. Martin. 1996. Left-insular cortex lesions perturb cardiac autonomic tone in humans. Clin Auton Res 6 (3): 131–140. Oppenheimer, S. M., J. X. Wilson, C. Guiraudon, and D. F. Cechetto. 1991. Insular cortex stimulation produces lethal cardiac arrhythmias: A mechanism of sudden death? Brain Res 550 (1): 115–121.

Sudden Epileptic Death in Experimental Animal Models

611

O’Regan, M. E., and J. K. Brown. 2005. Abnormalities in cardiac and respiratory function observed during seizures in childhood. Dev Med Child Neurol 47 (1): 4–9. Papez, J. W. 1937. A proposed mechanism of emotion. Arch Neurol Psychiatry 38: 725–743. Parent, J. M. 2007. Adult neurogenesis in the intact and epileptic dentate gyrus. Prog Brain Res 163: 529–540. Parent, J. M., R. C. Elliott, S. J. Pleasure, N. M. Barbaro, and D. H. Lowenstein. 2006. Aberrant seizure-induced neurogenesis in experimental temporal lobe epilepsy. Ann Neurol 59 (1): 81–91. Parent, J. M., T. W. Yu, R. T. Leibowitz, D. H. Geschwind, R. S. Sloviter, and D. H. Lowenstein. 1997. Dentate granule cell neurogenesis is increased by seizures and contributes to aberrant network reorganization in the adult rat hippocampus. J Neurosci 17 (10): 3727–3738. Parker, Jr., I. T., C. G. Gunn, and T. N. Lynn. 1962. Experimental centrogenic arrhythmias. Clin Res 10: 179. Pericic, D., J. Lazic, and D. Svob Strac. 2005. Anticonvulsant effects of acute and repeated fluoxetine treatment in unstressed and stressed mice. Brain Res 1033 (1): 90–95. Phizackerley, P. J., E. W. Pool, and C. W. Whitty. 1954. Sino-auricular heart block as an epileptic manifestation; a case report. Epilepsia 3: 89–91. Polenov, A. L., and V. K. Chetverukhin. 1993. Ultrastructural radioautographic analysis of neurogenesis in the hypothalamus of the adult frog, Rana temporaria, with special reference to physiological regeneration of the preoptic nucleus: II. Types of neuronal cells produced. Cell Tissue Res 271 (2): 351–362. Pool, J. L., and J. Ransohoff. 1949. Autonomic effects on stimulating rostral portion of cingulate gyri in man. J Neurophysiol 12 (6): 385–392. Prince, D. A. 1969. Microelectrode studies of penicillin foci. In Basic Mechanisms of the Epilepsies, ed. H. H. Jasper, A. A. Ward, and A. Pope. Boston, MA: Little Brown. Prince, D. A. 1972. Topical convulsant drug and metabolic antagonists. In Experimental Model of Epilepsy. A Manual for the Laboratory Worker, ed. D. P. Purpura, J. K. Penry, D. B. Tower, D. M. Woodbury, and R. D. Walter. New York, NY: Raven Press. Purpura, D. P., J. L. Pool, E. M. Housepian, M. Girado, and S. A. Jacobson. 1958. Hypothermic potentiation of centrally induced cardiac irregularities. Anesthesiology 19 (1): 27–37. Ramabadran, K., and M. Bansinath. 1990. Endogenous opioid peptides and epilepsy. Int J Clin Pharmacol Ther Toxicol 28 (2): 47–62. Saper, C. B., A. D. Loewy, L. W. Swanson, and W. M. Cowan. 1976. Direct hypothalamo-autonomic connections. Brain Res 117 (2): 305–312. Sathyaprabha, T. N., P. Satishchandra, K. Netravathi, S. Sinha, K. Thennarasu, and T. R. Raju. 2006. Cardiac autonomic dysfunctions in chronic refractory epilepsy. Epilepsy Res 72 (1): 49–56. Scharfman, H. E., J. H. Goodman, and A. L. Sollas. 2000. Granule–like neurons at the hilar/CA3 border after status epilepticus and their synchrony with area CA3 pyramidal cells: Functional implications of seizure-induced neurogenesis. J Neurosci 20 (16): 6144–6158. Schernthaner, C., G. Lindinger, K. Potzelberger, K. Zeiler, and C. Baumgartner. 1999. Autonomic epilepsy—The influence of epileptic discharges on heart rate and rhythm. Wien Klin Wochenschr 111 (10): 392–401. Schraeder, P. L., K. Delin, R. L. McClelland, and E. L. So. 2006. Coroner and medical examiner documentation of sudden unexplained deaths in epilepsy. Epilepsy Res 68 (2): 137–143. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32 (12): 1371–1382. Schraeder, P. L., and C. M. Lathers. 1989. Paroxysmal autonomic dysfunction, epileptogenic activity and sudden death. Epilepsy Res 3 (1): 55–62. Schwartz, R. D., and C. M. Lathers. 1990. GABA neurotransmission, epileptogenic activity, and cardiac arrhythmias. In Epilepsy and Sudden Death, Chap. 17, C. M. Lathers and P. Schraeder. New York, NY: Marcel Dekker.

612 Sudden Death in Epilepsy: Forensic and Clinical Issues Scorza, F. A., R. M. Cysneiros, R. M. Arida, C. A. Scorza, A. C. de Almeida, B. Schmidt, and E. A. Cavalheiro. 2008. Adult hippocampal neurogenesis and sudden unexpected death in epilepsy: Reality or just an attractive history? Med Hypotheses 71 (6): 914–922. Seeck, M., O. Blanke, P. Jallon, F. Picard, and S. Zaim. 2001. Symptomatic postictal cardiac asystole in a young patient with partial seizures. Europace 3 (3): 247–252. Simon, R. P., M. J. Aminoff, and N. L. Benowitz. 1984. Changes in plasma catecholamines after tonic–clonic seizures. Neurology 34 (2): 255–257. Smith, W. S., and M. A. Matthay. 1997. Evidence for a hydrostatic mechanism in human neurogenic pulmonary edema. Chest 111 (5): 1326–1333. Sperling, M. R., H. Feldman, J. Kinman, J. D. Liporace, and M. J. O’Connor. 1999. Seizure control and mortality in epilepsy. Ann Neurol 46 (1): 45–50. St. John, W. M., A. H. Rudkin, G. L. Homes, and J. C. Leiter. 2006. Changes in respiratory-modulated neural activities, consistent with obstructive and central apnea, during ἀctive seizures in an in situ anaesthetized rat preparation. Epilepsy Res 70 (2–3): 218–228. Standish, A., L. W. Enquist, J. A. Escardo, and J. S. Schwaber. 1995. Central neuronal circuit innerÂ� vating the rat heart deἀned by transneuronal transport of pseudorabies virus. J Neurosci 15 (3€Pt 1): 1998–2012. Standish, A., L. W. Enquist, and J. S. Schwaber. 1994. Innervation of the heart and its central medullary origin deἀned by viral tracing. Science 263 (5144): 232–234. Stanἀeld, B. B., and J. E. Trice. 1988. Evidence that granule cells generated in the dentate gyrus of adult rats extend axonal projections. Exp Brain Res 72 (2): 399–406. Sterns, R. H., J. Guzzo, and P. U. Feig. 1981. The disposition of intravenous potassium in normal man: The role of insulin. Clin Sci (Lond) 61 (1): 23–28. Suter, L. E., and C. M. Lathers. 1984. Modulation of presynaptic gamma aminobutyric acid release by prostaglandin E2: Explanation for epileptogenic activity and dysfunction in autonomic cardiac neural discharge leading to arrhythmias? Med Hypotheses 15 (1): 15–30. Taggart, P., M. Carruthers, and W. Somerville. 1973. Electrocardiogram, plasma catecholamines and lipids, and their modiἀcation by oxyprenolol when speaking before an audience. Lancet 2 (7825): 341–346. Tavee, J., and H. Morris, 3rd. 2008. Severe postictal laryngospasm as a potential mechanism for sudden unexpected death in epilepsy: A near-miss in an EMU. Epilepsia 49 (12): 2113–2117. Taylor, E. W., D. Jordan, and J. H. Coote. 1999. Central control of the cardiovascular and respiratory systems and their interactions in vertebrates. Physiol Rev 79 (3): 855–916. Tennis, P., T. B. Cole, J. F. Annegers, J. E. Leestma, M. McNutt, and A. Rajput. 1995. Cohort study of incidence of sudden unexplained death in persons with seizure disorder treated with antiepileptic drugs in Saskatchewan, Canada. Epilepsia 36 (1): 29–36. Terrence, C. F., G. R. Rao, and J. A. Perper. 1981. Neurogenic pulmonary edema in unexpected, unexplained death of epileptic patients. Ann Neurol 9 (5): 458–464. Terrence, Jr., C. F., H. M. Wisotzkey, and J. A. Perper. 1975. Unexpected, unexplained death in epileptic patients. Neurology 25 (6): 594–598. Tinuper, P., F. Bisulli, A. Cerullo, R. Carcangiu, C. Marini, G. Pierangeli, and P. Cortelli. 2001. Ictal bradycardia in partial epileptic seizures: Autonomic investigation in three cases and literature review. Brain 124 (Pt 12): 2361–2371. Toichi, M., T. Murai, A. Sengoku, and K. Miyoshi. 1998. Interictal change in cardiac autonomic function associated with EEG abnormalities and clinical symptoms: A longitudinal study following acute deterioration in two patients with temporal lobe epilepsy. Psychiatry Clin Neurosci 52 (5): 499–505. Tomson, T. 2000. Mortality in epilepsy. J Neurol 247 (1): 15–21. Tomson, T., T. Walczak, M. Sillanpaa, and J. W. Sander. 2005. Sudden unexpected death in epilepsy: A review of incidence and risk factors. Epilepsia 46 (Suppl 11): 54–61. Tupal, S., and C. L. Faingold. 2006. Evidence supporting a role of serotonin in modulation of sudden death induced by seizures in DBA/2 mice. Epilepsia 47 (1): 21–26. Ueda, H. 1962. Arrhythmias produced by cerebral stimulation. Jpn Circ J 26: 225–230.

Sudden Epileptic Death in Experimental Animal Models

613

Van Bogaert, L. 1936. Hypothamaus et reacations cardiovascularies d’origine centrale. Arch Int Pharmacodyn 52: 137–174. Van Buren, J. M. 1958. Some autonomic concomitants of ictal automatism; a study of temporal lobe attacks. Brain 81 (4): 505–528. Venit, E. L., B. D. Shepard, and T. N. Seyfried. 2004. Oxygenation prevents sudden death in seizureprone mice. Epilepsia 45 (8): 993–996. Vick, R. L., E. P. Todd, and D. W. Luedke. 1972. Epinephrine-induced hypokalemia: Relation to liver and skeletal muscle. J Pharmacol Exp Ther 181 (1): 139–146. Walczak, T. S., I. E. Leppik, M. D’Amelio, J. Rarick, E. So, P. Ahman, K. Ruggles, G. D. Cascino, J. F. Annegers, and W. A. Hauser. 2001. Incidence and risk factors in sudden unexpected death in epilepsy: A prospective cohort study. Neurology 56 (4): 519–525. Walker, A. E. 1966. Internal structure and afferent–efferent relations of the thalamus. In The Thalamus, ed. D. P. Purpura and M. D. Yahr. New York, NY: Columbia University Press. Wall, P. D., and G. D. Davis. 1951. Three cerebral cortical systems affecting autonomic function. J Neurophysiol 14 (6): 507–517. Wang, J., Y. Chen, K. Li, and L. Hou. 2006. Blockade of inhibitory neurotransmission evoked seizurelike ἀring of cardiac parasympathetic neurons in brainstem slices of newborn rats: Implications for sudden deaths in patients of epilepsy. Epilepsy Res 70 (2–3): 172–183. Weinberg, S. J., and J. M. Fuster. 1960. Electrocardiographic changes produced by localized hypothalamic stimulations. Ann Intern Med 53: 332–341. Weinstein, M. D., and C. Albertario. 2000. Cardiac asystole and bradycardia as a manifestation of left temporal lobe complex partial seizure. Ann Intern Med 132 (2): 165–166. Wilder-Smith, E. 1992. Complete atrio-ventricular conduction block during complex partial seizure. J Neurol Neurosurg Psychiatry 55 (8): 734–736. Williams, M. E., R. M. Rosa, P. Silva, R. S. Brown, and F. H. Epstein. 1984. Impairment of extrarenal potassium disposal by alpha-adrenergic stimulation. N Engl J Med 311 (3): 145–149. Willis, W. L., and R. G. Grossman. 1981. Medical Neurobiology. St. Louis, MO: Mosby. Willott, J. F., and K. R. Henry. 1976. Roles of anoxia and noise-induced hearing loss in the postictal refractory period for audiogenic seizures in mice. J Comp Physiol Psychol 90 (4): 373–381. Zhang, Z. H., and S. M. Oppenheimer. 2000. Baroreceptive and somatosensory convergent thalamic neurons project to the posterior insular cortex in the rat. Brain Res 861 (2): 241–256. Zhang, Z. H., S. Rashba, and S. M. Oppenheimer. 1998. Insular cortex lesions alter baroreceptor sensitivity in the urethane-anesthetized rat. Brain Res 813 (1): 73–81. Zhao, C., W. Deng, and F. H. Gage. 2008. Mechanisms and functional implications of adult neurogenesis. Cell 132 (4): 645–660.

Sympathetic Nervous System Dysregulation of Cardiac Function and Myocyte Potassium Channel Remodeling in Rodent Seizure Models Candidate Mechanisms for SUDEP

38

Steven L. Bealer Cameron S. Metcalf Jason G. Little Matteo Vatta Amy Brewster Anne E. Anderson

Contents 38.1 Introduction 38.2 Effects of Seizure Activity on Cardiac Sympathovagal Balance 38.3 Effects of Seizures on Cardiac Responses to Environmental Stress 38.4 Effects of Seizure Activity on Cardiac Ion Channels 38.5 Discussion References

615 617 620 621 622 623

38.1â•…Introduction Although the precise causes of SUDEP have not been completely deἀned, a number of studies indicate that cardiac ventricular abnormalities, and the resulting arrhythmias and sudden cardiac death may contribute (Dasheiff 1991; Lathers and Schraeder 2002; Leung et al. 2006; Nei et al. 2004; P-Codrea Tigaran et al. 2005). Indeed, it has been proposed that cumulative ischemic damage resulting from sympathetic nervous system activation during the repeated seizures characteristic of intractable epilepsy may progressively increase susceptibility to ventricular arrhythmias and SUDEP (McGugan 1999; Shimizu et al. 2008; Tigaran et al. 1997). The proposal that multiple seizures enhance the risk of cardiac-related death is supported by numerous reports that patients with uncontrolled, intractable seizures are at the greatest risk of SUDEP (Nilsson et al. 1999; Opeskin et al. 1999; Opeskin and Berkovic 2003; Tennis et al. 1995).

615

616 Sudden Death in Epilepsy: Forensic and Clinical Issues

Lethal ventricular arrhythmias can be induced by repeated, acute, transient elevations in sympathetic nervous system activity, as well as by chronic increases in sympathetic dominance of cardiac function, particularly acting on a background of cardiac structural changes (Anderson 2003; Campbell 1991). For example, a number of recent studies demonstrate that both the magnitude and duration of sympathetic nervous system responses to experimental stressors are enhanced in pathological conditions characterized by increased vulnerability to lethal arrhythmias, including postmyocardial infarction (CudnochJedrzejewska et al. 2007; Dobruch, Cudnoch-Jedrzejewska, and Szczepanska-Sadowska 2005), experimental obesity (D’Angelo et al. 2006), depression (Grippo et al. 2003, 2006), and hypertension (Giulumian et al. 1999; McDougall et al. 2005). Consequently, it is evident that both enhanced basal cardiac sympathetic nervous system tone, as well as large, transient increases in activation can increase risk of sudden cardiac death. Seizures in epilepsy are accompanied by intense stimulation of the sympathetic nervous system (Simon et al. 1984), with a resulting increase in both heart rate and blood pressure (Di Gennaro et al. 2004; Mayer et al. 2004; Rugg-Gunn et al. 2004). In addition to sympathetic nervous system activity during seizures, it is possible that responses to other environmental stressors are enhanced in patients with epilepsy. Although one recent epidemiological study reported mental stress as a risk factor for SUDEP (Lear-Kaul et al. 2005), the relationship between sympathetic nervous system activation in response to environmental stress and seizure activity has not been determined. Recently, Lathers and Schraeder (2006) elucidated the lack of knowledge in this area and the potential importance of evaluating the relationship between stress and SUDEP. The repeated activation of the sympathetic nervous system during seizures, particularly in conjunction with exaggerated autonomic responses to environmental stressors, may induce progressive cardiac deterioration and contribute to cardiac arrhythmic activity observed during the ictal and immediate postictal period (Rugg-Gunn et al. 2004), which could progress to SUDEP. However, the relationship between seizure activity, sympathetic nervous system tone, and reactivity, in control of the heart in patients with seizure disorders, has not been completely elucidated. The balance between the influence of the sympathetic (sympathetic nervous system) and parasympathetic components of the autonomic nervous system, known as sympathovagal balance or the vagal-sympathetic effect (sympathovagal balance) (Goldberger 1999), is a major regulator of cardiac function. Normally, parasympathetic nervous system (or vagal) tone predominates and reduces the proarrhythmic effects of the sympathetic nervous system activity (Anderson 2003), and activation of the sympathetic nervous system, which occurs during stress or exercise, is normally short-lived and highly regulated. However, pathological conditions such as myocardial infarction, congestive heart failure, and coronary artery disease are associated with chronic alterations in normal sympathovagal balance, characterized by dominance of sympathetic nervous system activity, and an associated increased risk of cardiac arrhythmias (La Rovere et al. 2001; La Rovere and Schwartz 1997; Stein and Kleiger 1999; Stein et al. 1994; Vanoli et al. 2008; Liao et al. 1996; Verrier and Antzelevitch 2004). Therefore, a shift in sympathovagal balance toward sympathetic nervous system dominance, either by increased sympathetic nervous system or decreased parasympathetic nervous system activity, can contribute to increased cardiac risk. Sympathovagal balance can be determined by comparing the normal, control heart rate to the intrinsic heart rate. Intrinsic heart rate is deἀned as the heart rate in the absence of autonomic influence and represents the intrinsic rate of the cardiac pacemaker. If the intrinsic heart rate is greater than the control heart rate, the heart is being predominantly

Sympathetic Nervous System Dysregulation of Cardiac Function

617

influenced by the parasympathetic nervous system. However, if the intrinsic heart rate is less than the control heart rate, the sympathetic nervous system is providing the dominant influence on cardiac rate. There are some indirect indications that sympathovagal balance may be altered in the direction of sympathetic nervous system dominance as a result of seizure activity. For example, heart rate variability, the spontaneous fluctuations in the intervals between heart beats, is decreased in patients with established epilepsy (Ansakorpi et al. 2002; Ronkainen et al. 2006; Tomson et al. 1998). Heart rate variability represents the interaction between the sympathetic nervous system and parasympathetic nervous system on the cardiac pacemaker (Task Force of the European Society of Cardiology the North American Society of Pacing Electrophysiology 1996; Stein and Kleiger 1999). Decreased heart rate variability is indicative of increased sympathetic dominance of cardiac control (Stein and Kleiger 1999; van Ravenswaaij-Arts et al. 1993), and is associated with cardiomyopathies that increase risk of sudden cardiac death including coronary artery disease (Krittayaphong et al. 1997), congestive heart failure (van Ravenswaaij-Arts et al. 1993), depression (Grippo et al. 2002), and following recovery myocardial infarction (Kleiger et al. 1987). Although heart rate variability is indicative of increased sympathetic dominance of cardiac function, the effects of seizures on sympathovagal balance have not been previously deἀned. As noted earlier, enhanced basal sympathetic nervous system tone and transient increases in catecholamine stimulation can be a facilitatory factor initiating ventricular arrhythmias, particularly when acting on abnormal cardiac tissue (Anderson 2003; Chen et al. 2007; Dorian 2005). However, the cardiac effects of seizures that provide the arrhythmogenic substrate for adrenergic facilitation of potentially lethal arrhythmias has not been identiἀed. One mechanism that increases susceptibility to sudden cardiac death is ion channel remodeling in cardiac myocytes. Ion channel remodeling, secondary to enhanced sympathetic nervous system activation, mimics what occurs in ion channel dysfunction caused by primary mutations in the genes encoding for the ion channel subunits or for channel interacting proteins (Schimpf et al. 2009; Ueda et al. 2008; Vatta et al. 2006; Wu et al. 2008). In diseases such as long QT syndrome, loss of function of potassium channels occurs due to genetic mutations or acquired inactivation subsequent to blood alkalosis, antibiotics, or other drug-induced inactivation. These ion channel defects caused by primary gene defects in ion channels or channel-interacting proteins deἀne the underlying diseases as a channelopathy (Schimpf et al. 2009). We propose that a similar remodeling may occur after acquired prolonged or recurrent seizures, which would predispose to sudden cardiac death. In summary, it is possible that repeated seizure activity may increase basal sympathetic nervous system dominance of cardiac function and enhance sympathetic nervous system responsiveness, inducing ion channel remodeling in cardiac myocytes. These effects would result in increased susceptibility to ventricular arrhythmias, which would contribute to SUDEP.

38.2â•… Effects of Seizure Activity on Cardiac Sympathovagal Balance To determine the effects of seizures on the relative sympathetic nervous system and parasympathetic nervous system control of cardiac function, we evaluated sympathovagal balance after recovery from a single, long-lasting seizure in male, Sprague–Dawley rats.

618 Sudden Death in Epilepsy: Forensic and Clinical Issues

Seizures were induced with lithium–pilocarpine injections and were terminated after 90 min (Metcalf et al. 2009). Either 1 or 2 weeks after seizures, cardiac sympathovagal balance (sympathovagal balance) was determined by comparing intrinsic heart rate to control heart rate in animals that had experienced seizures (Pilo) or control procedures (saline injection, Cont). Furthermore, in addition to evaluating cardiac function in the absence of all autonomic influences, the individual contributions of the sympathetic nervous system and parasympathetic nervous system were individually determined by measuring changes in heart rate after blockade of the sympathetic nervous system with atenolol and blockade of the parasympathetic nervous system with atropine (Goldberger 1999). All cardiovascular measures were obtained from conscious, unrestrained animals. The relationships between intrinsic heart rate and control heart rate, that is, the sympathovagal balance, which represents a measure of sympathovagal balance (Goldberger 1999), observed in pilocarpine-treated and control-treated rats 1 week (panel a) and 2 weeks (panel b) after treatment are shown in Figure 38.1. As mentioned earlier, a lower sympathovagal balance is indicative of increased sympathetic nervous system dominance of cardiac function. At both periods, the sympathovagal balance was signiἀcantly lower in animals that underwent seizures (Figure 38.1). This observation suggests seizure activity resulted in a long-lasting shift in sympathovagal control of cardiac function toward sympathetic nervous system dominance. However, increased sympathetic nervous system influence on cardiac function may result from enhanced sympathetic nervous system tone, decreased parasympathetic nervous system tone, or both. To elucidate the origin of the shift in cardiac sympathovagal balance toward sympathetic nervous system dominance, we evaluated sympathetic nervous system tone (i.e., the fall in heart rate in response to sympathetic nervous system blockade in the presence of parasympathetic nervous system blockade) and parasympathetic nervous system tone (i.e., the rise in heart rate in response to parasympathetic nervous system blockade in the presence of sympathetic nervous system blockade) (Goldberger 1999). The increase in heart rate in response to atropine (parasympathetic nervous system antagonist) in the presence of sympathetic nervous system blockade was signiἀcantly smaller in animals experiencing seizures (pilocarpine) (Figure 38.2) both 1 week (panel a) and 2 weeks (panel c) after treatment compared to control-treated rats. These data indicate that parasympathetic nervous system tone was signiἀcantly lower in animals undergoing seizures, compared to control rats, for at least 2 weeks after treatment. However, the decrease in

1.10 1.05 1.00 0.95 0.90 0.85 0.80

1.10 1.05 1.00 0.95 0.90 0.85 0.80

VSE (iHR/HR)

(b)

VSE (iHR/HR)

(a)

Cont

Pilo

Cont

Pilo

Figure 38.1╇ Vagal-sympathetic effect (sympathovagal balance) in animals undergoing con-

trol procedures (Cont) or seizure (SE) animals 1 week (a) and 2 weeks (b) after treatment. *p < 0.05, compared to Cont. (From Metcalf, C. S., et al., Epilepsia, 50 (4), 747–754, 2009. With permission.)

45 40 35 30 25 20 15 10 5 0

(a) 0 –25 –50 –75

–100

(c)

(d)

50

HR change (beats/min)

HR change (beats/min)

60

40 30 20 10 0

619

(b) HR change (beats/min)

HR change (beats/min)

Sympathetic Nervous System Dysregulation of Cardiac Function

Cont

Pilo

–10 –35 –60 –85 –110

Cont

Pilo

Figure 38.2╇ Changes in heart rate (heart rate) resulting from atropine after atenolol (parasympathetic nervous system tone) 1 week (a) and 2 weeks (c) after treatment. Changes in heart rate resulting from atenolol after atropine (sympathetic nervous system tone) 1 (b) and 2 (d) weeks after treatment. *p < 0.05 compared to Cont. (From Metcalf, C. S., et al., Epilepsia, 50 (4), 747–754, 2009. With permission.)

heart rate in response to atenolol (sympathetic nervous system antagonist) during parasympathetic nervous system blockade was equivalent between pilocarpine and control rats both 1 and 2 weeks after treatment (Figure 38.2, panels b and d, respectively). This ἀnding suggests that seizures did not alter basal cardiac sympathetic nervous system. Taken together, these data demonstrate that seizure activity resulted in a prolonged increase in sympathetic nervous system dominance of cardiac function, resulting primarily from decreased parasympathetic nervous system tone, with little or no alteration in sympathetic nervous system activity. This chronic increase in sympathetic nervous system influence on the heart after seizures provides support for the proposal that decreased heart rate variability observed in patients with established epilepsy (Ansakorpi et al. 2002; Ronkainen et al. 2006; Tomson et al. 1998) results from increased cardiac sympathetic nervous system tone (Stein and Kleiger 1999; van Ravenswaaij-Arts et al. 1993). In addition, consistent with the ἀndings in this animal model, recent clinical studies have reported patients with intractable epilepsies exhibited higher sympathetic nervous system tone, lower parasympathetic nervous system tone, and signiἀcant autonomic dysregulation (Mukherjee et al. 2009; Sathyaprabha et al. 2006). Because susceptibility to lethal ventricular arrhythmias is enhanced by chronic, sustained increases in sympathetic nervous system control of cardiac function (Airaksinen 1999; Anderson 2003; Chen et al. 2007), these data support the proposal that seizure-induced changes in basal sympathovagal balance predispose patients to SUDEP.

620 Sudden Death in Epilepsy: Forensic and Clinical Issues

38.3â•…Effects of Seizures on Cardiac Responses to€Environmental Stress In addition to the arrhythmogenic effects of chronic, basal sympathetic nervous system dominance on cardiac function, lethal ventricular arrhythmias can also be precipitated by more acute and transient elevations in sympathetic nervous system activity (Anderson 2003; Campbell 1991). Enhanced sympathetic nervous system responses to environmental stressors, and the resulting increase in catecholamine stimulation of the heart, is characteristic of cardiac pathologies that often result in sudden cardiac death (CudnochJedrzejewska et al. 2007; D’Angelo et al. 2006; Dobruch et al. 2005; Grippo et al. 2003, 2006; Giulumian et al. 1999; McDougall et al. 2005). Indeed, it has been proposed that enhanced responses to mental stress may be a risk factor for SUDEP (Lear-Kaul et al. 2005). This would suggest that in addition to the documented activation of the sympathetic nervous system (Simon et al. 1984) and associated increase in heart rate and blood pressure (Di Gennaro et al. 2004; Mayer et al. 2004; Rugg-Gunn et al. 2004) that occurs during seizures, patients with epilepsy may experience more intense sympathetic nervous system activation to non-seizure-related environmental stressors, further increasing the risk of lethal arrhythmias (Anderson 2003; Campbell 1991; Cudnoch-Jedrzejewska et al. 2007). We suggest that environmental stress produces enhanced sympathetic nervous system–mediated cardiac responses in patients with epilepsy that, in conjunction with basal sympathetic nervous system dominance and acute sympathetic nervous system activation during seizures, contributes to progressive deterioration in cardiac function and potentially results in SUDEP. To determine if seizures affect autonomic responses to stress, we measured changes in heart rate and blood pressure before and during administration of a moderate, environmental stressor in animals 10 days to 2 weeks after seizures induced by lithium– pilocarpine treatment. Blood pressure and heart rate were measured in rats subjected to the moderate stress of an air jet aimed at the animal’s head for 3 min. Figure 38.3 shows control heart rate recorded before administration of the stress (Pre), and the maximum heart rate measured during the 3 min of air jet administration (Stress) in animals undergoing seizures (Pilo) and in control-treated rats. As can be seen, there were no differences in basal heart rate between control and pilocarpine rats before administration of the air jet. Furthermore, although mean heart rate tended to increase in control rats in response to the stress, this tachycardia was not statistically signiἀcant. However, in rats that had experienced seizures, air jet stress increased heart rate signiἀcantly above both prestress levels and the values observed in control animals during stress. In contrast, blood pressure was not elevated by air jet stress in either group of animals (data not shown). These ἀndings suggest that seizure activity produces a chronic increase in cardiac sympathetic nervous system responses to environmental stressors, which is reflected in a signiἀcant tachycardia. This increased cardiac reactivity during stress may contribute to SUDEP in patients with recurrent seizures because a similar relationship between enhanced responses to stress and increased risk of lethal arrhythmias has been reported in other pathological conditions characterized by sudden cardiac death, such as post-myocardial infarction (Cudnoch-Jedrzejewska et al. 2007; Dobruch, Cudnoch-Jedrzejewska, and Szczepanska-Sadowska 2005), experimental obesity (D’Angelo et al. 2006), depression (Grippo et al. 2003, 2006), and hypertension (Giulumian et al. 1999; McDougall et al. 2005).

Sympathetic Nervous System Dysregulation of Cardiac Function 500

621

*#

Cont Heart rate (beats/min)

Pilo 450

400

350

0

Stress

Pre

Pre

Stress

Figure 38.3╇ Basal heart rate measured before stress (Pre), and the maximum heart rate

observed during 3 min of air jet stress (Stress) in control (Cont) rats and animals undergoing seizures induced with lithium–pilocarpine (Pilo). *p < 0.05 compared to Pilo-Pre; #p < 0.05 compared to Control-Stress.

38.4â•… Effects of Seizure Activity on Cardiac Ion Channels We have observed sudden death after kainate convulsant stimulation in Kv4.2 knockout mice (Barnwell et al. 2009). These mice do not have a cardiac phenotype at rest. However, after prolonged seizure stimulation with presumed ongoing sympathetic nervous system dominance based on the above ἀndings, these animals suffered sudden death. Given that Kv4.2 subunits contribute to channels that underlie the transient outward current (Ito,f ) encoded in rodent myocytes, one possible mechanism underlying sudden death in these animals after seizure stimulation is a lethal cardiac arrhythmia. Subsequently, we have 125 Optical density (% Control)

100 75 50 25 0

Control

KA

Kv4.2 Actin

Figure 38.4╇ Reduction in cardiac Kv4.2 channel protein levels after kainate-induced seizures.

Immunoblotting of normalized cardiac membranes prepared from animals with sham treatment (Control) vs. kainate-induced seizures (KA) was performed using Kv4.2 and actin (lane loading control) antibodies. *p < 0.05, kainate-treated animals (KA) compared to Control.

622 Sudden Death in Epilepsy: Forensic and Clinical Issues

begun to evaluate whether remodeling of Kv4 channels occurs in models of acquired seizures or epilepsy. As a ἀrst step in these studies, we evaluated Kv4.2 protein levels in rats after kainate- or pilocarpine-induced seizures. We found signiἀcant decreases in Kv4.2 channel levels in both of these models (Figure 38.4 for kainate model data; pilo model data not shown). Thus, we conclude that ion channel remodeling occurs after seizures and may contribute to sudden death in epilepsy. Although the link remains to be shown, one candidate mechanism underlying ion channel remodeling in acquired epilepsy is through altered sympathetic tone.

38.5â•…Discussion These studies have demonstrated that convulsant-induced seizures in rodents result in long-lasting alterations in autonomic control of cardiac function, sympathetic nervous system responses to stress, and cardiomyocyte potassium channels. Speciἀcally, seizures produce an increase in sympathetic nervous system dominance in cardiac control, resulting from diminished parasympathetic nervous system tone, enhanced sympathetic nervous system–induced tachycardia during environmental stress, and reduced expression of Kv4.2 channels in cardiomyocytes. These effects of seizures would be expected to increase susceptibility to ventricular arrhythmias and sudden cardiac death contributing to SUDEP. Lethal ventricular arrhythmias can be produced when a physiological facilitator interacts with abnormal anatomical and electrical substrates in cardiac tissue (Anderson 2003; Campbell 1991). A number of previous studies demonstrated that sympathetic nervous system dominance of cardiac function and enhanced sympathetic nervous system responses to environmental stimuli are associated with increased risk of sudden death in several cardiac pathologies including post-myocardial infarction (Cudnoch-Jedrzejewska et al. 2007; Dobruch, Cudnoch-Jedrzejewska, and Szczepanska-Sadowska 2005), experimental obesity (D’Angelo et al. 2006), depression (Grippo et al. 2003, 2006), and hypertension (Giulumian et al. 1999; McDougall et al. 2005). These data suggest that excessive catecholaminergic stimulation of cardiac tissue is an important facilitator of arrhythmogenesis in these conditions. Our studies extend these ἀndings by demonstrating that seizures that can result in sudden cardiac death are similarly associated with increases in sympathetic nervous system dominance of cardiac death and exaggerated responses to environmental stressors lasting well beyond the period of seizure activity. In addition, these studies suggest that seizure-induced decreases in Kv4.2 channels may provide the electrical substrate for arrhythmogenic activity and contribute to sudden cardiac death in epilepsy. It has previously been shown that severe reduction in channel function caused by the dominant negative W362F mutation leads to prolonged QT interval, enhanced dispersion of repolarization, and refractoriness (London et al. 2007). In addition, a reduction in Kv4.2 levels may also affect the transmural gradient between epicardium and endocardium, leading to an imbalance in homogeneity of cardiac depolarization and thereby providing a setup for risk of cardiac dysrhythmia. Cardiac ion channel remodeling has been previously associated with autonomic imbalance or primary genetic mutations, which are both important causes of cardiac arrhythmias; however, ion channel remodeling acquired in association with various cardiac pathologies is an additional important cause of cardiac arrhythmias (for review, see Shah et al. 2005). Experimentally induced tachycardia through rapid cardiac pacing has been

Sympathetic Nervous System Dysregulation of Cardiac Function

623

associated with K+ channel remodeling. Speciἀcally, levels of Kv4.2 and Kv4.3 decreased in myocardium after prolonged cardiac pacing in rats (Yamashita et al. 2000). Thus, a disruption of the ἀnely tuned balance of ion channels normally expressed in the heart, which may occur through a variety of candidate mechanisms, is a risk for cardiac arrhythmia. Future studies will no doubt begin to shed light on the role of cardiac ion channel remodeling and the link to altered sympathetic tone in epilepsy and SUDEP. In summary, these animal studies suggest that lethal ventricular arrhythmias that contribute to SUDEP may occur in response to chronic sympathetic nervous system dominance of cardiac function, combined with periodic, transient, and exaggerated catecholaminergic stimulation, acting on cardiac myocytes that are predisposed to arrhythmogenesis due to altered potassium channel function. These data support the proposal that single intense seizures and/or repeated seizure activity may increase the risk of SUDEP in patients with intractable epilepsy by (1) increasing exposure of cardiac tissue to an arrhythmogenic facilitator, catecholamines, and (2) providing the electrical substrate, decreased Kv4.2 channels, on which the substrate acts to induce arrhythmic activity. These data further indicate that patients with epilepsy who are at increased risk of SUDEP could beneἀt from cardioprotective agents shown to diminish the incidence of lethal arrhythmias in other pathological conditions. For example, beta-adrenergic antagonists prevent lethal arrhythmias and detrimental cardiac adaptations associated with several cardiomyopathies, including coronary artery disease, heart failure, and myocardial infarction (Adamson and Gilbert 2006; Dorian 2005; Hohnloser 2005).

References Adamson, P. B., and E. M. Gilbert. 2006. Reducing the risk of sudden death in heart failure with betablockers. J Card Fail 12 (9): 734–746. Airaksinen, K. E. 1999. Autonomic mechanisms and sudden death after abrupt coronary occlusion. Ann Med 31 (4): 240–245. Anderson, K. P. 2003. Sympathetic nervous system activity and ventricular tachyarrhythmias: Recent advances. Ann Noninvasive Electrocardiol 8 (1): 75–89. Ansakorpi, H., J. T. Korpelainen, H. V. Huikuri, U. Tolonen, V. V. Myllyla, and J. I. Isojarvi. 2002. Heart rate dynamics in refractory and well controlled temporal lobe epilepsy. J Neurol Neurosurg Psychiatry 72 (1): 26–30. Barnwell, L. F., J. N. Lugo, W. L. Lee, S. E. Willis, S. J. Gertz, R. A. Hrachovy, and A. E. Anderson. 2009. Kv4.2 knockout mice demonstrate increased susceptibility to convulsant stimulation. Epilepsia 50: 1741–1751. Campbell, R. W. 1991. Predisposing factors for ventricular arrhythmias. J Cardiovasc Pharmacol 17 (Suppl 6): S9–S12. Chen, L. S., S. Zhou, M. C. Fishbein, and P. S. Chen. 2007. New perspectives on the role of autonomic nervous system in the genesis of arrhythmias. J Cardiovasc Electrophysiol 18 (1): 123–127. Cudnoch-Jedrzejewska, A., J. Dobruch, L. Puchalska, and E. Szczepanska-Sadowska. 2007. Interaction of AT1 receptors and V1a receptors-mediated effects in the central cardiovascular control during the post-infarct state. Regul Pept 142 (3): 86–94. D’Angelo, G., J. D. Mintz, J. E. Tidwell, A. M. Schreihofer, D. M. Pollock, and D. W. Stepp. 2006. Exaggerated cardiovascular stress responses and impaired beta-adrenergic-mediated pressor recovery in obese Zucker rats. Hypertension 48 (6): 1109–1115. Dasheiff, R. M. 1991. Sudden unexpected death in epilepsy: A series from an epilepsy surgery program and speculation on the relationship to sudden cardiac death. J Clin Neurophysiol 8 (2): 216–222.

624 Sudden Death in Epilepsy: Forensic and Clinical Issues Di Gennaro, G., P. P. Quarato, F. Sebastiano, V. Esposito, P. Onorati, L. G. Grammaldo, G. N. Meldolesi, et al. 2004. Ictal heart rate increase precedes EEG discharge in drug-resistant mesial temporal lobe seizures. Clin Neurophysiol 115 (5): 1169–1177. Dobruch, J., A. Cudnoch-Jedrzejewska, and E. Szczepanska-Sadowska. 2005. Enhanced involvement of brain vasopressin V1 receptors in cardiovascular responses to stress in rats with myocardial infarction. Stress 8 (4): 273–284. Dorian, P. 2005. Antiarrhythmic action of beta-blockers: Potential mechanisms. J Cardiovasc Pharmacol Ther 10 (Suppl 1): S15–S22. Giulumian, A. D., S. G. Clark, and L. C. Fuchs. 1999. Effect of behavioral stress on coronary artery relaxation altered with aging in BHR. Am J Physiol 276 (2 Pt 2): R435–R440. Goldberger, J. J. 1999. Sympathovagal balance: How should we measure it? Am J Physiol. 276 (4 Pt 2): H1273–H1280. Grippo, A. J., T. G. Beltz, and A. K. Johnson. 2003. Behavioral and cardiovascular changes in the chronic mild stress model of depression. Physiol Behav 78 (4–5): 703–710. Grippo, A. J., T. G. Beltz, R. M. Weiss, and A. K. Johnson. 2006. The effects of chronic fluoxetine treatment on chronic mild stress-induced cardiovascular changes and anhedonia. Biol Psychiatry 59 (4): 309–316. Grippo, A. J., J. A. Moffitt, and A. K. Johnson. 2002. Cardiovascular alterations and autonomic imbalance in an experimental model of depression. Am J Physiol Regul Integr Comp Physiol 282 (5): R1333–R1341. Hohnloser, S. H. 2005. Ventricular arrhythmias: Antiadrenergic therapy for the patient with coronary artery disease. J Cardiovasc Pharmacol Ther 10 (Suppl 1): S23–S31. Kleiger, R. E., J. P. Miller, J. T. Bigger Jr., and A. J. Moss. 1987. Decreased heart rate variability and its association with increased mortality after acute myocardial infarction. Am J Cardiol 59 (4): 256–262. Krittayaphong, R., W. E. Cascio, K. C. Light, D. Sheffield, R. N. Golden, J. B. Finkel, G. Glekas, G.€G. Koch, and D. S. Sheps. 1997. Heart rate variability in patients with coronary artery disease: Differences in patients with higher and lower depression scores. Psychosom Med 59 (3): 231–235. La Rovere, M. T., G. D. Pinna, S. H. Hohnloser, F. I. Marcus, A. Mortara, R. Nohara, J. T. Bigger Jr., A. J. Camm, P. J. Schwartz, on behalf of the Autonomic Tone and Reflexes After Myocardial Infarction (ATRAMI) Investigators. 2001. Baroreflex sensitivity and heart rate variability in the identiἀcation of patients at risk for life-threatening arrhythmias: Implications for clinical trials. Circulation 103 (16): 2072–2077. La Rovere, M. T., and P. J. Schwartz. 1997. Baroreflex sensitivity as a cardiac and arrhythmia mortality risk stratiἀer. Pacing Clin Electrophysiol 20 (10 Pt 2): 2602–2613. Lathers, C. M., and P. L. Schraeder. 2002. Clinical pharmacology: Drugs as a beneἀt and/or risk in sudden unexpected death in epilepsy? J Clin Pharmacol 42 (2): 123–136. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9 (2): 236–242. Lear-Kaul, K. C., L. Coughlin, and M. J. Dobersen. 2005. Sudden unexpected death in epilepsy: A retrospective study. Am J Forensic Med Pathol 26 (1): 11–17. Leung, H., P. Kwan, and C. E. Elger. 2006. Finding the missing link between ictal bradyarrhythmia, ictal asystole, and sudden unexpected death in epilepsy. Epilepsy Behav 9 (1): 19–30. Liao, D., J. Cai, R. W. Barnes, H. A. Tyroler, P. Rautaharju, I. Holme, and G. Heiss. 1996. Association of cardiac autonomic function and the development of hypertension: The ARIC study. Am J Hypertens 9 (12 Pt 1): 1147–1156. London, B., L. C. Baker, P. Petkova-Kirova, J. M. Nerbonne, B. R. Choi, and G. Salama. 2007. Dispersion of repolarization and refractoriness are determinants of arrhythmia phenotype in transgenic mice with long QT. J Physiol 578 (Pt 1): 115–129. Mayer, H., F. Benninger, L. Urak, B. Plattner, J. Geldner, and M. Feucht. 2004. EKG abnormalities in children and adolescents with symptomatic temporal lobe epilepsy. Neurology 63 (2): 324–328.

Sympathetic Nervous System Dysregulation of Cardiac Function

625

McDougall, S. J., R. E. Widdop, and A. J. Lawrence. 2005. Differential gene expression in WKY and SHR brain following acute and chronic air-puff stress. Brain Res Mol Brain Res 133 (2): 329–336. McGugan, E. A. 1999. Sudden unexpected deaths in epileptics—A literature review. Scott Med J 44 (5): 137–139. Metcalf, C. S., P. B. Radwanski, and S. L. Bealer. 2009. Status epilepticus produces chronic alterations in cardiac sympathovagal balance. Epilepsia 50 (4): 747–754. Mukherjee, S., M. Tripathi, P. S. Chandra, R. Yadav, N. Choudhary, R. Sagar, R. Bhore, R. M. Pandey, and K. K. Deepak. 2009. Cardiovascular autonomic functions in well-controlled and intractable partial epilepsies. Epilepsy Res 85 (2–3): 261–269. Nei, M., R. T. Ho, B. W. Abou-Khalil, F. W. Drislane, J. Liporace, A. Romeo, and M. R. Sperling. 2004. EEG and ECG in sudden unexplained death in epilepsy. Epilepsia 45 (4): 338–345. Nilsson, L., B. Y. Farahmand, P. G. Persson, I. Thiblin, and T. Tomson. 1999. Risk factors for sudden unexpected death in epilepsy: A case-control study. Lancet 353 (9156): 888–893. Opeskin, K., M. P. Burke, S. M. Cordner, and S. F. Berkovic. 1999. Comparison of antiepileptic drug levels in sudden unexpected deaths in epilepsy with deaths from other causes. Epilepsia 40 (12): 1795–1798. Opeskin, K., and S. F. Berkovic. 2003. Risk factors for sudden unexpected death in epilepsy: A controlled prospective study based on coroners cases. Seizure 12 (7): 456–464. P-Codrea Tigaran, S., S. Dalager-Pedersen, U. Baandrup, M. Dam, and A. Vesterby-Charles. 2005. Sudden unexpected death in epilepsy: Is death by seizures a cardiac disease? Am J Forensic Med Pathol 26 (2): 99–105. Ronkainen, E., J. T. Korpelainen, E. Heikkinen, V. V. Myllyla, H. V. Huikuri, and J. I. Isojarvi. 2006. Cardiac autonomic control in patients with refractory epilepsy before and during vagus nerve stimulation treatment: A one-year follow-up study. Epilepsia 47 (3): 556–562. Rugg-Gunn, F. J., R. J. Simister, M. Squirrell, D. R. Holdright, and J. S. Duncan. 2004. Cardiac arrhythmias in focal epilepsy: A prospective long-term study. Lancet 364 (9452): 2212–2219. Sathyaprabha, T. N., P. Satishchandra, K. Netravathi, S. Sinha, K. Thennarasu, and T. R. Raju. 2006. Cardiac autonomic dysfunctions in chronic refractory epilepsy. Epilepsy Res 72 (1): 49–56. Schimpf, R., C. Veltmann, C. Wolpert, and M. Borggrefe. 2009. Channelopathies: Brugada syndrome, long QT syndrome, short QT syndrome, and CPVT. Herz 34 (4): 281–288. Shah, M., F. G. Akar, and G. F. Tomaselli. 2005. Molecular basis of arrhythmias. Circulation 112 (16): 2517–2529. Shimizu, M., A. Kagawa, T. Takano, H. Masai, and Y. Miwa. 2008. Neurogenic stunned myocardium associated with status epileptics and postictal catecholamine surge. Intern Med 47 (4): 269–273. Simon, R. P., M. J. Aminoff, and N. L. Benowitz. 1984. Changes in plasma catecholamines after tonic–clonic seizures. Neurology 34 (2): 255–257. Stein, P. K., M. S. Bosner, R. E. Kleiger, and B. M. Conger. 1994. Heart rate variability: A measure of cardiac autonomic tone. Am Heart J 127 (5): 1376–1381. Stein, P. K., and R. E. Kleiger. 1999. Insights from the study of heart rate variability. Annu Rev Med 50: 249–261. Task Force of the European Society of Cardiology the North American Society of Pacing Electrophysiology. 1996. Heart rate variability: Standards of measurement, physiological interpretation and clinical use. Circulation 93 (5): 1043–1065. Tennis, P., T. B. Cole, J. F. Annegers, J. E. Leestma, M. McNutt, and A. Rajput. 1995. Cohort study of incidence of sudden unexplained death in persons with seizure disorder treated with antiepileptic drugs in Saskatchewan, Canada. Epilepsia 36 (1): 29–36. Tigaran, S., V. Rasmussen, M. Dam, S. Pedersen, H. Hogenhaven, and B. Friberg. 1997. ECG changes in epilepsy patients. Acta Neurol Scand 96 (2): 72–75. Tomson, T., M. Ericson, C. Ihrman, and L. E. Lindblad. 1998. Heart rate variability in patients with epilepsy. Epilepsy Res 30 (1): 77–83.

626 Sudden Death in Epilepsy: Forensic and Clinical Issues Ueda, K., C. Valdivia, A. Medeiros-Domingo, D. J. Tester, M. Vatta, G. Farrugia, M. J. Ackerman, and J. C. Makielski. 2008. Syntrophin mutation associated with long QT syndrome through activation of the nNOS-SCN5A macromolecular complex. Proc Natl Acad Sci U S A 105 (27): 9355–9360. van Ravenswaaij-Arts, C. M., L. A. Kollee, J. C. Hopman, G. B. Stoelinga, and H. P. van Geijn. 1993. Heart rate variability. Ann Intern Med 118 (6): 436–447. Vanoli, E., P. B. Adamson, R. D. Foreman, and P. J. Schwartz. 2008. Prediction of unexpected sudden death among healthy dogs by a novel marker of autonomic neural activity. Heart Rhythm 5 (2): 300–305. Vatta, M., M. J. Ackerman, B. Ye, J. C. Makielski, E. E. Ughanze, E. W. Taylor, D. J. Tester, R. C. Balijepalli, J. D. Foell, Z. Li, T. J. Kamp, and J. A. Towbin. 2006. Mutant caveolin-3 induces persistent late sodium current and is associated with long-QT syndrome. Circulation 114 (20): 2104–2112. Verrier, R. L., and C. Antzelevitch. 2004. Autonomic aspects of arrhythmogenesis: The enduring and the new. Curr Opin Cardiol 19 (1): 2–11. Wu, G., T. Ai, J. J. Kim, B. Mohapatra, Y. Xi, Z. Li, S. Abbasi, E. Purevjav, K. Samani, M. J. Ackerman, M. Qi, A. J. Moss, W. Shimizu, J. A. Towbin, J. Cheng, and M. Vatta. 2008. alpha-1-syntrophin mutation and the long-QT syndrome: A disease of sodium channel disruption. Circ Arrhythm Electrophysiol 1 (3): 193–201. Yamashita, T., Y. Murakawa, N. Hayami, E. Fukui, Y. Kasaoka, M. Inoue, and M. Omata. 2000. Shortterm effects of rapid pacing on mRNA level of voltage-dependent K(+) channels in rat atrium: Electrical remodeling in paroxysmal atrial tachycardia. Circulation 101 (16): 2007–2014.

The Urethane/Kainate Seizure Model as a Tool to Explore Physiology and Death Associated with Seizures

39

Mark Stewart

Contents 39.1 Introduction 39.2 Autonomic Consequences of Seizures 39.3 The Urethane/Kainate Model 39.4 Results Relevant to Sudden Death in Epilepsy 39.5 Quantitative Activity Differences 39.6 Pathways 39.7 Insights into the Mechanism for Sudden Death in Epilepsy 39.8 Closing Comments on This and Other Animal Models References

627 628 629 630 636 637 638 639 639

39.1â•…Introduction The anesthetized whole animal or acute preparation has been supplanted in recent decades by a choice of preparations that offer different advantages. Brain slices and cell cultures offer much better access to the neurons for physiological and pharmacological studies; and provided the questions addressed relate to relatively simple neuronal circuits, the preparations are superior to an intact animal preparation—the specimen visualization and recording advantages (intracellular, whole cell patch, calcium and voltage sensitive dyes, neuronal imaging) are signiἀcant. In preparations where the behavior is important (e.g., to observe convulsions or maze learning), the freely moving, freely behaving animal preparation is the only choice. The acute animal preparation in fact has signiἀcant advantages for neurophysiology and physiology studies. Among these is access to measurements and recordings that can be impossible in freely moving animals. In the whole animal preparation, one has access to the entire central nervous system circuitry, something that cannot be preserved in a brain slice or cell culture. The acute animal preparation has been reexamined by us for studies of the autonomic consequences of seizures because we need access to the entire physiology of the animal. One has options for a host of invasive measures or procedures and considerable control of an animal’s condition (e.g., ventilation, temperature, blood pressure, and even blood volume). It is the preparation to explore the limits of the physiology associated with functional or anatomical neuropathology. One can directly address questions such as, Can a seizure 627

628 Sudden Death in Epilepsy: Forensic and Clinical Issues

cause death? If so, how? If not, why not? Can autonomic nervous system activity itself cause death? These questions and many others can be explored in detail to address fundamental questions of physiology. In addition to its flexibility, the acute animal preparation offers efficiency that comes from the control of conditions such as seizure activity to eliminate issues of “capturing” the right conditions. The trade-off, of course, is that although the animal is intact, the conditions of the animal are always a subset of the conditions that would be observed during “captured” events in behaving animals. In short, the acute whole animal preparation is ideally suited for efficiently deἀning boundary conditions and making certain kinds of manipulations or measurements that would later guide studies of activity such as seizures in behaving animals.

39.2â•… Autonomic Consequences of Seizures In the context of sudden death in epilepsy, most would agree that the autonomic nervous system provides the link between seizures and death. Autonomic dysfunction during seizures can have serious clinical consequences (Devinsky 2004; Goodman et al. 2008). Changes in heart rate and rhythm and blood pressure can occur during complex partial seizures (Baumgartner et al. 2001; Lathers et al. 1998; Locatelli et al. 1999; Nei et al. 2000; Wilder-Smith 1992). Generalized tonic–clonic seizures are sometimes associated with severe increases in blood pressure and arrhythmias (Schraeder and Lathers 1989), and some individuals experience more ominous autonomic derangements such as marked bradycardia or asystole (Smith-Demps and Jagoda 1998; Cheung and Hachinski 2000; Mameli et al. 2001). The causes of sudden unexpected death in epilepsy (SUDEP) are suspected to involve cardiovascular or respiratory dysfunction provoked by seizures (Cheung and Hachinski 2000; Langan 2000; Liedholm and Gudjonsson 1992; Mameli et al. 2001; SmithDemps and Jagoda 1998; Tinuper et al. 2001). Animal experimentation has replicated features of human epilepsy and autonomic dysfunction associated with seizures (reviewed by Lathers and Schraeder 2006). Amygdalakindled seizures in rats coactivate sympathetic and parasympathetic systems and were shown to produce ictal hypertension and bradycardia (Goodman et al. 1990, 1999). Left insular cortex stimulation in rats caused degrees of heart block (Oppenheimer et al. 1991), resembling ἀndings from a study of epilepsy surgery patients (Oppenheimer et al. 1992; Hilz et al. 2001). In our own model (described in more detail below), left-sided limbic seizure activity was associated with a more pronounced vagus nerve activity increase and associated blood pressure decrease (Saito et al. 2006) than right-sided limbic seizure activity. Focal and generalized seizures in anesthetized cats (Schraeder and Lathers 1983; Lathers et al. 1998, 1993) were associated with cardiac sympathetic and parasympathetic activity that was intermittently synchronized with interictal spike activity (“lockstep phenomena”) (Lathers et al. 1987; Stauffer et al. 1989). Application of penicillin G to the hypothalamus and mesencephalic centers of hemispherectomized rats produced increases in vagus nerve activity and various cardiovascular disturbances (Mameli et al. 2001, 1993) including death (Mameli et al. 2001, 2006). Mice that exhibit audiogenic seizures frequently display respiratory arrest that can be fatal and prevented by increasing oxygen in the environment (Venit et al. 2004; Willott and Henry 1976). Hypoventilation was reported as an important correlate of death during seizures in a sheep model (Johnston et€al. 1997, 1995).

The Urethane/Kainate Seizure Model A

Urethane

B

Ketamine+xylazine

Hippo EEG Face EMG Leg EMG

Hippo EEG Face EMG Leg EMG

629

5s

1s

Figure 39.1╇ Urethane permits limbic cortical seizure activity with minimal motor convul-

sions. (A) Urethane-anesthetized rat (1.5 g/kg i.p. with intra-arterial (i.a.) supplements). Three 50-s sweeps of simultaneously recorded (a) EEG (dorsal hippocampus area CA1), (b) EMG from face (in the area of vibrissae), and (c) EMG from the lower leg (gastrocnemius and soleus muscles). Seizure activity after systemic kainic acid (10 mg/kg) is present throughout all three sweeps, with some changes in frequency components. Very low EMG activity levels from leg recordings (third trace in each set) did not change during seizures. Minimal EMG activity is apparent in the face EMG recordings, but rarely (indicated by asterisks in sweeps 2 and 3). (B) Ketamine and xylazine–anesthetized rat (60 mg/kg ketamine + 10 mg/kg xylazine i.p. with i.a. supplements). Similar set of three 50-s sweeps with simultaneously recorded dorsal hippocampal EEG, and EMG from face and hind leg. Seizure activity is more pronounced in some, but not all, of the time shown. Most significantly, there was substantial EMG activity in all recording locations. Thick black bar indicates time period for faster sweep shown below. Synchronous phasic activity in both the face and leg EMG electrodes is clearly apparent. Motor convulsions were stopped by Nembutal (10 mg i.a.). EEG spikes are clipped in the display. Calibrations: time, 5 s for both A and B; voltage, 400 µV for EEG A and B, 600 µV for EMG A face and B leg, 400 µV for EMG A leg, and 300 µV for EMG B face. (From Figure 2 in Saito, T., et al., J Neurosci Methods, 155, 241–50, 2006.)

39.3â•… The Urethane/Kainate Model We have developed an acute rat preparation based on urethane anesthesia and systemic or focal kainic acid to deἀne the autonomic consequences of seizure activity. Initially, we€compared anesthetics such as urethane and ketamine/xylazine combinations for acute seizure studies because these anesthetics permit several kinds of polysynaptic activity (Flecknell

630 Sudden Death in Epilepsy: Forensic and Clinical Issues

1996). Hippocampal theta rhythm is one example of a complex polysynaptic brain oscillation that is readily and well studied in urethane-anesthetized animals (Brankack et€al. 1993; Stewart and Fox 1990). In urethane-anesthetized rats, systemic kainic acid can cause repeated long episodes of seizure activity ranging in duration from 10 s to more than 1€min€(Saito et al. 2006; Sakamoto et al. 2008; Hotta et al. 2009a). Seizures are typically characterized by an initial period of high-frequency activity that progresses to slower, larger amplitude waves until the seizure episode terminates leaving low-voltage activity in the EEG until the next seizure episode. Remarkably, the EMG during seizures in urethaneanesthetized rats is essentially flat from all locations except the face, even during limbic status epilepticus (Saito et al. 2006). Animals breathe spontaneously, but motor convulsions during seizure activity are absent. By contrast, motor convulsions accompany electroencephalographic seizures in animals anesthetized with ketamine/xylazine (Saito et al. 2006). Signiἀcant EMG activity can be recorded from multiple locations, and rhythmic muscle contractions corresponded to periodic activity in the EEG. These differences are highlighted in Figure 39.1. The fact that seizure activity in the urethane/kainate model was conἀned to limbic cortical areas without spread to neocortical areas meant that we had access to the autonomic consequences of seizure activity in an animal preparation that was breathing spontaneously and could be mechanically stable enough for many kinds of recordings. There was no need to paralyze the animal. Titration of the urethane level and the amount of kainic acid ensured mechanical stability and offered some flexibility in the durations of seizure activity, ranging from very brief to frank limbic status epilepticus. The reason for the limbic cortical–neocortical disconnect is unclear. Although urethane has been used for studies of activity in limbic cortical structures (e.g., hippocampal theta rhythm), including cingulate cortex (Feenstra and Holsheimer 1979), urethane has been described as anticonvulsant in somatosensory cortex (Heltovics et al. 1995). There is evidence that urethane can suppress glutamate release at some cortical synapses (Moroni et al. 1981).

39.4â•…Results Relevant to Sudden Death in Epilepsy Using the urethane/kainate model, we have deἀned massive increases in parasympathetic and sympathetic outflow that occur during limbic cortical seizures and the cardiovascular consequences of these changes (Saito et al. 2006; Hotta et al. 2009a; Sakamoto et al. 2008; Stewart 2008, Hotta et al. 2009b). Increases were 8–10 times baseline rates for parasympathetic (vagus nerve) activity and somewhat less in sympathetic pre-€and postganglionic nerve recordings (Sakamoto et al. 2008; Hotta et al. 2009a). Peripheral nerve activity increases during seizures were greater than increases induced by nitroprusside or phenylephrine that produced mean arterial pressure changes of >50 mm Hg. Increases in c-fos expression were found in both sympathetic and parasympathetic medullary regions (as well as hypothalamic areas) (Sakamoto et al. 2008). Changes in nerve activity induced by blood pressure increases or decreases were smaller or absent during tests made in€the€seizure state, indicating that baroreceptor reflex function was impaired during seizures. The normal relations of cardiac sympathetic nerve activity to ventilation (Kollai and Koizumi 1979) were lost when ventilation rate changes were tested during limbic seizures (Hotta et al. 2009a). Both the direction of the nerve activity change and the magnitudes of changes

The Urethane/Kainate Seizure Model

631

were unpredictable during seizures. Acute cardiac dilatation accompanied the bradyarrhythmia that occurred during seizures or seizures plus hypercapnia/hypoxia (Sakamoto et al. 2008). Cardiac dilatation depended on autonomic coactivation (Hotta et al. 2009b) (Figures 39.2 through 39.4). A parallel, nonreciprocal activation of the sympathetic and parasympathetic outflow€to€the heart can occur physiologically. Koizumi and others showed such activity in response to hypoxia (Koizumi et al. 1986; Fukuda et al. 1989) and during hypothalamic stimulation€(Koizumi and Kollai 1981). Coactivation enhanced cardiac output (Koizumi Sweep 1

BP Hippo EEG

Seizure begins

Vagus Sweep 2

0.5 mV 2 mV 20 mm Hg

Sweep 3

Sweep 4 Seizure ends

1s

Figure 39.2╇ Vagus nerve activity during a complete limbic seizure episode. Sweeps from a single kainic acid–induced seizure in a urethane-anesthetized rat. Each individual seizure episode typically begins with a period of very high-frequency activity that progressively slows and synchronizes (making larger amplitude EEG waves). Slowing continues until the large amplitude activity stops. Each panel shows BP, dorsal hippocampal EEG, and vagus nerve activity. Note the changes in frequency components of the seizure as it progresses, and the increase in vagus nerve activity, especially during the middle portion (sweeps 2 and 3). Bursts in the vagus nerve recording associated with late slow spikes in the seizure episode are less apparent than those shown in Figure 39.3. BP increases as the seizure begins, and again as the seizure ends. As the vagus nerve activity increases, there is also a decrease in the pulse pressure. Calibrations (and major divisions) are x-axis (time) = 1 s; y-axis (top, BP) = 20 mm Hg, y-axis (middle, hippocampal EEG) = 2.0 mV, and y-axis (bottom, vagus nerve recording) = 0.5 mV. Note that major divisions in the BP records for sweeps 3 and 4 are 10 mm Hg. Triangles at left indicate 180 mm Hg. Horizontal lines at left edge of some sweeps are where traces were cropped by removing trace labels. Each sweep is 30 s long. (From Figure 1 in Sakamoto, K., et al., Epilepsia, 49, 982–996, 2008.)

632 Sudden Death in Epilepsy: Forensic and Clinical Issues Cardiac sympathetic nerve

First half of seizure episode

EEG ECG

Ventilation 0.2 mV

Second half of seizure episode 0.4 mV 2 mV 10 mm Hg 10 s

Figure 39.3╇ Cardiac sympathetic nerve activity during a limbic seizure episode. Top to bot-

tom: cardiac sympathetic multiunit nerve activity, EEG, ECG, and a record ventilator air movements. There is a modest increase in the cardiac sympathetic nerve activity during the seizure episode. The entire time period is split in half. The first part of the recording is shown above and the second part is shown below. The seizure episode begins approximately halfway through the top set of traces and ends approximately halfway through the bottom set of traces. Pre-seizure vs. seizure comparisons can be made by looking across the top row or comparing the first part of the top traces with the first part of the bottom traces. Similarly, seizure vs. postseizure comparisons can be made by comparing across the bottom set of traces or the second parts of the top and bottom. Calibrations are given for all traces and labeled in the bottom set of traces. (From Figure 1 in Hotta, H., et al., Epilepsia, 50, 923–927, 2009a.)

et€al.€1982).€During periods of intense parasympathetic activity, sympathetic coactivation helped to preserve blood pressure in the face of bradyarrhythmia and thereby facilitate venous return and better ventricular ἀlling (Hotta et al. 2009b). During severe vagus nerve stimulation (≥20 Hz), the baroreceptor reflex response was not sufficient to support ventricular ἀlling, evidenced by the small cavity size in echocardiographic images. Strong sympathetic coactivation, either with pharmacological activation of adrenergic receptors or indirect sympathetic activation via hypercapnia/hypoxia, improved blood pressure, which improved venous return and ventricular ἀlling. The drawback of sympathetic coactivation seems to be impaired contractility (Hotta et al. 2009b). Catecholamines are generally viewed as compounds that increase myocardial contractility but accelerate myocardial relaxation (lusitropy). On echocardiography, however, it appeared that the contractions were actually weaker, suggesting that there is an element of sympathetic enhancement of parasympathetic activity. This is consistent with ἀndings that a background of sympathetic activity intensiἀes cardiac responses to parasympathetic input (Randall et al. 2003) either directly at cardiac myocytes or via interactions within the intrinsic cardiac plexus (Ardell 2004).

The Urethane/Kainate Seizure Model

633

Parasympathetics 6 5 4

KA

KA

KA

3 2 1 0

PE

PE

NP

PE NP

NP

Vagus (all)

Vagus (left)

Vagus (right)

Sympathetics 6 5 4 3 2 1 0

NP

NP PE

PE Cerv gang (all)

KA

KA

KA

Cerv gang (left)

PE NP Cerv gang (right)

6 5 4 3 2

KA NP

1 0

PE Renal (left)

Figure 39.4╇ Relative activity in autonomic nerve recordings. Summary of activity changes using a normalized activity scale (1 = baseline, 2 = 2× baseline, 3 = 4× baseline, 4 = 6× baseline, . . . , 7 = 12× baseline and above) for all recorded nerves. Plots are for parasympathetics (vagus nerve) and sympathetics (cervical sympathetic ganglion, renal sympathetic nerve, grater splanchnic nerve). For the vagus nerve recordings and cervical sympathetic ganglia recordings, right or left side data are also shown. Some recordings from each of these groups did not have the side recorded, so they appear only in the pooled totals. Nitroprusside (NP) and phenylephrine (PE) tested before kainic acid (KA) are shown to the left of the KA data. NP and PE tests made during seizures are shown to the right of the KA data. Stars indicate statistically significant increases or decreases (pre vs. post for that particular treatment). p < 0.0001 for vagus nerve significance levels except PE pre/post for the right vagus nerve data where p < 0.0005. p < 0.0001 for sympathetic nerves and ganglia (all), KA pre/post for cervical sympathetic ganglion (all, left). p < 0.01 for all other significance levels. (Adapted from Table 1 in Sakamoto, K., et al., Epilepsia, 49, 982–996, 2008.)

634 Sudden Death in Epilepsy: Forensic and Clinical Issues BP ECG EEG Ventilation

BP ECG EEG Ventilation 1

2

3

4

5

Figure 39.5╇ “Flat line” EEG in a rat model of death during seizures. Urethane-anesthetized rats receiving kainic acid as a convulsant have recurring limbic cortical seizures. (From Saito, T., et al., J Neurosci Methods, 155, 241–250, 2006; Sakamoto, K., et al., Epilepsia, 49, 982–996, 2008. With permission.) Intense seizure activity or seizures combined with asphyxia produce a profound bradycardia and cardiac dilatation leading to hypoperfusion of the brain, the termination of ongoing seizure activity, and eventually a flatline EEG. The top panel shows the full experiment and specific times are highlighted below. The set of sweeps in each panel are, from top to bottom, arterial blood pressure, ECG, EEG, and a pressure transducer on the ventilator tubing. The output of the ventilator produces upward deflections, and spontaneous “pulling” against the closed air system produces downward deflections. The horizontal bar at the bottom of the top panel shows the time during which the ventilator was stopped. This is also apparent as the switch from positive to negative deflections on the ventilation record. The seizure continues for a short time after the ventilator has stopped (panel 2, bottom), but is arrested in less than 1 min (panel 3, bottom), and the EEG is a flat line by panel 4. Bradycardia and AV block are evident in panel 4. As the force of the attempts to inspire increased (larger negative deflections), there were larger movement artifacts in the ECG and EEG recordings (panel 4, bottom). The combination of restored ventilation and seizure suppression permitted recovery of cardiac performance and brain perfusion, with recovery of brain activity (in this case, additional seizures). Calibrations are BP, 20 mm Hg; ECG, 1 mV; EEG, 2 mV; ventilator, 20 mm Hg; and time, 5 s. Data were collected together with Dr. Harumi Hotta of the Tokyo Metropolitan Institute of Gerontology, Tokyo, Japan. (Taken from Stewart, M., J Neurol Neurosurg Psychiatry, e-letter, February 12, 2008.)

The mechanism of death in these animals was deἀned through ECG, blood pressure, and echocardiographic measures to be profound mechanical dysfunction coupled with sinus bradycardia and AV nodal block leading to hypoperfusion of the brain and ἀnally hypoperfusion of the heart itself (Sakamoto et al. 2008). The physiological picture we ἀnd during seizures resembles the picture found during asphyxiation. The time scale for seizures and/or hypoxia and hypercapnia to impact the heart is seconds, and the development of cardiac dilatation with bradyarrhythmia and eventual death can occur on a time scale of minutes. These changes result from a massive combined activation of the parasympathetic

The Urethane/Kainate Seizure Model Baseline

635

During 2 Hz VNS 0.6

During 5 Hz VNS 0.6

0.7

During 50 Hz VNS

0.4

0.4

1 cm 0.2 s

Baseline

During 50 Hz VNS + re-breathing

0.5

0.8

0.9

Figure 39.6╇ Electrocardiogram and M-mode echocardiographic images during VNS. Top left is a series of beats in the baseline condition. Dots below the panel indicate the times for p waves in the ECG (gray trace toward the bottom of each panel). Low-frequency VNS (2, 5 Hz) is shown in the top row. To the right edge of each panel, the maximal ventricular cavity size is marked. A modest increase in cavity size is evident as the heart rate goes down. In the middle row are two examples of isolated beats during 50 Hz VNS. The smaller size of the left ventricular cavity€is evident and indicated to the right of each panel. VNS (50 Hz) during hypercapnia/hypoxia is associated with cardiac dilatation. A series of tiny vertical white dots at about one-fourth of the way along the sweep (between the left edge of the image and the QRS complex) is the distance calibration of 1 cm between successive dots. Horizontal distance between successive tick marks along the top of each panel is 200 ms. (From Hotta, H., et al., Epilepsia, 50, 923–927, 2009b.)

and sympathetic divisions of the autonomic nervous system that is driven by seizures and compounded by respiratory obstruction or impaired ventilation. One question is whether seizures alone are enough to cause death. In these experiments and in our own kindling studies in freely moving rats, seizures can clearly cause serious arrhythmias, but each episode has been transient and therefore not life-Â�threatening. Changing blood flow (e.g., carotid occlusion) and lowering blood oxygen content (e.g., respiratory obstruction) are both processes that can rapidly terminate seizure activity. Transient unilateral or bilateral carotid occlusion terminated seizure activity in the ipsilateral hemisphere or bilaterally in seconds (Saito et al. 2006). Seizures combined with respiratory obstruction or rebreathing air from a balloon caused cardiovascular dysfunction that terminated seizure activity on a time scale of many seconds to minutes (Sakamoto et al. 2008) (Figure 39.5). Because a purely seizure-driven autonomic “storm” would stop when the brain blood flow was compromised, we used vagus nerve stimulation (VNS) in urethane-anesthetized rats to explore the limits of cardiac dysfunction during autonomic overactivity (Hotta et al. 2009b). Based on our previous ἀndings, vagus nerve ἀring rates about 10 times over baseline can occur during seizures (in a combined sympathetic and parasympathetic activity surge) and the highest rates can be achieved during the combination of seizure

636 Sudden Death in Epilepsy: Forensic and Clinical Issues

and asphyxia, or asphyxia alone (Sakamoto et al. 2008). With VNS, we deἀned a baseline parasympathetic “tone” of about 2 Hz and showed that the most severe bradyarrhythmia was produced by 50 Hz VNS. The cardiac consequences of VNS were enhanced by decreasing body temperature, and we could achieve cardiac standstill for many seconds before ventricular escape rhythms developed in ventilated animals. We showed that blood pressure during VNS is relatively maintained during lower-frequency VNS, but collapses at frequencies ≥20 Hz to dramatically impair ventricular ἀlling. Absolutely critical was the ἀnding that heart rate and blood pressure returned to near baseline levels within several seconds after VNS stopped, even after 5-min stimulus trains. When additional sympathetic activity was produced by systemic alpha- or beta-receptor agonists or by hypercapnia, hypoxia, or hypercapnia plus hypoxia, there was a small improvement in mean arterial pressure, much better venous return leading to cardiac dilatation, but even worse cardiac performance (e.g., fractional shortening, ejection fraction) (Figure 39.6). Abrupt increases in parasympathetic activity on the order of 5 times the background parasympathetic tone were sufficient to produce transient bradyarrhythmias, and abrupt increases on the order of 20 times can produce cardiac standstill, sometimes accompanied by apnea. This can only be the mechanism for death when there is airway obstruction because cardiac performance recovery was always rapid after the parasympathetic discharge stopped and, in the case of seizure activity, the decreased cardiac output during intense parasympathetic activity cannot support seizure activity (Sakamoto et al. 2008; Stewart 2008; Saito et al. 2006). More severe or entirely different autonomic disturbances might result when seizures occur on the background of a damaged brain rather than on the background of a normal brain. Another possibility is that episodes of arrhythmia predispose to more serious arrhythmias (J. A. Armour, personal communication, 2008)

39.5â•… Quantitative Activity Differences The notion that autonomic dysfunction arises from differences in activity of the two hemispheres predates suggestions that asymmetric seizure activity may contribute to sudden death in epileptic patients. The “brain-laterality hypothesis” was proposed to account for cases of sudden cardiac death, wherein patients died during extreme stress (Lane and Jennings 1998; Galin 1974). The brain-laterality hypothesis was that emotional arousal could trigger ventricular ἀbrillation and sudden death by inducing a net lateralized imbalance in the sympathetic outflow to the heart. Hemispheric inactivation studies have supported this idea (Hachinski et al. 1992; Zamrini et al. 1990). Differences in activity in the two hemispheres have also been proposed as a contributor or cause for autonomic disturbances in seizure patients, including death (Hilz et al. 2001; Oppenheimer 2001). Controlled lateralized differences in activity are difficult to ἀnd or cause in studies of seizure activity because seizures can spread rapidly in the brain, and most examples of death during or after a seizure involve generalized seizure activity. A different kind of asymmetry involves differential activation of dorsal and ventral portions of the limbic cortices. Part of this notion comes from data that ventral hippocampal areas have a greater tendency to generate seizure activity than dorsal areas (Gilbert et al. 1985). The connectivity of hippocampal formation areas with brain regions having autonomic impact, for example, insular cortex, is also heaviest from ventral areas (Saper 1982a; Witter et al. 1988).

The Urethane/Kainate Seizure Model

637

Other studies suggest that the volume or extent of the cortical involvement, not lateralization of seizure activity, is important for signiἀcant heart rate changes (Epstein et al. 1992; Britton et al. 2006). The Epstein study not only describes predominantly heart rate increases in patients but also mentions patients who deviated from the majority, showing heart rate slowing and sinus pauses. Ictal bradycardia was found to most often occur in patients with temporal lobe seizures when the seizure activity was bilateral (Britton et al. 2006). Furthermore, the current data suggest that sudden death is not consistently associated with the most severe seizures (Goodman et al. 2008). In fact, in kindling animal models (Goodman et al. 1990; Kunitake and Stewart, unpublished data), arrhythmias were often seen early in the kindling progression (e.g., stage 2 seizures); however, none of these were fatal arrhythmias. Although we have seen autonomic nervous system activity differences depending on the hemisphere exhibiting seizure activity (Saito et al. 2006), our nerve recordings and c-fos data indicate that both divisions of the autonomic nervous system are strongly activated during kainic acid–induced limbic seizure activity in rats. Furthermore, our data indicate that death results from massive combined autonomic nervous system overactivity, with the overall result being a parasympathetic-dominated cardiovascular state. In our opinion, cardiovascular complications depend on the degree and duration of increased autonomic activity, and therefore likely depend on the extent and severity of the seizure plus associated respiratory/ventilation complications. Based on our data, the predominantly parasympathetic effect at the heart is due to quantitative differences between sympathetic and parasympathetic outflow. This aspect of€the physiology appears to be quite consistent, even across various seizure models and various animal species, including rats with chronic seizure conditions. That said, there are a number of reports of seizures associated with a predominantly sympathetic state, even in€the same animal preparation during continuous mechanical ventilation (Lathers and Schraeder 1982). What can account for such autonomic variability? We have shown that sympathetic and parasympathetic activity changes accompanying ventilation rate (Hotta et al. 2009a) or blood pressure (Sakamoto et al. 2008) changes can become unpredictable. These become signiἀcant sources for altering the nerve activity patterns that occur during a seizure. An additional source of variability can be speciἀc seizure activity patterns that result in speciἀc autonomic responses, but we are just beginning to explore speciἀc seizure activity patterns.

39.6â•… Pathways Neocortex (Allen et al. 1991; Saper 1982b, 2000; Yasui et al. 1991) and limbic cortices (Witter and Amaral 2004; Witter et al. 1989) are heavily connected with structures that constitute the autonomic nervous system or with regions that project into autonomic structures. The central nucleus of the amygdala and bed nucleus of the stria terminalis are major inputs to hypothalamic, pontine, and medullary structures that control preganglionic autonomic neurons of both the sympathetic and parasympathetic branches of the autonomic nervous system (Alheid et al. 1995; Loewy 1990). They are positioned to relay cortical activity into the autonomic nervous system (Cullian et al. 1993; Pitakanen et al. 2000; Westerhaus and Loewy 2001; Witter et al. 1989). Neocortex (Saper 1985, 2000) and limbic cortices, especially subiculum (Kishi et al. 2000; Witter et al. 1989; Wouterlood and Tuinhof 1992), also have

638 Sudden Death in Epilepsy: Forensic and Clinical Issues

direct projections to hypothalamus. These latter projections are likely the route by which seizure activity spreads from limbic cortical areas into autonomic areas in the urethane/ kainate model. Autonomic pathways are integrated in the hypothalamus and influence all visceral systems, including cardiovascular, genitourinary, endocrine, and digestive, as well as peripheral involuntary muscles such as pilomotor and pupillary sphincters. Projections of hypothalamic, pontine, and medullary neurons to preganglionic parasympathetic and sympathetic neurons controlling cardiac and respiratory function have been studied anatomically (Loewy et al. 1979; Moga et al. 1990, 1989; Saper 1979; Saper et al. 1976; Tucker and Saper 1985) and physiologically (Chamberlin and Saper 1992, 1998; Gebber et al. 1990, 1987; Gebber and Snyder 1970; Gebber et al. 1995; Spyer 1979; Yamashita et al. 1983). Hypothalamic cells project onto medullary structures (ventrolateral medulla and periaqueductal gray) that project directly into the intermediolateral horn of the spinal cord where the sympathetic preganglionic cell bodies are located. Activity of hypothalamic and medullary neurons correlates with sympathetic nerve activity (Barman and Gebber 1982, 1998; Barman et al. 1999, 2000, 1995). Some hypothalamic cells project directly into the spinal cord (Cechetto and Saper 1988). For the parasympathetic system, central nucleus of the amygdala (Loewy 1990) and hypothalamus (PVN and SON) (Yamashita et al. 1983) project directly into the dorsal motor nucleus of the vagus and the nucleus ambiguus, the sites of parasympathetic preganglionic cell bodies. Furthermore, the sympathetic and parasympathetic paths can influence each other. For example, efferents of the nucleus ambiguus and dorsal nucleus of the vagus (preganglionic parasympathetic neurons) project to the nucleus of the solitary tract that has a relayed projection to intermediolateral horn (sympathetic preganglionic neurons).

39.7â•…Insights into the Mechanism for Sudden Death in Epilepsy In our urethane/kainic acid rat model of seizures, the mechanism for death during autonomic overactivity was a profound bradycardia coupled with cardiac dilatation (mechanical weakness). However, some form of persistent respiratory obstruction was necessary for death because prolonged severe bradycardic episodes were associated with a “flatlining” of the EEG; that is, the seizure stops (Stewart 2008). When seizures are the only stimulus for the autonomic overactivity, impaired cardiac function suppresses seizure activity that eliminates the autonomic overactivity. Cardiac function is rapidly restored. This suggests that a seizure cannot by itself cause death via an autonomically mediated heart failure. Continued hypercapnia/hypoxia appears to be essential, ἀrst, to sustain autonomic overactivity after a seizure stops and, second, to ultimately cause heart failure from hypoxia. It is interesting that the picture we ἀnd during seizures resembles the picture found during asphyxiation. The respiratory distress that can occur during a seizure (e.g., airway mucous overproduction) is a form of asphyxiation and works pathophysiologically in the same direction as the seizures to drive the sympathetic and parasympathetic divisions simultaneously. Seizures drive both the parasympathetics and the sympathetics. Respiratory distress causes an increase in vagal activity because of the increase in arterial carbon dioxide, and it increases sympathetic activity (and blood pressure) that also favors an increase in vagal outflow via any functional baroreceptor activity. The combination is clearly rapidly lethal in our rats. Some “jumps” in parasympathetic outflow were likely large enough to make nodal block the initial or predominant ἀrst event. In the absence of

The Urethane/Kainate Seizure Model

639

sustained seizure activity or respiratory distress, a transient arrhythmia may be all that occurs. We have seen such transient arrhythmias in freely moving rats that were receiving kindling stimulation (Kunitake and Stewart, unpublished data). Death in our animal model does not resemble the vast majority of cases of sudden cardiac death. Although the literature spans some three decades, it is written almost universally that 80–85% of cases of sudden cardiac death are due to ventricular arrhythmias (reviewed by Kirchhof et al. 2006). Sudden death from fatal arrhythmia is most commonly a ventricular tachyarrhythmia. In some patients, a ventricular tachyarrhythmia is triggered by an acute myocardial ischemic event, and in other patients, a ventricular tachyarrhythmia is related to an anatomical substrate (e.g., scarring from a previous myocardial infarct) (Huikuri et al. 2001; Kiviniemi et al. 2007; Lombardi et al. 2001). Such structural cardiac changes may be part of the disease spectrum in patients with epilepsy, exposing them to increased risk of ventricular ἀbrillation. We have seen similar changes in our chronically epileptic rats, but intense seizure activity in these animals leads to the same parasympathetically predominated bradyarrhythmias. Although we have referred to the circumstances surrounding death in our animals as a parasympathetic “point-of-noÂ�return,” an alternative view of the massive parasympathetic outflow is that it is an attempt to resist a ἀbrillation state. Beyond cardiac changes that would alter the heart’s response to autonomic activity are central nervous system changes that could potentially alter the quantitative balances of activity discussed earlier. Seizures spreading through an altered autonomic nervous system could have any number of different consequences on the heart. Finally, some of these same changes in cardiac and neural activity can be produced by chronic anticonvulsant use. Studies in animal models will be critical to address the range of impact of seizures on cardiac function in chronically epileptic animals.

39.8â•… Closing Comments on This and Other Animal Models There are a number of interesting animal models that have been used to study seizures and their cardiorespiratory consequences. Unfortunately, we are still in the very early stages of understanding brain–heart interactions. Our model has the speciἀc advantage of seizures without motor convulsions, but also without chemical paralysis of the animal. This offers access to the full spectrum of cardiorespiratory consequences in an acute preparation. The general advantages of animal models, efficiency plus measurement and manipulation options, make the acute preparation a flexible tool that should be considered for a variety of epilepsy studies. Each preparation has certain advantages and disadvantages. The choice of model should start with its advantages for studies of particular questions, but most importantly, the flexibility of animal models cannot be overstated. These can literally be “tailored” to address questions from the clinical world that the animal investigators may have never considered.

References Alheid, G., J. De Olmos, and C. Beltramino. 1995. Amygdala and extended amygdala. In The Rat Nervous System, 2nd ed., ed. G. Paxinos, 495–578. San Diego, CA: Academic Press. Allen, G. V., C. B. Saper, K. M. Hurley, and D. F. Cechetto. 1991. Organization of visceral and limbic connections in the insular cortex of the rat. J Comp Neurol 311: 1–16.

640 Sudden Death in Epilepsy: Forensic and Clinical Issues Ardell, J. L. 2004. Intrathoracic neuronal regulation of cardiac function. In Basic and Clinical Neurocardiology, ed. J. A. Armour and J. L. Ardell, 118–152. New York, NY: Oxford University Press. Barman, S. M., and G. L. Gebber. 1982. Hypothalamic neurons with activity patterns related to sympathetic nerve discharge. Am J Physiol 242: R34–R43. Barman, S. M., and G. L. Gebber. 1998. Classiἀcation of caudal ventrolateral pontine neurons with sympathetic nerve-related activity. J Neurophysiol 80: 2433–2445. Barman, S. M., G. L. Gebber, and H. Kitchens. 1999. Rostral dorsolateral pontine neurons with sympathetic nerve-related activity. Am J Physiol 276: H401–H412. Barman, S. M., G. L. Gebber, and H. S. Orer. 2000. Medullary lateral tegmental ἀeld: An important source of basal sympathetic nerve discharge in the cat. Am J Physiol Regul Integr Comp Physiol 278: R995–R1004. Barman, S. M., H. S. Orer, and G. L. Gebber. 1995. Axonal projections of caudal ventrolateral medullary and medullary raphe neurons with activity correlated to the 10-Hz rhythm in sympathetic nerve discharge. J Neurophysiol 74: 2295–2308. Baumgartner, C., S. Lurger, and F. Leutmezer. 2001. Autonomic symptoms during epileptic seizures. Epileptic Disord 3: 103–116. Brankack, J., M. Stewart, and S. E. Fox. 1993. Current source density analysis of the hippocampal theta rhythm: Associated sustained potentials and candidate synaptic generators. Brain Res 615: 310–327. Britton, J. W., G. R. Ghearing, E. E. Benarroch, and G. D. Cascino. 2006. The ictal bradycardia syndrome: Localization and lateralization. Epilepsia 47: 737–744. Cechetto, D. F., and C. B. Saper. 1988. Neurochemical organization of the hypothalamic projection to the spinal cord in the rat. J Comp Neurol 272: 579–604. Chamberlin, N. L., and C. B. Saper. 1992. Topographic organization of cardiovascular responses to electrical and glutamate microstimulation of the parabrachial nucleus in the rat. J Comp Neurol 326: 245–262. Chamberlin, N. L., and C. B. Saper. 1998. A brainstem network mediating apneic reflexes in the rat. J Neurosci 18: 6048–6056. Cheung, R. T., and V. Hachinski. 2000. The insula and cerebrogenic sudden death. Arch Neurol 57: 1685–1688. Cullian, W. E., J. P. Herman, and S. J. Watson. 1993. Ventral subicular interaction with the hypothalamic paraventricular nucleus: Evidence for a relay in the bed nucleus of the stria terminalis. J Comp Neurol 332: 1–20. Devinsky, O. 2004. Effects of seizures on autonomic and cardiovascular function. Epilepsy Curr 4: 43–46. Epstein, M. A., M. R. Sperling, and M. J. O’Connor. 1992. Cardiac rhythm during temporal lobe seizures. Neurology 42: 50–53. Feenstra, B. W., and J. Holsheimer. 1979. Dipole-like neuronal sources of theta rhythm in dorsal hippocampus, dentate gyrus and cingulate cortex of the urethane-anesthetized rat. Electroencephalogr Clin Neurophysiol 47: 532–538. Flecknell, P. A. 1996. Laboratory Animal Anaesthesia: A Practical Introduction for Research Workers and Technicians. London: Academic Press. Fukuda, Y., A. Sato, A. Suzuki, and A. Trzebski. 1989. Autonomic nerve and cardiovascular responses to changing blood oxygen and carbon dioxide levels in the rat. J Auton Nerv Syst 28: 61–74. Galin, D. 1974. Implications for psychiatry of left and right cerebral specialization. A neurophysiological context for unconscious processes. Arch Gen Psychiatry 31: 572–583. Gebber, G. L., S. M. Barman, and B. Kocsis. 1990. Coherence of medullary unit activity and sympathetic nerve discharge. Am J Physiol 259: R561–R571. Gebber, G. L., S. M. Barman, and S. F. Morrison. 1987. Electrophysiological evidence for the modular organization of the reticular formation: Sympathetic controlling circuits. Brain Res 410: 106–110. Gebber, G. L., and D. W. Snyder. 1970. Hypothalamic control of baroreceptor reflexes. Am J Physiol 218: 124–131.

The Urethane/Kainate Seizure Model

641

Gebber, G. L., S. Zhong, and S. M. Barman. 1995. Synchronization of cardiac-related discharges of sympathetic nerves with inputs from widely separated spinal segments. Am J Physiol 268: R1472–R1483. Gilbert, M., R. J. Racine, and G. K. Smith. 1985. Epileptiform burst responses in ventral vs dorsal hippocampal slices. Brain Res 361: 389–391. Goodman, J. H., R. W. Homan, and I. L. Crawford. 1990. Kindled seizures elevate blood pressure and induce cardiac arrhythmias. Epilepsia 31: 489–495. Goodman, J. H., R. W. Homan, and I. L. Crawford. 1999. Kindled seizures activate both branches of the autonomic nervous system. Epilepsy Res 34: 169–176. Goodman, J. H., M. Stewart, and F. W. Drislane. 2008. Autonomic disturbances. In Epilepsy: A Comprehensive Textbook, 2nd ed., ed. J. Engel and T. A. Pedley, 1999–2005. Philadelphia, PA: Lippincott Williams and Wilkins. Hachinski, V. C., S. M. Oppenheimer, J. X. Wilson, C. Guiraudon, and D. F. Chetto. 1992. Asymmetry of sympathetic consequences of experimental stroke. Arch Neurol 49: 697–702. Heltovics, G., B. Boda, and M. Szente. 1995. Anticonvulsive effect of urethane on aminopyridineinduced epileptiform activity. Neuroreport 6: 577–580. Hilz, M. J., M. Dutsch, K. Perrine, P. K. Nelson, U. Rauhut, and O. Devinsky. 2001. Hemispheric influence on autonomic modulation and baroreflex sensitivity. Ann Neurol 49: 575–584. Hotta, H., K. Koizumi, and M. Stewart. 2009a. Cardiac sympathetic nerve activity during kainic acid–induced limbic cortical seizures in rats. Epilepsia 50: 923–927. Hotta., H., J. Lazar, R. Orman, et al. 2009b. Vagus nerve stimulation-induced bradyarrhythmias in rats. Auton Neurosci 151: 98–105. Huikuri, H. V., A. Castellanos, and R. J. Myerburg. 2001. Sudden death due to cardiac arrhythmias. N Engl J Med 345: 1473–1482. Johnston, S. C., J. K. Horn, J. Valente, and R. P. Simon. 1995. The role of hypoventilation in a sheep model of epileptic sudden death. Ann Neurol 37: 531–537. Johnston, S. C., R. Siedenberg, J. K. Min, E. H. Jerome, and K. D. Laxer. 1997. Central apnea and acute cardiac ischemia in a sheep model of epileptic sudden death. Ann Neurol 42: 588–594. Kirchhof, P., G. Breithardt, and L. Eckardt. 2006. Primary prevention of sudden cardiac death. Heart 92: 1873–1878. Kishi, T., T. Tsumori, K. Ono, S. Yokota, H. Ishino, and Y. Yasui. 2000. Topographical organization of projections from the subiculum to the hypothalamus in the rat. J Comp Neurol 419: 205–222. Kiviniemi, A. M., M. P. Tulppo, D. Wichterle, et al. 2007. Novel spectral indexes of heart rate variability as predictors of sudden and non-sudden cardiac death after an acute myocardial infarction. Ann Med 39: 54–62. Koizumi, K., and M. Kollai. 1981. Control of reciprocal and non-reciprocal action of vagal and sympathetic efferents: Study of centrally induced reactions. J Auton Nerv Syst 3: 483–501. Koizumi, K., M. Kollai, and N. Turei. 1986. The physiology of cardiac innervation: Relationships between cardiac vagal and sympathetic nerve activities. In Neurovegetative Control Systems: Basic Functions, Integration, and Disorders: Proceedings of the XIXth International Congress of Neurovegetative Research, Berlin (West), FRG, March 1984, ed. H. P. Koepchen, C. M. Brooks, K. Koizumi, and conf. author, 161–171. Amsterdam, NY: International Congress of Neurovegetative Research. Koizumi, K., N. Terui, M. Kollai, and C. M. Brookis. 1982. Functional signiἀcance of coactivation of vagal and sympathetic cardiac nerves. Proc Natl Acad Sci U S A 79: 2116–2120. Kollai, M., and K. Koizumi. 1979. Reciprocal and non-reciprocal action of the vagal and sympathetic nerves innervating the heart. J Auton Nerv Syst 1: 33–52. Lane, R. D., and J. R. Jennings. 1998. Hemispheric asymmetry, autonomic asymmetry, and the problem of sudden cardiac death. In Brain Asymmetry, 2nd printing ed., ed. R. J. Davidson and K. Hugdahl, 271–304. Cambridge, MA: MIT Press. Langan, Y. 2000. Sudden unexpected death in epilepsy (SUDEP): Risk factors and case control studies. Seizure 9: 179–183.

642 Sudden Death in Epilepsy: Forensic and Clinical Issues Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23: 633–647. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9: 236–242. Lathers, C. M., P. L. Schraeder, and J. G. Boggs. 1998. Sudden unexplained death and autonomic dysfunction. In Epilepsy: A Comprehensive Textbook, vol. 2, chap. 183, ed. J. Engel and T. A. Pedley, 1943–1956. Philadelphia: Lippincott-Raven. Lathers, C. M., P. L. Schraeder, and N. Tumer. 1993. The effect of phenobarbital on autonomic function and epileptogenic activity induced by the hippocampal injection of penicillin in cats. J Clin Pharmacol 33: 837–844. Lathers, C. M., P. L. Schraeder, and F. W. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67: 247–259. Liedholm, L. J., and O. Gudjonsson. 1992. Cardiac arrest due to partial epileptic seizures. Neurology 42: 824–829. Locatelli, E. R., J. P. Varghese, A. Shuabib, and S. J. Potolicchio. 1999. Cardiac asystole and bradycardia as a manifestation of left temporal lobe complex partial seizure. Ann Intern Med 130: 581–583. Loewy, A. D. 1990. Central autonomic pathways. In Central Regulation of Autonomic Functions, ed. A. D. Loewy and K. M. Spyer, 88–103. New York, NY: Oxford University Press. Loewy, A. D., S. McKellar, and C. B. Saper. 1979. Direct projections from the A5 catecholamine cell group to the intermediolateral cell column. Brain Res 174: 309–314. Lombardi, F., T. H. Makikallio, R. J. Myerburg, and H. V. Huikui. 2001. Sudden cardiac death: Role of heart rate variability to identify patients at risk. Cardiovasc Res 50: 210–217. Mameli, O., M. A. Caria, F. Melis et al. 2001. Autonomic nervous system activity and life threatening arrhythmias in experimental epilepsy. Seizure 10: 269–278. Mameli, O., M. A. Caria, A. Pintus, G. Padua and S. Mameli. 2006. Sudden death in epilepsy: An experimental animal model. Seizure 15: 275–287. Mameli, O., F. Melis, D. Giraudi et al. 1993. The brainstem cardioarrhythmogenic triggers and their possible role in sudden epileptic death. Epilepsy Res 15: 171–178. Moga, M. M., H. Herbert, K. M. Hurley, Y. Yasumi, T. S. Gray, and C. B. Saper. 1990. Organization of cortical, basal forebrain, and hypothalamic afferents to the parabrachial nucleus in the rat. J Comp Neurol 295: 624–661. Moga, M. M., C. Saper, and T. S. Gray. 1989. Bed nucleus of the stria terminalis: Cytoarchitecture, immunohistochemistry, and projection to the parabrachial nucleus in the rat. J Comp Neurol 283: 315–332. Moroni, F., R. Corradetti, F. Casementi, G. Moneti, and G. Pepue. 1981. The release of endogenous GABA and glutamate from the cerebral cortex in the rat. Naunyn Schmiedeberg’s Arch Pharmacol 316: 235–239. Nei, M., R. T. Ho, and M. R. Sperling. 2000. EKG abnormalities during partial seizures in refractory epilepsy. Epilepsia 41: 542–548. Oppenheimer, S. 2001. Forebrain lateralization and the cardiovascular correlates of epilepsy. Brain 124: 2345–2346. Oppenheimer, S. M., A. Gelb, J. P. Girvin, and V. C. Hachinski. 1992. Cardiovascular effects of human insular cortex stimulation. Neurology 42: 1727–1732. Oppenheimer, S. M., J. X. Wilson, C. Guiraudon, and D. F. Chechetto. 1991. Insular cortex stimulation produces lethal cardiac arrhythmias: A mechanism of sudden death? Brain Res 550: 115–121. Pitakanen, A., M. Pikkarainen, N. Nurminen, and A. Ylinen. 2000. Reciprocal connections between the amygdala and the hippocampal formation, perirhinal cortex, and postrhinal cortex in rat. A review. Ann N Y Acad Sci 911: 369–391. Randall, D. C., D. R. Brown, A. S. McGuirt, G. W. Thompson, J. A. Armour, and J. L. Ardell. 2003. Interactions within the intrinsic cardiac nervous system contribute to chronotropic regulation. Am J Physiol Regul Integr Comp Physiol 285: R1066–R1075.

The Urethane/Kainate Seizure Model

643

Saito, T., K. Sakamoto, K. Koizumi, and M. Stewart. 2006. Repeatable focal seizure suppression: A rat preparation to study consequences of seizure activity based on urethane anesthesia and reversible carotid artery occlusion. J Neurosci Methods 155: 241–250. Sakamoto, K., T. Saito, R. Orman, et al. 2008. Autonomic consequences of kainic acid–induced limbic cortical seizures in rats: Peripheral autonomic nerve activity, acute cardiovascular changes, and death. Epilepsia 49: 982–996. Saper, C. B. 1979. Anatomical substrates for the hypothalamic control of the autonomic nerÂ� vous system. In Integrative Functions of the Autonomic Nervous System: An Analysis of the Interrelationships and Interactions of the Sympathetic and Parasympathetic Divisions of the Autonomic System in the Control of Body Function, ed. C. M. Brooks, K. Koizumi, and A. Sato, 333–341. Tokyo: University of Tokyo Press. Saper, C. B. 1982a. Convergence of autonomic and limbic connections in the insular cortex of the rat. J Comp Neurol 210 (2): 163–173. Saper, C. B. 1982b. Reciprocal parabrachial–cortical connections in the rat. Brain Res 242 (1): 33–40. Saper, C. B. 1985. Organization of cerebral cortical afferent systems in the rat: II. Hypothalamocortical projections. J Comp Neurol 237: 21–46. Saper, C. B. 2000. Hypothalamic connections with the cerebral cortex. In Cognition, Emotion and Autonomic Responses: The Integrative Role of the Prefrontal Cortex and Limbic Structures, ed. H.€B. M. Uylings, G. G. Van Eden, J. P. C. De Bruin, M. G. P. Feenstra, and C. M. A. Pennartz, 39–48. Amsterdam: Elsevier. Saper, C. B., A. D. Loewy, L. W. Swanson, and W. M. Cowan. 1976. Direct hypothalamo-autonomic connections. Brain Res 117: 305–312. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32: 1371–1382. Schraeder, P. L., and C. M. Lathers. 1989. Paroxysmal autonomic dysfunction, epileptogenic activity and sudden death. Epilepsy Res 3: 55–62. Smith-Demps, C., and A. Jagoda. 1998. A case of seizure-related bradycardia and asystole. Am J Emerg Med 16: 582–584. Spyer, K. M. 1979. Baroreceptor control of vagal preganglionic activity. In Integrative Functions of the Autonomic Nervous System: An Analysis of the Interrelationships and Interactions of the Sympathetic and Parasympathetic Divisions of the Autonomic System in the Control of Body Function, ed. C. M. Brooks, K. Koizumi, and A. Sato, 287–292. Tokyo: University of Tokyo Press. Stauffer, A. Z., J. Dodd-O, and C. M. Lathers. 1989. The relationship of the lock-step phenomenon and precipitous changes in mean arterial blood pressure. Electroencephalogr Clin Neurophysiol 72: 340–345. Stewart, M. 2008. Is an abrupt “cerebral electrical shutdown” during a seizure the mechanism of SUDEP? J Neurol Neurosurg Psychiatry. E letter. http://jnnp.bmj.com/cgi/eletters/78/12/1395#╉ 3299, accessed on 12 Feb 2008. Stewart, M., and S. E. Fox. 1990. Do septal neurons pace the hippocampal theta rhythm? Trends Neurosci 13: 163–168. Tinuper, P., F. Bisulli, A. Cerullo, et al. 2001. Ictal bradycardia in partial epileptic seizures: Autonomic investigation in three cases and literature review. Brain 124: 2361–2371. Tucker, D. C., and C. B. Saper. 1985. Speciἀcity of spinal projections from hypothalamic and brainstem areas which innervate sympathetic preganglionic neurons. Brain Res 360: 159–164. Venit, E. L., B. D. Shepard, and T. N. Seyfried. 2004. Oxygenation prevents sudden death in seizureprone mice. Epilepsia 45: 993–996. Westerhaus, M. J., and A. D. Loewy. 2001. Central representation of the sympathetic nervous system in the cerebral cortex. Brain Res 903: 117–127. Wilder-Smith, E. 1992. Complete atrio-ventricular conduction block during complex partial seizure. J Neurol Neurosurg Psychiatry 55: 734–736. Willott, J. F., and K. R. Henry. 1976. Roles of anoxia and noise-induced hearing loss in the postictal refractory period for audiogenic seizures in mice. J Comp Physiol Psychol 90: 373–381.

644 Sudden Death in Epilepsy: Forensic and Clinical Issues Witter, M. P., and D. G. Amaral. 2004. Hippocampal formation. In The Rat Nervous System, 3rd ed., ed. G. Paxinos, 635–704. San Diego, CA: Elsevier Academic Press. Witter, M. P., H. L. Groenewegen, F. H. Lopes Da Silva, and A. H. Lohman. 1989. Functional organization of the extrinsic and intrinsic circuitry of the parahippocampal region. Prog Neurobiol 33: 161–253. Witter, M. P., P. Room, H. L. Groenewegen, and A. H. Lohman. 1988. Reciprocal connections of the insular and piriform claustrum with limbic cortex: An anatomical study in the cat. Neuroscience 24: 519–539. Wouterlood, F. G., and R. Tuinhof. 1992. Subicular efferents to histaminergic neurons in the posterior hypothalamic region of the rat studied with PHA-L tracing combined with histidine decarboxylase immunocytochemistry. J Hirnforsch 33: 451–465. Yamashita, H., H. Kannah, K. Inenega, and K. Koizumi. 1983. Role of neurones in the supraoptic and paraventricular nuclei in cardiovascular control. Prog Brain Res 60: 459–468. Yasui, Y., C. D. Breder, C. B. Saper, and D. F. Cechetto. 1991. Autonomic responses and efferent pathways from the insular cortex in the rat. J Comp Neurol 303: 355–374. Zamrini, E. Y., K. J. Meador, D. W. Loring, et al. 1990. Unilateral cerebral inactivation produces differential left/right heart rate responses. Neurology 40: 1408–1411.

Acute Cardiovascular Response during Kindled Seizures Jeffrey H. Goodman Richard W. Homan IsAac L. Crawford

40

Contents 40.1 Introduction 40.2 Methods 40.3 Results 40.3.1 Effect on Blood Pressure and Heart Rate 40.3.2 Effect of Seizure on the Electrocardiogram 40.3.3 Seizure Dependency 40.3.4 Autonomic Changes in Non-Amygdaloid-Initiated Seizures 40.3.5 Cardiovascular Changes during Kindling Acquisition 40.4 Discussion 40.4.1 Role of the Amygdala 40.4.2 The Pressor Response 40.4.3 The Arrhythmogenic Response 40.4.4 The Cerebral Ischemic Response 40.4.5 Clinical Relevance 40.4.6 Kindling as a Model of Seizure-Induced Arrhythmias and Sudden Unexplained Death 40.5 Summary References

645 647 648 648 648 649 651 651 652 652 653 654 654 654 655 655 655

40.1â•…Introduction The high incidence of sudden unexplained death in the epileptic population not only poses a serious problem for the epileptic patient but is also a challenge to the clinician and an unsolved mystery for the basic scientist. A reasonable hypothesis is that if seizures disrupt autonomic regulation of the cardiovascular system, then the consequent generation of cardiac arrhythmias may underlie sudden unexplained death. Clearly, spontaneous seizures in man (Gilchrist 1985; Marshall et al. 1983; Pritchett et al. 1980) and experimentally induced seizures in animals (Doba et al. 1975; Lathers and Schraeder 1982) can be accompanied by cardiac arrhythmias. There is also a case report of sudden cardiac death immediately after a complex partial seizure (Dasheiff and Dickinson 1986). However, no direct evidence has been provided that deἀnitely links sudden unexplained death to seizure occurrence. A recent report by Keilson et al. (1987) revealed that the incidence of serious cardiac arrhythmias is no greater in epileptic patients than in the general population. Because there is 645

646 Sudden Death in Epilepsy: Forensic and Clinical Issues

no clinical marker that identiἀes those epileptic patients most at risk for sudden death, a greater research emphasis placed on basic experimental models of epilepsy may reveal new directions for clinical studies. Cardiovascular changes occur during several types of experimentally induced seizures: electroconvulsive treatment (Petito et al. 1977; Plum et al. 1968; Wasterlain 1974; Westergaard et al. 1978), pentylenetetrazol (Doba et al. 1975; Lathers and Schraeder 1982; Plum et al. 1968), bicuculline (Meldrum and Horton 1973), and penicillin (Lathers and Schraeder, this book, Chapter 10; Mameli et al. 1988). However, these models have inherent limitations because of the presence of anesthetics, paralytic agents with artiἀcial ventilation, or widespread exposure of the CNS to chemical convulsants or electrical current. In the kindling seizure model, repeated spaced presentations of an initially subconvulsive electrical stimulus eventually led to a permanent change in brain function (Goddard et al. 1969; Homan and Goodman 1988). This change is characterized by the occurrence of a generalized motor seizure each time the stimulus is presented. During the kindling process, there is a progressive increase in electrical afterdischarge activity (Figure 40.1), with spread of the afterdischarge to remote brain regions. Eventually, clinically apparent seizures develop, progressing through ἀve distinct behavioral stages (Racine 1972; Figure 40.2). Early stages in the kindling process (stages 1 and 2) are equivalent to partial seizures. Spread of the seizure ἀrst occurs at stage 3, whereas a stage 5 seizure is equivalent to a generalized tonic–clonic convulsion. This is a key feature of kindling because partial and generalized seizures can be examined in the same animal. A stage 5 generalized seizure usually occurs after 9–10 days of amygdaloid stimulation. To elicit spontaneous interictal spiking or spontaneous seizures requires 100–300 days of daily stimulation (Pinel and Rovner 1978). For this reason, these types of epileptiform activity are difficult to investigate in the kindling model. However, in contrast to chemical models, no drug is injected, and therefore one does not have to distinguish the direct from the nondirect effects of nonexperimental drugs. Thus far, few major morphological alterations have been found in the kindled focus, in contrast to injection of heavy metals Day 1

Day 2

1 mV 1s

Day 5

Gate artifact 400 μA, 60 Hz, 1 s BLA stimulation

Figure 40.1╇ Growth of seizure afterdischarge during the kindling process. The lower elec-

trograph for day 5 is a continuation of that shown before and after the fifth stimulus. (From Goodman, J. H., et al., Epilepsia, 31, 489–495, 1990. With permission.)

Acute Cardiovascular Response during Kindled Seizures

IV

647

V

III

II I

Figure 40.2╇ Behavioral stages of kindled seizures. The five stages are those identified by

Racine (1972).

or thermal injury to produce convulsions; both of the latter procedures result in neuronal loss and glial proliferation. The kindling model is well suited for an examination of the relationship between seizures and the cardiovascular system without the presence of drugs. Kindled seizures can be initiated in a discrete region of the brain of unanesthetized, unrestrained animals (Racine 1972, 1978). We have used this model to explore the acute cardiovascular changes that occur during amygdaloid-kindled seizures. We will review the results of our studies and also discuss the signiἀcance of these changes with respect to the potential value of kindling as a model of seizure-induced cardiac arrhythmias. Speculation on the relationship of our ἀndings to the clinical phenomenon of sudden unexplained death will also be presented.

40.2â•… Methods To examine the effect of kindled seizures on the cardiovascular system of the rat, male Sprague–Dawley rats (275–325 g) were anesthetized with ketamine (90 mg/kg) and xylazine (15 mg/kg) by intramuscular injection. Teflon-coated stainless steel electrodes (100€µm) were stereotaxically implanted in the basolateral amygdalae or in the olfactory bulbs according to previously reported methods used in our laboratory (Campbell and Crawford 1980; Crawford 1986; Walker et al. 1981). Each animal was allowed to recover a minimum of 1 week before the kindling process was initiated. During the kindling process, each animal was stimulated once daily for 1 s (400 µA, 60 Hz, 1-ms square wave, bipolar pulse) until three consecutive stage 5 generalized seizures occurred. A chronic femoral artery catheter ἀlled with heparinized saline was then surgically implanted to allow for the direct measurement of blood pressure in unanesthetized rats during the kindled seizure. After recovery from catheterization (a minimum of 24 h), ECG electrodes were placed across the chest; a kindled seizure was initiated; and changes in blood pressure, heart rate,

648 Sudden Death in Epilepsy: Forensic and Clinical Issues

and ECG were measured. Controls, with electrodes and catheter, were treated in a similar manner without undergoing the kindling process. In a separate group of rats, blood pressure and heart rate were measured during amygdaloid kindling acquisition. Amygdaloid electrodes and the arterial catheter were implanted as described above. These animals did not have ECG electrodes, so changes in the ECG were not measured. Because the arterial catheter remained patent for only 5 days, these animals were stimulated twice a day during the kindling process. The other stimulus parameters were the same as previously described. After each stimulus, changes in behavioral seizure score, afterdischarge duration, blood pressure, and heart rate were measured.

40.3â•…Results 40.3.1â•… Effect on Blood Pressure and Heart Rate Figure 40.3 shows the typical cardiovascular response during a stage 5 generalized amygdaloid-kindled seizure. The response was characterized by an abrupt increase in systolic and diastolic pressures, which lasted 20–30 s after initiation of the seizure. Superimposed on this change in pressure was a profound bradycardia occasionally accompanied by premature ventricular contractions. A summary of the effect of the seizure on mean arterial pressure (MAP) and heart rate is shown in Figure 40.4. MAP was signiἀcantly elevated for 20 s after initiation of the kindled seizure. During the ἀrst 10 s of the seizure, MAP approximated 150 mm Hg, whereas the mean heart rate decreased 50%. The increase in MAP was concomitant with increases in systolic, diastolic, and pulse pressures (Table 40.1). Changes in pressure and heart rate were greatest in the ἀrst 10–15 s of the seizure; however, both parameters returned to baseline before the end of the seizure afterdischarge. 40.3.2â•… Effect of Seizure on the Electrocardiogram The effect of the kindled seizure on the heart was determined by making simultaneous recordings of blood pressure and the ECG from animals during kindled seizures. During the seizure the bradycardia present in the pressure recording was consistently present in€the ECG of each animal. In most of the animals, additional irregularities were detected in the ECG during the bradycardia. Occasionally, premature ventricular contractions were also observed. These changes, as illustrated in Figure 40.5, included nonconducted P waves€and increases in the P–R interval, which are indicative of conduction heart block (Berne and Levy 1981). 200 mm Hg 150 100 50 0

Kindling stimulus

5s

Figure 40.3╇ Typical blood pressure response during an amygdaloid-kindled seizure. Arrow indicates when stimulus was delivered.

Acute Cardiovascular Response during Kindled Seizures

Mean arterial pressure (mean ± SEM) (mm Hg)

125

*

*

* *

100 75

0

5

N=9

Map

400

* *

*

300 200

Heart rate

10 15 20 Time from seizure onset (s)

30

Heart rate (mean ± SEM) (beats/min)

*

150

649

100

Figure 40.4╇ Graphic comparison of changes in mean arterial pressure (MAP) and heart rate during the first 30 s after initiation of stage 5 amygdaloid-kindled seizures (*p < 0.01). SEM, standard error of the mean. (From Goodman, J. H., et al., Epilepsia, 31, 489–495, 1990. With permission.)

40.3.3â•… Seizure Dependency The possibility existed that the cardiovascular response observed during the kindled seizure was simply the result of electrical stimulation of the amygdala, a known cardiovascular control center. To test this hypothesis, rats not previously kindled were stimulated with a single kindling train. In each instance the stimulus elicited a small seizure afterdischarge; however, no change in blood pressure or heart rate was detected. This suggests the cardiovascular changes that occurred in kindled animals in response to the same stimulus were seizure dependent. Electrically kindled seizures can be followed by postictal spiking. When this occurred in several animals, postictal spikes were accompanied by increases in blood pressure (Figure 40.6). These postictal spikes often develop into spontaneous seizures 2 or 3 min after the stimulus-induced seizure. These seizures, electrographically and behaviorally, mimicked stimulus-induced seizures. The cardiovascular responses during the secondary spontaneous seizures (Figure 40.7b) were qualitatively the same as those recorded during amygdaloid-kindled seizures (Figure 40.7a). Table 40.1â•… Effect of Amygdaloid-Kindled Seizures on Blood Pressure a Mean ± SEM (mm€Hg) Systolic pressure (mean ± SEM) Diastolic pressure (mean ± SEM) Pulse pressure (mean ± SEM)

Time from Seizure Initiation (s) n

Baseline

5

10

15

20

30

60

9

131.1 ± 1.4 85.0 ± 1.1 46.1 ± 1.0

192.8** ± 2.6 126.1** + 2.2 66.7** ± 2.3

197.8** ± 3.3 121.1** ± 1.5 76.7** ± 2.2

171.1** ± 1.5 103.9* ± 1.6 67.2** ± 1.6

163.3** ± 1.9 108.3** ± 2.2 55.0 ± 1.3

142.2 ± 1.4 90.6 ± 1.9 51.7 ± 1.0

131.4 ± 2.1 85.9 ± 1.5 45.6 ± 1.6

9 9

Source: Goodman, J. H., et al., Epilepsia, 31, 489–495, 1990. With permission. a One repeated-measure analysis of variance followed by Neuman–Keuls test, compared to baseline value. *p < 0.05; **p < 0.01.

650 Sudden Death in Epilepsy: Forensic and Clinical Issues

2 mV

ECG

Blood pressure 200 mm Hg 150 100 1s

50 0

EEG

1 mV

Stimulus

Figure 40.5╇ Electrograph of the effect of stage 5 amygdaloid-kindled seizure on the electro-

cardiogram (ECG) and blood pressure. Notice the isolated P wave (*) and increased P–R interval (small arrow) that correspond with irregular changes in the blood pressure recording and the afterdischarge in the EEG. The up and down arrows under the EEG indicate the beginning and end of the kindling stimulus. This is followed by a period of amplifier block. (From Goodman, J. H., et al., Epilepsia, 31, 489–495, 1990. With permission.) ECG

(a)

5 mV 180

BP

100 0

EEG 1 mV

ECG

(b)

5 mV 180

BP

100 0 EEG

1 mV 5s

Figure 40.6╇ Electrograph of blood pressure. EEG, electroencephalogram; ECG, electrocardiogram in a kindled rat before (a) and 3 min after a kindling stimulus (b). Note the transient rise in pressure after each interictal, epileptiform spike-wave complex.

Acute Cardiovascular Response during Kindled Seizures

(a)

Blood pressure 200 mm Hg 150 100 50 0

(b)

651

Amygdala Kindling stimulus

200 mm Hg 150 100 50 0

Spontaneous seizure

Amygdala

(c) 200 mm Hg 150 100 50 0

Olfactory bulb Kindling stimulus

5s

Figure 40.7╇ Effect of kindled seizures on blood pressure. (a) Change in blood pressure during

an amygdaloid-kindled stage 5 generalized seizure. (b) Change in blood pressure during a spontaneous generalized seizure that occurred 3 min after a stimulus-induced amygdaloid-kindled seizure. (c) Blood pressure response during a stage 5 generalized seizure initiated in the olfactory bulb.

40.3.4â•… Autonomic Changes in Non-Amygdaloid-Initiated Seizures To determine whether kindled seizure-induced autonomic changes were restricted to seizure initiation in the amygdala, we kindled animals in the olfactory bulb. Figure 40.7c shows the cardiovascular changes that occurred during an olfactory bulb–kindled seizure. These changes were qualitatively the same as those recorded during amygdaloid-kindled seizures. 40.3.5â•… Cardiovascular Changes during Kindling Acquisition The initial stimulus in the kindling process did not elicit a pressor response or a change in heart rate. However, as kindling was continued, several changes were observed. Figure 40.8 shows the progressive changes in the pressure and arrhythmic responses during the kindling process. The ἀrst cardiovascular change was a small rise in pressure with no change in heart rate. With subsequent stimulations the pressor response continued to increase. In contrast, the arrhythmic response developed at a different rate. The arrhythmia did not occur until after the sixth stimulation. Continued stimulation resulted in an increase in the duration and the severity of the arrhythmia while the magnitude of

652 Sudden Death in Epilepsy: Forensic and Clinical Issues Blood pressure 180 mm Hg Stimulus * 3

100 0 180

Stimulus * 5

100 0 180

Stimulus * 6

100 0 180

Stimulus * 7

20 s

100

5s

0 Kindling stimulus

Figure 40.8╇ Changes in blood pressure during kindling acquisition. Notice the delayed appearance of the arrhythmic response after the sixth and seventh stimulations. Arrows indicate when stimulus was delivered. (From Goodman, J. H., et al., Epilepsia, 31, 489–495, 1990. With permission.)

the pressor response remained the same. The cardiovascular response present after the seventh stimulation occurred during a stage 2 seizure, before seizure generalization had developed.

40.4â•…Discussion The results from the foregoing experiments identify an experimental seizure model that simultaneously evaluates altered cardiovascular function. The cardiovascular response during amygdaloid-kindled seizures was characterized by a large increase in blood pressure accompanied by a profound bradycardia. These changes were present during partial (Figure 40.8) and generalized seizures. It is important to recognize that these results were obtained from unanesthetized animals, thereby eliminating one of the confounding variables associated with other experimental seizure models. 40.4.1â•…Role of the Amygdala The amygdala is an important cardiovascular control center with numerous connections to brain stem (Hopkins and Holstege 1978; Schwaber et al. 1980; Takeuchi et al. 1982) and

Acute Cardiovascular Response during Kindled Seizures

653

limbic (De Olmos 1972) structures that regulate the autonomic nervous system. The connections to the hypothalamus through the stria terminals (De Olmos and Ingram 1972) and ventral amygdalofugal pathways (Millhouse 1969) are particularly relevant because these pathways mediate amygdala-induced changes in cardiovascular function (Faiers et al. 1975; Galosy et al. 1982; Hilton and Zbrozyna 1963). Stimulation of the amygdala in the rat (Faiers et al. 1975; Iwata et al. 1987; Le Doux et al. 1982; Mogenson and Calaresu 1973), cat (Heinemann et al. 1973; Stock et al. 1978), and monkey (Reis and Oliphant 1964) has been shown to affect blood pressure and heart rate. Speciἀc changes with basal amygdaloid stimulation in unanesthetized animals include a decrease in blood pressure with no change or an increase in heart rate (Heinemann et al. 1973; Stock et al. 1978) or a bradycardia followed by an increase in blood pressure (Reis and Oliphant 1964). With these observations in mind, a predictable response to an electrical stimulus in the amygdala of kindled rats would be an alteration of cardiovascular function. However, a single stimulus train in the amygdala of implanted control animals had no effect on blood pressure or heart rate. The difference between this result and those of past studies may be due to the use of different stimulus parameters, speciἀcally stimulus frequency and duration. The observation that a single stimulus train (1 s, 400 µA, 60 Hz) did not elicit cardiovascular changes in control animals suggests that the cardiovascular changes in kindled animals are seizure dependent. This is supported by the observations that postictal spikes are accompanied by an increase in pressure and that secondary spontaneous seizures, in the absence of a kindling stimulus, result in the same abnormal cardiovascular changes that occur during stimulus-induced seizures. Similar changes observed during olfactory bulb–kindled seizures suggest that this response is not limited to seizures initiated in the amygdala. 40.4.2â•… The Pressor Response The hypertension observed during generalized kindled seizures was similar to pressor changes reported to occur in several other experimental seizure models: ECT (Petito et al. 1977; Plum et al. 1968; Wasterlain 1974; Westergaard et al. 1978), PTZ (Doba et al. 1975; Lathers and Schraeder 1982; Plum et al. 1968), and bicuculline (Johansson and Nilsson 1977; Meldrum and Horton 1973). The pressor response during the kindled seizure was characterized by a rapid elevation in MAP to approximately 150 mm Hg. The normal response to an increase in MAP is an increase in vascular resistance, minimizing the increase in cerebral blood flow. However, increases in MAP of this magnitude were observed by Plum et al. (1968) during ECT- and PTZ-induced seizures to interrupt cerebral autoregulation, thereby causing cerebral blood flow to become pressure dependent. Such major MAP elevations can also compromise the blood–brain barrier (Johansson and Nilsson 1977; Suzuki et al. 1984; Westergaard et al. 1978). The sudden rise in blood pressure during the kindled seizure may be due to activation of sympathetic pathways. Plum et al. (1968) demonstrated that severing the spinal cord in dogs eliminated the increase in blood pressure and cerebral blood flow that occurred during seizures initiated with ECT or PTZ. A massive autonomic discharge has been reported to accompany generalized seizures (Meldrum et al. 1979), and in other seizure models, the seizure-induced rise in pressure can be prevented with ganglionic blockers (Meldrum 1976), alpha-adrenergic blockers, or chemical sympathectomy (Doba et al. 1975).

654 Sudden Death in Epilepsy: Forensic and Clinical Issues

40.4.3â•… The Arrhythmogenic Response It is not surprising that kindled seizures are accompanied by cardiac arrhythmias. There are several reports of arrhythmias associated with other types of experimentally induced seizures in animals (Delgado et al. 1960; Doba et al. 1975; Lathers and Schraeder 1982; Wasterlain 1974). These arrhythmias were probably due to abnormal autonomic activity induced by interictal or ictal seizure activity. The bradycardic response that occurred during the kindled seizure was probably mediated by the parasympathetic system. The irregularities observed in the ECG during the seizure indicate the bradycardia was due to a conduction heart block (Berne and Levy 1981). The parasympathetic system could be activated by two different mechanisms. The most obvious would be the activation of baroreceptor reflexes by the large pressor response (Frysinger et al. 1984). However, the parasympathetic system could also be activated centrally by the seizure. Lathers et al. (1987) demonstrated that cardiac arrhythmias associated with PTZ-induced interictal or ictal seizure activity result from autonomic dysfunction, that is, an imbalance in neural activity within and between sympathetic and parasympathetic nerves to the heart. Seizure-induced coactivation of the sympathetic and parasympathetic systems is another possibility (Doba et al. 1975). 40.4.4â•… The Cerebral Ischemic Response The hypertension and bradycardia that occurred during the kindled seizure resemble cardiovascular changes that occur in response to an increase in intracranial pressure, the Cushing response (Cushing 1902; Doba and Reis 1972; Hoff and Reis 1970), or to cerebral ischemia (Dampney et al. 1979; Guyton 1948). The rapid onset of the seizure-induced cardiovascular response makes it unlikely that these changes were the result of an increase in intracranial pressure or cerebral ischemia. However, the kindled changes may be mediated through the same central pathways. The cardiovascular changes during cerebral ischemia are mediated by coactivation of sympathetic and parasympathetic systems at the level of the lower brain stem (Kumada et al. 1979). The kindled seizure may activate the same pathways to induce changes in cardiovascular function. 40.4.5â•… Clinical Relevance The etiology of sudden death in epileptic patients remains undeἀned and may not be the same in all patients. Evidence suggests that it may be associated with events occurring at the time of a seizure and perhaps is precipitated by a seizure. The ἀnding that many patients who are found dead have subsequently been shown to have subtherapeutic anticonvulsant levels supports this hypothesis (Dasheiff and Dickinson 1986). Of particular importance is the possibility that the interaction between seizures and the cardiovascular system underlies the sudden unexplained death that has been estimated to occur in 5–17% of epileptic patients (Leestma et al. 1984). Our observations obtained during kindled seizures may be relevant to similar cardiovascular changes seen during seizures in epileptic patients. Hypertension has been observed during spontaneous seizures (Meyer et al. 1966; Van Buren 1958; Van Buren and Ajmone-Marsan 1960; White et al. 1961) and ECT (Elliot et al. 1982). There are also case reports of ictal tachycardia (Marshall et al. 1983; Metz et al. 1978; Pritchett et al. 1980) and bradycardia (Devinsky et al. 1986; Coulter 1984; Gilchrist 1985; Kiok et al. 1986).

Acute Cardiovascular Response during Kindled Seizures

655

40.4.6â•…K indling as a Model of Seizure-Induced Arrhythmias and Sudden Unexplained Death In the past, a good experimental model to study this relationship was lacking. Experimental seizure models such as ECT, PTZ, and bicuculline have inherent limitations because the presence of anesthetics, paralytic agents, or widespread exposure of the CNS to chemical convulsants. In addition, these are models of generalized seizures, which precludes the examination of the effect of partial seizures on the cardiovascular system. The kindling model does not have these limitations. However, it is difficult to study the effect of interictal seizure activity on cardiovascular function in kindled animals. To study these changes, the PTZ model in anesthetized cats has been useful (Carnel et al. 1985; Lathers and Schraeder 1982; Lathers et al. 1984, 1987). Kindled seizures are initiated in a discrete region of the brain of unanesthetized, unrestrained animals. Kindling also progresses through a series of distinct stages that allows for the examination of the effect of partial and generalized seizures on the cardiovascular system. In light of our results, the kindling seizure model appears to be well suited for studies that examine the relationship between seizures and the cardiovascular system. Several questions remain to be answered. Do seizures have a progressive effect on the cardiovascular system? Is the severity of the seizure related to the degree of cardiac involvement? Is the quality of the arrhythmic response dependent on the site of origin of the seizure? Can repeated kindled seizures increase the susceptibility of kindled animals to cardiovascular changes associated with minimal discharges, that is, interictal spikes? All of these questions have yet to be addressed in the kindling seizure model.

40.5â•… Summary The effect of seizures on cardiovascular function was evaluated during kindled seizures in conscious rats. This relationship was also examined during kindling acquisition. The typical cardiovascular response during a generalized kindled seizure consisted of a large increase in blood pressure accompanied by a profound bradycardia during the ἀrst 20–30€ s of the seizure. Similar changes in heart rate and blood pressure were observed during amygdaloid- and olfactory bulb–kindled seizures as well as secondary spontaneous siezures, suggesting these changes were seizure dependent but not limited to seizures initiated in the amygdala. These cardiovascular changes were also present during partial seizures early in the kindling process. These results suggest that kindling is a useful seizure model in which to study the underlying mechanism of seizure-induced arrhythmias and possibly the clinical phenomenon of sudden unexplained death.

References Berne, R. M., and M. N. Levy. 1981. Cardiovascular Physiology, 46. St. Louis, MO: C. V. Mosby. Campbell, G. A., and I. L. Crawford. 1980. A gated electronic switch for stimulation and recording with a single electrode. Brain Res Bull 5: 485–486. Carnel, S. B., P. L. Schraeder, and C. M. Lathers. 1985. The effect of phenobarbital pretreatment on cardiac neural discharge and pentylenetetrazol-induced epileptogenic activity. Pharmacology 20: 225–240.

656 Sudden Death in Epilepsy: Forensic and Clinical Issues Coulter, D. 1984. Partial seizures with apnea and bradycardia. Arch Neurol 41: 173–174. Crawford, I. L. 1986. Relationship of glutamic acid and zinc to kindling of the rat amygdala: Afferent transmitter systems and excitability in a model of epilepsy. In Excitatory Amino Acids and Epilepsy, ed. R. Schwarcz and Y. Ben-Ari, 611–623. New York, NY: Plenum. Cushing, H. 1902. Some experimental and clinical observations concerning states of increased intracranial tension. Am J Med Sci 124: 375–400. Dampney, R. A. L., M. Kumada, and D. J. Reis. 1979. Central neural mechanisms of the cerebral ischemic response. Circ Res 45: 48–62. Dasheiff, R. M., and L. J. Dickinson. 1986. Sudden unexpected death of epileptic patient due to cardiac arrhythmia after seizure. Arch Neurol 43: 194–196. Delgado, J. M., L. Mihailovic, and M. Sevillano. 1960. Cardiovascular phenomena during seizure activity. J Nerv Ment Dis 130: 477–487. De Olmos, J. S. 1972. The amygdaloid projection ἀeld in the rat as studied with the cupric-silver method. In The Neurobiology of the Amygdala, ed. B. Eleftheriou, 145–204. New York, NY: Plenum. De Olmos, J. S., and W. R. Ingram. 1972. The projection ἀeld of the stria terminalis in the rat brain: An experimental study. J Comp Neurol 146: 303–334. Devinsky, O., B. H. Price, and S. I. Cohen. 1986. Cardiac manifestations of complex partial seizures. Am J Med 80: 195–202. Doba, N., and D. J. Reis. 1972. Localization within the lower brainstem of a receptive area mediating the pressure response to increased intracranial pressure (the Cushing response). Brain Res 47: 487–491. Doba, N., H. R. Beresford, and D. J. Reis. 1975. Changes in regional blood flow and cardiodynamics associated with electrically and chemically induced epilepsy in the cat. Brain Res 90: 115–132. Elliot, D., D. Linz, and J. Kane. 1982. Electroconvulsive therapy. Arch Intern Med 142: 919–981. Faiers, A. A., F. R. Calaresu, and G. J. Mogenson. 1975. Pathway mediating hypotension elicited by stimulation of the amygdala in the rat. Am J Physiol 228: 1358–1366. Frysinger, R. C., J. D. Marks, R. B. Trelease, V. L. Schechtman, and R. M. Harper. 1984. Sleep states attenuate the pressor response to central amygdala stimulation. Exp Neurol 83: 604–611. Galosy, R. A., I. L. Crawford, and M. E. Thompson. 1982. Behavioral stress and cardiovascular regulation: Neural mechanisms. In Circulation, Neurobiology, and Behavior, ed. O. A. Smith, R. A. Galosy, and S. M. Weiss, 109–120. New York, NY: Elsevier. Gilchrist, J. M. 1985. Arrhythmogenic seizures: Diagnosis by simultaneous EEG/ECG recording. Neurology 35: 1503–1506. Goddard, G. V., D. C. McIntyre, and C. K. Leech. 1969. A permanent change in brain function resulting from daily electrical stimulation. Exp Neurol 25: 295–300. Goodman, J. H., R. W. Homan, and I. L. Crawford. 1990. Kindled seizures elevate blood pressure and induce cardiac arrhythmias. Epilepsia 31: 489–495. Guyton, A. C. 1948. Acute hypertension in dogs with cerebral ischemia. Am J Physiol 154: 45–54. Heinemann, H., G. Stock, and H. Schaefer. 1973. Temporal correlation of responses in blood pressure and motor reaction under electrical stimulation of limbic structures in unanesthetized, unrestrained cats. Pflügers Arch 343: 27–40. Hilton, S. M., and A. W. Zbrozyna. 1963. Amygdaloid region for defense reactions and its efferent pathway to the brainstem. J Physiol 165: 160–173. Hoff, J. T., and D. J. Reis. 1970. Localization of the regions mediating the Cushing response in CNS of the cat. Arch Neurol 23: 228–240. Homan, R. W., and J. H. Goodman. 1988. Endurance of the kindling effect is independent of the degree of generalization. Brain Res 447: 404–406. Hopkins, D. A., and G. Holstege. 1978. Amygdaloid projections to the mesencephalon, pons and medulla oblongata in the cat. Exp Brain Res 32: 529–547.

Acute Cardiovascular Response during Kindled Seizures

657

Iwata, J., K. Chida, and J. E. LeDoux. 1987. Cardiovascular responses elicited by stimulation of neurons in the central amygdaloid nucleus in awake but not anesthetized rats resemble conditioned emotional responses. Brain Res 418: 183–188. Johansson, B., and B. Nilsson. 1977. The pathophysiology of the blood brain barrier dysfunction induced by severe hypercapnia and by epileptic brain activity. Acta Neuropathol 38: 153–158. Keilson, M. J., W. A. Hauser, J. P. Magrill, and M. Goldman. 1987. ECG abnormalities in patients with epilepsy. Neurology 37: 1624–1626. Kiok, M. C., C. F. Terrence, G. H. Fromm, and S. Lavier. 1986. Sinus arrest in epilepsy. Neurology 36: 115–116. Kumada, M., R. A. L. Dampney, and D. J. Reis. 1979. Profound hypotension and abolition of the vasomotor component of the cerebral ischemic response produced by restricted lesions of the medulla oblongata in rabbit. Circ Res 45: 63–10. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23: 633–641. Lathers, C. M., P. L. Schraeder, and S. B. Carnel. 1984. Neural mechanisms in cardiac arrhythmias associated with epileptogenic activity: The effect of phenobarbital. Life Sci 34: 1919–1936. Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of autonomic neural discharge with epileptogenic activity: The lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67: 247–259. Le Doux, J. E., A. Delbo, L. W. Tucker, G. Harshἀeld, W. T. Talman, and D. J. Reis. 1982. Hierarchic organization of blood pressure responses during the expression of natural behaviors in rat: Mediation by sympathetic nerves. Exp Neurol 78: 121–133. Leestma, J. E., M. B. Kalelkar, S. S. Teas, G. W. Jay, and J. R. Hughes. 1984. Sudden unexpected death associated with seizures: Analysis of 66 cases. Epilepsia 25: 84–88. Mameli, P., O. Mameli, E. Tolu, G. Padua, D. Giraudi, M. A. Caria, and F. Melis. 1988. Neurogenic myocardial arrhythmias in experimental focal epilepsy. Epilepsia 29: 74–82. Marshall, D. W., B. F. Westmoreland, and F. W. Sharbrough. 1983. Ictal tachycardia during temporal lobe seizures. Mayo Clin Proc 58: 443–446. Meldrum, B. S. 1976. Neuropathology and pathophysiology. In A Textbook of Epilepsy, ed. J. Laidlaw and A. Richens, 314–354. Edinburgh: Churchill-Livingstone. Meldrum, B. S., and R. W. Horton. 1973. Physiology of status epilepticus in primates. Arch Neurol 28: 1–9. Meldrum, B. S., R. W. Horton, S. R. Bloom, J. Butler, and J. Keenan. 1979. Endocrine factors and glucose metabolism during prolonged seizures in baboons. Epilepsia 20: 527–534. Metz, S. A., J. B. Halter, D. Porte, and R. P. Robertson. 1978. Autonomic epilepsy. Ann Intern Med 88: 189–193. Meyer, J. S., F. Gotch, and E. Favale. 1966. Cerebral metabolism during epileptic seizures in man. Electroencephalogr Clin Neurophysiol 21: 10–22. Millhouse, O. E. 1969. A Golgi study of the descending medial forebrain bundle. Brain Res 15: 341–363. Mogenson, G. J., and F. R. Calaresu. 1973. Cardiovascular responses to electrical stimulation of the amygdala in the rat. Exp Neurol 39: 166–180. Petito, C. K., J. A. Schaeffer, and F. Plum. 1977. Ultrastructural characteristics of the brain and blood brain barrier in experimental seizures. Brain Res 127: 251–267. Pinel, J., and L. Rovner. 1978. Experimental epileptogenesis: Kindling-induced epilepsy in rats. Exp Neurol 58: 190–202. Plum, F., J. B. Posner, and B. Troy. 1968. Cerebral metabolic and circulatory responses to induced convulsions in animals. Arch Neurol 18: 1–13. Pritchett, E. L. C., J. O. McNamara, and J. J. Gallagher. 1980. Arrhythmogenic epilepsy: An hypothesis. Am Heart J 100: 683–688.

658 Sudden Death in Epilepsy: Forensic and Clinical Issues Racine, R. J. 1972. Modiἀcation of seizure activity by electrical stimulation. II. Motor seizure. Electroencephalogr Clin Neurophysiol 32: 281–294. Racine, R. J. 1978. Kindling, the ἀrst decade. Neurosurgery 3: 234–252. Reis, D. J., and M. C. Oliphant. 1964. Bradycardia and tachycardia following electrical stimulation of the amygdaloid region in monkey. J Neurophysiol 27: 893–912. Schwaber, J. S., B. S. Kapp, and G. Higgins. 1980. The origin and extent of direct amygdala projections to the region of the dorsal motor nucleus of the vagus and the nucleus of the solitary tract. Neurosci Lett 20: 15–20. Stock, G., K. H. Schlor, H. Heidt, and J. Buss. 1978. Psychomotor behavior and cardiovascular patterns during stimulation of the amygdala. Pflügers Arch 376: 177–184. Suzuki, R., C. Nitsch, K. Fujiwara, and I. Klatzo. 1984. Regional changes in cerebral blood flow and blood brain barrier permeability during epileptic seizures and in acute hypertension in rabbits. J Cereb Blood Flow Metab 4: 96–102. Takeuchi, Y., J. H. Mclean, and D. A. Hopkins. 1982. Reciprocal connections between the amygdala and parabrachial nuclei: Ultrastructural demonstration by degeneration and axonal transport of horseradish peroxidase in the cat. Brain Res 239: 583–588. Van Buren, J. M. 1958. Some autonomic concomitants of ictal automatism. Brain 81: 505–528. Van Buren, J. M., and C. Hopkins. 1960. Correlations of autonomic and EEG components in temporal lobe epilepsy. Arch Neurol 3: 683–703. Walker, J. E., J. A. Mikeska, and I. L. Crawford. 1981. Cyclic nucleotides in the amygdala of the kindled rat. Brain Res Bull 6: 1–3. Wasterlain, C. G. 1974. Mortality and morbidity from serial seizures. Epilepsia 15: 155–176. Westergaard, E., M. N. Hertz, and T. G. Bolwig. 1978. Increased permeability to horseradish peroxidase across cerebral vessels evoked by electrically-induced seizures in the rat. Acta Neuropathol 41: 73–80. White, P. T., P. Grant, J. Mosier, and A. Craig. 1961. Changes in cerebral dynamics associated with seizures. Neurology 11: 354–361.

DBA Mice as Models of Sudden Unexpected Death in Epilepsy Carl L. Faingold Srinivasan Tupal Yashanad Mhaskar Victor V. Uteshev

41

Contents 41.1 41.2 41.3 41.4

Introduction Mechanisms of SUDEP—Cardiac Mechanisms of SUDEP—Respiratory SUDEP Animal Model in DBA/2 Mice 41.4.1 Duration of Susceptibility to Respiratory Arrest in DBA/2 Mice 41.4.2 Genetic Basis of Seizure Susceptibility in DBA/2 Mice 41.5 Regulation of Respiration by Serotonin 41.6 Expression of 5-HT Receptor Proteins and Transporter in C57BL/6J and DBA/2 Mice 41.7 Sudden Infant Death Syndrome and SUDEP 41.8 Regulation of Seizure Susceptibility by Serotonin 41.9 Enhanced Serotonin Action Reduces SUDEP in DBA/2 Mice 41.10 Reduced Serotonin Action Increases SUDEP in DBA/2 Mice: Serotonin Antagonists 41.11 DBA/1 Mice as a Chronic Model of SUDEP 41.12 Discussion Acknowledgments References

659 660 661 661 663 663 663 665 666 667 667 669 669 671 672 672

41.1â•…Introduction Published clinical observations indicate that a major proposed cause for sudden unexpected death in epilepsy (SUDEP) patients is respiratory arrest that is commonly associated with a generalized convulsive seizure. DBA mice were known to exhibit generalized convulsive seizures in response to high intensity acoustic stimuli that lead to respiratory arrest, whereas recent data indicate that the electrocardiogram of these mice remains active for several minutes beyond respiratory arrest onset. These data have led to the proposal that DBA/2 mice are a useful model of respiratory arrest–mediated SUDEP. We have documented that SUDEP in DBA/2 mice can be prevented in >90% of these mice by rapid intervention with mechanical respiratory support, allowing multiple testing of individual mice for investigating possible preventative measures.

659

660 Sudden Death in Epilepsy: Forensic and Clinical Issues

Serotonin (5-HT) is a critically important neurotransmitter in brainstem respiratory centers that is known to play an important role in controlling normal respiration and in enhancing respiratory rate in response to elevated carbon dioxide levels. Our studies indicate drugs that enhance the activation of 5-HT receptors directly or indirectly, including a selective serotonin reuptake inhibitor and selective 5-HT receptor agonists, reduce or block respiratory arrest in DBA/2 mice at doses that do not block seizures. By contrast, a 5-HT receptor antagonist will induce SUDEP in the small percentage of DBA/2 mice that exhibit tonic seizures without respiratory arrest. Our recent data indicate that the expression of speciἀc 5-HT2C (which is inhibitory), 5-HT3, and 5-HT4 receptor proteins in the rostral ventral medulla respiratory region of DBA/2 mice are signiἀcantly diminished, whereas expression of the excitatory 5-HT2B receptors is signiἀcantly enhanced compared to seizure-resistant control (C57BL/6J) mice. By contrast, no statistically signiἀcant difference in expression of serotonin transporter protein between control and DBA/2 mice is seen. DBA/2 mice exhibit a relatively short duration of consistent susceptibility to respiratory arrest (~7 days), but we have recently observed that a closely related strain, DBA/1 mice, exhibits a >4-week susceptibility to seizure-induced respiratory arrest. The susceptibility of DBA/1 mice to respiratory arrest can be blocked, both acutely and after semichronic (5-day) treatment, with a selective serotonin reuptake inhibitor (fluoxetine, Prozac), with the mice subsequently returning to respiratory arrest susceptibility. These ἀndings indicate the potential usefulness of DBA/1 and DBA/2 mice as models of human SUDEP and drugs, such as selective serotonin reuptake inhibitors and selective 5-HT receptor agonists, that have proven effective in preventing respiratory arrest in these mice have the potential to be effective preventative agents for human SUDEP.

41.2â•… Mechanisms of SUDEP—Cardiac SUDEP is a devastating consequence of epilepsy and has long been recognized as an important concern in both adult and pediatric patients with epilepsy, particularly those who experience generalized convulsive seizures (Langan et al. 2000; Donner et al. 2001; Nilsson et al. 2001; McGregor and Wheless 2006; Tomson et al. 2008). A case-control study concluded that SUDEP is a seizure-related phenomenon and that control of generalized convulsive seizure and nocturnal supervision may be protective against SUDEP (Langan et al. 2005). The overwhelming seizure pattern (>80%) exhibited by SUDEP patients before death was syndromes that involved a generalized convulsive seizure, particularly tonic–clonic seizures (Opeskin and Berkovic 2003; Tomson et al. 2008). Diminished respiratory function and irregular cardiac rhythm in patients and animals are commonly associated with generalized convulsive seizure (Schraeder and Lathers 1989; Nashef et al. 1996; Langan et al. 2000; So et al. 2000; Ryvlin et al. 2006; Bateman et al. 2008; Lathers et al. 2008). Evidence has been presented that implicates cardiac events, including arrhythmias and asystole, as a major cause for SUDEP with varying degrees of clinical validation (Nashef et al. 1995; Langan et al. 2000; Darbin et al. 2002; Davis and McGwin 2004; So 2006, 2008; St-John et al. 2006; Tomson et al. 2008).

DBA Mice as Models of Sudden Unexpected Death in Epilepsy

661

41.3╇ Mechanisms of SUDEP—Respiratory The second major proposed cause for SUDEP is respiratory malfunction. Thus, in a number of witnessed cases of human SUDEP, respiratory difficulties were observed before death in ~75% of cases (Langan et al. 2000; Lear-Kaul et al. 2005). A recent report on human SUDEP indicated that death occurring in the postictal period primarily resulted from respiratory arrest (Lear-Kaul et al. 2005). SUDEP and near-SUDEP cases have been observed in epilepsy monitoring units [in the ongoing mortality in Epilepsy Monitoring Unit Study (MORTEMUS)], and a signiἀcant percentage of these cases exhibited respiratory difficulties (Ryvlin et al. 2006; Bateman et al. 2010). A recent study in an epilepsy monitoring unit indicated that signiἀcantly compromised respiratory function was found in most observed seizures, and patients exhibiting generalized convulsive seizure showed ~50% decline in respiration (Bateman et al. 2008; Seyal et al. 2010). Thus, the preponderance of clinical observations suggests an important role of respiratory difficulties during and after generalized convulsive seizure and respiratory arrest in observed cases of SUDEP and near-SUDEP cases. A recent report observed that two human SUDEP cases in an epilepsy monitoring unit that began with respiratory difficulties subsequently led to cardiac arrhythmias prior to death (Bateman et al. 2010), which is similar to what is seen in the DBA mouse models of SUDEP, as discussed below (Faingold et al. 2010a).

41.4â•… SUDEP Animal Model in DBA/2 Mice Animal studies of SUDEP had been hampered by the paucity of suitable models, but in recent years, the dilute brown non-Agouti (DBA/2) mouse has been proposed as a model of SUDEP (Venit et al. 2004; Tupal and Faingold 2006). DBA mice are genetically susceptible to generalized convulsive seizures evoked by acoustic stimulation (audiogenic seizures) that results in sudden death. Signiἀcant evidence indicates that the sudden death in the DBA/2 mice is caused by generalized convulsive seizure–associated respiratory arrest and that several other inbred rodent strains exhibit generalized convulsive seizure when exposed to intense acoustic stimulation. Susceptibility to audiogenic seizures can be induced in previously normal rodents with several types of treatments (Faingold 2002). In our initial study of 168 DBA/2 mice, behavioral responses to intense acoustic stimulation (122 dB SPL re: 0.0002 dyn/cm2) using broadband acoustic stimuli (electrical bell) were evaluated. All of the DBA/2 mice exhibited generalized convulsive seizures, which were followed by postictal depression of behavior in 85%. Speciἀc behaviors during audiogenic seizures included wild running (100% incidence), generalized clonus (100%), and tonic seizures culminating in tonic hind limb extension (85%). During the postictal period, most DBA/2 mice (75%) displayed respiratory arrest (Figure 41.1). Ten percent of DBA/2 mice exhibited tonic extension without respiratory arrest. The remaining (15%) did not exhibit tonic extension or respiratory arrest. Respiratory arrest in DBA/2 mice occurred almost exclusively after convulsive behavior (Tupal and Faingold 2006). After audiogenic seizures, a large proportion of DBA/2 mice exhibited sudden death unless respiration was supported (Willott and Henry 1976; Seyfried and Glaser 1985; Tupal and Faingold 2006). DBA/2 mice that exhibited respiratory arrest could be consistently resuscitated if resuscitation was initiated within 2–5 s after the ἀnal deep respiratory gasp (Tupal and Faingold

662 Sudden Death in Epilepsy: Forensic and Clinical Issues DBA/1 100

% Seizure + RA

80 60

DBA/2

40 20 0

21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Age of mice (days)

Figure 41.1╇ Susceptibility of DBA/2 and DBA/1 mice to audiogenic seizures with respiratory

arrest changes differentially with age and audiogenic seizure experience. The two strains of DBA mice (DBA/2 and DBA/1) display differing patterns of susceptibility to respiratory arrest (RA) after audiogenic seizures. The susceptibility to respiratory arrest after audiogenic seizures was seen in 75% of DBA/2 mice on postnatal day 21, which declined progressively to 0% on day 35 due to progressive hearing loss. (Note: no change in respiratory arrest susceptibility in DBA/2 mice was noted with retest and not all tests are shown.) Each individual DBA/2 mouse that exhibited audiogenic seizures with respiratory arrest showed consistent susceptibility for ~7 days. The incidence of audiogenic seizures with respiratory arrest in DBA/1 mice increased from 33% on postnatal day 21 to 100% by the third test on day 23. If testing began at later ages, it still required three tests for DBA/1 mice to exhibit 100% susceptibility. Unlike the DBA/2 mice, the DBA/1 mice continued to display respiratory arrest susceptibility for >35 (up to 50) days.

2006). This involves placing the mouse in a supine position and placing a polyethylene tube connected to outflow of a rodent respirator over the nostrils of the mouse for 90% of DBA/2 mice that are given prompt respiratory support by an experienced exÂ�perimenter (Tupal and Faingold 2006). However, each experimenter requires practice to reach this level of success. Thus, for one of the authors (S. T.), in the initial month of experimentation, only 75% of the DBA/2 mice were successfully resuscitated, but by the second month, the rate exceeded 90%, and by the seventh month, it approached 99%. Increasing oxygen supply in the vicinity of DBA/2 mice also increases survival rates (Venit et al. 2004). Recent data indicate that during respiratory arrest in DBA/2 mice when the animal is not breathing, the electrocardiogram remains active for 4–6 min with decreasing and increasing heart rates and abnormal rhythms that alternate before cessation of the electrocardiogram (Figure 41.2), suggesting that some cardiac function remains after respiratory arrest.

DBA Mice as Models of Sudden Unexpected Death in Epilepsy (a)

(b)

(c)

663

(d)

Figure 41.2╇ A typical example of electrocardiogram findings in DBA/2 mice under nonseizure (anesthetized) conditions and electrocardiogram changes associated with seizure-induced respiratory arrest. (a) Electrocardiogram of a DBA/2 mouse under ketamine/xylazine anesthesia (85/5 mg/kg). (b–d) Taken from a different DBA/2 mouse; electrocardiogram changes after an audiogenic generalized convulsive seizure. Reduced electrocardiogram rate immediately after respiratory arrest is shown in (b), whereas (c) shows the electrocardiogram rate and pattern ~1 min after respiratory arrest onset from the same mouse, illustrating evidence of abnormal cardiac rhythm. In d, the electrocardiogram can be seen to be nearly absent in the same mouse ~4 min after respiratory arrest onset. The electrocardiogram was completely absent subsequent to this trace (trace length = 5 s, amplitude in (a) = 0.5 mv, (b–d) = 0.4 mv, two forelimb needle electrodes plus ground configuration).

41.4.1â•… Duration of Susceptibility to Respiratory Arrest in DBA/2 Mice Seizure susceptibility is known to decrease with age in DBA/2 mice (Collins 1972; Turner and Willott 1998). When the seizure-inducing stimulus was presented daily at 21–25 days of age, resuscitated DBA/2 mice subsequently exhibit seizures ending with respiratory arrest each time (Tupal and Faingold 2006). However, after 25 days of age, the occurrence of respiratory arrest declined; and when the DBA/2 mice reached 29 days of age, a loss of seizure susceptibility occurred in ~30% of mice, and ~40% of the mice that remained susceptible to seizures no longer exhibited tonic extension with a greatly reduced incidence of respiratory arrest (to ~25%). The incidence of audiogenic seizures with respiratory arrest reached zero by 35 days of age (Figure 41.1). 41.4.2â•… Genetic Basis of Seizure Susceptibility in DBA/2 Mice Seizure susceptibility of DBA/2 mice is associated with the Asp gene family (Collins and Fuller 1968) and is influenced by three gene loci. One locus, Asp-1, is located on chromosome 12 between Ah and D12Nyu1; another locus, Asp-2, is on chromosome 4, located within an eight-centimorgan segment distal to b; and Asp-3 is linked to c on chromosome 7 (Neumann and Collins 1991, 1992). These three loci account for most of the heritable variation in susceptibility to audiogenic seizures in crosses of DBA/2 and C57BL/6J mice, and susceptibility to audiogenic seizures is influenced by genomic imprinting (Neumann and Collins 1991).

41.5â•…Regulation of Respiration by Serotonin Considerable basic research has established the rostral ventral medulla, including the nucleus of the solitary tract and the pre-Bötzinger complex of the brainstem, as principal

664 Sudden Death in Epilepsy: Forensic and Clinical Issues

centers of respiration control (Bonham et al. 2006; Kubin et al. 2006; Potts 2006). The pre-Bötzinger complex is a group of neurons that has been localized in the ventrolateral medulla that exerts critical control of respiratory rhythm, and the intrinsic pacemaker currents in these neurons are modulated by several neurotransmitters (Ramirez and Viemari 2005; Brisson-Thoby and Greer 2008), including 5-HT. The respiratory sensory afferents from the respiratory tract (lungs and airways) enter the brainstem via the vagus nerve and terminate in the nucleus of the solitary tract (Boscan et al. 2002; Potts 2002). The nucleus of the solitary tract receives and integrates afferents from the respiratory system and projects to the medullary respiratory nuclei (Richerson and Getting 1992; Li et al. 1999). Several thalamic, hypothalamic, cortical, medullary, pontine, and midbrain nuclei send modulatory projections to the brainstem respiratory centers (Terreberry and Neafsey 1987; Agarwal and Calaresu 1990; Otake et al. 1994). Sensory signals are also modulated within the brainstem, as the excitatory and inhibitory synaptic inputs to second-order neurons as well as neuronal properties exhibit plastic changes as a result of stimulation, neuronal activity, or action of neuromodulators, particularly 5-HT (Chen and Bonham 1998; Kline et al. 2002; Bonham et al. 2006). As noted above, 5-HT is one of the major brain neurotransmitters and is involved in generating and transmitting respiratory rhythm (Bonham 1995; Pena and Ramirez 2002; Wong -Riley and Liu 2005). Endogenous 5-HT is known to play an important role in modulating the activity of neuronal networks in the brainstem and is required for normal respiration (Al-Zubaidy et al. 1996). Speciἀc subtypes of 5-HT receptors are thought to be selectively relevant to control of respiration and to epilepsy (Carley and Radulovacki 1999; Richter et al. 2003; Toczek et al. 2003). Pharmacological, electrophysiological, and immunohistochemical studies indicate that expression of at least 7 of the 14 known 5-HT receptor subtypes occurs in the brainstem (Jordan 2005; Raul 2003). Of those subtypes, the subfamily of 5-HT2 receptors (i.e., 5-HT2A, 5-HT2B, and 5-HT2C) is especially relevant to epilepsy and seizures (see below). Although all three 5-HT2 receptors act via activation of Gq/11 receptors coupled to phospholipase C, their net effects are not identical and stimulation of 5-HT2 receptors could produce either excitation or inhibition. In the brainstem and speciἀcally in the nucleus of the solitary tract, activation of 5-HT2C receptors is usually inhibitory (Jordan 2005). By contrast, activation of 5-HT2A and 5-HT2B receptors exhibited predominantly excitatory effects (Jordan 2005). Consistent with this mechanism, the absence of 5-HT2C receptors in transgenic mice is associated with audiogenic seizure susceptibility that can result in death (Applegate and Tecott 1998). Vagus nerve stimulation is an effective antiepileptic therapy in patients (Theodore 2005; Nemeroff 2007) and has also been shown to increase the ἀring of 5-HT neurons in the dorsal raphe, which is mediated, in part, via projections from nucleus of the solitary tract (Dorr and Debonnel 2006). Modulation of respiratory network function by 5-HT may result, in part, from activation of 5-HT3 receptors (Kubin et al. 2006). Endogenous activation of 5-HT2A receptors is required for the generation of the respiratory rhythm in vitro (Pena and Ramirez 2002). Enhanced activation of 5-HT receptors after generalized convulsive seizure may be an effective approach to prevention of seizure-induced respiratory arrest. Activation of 5-HT receptors (particularly 5-HT4 receptors) (Manzke et al. 2003; Meyer et al. 2006) effectively controls respiratory disturbances, and activation of 5-HT pathways may be an important approach to treatment of life-threatening respiratory disorders in patients (Richter et al. 2003). Blockade of 5-HT1 and 5-HT2 receptors by the 5-HT antagonist, methysergide, decreased respiratory frequency in vivo and in vitro

DBA Mice as Models of Sudden Unexpected Death in Epilepsy

665

(Bodineau et al. 2004). A selective serotonin reuptake inhibitor, paroxetine, signiἀcantly increased respiratory rates in rodents (Olsson et al. 2004).

41.6â•…Expression of 5-HT Receptor Proteins and Transporter in C57BL/6J and DBA/2 Mice The deἀcit in serotonergic activation in DBA/2 mice, which are susceptible to audiogenic seizures and respiratory arrest, may result from altered expression of speciἀc subtypes of 5-HT receptors or the 5-HT transporter in respiratory centers of the brainstem. To test this hypothesis, antibodies against 5-HT2B, 5-HT2C, 5-HT3B, and 5-HT4 receptor proteins and the 5-HT transporter were used in Western blot experiments to compare the expression levels of these proteins in the caudal brainstem of DBA/2 vs. control (C57BL/6J) mice. C57BL/6J mice are genetically related to DBA/2 mice but do not exhibit audiogenic seizure susceptibility. The presence of 5-HT2B, 5-HT2C, 5-HT3B, and 5-HT4 receptors in the brainstem has been previously observed, and 5-HT neurotransmission in brainstem respiratory networks is known to contribute to increasing frequency of respiration in response to elevated carbon dioxide levels in normal animals (Feldman et al. 2003; Di Giovanni et al. 2006; Benarroch 2007; Hodges and Richerson 2008). Selective serotonin reuptake inhibitors increase the frequency of respiratory rhythm in medullary brain slices (Doi and Ramirez 2008). Null mutant mice lacking serotonin 5-HT2C receptors are susceptible to generalized convulsive seizure induced by sound, and these audiogenic seizures also result in postictal respiratory arrest (Brennan et al. 1997), supporting the hypothesis that an alteration in the activation of speciἀc subtypes of 5-HT receptors may result in respiratory deἀcits. In our experiments, the tissue extracts from the caudal brainstem of C57BL/6J and DBA/2 mice were immunostained on the same blots. Normalized (to the amount of actin) densities of the receptor protein from the two strains of mice were compared. In each experiment, caudal brainstems of DBA/2 and C57BL/6J mice of equal age were collected. Possible changes in serotonergic activation in DBA/2 mice may result from reduced levels of speciἀc inhibitory 5-HT receptors whose deἀciency is associated with occurrence of seizures, for example, 5-HT2C receptors (Brennan et al. 1997; Applegate and Tecott 1998). Possible differences of expression levels of 5-HT receptor proteins (5-HT2B, 5-HT2C, 5-HT3B, and 5-HT4) in the caudal brainstem of DBA/2 and C57BL/6J (control) mice were compared (Figure 41.3 and Table 41.1). The results indicated that DBA/2 mice express€signiἀcantly lower levels of 5-HT2C, 5-HT3B, and 5-HT4 receptors than control mice. Signiἀcant differences, 26% (n = 10; p < 0.02), 21% (n = 9; p < 0.035), and 28% (n = 7; p < 0.025) in the expression of 5-HT2C, 5-HT3B, and 5-HT4 receptor proteins, respectively, were detected in the caudal brainstem of DBA/2 relative to control mice (Uteshev et al. 2010). These observations are consistent with the hypothesis that a deἀciency in serotonergic activation in the brainstem of DBA/2 mice contributes to susceptibility to audiogenic seizures and respiratory arrest. By contrast, excitatory 5-HT2B receptors were expressed at signiἀcantly higher (n = 9; p < 0.001) levels in DBA/2 mice (143%) than in C57BL/6J mice (100%). The signiἀcant differences in expression of 5-HT receptor subtypes in the brainstem of audiogenic seizure-non-susceptible mice vs. seizure-prone DBA/2 mice suggest an important role for 5-HT receptors (including excitatory 5-HT2B and inhibitory 5-HT2C) in susceptibility to seizures and respiratory arrest in this DBA/2 SUDEP model. Understanding the molecular

666 Sudden Death in Epilepsy: Forensic and Clinical Issues (a)

(b)

5-HT2B 5-HT2B

5-HT2C

C57 DBA/2

(56 kDa)

Actin (43 kDa)

100%

100 75 50 25 0

100%

100 Relative expression (%)

Relative expression (%)

125

Actin (43 kDa) 143.5±24.2% p < 0.01 n=7

150

C57 DBA/2

5-HT2C (46 kDa)

74.3±32.4% p < 0.02 n = 10

75 50 25 0 C57BL/6J DBA/2

C57BL/6J DBA/2

Figure 41.3╇ Western blot analysis of 5-HT2B and 5-HT2C receptor proteins in C57BL/6J and DBA/2 mice. In each set of experiments, the caudal portions of the brainstem, containing the respiratory centers, of three DBA/2 and three C57BL/6J (control) mice were collected for Western blot analysis. Fresh brain tissue from each group of mice was homogenized separately, and specific antibodies against inhibitory 5-HT2C and excitatory 5-HT2B were used for labeling. The results indicate that the caudal brainstems of DBA/2 mice express significantly greater amounts of 5-HT 2B (143.5 ± 24.2% of control mice (a) receptors, which are excitatory, and significantly lower amounts of 5-HT 2C (74.3 ± 32.4% of control mice (b) receptors, which are inhibitory, compared to control mice. These results are consistent with the behavioral differences between DBA/2 (seizure-Â�susceptible) and C57BL/6J (seizure-resistant) and the previous finding that 5-HT2C knockout mice exhibit susceptibility to audiogenic seizures and respiratory arrest. (Reprinted by permission from Macmillan Publishers Ltd: Brennan, T. J., et al. Nat Genet. 16:387–390, 1997. Copyright 1997.)

basis of susceptibility to SUDEP can assist in developing more selective 5HT-based treatments for prevention of SUDEP.

41.7╅ Sudden Infant Death Syndrome and SUDEP The relevance of serotonergic neurotransmission to respiration is also supported by the abnormalities in 5-HT neurotransmission, particularly polymorphisms in the 5-HT Table 41.1╅ Expression of Serotonergic Receptor and Transporter Proteins in the Caudal Brainstem of C57BL/6J and DBA/2 Mice C57BL/6J DBA/2 No. of groups (three€animals/group) p value

5-HT2B

5-HT2C

5-HT3B

5-HT4

Transporter

100 143.5 ± 24.2 n=9

100 74.3 ± 32.4 n = 10

100 78.5 ± 25.1 n=9

100 72.0 ± 24.2 n=7

100 116.0 ± 21.0 n=9

C (Hindocha et al. 2008). The mutation resulted in the amino acid exchange I1867T within the intracellular C-terminal domain. Two of the family members, a 23-yearold woman and a 19-year-old woman, had died of SUDEP. The 23-year-old woman had afebrile, generalized tonic–clonic seizures throughout her life and the 19-year-old woman had febrile, generalized tonic–clonic seizures, absences, and focal-onset seizures since she was 14 months old and afebrile generalized tonic–clonic seizures since she was 2 years old. Arguments for a causative role of the mutation in these two patients are that two SUDEP cases occurred within the same family and that the SCN1A gene product was also detected in the sinoatrial node of mice (Hindocha et al. 2008). The latter ἀnding suggests that triggering of arrhythmias by a seizure may be facilitated by the presence of ion channel mutations within the cardiac conduction system. The mutation may also affect brainstem control of respiration or other autonomic functions, as in patients with severe myoclonic epilepsy at infancy, which is also caused by SCN1A mutations (Kimura et al. 2005). In a 25-yearold patient with SUDEP, postmortem DNA analysis revealed a novel missense mutation in the cardiac voltage-gated sodium-channel alpha-subunit gene SCN5A (Aurlien et al. 2009). Because epilepsy in this patient was treated with lamotrigine, it could not be decided whether SUDEP was causally related to the SCN5A mutation or to the side effects of lamotrigine, which is known to inhibit cardiac potassium channels (Aurlien et al. 2009). 42.2.2â•… Cardiotoxic Drugs All drugs known to have a cardiotoxic effect may be potentially dangerous for patients at risk for SUDEP. Most investigations, however, were directed toward the arrhythmogenic side effects of antiepileptic drugs (Table 42.2). Unfortunately, these studies only investigated the side effects of antiepileptic drugs but not of other concomitant cardiotoxic Table 42.2â•… Antiepileptic Drugs and Their Potential Cardiotoxicity Phenobarbital Phenytoin Carbamazepine Oxcarbazepine Valproic acid Lorazepam Clobazam Clonazepam Felbamate Gabapentin Tiagabine Lamotrigine Topiramate Vigabatrin Levetiracetam Pregabalin Zonisamide Lacosamide

Arrhythmias

Negative Inotropy

No Yes Yes No No Yes Yes No No Yes No Yes Yes No No Yes No No

No Yes Yes No Yes No No No No No No No No No No Yes No No

682 Sudden Death in Epilepsy: Forensic and Clinical Issues

drugs, which may also have arrhythmogenic side effects, such as antibiotics or neuroleptics (Stöllberger and Finsterer 2004). Antibiotics, in which an arrhythmogenic effect has been reported, include penicillin (AV blocks I and II), ampicillin (QT prolongation), cefotaxime (arrhythmia), ciprofloxazine (QT prolongation), moxifloxacine (sinus tachycardia), sulfametoxazole (QT prolongation), erythromycine (prolonged depolarization, QT prolongation, torsades des pointes), and doxycycline (supraventricular tachycardia, sporadic Wenkebach block). Concerning the question of whether or not antiepileptic drugs€predisÂ� pose to SUDEP, the results are conflicting (Kloster and Engelskjøn 1999; Langan et al. 2005; Lathers and Schraeder 2002; Nilsson et al. 2001; Walczak 2003). However, it is generally accepted in most to studies that the risk of SUDEP increases with the increased number of antiepileptic drugs being taken by the patient (Nilsson et al. 1999; Walczak et al. 2001). Most data on cardiotoxicity of antiepileptic drugs are available for carbamazepine. Carbamazepine has been shown to slow atrioventricular conduction, to increase the sympathetic tone, and to suppress parasympathetic functions (Isojärvi et al. 1998). In addition, abrupt withdrawal of carbamazepine resulted in enhanced sympathetic activity in sleep, as evidenced by reduced heart rate variability (Hennessy et al. 2001). This effect was explained by a rebound to the withdrawal of the sedating, relaxing, and antidepressive effects of carbamazepine. Heart rate variability was also generally reduced during sleep in patients taking carbamazepine in another study (Persson et al. 2007) and increased when carbamaÂ� zepine was withdrawn (Lossius et al. 2007). In a series of 14 SUDEP patients, 85% had been treated with carbamazepine compared to 38% of the non-SUDEP cases (Timmings 1998). In three larger controlled studies, however, carbamazepine was used with equal frequency in SUDEP patients and controls (Walczak et al. 2001; Kloster and Engelskjøn 1999; Nilsson et al. 1999). A further study found frequent changes of carbamazepine dosages with concentrations above the upper therapeutic range to be a risk factor for SUDEP (Nilsson et al. 2001). Oral phenytoin rarely induced AV block (Stöllberger and Finsterer 2004), whereas phenobarbital and valproic acid were not arrhythmogenic (Walczak 2003) (Table 42.2). In studies associated with the development of new antiepileptic drugs and reviewing data on the risk of SUDEP, lamotrigine, gabapentine, tiagabine, topiramate, and zonisamide were not associated with an increased risk of SUDEP in the young epilepsy population (Leestma et al. 1997; Walczak 2003). In addition, in a study of 4700 patients on lamotrigine, 45 deaths occurred, of which 18 were classiἀed as SUDEP. The rate of SUDEP was not increased compared to young adults with severe epilepsy and taking antiepileptic drugs other than lamotrigine (Leestma et al. 1997). Recently, however, it has been reported that SUDEP occurred in four females with idiopathic epilepsy who were under a monotherapy with lamotrigine (Aurlien et al. 2007). Lamotrigine was suspected of being a risk factor because it inhibits the cardiac rapid delayed rectiἀer potassium ion current (Ikr), which has been shown to increase the risk of cardiac arrhythmias and sudden cardiac death (Aurlien et al. 2007). Arrhythmogenic side effects of other antiepileptic drugs have been studied only in small series. 42.2.3â•… Other Possible Cardiac Risk Factors Although every cardiac disease has the potential to be a risk factor of SUDEP, only a few were thought to increase the risk of epilepsy patients dying from SUDEP. One of these presumed risk factors is cold (outside temperatures below 0°C), which has been identiἀed as a risk factor for sudden cardiac death (Colugnati et al. 2008). In rats with epilepsy,

Cardiac and Pulmonary Risk Factors and Pathomechanisms of SUDEP

683

exposure to cold increased the heart rate, causing the authors to suggest that cold may be a cardiovascular risk factor for SUDEP (Sonoda et al. 2008). Another potential risk factor could be a short QTc interval. In a recent cohort study, it has been shown that QTc is shortened in patients with epilepsy, particularly in patients with cryptogenic epilepsy (The et al. 2007). Whether shortened QTc interval is indeed a risk factor for SUDEP, however, remains speculative. There is a lack of data indicating that the cardiac conduction system is affected in SUDEP patients. There was no morphological difference of the cardiac conduction system in an autopsy study of 10 SUDEP patients and 10 matched controls (Opeskin et al. 2000). However, in another autopsy study on 15 SUDEP patients, 6 showed ἀbrotic changes in the subendocardial and deep myocardium (P-Codrea Tigaran et al. 2005), which were attributed to epilepsy. Some authors also consider electrolyte imbalances, particularly hyponatremia, known to trigger arrhythmias, as a risk factor of SUDEP (Ruis Ginez et al. 2007). So far, there is no indication that arterial hypertension has any influence on the outcome of patients with uncontrolled epilepsy. Whether previous Takotsubo syndrome, also known as reversible left ventricular apical ballooning, is a potential risk factor of SUDEP is so far unknown because this question has not been addressed in an appropriate study, most likely because of the low number of patients with this abnormality. Takotsubo syndrome, also known as reversible left ventricular apical ballooning, is a stress-induced myocardial contraction failure due to supposed nonresponsiveness to adrenergic stimulation (Dorfman and Iskandrian 2009).

42.3â•… Cardiac Pathomechanisms Although a number of potential pathomechanism for SUDEP are discussed (Table 42.3), no ἀnal decision has been made as to whether a single or several pathomechanism(s) lead(s) Table 42.3â•… Potential Cardiac or Pulmonary Pathomechanisms of SUDEP Heart (Hitiris et al. 2007; Scorza et al. 2008) (Almansori et al. 2006; Devinsky et al. 1997; Leung et€al. 2006) High-grade AV block (Altenmüller et al. 2004) Ictal asystole (Carinci et al. 2007; Devinsky et al. 1997; Leung et al. 2006; Schuele et al. 2007; So and Sperling 2007) Tachycardia (Langan et al. 2000; Zaidi et al. 1997) Transmission of epileptic activity to the heart via the (Scorza et al. 2008) autonomic system Impaired sympathetic cardiac innervation (Kerling et al. 2009)

Arrhythmias during or between seizures Bradycardia

Postictal central apnea Postictal obstructive apnea Acute neurogenic pulmonary edema Postictal laryngospasm Ictal hypoxemia, hypercapnia

Lung (Jehi and Najm 2008; Nashef et al. 1996; Ryvlin et al. 2009) (Ryvlin et al. 2009) (Jehi and Najm 2008; Terrence et al. 1981) (Tavee and Morris 2008) (Bateman et al. 2008)

684 Sudden Death in Epilepsy: Forensic and Clinical Issues

to SUDEP (Bell and Sander 2006). In general, potential pathomechanism of SUDEP may be classiἀed as those in which a preexisting, subclinical pathology becomes symptomatic during the seizure and those in which the seizure induces a pathologic reaction in an otherwise healthy individual. Actually, the most widely discussed pathomechanism assumes preexisting cardiopulmonary abnormalities to manifest during tonic–clonic seizures. The most frequently suggested cardiac pathomechanism of SUDEP is induction of asystole, ventricular tachycardia, or ventricular ἀbrillation. There is no deἀnitive proof that cardiac arrhythmias during or between seizures or transmission of epileptic activity to the heart via the central and peripheral autonomic nervous system plays a potential pathogenetic role in SUDEP, but animal studies suggest autonomic dysfunction during seizures to contribute to the fatal outcome of such events (Lathers and Schraeder 1982; Schraeder and Lathers 1983; Scorza et al. 2008). Other authors propose postictal central or obstructive apnea (respiratory arrest) to represent the most likely pathomechanism of SUDEP (Ryvlin et al. 2009). 42.3.1â•… Ictal Arrhythmias Epileptic seizures are potentially arrhythmogenic (So et al. 2000). In a previous study of 43 patients with intractable epilepsy and a total of 51 seizures, seizures were associated with ECG abnormalities such as asystole (n = 1), atrial ἀbrillation (n = 1), marked or moderate sinus arrhythmia (n = 6), supraventricular tachycardia (n = 1), atrial premature depolarization (n = 8), ventricular premature depolarization (n = 2), or bundle-branch block (n = 3) (Nei et al. 2000). ECG abnormalities in patients with intractable epilepsy were also reported by others (Zijlmans et al. 2002). In a study on 11 SUDEP patients and 11 matched controls with epilepsy, the QTc interval increased ictally in both groups (Tavernor et al. 1996). In rats, repeated electrical stimulation of the insula induced progressive heart blocks, resulting in escape rhythms, premature ventricular contraction, or death (Oppenheimer et al. 1991). In a study on 21 SUDEP patients, maximal ictal heart rate was higher than in controls and ictal ECG abnormalities occurred in 56% of the patients: atrial ἀbrillation (n = 2), premature ventricular depolarization (n = 2), sinus arrhythmia (n = 2), atrial premature depolarization (n = 2), junctional escape rhythm (n = 1), and ST elevation (n = 1) (Nei et al. 2004). The authors supposed these ECG abnormalities to be contributory to the occurrence of SUDEP. 42.3.1.1â•… Bradyarrhythmias and Asystole Bradyarrhythmias have been observed repeatedly during seizures and may be explained by the influence of the central autonomic network on cardiac impulse generation, although localization and lateralization issues need to be considered in the light of patterns of seizure spread, hand dominance, or presence of lesions (Leung et al. 2006). Epileptic seizures may particularly induce bradycardia and asystole (Carinci et al. 2007). During a study using loop recording over 2 years, 21% of the patients with intractable epilepsy had at least one ictal bradycardia (Rugg-Gunn et al. 2004). Altogether, 2% of the seizures were associated with bradycardia. Two thirds of the patients with ictal bradycardia had temporal lobe epilepsy and the remaining had frontal lobe epilepsy (Tinuper et al. 2001). Although the literature about asystole in epilepsy is replete with single case reports and small case series (Strzelczyk et al. 2008), there is a consensus that ictal asystole is a rare event and is reported to occur in only 0.27% of the epilepsy patients experiencing a seizure

Cardiac and Pulmonary Risk Factors and Pathomechanisms of SUDEP

685

under video-EEG monitoring (Schuele et al. 2007). In another video-EEG study, 0.4% of the investigated patients had ictal bradycardia or asystole (Rocamora et al. 2003). Asystole during or after seizures most frequently occurs in patients with temporal lobe epilepsy, rarely in patients with extratemporal lobe epilepsy, and not in patients with generalized epilepsy (Schuele et al. 2007). Asystole may be preceded by a loss of muscle tone (Schuele et al. 2007; So and Sperling 2007). Seizures and cardiac syncope are often clinically indistinguishable. Cerebral hypoxemia due to asystole or ventricular ἀbrillation may induce clinical features imitating partial or secondarily generalized tonic–clonic seizures, and thus may be misinterpreted as epilepsy. Because such cardiac abnormalities are potentially fatal, and because a cardiac syncope may be misdiagnosed as epilepsy, the frequency of SUDEP may be difficult to ascertain (Venkataraman et al. 2001). 42.3.1.2â•… Tachyarrhythmias Generally, tachyarrhythmias may be induced by ischemia, metabolic changes, electrolyte disturbances, myocardial ἀbrosis, or abnormalities of the cardiac conduction system. In a loop-recorder study on 20 patients with refractory epilepsy, 16 had ictal tachycardia (RuggGunn et al. 2004). In addition, patients with epilepsy have been shown to develop torsades de pointes from long-QT syndrome, atrioventricular nodal reentry tachycardia, or sinus arrest (Langan et al. 2000; Zaidi et al. 1997; Venkataraman et al. 2001). In another study, it turned out that 50% of seizures resulted in tachycardia within the ἀrst 10 s of EEG recordings. Overall, there are conflicting results about whether bradyarrhythmias, including asystole (Nabbout 2008; Rugg-Gunn et al. 2004), or tachyarrhythmias (Nashef et al. 2007) contribute more frequently to the occurrence of SUDEP. 42.3.1.3â•… Autonomic Mechanisms There are some indications that epileptogenic activity may induce impaired cardiac impulse generation or propagation abnormalities, being explained by the fact that cardiovascular responses, including heart rate variability, have a cortical representation in the insular cortex, cingulate gyrus, and prefrontal cortex (Jehi and Najm 2008; Leung et al. 2006). At the subcortical level, the hypothalamus manages autonomic function to maintain homeostasis. Cortical and subcortical autonomic centers are linked by the amygdala, an integral component of the limbic system, which mediates the autonomic response to emotions (Jehi and Najm 2008). Interestingly, the amygdala, gyrus cinguli, insular cortex, and frontopolar and frontoorbital regions are the most frequent foci of partial epileptic activity, explaining why autonomic disturbances are frequently associated with the occurrence of supraventricular tachycardia, sinus tachycardia, sinus bradycardia, sinus arrest, atrioventricular block, or asystole (Devinsky et al. 1997). Direct transmission of epileptogenic discharges to the central or peripheral sympathetic cardiovascular system has been conἀrmed in various animal studies (Lathers et al. 1987; Lathers and Schraeder 1982, 1987; Schraeder and Lathers 1983). Additional arguments for involvement of the autonomic nervous system in the pathogenesis of SUDEP have been raised. After successful temporal lobe epilepsy surgery, the risk of sympathetically mediated tachyarrhythmias was lessened (Hilz et al. 2002). Physiologic heart rate and blood pressure modulation, mediated by both the parasympathetic and sympathetic nervous systems, were reduced in patients with epilepsy (Isojärvi et al. 1998). Patients with temporal lobe epilepsy also had reduced low-frequency power

686 Sudden Death in Epilepsy: Forensic and Clinical Issues

(low-frequency power represents sympathetic activity and high-frequency power represents parasympathetic activity) on spectral analysis of 24 h-ECG (Tomson et al. 1998). Other studies found both power of low and high frequencies to be reduced in patients with refractory epilepsy (Ansakorpi et al. 2002). Heart rate increased more during sleep seizures than during wake seizures (Nei et al. 2004), possibly reflecting the common occurrence of SUDEP at night (Langan et al. 2005). Overall, there is increased acknowledgment of the role of the autonomic nervous system for “epileptogenic” arrhythmias in SUDEP. 42.3.2â•… Systolic Dysfunction Epileptic activity may not only induce arrhythmia but may also lead to systolic dysfunction. Autopsy of SUDEP patients has shown that hearts are heavier and more dilated compared to controls (Leestma et al. 1997, 1989) and that pulmonary edema is present in 50–86% of the SUDEP cases (Kloster and Engelskjøn 1999; Leestma et al. 1997, 1989). In a sheep model of status epilepticus, animals that died suddenly were found to have developed pulmonary hypertension, an enlarged left atrium, and pulmonary edema. Autonomic dysfunction with altered pulmonary vascular tone and/or cardiac dysfunction is thought to be responsible for pulmonary congestion and edema in SUDEP patients (Pezzella et al. 2009). Whether the pathomechanism of left ventricular dysfunction in epilepsy is the same as that in subarachnoid hemorrhage (Deibert et al. 2003) is unknown. Possibly, stress from seizures or subarachnoid hemorrhage triggers a Takotsubo syndrome, which is characterized not only by systolic dysfunction but also by severe arrhythmias (Stöllberger et al. 2005). Although Takotsubo syndrome has not been described thus far in SUDEP, it has been repeatedly described in patients with epilepsy (Legriel et al. 2008; Stöllberger et al. 2009).

42.4â•… Pulmonary Risk Factors of SUDEP So far, there is minimal evidence that any primary pulmonary disease could be a deἀnite risk factor for SUDEP. For example, it is unknown if patients with respiratory chain defects, which frequently present with muscle disease and epilepsy and may develop muscular respiratory insufficiency, have an increased risk not to survive a tonic–clonic seizure. These patients appear particularly endangered because they often have epilepsy and because their epilepsy is often difficult to treat. Despite this uncertainty about pulmonary risk factors, there are frequent reports about patients who develop severe pulmonary problems during or after seizures, such as ictal hypoxemia or hypercapnia (Bateman et al. 2008), apnea (Bell and Sander 2006; Jehi and Najm 2008; Ryvlin et al. 2009), acute neurogenic pulmonary edema (Jehi and Najm 2008), or postictal laryngospasm (Tavee and Morris 2008) (Table 42.1). In a prospective autopsy series on 52 SUDEP patients, 80% had pulmonary congestion and edema (Leestma et al. 1989). It is uncertain if these abnormalities are due to primary pulmonary, cardiac, or laryngeal mechanisms. It is unclear if only patients with previous lung disease, as opposed to previously healthy subjects, develop such problems. There is also little information available about the effects of a tonic–clonic seizure on the respiratory system in general. Do tonic–clonic seizures generally induce bronchospasm, or loss of tone of the muscles involved in respiration? Recent investigations have shown that at least the vital capacity, forced vital capacity, and forced expiratory volume within the

Cardiac and Pulmonary Risk Factors and Pathomechanisms of SUDEP

687

ἀrst second, are not signiἀcantly different between healthy subjects and epilepsy patients (Scorza et al. 2007).

42.5â•…Respiratory Pathomechanisms Suspected pulmonary pathomechanisms for SUDEP include ictal or postictal central or obstructive apnea (Table 42.3) (Pezzella et al. 2009; Ryvlin et al. 2009). However, seizureinduced apnea and asystole might promote each other through cardiorespiratory reflexes or cerebral hypoxia (Tomson et al. 2008). Seizure-associated respiratory compromise, due to hypoventilation or obstructive apnea, is a prominent feature in witnessed SUDEP cases (Langan et al. 2000; Nashef et al. 1998; So et al. 2000). Polysomnography demonstrated central as well as obstructive ictal apnea to be sometimes associated with bradycardia (Nashef et al. 1996). Ictal apnea was also reported in 60% of patients undergoing video-EEG monitoring (Nashef et al. 1996). In other studies, however, postictal pulmonary edema is a rare ἀnding in SUDEP patients (Swallow et al. 2002). Postictal apnea was also reported in a number of animal studies (So 2008). In one experimental animal study, one-third of the animals died of hypoventilation and showed pulmonary edema at autopsy (Johnston et al. 1995). A role for mechanical obstruction in SUDEP is substantiated by a study that reported SUDEP cases to have been found more often in the prone rather than in the supine position (Kloster and Engelskjøn 1999). The prone position may cause obstruction of the nose and mouth by exerting pressure against the bed clothing. In addition, the prone position may impede ventilation by decreasing vital capacity and tidal volume, particularly after exercise (Röggla and Röggla 1999). Other studies, however, found prone position only to be a weak risk factor for SUDEP (Monté et al. 2007). Severe postictal laryngospasm was reported in a single patient during video-EEG monitoring and assumed to be causal for SUDEP (Tavee and Morris 2008). SUDEP resulting from apnea could result from epileptogenic activity transmitted to the medullar respiratory centers resulting in suppression of brainstem respiratory activity (So 2008). Apnea could also result from cardiorespiratory reflexes triggered by the seizure (Nashef et al. 1996; So 2008). These reflexes are more intense in the young population compared to the elderly, possibly reflecting the higher incidence of SUDEP in young adults (So 2008). In addition, hypotonia of the respiratory muscles, laryngospasm, or bronchospasm may be further involved in seizure-related apnea. Overall, it is unknown whether patients who develop apnea during or after seizures have a higher risk of SUDEP than patients without apnea. Furthermore, no information is available about whether apnea during seizures occurs more often in patients with preexisting pulmonary diseases, such as chronic obstructive pulmonary disease, emphysema, bronchial asthma, in smokers or nonsmokers, or in patients without pulmonary disease.

42.6â•… Clinical Implications 42.6.1â•… Interictal As a consequence of the data so far available about cardiac and pulmonary risk factors and potential pathomechanisms of SUDEP, we propose that patients at risk for SUDEP

688 Sudden Death in Epilepsy: Forensic and Clinical Issues

should undergo comprehensive cardiac and pulmonary investigations (Finsterer and Stöllberger 2009). These evaluations should not only include individual and family history and physical examination for cardiac and pulmonary disease and basic cardiac and pulmonary instrumental investigations but also echocardiography, stress test, 24-h ECG or loop recording, lung function testing, otorhinolaryngological investigations, and polysomnography (Finsterer and Stöllberger 2009). Antiepileptic drugs or other drugs with known or presumed cardiac or pulmonary toxicity should, if possible and arguable, be replaced by nontoxic medication. In addition, interactions between antiepileptic drugs and drugs given for comorbidities should be considered. The family history should be directed toward eliciting histories of syncope, sudden cardiac death, or hereditary arrhythmias, such as Wolf–Parkinson–White syndrome, AV blocks, Brugada syndrome, hereditary atrial ἀbrillation, or short- and long-QT syndrome. In cases where these investigations discover premorbid cardiac or pulmonary disease, optimal antiepileptic, cardiac, and pulmonary treatment should be provided to these patients, particularly when new comorbidities appear. 42.6.2â•… Ictal Continuous ECG recording during video-EEG monitoring would be of additional help in clarifying the pathogenesis of SUDEP. Even better would be a continuous ictal and interictal ECG recording by long-term loop recording in patients at risk to develop SUDEP (Rugg-Gunn et al. 2004). Only with such interventions would ECG changes such as bradyarrhythmias, asystole, or ventricular tachycardia actually be documented during seizures. Such observations could indicate the need for the implantation of pacemakers or even implantable cardioverter deἀbrillators. 42.6.3â•… Postictal Patients at risk for SUDEP should speciἀcally undergo cardiac and pulmonary investigations immediately after a seizure, irrespective of its phenotype and clinical outcome. Acute postictal investigations should include immediate physical examination, blood chemical investigations (creatine kinase, electrolytes, troponin, glucose, brain natriuretic peptide), ECG, x-ray of the lung, and echocardiography. To detect Takotsubo syndrome, a postictal ECG should be evaluated for ST- and T-wave changes and the echocardiogram evaluated for systolic dysfunction. If Takotsubo syndrome is detected, the patient should be intensively observed for potentially life-threatening complications of this syndrome (Stöllberger et al. 2005).

42.7â•… Conclusion There is general agreement that more intensive ictal, interictal, and postictal investigations for clinical and subclinical cardiac and pulmonary disease in patients at risk for SUDEP are required. For this purpose, international organizations should provide a common deἀnition of patients at risk for SUDEP and these patients should then be identiἀed and systematically investigated for clinical and subclinical cardiac and pulmonary disease. More comprehensive cardiac and pulmonary investigations during seizures and

Cardiac and Pulmonary Risk Factors and Pathomechanisms of SUDEP

689

postictally, development of more non-cardiotoxic and non-pulmonary-toxic antiepileptic drugs, research on interactions between antiepileptic drugs and drugs taken for comorbidities, and adequate identiἀcation and treatment of cardiac and pulmonary disease and epilepsy may prevent the fatal outcome in patients at risk for SUDEP.

References Almansori, M., M. Ijaz, and S. N. Ahmed. 2006. Cerebral arrhythmia influencing cardiac rhythm: A case of ictal bradycardia. Seizure 15: 459–461. Altenmüller, D. M., M. Zehender, and A. Schulze-Bonhage. 2004. High-grade atrioventricular block triggered by spontaneous and stimulation-induced epileptic activity in the left temporal lobe. Epilepsia 45: 1640–1644. Ansakorpi, H., J. T. Korpelainen, H. V. Huikuri, U. Tolonen, V. V. Myllylä, and J. I. Isojärvi. 2002. Heart rate dynamics in refractory and well controlled temporal lobe epilepsy. J Neurol Neurosurg Psychiatry 72: 26–30. Aurlien, D., T. P. Leren, E. Taubøll, and L. Gjerstad. 2009. New SCN5A mutation in a SUDEP victim with idiopathic epilepsy. Seizure 18: 158–160. Aurlien, D., E. Taubøll, and L. Gjerstad. 2007. Lamotrigine in idiopathic epilepsy—Increased risk of cardiac death? Acta Neurol Scan 115: 199–203. Bateman, L. M., C. S. Li, and M. Seyal. 2008. Ictal hypoxemia in localization-related epilepsy: Analysis of incidence, severity and risk factors. Brain 131: 3239–3245. Bell, G. S., and J. W. Sander. 2006. Sudden unexpected death in epilepsy. Risk factors, possible mechanisms and prevention: A reappraisal. Acta Neurol Taiwan 15: 72–83. Carinci, V., G. Barbato, A. Baldrati, and G. Di Pasquale. 2007. Asystole induced by partial seizures: A rare cause of syncope. Pacing Clin Electrophysiol 30: 1416–1419. Colugnati, D. B., R. M. Arida, S. L. Cravo, et al. 2008. Hibernating mammals in sudden cardiac death in epilepsy: What do they tell us? Med Hypotheses 70: 929–932. Deibert, E., B. Barzilai, A. C. Braverman, et al. 2003. Clinical signiἀcance of elevated troponin I levels in patients with nontraumatic subarachnoid hemorrhage. J Neurosur 98: 741–746. Devinsky, O., S. Pacia, and G. Tatambhotla. 1997. Bradycardia and asystole induced by partial seizures: A case report and literature review. Neurology 6: 1712–1714. Dorfman, T. A., and A. E. Iskandrian. 2009. Takotsubo cardiomyopathy: State-of-the-art review. J Nucl Cardiol 16: 122–134. Finsterer, J., and C. Stöllberger. 2009. Cardiopulmonary surveillance to prevent SUDEP. Lancet€Neurol 8: 131–132. Hennessy, M. J., M. G. Tighe, C. D. Binnie, and L. Nashef. 2001. Sudden withdrawal of carbaÂ� mazepine€increases cardiac sympathetic activity in sleep. Neurology 57: 1650–1654. Hilz, M. J., O. Devinsky, W. Doyle, A. Mauerer, and M. Dütsch. 2002. Decrease of sympathetic cardiovascular modulation after temporal lobe epilepsy surgery. Brain 125: 985–995. Hindocha, N., L. Nashef, F. Elmslie, et al. 2008. Two cases of sudden unexpected death in epilepsy in a GEFS+ family with an SCN1A mutation. Epilepsia 49: 360–365. Hitiris, N., R. Mohanraj, J. Norrie, and M. J. Brodie. 2007. Mortality in epilepsy. Epilepsy Behav 10: 363–376. Hughes, J. R. 2009. A review of sudden unexpected death in epilepsy: Prediction of patients at risk. Epilepsy Behav 14: 280–287. Isojärvi, J. I., H. Ansakorpi, K. Suominen, U. Tolonen, M. Repo, and V. V. Myllyla. 1998. Interictal cardiovascular autonomic responses in patients with epilepsy. Epilepsia 39: 420–426. Jehi, L., and I. M. Najm. 2008. Sudden unexpected death in epilepsy: Impact, mechanisms, and prevention. Cleve Clin J Med 75 (s2): S66–S70. Johnston, A., and P. Smith. 2007. Sudden unexpected death in epilepsy. Expert Rev Neurother 7: 1751–1761.

690 Sudden Death in Epilepsy: Forensic and Clinical Issues Johnston, S. C., J. K. Horn, J. Valente, and R. P. Simon. 1995. The role of hypoventilation in a sheep model of epileptic sudden death. Ann Neurol 37: 531–537. Kerling, F., M. Dütsch, R. Linke, T. Kuwert, H. Stefan and M. J. Hilz. 2009. Relation between ictal asystole and cardiac sympathetic dysfunction shown by MIBG-SPECT. Acta Neurol Scand 120 (2): 123–129. Kimura, K., T. Sugawara, E. Mazaki-Miyazaki, et al. 2005. A missense mutation in SCN1A in brothers with severe myoclonic epilepsy in infancy (SMEI) inherited from a father with febrile seizures. Brain Dev 27: 424–430. Kloster, R., and T. Engelskjon. 1999. Sudden unexpected death in epilepsy (SUDEP): A clinical perspective and a search for risk factors. J Neurol Neurosurg Psychiatry 67: 439–444. Langan, Y., L. Nashef, and J. W. Sander. 2005. Case-control study of SUDEP. Neurology 64: 1131–1133. Langan, Y., L. Nashef, and J. W. A. S. Sander. 2000. Sudden unexpected death in epilepsy: A series of witnessed deaths. J Neurol Neurosurg Psychiatry 68: 211–213. Lathers, C. M., and P. L. Schraeder. 2002. Clinical pharmacology: Drugs as a beneἀt and/or risk in sudden unexpected death in epilepsy? J Clin Pharmacol 42: 123–136. Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67: 247–259. Lathers, C. M., and P. L. Schraeder. 1987. Review of autonomic dysfunction, cardiac arrhythmias, and epileptogenic activity. J Clin Pharmacol 27: 346–356. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23: 633–647. Leestma, J. E., J. F. Annegers, M. J. Brodie, et al. 1997. Sudden unexplained death in epilepsy: Observations from a large clinical development program. Epilepsia 38: 47–55. Leestma, J. E., T. Walczak, J. R. Hughes, M. B. Kalelkar, and S. S. Teas. 1989. A prospective study on sudden unexpected death in epilepsy. Ann Neurol 26: 195–203. Legriel, S., F. Bruneel, L. Dalle, et al. 2008. Recurrent takotsubo cardiomyopathy triggered by convulsive status epilepticus. Neurocrit Care 9: 118–121. Leung, H., P. Kwan, and C. E. Elger. 2006. Finding the missing link between ictal bradyarrhythmia, ictal asystole, and sudden unexpected death in epilepsy. Epilepsy Behav 9: 19–30. Lossius, M. I., J. E. Erikssen, P. Mowinckel, P. Gulbrandsen, and L. Gjerstad. 2007. Changes in autonomic cardiac control in patients with epilepsy after discontinuation of antiepileptic drugs: A randomized controlled withdrawal study. Eur J Neurol 14: 1022–1028. Monté, C. P., J. B. Arends, I. Y. Tan, A. P. Aldenkamp, M. Limburg, and M. C. de Krom. 2007. Sudden unexpected death in epilepsy patients: Risk factors. A systematic review. Seizure 16: 1–7. Nabbout, R. 2008. Can SCN1A mutations account for SUDEP? Commentary on Hindocha et al. Epilepsia 49: 367–368. Nashef, L., N. Hindocha, and A. Makoff. 2007. Risk factors in sudden death in epilepsy (SUDEP): The quest for mechanisms. Epilepsia 48: 859–871. Nashef, L., S. Garner, J. W. A. S. Sander, D. R. Fish, and S. D. Shorvon. 1998. Circumstances of death in sudden death in epilepsy: Interviews of bereaved relatives. J Neurol Neurosurg Psychiatry 64: 349–352. Nashef, L., F. Walker, P. Allen, J. W. A. S. Sander, S. D. Shorvon, and D. R. Fish. 1996. Apnoea and bradycardia during epileptic seizures: Relation to sudden death in epilepsy. J Neurol Neurosurg Psychiatry 60: 297–300. Nei, M., R. T. Ho, B. W. Abou-Khalil, et al. 2004. EEG and ECG in sudden unexplained death in epilepsy. Epilepsia 45: 338–345. Nei, M., R. T. Ho, and M. R. Sperling. 2000. EKG abnormalities during partial seizures in refractory epilepsy. Epilepsia 41: 542–548.

Cardiac and Pulmonary Risk Factors and Pathomechanisms of SUDEP

691

Nilsson, L., U. Bergman, V. Diwan, B. Y. Farahmand, P. G. Persson, and T. Tomson. 2001. Antiepileptic drug therapy and its management in sudden unexpected death in epilepsy: A case-control study. Epilepsia 42: 667–673. Nilsson, L., B. Y. Farahmand, P. G. Persson, I. Thiblin, and T. Tomson. 1999. Risk factors for sudden unexpected death in epilepsy: A case-control study. Lancet 353: 888–893. Opeskin, K., A. Thomas, and S. F. Berkovic. 2000. Does cardiac conduction pathology contribute to sudden unexpected death in epilepsy? Epilepsy Res 40: 17–24. Oppenheimer, S. M., J. X. Wilson, C. Guiraudon, and D. F. Cechetto. 1991. Insular cortex stimulation produces lethal cardiac arrhythmias: A mechanism of sudden death? Brain Res 550: 115–121. P-Codrea Tigaran, S., S. Dalager-Pedersen, U. Baandrup, M. Dam, and A. Vesterby-Charles. 2005. Sudden unexpected death in epilepsy: Is death by seizures a cardiac disease? Am J Forensic Med Pathol 26: 99–105. Persson, H., E. Kumlien, M. Ericson, and T. Tomson. 2007. Circadian variation in heart-rate variability in localization-related epilepsy. Epilepsia 48: 917–922. Pezzella, M., P. Striano, C. Ciampa, L. Errichiello, P. Penza, and S. Striano. 2009. Severe pulmonary congestion in a near-miss at the ἀrst seizure: Further evidence for respiratory dysfunction in sudden unexpected death in epilepsy. Epilepsy Behav 14: 701–702. Rocamora, R., M. Kurthen, L. Lickfett, J. von Oertzen, and C. E. Elger. 2003. Cardiac asystole in epilepsy: Clinical and neurophysiologic features. Epilepsia 44: 179–185. Röggla, G., and M. Röggla. 1999. Death in a hobble restraint. Can Med Assoc J 161: 21. Rugg-Gunn, F. J., R. J. Simister, M. Squirrell, D. R. Holdright, and J. S. Duncan. 2004. Cardiac arrhythmias in focal epilepsy: A prospective long-term study. Lancet 364: 2212–2219. Ryvlin, P., T. Tomson, and A. Montavont. 2009. Excess mortality and sudden unexpected death in epilepsy. Presse Med 38 (6): 905–910. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32: 1371–1382. Schuele, S. U., A. C. Bermeo, A. V. Alexopoulos, et al. 2007. Video-electrographic and clinical features in patients with ictal asystole. Neurology 69: 434–441. Scorza, F. A., R. M. Cysneiros, R. M. Arida, et al. 2008. Adult hippocampal neurogenesis and sudden unexpected death in epilepsy: Reality or just an attractive history? Med Hypotheses 71: 914–922. Scorza, F. A., A. M. Abreu, M. Albuquerque, et al. 2007. Quantiἀcation of respiratory parameters in patients with temporal lobe epilepsy. Arq Neuropsiquiatr 65: 450–453. So, E. L. 2008. What is known about the mechanisms underlying SUDEP? Epilepsia 49 (s9): 93–98. So, N. K., and M. R. Sperling. 2007. Ictal asystole and SUDEP. Neurology 69: 423–424. So, E. L., M. C. Sam, and T. L. Lagerlund. 2000. Postictal central apnea as a cause of SUDEP: Evidence from near-SUDEP incident. Epilepsia 41: 1494–1497. Sonoda, E. Y., D. B. Colugnati, and C. A. Scorza. 2008. Is cold the new hot in sudden unexpected death in epilepsy? Effect of low temperature on heart rate of rats with epilepsy. Arq Neuropsiquiatr 66: 848–852. Stöllberger, C., J. O. Huber, B. Enzelsberger, and J. Finsterer. 2009. Fatal outcome related to postictal Takotsubo-syndrome with left ventricular rupture. Eur J Neurol 16: e116–e117. Stöllberger, C., J. Finsterer, and B. Schneider. 2005. Takotsubo-like left ventricular dysfunction: Clinical presentation, instrumental ἀndings, additional cardiac and non-cardiac diseases and potential pathomechanisms. Minerva Cardioangiol 53: 139–145. Stöllberger, C., and J. Finsterer. 2004. Cardiorespiratory ἀndings in sudden unexplained/unexpected death in epilepsy (SUDEP). Epilepsy Res 59: 51–60. Strzelczyk, A., S. Bauer, S. Knake, W. H. Oertel, H. M. Hamer, and F. Rosenow. 2008. Ictal asystole in€temporal lobe epilepsy before and after pacemaker implantation. Epileptic Disord 10: 39–44.

692 Sudden Death in Epilepsy: Forensic and Clinical Issues Swallow, R. A., C. E. Hillier, and P. E. Smith. 2002. Sudden unexplained death in epilepsy (SUDEP) following previous seizure-related pulmonary oedema: Case report and review of possible preventative treatment. Seizure 11: 446–448. Tavee, J., and H. Morris III. 2008. Severe postictal laryngospasm as a potential mechanism for sudden unexpected death in epilepsy: A near-miss in an EMU. Epilepsia 49: 2113–2117. Tavernor, S. J., S. W. Brown, R. M. Tavernor, and C. Gifford. 1996. Electrocardiograph QT lengthening associated with epileptiform EEG discharges—A role in sudden unexplained death in epilepsy? Seizure 5: 79–83. Terrence, C. F., G. R. Rao, and J. A. Perper. 1981. Neurogenic pulmonary edema in unexpected, unexplained death of epileptic patients. Ann Neurol 9: 458–464. The, H. S., H. L. Tan, C. Y. Loo, and A. A. Raymond. 2007. Short QTc in epilepsy patients without cardiac symptoms. Med J Malaysia 62: 104–108. Timmings, P. L. 1998. Sudden unexpected death in epilepsy: Is carbamazepine implicated? Seizure 7: 289–291. Tinuper, P., F. Bisulli, A. Cerullo, R. Carcangiu, C. Marini, G. Pierangeli, and P. Cortelli. 2001. Ictal bradycardia in partial epileptic seizures: Autonomic investigation in three cases and literature review. Brain 124: 2361–2371. Tomson, T., L. Nashef, and P. Ryvlin. 2008. Sudden unexpected death in epilepsy: Current knowledge and future directions. Lancet Neurol 7: 1021–1031. Tomson, T., A. C. Sköld, P. Holmgen, L. Nilsson, and B. Danielsson. 1998. Postmortem changes in blood concentrations of phenytoin and carbamazepine: An experimental study. Ther Drug Monit 20: 309–312. Venkataraman, V., J. W. Wheless, L. J. Willmore, and H. Motookal. 2001. Idiopathic cardiac asystole presenting as an intractable adult onset partial seizure disorder. Seizure 10: 359–364. Vlooswijk, M. C., H. J. Majoie, M. C. De Krom, I. Y. Tan, and A. P. Aldenkamp. 2007. SUDEP in the Netherlands: A retrospective study in a tertiary referral center. Seizure 16: 153–159. Walczak, T. 2003. Do antiepileptic drugs play a role in sudden unexpected death in eplilepsy? Drug Safety 26: 673–683. Walczak, T. S., I. E. Leppik, M. D’Amelio, et al. 2001. Incidence and risk factors in sudden unexpected death in epilepsy. A prospective cohort study. Neurology 56: 519–525. Zaidi, A., L. Cotter, and A. Fitzpatrick. 1997. Sudden unexplained death (Letter). Lancet 349: 135. Zijlmans, M., D. Flanagan, and J. Gotman. 2002. Heart rate changes and ECG abnormalities during epileptic seizures: Prevalence and deἀnition of an objective clinical sign. Epilepsia 43: 847–854.

Neurocardiac Interactions in Sudden Unexpected Death in Epilepsy Can Ambulatory Electrocardiogram-Based Assessment of Autonomic Function and T-Wave Alternans Help to Evaluate Risk?

43

Richard L. Verrier Steven C. Schachter

Contents 43.1 Scope of the Problem and General Features 43.2 Importance of Neurocardiac Interactions in Triggering Life-Threatening Arrhythmias 43.3 Neurocircuitry of Cardiac Rhythm Control 43.4 Autonomic Mechanisms in Arrhythmogenesis 43.4.1 Adrenergic Influences 43.4.2 Alpha-Adrenergic Receptors 43.4.3 Sympathetic–Parasympathetic Interactions 43.4.4 Baroreflexes and Arrhythmias 43.5 Behavioral Stress and Arrhythmias 43.6 Ambulatory Electrocardiogram-Based Tools for Evaluation of Autonomic Function 43.7 Ambulatory Electrocardiogram-Based T-Wave Alternans to Assess Cardiac Electrical Instability and Risk for Sudden Cardiac Death 43.8 Clinical Evidence of Ambulatory Electrocardiogram-Based T-Wave Alternans for Prediction of Sudden Cardiac Death 43.9 Update on Vagus Nerve Stimulation in Epilepsy 43.10 Conclusions and Implications References

693 694 696 697 697 699 699 700 700 701 702 703 705 705 706

43.1â•… Scope of the Problem and General Features Sudden unexpected death in epilepsy (SUDEP) accounts for nearly 17% of premature deaths among individuals with this condition (Lathers et al. 2008). Witnessed cases indicate that death is contemporaneous with seizure activity (Langan et al. 2005; Tomson 693

694 Sudden Death in Epilepsy: Forensic and Clinical Issues

et al. 2005; Kloster and Engelskjøn 1999). Cardiac dysfunction, primarily in the form of enhanced arrhythmia susceptibility, has been implicated as a critical factor (Opherk et al. 2002; Opeskin et al. 2000; Ryvlin et al. 2006). Whereas decreases in heart rate can occur during epileptic seizures, pronounced sinus tachycardia is the most common electrocardiographic abnormality. Other relatively severe cardiac rhythm, conduction, and repolarization abnormalities have been reported in association with seizures, including bradycardia, asystole, bundle-branch block, ST-segment changes indicative of myocardial ischemia, and T-wave inversion (Opherk et al. 2002; Opeskin et al. 2000; Nei et al. 2004; Tigaran et al. 2003; Jallon et al. 2004). When these abnormalities occur, the length of the seizure is typically prolonged (Zijlmans et al. 2002). The occurrence of bradycardia and asystole in some cases (Rugg-Gunn et al. 2004; Rocamora et al. 2003; Ryvlin et al. 2006) may reflect the indirect influences of seizure-related respiratory disturbances and hypoxia (Lathers et al. 2008) with prone position during sleep a possible factor (Kloster and Engelskjøn 1999). Identiἀcation of the influences contributing to enhanced myocardial risk in SUDEP has been elusive, although a number of plausible cardiac pathologic changes have been suggested. Importantly, SUDEP is not signiἀcantly associated with any speciἀc type of seizure (Annegers et al. 1984), but enhanced adrenergic state associated with seizures constitutes a major component of arrhythmia risk. The goals of this chapter are twofold. The ἀrst is to review the basic mechanisms of autonomic control of cardiac electrical function, which could potentially contribute to precipitation of SUDEP. The second is to draw attention to new ambulatory electrocardiogram-based tools for assessing autonomic function with heart rate turbulence and cardiac arrhythmia vulnerability with T-wave alternans.

43.2â•…Importance of Neurocardiac Interactions in Triggering Life-Threatening Arrhythmias A major theme in the ἀeld of sudden cardiac death research is that whereas autonomic factors exert potent influences on the stability of heart rhythm, precipitation of lifethreatening arrhythmias requires the coexistence of a vulnerable coronary vasculature and myocardial substrate (Figure 43.1). In the general population, enhanced vulnerability of the myocardium to arrhythmic death results from the presence of relatively advanced coronary artery disease due to atherosclerosis, which is an underlying condition in ~80% of cases. The mean age of sudden cardiac death victims is 60 years (Goldberger et al. 2008). However, patients with epilepsy who die suddenly are generally younger, in the range of 20 to 40 years of age (Tomson et al. 2005), and have ischemic heart disease in only 19% of cases (Annegers et al. 1984). Thus, coronary atherosclerosis does not appear to be a major predisposing condition for SUDEP. The hyperadrenergic state associated with seizures constitutes a major component of arrhythmia risk because it is capable not only of directly triggering rhythm disturbances but also of provoking changes such as cardiac ἀbrosis (Kloster and Engelskjøn 1999), perivascular and interstitial myocardial ἀbrosis (Natelson et al. 1998) (Figure 43.2), or cardiac conduction system abnormalities (Opeskin et al. 2000), which could constitute a vulnerable myocardial substrate for arrhythmia. Indeed, the incidence of myocardial infarction is increased in patients with epilepsy over the general population (Annegers et

Neurocardiac Interactions in Sudden Unexpected Death in Epilepsy

695

Central nervous system and autonomic activity

Adrenergic and muscarinic receptors

Coronary vascular resistance

Platelet aggregability

Myocardial perfusion

Second messenger Myocardial electrical stability

Figure 43.1╇ Interaction between neural triggers and cardiovascular substrate during auto-

nomic activation. Stimulation of beta1-adrenergic receptors can decrease electrical stability directly as a result of changes in second messenger formation and alterations in ion fluxes. This deleterious influence is opposed by muscarinic receptor stimulation, which inhibits presynaptically the release of norepinephrine and opposes its action at the receptor level. Catecholamines may also alter myocardial perfusion by complex means, including alphareceptor stimulation of coronary vessels and platelets and by impairing diastolic perfusion time due to adrenergically mediated sinus tachycardia. (From Verrier, R.L., in Rosen, M.R., and Palti, Y., ed.,€Lethal Arrhythmias Resulting from Myocardial Ischemia and Infarction, Kluwer Academic Publishers, Boston, MA, 1988. With permission.)

al. 1984), and increased serum levels of cardiac troponin I, indicative of myocardial injury, have been documented after uncomplicated epileptic seizures (Hajsadeghi et al. 2009). The fact that neural influences play a major role in the stability of heart rhythm and that€a number of indicators of autonomic function reveal signiἀcant alterations provide a basis for considering the important role of neural triggering of arrhythmias during epileptic seizures. Increased cardiac sympathetic nerve activity during a seizure is indicated by surges in heart rates to greater than 150 beats/min (Opherk et al. 2002; Nei 2004) and by heart rate variability analysis (Evrengul et al. 2003; Ronkainen et al. 2006; Persson et al. 2007). The higher maximum heart rates and tachycardia associated with seizures are indicative of heightened adrenergic activity and are more frequent during sleep in SUDEP victims by comparison with other patients with epilepsy (Nei et al. 2004). This proἀle suggests a link to sleep-state related autonomic activity (Verrier and Harper 2005; Verrier and Mittleman 2005; Verrier and Josephson 2009). Although it is generally viewed that sleep is a protected state, bursts in cardiac sympathetic nerve activity have been reported in healthy individuals during rapid eye movement (REM) sleep (Somers et al. 1993) (Figure 43.3). These surges are concentrated in short, irregular periods, reach levels higher than during wakefulness, trigger intermittent increases in heart rate and blood pressure, and have been implicated in nocturnal ventricular arrhythmias and myocardial ischemia (Kales and Kales 1970). No sleep studies with electroencephalographic data have been published in patients with epilepsy, and the single report of nocturnal heart rate variability analysis results indicated no differences across the nocturnal period from healthy control subjects (Persson et al. 2007). However, detailed analyses of short-term autonomic activity especially during sleep may provide important clues regarding the risk of SUDEP.

696 Sudden Death in Epilepsy: Forensic and Clinical Issues

(a)

(b) B

(c)

(d)

Figure 43.2╇ Reversible (a, b) and irreversible (c, d) pathologic conditions found at autopsy in patients with epilepsy who died suddenly. (a) Myocytes with vacuolization; note in lower right the nucleus being displaced by vacuole (original magnification ×400). (b) Several longitudinally cut myocytes showing areas of vacuolization (original magnification ×400). (c) Diffuse interstitial fibrosis with multiple areas of myocyte replacement by connective tissue (original magnification ×40). (d) Perivascular fibrosis (original magnification ×100). All specimens were stained with hematoxylin–eosin. (From Natelson, B. H., et al., Arch Neurol, 55, 857, 1998. With€permission.)

43.3â•…Neurocircuitry of Cardiac Rhythm Control Regulation of cardiac neural activity is highly integrated and is achieved by circuitry at multiple levels (Figure 43.4). Higher brain centers operate through elaborate pathways between and within the hypothalamus and medullary cardiovascular regulatory sites. Baroreceptor mechanisms are integral to autonomic control of the cardiovascular system, as evidenced by heart rate variability and baroreceptor sensitivity testing of both cardiac patients and normal subjects. The intrinsic cardiac nerves and fat pads provide local neural coordination independent of higher brain centers (Armour 1999). Electrical remodeling of the myocardium by nerve growth and degeneration is newly recognized (Zhou et al. 2004). At the level of the myocardial cell, autonomic receptors influence G proteins to control ionic channels, pumps, and exchangers (Verrier and Antzelevitch 2004). Studies of behavioral state provide evidence that markers of arrhythmia vulnerability and autonomic

Neurocardiac Interactions in Sudden Unexpected Death in Epilepsy Awake

Stage 4 SNA

SNA

125 BP 0

125 0

REM

Stage 2 SNA 125 BP 0

697

K

Stage 3 SNA

SNA 125 BP 0

T

10 s

125 BP 0

Figure 43.3╇ Recordings of sympathetic nerve activity (SNA) and mean blood pressure (BP) in a single subject while awake and while in stages 2, 3, 4, and rapid eye movement sleep. As non– rapid eye movement sleep deepened (stages 2 through 4), sympathetic nerve activity gradually decreased and blood pressure (measured in millimeters of mercury) and variability in blood pressure were gradually reduced. Arousal stimuli elicited K complexes on the electroencephalogram (not shown), which were accompanied by increases in sympathetic nerve activity and blood pressure (indicated by the arrows, stage 2 sleep). In contrast to the changes during non– rapid eye movement sleep, heart rate, blood pressure, and blood pressure variability increased during rapid eye movement sleep, together with a profound increase in both the frequency and the amplitude of sympathetic nerve activity. There was a frequent association between rapid eye movement twitches (momentary periods of restoration of muscle tone, denoted by “T” on the tracing) and abrupt inhibition of sympathetic nerve discharge and increases in blood pressure. (From Somers, V. K., et al., N Engl J Med, 328, 303–307, 1993. With permission.)

parameters can be monitored noninvasively during emotional and physical stressors as well as sleep states to identify individuals at heightened risk of lethal cardiac arrhythmias (Kovach et al. 2001; Kop et al. 2005; Lampert et al. 2009).

43.4â•… Autonomic Mechanisms in Arrhythmogenesis 43.4.1â•… Adrenergic Influences It is well established that adrenergic inputs constitute the primary neural trigger for arrhythmias (Verrier and Antzelevitch 2004). Activation of sympathetic nerve structures, including the posterior hypothalamus or stellate ganglia, increases susceptibility to ventricular ἀbrillation. A striking surge in sympathetic nerve activity also occurs within a few minutes of induction of myocardial ischemia by left anterior descending coronary artery occlusion in experimental animals, as documented by direct nerve recording (Lombardi et al. 1983). This enhancement in sympathetic nerve activity is associated with a marked increase in susceptibility to ventricular ἀbrillation, as evidenced by a fall in ventricular

698 Sudden Death in Epilepsy: Forensic and Clinical Issues Integration Level

Cerebral cortex Hypothalamus

Central

Medulla Dorsal root ganglia

Spinal cord

Spinal

Stellate and mid-cervical ganglia

Cervical and Thoracic Ganglia

Interneuron Afferent

Sympathetic

Intrinsic Cardiac Nerves

Intrinsic cardiac ganglionated plexus Interneuron Afferent

Arterial baroreceptors

Sympathetic

Epinephrine β1

N. terminal

Parasymp.

AC Gs

M2 Gi

Acetylcholine

Myocardial Cell

AMP cAMP

Figure 43.4╇ Synthesis of new and present views on levels of integration important in neural control of cardiac electrical activity. More traditional concepts focused on afferent tracts (dashed lines) arising from myocardial nerve terminals and reflex receptors (e.g., baroreceptors) that are integrated centrally within hypothalamic and medullary cardiostimulatory and cardioinhibitory brain centers and on central modulation of sympathetic and parasympathetic outflow (solid lines) with little intermediary processing at the level of the spinal cord and within cervical and thoracic ganglia. More recent views incorporate additional levels of intricate processing within the extraspinal cervical and thoracic ganglia and within the cardiac ganglionic plexus, where recently described interneurons are envisioned to provide new levels of noncentral integration. Release of neurotransmitters from postganglionic sympathetic neurons is believed to enhance excitation in the sinoatrial node and myocardial cells through norepinephrine binding to beta1-receptors, which enhances adenyl cyclase (AC) activity through intermediary stimulatory G proteins (Gs). Increased parasympathectomy outflow enhances postganglionic release and binding of acetylcholine to muscarinic (M 2) receptors, and through coupled inhibitory G proteins (Gi), inhibits cyclic AMP production (cAMP). The latter alters electrogenesis and pacemaking activity by affecting the activity of specific membrane Na, K, and Ca channels. New levels of integration are shown superimposed on previous views and are emphasized here to highlight new possibilities for intervention. (From Lathrop, D. A. and Spooner, P. M., J Cardiovasc Electrophysiol, 12, 841–844, 2001. With permission.)

Neurocardiac Interactions in Sudden Unexpected Death in Epilepsy

699

ἀbrillation threshold and by spontaneous occurrence of the arrhythmia. Stellectomy signiἀcantly blunts this occlusion-induced surge in vulnerability to ventricular ἀbrillation (Lombardi et al. 1983; Nearing et al. 1991). Enhanced sympathetic nerve activity increases cardiac vulnerability in the normal and ischemic heart by complex processes. The major indirect effects include impairment of oxygen supply/demand ratio due to increased cardiac metabolic activity, alpha-adrenergically mediated coronary vasoconstriction, especially in vessels with damaged endothelium, and changes in preload and afterload. The direct arrhythmogenic effects on cardiac electrophysiologic function, which are primarily mediated through beta1-adrenergic receptors, include derangements in impulse formation, conduction, repolarization alternans, and heterogeneity of repolarization (Janse and Wit 1989; Nearing and Verrier 2003). Increased levels of catecholamines stimulate beta-adrenergic receptors, which in turn alter adenylÂ� ate cyclase activity and intracellular calcium flux (Opie 2004; Verrier et al. 2009). These effects are probably mediated by the cyclic nucleotide and protein kinase regulatory cascade, which can alter spatial heterogeneity of calcium transients and consequently provoke T-wave alternans and dispersion of repolarization. The effects of increased intracellular calcium, with the potential for overload and impaired intracellular calcium cycling by the sarcoplasmic reticulum, may be compounded and become especially arrhythmogenic during concurrent myocardial ischemia, which further predisposes to intracellular calcium excess (Verrier et al. 2009). Cardiac beta1-adrenergic receptor blockade is capable of negating the proἀbrillatory effect of direct sympathetic nerve stimulation by an action at the neurocardiac effector junction. However, cardiac beta2-adrenergic receptors, which control vasodilation in the vasculature and bronchi (Brodde et al. 1991), do not appear to play a signiἀcant role in modulating ventricular excitable properties. 43.4.2â•… Alpha-Adrenergic Receptors In the normal heart, alpha-adrenergic receptor stimulation or blockade does not appear to affect ventricular electrical stability, as evidenced by the fact that administration of an alpha-adrenergic agonist such as phenylephrine or methoxamine does not influence excitable properties when the pressor response is controlled to prevent reflex changes in autonomic tone (Kowey et al. 1983). In the setting of myocardial ischemia, alpha-adrenergic blockade may alleviate coronary vasoconstriction and reduce platelet aggregability. Thus, alpha-adrenergic receptor activity exerts direct actions not only on myocardial excitable properties but also on platelet aggregability and coronary hemodynamic function. 43.4.3â•… Sympathetic–Parasympathetic Interactions Vagus nerve influences are contingent on the prevailing level of adrenergic tone. When sympathetic tone to the heart is augmented by thoracotomy, sympathetic nerve stimulation, myocardial ischemia, or catecholamine infusion, vagal activation exerts a protective effect on ventricular vulnerability. Vagus nerve stimulation is without effect on ventricular vulnerability when adrenergic input to the heart is ablated by beta-adrenergic blockade, a phenomenon termed accentuated antagonism (Verrier and Antzelevitch 2004). The basis for this antagonism of adrenergic effects is presynaptic inhibition of norepinephrine release from nerve endings and a muscarinically mediated action at the second messenger level,

700 Sudden Death in Epilepsy: Forensic and Clinical Issues

which attenuates the response to catecholamines at receptor sites. In addition, importantly, vagal influences provide indirect protection against ventricular ἀbrillation by reducing excess heart rates, which can otherwise increase ischemic insult by critically compromising diastolic perfusion time during acute myocardial ischemia. However, beneἀcial effects of vagus nerve activity may be annulled if profound bradycardia and hypotension ensue. Vagus nerve stimulation has been shown in experimental studies to protect against ventricular arrhythmias during myocardial ischemia. Myocardial infarction may damage nerve pathways by sustained reduction in supply of oxygen and nutrients to the nerve, thereby limiting the potential of the vagus nerve to be activated. Seizure activity causes local vasospasm and accumulation of microthrombi and could also thereby lead to impaired blood flow and damage to the nerve supply. Vanoli and colleagues (1991) demonstrated the antiἀbrillatory effect of vagus nerve stimulation during exercise-induced ischemia in canines with a healed myocardial infarction. Direct stimulation of the right cervical vagus through a chronically implanted electrode at 15 s after onset of exercise-induced acute myocardial ischemia reduced the incidence of ventricular ἀbrillation by 92%. This effect was only partly due to the attendant heart rate reduction, as in half of the animals, the antiarrhythmic efficacy of vagus nerve stimulation persisted despite maintenance of constant heart rate by atrial pacing. Vagus nerve stimulation in humans to elicit beneἀcial cardiac effects has recently been explored by Schwartz et al. (2008), who demonstrated that this procedure is safe and well tolerated in patients with heart failure, as indicated by some improvement in heart failure class and a decrease in left ventricular end-systolic volume. The exciting possibility that chronically implanted vagus nerve stimulation devices may suppress arrhythmias clinically remains to be explored. The potential for this form of therapy is underscored by the extensive experimental evidence and the clinical demonstration that vagomimetic maneuvers such as carotid sinus massage and administration of agents such as phenylephrine or edrophonium can terminate ventricular tachyarrhythmias (Waxman et al. 1994). 43.4.4â•… Baroreflexes and Arrhythmias The classic studies by Billman et al. (1982) drew attention to the importance of baroreceptor function in susceptibility to life-threatening arrhythmias associated with myocardial isÂ�Â� chemia and infarction. In their initial investigations in canines, they demonstrated that the more intense the baroreflex response was, the less vulnerable animals were to ventricular ἀbrillation during myocardial ischemia superimposed on prior myocardial infarction. The protective effect of the baroreceptor mechanism has been linked primarily to the antiἀbrillatory influence of vagus nerve activity. The latter effect improves diastolic coronary perfusion, minimizing the ischemic insult from coronary artery occlusion. The importance of baroreceptor sensitivity was subsequently documented in human subjects in whom baroreceptor function was evaluated with the pressor agent phenylephrine. La Rovere et al. (1998) demonstrated that following myocardial infarction, patients were less likely to experience sudden cardiac death if their baroreceptor function was preserved.

43.5â•… Behavioral Stress and Arrhythmias Because epilepsy is associated with intense emotions such as anxiety and depression (Lathers and Schraeder 2006), it is relevant to consider experimental evidence of the impact

Neurocardiac Interactions in Sudden Unexpected Death in Epilepsy

701

of behavioral state on cardiac electrical stability. In early studies, we developed aversive behavioral conditioning paradigms and models eliciting natural emotions, notably anger and fear. Aversive conditioning of dogs in a Pavlovian sling with mild chest shock on 3 consecutive days resulted in a reduction in the repetitive extrasystole threshold greater than 30% during subsequent exposure to the environment without shock. The same paradigm elicited a threefold increase in the occurrence of spontaneous ventricular ἀbrillation when coronary artery occlusion was carried out in the aversive sling compared to the nonaversive cage environment (Lown et al. 1973). In dogs recovering from myocardial infarction, exposure to the aversive environment consistently elicited ventricular tachycardia for several days during the healing process. After this time, the animals continued to exhibit signs of behavioral stress in the aversive environment but no longer experienced ventricular arrhythmias, indicating that the arousal state required a substrate of cardiac electrical instability for the induction of rhythm disturbances. The stress-induced changes in cardiac excitable properties were largely obtunded by beta-adrenergic receptor blockade with propranolol or metoprolol. In a separate series of experiments, an experimental canine model was developed to emulate anger (Verrier et al. 1987; Kovach et al. 2001), which is the emotion most commonly associated with myocardial infarction and sudden death (Mittleman et al. 1995; Verrier and Mittleman 1996). A standardized food-access-denial paradigm provoked intense arousal and pronounced myocardial ischemia in territories of stenosed coronary vessels. The angerlike state also elicited severe repolarization abnormalities conducive to life-Â�threatening cardiac€arrhythmias.

43.6â•…Ambulatory Electrocardiogram-Based Tools for Evaluation of Autonomic Function Several studies have indicated that epileptic seizures are associated with surges in heart rate (Opherk et al. 2002; Nei et al. 2004) and reduced heart rate variability (Ronkainen et al. 2006; Persson et al. 2007; Evrengul et al. 2005), indicative of a hyperadrenergic state. This ἀnding is important because depressed heart rate variability is associated with heightened risk for sudden cardiac death in patients with ischemic heart disease (Lombardi 2002; Tulppo et al. 1996). Because depressed baroreceptor sensitivity is associated with heightened risk for arrhythmias, it will be crucial to establish whether baroreceptor reflexes might be altered by recurring epileptic seizures. Baroreceptor sensitivity can be noninvasively measured from ambulatory electrocardiograms by monitoring heart rate turbulence, a tool that has not been explored in patients with epilepsy. Heart rate turbulence refers to fluctuations of sinus-rhythm cycle length after a single ventricular premature beat. These effects are a direct function of baroreceptor responsiveness because reflex activation of the vagus nerve controls the pattern of sinus rhythm. Several studies conἀrm that in low-risk patients with cardiovascular disease, after a ventricular premature beat, sinus rhythm exhibits a characteristic pattern of early acceleration and subsequent deceleration. By contrast, patients at high risk for cardiovascular events exhibit essentially a flat, nonvarying response to the ventricular premature beat, indicating inability to activate vagus nerves and their cardioprotective effect. The method is an independent predictor of total mortality in patients with ischemic heart disease or heart failure (Schmidt et al. 1999; Bauer et al. 2008). Heart

702 Sudden Death in Epilepsy: Forensic and Clinical Issues

rate deceleration capacity, a related and even more comprehensive marker of autonomic control than heart rate turbulence, may be of considerable clinical value in assessing overall autonomic regulation of the heart in patients with diverse types of cardiovascular disease (Bauer et al. 2006) and could also be used to evaluate patients with epilepsy.

43.7â•…Ambulatory Electrocardiogram-Based T-Wave Alternans to Assess Cardiac Electrical Instability and Risk for Sudden Cardiac Death Extensive scientiἀc evidence spanning more than a decade points to a fundamental link between T-wave alternans, a beat-to-beat fluctuation in the morphology of the T wave of the electrocardiogram, and susceptibility to malignant ventricular tachyarrhythmias (VerÂ� rier et al. 2009). The basic principle is that T-wave alternans is an indicator of temporal– spatial heterogeneity of repolarization, a precondition for reentrant arrhythmias. T-wave alternans also detects derangements in intracellular calcium cycling, which is subject to the influences of both the autonomic nervous system and alterations in myocardial substrate, particularly ischemia and scar associated with infarction. Similar mechanisms may operate in association with cardiac ἀbrosis, contraction band necrosis, or other types of myocardial injury that are known to occur in patients with epilepsy. Diverse physiologic interventions have been shown to alter T-wave alternans magnitude in parallel with their influence on vulnerability to ventricular tachyarrhythmias. Speciἀcally, these include elevations in heart rate, coronary artery occlusion and reperfusion, and sympathetic nerve stimulation (Nearing et al. 1991). In the chronically instrumented canine model developed to emulate anger (Verrier et al. 1987; Kovach et al. 2001), Alternans (mV ms) Baseline

0.04

Occlusion

0.47

Anger

2.88

Anger and occlusion Ventricular fibrillation Lead: V4 No drug

3.97 1 mV 1s

Figure 43.5╇ Flow chart of the major components of the modified moving average method of

T-wave alternans analysis. The left ventricular electrocardiogram was obtained from a representative experiment in which coronary artery occlusion subsequently resulted in ventricular fibrillation. Electrocardiograms are filtered to reduce high-frequency noise and to remove baseline wander. Ventricular and supraventricular premature beats as well as beats with a high noise level are removed. The even beats in the sequence are then assigned to group A and the odd beats to group B. Modified moving average computed beats of types A and B are computed continuously. The nth computed beat is developed from the n − 1th electrocardiographic beat and the n − 1th computed beat. The amplitude of the effect of any one beat on the computed beat is also limited by the bounds on ΔA and ΔB. The alternans estimate is determined as the maximum absolute difference between A and B computed beats within the ST-segment and T-wave region. The output period can be adjusted as desired. (From Nearing, B. D. and Verrier,€R.€L., J€Appl Physiol, 92, 541–549, 2002. With permission.)

Neurocardiac Interactions in Sudden Unexpected Death in Epilepsy

703

a pronounced increase in T-wave alternans was observed during behavioral arousal. A 3-min period of coronary artery occlusion potentiated arrhythmia risk, more than doubling the magnitude of anger-induced T-wave alternans (Figure 43.5). The stress-related effects were signiἀcantly lessened by metoprolol, further implicating a major role of beta1adrenergic receptors in sympathetic nerve induction of cardiac vulnerability and T-wave alternans (Kovach et al. 2001; Lathers et al. 2008). Vagus nerve stimulation, blockade of beta-adrenergic receptors, and sympathetic denervation, which reduce susceptibility to ventricular tachyarrhythmias, have been shown to decrease T-wave alternans magnitude (Verrier et al. 2009). These series of observations underscore the fundamental link between T-wave alternans and vulnerability to lethal arrhythmias, which underlies the utility of this parameter in assessing propensity for life-threatening ventricular arrhythmias.

43.8â•…Clinical Evidence of Ambulatory Electrocardiogram-Based T-Wave Alternans for Prediction of Sudden Cardiac Death A few years ago, we developed a time-domain method, termed Modified Moving Average analysis was developed to quantify T-wave alternans during both routine exercise stress testing and ambulatory electrocardiographic monitoring (Nearing and Verrier 2002). The technique is based on the powerful noise-rejection principle of recursive averaging (Figure

A B A B A B Noise reduction Baseline wander removal and separation of beats

A

A

B

A

B

B

Median beat B

Median beat A

Alternans measurement

Figure 43.6╇ Segments of electrocardiogram tracings with measurements of maximum T-wave alternans magnitude at baseline, during myocardial ischemia, during angerlike state, and with simultaneous angerlike state and myocardial ischemia in one dog that experienced ventricular fibrillation at 42 s after provocation of angerlike response was superimposed at 1 min of coronary artery occlusion. (From Kovach, J. A., et al., J Am Coll Cardiol, 37, 1719–1725, 2001. With permission.)

704 Sudden Death in Epilepsy: Forensic and Clinical Issues

43.6), and respiration and motion artifacts have been further reduced by cubic alignment and other ἀlters. Modiἀed moving average analysis computes T-wave alternans magnitude as the peak difference between A and B beats in an ABAB beat stream at any point within the JT interval of the electrocardiogram. The predictive capacity of this technique has been examined in consecutive patients referred for routine exercise stress testing enrolled in the Finnish Cardiovascular Study (FINCAVAS) (Nieminen et al. 2007; Minkkinen et al. 2009). Nieminen et al. (2007) found in multivariate analysis in this low-risk population that the relative risk of T-wave alterÂ�nans ≥65 µV for sudden cardiac death was 7.4 (95% CI, 2.8–19.4; P < 0.001), for cardiovascular mortality was 6.0 (95% CI, 2.8–12.8; P < 0.001), and for all-cause mortality was 3.3 (95% CI, 1.8–6.3; P = 0.001), indicating that T-wave alternans exhibits speciἀcity for sudden cardiac death. The modiἀed moving average method also proved suitable for analyzing T-wave alternans from ambulatory electrocardiographic monitoring records to investigate the effects of physical activity (Verrier et al. 2003), circadian factors (Verrier et al. 2003; Zanobetti et al. 2009), mental stress (Kop et al. 2004; Lampert et al. 2009), and sleep states in arrhythmogenesis in patients with diverse conditions including myocardial ischemia, infarction, and heart failure (Verrier et al. 2009). Recently, in a prospective trial, Sakaki et al. (2009) demonstrated the remarkable predictive capacity of T-wave alternans in ambulatory electrocardiograms in both ischeÂ�mic and nonischemic patients with left ventricular dysfunction (Figures 43.7 and 43.8). As mental stress has been implicated in seizure-induced risk, it is important to recognize that T-wave alternans has been shown to detect cardiac electrical instability induced by mental arithmetic and anger recall in patients with implantable cardioverter deἀbrillators. V5 (CM5) Max TWA 67 μV

V1 (NASA) Max TWA 98 μV

Figure 43.7╇ An example of a positive Modified Moving Average T-wave alternans test. Superimposed complexes from precordial leads V5 (CM5) and V1 (NASA) illustrate peak TWA. In this patient, TWA was determined to be positive because the peak TWA voltage was 98 μV in lead V1. (From Sakaki, K., et al., Heart Rhythm, 6, 332–337, 2009. With permission.)

Neurocardiac Interactions in Sudden Unexpected Death in Epilepsy

705

100

Survival rate (%)

80 60

Hazard ratio = 15.9 95% Cl 6.7–37.8, P < 0.001

40

TWA negative

20

TWA positive

0 0

50

100

150

200

250

300

350

350

Time (days)

Figure 43.8╇ Event-free curves for cardiac mortality using peak voltage of Modified Moving Average T-wave alternans from 24-h ambulatory electrocardiogram recordings in either lead V1 or V5. (From Sakaki, K., et al., Heart Rhythm, 6, 332–337, 2009. With permission.)

Kop et al. (2004) and Lampert et al. (2009) reported that during these routine challenges, patients with implantable cardioverter deἀbrillators exhibited signiἀcant increases in T-wave alternans level compared to age-matched healthy control subjects. Furthermore, Lampert et al. (2009) determined that the rise in T-wave alternans level appears to coincide with enhanced risk for life-threatening arrhythmias. Testing patients with epilepsy with simple mental tasks such as mental arithmetic could be used to disclose latent cardiac electrical instability attributable to heightened adrenergic activity, behavioral state, or substrate changes, as manifested by elevated levels of T-wave alternans.

43.9â•…Update on Vagus Nerve Stimulation in Epilepsy Vagus nerve stimulation has been approved by the FDA for the adjunctive treatment of medically refractory partial seizures in patients older than 12 years since 1997 (Schachter 2002). Side effects related to stimulation are usually mild; almost always resolve with adjustment in the stimulation settings; and include hoarseness, throat pain, coughing, shortness of breath, and tingling. Although there are mixed reports of measurable autonomic effects of vagus nerve stimulation when used for epilepsy (Banzett et al. 1999; Binks et al. 2001; Schachter 2008), SUDEP rates halved after more than 2 years of treatment (Annegers et al. 2000). Therefore, the possibility exists but remains to be proven that vagus nerve stimulation may reduce cardiac arrhythmias and risk for sudden death, potentially mediated by the efferent cardiac vagus nerve (Vanoli et al. 1991; Verrier and Antzelevitch 2004). This potential beneἀt in patients with epilepsy is intriguing and unexplored but, given the scientiἀc rationale, deserves pursuit.

43.10â•… Conclusions and Implications Extensive evidence implicates the interaction between autonomic factors and a vulnerable myocardial substrate in predisposing patients to life-threatening cardiac arrhythmias. Adrenergic influences constitute the primary trigger, and vagus nerve stimulation can

706 Sudden Death in Epilepsy: Forensic and Clinical Issues

antagonize the arrhythmogenic effect of catecholamines to exert a cardioprotective action. Recently, new ambulatory electrocardiographic tools have been developed to assess autonomic function and vulnerability to cardiac arrhythmias, namely, heart rate turbulence and T-wave alternans, respectively. Although these methodologies have been extensively tested in populations with ischemic and nonischemic heart disease, their use in patients with epilepsy should be explored. Because it is likely that arrhythmic events result from an interaction between autonomic triggers and a vulnerable myocardial substrate, there is a great need for systematic longitudinal studies of electrocardiogram-based pathophysiologic markers including ST-segment changes, QRS interval duration, and R-wave changes and T-wave alternans in patients with epilepsy to provide insights into the factors leading to SUDEP. Increasing use of implantable loop recorders in patients with epilepsy promises to expand the opportunity to understand disturbances in autonomic function and cardiac rhythm associated with ictal events (Zaidi et al. 2000; Rugg-Gunn et al. 2004; Ronkainen et al. 2006). Thus, in the future, these parameters could not only play a role in advancing our understanding of the underlying cardiac pathologies responsible for SUDEP but also enable evaluation of cardiac antiarrhythmic as well as proarrhythmic effects of medical antiepileptic therapy and provide the basis for considering patients with epilepsy as potential candidates for antiarrhythmic therapies. Especially intriguing is the possibility that vagus nerve stimulation (Schachter 2009) could not only reduce seizure activity but may also protect the heart (Schwartz et al. 2008), as activation of this neural pathway may be antiarrhythmic. Ambulatory electrocardiogram-based T-wave alternans can detect vagus nerve influences on cardiac electrical instability, providing a tool for quantiἀcation of the potential cardioprotective effect of this potent intervention.

References Annegers, J. F., W. A. Hauser, and S. B. Shirts. 1984. Heart disease mortality and morbidity in patients with epilepsy. Epilepsia 25: 699–704. Annegers, J. F., S. P. Coan, W. A. Hauser, and J. Leestma. 2000. Epilepsy, vagal nerve stimulation by the NCP system, all-cause mortality, and sudden, unexpected, unexplained death. Epilepsia 41: 549–553. Armour, J. A. 1999. Myocardial ischaemia and the cardiac nervous system. Cardiovasc Res 41: 41–54. Banzett, R. B., A. Guz, D. Paydarfar et al. 1999. Cardiorespiratory variables and sensation during stimulation of the left vagus in patients with epilepsy. Epilepsy Res 35: 1–11. Bauer, A., J. W. Kantelhardt, P. Barthel et al. 2006. Deceleration capacity of heart rate as a predictor of mortality after myocardial infarction: Cohort study. Lancet 367: 1674–1681. Bauer, A., M. Malik, G. Schmidt et al. 2008. Heart rate turbulence: Standards of measurement, physiological interpretation, and clinical use: International Society for Holter and Noninvasive Electrocardiology Consensus. J Am Coll Cardiol 52: 1353–1365. Billman, G. E., P. J. Schwartz, and H. L. Stone. 1982. Baroreceptor reflex control of heart rate: A predictor of sudden cardiac death. Circulation 66: 874–880. Binks, A. P., D. Paydarfar, S. C. Schachter, A. Guz, and R. B. Banzett. 2001. High strength stimulation of the vagus nerve in awake humans: A lack of cardiorespiratory effects. Respirat Physiol 127: 125–133. Brodde, O. E. 1991. Beta1- and beta2-adrenoceptors in the human heart: Properties, function and alterations in chronic heart failure. Pharmacol Rev 43: 203–242.

Neurocardiac Interactions in Sudden Unexpected Death in Epilepsy

707

Evrengul, H., H. Tanriverdi, D. Dursunoglu et al. 2005. Time and frequency domain analyses of heart rate variability in patients with epilepsy. Epilepsy Res 63: 131–139. Goldberger, J. J., M. E. Cain, S. H. Hohnloser et al. 2008. American Heart Association/American College of Cardiology Foundation/Heart Rhythm Society Scientiἀc Statement on Noninvasive Risk Stratiἀcation Techniques for Identifying Patients at Risk for Sudden Cardiac Death. A scientiἀc statement from the American Heart Association Council on Clinical Cardiology Committee on Electrocardiography and Arrhythmias and Council on Epidemiology and Prevention. Circulation 118: 1497–1518. Hajsadeghi, S., S. Afsharian, S. M. Fereshtehnejad, M. R. Keramati, and R. Mollahoseini. 2009. Serum levels of cardiac troponin I in patients with uncomplicated epileptic seizure. Arch Med Res 40: 24–28. Jallon, P. 2004. Mortality in patients with epilepsy. Curr Opin Neurol 17: 141–146. Janse, M. J., and A. L. Wit. 1989. Electrophysiological mechanisms of ventricular arrhythmias resulting from myocardial ischemia and infarction. Physiol Rev 69: 1049–1169. Kales, A., and J. D. Kales. 1970. Evaluation, diagnosis, and treatment of clinical conditions related to sleep. JAMA 213: 2229–2232. Kloster, R., and T. Engelskjøn. 1999. Sudden unexpected death in epilepsy (SUDEP): A clinical perspective and a search for risk factors. J Neurol Neurosurg Psychiatry 67: 439–444. Kop, W. J., D. S. Krantz, B. D. Nearing et al. 2004. Effects of acute mental stress and exercise on T-wave alternans in patients with implantable cardioverter deἀbrillators and controls. Circulation 109: 1864–1869. Kovach, J. A., B. D. Nearing, and R. L. Verrier. 2001. An anger-like behavioral state potentiates myocardial ischemia-induced T-wave alternans in canines. J Am Coll Cardiol 37: 1719–1725. Kowey, P. R., R. L. Verrier, and B. Lown. 1983. Effect of alpha-adrenergic receptor stimulation on ventricular electrical properties in the normal canine heart. Am Heart J 105: 366–371. La Rovere, M. T., J. T. Bigger Jr., F. I. Marcus et al. 1998. Baroreflex sensitivity and heart-rate variability in prediction of total cardiac mortality after myocardial infarction. ATRAMI (Autonomic Tone and Reflexes After Myocardial Infarction) Investigators. Lancet 351: 478–484. Lampert, R., V. Shusterman, M. M. Burg et al. 2009. Anger-induced T-wave alternans predicts future ventricular arrhythmias in patients with implantable cardioverter-deἀbrillators. J Am Coll Cardiol 53: 774–778. Langan, Y., L. Nashef, and J. W. Sander. 2005. Case-control study of SUDEP. Neurology 64: 1131–1133. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9: 236–242. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008. The mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12: 3–24. Lathrop, D. A., and P. M. Spooner. 2001. On the neural connection. J Cardiovasc Electrophysiol 12: 841–844. Lombardi, F., R. L. Verrier, and B. Lown. 1983. Relationship between sympathetic neural activity, coronary dynamics, and vulnerability to ventricular ἀbrillation during myocardial ischemia and reperfusion. Am Heart J 105: 958–965. Lombardi, F. 2002. Clinical implications of present physiological understanding of HRV components. Card Electrophysiol Rev 6: 245–249. Lown, B., R. L. Verrier, and R. Corbalan. 1973. Psychologic stress and threshold for repetitive ventricular response. Science 182: 834–836. Minkkinen, M., M. Kähönen, J. Viik et al. 2009. Enhanced predictive power of quantitative TWA during routine exercise testing in the Finnish Cardiovascular Study. J Cardiovasc Electrophysiol 20: 408–415. Mittleman, M. A., M. Maclure, J. B. Sherwood et al. 1995. Triggering of acute myocardial infarction onset by episodes of anger. Circulation 92: 1720–1725. Natelson, B. H., R. V. Suarez, C. F. Terrence, and R. Turizo. 1998. Patients with epilepsy who die suddenly have cardiac disease. Arch Neurol 55: 857.

708 Sudden Death in Epilepsy: Forensic and Clinical Issues Nearing, B. D., A. H. Huang, and R. L. Verrier. 1991. Dynamic tracking of cardiac vulnerability by complex demodulation of the T-wave. Science 252: 437–440. Nearing, B. D., and R. L. Verrier. 2002. Modiἀed moving average method for T-wave alternans analysis with high accuracy to predict ventricular ἀbrillation. J Appl Physiol 92: 541–549. Nearing, B. D., and R. L. Verrier. 2003. Tracking heightened cardiac electrical instability by computing interlead heterogeneity of T-wave morphology. J Appl Physiol 95: 2265–2272. Nei, M., R. T. Ho, and B. W. Abou-Khalil et al. 2004. EEG and ECG in sudden unexplained death in epilepsy. Epilepsia 45: 338–345. Nieminen, T., T. Lehtimäki, J. Viik et al. 2007. T-wave alternans predicts mortality in a population undergoing a clinically indicated exercise test. Eur Heart J 28: 2332–2337. Opeskin, K., A. Thomas, and S. F. Berkovic. 2000. Does cardiac conduction pathology contribute to sudden unexpected death in epilepsy? Epilepsy Res 40: 17–24. Opherk, C., J. Coromilas, and L. J. Hirsch. 2002. Heart rate and EKG changes in 102 seizures: Analysis of influencing factors. Epilepsy Res 52: 117–127. Opie, L. H. 2004. Heart Physiology: From Cell to Circulation, 4th ed. Philadelphia, PA: Lippincott, Williams, and Wilkins. Persson, H., E. Kumlien, M. Ericson, and T. Tomson. 2007. Circadian variation in heart-rate variability in localization-related epilepsy. Epilepsia 48: 917–922. Rocamora, R., M. Kurthen, L. Lickfett, J. von Oertzen, and C. E. Elger. 2003. Cardiac asystole in epilepsy: Clinical and neurophysiologic features. Epilepsia 44: 179–185. Ronkainen, E., J. T. Korpelainen, E. Heikkinen et al. 2006. Cardiac autonomic control in patients with refractory epilepsy before and during vagus nerve stimulation treatment: A one-year follow-up study. Epilepsia 47: 556–562. Rugg-Gunn, F. J., R. J. Simister, M. Squirrell, D. R. Holdright, and J. S. Duncan. 2004. Cardiac arrhythmias in focal epilepsy: A prospective long-term study. Lancet 364 (9452): 2212–2219. Ryvlin, P., A. Montavont, and P. Kahane. 2006. Sudden unexpected death in epilepsy: From mechanisms to prevention. Curr Opin Neurol 19: 194–199. Sakaki, K., T. Ikeda, Y. Miwa et al. 2009. Time-domain T-wave alternans measured from Holter electrocardiograms predicts cardiac mortality in patients with left ventricular dysfunction: A prospective study. Heart Rhythm 6: 332–337. Schachter, S. C. 2002. Vagus nerve stimulation: Where are we? Curr Opin Neurol 15: 201–206. Schachter, S. C. 2008. Review of “Effects of vagus nerve stimulation on cardiovascular regulation in patients with epilepsy.” J Watch Neurol 10: 52. Schachter, S. C. 2009. Vagal nerve stimulation. In The Treatment of Epilepsy, 3rd ed., ed. S. Shorvon, E. Perucca, and J. Engle, 1017–1023. London: Blackwell Publishing. Schmidt, G., M. Malik, P. Barthel et al. 1999. Heart-rate turbulence after ventricular premature beats as a predictor of mortality after acute myocardial infarction. Lancet 353: 1390–1396. Schwartz, P. J., G. De Ferrari, A. Sanzo et al. 2008. Long term vagal stimulation in patients with advanced heart failure: First experience in man. Eur J Heart Failure 10: 884–891. Somers, V. K., M. E. Dyken, A. L. Mark et al. 1993. Sympathetic nerve activity during sleep in normal subjects. N Engl J Med 328: 303–307. Tigaran, S., H. Molgaard, R. McClelland, M. Dam, and A. S. Jaffe. 2003. Evidence of cardiac ischemia during seizures in drug refractory epilepsy patients. Neurology 60: 492–495. Tomson, T., T. Walczak, M. Sillanpaa, and J. Sander. 2005. Sudden unexpected death in epilepsy: A review of incidence and risk factors. Epilepsia 46 (Suppl 11): 54–61. Tulppo, M. P., T. H. Makikallio, T. E. S. Takala, T. Seppanen, and H. V. Huikuri. 1996. Quantitative beat-to-beat analysis of heart rate dynamics during exercise. Am J Physiol 271: H244–H252. Vanoli, E., G. M. De Ferrari, M. Strambaâ•‚Badiale et al. 1991. Vagal stimulation and prevention of sudden death in conscious dogs with a healed myocardial infarction. Circ Res 68: 1471–1481. Verrier, R. L., E. L. Hagestad, and B. Lown. 1987. Delayed myocardial ischemia induced by anger. Circulation 75: 249–254.

Neurocardiac Interactions in Sudden Unexpected Death in Epilepsy

709

Verrier, R. L. 1988. Central nervous system modulation of cardiac rhythm. In Lethal Arrhythmias Resulting from Myocardial Ischemia and Infarction, ed. M. R. Rosen and Y. Palti, 149–164. Boston, MA: Kluwer Academic Publishers. Verrier, R. L., and M. A. Mittleman. 1996. Life-threatening cardiovascular consequences of anger in patients with coronary heart disease. Cardiol Clin 14: 289–307. Verrier, R. L., B. D. Nearing, M. T. LaRovere et al. 2003. Ambulatory ECG-based tracking of T-wave alternans in post-myocardial infarction patients to assess risk of cardiac arrest or arrhythmic death. J Cardiovasc Electrophysiol 14: 705–711. Verrier, R. L., and C. A. Antzelevitch. 2004. Autonomic aspects of arrhythmogenesis: The enduring and the new. Curr Opin Cardiol 19: 2–11. Verrier, R. L., R. M. Harper, and J. A. Hobson. 2005. Central and autonomic mechanisms regulating cardiovascular function. In Principles and Practice of Sleep Medicine, 4th ed., eds. M. H. Kryger, T. Roth, and W. C. Dement, 192–202. Philadelphia, PA: WB Saunders. Verrier, R. L., and M. A. Mittleman. 2005. Sleep-related cardiac risk. In Principles and Practice of Sleep Medicine, 4th ed., ed. M. H. Kryger, T. Roth, and W. C. Dement, 1161–1170. Philadelphia, PA: WB Saunders. Verrier, R. L., and M. E. Josephson. 2009. Impact of sleep on arrhythmogenesis. Circ Arrhythmia Electrophysiol 2: 450–459. Verrier, R. L., K. Kumar, and B. D. Nearing. 2009. Basis for sudden cardiac death prediction by T-wave alternans from an integrative physiology perspective. Heart Rhythm 6 (3): 416–422. Waxman, M. B., D. Cameron, and R. W. Wald. 1994. Vagal activity and ventricular tachyarrhythmias. In Vagal Control of the Heart, eds. M. N. Levy, and P. J. Schwartz, 579–612. Mt. Kisco, NY: Futura. Zaidi, A., P. Clough, P. Cooper, B. Scheepers, and A. P. Fitzpatrick. 2000. Misdiagnosis of epilepsy: Many seizure-like attacks have a cardiovascular cause. J Am Coll Cardiol 36: 181–184. Zanobetti, A., P. H. Stone, F. E. Speizer et al. 2009. T-wave alternans, air pollution and traffic in highrisk subjects. Am J Cardiol 104 (5): 665–670. Zhou, S., L. S. Chen, Y. Miyauchi et al. 2004. Mechanisms of cardiac nerve sprouting after myocardial infarction in dogs. Circ Res 95: 76–83. Zijlmans, M., D. Flanagan, and J. Gotman. 2002. Heart rate changes and ECG abnormalities during epileptic seizures: prevalence and deἀnition of an objective clinical sign. Epilepsia 43: 847–854.

Arrhythmogenic, Respiratory, and Psychological Risk Factors for Sudden Unexpected Death and Epilepsy Case Histories

44

Claire M. Lathers

Contents 44.1 Introduction 44.2 Case Histories of SUDEP or “Near Miss” SUDEP 44.3 Discussion 44.3.1 Analysis of the Electrocardiogram Is Essential for All Patients Who Present with Seizures References

711 712 714 718 718

44.1â•…Introduction The classiἀcation of deaths in the context of SUDEP by coroners and medical examiners in the United States is underreported as per recent surveys conducted by Schraeder et al. (2006, 2009). In practice, the term SUDEP is not routinely used on death certiἀcate diagnoses. Thus, the incidence for SUDEP may actually be higher than reported. In addition, SUDEP is most often an unwitnessed event (Langan et al. 2000). These facts make the prevention of SUDEP more difficult. Physicians and caregivers responsible for treating persons with epilepsy must know the risk factors for SUDEP and evaluate these potential risks for each and every patient. Lathers et al. (2008a, 2008b) and Scorza et al. (2008) have recently summarized the proposed mechanistic factors in SUDEP with the three major risk categories of arrhythmogenic factors, respiratory factors and hypoxia, and psychological factors. The SUDEP case histories discussed by Langan et al. (2000) and the two cases of “near miss” sudden death in persons with epilepsy published by Tavee and Morris (2008) and by So et al. (2000) are presented below. The discussion emphasizes the mystery of sudden death and the difficulty of identifying given risk factors and associated mechanisms for risks (Lathers et al. 2008a, 2008b; Scorza et al. 2008). Lack of complete understanding of the risk factors and mechanisms of SUDEP make prevention of SUDEP more difficult. These case histories introduce the reader to problems associated with treating persons with epilepsy and problems associated with the objective of preventing the occurrence of SUDEP.

711

712 Sudden Death in Epilepsy: Forensic and Clinical Issues

44.2â•… Case Histories of SUDEP or “Near Miss” SUDEP A Series o f Wit nessed SUDEP Deat hs Fifteen witnessed deaths were included in a total group of 125 cases of SUDEP identiἀed by coroners, neurologists, and families of victims (Langan et al. 2000). Twelve of the deaths occurred in association with convulsive seizures. One victim exhibited a generalized seizure and then collapsed about 5 min later. Of the last two witnessed deaths, one victim died during what was thought to be a “probable postictal state” and one died after experiencing an aura. The witnesses reported that 12 of the 15 cases experienced respiratory difficulty. Co mment s by L ang an et al. (2000) Most of the sudden deaths in persons with epilepsy were not witnessed, that is, only 15 of 125 SUDEP cases. Most of the deaths that were witnessed occurred in conjunction with a seizure. Respiratory difficulty/compromise was reported to be the prominent problem observed. The authors suggest that repositioning of the patient and/or stimulation of respiration may be important to help prevent sudden deaths in persons with epilepsy. Co mment s by L at hers The possible mechanisms of SUDEP are several (see discussion and Tables 1.1, 1.2, and 1.4 in Chapter 1) and include central and obstructive apnea and cardiovascular, including cardiac arrhythmias. Caution must be exerted when concluding respiratory changes alone are the primary mechanism of death. At the ἀrst onset of the clinical problem, cardiac arrhythmias may be felt by the patient but may not be visually detected by a witness. “Invisible” cardiac arrhythmias may be initiated and then followed by “visible” respiratory distress. Therefore, in addition to repositioning the patient to ensure ease of respiration and/or stimulation of respiration, it is important, if possible, also to simultaneously monitor and medically support cardiac rate and rhythm. The answer to the question of “Which is the cart and which is the horse?” or “Which is the egg and which is the chicken?” will vary from patient to patient depending on the type of initiating clinical event. However, if an event is witnessed, it is of the upmost importance to support all vital systems, that is, respiratory, cardiac, and circulatory. If intravenous access is difficult to obtain in an adult, and even more so in a pediatric patient during a witnessed event, it is possible that the intraosseous route of drug administration would allow rapid access to the circulation to provide anticonvulsant drugs and/or drugs used to treat cardiac arrest (Jim et al. 1988, 1989; Lathers et al. 1989a, 1989b, 1989c; Schoffstall et al. 1989; Spivey et al. 1987a, 1987b). Cases summarized from Langan et al. (2000).

Arrhythmogenic, Respiratory, and Psychological Risk Factors for SUDEP

713

Po st ict al Lary ng o spasm as One€Po t ent ial Mechanism fo r SUDEP A 1-min generalized tonic–clonic seizure occurred as a 42-year-old male patient diagnosed with refractory epilepsy was being monitored in an epilepsy-monitoring unit (Tavee and Morris 2008). This episode was followed by persistent inspiratory stridor and cyanosis. Cardiac parameters remained stable during the event. However, the respiratory status declined rapidly, despite the administration of oxygen via bagvalve-mask. As the emergency code team proceeded with intubation, they noted severe laryngospasm as the endotracheal tube was being inserted. Resuscitation was successful. This case demonstrated that postictal laryngospasm may be one potential cause of sudden unexpected death in persons with epilepsy. Co mment s by L at hers As discussed by Abu-Shaweesh (2007), activation of the laryngeal mucosa results in apnea mediated through, and elicited via, electrical stimulation of the superior laryngeal nerve. The inhibitory reflex is thought to be involved in the pathogenesis of apnea of prematurity and sudden infant death syndrome. Theophylline and block of GABA(A) receptors attenuate the inhibitory reflex. Phrenic nerve response to increasing levels of superior laryngeal nerve stimulation was examined in ventilated, vagotomized, decerebrate, and paralyzed newborn piglets. Phrenic activity decreased with increased stimulation of the superior laryngeal nerve and resulted in apnea and hypotension with higher levels of stimulation. It was concluded that activation of adenosine A (2A) receptors enhances superior laryngeal nerve stimulationinduced apnea. This may occur via a GABAergic pathway. The authors hypothesize that superior laryngeal nerve stimulation may cause endogenous release of adenosine to activate A (2A) receptors on GABAergic neurons, resulting in release of GABA at inspiratory neurons and subsequent respiratory inhibition. Case summarized from Tavee and Morris (2008).

Po st ict al Cent ral Apnea as One€Po t ent ial Mechanism fo r SUDEP A 55-s convulsive seizure occurred in a 20-year-old woman as she underwent videoEEG monitoring (So et al. 2000). Persistent apnea then developed. Electrocardiogrammonitored rhythm was not altered for the ἀrst 10 s, then it gradually and progressively slowed and stopped 57 s later. Cardiorespiratory resuscitation was successful. No evidence of airway obstruction or pulmonary edema was noted. Co mment s by S o , Sam, and Lag erlund One previous cardiorespiratory arrest after a complex partial seizure without secondary generalization had been reported for this patient. So et al. (2000) note that although epileptic seizures may be associated with arrhythmogenic actions at the

714 Sudden Death in Epilepsy: Forensic and Clinical Issues

heart, in this patient the mechanism of marked central suppression of respiratory activity after seizures was clearly involved and almost resulted in sudden death. Co mment s by L at hers Psychological factors, including stress, are risk factors for SUDEP (Schraeder and Lathers 1983; Fenwick 1994; Lathers and Schraeder 2006; Stopper et al. 2007; Lathers et al. 2008a, 2008b; Scorza et al. 2008). One must address the issue of the stress experienced by some persons with epilepsy as they undergo video-EEG monitoring, especially because monitoring protocols often include reduction or withdrawal of antiepileptic drugs to increase the likelihood of occurrence of a clinical event. The stress of having the actual procedure itself may trigger seizures and the unwanted respiratory and/ or cardiac events. Fortunately, in this patient and in the patient of Tavee and Morris (2008), successful resuscitation occurred and SUDEP appears to have been prevented. Unlike the ἀnding of Tavee and Morris (2008) discussed above, So et al. (2000) do not mention any observation of severe laryngospasm as the endotracheal tube was being inserted. We do not know if severe laryngospasm did occur in this 20-year-old woman with epilepsy and the emergency code team did not detect it and/or if it did actually occur. In the future, it will be important for emergency teams to note if a given person with known epilepsy exhibits severe laryngospasm. In any event, these case histories highlight that both respiratory and cardiac changes do occur in persons with epilepsy. The timing of events such as seizures, respiratory and/or laryngospasm, and cardiac ECG changes does vary in different patients. The physician must consider risk factors for a given individual patient and protective procedures to prescribe to protect the person from future unwanted events that may result in SUDEP. The question must be asked as to whether a person ἀrst experienced seizures and respiratory events and then cardiac events or if the person experienced seizures and arrhythmia and then respiratory events. There is documented evidence in the literature to support both cardiac and respiratory events as initiating mechanisms of sudden death. Obviously, rapid reversal of these changes is essential, and the “availability of resuscitation methods on the spot where the victim is located” certainly increases the likelihood that SUDEP will be prevented. It is also important to note that interictal discharges, just like ictal discharges, have been reported to be associated with cardiac arrhythmias and/or respiratory changes and/or sudden death (Lathers and Schraeder 1982; Schraeder and Lathers 1983). The persons administering the EEG/ECG video monitoring must watch for the appearance of both interictal and ictal discharges as warning signs of unwanted events that may subsequently be triggered. Case summarized from So et al. (2000).

44.3â•…Discussion The evidence presented in these two “near miss” sudden death patient case histories highlight two potential mechanisms of SUDEP. In one person with epilepsy, severe laryngoÂ� spasm was detected as the endotracheal tube was being inserted, suggesting that postictal laryngospasm may be one potential cause of sudden unexpected death. In another person

Arrhythmogenic, Respiratory, and Psychological Risk Factors for SUDEP

715

with epilepsy, postictal central apnea occurred and suggested that this is another potential mechanism of sudden death (So et al. 2000). Clearly, there are different mechanisms of SUDEP involved in different patients. Lathers et al. (2008a) recently summarized proposed mechanistic factors in SUDEP with the three major risk categories of arrhythmogenic factors, respiratory factors and hypoxia, and psychological factors (see Tables 1.1 through 1.8 in Lathers et al. 2010, Chapter 1). The reader is also referred to Monte et al. (2007), Lathers and Schraeder (2010, Chapter 28), Walczak (2010, Chapter 12), Tomson (2010, Chapter 51), and Chapters 3, 4, and 13–19 in this book for further discussion about the risk factors and mechanisms of sudden death in persons with epilepsy. Strong risk factors for SUDEP include being a young adult male, having generalized tonic seizures, and lying in bed. Less vigorous risk factors include being in prone posture, having subtherapeutic AED blood levels, being in a bedroom (not in bed), and having a structural brain lesion. Nilsson et al. (2001) investigated the association between clinical variables and SUDEP in an effort to identify risk factors in a Swedish population and reported that 91% of the 57 SUDEP cases studied had autopsies performed. The associated risk factors for SUDEP were: 1. Higher number of seizures per year (a relative risk of 10.16 in patients having more than 50 seizures per year compared to those with up to two per year) 2. Increased number of antiepileptic drugs (9.89 for three antiepileptic drugs vs. monotherapy) 3. Early- vs. late-onset epilepsy (7.72) 4. Frequent changes in antiepileptic drug dosage vs. unchanged dosage (6.08) 5. Male gender The study found the association between SUDEP risk and early onset, and SUDEP risk and seizure frequency was weaker for females vs. frequent dosage changes, which had a stronger association in females. This study conἀrmed early age of onset of epilepsy and male gender as risk factors (Nilsson et al. 2001). In a more recent report (McHugh and Delanty 2008), the marginally lower incidence of epilepsy and unprovoked seizures in females is attributed to the fact that males have a greater exposure to risk factors for lesional epilepsy and acute symptomatic seizure. Females have a greater incidence of idiopathic generalized epilepsies that represent approximately 15–20% of all epilepsies. Of interest, common epilepsy syndromes such as mesial temporal sclerosis may differ between the genders with isolated auras more common among females and secondary seizure spread more likely to occur in males. Gender differences are observed in the incidence of status epilepticus (more common in men), the incidence of SUDEP, prognosis, and mortality (McHugh and Delanty 2008). Bateman et al. (2008) examined the incidence and severity of ictal hypoxemia in patients with localization-related epilepsy undergoing video-EEG telemetry. They measured seizure associated oxygen desaturation and hypoventilation. Pulse oximetry revealed oxygen desaturations below 90% in 33.2% of all 304 seizure events. The degree of desaturation was signiἀcantly correlated with seizure duration and with electrographic evidence of seizure spread to the contralateral hemisphere. Central apneas or hypopneas occurred with 50% of all seizures. Ictal hypoxemia occurred often in these patients with localization related epilepsy, and may be pronounced and prolonged, even if the seizures do not progress to generalized convulsions. End-tidal carbon dioxide increase occurred in oxygen desaturation and supports the assumption that ictal oxygen desaturation is a consequence of

716 Sudden Death in Epilepsy: Forensic and Clinical Issues

hypoventilation. Both ictal hypoxemia and hypercapnia may be contributing risk factors to SUDEP occurrence. Thus, mechanisms of SUDEP are various (Tables 1.1, 1.2, and 1.4 in Chapter 1) and include central and obstructive apnea and cardiovascular, including cardiac arrhythmias. Caution must be exerted when concluding that respiratory changes are the primary mechanism of death. In some victims, SUDEP has been attributed to a mechanism of cardiac arrhythmia (Lathers and Schraeder 1982, 1990; Schraeder and Lathers 1983; Drake et al. 1993). In some cases, the arrhythmias may be triggered by cerebral events. Data from the animal studies of Lathers and Schraeder evaluated ECG records obtained in animals before, during, and after epileptogenic activity was induced. Study results monitored intrathoracic cardiac sympathetic and vagal branch discharges and showed that discharge patterns varied widely with different degrees of epileptiform discharges. The cardiac sympathetic nerve discharges were time-locked to the bilateral cortical interictal spikes induced by pentylenetetrazol (Lathers et al. 1987; Stauffer et al. 1989). During interictal discharges, this time-locked event called the lockstep phenomenon waxed and waned among all of the monitored sympathetic branches. Latency between electrocorticogram spike and discharges in the cardiac sympathetic branches ranged from 30 to 100€ms. This suggests the cortical discharges preceded the cardiac neural discharges and were probably conducted via a multisynaptic pathway within the central nervous system. Only one of ἀve cats in which vagal branches were recorded manifested this phenomenon with interictal spikes. During the lockstep phenomenon, the ECG exhibited peaking, inversion, and biphasic T waves as well as P-wave changes consisting of peaking, flattening, and biphasic conἀguration; variable Q-wave changes and widened QRS intervals suggested neurologically mediated interference in myocardial conduction time. Premature ventricular contractions occurred in one animal in association with interictal spikes. The lockstep phenomenon was also signiἀcantly correlated with variability in mean arterial blood pressure. Resting ECGs obtained from 75 persons with epilepsy were reviewed (Drake et al. 1993). Ventricular rate, P–R interval, QRS duration, and QT interval corrected for heart rate were compared with normal ECGs obtained from age-matched patients with no cardiac or neurological disease. No potentially lethal arrhythmias were found in the persons with epilepsy. Those who appeared to be in a risk category for SUDEP did have more abnormal ECGs and the ventricular rate was faster than in other epileptic persons. Those with complex partial and secondary generalized seizures exhibited a faster ventricular rate than other epileptics. The QT interval was longer in patients with complex partial seizures vs. control or other epileptics while no changes in QRS duration or P–R interval were found. The authors conclude that resting ECG has low diagnostic yield in persons with epilepsy who do not exhibit cardiac symptoms. They suggest that factors predisposing to SUDEP may be associated with a relative increase in the resting heart rate. The relative increase in the QT interval with complex partial seizures may indicate some difference in cardiac excitability. This cardiac excitability possibly may be neurally mediated and could contribute to the occurrence of SUDEP. Nei et al. (2000) noted that SUDEP is responsible for excess mortality in those with refractory epilepsy and cardiac arrhythmias during seizures may be responsible. They examined the frequency of cardiac abnormalities during seizures in this patient group. Arrhythmias, repolarization abnormalities, and PR and QTc intervals were examined in the preictal, ictal, and postictal periods for one or more seizures per patient. Thirty-nine

Arrhythmogenic, Respiratory, and Psychological Risk Factors for SUDEP

717

percent (39%) or more exhibited one or more abnormalities of rhythm and/or repolarization during or immediately after seizures. Abnormalities included asystole, atrial ἀbrillation, mild or moderate sinus arrhythmia, supraventricular tachycardia, atrial premature depolarization, ventricular premature depolarization, and bundle-branch block. The duration of seizures was longer in persons with epilepsy and cardiac abnormalities. These abnormalities may be a contributing cause to SUDEP. These data obtained from persons with epilepsy document the ἀndings that Lathers and Schraeder (1982) and Schraeder and Lathers (1983) obtained in an animal model for SUDEP. Toth et al. (2008) examined the heart rate 6 h before and after seizure in 18 patients with focal epilepsy before epilepsy surgery. Studies were done for 2–10 days and 32 seizures were recorded. Heart rate was analyzed before and after seizures. Post-seizures heart rate was increased and was even higher 3 h later. One patient exhibited an ectopic cardiac rhythm after a generalized tonic–clonic seizure and none exhibited severe peri-ictal bradycardia. It was concluded that sympathetic activity increased while parasympathetic activity decreased after seizures. Because the changes persisted for a long time and may predict fatal arrhythmia, sudden death in epilepsy may be induced by cardiac arrhythmias connected with epileptic seizures. In some patients, seizures contribute to initiation of peripheral cardiac arrhythmias that may contribute to sudden death if not corrected. Such a case in which cerebral arrhythmia influenced cardiac rhythm was reported by Almansori et al. (2006). They noted that partial seizures of temporal origin may be associated with clinically signiἀcant tachycardia or bradycardia. Both ictal bradycardia and asystole have been designated as risk factors for SUDEP. Many patients exhibit symptomatic ictal bradycardia. In this case report, a patient being monitored via video EEG telemetry was diagnosed with asymptomatic ictal bradycardia. A cardiac pacemaker was implanted. As discussed by Zaidi et al. (2000), many seizure-like attacks have a cardiovascular cause and result in a misdiagnosis of epilepsy. Early on in the diagnosis, simple noninvasive cardiovascular evaluation with a head-up tilt test (Lathers et al. 1990, 1991) and carotid sinus massage during continuous electrocardiography, electroencephalography, and blood pressure monitoring should be done to identify patients with apparent epilepsy presenting with convulsive syncopal blackouts. Linzer et al. (1994) note that syncope and seizures are often clinically indistinguishable. Use of Holter monitoring, long-term ambulatory loop electrocardiographic recording, or tilt-table studies may be done to diagnose arrhythmic or neurally mediated syncope and to eliminate misdiagnosis of these patients as persons with epilepsy and treatment with anticonvulsant medications. Correct diagnosis is necessary to prevent potentially fatal events. An incorrect incidence of SUDEP may occur if undiagnosed cardiac syncope contributes to the documented increased sudden death rate in patients with presumed epilepsy. Likewise, Akhtar (Akhtar) recommends a simple noninvasive cardiovascular evaluation to diagnose convulsive syncope in children with apparent treatment-resistant epilepsy and to assess the extent of misdiagnosis of epilepsy in children. Long QT syndrome can masquerade as epilepsy. Hunt and Tang (2005) present the case of a female presenting to the emergency department with a generalized seizure. She had been previously diagnosed with epilepsy. Her ECG demonstrated QT prolongation secondary to bradycardia. A subsequent seizure documented the events were secondary to cerebral hypoperfusion during episodes of torsades de pointes. When the long QT syndrome does masquerade as epilepsy, the correct cardiac treatment is delayed and the patient is placed in a high risk of sudden cardiac death.

718 Sudden Death in Epilepsy: Forensic and Clinical Issues

44.3.1â•…A nalysis of the Electrocardiogram Is Essential for All Patients Who Present with Seizures In spite of the high proἀle of SUDEP, evidence for speciἀc risk factors and pathophysiology is not yet ἀrmly established. Inaccurate death certiἀcation, fewer postmortem examinations, and poor incident case reporting limit the value of epidemiological data on SUDEP (Johnston and Smith 2007). The importance of case histories is to build the “bigger medical picture” to understand the mechanisms of risk factors for sudden death in persons with epilepsy. Case histories also teach physicians and caregivers how to diagnose and how to treat patients with epilepsy who present with symptoms and risk factors for sudden death to prevent the occurrence of sudden death. Ultimately, we anticipate that the best medical/surgical practices will emerge to provide the best quality of life for those individuals who are at risk for SUDEP.

References Abu-Shaweesh, J. M. 2007. Activation of central adenosine A(2A) receptors enhances superior laryngeal nerve stimulation-induced apnea in piglets via a GABAergic pathway. J Appl Physiol 103 (4): 1205–1211. Akhtar, M. J. 2002. All seizures are not epilepsy: Many have a cardiovascular cause. J Pak Med Assoc 52 (3): 116–120. Almansori, M., M. Ijaz, and S. N. Ahmed. 2006. Cerebral arrhythmia influencing cardiac rhythm: A case of ictal bradycardia. Seizure 15 (6): 459–461. Bateman, L. M., C. S. Li, and M. Seyal. 2008. Ictal hypoxemia in localization-related epilepsy: Analysis of incidence, severity and risk factors. Brain 131 (Pt 12): 3239–3245. Drake, M. E., C. R. Reider, and A. Kay. 1993. Electrocardiography in epilepsy patients without cardiac symptoms. Seizure 2 (1): 63–65. Fenwick, P. 1994. The behavioral treatment of epilepsy generation and inhibition of seizures. Neurol Clin 12 (1): 175–202. Hunt, D. P., and K. Tang. 2005. Long QT syndrome presenting as epileptic seizures in an adult. Emerg Med J 22 (8): 600–601. Jim, K. F., C. M. Lathers, V. L. Farris, L. F. Pratt, and W. H. Spivey. 1989. Suppression of pentylenetetrazol-elicited seizure activity by intraosseous lorazepam in pigs. Epilepsia 30 (4): 480–486. Jim, K. F., C. M. Lathers, W. H. Spivey, W. D. Matthews, C. Kahn, and K. Dolce. 1988. Suppression of pentylenetetrazol-elicited seizure activity by intraosseous propranolol in pigs. J Clin Pharmacol 28 (12): 1106–1111. Johnston, A., and P. Smith. 2007. Sudden unexpected death in epilepsy. Expert Rev Neurother 7 (12): 1751–1761. Langan, Y., L. Nashef, and J. W. Sander. 2000. Sudden unexpected death in epilepsy: A series of witnessed deaths. J Neurol Neurosurg Psychiatry 68 (2): 211–213. Lathers, C. M., K. F. Jim, W. B. High, W. H. Spivey, W. D. Matthews, and T. Ho. 1989a. An investigation of the pathological and physiological effects of intraosseous sodium bicarbonate in pigs. J Clin Pharmacol 29 (4): 354–359. Lathers, C. M., K. F. Jim, and W. H. Spivey. 1989b. A comparison of intraosseous and intravenous routes of administration for antiseizure agents. Epilepsia 30 (4): 472–479. Lathers, C. M., N. Tumer, and J. M. Schoffstall. 1989c. Plasma catecholamines, pH, and blood pressure during cardiac arrest in pigs. Resuscitation 18 (1): 59–74. Lathers, C. M., P. H. Diamandis, J. M. Riddle, C. Mukai, K. F. Elton, M. W. Bungo, and J. B. Charles. 1990. Acute and intermediate cardiovascular responses to zero gravity and to fractional gravity levels induced by head-down or head-up tilt. J Clin Pharmacol 30 (6): 494–523.

Arrhythmogenic, Respiratory, and Psychological Risk Factors for SUDEP

719

Lathers, C. M., P. H. Diamandis, J. M. Riddle, C. Mukai, K. F. Elton, M. W. Bungo, and J. B. Charles. 1991. Orthostatic function during a stand test before and after head-up or head-down bed rest. J Clin Pharmacol 31 (10): 893–903. Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67 (3): 247–259. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008a. The mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12 (1): 3–24. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23 (6): 633–647. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9 (2): 236–242. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2010. Neurocardiologic mechanistic risk factors in sudden unexpected death in epilepsy. Chapter 1 in Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M.€W.€Bungo, and J. Leestma. Boca Raton: CRC Press. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008b. Chapter 13. Sudden death: Neurocardiologic mystery. In Psychological Factors and Cardiovascular Disorders, ed. L. Sher. Hauppauge, NY: Nova Science. Lathers, C. M., and P. L. Schraeder. 1990. Chapter 9. Arrhythmias associated with epileptogenic activity elicited by penicillin. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. Schraeder. New York, NY: Marcel Dekker. Lathers, C. M., and P. L. Schraeder. 2010. Antiepileptic drugs beneἀt/risk clinical pharmacology: Possible role in cause and/or prevention of SUDEP. Chapter 37 in Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. Leestma. Boca Raton: CRC Press. Linzer, M., B. P. Grubb, S. Ho, L. Ramakrishnan, E. Bromἀeld, and N. A. Estes 3rd. 1994. CardioÂ� vascular causes of loss of consciousness in patients with presumed epilepsy: A cause of the increased sudden death rate in people with epilepsy? Am J Med 96 (2): 146–154. McHugh, J. C., and N. Delanty. 2008. Epidemiology and classiἀcation of epilepsy: Gender comparisons. Int Rev Neurobiol 83: 11–26. Monte, C. P., J. B. Arends, I. Y. Tan, A. P. Aldenkamp, M. Limburg, and M. C. de Krom. 2007. Sudden unexpected death in epilepsy patients: Risk factors. A systematic review. Seizure 16 (1): 1–7. Nei, M., R. T. Ho, and M. R. Sperling. 2000. EKG abnormalities during partial seizures in refractory epilepsy. Epilepsia 41 (5): 542–548. Nilsson, L., U. Bergman, V. Diwan, B. Y. Farahmand, P. G. Persson, and T. Tomson. 2001. Antiepileptic drug therapy and its management in sudden unexpected death in epilepsy: A case-control study. Epilepsia 42 (5): 667–673. Schoffstall, J. M., W. H. Spivey, S. Davidheiser, and C. M. Lathers. 1989. Intraosseous crystalloid and blood infusion in a swine model. J Trauma 29 (3): 384–387. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32 (12): 1371–1382. Schraeder, P. L., K. Delin, R. L. McClelland, and E. L. So. 2006. Coroner and medical examiner documentation of sudden unexplained deaths in epilepsy. Epilepsy Res 68 (2): 137–143. Schraeder, P. L., K. Delin, R. L. McClelland, and E. L. So. 2009. A nationwide survey of the extent of autopsy in sudden unexplained death in epilepsy. Am J Forensic Med Pathol 30 (2): 123–126. Scorza, F. A., R. M. Arida, and E. A. Cavalheiro. 2008. Preventive measures for sudden cardiac death in epilepsy beyond therapies. Epilepsy Behav 13 (1): 263–264; author reply 265–269. So, E. L., M. C. Sam, and T. L. Lagerlund. 2000. Postictal central apnea as a cause of SUDEP: Evidence from near-SUDEP incident. Epilepsia 41 (11): 1494–1497. Spivey, W. H., H. D. Unger, C. M. Lathers, and R. M. McNamara. 1987a. Intraosseous diazepam suppression of pentylenetetrazol-induced epileptogenic activity in pigs. Ann Emerg Med 16 (2): 156–159.

720 Sudden Death in Epilepsy: Forensic and Clinical Issues Spivey, W. H., H. D. Unger, R. M. McNamara, M. M. LaManna, T. Ho, and C. M. Lathers. 1987b. The effect of intraosseous sodium bicarbonate on bone in swine. Ann Emerg Med 16 (7): 773–776. Stauffer, A. Z., J. Dodd-O, and C. M. Lathers. 1989. The relationship of the lock-step phenomenon and precipitous changes in mean arterial blood pressure. Electroencephalogr Clin Neurophysiol 72 (4): 340–345. Stopper, M., T. Joska, M. M. Burg, W. P. Batsford, C. A. McPherson, D. Jain, and R. Lampert. 2007. Electrophysiologic characteristics of anger-triggered arrhythmias. Heart Rhythm 4 (3): 268–273. Tavee, J., and H. Morris 3rd. 2008. Severe postictal laryngospasm as a potential mechanism for sudden unexpected death in epilepsy: A near-miss in an EMU. Epilepsia 49 (12): 2113–2117. Tomson, T. 2010. Compliance with antiepileptic drug treatment and the risk of sudden unexpected death in epilepsy. Chapter 51 in Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. Schraeder, M. W. Bungo, and J. Leestma. Boca Raton: CRC Press. Toth, V., L. Hejjel, Z. Kalmar, A. Fogarasi, T. Auer, C. Gyimesi, A. Szucs, and J. Janszky. 2008. Effect of epileptic seizures on the heart rate. Ideggyogy Sz 61 (5–6): 155–161. Walczak, T. 2010. Risk factors for sudden death in epilepsy. Chapter 12 in Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. Schraeder, M. W. Bungo, and J. Leestma. Boca Raton: CRC Press. Zaidi, A., P. Clough, P. Cooper, B. Scheepers, and A. P. Fitzpatrick. 2000. Misdiagnosis of epilepsy: Many seizure-like attacks have a cardiovascular cause. J Am Coll Cardiol 36 (1): 181–184.

Sudden Arrhythmic Death Syndrome Underlying Cardiac Etiologies, Their Implications, and the Overlap with SUDEP

45

Paramdeep S. Dhillon Elijah R. Behr

Contents 45.1 Introduction 45.2 Illustrative Case 45.3 SADS: A Deἀnition and Characteristics 45.3.1 Etiology 45.3.1.1 The Channelopathies 45.3.2 Structural Heart Disease 45.3.2.1 Arrhythmogenic Right Ventricular Cardiomyopathy 45.3.2.2 Hypertrophic Cardiomyopathy 45.3.2.3 Other Structural Cardiac Diseases 45.4 The Familial Basis of SADS 45.5 Non-Mendelian Basis to SADS? 45.6 The Overlap between SUDEP and SADS 45.7 Conclusion References

721 722 724 724 724 730 731 731 732 733 733 733 735 735€

45.1â•…Introduction Approximately 300,000 adults die suddenly each year in the United States (Virmani et al. 2001) mainly as the result of fatal ventricular arrhythmias related to coronary artery and structural heart disease (Davies and Thomas 1984). Unexplained sudden death in children, adolescents, and young adults are much less common with an incidence of between 1.3 and 8.5 deaths per 100,000 patient-years (Liberthson 1996) equivalent to approximately 500 deaths annually in the UK (Behr et al. 2007), or 2500 deaths per annum in the United States. Cardiac fatalities in this younger group of individuals are caused predominantly by cardiomyopathic disease, although other heart diseases are recognized (Table 45.1) (Corrado et al. 2001; Fabre and Sheppard, 2006). Up to 30% of deaths remain unexplained despite a thorough postmortem examination and toxicology screen (Behr et al. 2003; Chugh et al. 2000; Corrado et al. 2001; Maron et al. 1996; Morentin et al. 2003; Puranik et al. 2005). Although the terms sudden adult death syndrome and sudden death syndrome have been used to describe unexpected death occurring within 1 h of symptoms (de la Grandmaison 721

722 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 45.1â•… Principal Causes of Sudden Death in Young Adults That May Be Identified at Autopsy Toxicology (Causes with a Predominant Genetic Basis Are Marked with Asterisks) Cardiomyopathy

Coronary artery disease Infective and inἀltrative Valvular heart disease Conduction disease Congenital heart disease Connective tissue disease Illicit drug use Unknown cause

Hypertrophic cardiomyopathy* ARVC* Dilated cardiomyopathy* Left ventricular non-compaction* Glycogen storage disease and mitochondrial myopathy* Premature coronary artery disease* Anomalous coronary arteries Focal myocarditis Cardiac sarcoidosis Mucoid degeneration of the mitral valve Lev-Lenègre’s disease Wolf–Parkinson–White syndrome Various Marfan’s syndrome* Cocaine, amphetamines Idiopathic ἀbrosis Idiopathic left ventricular hypertrophy

and Durigon 2002), the term sudden arrhythmic death syndrome (SADS) is now preferred to describe unexpected and unheralded deaths in adults as well as children (excluding infants) where no deἀnite cause of death can be found at postmortem (Behr et al. 2003; Bowker et al. 2003). Familial evaluation studies have shown that up to half of sudden unexplained deaths in young adults may have been the result of hereditary cardiac conditions including ion channel disease (Behr et al. 2003, 2008; Tan et al. 2005). These diseases are associated with mutations in genes coding for proteins that form or interact with cardiac ion channels that conduct sodium, potassium, and calcium ions (Ackerman 2004). The generation and propagation of the cardiac action potential are then disturbed, predisposing those affected to cardiac arrhythmias (Marban 2002). Epileptics are also prone to sudden unexpected death (Morentin and Alcaraz 2002), despite a normal heart at autopsy, and epilepsy-related cardiac rhythm abnormalities (Al-Hillawi et al. 1984; Lim et al. 1989; Rocamore et al. 2003; Tigaran et al. 2002) may offer a potential explanation to this phenomenon. In addition, ventricular arrhythmias and asystole may cause symptoms that mimic seizure disorders raising the possibility that the patient is not epileptic. In a recent study of patients diagnosed with the congenital long QT syndrome (LQTS), 7% of patients were at some stage managed with antiepileptic drugs (Johnson et al. 2009). The purpose of this chapter, therefore, is to describe the features of SADS and the underlying etiologies. In particular, we explore whether there are potential overlaps with the phenomenon of sudden unexpected death in epilepsy (SUDEP) and whether there may be common underlying mechanisms.

45.2â•…Illustrative Case A 25-year-old woman was referred by her neurologist. From the age of 15, she had suffered recurrent episodes of witnessed loss of consciousness. She would become vacant and

Sudden Arrhythmic Death Syndrome

723

unresponsive before falling to the floor and becoming rigid and motionless. She would then develop limb twitching, hyperventilation, and was unresponsive for approximately 1 min. On one occasion she bit her tongue and was incontinent of urine. There was no family history of epilepsy or personal history of neonatal trauma or febrile convulsion, although a male cousin had died in his sleep unexpectedly at age 32. She was diagnosed with complex partial and secondary generalized epilepsy but failed to respond to a variety of medications including lamotrogine, levitiracetam, and subsequently oxcarbazepine. Standard and sleep-deprived electroencephalography and brain magnetic resonance imaging did not demonstrate an abnormality. A 12-lead electrocardiogram (ECG) performed while on levitiracetam demonstrated QT interval prolongation with broad-based and flattened T-wave morphology that persisted despite discontinuation of the drug (Figure 45.1a). Apart from clear repolarization abnormality, all other cardiac (a)

(b)

Figure 45.1╇ (a) Twelve-lead ECG demonstrating QT interval prolongation with a broad-based and flattened T-wave morphology. (b) Polymorphic ventricular tachycardia.

724 Sudden Death in Epilepsy: Forensic and Clinical Issues

investigations were within normal limits. Further familial evaluation revealed that her mother had suffered a single seizure as a young women after childbirth. An ECG demonstrated mild QT interval prolongation. An implantable internal-loop recorder was able to demonstrate polymorphic ventricular tachycardia (Figure 45.1b) that correlated with her next episode of loss of consciousness. She was therefore diagnosed with the congenital LQTS and commenced on propranolol and her antiepileptic medication discontinued.

45.3â•… SADS: A Definition and Characteristics SADS is an umbrella term for unexpected and unexplained sudden death and has been deἀned for research purposes as a sudden death, age 4–64 years, last seen alive and well within 12 h of being found dead, no prior recorded cardiac disease, a normal full coroner’s postmortem, negative toxicology results, and a normal expert cardiac pathologist’s examination (Behr et al. 2003, 2008). Most of the victims are male, have antecedent symptoms (particularly syncope), and die in their sleep. A family history of sudden death can be elicited in 30% of cases (Behr et al. 2003, 2008). 45.3.1â•… Etiology The underlying etiologies of SADS, where identiἀable, have been shown to be predominantly inherited cardiac diseases. In particular, “molecular autopsy,” the postmortem study of unexplained sudden death victims, has established the presence of cardiac ion channel diseases or “channelopathies” (Tester and Ackerman 2007). In a British study of 57 families of SADS victims, such a diagnosis was established in 30 (53%) families after a comprehensive investigative algorithm (Figure 45.2) (Behr et al. 2008). This was an improvement over prior research from the same group that detected inherited heart disease in 7 of 32 families (22%) but relied upon a much more limited investigative protocol (Behr et al. 2003). Channelopathies were the most common cause, but subtle structural heart disease was also noted, in particular arrhythmogenic right ventricular cardiomyopathy (ARVC). Similar results were described in a Dutch study of 43 families with a high frequency of unexplained sudden deaths that showed that 17 (40%) deaths were due to inherited cardiac diseases (Tan et al. 2005). The main causes of SADS are shown in Figure 45.3. 45.3.1.1â•… The Channelopathies The cardiac ion channel diseases have several common features: a genetic etiology; an absence of structural disease; and the risk of arrhythmias causing syncope, seizures, and/ or sudden death. This group of diseases include LQTS, Brugada syndrome (BrS), and catecholaminergic polymorphic ventricular tachycardia that have been recognized as inherited causes of SADS (Behr et al. 2008; Tan et al. 2005; Tester and Ackerman 2006). Other more rare entities such as short QT syndrome (SQTS), progressive cardiac conduction defect (PCCD), and the recently described “early repolarization syndrome” may also play a role but have yet to be formally described in this group. 45.3.1.1.1â•… ἀ e Long QT Syndromeâ•… Congenital LQTS affects 1 in 5000 persons and is characterized by delayed myocardial repolarization that may manifest itself on the surface

Sudden Arrhythmic Death Syndrome

725

Proband evaluation: Mutation analysis in probands: “the molecular autopsy”

Sudden death Normal coroner’s autopsy Negative toxicology ± Normal expert pathologist’s assessment

NB concurrent process without delay to familial evaluation

Familial evaluation Initial relative evaluation: Historical assessment Physical examination Resting ECG 24 h ECG Exercise ECG Echocardiogram Normal heart

Normal ECG

Ajmaline test

Abnormal or equivocal cardiac morphology

Right ventricular lead repolarization changes

Cardiac MRI

Inherited heart disease diagnosed? Mutation analysis in relatives: If a relative is diagnosed or suspected of carrying an inherited cardiac disease, proceed to mutation analysis. If an unequivocal mutation is identified, offer familial cascade genetic testing and exclude noncarriers. If an equivocal mutation is detected, offer familial cascade testing and clinical evaluation.

Figure 45.2╇ An algorithm for investigating families with a SADS death. (Adapted from Behr, E. R., et al., Eur Heart J, 29, 1670–1680, 2008.)

ECG as QT interval prolongation and T-wave abnormalities (Figure 45.1a). Early afterdepolarizations may trigger ventricular arrhythmias (Roden 1993), speciἀcally polymorphic ventricular tachycardia (torsades de pointes) (Figure 45.1b) and ventricular ἀbrillation. If these are self-limiting, they result in syncope and/or secondary anoxic seizures, but if sustained, they result in sudden death in the setting of a structurally normal heart (Keating and Sanguinetti 2001). The Schwartz score (Schwartz et al. 1993) encompasses the accepted clinical criteria for the diagnosis of LQTS and relies on phenotypic features of risk such as the degree of QT

726 Sudden Death in Epilepsy: Forensic and Clinical Issues LQTS — definite 23%

(a) No diagnosis 47%

LQTS — possible 5%

Brugada Syndrome 9%

HCM 2% (b)

DCM 2%

LV noncompaction

ARVC 9%

3%

LQTS 21%

CPVT 14%

No diagnosis 65%

Figure 45.3╇ The causes of sudden arrhythmic death syndrome. CPVT, catecholaminergic

polymorphic ventricular tachycardia; LQTS, long QT syndrome; ARVC, arrhythmogenic right ventricular cardiomyopathy; DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy. (a) Adapted from Behr et al., Eur Heart J, 29, 1670–1680, 2008. (b) Adapted from Tester, D.€J., and Ackerman, M. J., J Am Coll Cardiol, 49, 240–246, 2007.

Sudden Arrhythmic Death Syndrome

727

prolongation, other ECG abnormalities, documented ventricular arrhythmia, prior syncope, prior cardiac arrest, and a family history of sudden death and/or LQTS. A score of ≥4 is highly speciἀc for LQTS in an index case, although its sensitivity is low, particularly in relatives of the index case (Hofman et al. 2007). This is because LQTS is a genetically heterogeneous disease with variable penetrance, meaning that some patients demonstrate little or no QT prolongation on the 12-lead ECG, the forme fruste (Priori et al. 1999), until an additional event such as exposure to a QT-prolonging drug or electrolyte imbalance occurs (Yang et al. 2002). The natural history of LQTS patients varies greatly from sudden death in childhood to asymptomatic longevity even among members of the same family (Ackerman and Clapham 1997). The more common form of LQTS (99%) is autosomal dominant and known as the Romano–Ward syndrome (Romano et al. 1963; Ward 1964). The rarer autosomal recessive form, known as the Jervell and Lange–Nielsen syndrome, is associated with sensorineural deafness (Jervell and Lange-Nielsen 1957). The genetic basis of LQTS has become better understood over the past 15 years with the identiἀcation of disease-Â�associated mutations in 12 different genes linked to cardiac repolarization and depolarization (Table 45.2). These mutations lead either to a net reduction or “loss-of-function” of outward potassium rectifying currents (KCNQ1, KCNH2, KCNE1, KCNJ2, and KCNE2) or a net increase or “gainof-function” in the inward sodium current (SCN5A) or calcium currents (CACNA1C) (Noseworthy and Newton-Cheh 2008; Rodriguez-Calvo et al. 2008). Mutations of the ἀrst ἀve identiἀed genes (KCNQ1, KCNH2, SCN5A, KCNE1, and KCNE2) account for up to 70% of unrelated deἀnite LQTS probands undergoing genetic testing, with KCNQ1- and KCNH2-associated LQTS (LQT1 and LQT2, respectively) accounting for approximately 40–45% of genotyped patients each and SCN5A-associated disease (LQT3) being linked in 8–15% (Splawski et al. 2000; Tester et al. 2005). Also recognized are mutations in cellular

Table 45.2â•… Congenital LQTS: Subtypes, Their Genetic Basis, and Frequencies Subtype

Chromosome

Gene

Product

Population Frequency

Long QT syndrome 1 Long QT syndrome 2 Long QT syndrome 3 Long QT syndrome 4

11 7 3 4

KCNQ1 KCNH2 SCN5A ANKB

~30% ~30% ~5% ~2%

Long QT syndrome 5 Long QT syndrome 6 Long QT syndrome 7 Long QT syndrome 8 Long QT syndrome 9 Long QT syndrome 10 Long QT syndrome 11 Long QT syndrome 12

21 21 23 12 7 11 7 20

KNCE1 KNCE2 KCNJ2 CACNA1c CAV3 SCN4B AKAP9 SNTA1

IKs α subunit IKr α subunit INa α subunit Submembrane component IKs β subunit IKr β subunit IK1 Cav1.2 Caveola component INa β subunit AKAP9/yotiao a1-syntrophin

~2% ~2% – – ~2% – – –

Sources: Chen, L., et al., Proc Natl Acad Sci U S A, 104 (52), 20990–20995, 2007. Hofman, N., et al. Eur Heart J, 28 (11), 1399, 2007. Splawski, I., et al., Circulation, 102 (10), 1178–1185, 2000. Tester, D. J., et al., Heart Rhythm, 2 (5), 507–517, 2005. Ueda, K., et al., Proc Natl Acad Sci U S A, 105 (27), 9355–9360, 2008. Note: IKs, slow rectifying potassium current; IKr, rapid rectifying potassium current; INa, inward sodium current; IK1, Kir 2.1 inward rectifying current; Cav1.2, L-type calcium channel current.

728 Sudden Death in Epilepsy: Forensic and Clinical Issues

proteins required for cellular localization of ion channels, otherwise known as channel interacting proteins such as ankyrinB and caveolin3 (Rodriguez-Calvo et al. 2008). The incidence of sudden death or resuscitated cardiac arrest from birth to age 40 years ranges between 30% for KCNQ1 mutations and 46% in KCNH2 mutations (Priori et al. 2003). Interestingly, several phenotypical variations in terms of triggers for ventricular arrhythmias have been described. Swimming and exertion-related arrhythmias are associated with the LQT1 phenotype (Schwartz et al. 2001). In contrast, the LQTS3 phenotype is associated with cardiac events predominantly during sleep and rest. Auditory triggers may precipitate cardiac events, which also occur in the period after childbirth in LQT2 patients (Moss et al. 2002). Risk assessment of LQTS patients is predominantly based on symptoms (in particular syncope and prior cardiac arrest), gender, and the length of the corrected QT interval; a corrected QT interval >500 ms on the resting ECG carrying a higher risk (Hobbs et al. 2006; Sauer et al. 2007). The LQT1 genotype has been thought to be less hazardous than LQT2 and LQT3 (Priori et al. 2003), but the latest evidence from the international long QT registry does not support the utility of genetic locus as a marker for risk apart from in female adult LQT2 carriers (Goldenberg and Moss 2008). Beta-blocker therapy appears to be effective for patients with the LQT1 and LQT2 phenotypes, but is less effective for LQT3 patients. An implantable cardiac deἀbrillator (ICD) is indicated for survivors of cardiac arrest and those deemed particularly high risk for sudden death (Goldenberg et al. 2008). Surgical left cervicothoracic sympathetic denervation has been shown to reduce the mean annual number of LQTS-related cardiac events and ICD discharges in high-risk patients (Schwartz et al. 2004). 45.3.1.1.2â•… ἀ e Brugada Syndromeâ•… Brugada syndrome is an inherited condition that is characterized by the presence of coved ST elevation and “J point” elevation of at least 2 mm in at least two of the right precordial ECG leads (“type-1” ECG) (Figure 45.4) in the absence of cardiac structural disease (Antzelevitch et al. 2005; Brugada and Brugada 1992). The prevalence of the Brugada ECG pattern is between 0.05% in Europe and 0.6% in Japan (Hermida et al. 2000; Matsuo et al. 2004) and appears to increase with age (Brugada et al. 2001). BrS is thought to be the same disease as the “sudden unexpected nocturnal death syndrome” in Southeast Asia, also known as “Pokkuri” in Japan and “Lai Tai” in Thailand. It is characterized by ventricular arrhythmias, conduction abnormality, atrial arrhythmias, syncope, and sudden death. The diagnosis in an individual requires the presence of the Brugada ECG pattern with at least one of the recognized diagnostic criteria: syncope, prior cardiac arrest, documented or inducible polymorphic ventricular tachycardia or ventricular ἀbrillation, a family history of sudden death G) and p.P308L (c.923C>T), for the CASQ2 gene. Subsequent familial analysis indicated a compound heterozygous form of inheritance (De la Fuente et al. 2008). Lehnart et al. (2008) provided data demonstrating leaky Ca2+ release channel/ryanodine receptor 2 (RyR2) can trigger seizures and sudden cardiac death in mice. This receptor is required for excitation–contraction coupling in the heart. The receptor is also located in the brain. Mutations of this receptor are linked to exercise-induced sudden cardiac death catecholaminergic polymorphic ventricle tachycardia. Catecholaminergic polymorphic ventricular tachycardia–associated mutations in the receptor may decrease binding to a subunit [calstabin (FKBP12.6)] responsible for stabilizing the closed state of the Ca+ channel. Mice heterozygous for the R2474S mutation in Ryr2 mice exhibited spontaneous generalized tonic–clonic seizures in the absence of arrhythmia, exercise-induced ventricular arrhythmias, and sudden cardiac death. When treated with a compound to enhance the binding of calstabin2 to the mutant RyR2-speciἀc channel, they inhibited the channel leak, prevented cardiac arrhythmias, and increased the seizure threshold. It was concluded that catecholaminergic polymorphic ventricular tachycardia–associated mutant leaky Ryr2-R2474S channels in the brain can cause seizures in mice, independent of cardiac arrhythmias. Lehnart et al. (2008) propose that catecholaminergic polymorphic ventricular tachycardia is a combined neurocardiac disorder in which leaky RyR2 channels in the brain cause epilepsy and the same leaky channels in the heart cause exercise-induced sudden cardiac death. Finally, omega-3 polyunsaturated fatty acid (omega-3 fatty acids) dietary supplements appear to reduce sudden death after myocardial infarction (Dujardin et al. 2008). In a Langendorff-perfused rabbit heart preparation, the proarrhythmic effects of dofetilideinduced ventricular arrhythmias were examined. Hearts were obtained from animals pretreated with omega-3 fatty acids and compared with data obtained in control rabbits. Torsades de pointes occurred in ἀve of six control hearts and none were observed in the pretreated hearts. Dietary omega-3 fatty acids markedly reduced dofetilide-induced instability and dispersion of the cardiac action potential. Ultrafast sodium channel block by docosahexaenoic acid may explain the antiarrhythmic protection of dietary supplements of omega-3 fatty acids. Scorza et al (2008a) emphasized the value of omega-3 fatty acids to protect the heart. Lathers et al. (2008b) also discussed the possible beneἀt of dietary supplemental omega-3 fatty acids to reduce the incidence of sudden death in persons with epilepsy. It is well known that the beneἀcial effect of nutritional aspects of omega-3-fatty acids may decrease cardiac arrhythmias and sudden death in patients with cardiac disease.

782 Sudden Death in Epilepsy: Forensic and Clinical Issues

Does a nutritional deἀciency in omega-3 fatty acids become a risk factor for SUDEP? There is no answer for the question at this time. Whether omega-3 fatty acids deἀciency is a risk factor for the relatively young population prone to SUDEP is unknown and studies designed to address this question have yet to be done. The beneἀcial cardiovascular effect of omega-3 fatty acids in decreasing sudden death in cardiac patients may also be due, in part, to actions on cholesterol levels. Will this cardioprotective effect, independent of the effect on cholesterol levels, be of beneἀt for patients with epilepsy? Although at this time there is certainly nothing negative to be said for including omega-3 fatty acid–containing ἀsh in one’s diet to gain the nutritional beneἀts, Scorza et al. (2008b) emphasized that nutritional therapy, including omega-3 fatty acid supplementation, should never be a substitute for anticonvulsant medications. Collectively, all of the human and animal studies discussed above emphasize the importance of the clinical pharmacology question: “Are antiepileptic drugs a beneἀt and/ or a risk in sudden unexpected death in epilepsy?” (Lathers and Schraeder 2002). Clearly, each patient must be continually evaluated for his/her response to a given antiepileptic drug and/or combination of antiepileptic drugs. Personalized therapeutic interventions in persons with epilepsy are important to prevent unwanted antiepileptic drug side effects while optimally addressing the need to control the unwanted symptoms of epilepsy to minimize the risk of SUDEP.

References Aldenkamp, A. P. 2001. Cognitive and behavioral assessment in clinical trials: When should they be done? Epilepsy Res 45: 155–157. Aldenkamp, A. P. 1997. Effect of seizures and epileptiform discharges on cognitive function. Epilepsia 38 (S1): S52–S55. Aldenkamp, A., and J. Arends. 2004. The relative influence of epileptic EEG discharges, short nonconvulsive seizures, and type of epilepsy on cognitive function. Epilepsia 45: 54–63. Aldenkamp, A. P., J. Beitler, J. Arends, I. van der Linden, and L. Diepman. 2005. Acute effects of subclinical epileptiform EEG discharges on cognitive activation. Funct Neurol 20: 23–28. Aldenkamp, A. P., M. DeKrom, and R. Reijs. 2003. Newer antiepileptic drugs and cognitive issues. Epilepsia 44 (S4): 21–29. Aldenkamp, A. P., and N. Bodde. 2005. Behavior, cognition and epilepsy. Acta Neurol Scand Suppl 182: 19–25. Andermann, F., M. S. Duh, A. Goselin, and P. E. Paradis. 2007. Compulsory generic switching of antiepileptic drugs: High switchback rates to branded compounds compared with other drug classes. Epilepsia 48: 464–469. Austin, J. K., and D. W. Dunn. 2002. Progressive behavioral changes in children with epilepsy. Prog Brain Res 135: 419–427. Bachman, K. A., and R, J. Belloto. 1999. Differential kinetics of phenytoin in elderly patients. Drugs Aging 15: 235–250. Backiner, O., M. E. Ertorer, F. E. Haydardedeogiu, E. Bozkirli, N. B. Tutuncu, and N. G. Demirag. 2008. Subclinical hypothyroidism is characterized by increase QT interval dispersion among women. Med Princ Pract 17: 390–394. Besag, F. M. 1995. The therapeutic dilemma: Treating subtle seizures or indulging in electroencephalogram cosmetics? Semin Pediatr Neurol 2: 261–268. Beydoun, A. B., M. Uthman, A. R. Kugler, M. J. Greiner, L. E. Knapp, and E. A. Garofalo. 2005. Pregabalin 1008-009 Study Group. Safety and efficacy of two pregabalin regiments for add-on treatment or partial epilepsy. Neurology 64: 475–480.

Antiepileptic Drugs Benefit/Risk Clinical Pharmacology

783

Binnie, C. D. 2003. Cognitive impairment during epileptiform discharges: Is it ever justiἀable to treat the EEG? Lancet Neurol 2: 725–730. Binnie, C. D., and D. Marston. 1992. Cognitive correlates of interictal discharges. Epilepsia 33 (S6): S11–S17. Brown, S. 2006. Deterioration. Epilepsia 47 (S2): 19–23. Brunbech, L., and A. Sabers. 2002. Effect of antiepileptic drugs on cognitive function in individuals with epilepsy: A comparative review of newer versus older agents. Drugs 62: 593–604. De la Fuente, S., I. M. Van Langen, A. V. Postma, H. Bikker, and A. Meijer. 2008. A case of catecholaminergic polymorphic ventricular tachycardia caused by two calsequestrin 2 mutations. Pacing Clin Electgrophysiol 31: 916–919. Deng, P. Y., J. E. Porter, H. S. Shin, and S. Lei. 2006. Thyrotropin-releasing hormone increases GABA release in rat hippocampus. J Physiol 577 (Pt2): 497–511. Devinsky, O., K. Perrine, and W. H. Theodore. 1994. Interictal autonomic nervous system function in patients with epilepsy. Epilepsia 35: 199–204. Dixon, R., S. Job, R. Oliver et al. 2008. Lamotrigine does not prolong QTc in a thorough QT/QTc study in healthy subjects. Br J Clin Pharmacol 66: 396–404. Dodd-O, J., and C. M. Lathers. 1990. A characterization of the lockstep phenomenon in phenobarbitalÂ�pretreated cats. In Epilepsy and Sudden Death, Chapter 13, 199–220. ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Dong, X., I. E. Leppik, J. White, and J. Rarick. 2005. Hyponatremia from oxcarbazepine and carbaÂ� mazepine. Neurology 65: 1976–1978. Dooley, M., and G. L. Plosker. 2000. Levetiracetam. A review of its adjunctive use in the management of partial onset seizures. Drug 60: 871–893. Drane, D. L., and K. J. Meador. 1996. Epilepsy, anticonvulsant drugs and cognition. Baillieres Clin Neurol 5: 877–885. Druschky, A., M. J. Hilz, P. Hopp et al. 2001. Interictal cardiac autonomic dysfunction in temporal lobe epilepsy demonstrated by [(123)I]metaiodobenzylguanidine-SPECT. Brain 124 (Pt 12): 2372–2382. Dujardin, K. S., B. Dumotier, M. David, M. Guizy, C. Valenzuela, and L. M. Hondeghem. 2008. Ultra fast sodium channel block by dietary ἀsh oil prevents dofetilide-induced ventricular arrhythmias in rabbit hearts. Am J Physiol Heart Circ Physiol 295 (4): H1414–H1421. iEpub 2008 Aug 1. Duncan, S. 2007. Teratogenesis of sodium valproate. Curr Opin Neurol 20: 175–180. Engel, J., R. Bandler, N. C. Griffith, and S. Caldecott-Hazard. 1991. Neurobiological evidence for epilepsy-induced interictal disturbances. Adv Neurol 55: 97–111. Ficker, D. M. 2000. Sudden unexplained death and injury in epilepsy. Epilepsia 41 (Suppl2): S7–S12. Ficker, D. M., M. Privitera, G. Krauss, A. Kanner, J. L. Moore, and T. Glauser. 2005. Improved tolerability and efficacy in epilepsy patients with extended-release carbamazepine. Neurology 65: 593–595. Fukuoka, N. 2004. Influence of coadministered antiepileptic drugs on serum antiepileptic drug concentrations in epileptic patients—Quantitative analysis based on suitable transforming factor. Yakugaku Zasshi 124: 443–450. Fukuoka, N., T. Tsukamoto, J. Uno, M. Kimura, and S. Morita. 2003a. Effects of concomitant antiepileptic drugs on serum carbamazepine concentration in epileptic patients: Quantitative analysis based on extracellular water volume as a transforming factor. Yakugaku Zasshi 123: 35–42. Fukuoka, N., T. Tsukamoto, J. Uno, M. Kimura, and S. Morita. 2003b. Influence of coadministered antiepileptic drugs on serum zonisamide concentrations in epileptic patients: Quantitative analysis based on suitable transforming factor. Biol Pharm Bull 26: 1734–1738. Garaizar, C. 2000. Sudden unexpected and unexplained death in epilepsy. Rev Neurol 31: 436–441. George, J. R., and G. G. Davis. 1998. Comparison of antiepileptic drug levels in different cases of sudden death. J Forensic Sci 43: 595–603. Gilman, J. T., L. A. Alvarez, and M. Duchowny. 1993. Carbamazepine toxicity resulting from generic substitution. Neurology 43: 2696–2697.

784 Sudden Death in Epilepsy: Forensic and Clinical Issues Gissel, T., C. S. Poulsen, and P. Vestergaard. 2007. Adverse effects of antiepileptic drugs on bone mineral density in children. Expert Opin Drug Saf 6: 267–278. Glauser, T., E. Ben-Menachem, B. Bourgeois et al. 2006. International League against Epilepsy (ILAE). Treatment guidelines: Evidence-based analysis of antiepileptic drug efficacy and effectiveness as initial monotherapy for epileptic seizures and syndromes. Epilepsia 47: 1094–1112. Gomer, B., K. Wagner, L. Frings et al. 2007. The influence of antiepileptic drugs on cognition: A comparison of levetiracetam with topiramate. Epilepsy Behav 10: 486–494. Graves, N. M., G. B. Holmes, and I. E. Leppik. 1988. Compliant populations: Variability in serum concentrations. Epilepsy Res Suppl 1: 91–99. Hamandi, K., and J. W. Sander. 2006. Pregabalin: A new antiepileptic drug for refractory epilepsy. Seizure 15: 73–78. Harden, C. L., and I. Leppik. 2006. Optimizing therapy of seizures in women who use oral contraceptives. Neurology 67 (12 S 4): S56–S58. Harnod, T., C. C. Yang, Y. L. Hsin, P. J. Wang, K. R. Shieh and T. B. Kuo. 2009. Heart rate variability in patients with frontal lobe epilepsy. Seizure 18: 21–25. Hilz, M. J., G. Platsch, K. Druschky et al. 2003. Outcome of epilepsy surgery correlates with sympathetic modulation and neuroimaging of the heart. J Neurol Sci 15;21 (6): 153–162. Heaney, D. C., and J. W. Sander. 2007. Antiepileptic drugs: Generic versus branded treatments. Lancet Neurol 6: 465–468. Herold, T. J. 2006. Lamotrigine as a possible cause of QRS prolongation in a patient with known seizure disorder. CJEM 8: 361–364. Hoch, D. B., R. A. Hill, and K. H. Oas. 1994. Epilepsy and mental decline. Neurol Clin 12: 101–113. Holmes, G. L., and P. P. Lenck-Santini. 2006. Role of interictal epileptiform abnormalities in cognitive impairment. Epilepsy Behav 8: 504–514. Ide, A., and Y. Kamijo. 2007. Intermittent complete atrioventricular block after long term low dose carbamazepine therapy with a serum concentration less than the therapeutic level. Intern Med 46: 627–629. Isojarvi, J. I., A. J. Parkainen, and V. V. Myllyla. 1992. Thyroid function with antiepileptic drugs. Epilepsia 33: 142–148. Jay, G. W., and J. E. Leestma. 1981. Sudden death in epilepsy. A comprehensive review of the literature and proposed mechanisms. Acta Neurol Scand Suppl 82: 1–66. Jehi, L., and I. M. Namj. 2008. Sudden unexpected death in epilepsy: Impact, mechanisms, and prevention. Cleve Clin J Med 75 S2: S66–S70. Jobst, B. C., and G. L. Holmes. 2004. Prescribing antiepileptic drugs: Should patients be switched on the basis of cost? CNS Drugs 18: 617–628. Johannessen, S. I., and T. Tomson. 2006. Pharmacokinetic variability of newer antiepileptic drugs: When is monitoring needed? Clin Pharmacokinet 45: 1061–1075. Kasteleijn-Nolst Trenite D. G., D. J. Bakker, C. D. Binnie, A. Buerman and M. Van Raaij. 1988. Psychological effects of subclinical epileptiform EEG discharges: I. Scholastic skills. Epilepsy Res 2: 111–116. Kasteleijn-Nolst Trenite, D. G., and R. Vermeiren. 2005. The impact of subclinical epileptiform discharges on complex tasks and cognition: Relevance for aircrew and air traffic controllers. Epilepsy Behav 6: 31–34. Kelly, K., L. J. Stephen, G. J. Sills, and M. J. Brodie. 2002. Topiramate in patients with learning disability and refractory epilepsy. Epilepsia 43: 399–402. Kenneback, G., M. Ericson, T. Tomson, and L. Bergfeldt. 1997. Changes in arrhythmia proἀle and heart rate variability during abrupt withdrawal of antiepileptic drugs. Implications for sudden death. Seizure 6: 369–375. Kloster, R., H. C. Borresen, and P. Hoff-Olsen. 1998. Sudden death in two patients with epilepsy and the syndrome of inappropriate antidiuretic hormone secretion (SIADH). Seizure 7: 419–420.

Antiepileptic Drugs Benefit/Risk Clinical Pharmacology

785

Knudsen, J. F., and G. Sokol. 2006. Correlates of neurocognitive changes: Role of carbonic anhydrase. J Clin Pharmacol Annual meeting abstracts. Kockelmann, E., C. E. Elger, and C. Helmstaedter. 2003. Signiἀcant improvement in frontal lobe associated neuropsychological functions after withdrawal of topiramate in epilepsy patients. Epilepsy Res 54: 171–178. Kramer, G., A. Biraben, M. Carreno et al. 2007. Current approaches to the use of generic antiepileptic drugs. Epilepsy Behav 11: 46–52. Kuz, G. M., and A. Manssourian. 2005. Carbamapepine-induced hyponatremia: Assessment of risk factors. Ann Pharmacother 39: 1943–1946. Lathers, C. M. 1981. Models for studying sequelae to “Induced myocardial infarction.” In Mammalian Models for Research on Aging, 224–228. Washington, DC: National Academy Press. Lathers, C. M., R. F. Flax, and L. J. Lipka. 1986. The effect of C1 spinal cord transaction or bilateral adrenal vein ligation on thioridazine-induced arrhythmia and death in the cat. J Clin Pharmacol 26: 515–523. Lathers, C. M., and J. E. Leestma. 1990. Psychoactive agents, epilepsy, arrhythmia, and sudden death. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, Chapter 23, 447–484. New York, NY: Marcel Dekker. Lathers, C. M., R. M. Levin, and W. H. Spivey. 1986. Regional distribution of myocardial betaÂ�adrenoceptors in the cat. Eur J Pharmacol 130: 111–117. Lathers, C. M., and L. J. Lipka. 1987. Cardiac arrhythmia, sudden death, and psychoactive agents. J Clin Pharmacol 27: 1–14. Lathers, C. M., and L. J. Lipka. 1986. Chlorpromazine: Cardiac arrhythmogenicity in the cat. Life Sci 38: 521–538. Lathers, C. M., and P. L. Schraeder. 1990. Arrhythmias associated with epileptogenic activity elicited by penicillin. In Epilepsy and Sudden Death, ed. C. M Lathers and P. L. Schraeder, Chapter 10. New York, NY: Marcel Dekker. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23: 633–647. Lathers, C. M., and P. L. Schraeder. 2002. Clinical pharmacology: Drugs as a beneἀt and/or risk in sudden unexpected death in epilepsy? J Clin Pharmacol 42: 123–136. Lathers, C. M., and P. L. Schraeder. 1990. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Lathers, C. M., and P. L. Schraeder. 1987. Review of autonomic dysfunction, cardiac arrhythmias, and epileptogenic activity. J Clin Pharmacol 27: 346–356. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9: 236–242. Lathers, C. M., P. L. Schraeder, and M. Bungo. 2008a. Mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12: 3–24. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008b. Reply. Unanswered questions: SUDEP studies needed. Epilepsy Behav 13: 265–269. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008c. Neurocardiologic mystery. In Psychological Factors and Cardiovascular Disorders: Role of Stress and Psychosocial Influences, ed. Leo Sher, Chapter 13. New York, NY: Nova Science Publishers, Inc. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008d. Sudden death: Neurocardiologic mystery. In Psychological Factors and Cardiovascular Disorders, ed. L. Sher. New York, NY: Nova Biomedical Books. Lathers, C. M., P. L. Schraeder, and H. G. Claycamp. 2003. Clinical pharmacology of topiramate versus lamotrigine versus Phenobarbital: Comparison of efficacy and side effects using odds ratios. J Clin Pharmacol 43: 491–503. Lathers, C. M., P. L. Schraeder, and N. Tumer. 1993. The effect of phenobarbital on autonomic function and epileptogenic activity induced by the hippocampal injection of penicillin in cats. J Clin Pharmacol 33: 837–844.

786 Sudden Death in Epilepsy: Forensic and Clinical Issues Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67: 247–259. Lathers, C. M., W. H. Spivey, and R. M. Levin. 1988. The effect of chronic timolol in an animal model for myocardial infarction. J Clin Pharmacol 28: 736–745. Lathers, C. M., W. H. Spivey, R. M. Levin, and N. Tumer. 1990. The effect of dilevalol on cardiac autonomic neural discharge, plasma catecholamines, and myocardial beta receptor density associated with coronary occlusion. J Clin Pharmacol 30: 241–253. Lathers, C. M., W. H. Spivey, L. E. Suter, J. P. Lerner, N. Tumer, and R. M. Levin. 1986. The effect of acute and chronic administration of timolol on cardiac sympathetic neural discharge, arrhythmia, and beta adrenergic receptor density associated with coronary occlusion in the cat. Life Sci 39: 2121–2141. Lehnart, S. E., M. Mongiool, A. Bellinger et al. 2008. Leaky Ca2+ release channel/ryanodine receptor 2 causes seizures and sudden cardiac death in mice. J Clin Invest 118: 2230–2245. Lee, S., V. Sziklas, F. Andermann et al. 2003. The effects of adjunctive topiramate on cognitive function in patients with epilepsy. Epilepsia 44: 339–347. Leestma, J. E., J. F. Annegersm, M. J. Brodie et al. 1997. Sudden unexplained death in epilepsy: Observations from a large clinical development program. Epilepsia 38: 47–55. Levy, R. H., and C. Collins. 2007. Risk and predictability of drug interactions in the elderly. Int Rev Neurobiol 81: 235–251. Lhatoo, S. D., Y. Langan, and J. W. Sander. 1999. Sudden unexpected death in epilepsy. Postgrad Med J 75 (890): 706–709. Lipka, L. J., and C. M. Lathers. 1987. Psychoactive agents, seizure production, and sudden death in epilepsy. J Clin Pharmacol 27: 169–183. Lipka, L. J., C. M. Lathers, and J. Roberts. 1988. Does chlorpromazine produce cardiac arrhythmia via the central nervous system? J Clin Pharmacol 28: 968–983. Lund, A., and H. Gormsen. 1985. The role of antiepileptics in sudden death in epilepsy. Acta Neurol Scand 72: 444–446. Lutz, M. T., and C. Helmstaedter. 2005. EpiTrack: Tracking cognitive side effects of medication on attention and executive functions in patients with epilepsy. Epilepsy Behav 7: 708–714. May, T. W., B. Rambec, and U. Jurgens. 1996. Serum concentrations of lamotrigine in epileptic patients: The influence of dose and comedication. Ther Drug Monit 18: 523–531. McNaughton, N. C., C. H. Davies, and A. Randall. 2004. Inhibition of alpha(1E) Ca(2+) channels by carbonic anhydrase inhibitors. J Pharmacol Sci 95: 240–247. Meador, K. J. 2002. Cognitive outcomes and predictive factors in epilepsy. Neurology 58 (8 S5): S21–S26. Meador, K. J. 1994. Cognitive side effects of antiepileptic drugs. Can J Neurol Sci 21: s12–s16. Miller, J. M., and R. P. Kustra. 2005. Cognitive and behavioral effects of lamotrigine and topiramate in healthy volunteers. Neurology 64: 2108–2114. Mintzer, S., P. Boppana, J. Toquri, and A. DeSantis. 2006. Vitamin D levels and bone turnover in epilepsy patients taking carbamazepine or oxcarbazepine. Epilepsia 47: 1586. Nilsson, L., B. Y. Farahmand, P. G. Persson, I. Thiblin, and T. Tomson. 1999. Risk factors for sudden unexpected death in epilepsy: A case-control study. Lancet 353: 888–893. Opeskin, K., M. P. Burke, S. M. Cordner, and S. F. Berkovic. 1999. Comparison of antiepileptic drug levels in sudden unexpected deaths in epilepsy with deaths from other causes. Epilepsia 40: 1795–1798. O’Rourke, D. K., and C. M. Lathers. 1990. Interspike interval histogram characterization of synchronized cardiac sympathetic neural discharge and epileptogenic activity in the electrocorticogram of the cat. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, Chapter 15, 239–260. New York, NY: Marcel Dekker. Pack, A. M., and M. J. Morrell. 2004. Epilepsy and bone health in adults. Epilepsy Behav 5 (Suppl 2): S24–S29.

Antiepileptic Drugs Benefit/Risk Clinical Pharmacology

787

Patsalos, P. N., and J. W. Sander. 1994. Newer antiepileptic drugs. Towards an improved risk–beneἀt ratio. Drug Saf 11: 37–67. Perucca, E. 2006. Clinically relevant drug interactions with antiepileptic drugs. Br J Clin Pharmacol 61: 246–255. Physician’s Desk Reference web site, www.pdr.net. Porter, R. J., A. Partiot, R. Sachdeo, V. Nohria, and W. M. Alves. 2007. 205 Study Group. RandomÂ� ized,€ multicenter, dose-ranging trial of retigabine for partial-onset seizures. Neurology 68: 1197–1204. Ramsay, R. E., D. Q. McManus, A. Guterman et al. 1990. Carbamazepine metabolism in humans: Effect of concurrent anticonvulsant therapy. Ther Drug Monit 12: 235–241. Reingardiene, D., and J. Vilcinskaite. 2007. QTc-prolonging drugs and the risk of sudden death. Medicina (Kaunas) 43: 347–353. Ruiz Gines, M. A., S. Garcia Garcia, J. A. Ruiz Gines, E. Tze Kiong, and E. Ferrnadez Rodriquez. 2007. Symptomatic secondary hyponatremia due to combined treatment anticonvulsant and antidepressant: Risk of sudden death in epilepsy? An Med Interna 24: 335–338. Schachter, S. C. 1999. Antiepileptic drug therapy: General treatment principles and application for special patient populations. Epilepsia 40 (Suppl 9): S20–S25. Schachter, S. C. 2002. Drug-mediated antiepileptogenesis in humans. Neurology 59 (9s 5): S34–S35. Schraeder, P. L., and G. G. Celesia. 1977. The effects of epileptogenic activity on auditory evoked potentials in cats. Arch Neurol 34: 677–682. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32: 1371–1382. Schraeder, P. L., and C. M. Lathers. 1995. Clinical pharmacology of antiepileptic drug use: “Clinical pearls about the perils of Patty.” J Clin Pharmacol 35: 1120–1135. Schraeder, P. L., and C. M. Lathers. 1989. Paroxysmal autonomic dysfunction, epileptogenic activity and sudden death. Epilepsy Res 3: 55–62. Scorza, F. A., R. M. Arida, and E. A. Cavalheiro. 2008a. Preventive measures for sudden cardiac death in epilepsy beyond therapies. Epilepsy Behav 13: 263–264. Scorza, F. A., R. M. Cysneiros, R. M. Arida, V. C. Terra-Bustamante, M. de Albuquerque, and E. A. Cavalheiro. 2008b. The other side of the coin: Beneἀciary effect of omega-3 fatty acids in sudden unexpected death in epilepsy. Epilepsy Behav 13 (2): 279–283. Shinoda, M., M. Akita, M. Hasegawa, T. Hasegawa, and T. Nabeshima. 1996. The necessity of adjusting the dosage of zonisamide when coadministered with other anti-epileptic drugs. Biol Pharm Bull 19: 1090–1092. Sirsi, D., and J. E. Safdieh. 2007. The safety of levetiracetam. Expert Opin Drug Saf 6: 241–250. Stauffer, A. Z., J. Dodd-O, and C. M. Lathers. 1989. The relationship of the lock-step phenomenon and precipitous changes in mean arterial blood pressure. Electroencephalogr Clin Neurophysiol 72: 340–345. Stauffer, A. Z., J. Dodd-O, and C. M. Lathers. 1990. The relationship of the lock-step phenomenon and precipitous changes in mean arterial blood pressure. In Epilepsy and Sudden Death, ed. C.€M. Lathers and P. L. Schraeder, Chapter 14, 221–238. New York, NY: Marcel Dekker. Stephen, L. J. 2003. Drug treatment of epilepsy in elderly people: Focus on valproic acid. Drugs Aging 20: 141–152. Steinhoff, B. J. 2008. Pregnancy, epilepsy, and anticonvulsants. Dialogues Clin Neurosci 10: 63–75. Vainionpaa, L. K., K. Mikkonen, J. Rattya et al. 2004. Thyroid function in girls with epilepsy with carbamazepine, oxcarbazepine, or valproate monotherapy and after withdrawal of medication. Epilepsia 45: 197–203. Tennis, P., T. B. Cole, J. F. Annegers, J. E. Leestma, M. McNutt, and A. Rajput. 1995. Cohort study of incidence of sudden unexplained death in persons with seizure disorder treated with antiepileptic drugs in Saskatchewan, Canada. Epilepsia 36: 29–36. Timmings, P. L. 1998. Sudden unexpected death in epilepsy: Is carbamazepine implicated? Seizure 7: 289–291.

788 Sudden Death in Epilepsy: Forensic and Clinical Issues Timmings, P. L. 1993. Sudden unexpected death in epilepsy: A local audit. Seizure 2: 287–290. Tomson, T., A. C. Sköld, P. Holmgen, L. Nilsson, and B. Danielsson. 1998. Postmortem changes in blood concentrations of phenytoin and carbamazepine: An experimental study. Ther Drug Monit 20: 309–312. Toth, V., L. Hejjel, Z. Kalmar et al. 2008. Effect of epileptic seizures on the heart rate. Ideggyogy Sz 61: 155–161. Tremmel, L., M. Holtmann, M. H. Schmidt, and U. Brandl. 2006. Do subclinical epileptiform discharges really affect short-term memory in children? Z Kinder Jugendpsychiatr Psychother 34: 139–148. Vainionpää, L. K., K. Mikkonen, J. Rättyä et al. 2004. Thyroid function in girls with epilepsy with carbamazepine, oxcarbazepine, or valproate monotherapy and after withdrawal of medication. Epilepsia 45: 197–203. Warner, G., and D. P. Figgitt. 2005. Pregabalin: As adjunctive treatment of partial seizures. CNS Drugs 19: 265–272. Wilby, J., A. Kainth, N. Hawkins et al. 2005. Clinical effectiveness, tolerability and cost-effectiveness of newer drugs for epilepsy in adults: A systematic review and economic evaluation. Health Technol Assess 9: 1–157, iii–iv.

Clinical Pharmacology and SUDEP Claire M. Lathers Paul L. Schraeder

48

Contents 48.1 Clinical Problem of Sudden Death in Persons with Epilepsy 48.2 Preclinical Animal Studies Leading to Clinical Studies in Persons with€Epilepsy 48.2.1 Animal Models of Ouabain-Induced or Coronary Occlusion–Induced Arrhythmia and Death 48.3 Clinical Studies in Persons with Epilepsy 48.4 Criteria for Sudden Unexplained Death in Epileptic Persons 48.5 Incidence of Sudden Unexpected Death in Epileptic Persons 48.6 Risk Factors for SUDEP 48.6.1 Low Drug Levels due to Variations in Bioavailability, Drug Metabolism, and/or Compliance? 48.6.2 Patients with Epilepsy Who Do Not Receive Antiepileptic Drug due to a Dispensing Error 48.7 Conclusions References

789 790 790 791 792 793 794 794 795 796 797

48.1â•… Clinical Problem of Sudden Death in Persons with Epilepsy Kuller (1996) and Kuller and Lilienfeld (1996) deἀned the disease process and percentage of cases of sudden and unexpected death in a general forensic autopsy population as follows: heart and/or aorta, 56.1%; respiratory, 14.5%; brain and meninges, 15.8%; digestive/urogenital, 8%; and miscellaneous, 9.5%. Rodin (1968) has stated that life ends earlier, in general, for the person with epilepsy. The mortality ratio is two to three times that found in the general population (Hauser et al. 1980; Kurtzke 1972). As summarized by Wannamaker (1990), death may be the consequence of natural or unnatural causes, such as accidents, homicide, and suicide, that have no relationship to epilepsy. Direct causes of death include status epilepticus, and indirect causes may be head trauma or drowning subsequent to a seizure. When death occurs suddenly and without explanation, the term sudden unexpected unexplained death is used. Sudden is deἀned as death occurring within 1 h. Leestma (1990) has discussed additional time frames as deἀnitions of sudden because forensic pathologists prefer more precise limits such as less than 1 h, less than 2 h, less than 12 h, and so on. Unexpected notes that death was not imminent. To date, unexpected implies that there are no symptoms or antecedent illness that would predict that death may be imminent. Most persons with epilepsy have many seizures without lethal outcome, and thus it is difficult for the neurologist to ascribe the unexpected, unexplained death of 789

790 Sudden Death in Epilepsy: Forensic and Clinical Issues

a patient with epilepsy to a single convulsion. Unexplained is a term that clinicians and research scientists are working to clarify.

48.2â•…Preclinical Animal Studies Leading to Clinical Studies in Persons with Epilepsy 48.2.1â•…A nimal Models of Ouabain-Induced or Coronary Occlusion–Induced Arrhythmia and Death In 1964, Han and Moe established that sympathetic nerve stimulation, ouabain toxicity, or coronary occlusion increased temporal dispersion of recovery of ventricular excitability and concluded that this led to an underlying electrical instability that predisposed the ventricular myocardium to arrhythmia. Lathers et al. (1974, 1977, 1978) ἀrst reported cardiac arrhythmias in an animal model for ouabain-induced toxicity that were associated with neural autonomic dysfunction. All studies established that the neural discharges were characterized by increases, decreases, or no change in the discharge of postganglionic cardiac sympathetic nerves monitored simultaneously in the same cat. Lathers and Schraeder (1982) concluded that the peripheral neural nonuniform sympathetic discharge initiated cardiac arrhythmia in the manner described by Han and Moe (1964). The conclusion was also supported by the ἀnding of Randall et al. (1968) that stimulation of the sympathetic ventrolateral cardiac nerve produced a shift in the origin of the pacemaker and tachyarrhythmias. This ἀnding was attributed to the anatomical observation that the nerve is not uniformly distributed to the various regions of the heart but is speciἀcally localized to the atrioventricular junctional and ventricular regions. This nonuniform distribution of sympathetic nerves would also contribute to initiation of arrhythmia if a nonuniform neural discharge occurred regardless of the underlying cause. Lathers et al. (1975) extended the ἀndings of the ouabain-induced nonuniform neural discharge associated with the arrhythmia model to an animal model of arrhythmia and sudden coronary death and reported that the nonuniform autonomic neural discharge was also associated with the occurrence of cardiac arrhythmias and/or sudden death induced by coronary occlusion demonstrating that differences in the ability of different digitalis glycosides to alter neural discharge existed. For example, the digitalis glycoside digoxin did not initiate a nonuniform neural discharge. Some pharmacological agents were found to modify the neural nonuniform discharge and arrhythmogenic activity of the heart, whereas others did not exert a beneἀcial effect. The beta-blocking agents practolol and solotalol (Lathers 1975), metoprolol, timolol (Lathers 1980a, 1980b), quinidine (Lathers et al. 1977), and procainamide (Lathers et al. 1976) did modify the nonuniform neural discharge. Pharmacological agents such as lidocaine (Lathers et al. 1977) and methylprednisolone (Lathers 1983) likewise did not exert an effect on the nonuniform neural discharge and/or the times to arrhythmia and death. Studying the animal model of ouabain-induced toxicity clearly established that appropriate doses of ouabain have a beneἀcial antiarrhythmic effect, whereas higher doses pose the risk of ouabain toxicity–induced arrhythmia and/or death. Just as differences exist in the ability of different glycosides to trigger nonuniform neural discharge, arrhythmia, and death (e.g., the polar molecule ouabain does and the nonpolar digoxin does not), one may theorize that differences exist in the changes associated with various diseases such as myocardial infarction or epilepsy or with different types of epileptogenic disorders. One

Clinical Pharmacology and SUDEP

791

could theorize that in an epileptic patient during a given aberrant epileptogenic discharge, the extent of the contribution of the nonuniform peripheral sympathetic discharge may vary so that during some events, the epilepsy patient may exhibit an arrhythmia, yet not continue on to a fatal event. The same patient may be at greater risk for sudden death in persons with epilepsy (SUDEP) during another event characterized by an efferent pattern of peripheral neural nonuniform discharge that is predisposed with a combination of other circumstances to actually elicit a “fatal event.” It is of interest, when considering persons with epileptogenic activity, to theorize that differences may exist in the status of the autonomic nervous system and its contribution to arrhythmia and/or death, just as differences were found in the effects of two digitalis glycosides, ouabain and digoxin. Furthermore, it may also be speculated that a given anticonvulsant drug at a given dose may be more beneἀcial in preventing SUDEP, an effect that perhaps may be determined in part by its interaction with the contribution of the autonomic nervous system at any given point in time. Thus, at some times, a potentially fatal event may be fortunately avoided, whereas at other times SUDEP may occur. The conclusions of all of the studies conducted in cats were reviewed in detail by Schraeder and Lathers (1989) and Lathers and Schraeder (1987). Lathers and Schraeder (1990) published the ἀrst book focusing on the clinical problem of epilepsy and sudden death. The ἀrst half of the book described the animal studies investigating possible mechanisms of SUDEP, whereas the second half of the book described the state of current clinical knowledge of SUDEP. All contributing authors acknowledged the paucity of clinical data addressing the mechanism of death. Bigger (1990) wrote in his forward to the book: “The current volume assembles a spectrum of international experts who discuss the latest experimental and clinical information available about sudden death and epilepsy. The contributors all struggle for insight into this problem, which resembles the ἀeld of sudden cardiac death 20 years ago while we seem a long way from understanding and controlling the problem of sudden death in epilepsy, the primary question is how shall we proceed toward those goals? Additional epidemiological inquiry should sharpen the focus on the high risk groups. We are on the threshold of major advances in addressing the problem of sudden death in epilepsy. The tools to advance our knowledge are at hand and we should energetically put them to use for the future beneἀt of patients with epilepsy. Knowledge gained from studies of sudden death in epileptics will very likely be useful for understanding sudden death in other situations as well.” To date, unlike the walls of Constantinople, the “threshold of major advances” has not been breached.

48.3â•…Clinical Studies in Persons with Epilepsy By the convening of a panel of experts to examine the question of what product labeling should be requested of the pharmaceutical industry, the FDA focused attention of practitioners and pharmaceutical manufacturers on the question of whether use of anticonvulsant drugs contributes to or prevents SUDEP. Consideration of the possibility of sudden unexpected death in epileptic persons when developing new anticonvulsant drugs was emphasized by the FDA-convened panel of scientists assigned to review data on the risk of sudden unexpected death in epileptic persons taking lamotrigine (Leestma et al. 1997). Other “new” anticonvulsant drugs, for example, gabapentin, topiramate, tigabine,

792 Sudden Death in Epilepsy: Forensic and Clinical Issues

and zonisamide, also have FDA-required warning labels with data on the risk of SUDEP in association with use of each drug. Another long-term beneἀcial effect of bringing SUDEP to the attention of epilepsy researchers has been the development of epidemiological studies devoted to this phenomenon. The original data from the Cook County coroner’s office have been supplemented by numerous other studies in various populations and have mostly conἀrmed the conclusions that SUDEP is a common cause of death in persons with epilepsy, with the risk increasing with increasing refractoriness of the seizure disorder. Some of the most recent studies found that the risk of sudden death in epilepsy was 24 times greater than in the general population (Ficker et al. 1998). The overall incident rate is 1:680/year but varies to some degree with the severity of seizure disorder with an incidence of 1:100 in populations most refractory to treatment (Langan et al. 1998). These epidemiological data conἀrm the conclusion that SUDEP is a risk facing any person with epilepsy.

48.4â•…Criteria for Sudden Unexplained Death in Epileptic Persons To avoid the problem of the lack of a standardized deἀnition of SUDEP associated with previous studies, Leestma et al. (1997) developed and then applied an algorithm for SUDEP in 1997. The criteria for SUDEP (deἀnite or highly probable) were as follows: 1. The subject had epilepsy, as deἀned by Gastaut (1973) and the World Health Organization (WHO): “a chronic disorder characterized by recurrent seizures due to excessive discharge of cerebral neurons.” Because all patients had had chronic and usually intractable epilepsy (according to their physicians who had prescribed one or more antiepileptic drugs for many years for their patients’ seizures), it was assumed that Gastaut/WHO criteria were met. 2. The subject died unexpectedly while in a reasonable state of health. 3. The fatal attack occurred suddenly. The complexity of this deἀnition was discussed. It was recognized that the ἀnal ictus must occur precipitously and unexpectedly but that death might not occur for several hours. Death may have occurred presumably from a seizure-associated cardiorespiratory arrest and its complications and not from status epilepticus. Sudden collapse and death may also have occurred without an observable seizure. 4. The death occurred during normal activities (e.g., at work, at home, in or around bed) in benign circumstances. 5. An obvious medical cause of death was not found. (An exception would be the presence of sudden cardiac arrhythmia, which may be related to the mechanism of SUDEP. Death in water if the victim does not show evidence of drowning may also be attributable to SUDEP.) 6. SUDEP was excluded in the presence of status epilepticus or acute trauma in the setting of a seizure. The classiἀcation of “possible SUDEP” was assigned when cases met most or all of these criteria for SUDEP but data suggested more than one possible cause of death [e.g., deaths associated with seizures while in the bath or swimming, or aspiration (conἀrmed or suspected) occurring concurrently with a seizure]. It was recognized that the classiἀcation of

Clinical Pharmacology and SUDEP

793

drowning deaths is difficult, and unless such deaths were witnessed or other information was available, a level of indeterminateness of classiἀcation was appropriate. The “other (non-SUDEP)” classiἀcation was assigned to cases in which the criteria for SUDEP (deἀnite or highly probable) or possible SUDEP were not met and an obvious other cause of death had been established. The classiἀcation “insufficient data” was assigned to cases that could not be properly interpreted because of missing or ambiguous data related to the circumstances of death or concurrent medical condition.

48.5â•…Incidence of Sudden Unexpected Death in Epileptic Persons Leestma et al. (1997) emphasized that even when using a well-deἀned population, confounding factors such as differences in the dose, duration of therapy, and concomitant medications must be addressed. For example, valproic acid was allowed as a concomitant medication in the UK-sponsored lamotrigine trials but was not allowed in those conducted in the United States. Although the resulting data indicated a lower SUDEP per patient-year of treatment in the lamotrigine database in the US clinical studies than that reported for the United Kingdom, it was not signiἀcantly different. Data obtained from those patients dying of SUDEP were compared with similarly treated patients who died of non-SUDEP causes. This analysis allowed assessment of how the data from these groups compared with data from the overall population. Males represented 52% of the total lamotrigine trial population, and men represented almost 75% of the SUDEP and possible SUDEP patients, suggesting that this observation might be of little signiἀcance because 63% of the non-SUDEP deaths also occurred in males. No differences were detected between the SUDEP group and the overall trial population in terms of the number of antiepileptic drugs or duration of epilepsy. No evidence was found to implicate high doses of lamotrigine or longer duration of exposure with any increase in the risk of SUDEP. Estimated SUDEP rates ranged from 9 in 1000 patient-years. Most studies investigated SUDEP rates in the general epilepsy population rather than in a population with severe epilepsy. Severe or intractable epilepsy (Tennis et al. 1995) as well as age in the 20- to 40-year range has been suggested to be one of the risk factors for SUDEP. Most patients in the lamotrigine clinical development program were categorized in the highrisk group for SUDEP. By deἀnition, they were refractory to currently available pharmacological treatments and exhibited more severe epilepsy than the general population of patients with epilepsy, with a mean of 6.6 seizures/week at baseline in controlled trials while receiving antiepileptic drug polytherapy. When compared to patients in the literature, the derived sudden death rates for patients in the lamotrigine clinical trial program compared favorably with the rate expected in young adults with severe epilepsy and with that reported in the gabapentin clinical trial population (Glaxowellcome 2000). Because the sudden death rate in the lamotrigine and gabapentin trials was low, even in patients with refractory epilepsy, the power to detect a relation between the severity of epilepsy, treatment, and a rare but catastrophic event of SUDEP was limited. Leestma et al. (1997) concluded that the rate of SUDEP in the lamotrigine trials was within the range of rates expected for this patient population. The rate of SUDEP did not appear to be affected by either lamotrigine dose or duration of therapy with lamotrigine. No evidence was found to suggest that lamotrigine alters the risk of SUDEP in patients

794 Sudden Death in Epilepsy: Forensic and Clinical Issues

with epilepsy. Fewer patients treated with lamotrigine died of SUDEP as compared with those treated with placebo. The lamotrigine SUDEP rate was found to be similar to that reported for a similar population of patients treated with gabapentin. This ἀnding suggests the need for additional investigation into the effect new antiepileptic drugs on the risk of SUDEP.

48.6â•…Risk Factors for SUDEP Wren et al. (2000) identiἀed the incidence, causes, and characteristics of sudden death in individuals ranging from 1 to 20 years of age. Their review of all deaths required obtaining death certiἀcates and additional information, where appropriate, from coroners, pediatricians, physicians, and pathologists. The study examined the resident population of one English health region from 1985 to 1994. It showed that in a population of 806,500 persons, medical causes explained 1017 (40%) deaths, 1236 (49%) were unnatural, and 270 (11%) were sudden. Of the sudden deaths, 142 had a previous diagnosis, with the most common being epilepsy for 49 patients (34%). Infections occurred in 17 (20%) and unsuspected cardiovascular abnormalities in 26 (30%). A total of 41 remained unexplained. It is unclear what proportion of the epilepsy-related deaths was attributable to SUDEP. Although no hard data are extant concerning the risk of SUDEP in persons with epilepsy, it is estimated that this phenomenon accounts for 10–15% of deaths in a general population of persons with epilepsy. In selected populations with risk factors such as mental retardation and frequent seizures, SUDEP may account for up to 22% of deaths in persons followed for many years (Chaney and Eyman 2000). Ficker (2000) concluded that risk factors for SUDEP might include poorly controlled seizures, early onset of epilepsy, and generalized tonic–clonic seizures. Of 57 SUDEP cases examined by Nilsson et al. (1999), 91% had undergone necropsy. It was found that the relative risk of SUDEP increased with the number of seizures per year. The estimated relative risk of SUDEP was 10.16 times greater in those patients with more than 50 seizures per year compared with those with up to two seizures per year. The risk also increased with increased number of antiepileptic drugs taken concomitantly, being as high as 9.89 with the use of three antiepileptic drugs compared with monotherapy. Other major risk factors were early-onset vs. late-onset epilepsy and frequent changes of antiepileptic drug dosage when compared with patients on an unchanged dosage. The associations between SUDEP risk and early onset and SUDEP risk and seizure frequency were weaker for female than for male patients, whereas frequent dose changes showed a stronger association in female patients. The authors suggest improvement of seizure control, and an attempt to avoid polytherapy may reduce the risk of SUDEP. 48.6.1â•…Low Drug Levels due to Variations in Bioavailability, Drug Metabolism, and/or Compliance? Lund and Gormsen (1985) examined the role of antiepileptics in sudden death in epilepsy and evaluated patients treated with one or more of the following anticonvulsants: phenobarbitone, phenytoin, and carbamazepine. They noted that variations in bioavailability and in drug metabolism by enzyme induction, resulting in low drug levels, might be contributing factors to the occurrence of SUDEP. They found subtherapeutic drugs levels

Clinical Pharmacology and SUDEP

795

in half of the cases and lethal concentrations, mainly of phenobarbitone, in one third of the cases. The results were interpreted to suggest that there were severe consequences of noncompliance in the treatment of epilepsy. The authors also emphasized the importance of the antiepileptic blood-drug level control as a monitor of compliance in patients with epilepsy, stressing the value of drug-blood level control. Graves et al. (1988) focused on the importance of preventing fluctuations of concomitant antiepileptic drugs in safety and efficacy trials and in patient care settings. Schachter et al. (1998) discussed observations of low postmortem blood concentrations of antiepileptic drugs in cases of SUDEP. These authors emphasized that although noncompliance may play a role in SUDEP, postmortem phenytoin concentrations should be interpreted with caution because animal data indicate that antemortem whole-blood concentrations were only 65% of the corresponding serum concentrations, and postmortem blood levels were even lower, being 35% of antemortem serum concentrations. The major implication of this study is that low postmortem concentrations do not necessarily imply poor compliance with treatment. In contrast, Opeskin et al. (1999) concluded that there were no data to support the idea that SUDEP patients were less compliant with antiepileptic drug treatment than the control group. They eliminated poor compliance with antiepileptic drug treatment as a risk factor for SUDEP. They also found no evidence that phenytoin or carbamazepine use was associated with a higher risk of SUDEP. May et al. (1999) also noted that the postmortem decrease in anticonvulsant serum concentrations, especially for phenobarbital and phenytoin, should be considered with caution to avoid misinterpretation in relation to the so-called “subtherapeutic” serum levels and noncompliance in relation to SUDEP.

48.6.2â•…Patients with Epilepsy Who Do Not Receive Antiepileptic Drug due to a Dispensing Error Clinical pharmacologists, neurologists, pharmacists, nurses, and all medical health professionals should also be aware of another issue that could contribute to the risks faced by persons with epilepsy, namely, the name of the drugs that they use. For example, the patent name for Lamictal (lamotrigine), an anticonvulsant, has been confused with Lamisil (terbinoἀne hydrochloride), an antifungal drug. Patients erroneously receiving either medication would be unnecessarily subjected to the risk of adverse events. In addition, patients with epilepsy who do not receive their antiepileptic drug because of a dispensing error would be inadequately treated and could experience serious consequences, including status epilepticus. Patients receiving Lamictal instead of Lamisil would not have the dose of Lamictal properly dispensed and would unnecessarily be subjected to a risk of potential side effects, including Stevens–Johnson syndrome. Clear communication of oral and written prescriptions will help avoid future dispensing errors. An earlier problem, related to drugs with similar names that could easily be confused when the pharmacist was dispensing, involved the antiepileptic drug clonazepam. One of the authors (PLS) had the experience many years ago of prescribing clonazepam only to ἀnd out that the patient was nearly in a coma associated with hypotension occurring when clonidine was accidentally dispensed by the pharmacist. This potential problem originated when “klonapin” was spelled “clonapin” as a brand name and was commonly confused with clonidine. This problem was simply corrected when the spelling of the name “clonapin” was changed from a “c” to a “k.”

796 Sudden Death in Epilepsy: Forensic and Clinical Issues

48.7â•… Conclusions In conclusion, the 1993 effort of the FDA (Leestma et al. 1997) subsequently focused attention of practitioners and pharmaceutical manufacturers on the question of whether use of anticonvulsant drugs contributes to or prevents sudden unexpected death in epileptic persons. Consideration of the possibility of sudden unexpected death in epileptic persons when developing new anticonvulsant drugs was emphasized by the FDA-convened panel of scientists assigned to review data on the risk of sudden unexpected death in epileptic persons in patients taking lamotrigine. The sudden unexpected death rate in persons with epilepsy in this cohort of patients was comparable to that expected in young persons with poorly controlled epilepsy. Estimated sudden unexpected death rates in patients receiving any of these newer anticonvulsant drugs are similar to those observed in patients receiving other anticonvulsant drugs. Cumulative evidence suggests that sudden unexpected death rates in epileptic persons reflect population rates and not a speciἀc drug effect. Five newer anticonvulsant drugs—lamotrigine, gabapentin, topiramate, tigabine, and zonisamide— have FDA-required warning labels with data on the risk of sudden unexpected death in epileptic persons in association with use of each drug. None of these ἀve newer anticonvulsant drugs have shown an associated increased risk of sudden unexpected death, nor did their use show a decreased risk. The status of clinical studies of SUDEP leaves much to be desired. Although some epidemiology studies have been done, many questions remain to be answered. 1. It is difficult to conduct an epidemiological study when lacking good postmortem examinations on persons with epilepsy who have died. 2. Even if postmortem studies have been done, there is a lack of standardized autopsy end points, including a lack of microscopic cardiac examinations emphasizing cardiac neural elements. 3. Most epidemiology studies of SUDEP lack a signiἀcant n to reach any deἀnitive conclusion regarding the risk factor for SUDEP. 4. Clinical pharmacological studies that determine the best use of drugs in terms of preventing SUDEP need to be designed. 5. The observation that poor seizure control is a risk factor for SUDEP should be an incentive for practitioners to work on improving patient compliance with antiepileptic drug use. 6. The fact that many other SUDEP victims have relatively low seizure frequency implies that leaving any seizure untreated places the patient at risk. Whenever a particular drug is selected to treat a given patient, consideration must be made of the fact that each pharmacological agent has the capability of providing the beneἀcial therapeutic effect if used in the correct dose range but may also produce unwanted, unexpected adverse effects (Lathers 1996; Lathers and Schraeder 1995). Some subpopulations may be more sensitive to a given drug than most of the patients. This problem is confounded in the epileptic population, especially because the individual may be at risk for sudden unexpected death from the disease state itself. Each person with epilepsy must be evaluated to determine if he or she may be at risk for SUDEP and to determine that the antiepileptic drug treatment regimen selected is the best therapy. It is essential that the drugs themselves do not place them at greater risk for SUDEP. At this time, the answer

Clinical Pharmacology and SUDEP

797

to the question of whether one or the other of the antiepileptic drugs, when used alone or in combination, have the effect of increasing or decreasing the risk of SUDEP is also unknown. One possible answer to the question posed appears to be that a given drug in a given patient with epilepsy during a given state of epileptogenic activity may be more beneἀcial or may lead to SUDEP, depending on the overall “mix” of these events during the given risk event. Good patient care is essential to provide continued regular evaluation and monitoring of each patient by his or her physician. Optimizing seizure control with antiepileptic drugs is the ἀrst line of defense in the prevention of SUDEP.

References Bigger, J. T. 1990. Foreword. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, v–vii. New York, NY: Marcel Dekker. Chaney, R. H., and R. K. Eyman. 2000. Patterns in mortality over 60 years among persons with mental retardation in a residential facility. Ment Retard 38: 289–293. Ficker, D. M. 2000. Sudden unexplained death and injury in epilepsy. Epilepsia 41 (S2): S7–S12. Ficker, D. M., E. L. So, W. K. Shen et al. 1998. Population-based study of the incidence of sudden unexplained death in epilepsy. Neurology 51: 1270–1274. Gastaut, H. 1973. Dictionary of Epilepsy. Geneva: World Health Organization. Glaxowellcome. 2000. Prescribing information for Lamictal (lamotrigine), 2000. In Physicians’ Desk Reference, 54th ed., 1208–1214. Montvale, NJ: Medical Economics. Graves, N. M., G. B. Holmes, and I. E. Leppik. 1988. Compliant populations: Variability in serum concentrations. Epilepsy Res Suppl 1: 91–99. Han, J., and G. K. Moe. 1964. Nonuniform recovery of excitability in ventricular vulnerability. Circ Res 14: 44–60. Hauser, W. A., J. F. Annegers, and L. R. Elveback. 1980. Mortality in patients with epilepsy. Epilepsia 21: 399–412. Kuller, L. 1966. Sudden and unexpected non-traumatic deaths in adults: Review of epidemiological and clinical studies. J Chron Dis 19: 1165–1192. Kuller, L., and A. Lilienfeld. 1966. Epidemiological study of sudden and unexpected deaths due to arteriosclerotic heart disease. Circulation 34: 1056–1068. Kurtzke, J. F. 1972. Mortality and Morbidity Data on Epilepsy. Pub. No. (NIH) 73-390. Washington, DC: Department of Health, Education and Welfare. Langan, Y., N. Nolan, and M. Hutchinson. 1998. The incidence of sudden unexpected death in epilepsy (SUDEP) in South Dublin and Wicklow. Seizure 7: 355–358. Lathers, C. M. 1980a. Effect of metoprolol on coronary occlusion–induced arrhythmia and autonomic neural discharge. Fed Proc 39: 771. Lathers, C. M. 1980b. Effect of timolol on autonomic neural discharge associated with ouabaininduced arrhythmia. Eur J Pharmacol 64: 95–106. Lathers, C. M. 1982. Lack of effect of methylprednisolone on cardiac neural discharge associated with coronary occlusion-induced arrhythmia and death. Eur J Pharmacol 85: 233–238. Lathers, C. M. 1983. Failure of methylprednisolone to prevent nonuniform cardiac accelerator nerve discharge associated with coronary occlusion-induced arrhythmia: Evidence against prostaglandin modulation of autonomic cardioaccelerator neural discharge in the anesthetized CAT. Med Hypothesis 10: 43–57. Lathers, C. M. 1996. Drug development: Role of academia, government, industry, and CRO’s. J Clin Pharmacol 36: 1–2. Lathers, C. M., G. J. Kelliher, J. Roberts, and A. B. Beasley. 1978. Nonuniform cardiac sympathetic nerve discharge: Mechanism of coronary occlusion and digitalis-induced arrhythmia. Circulation 57: 1058–1064.

798 Sudden Death in Epilepsy: Forensic and Clinical Issues Lathers, C. M., and J. Roberts. 1985. Are the sympathetic neural effects of digoxin and quinidine involved in their action on cardiac rhythm? J Cardiovasc Pharmacol 7: 350–360. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1974. Relationship between the effect of ouabain on arrhythmia and interspike intervals (I.S.I.) of cardiac accelerator nerves. The Pharmacologist 16: 201. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1975. Relationship of arrhythmias to digoxin and ouabain action on cardiac accelerator nerves. The Pharmacologist 17: 261. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1977. Correlation of ouabain-induced arrhythmia and nonuniformity in the histamine-evoked discharge of cardiac sympathetic nerves. J Pharmacol Exp Ther 203: 467–479. Lathers, C. M., J. Roberts, G. J. Kelliher, and A. B. Beasley. 1975. Comparison of the neural component in coronary occlusion and digitalis-induced arrhythmia. Clin Res 23: 566A. Lathers, C. M., J. Roberts, G. J. Kelliher, and A. B. Beasley. 1976. A comparison of procainamide and lidocaine effects on nerve discharge associated with coronary occlusion-induced arrhythmia in the cat. The Pharmacologist 18: 169. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23: 633–647. Lathers, C. M., and P. L. Schraeder. 1987. Review of autonomic dysfunction, cardiac arrhythmias, and epileptogenic activity. J Clin Pharmacol 27: 346–356. Lathers, C. M., and P. L. Schraeder. 1990. Epilepsy and Sudden Death. New York, NY: Marcel Dekker. Lathers, C. M., and P. L. Schraeder. 1995. Experience-based teaching of therapeutics and clinical pharmacology of antiepileptic drugs: Sudden unexplained death in epilepsy: Do antiepileptic drugs have a role? J Clin Pharmacol 35: 563–586. Lathers, C. M., A. J. Teres, A. B. Beasley, A. B. Malhotra, G. J. Kelliher, and J. Roberts. 1977. Cardiac accelerator nerve discharge and lidocaine blood levels after coronary occlusion. Fed Proc 36: 1002. Lathers, C. M., A. J. Teres, and G. J. Wetmore. 1977. Effect of quinidine on splanchnic and cardiac sympathetic neural discharge associated with digoxin-induced arrhythmia. Clin Res 25: 652A. Leestma, J. E. 1990. Sudden unexpected death associated with seizures: A pathological review. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, 61–88. New York, NY: Marcel Dekker. Leestma, J. E., A. F. Annegers, and M. J. Brodie, et al. 1997. Sudden unexplained death in epilepsy: Observations from a large clinical trial. Epilepsia 38: 47–55. Lund, A., and H. Gormsen. 1985. The role of antiepileptics in sudden death in epilepsy. Acta Neurol Scand 72: 444–446. May, T., U. Jurgens, B. Rambeck, and R. Schnabel. 1999. Comparison between premortem and postmortem serum concentrations of phenobarbital, phenytoin, carbamazepine and its 10,11epoxide metabolite in institutionalized patients with epilepsy. Epilepsy Res 33: 57–65. Nilsson, L., B. Y. Farahmand, P. G. Persson, I. Thiblin, and T. Tomson. 1999. Risk factors for sudden unexpected death in epilepsy: A case-control study. Lancet 13: 888–893. Opeskin, K., M. P. Burke, S. M. Cordner, and S. F. Berkovic. 1999. Comparison of antiepileptic drug levels in sudden unexpected deaths in epilepsy with deaths from other causes. Epilepsia 40: 1795–1798. Randall, W. C., M. Szentivanyi, J. B. Pace, J. S. Wechsler, and M. P. Kaye. 1968. Patterns of sympathetic nerve projections onto the canine heart. Circ Res 22: 315–323. Rodin, E. A. 1968. The Prognosis of Patients with Epilepsy, 326–329. Springἀeld, IL: Charles C Thomas. Schachter, S. C., G. W. Cramer, G. D. Thompson, R. J. Chaponis, M. A. Mendelson, and L. Lawhorne. 1998. An evaluation of antiepileptic drug therapy in nursing facilities. J Am Geriatr Soc 46: 1137–1141.

Clinical Pharmacology and SUDEP

799

Schraeder, P. L., and C. M. Lathers. 1989. Paroxysmal autonomic dysfunction, epileptogenic activity and sudden death. Epilepsy Res 3: 55–62. Tennis, P., T. H. B. Cole, J. F. Annegers, J. E. Leestma, M. McNutt, and A. Rajput. 1995. Cohort study of incidence of sudden unexplained death in persons with seizure disorder treated with antiepileptic drugs in Saskatchewan, Canada. Epilepsia 36 (1): 29–36. Wannamaker, B. B. 1990. A perspective on death of persons with epilepsy. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder, 26–37. New York, NY: Marcel Dekker. Wren, C., J. J. O’Sullivan, and C. Wright. 2000. Sudden death in children and adolescents. Heart 83: 410–413.

Experience-Based Teaching of Therapeutics and Clinical Pharmacology of Antiepileptic Drugs Sudden Unexplained Death in Epilepsy: Do Antiepileptic Drugs Have a Role?

49

Claire M. Lathers Paul L. Schraeder

Contents 49.1 49.2 49.3 49.4 49.5

Introduction Deἀnition and Classiἀcation of Epilepsy Epidemiology of Epilepsy Sudden, Unexpected, Unexplained Death in the General Population Sudden, Unexpected, Unexplained Death in the Epileptic Population 49.5.1 Cause of Death in Epilepsy 49.5.2 Cases of Sudden Death in Persons with Epilepsy 49.5.3 Cases of Near Sudden Death Associated with Seizures 49.5.4 Epidemiology of Sudden Death in Persons with Epilepsy 49.6 Possible Contributing Factors in the Pathogenesis of Cardiac Sudden Death 49.7 Possible Mechanisms for Autonomic Dysfunction and Sudden Death in Persons with Epilepsy 49.8 Some Clinical Pharmacologic Considerations in the Management of Epilepsy 49.8.1 Postmortem Toxicologic Findings in Victims of Sudden Unexplained Death in Epilepsy 49.8.2 The Ideal Pharmacologic Agent 49.8.3 Generic Substitutions of Antiepileptic Drugs 49.8.4 Federal Guidelines for Generic Antiepileptic Drugs 49.8.5 Consequences of Generic In Vivo Criteria 49.8.6 Generic Bioequivalence 49.8.7 Cases Illustrating Potential Problems Involved with the Use of Generic Antiepileptic Drugs 49.8.8 Drug Interactions 49.9 Conclusion References Appendix I. Self-Assessment Quiz

801

802 802 802 803 804 804 805 805 807 809 810 811 811 813 814 815 816 816 816 817 819 820 824

802 Sudden Death in Epilepsy: Forensic and Clinical Issues

49.1â•…Introduction Sudden unexplained death is an unfortunately frequent cause of mortality in young persons with epilepsy, accounting for up to 10% (Leestma 1990a) of all deaths in the epileptic population. Almost all persons with epilepsy take one or more antiepileptic drugs for many years. Yet many if not most victims of sudden unexplained death are found to have low or no measurable plasma levels of antiepileptic drugs. This latter observation suggests that low levels of antiepileptic drugs or possibly their withdrawal may predispose to sudden unexplained death. The risk/beneἀt of antiepileptic drugs in sudden unexplained death has recently become a concern of several investigators (American Academy of Neurology 1993). Although there are no ἀrm data, the appropriate plasma levels of antiepileptic drugs may exhibit a protective effect by decreasing the contribution of some of the factors, such as metabolism and bioavailability, that may be contributory to myocardial electrophysiologic instability and/or the associated sudden death.

49.2â•…Definition and Classification of Epilepsy Epilepsy has been deἀned by Gastaut (1973) as “chronic brain disorder of various etiologies characterized by recurrent seizures due to excessive discharge of cerebral neurons.” This deἀnition excludes single or occasional seizures that may be triggered by trauma, high fever, toxic states, syncopal attack, a ἀt of rage, somatic dysfunction, movement disorder, or behavioral reaction. In 1981, the International League against Epilepsy (1981) classiἀed epileptic seizures and syndromes into three major categories: (1) focal (partial or local); (2) generalized (convulsive or nonconvulsive); and (3) unclassiἀed. Abnormal electrical behavior of a limited population of neurons are focal seizures, whereas those involving large populations of neurons throughout the brain are designated as generalized. The latter group is subdivided into absence (petit mal), myoclonic, clonic, tonic, tonic–clonic, and atonic seizures, or combinations of the above. Seizures not included in the ἀrst two classiἀcations are designated as unclassiἀed. Repetitive or continuous seizures occurring in any category are termed status epilepticus.

49.3â•… Epidemiology of Epilepsy The incidence of epilepsy in the general population in the United States has been reported to vary from 30 to 54 in every 100,000 persons (Hauser and Kurland 1975; Kurland 1959). The likelihood of developing epilepsy is age related, with a gradually increasing prevalence with age until 50 years of age; thereafter, the number of cases begins to decline, and for the general population, begins to plateau (Schoenberg 1985). Hoffman (1988) reported that at least 73,000–131,000 individuals developed epilepsy in the United States in 1987. A lifetime prevalence rate for those with epilepsy that is clinically active and are living in the general population of major industrialized nations of the western hemisphere probably ranges between 0.5% and 0.78% (5.7–7.8/1000) (Zielinski 1982). Juul-Jensen and Foldspang (1983) concluded that approximately 1.2% of the population had epilepsy. The types of seizure disorders across the population are varied, with generalized tonic–clonic (grand mal) seizures being the most common in all age groups. Absence (petit mal) seizures are

Teaching of Therapeutics and Clinical Pharmacology of Antiepileptic Drugs

803

100

Percentage of cases

80

60

40

20

0 30

Age group GTC

Absence

Psychomotor

Other

Figure 49.1╇ Types of seizure disorders by percentage of incidence by age group. Note the relative commonness of absence seizures in the younger individuals and its relative rarity in older persons. (Adapted from Parke-Davis Company, Epilepsy: Patterns of Disease. Special Report. Chicago, 1958; and Epilepsy Foundation of America. Basic Statistics on the Epilepsies. F. A. Davis, Philadelphia, PA, 1975.)

the second most common type in young children, but diminish in prevalence with age. Psychomotor (complex-partial) seizures assume the secondary position in adolescents and adults (Figure 49.1). Note that approximately 50% of those with epilepsy will exhibit more than one type of seizure (Lennox 1960).

49.4â•…Sudden, Unexpected, Unexplained Death in the General Population As recently reviewed by Leestma (1990b), the component of sudden death that is “unexpected” is more readily deἀned than the time frame of the phenomenon. Some deἀne sudden death as that occurring between 1 and 24 h; others as less than 1 h, less than 2 h, or less than 12 h; some refer only to that occurring in min; and others distinguish between “instant” death, occurring in seconds, and “rapid” death, occurring in a few seconds up to about 10 min (Leestma et al. 1997). Because there are numerous deἀnitions of the time frame of unexpected sudden death, physicians should be careful to report the time frame in those cases where it is known to ultimately allow a clearer understanding of the contributory mechanisms involved and how to prevent their occurrence.

804 Sudden Death in Epilepsy: Forensic and Clinical Issues

However, some forensic pathologists deemphasize the importance of the time interval and focus on the unexpected aspect of the death. The aspect of unexplained death adds still yet another dimension to the deἀnition of sudden death (Leestma et al. 1985). The focus of this discussion is to present hypotheses on the mechanism(s) of sudden, unexpected, unexplained death in persons with epilepsy, and to discuss pharmacologic management of persons with epilepsy to attempt to avoid sudden death. In general, somewhat less than 1% of the population dies in any given year. In Cook County, Illinois, only about half of all deaths (15,000–17,000 per year) are reported to the Medical Examiner’s Office because (1)€there was no physician in attendance who could or would execute a death certiἀcate; (2)€foul play or criminal activity might have been involved; (3) the individual died within 24 h after admission to a hospital or during childbirth, surgery, or some other treatment; (4) the individual died in an institution or while incarcerated; or (5) the individual died in an accident (Leestma et al. 1985; Leestma 1990a). Sudden deaths account for at least 20% of all deaths in most urban areas and between 450,000 and 500,000 each year in the entire United States (Leestma 1990a). Approximately 65% of all reported cases are judged to have died of “natural” causes, from homicide, 4–5% of suicide, and 20% from accidents. In approximately 5% of the deaths, no “manner” of death could be determined. Although many individuals die suddenly by accident, homicide, or suicide, they could hardly be regarded as having died suddenly and unexpectedly or without obvious explanation. Most of the unexpected nontraumatic death cases fall into the “death-by-natural-causes” category (Leestma 1990a; Leestma et al. 1985). Anatomic classiἀcations of the autopsy ἀndings for 20,981 patients who were declared to have died from sudden death included the following: heart and/or aorta, respiratory, brain and meninges, digestive/urogenital, and miscellaneous. The miscellaneous category included sudden death occurring during no event or associated with surgery, obstetric delivery, bronchoscopy or radiologic procedures, hormone-secreting tumors such as pheochromocytomas and carcinoids, hemorrhaging tumors, or fat embolism (Luke and Helpern 1968; Kuller et al. 1967, 1974).

49.5â•…Sudden, Unexpected, Unexplained Death in the Epileptic Population 49.5.1â•… Cause of Death in Epilepsy In 1868, Bacon (1868), the medical superintendent of Cambridge County Asylum, classiἀed the causes of death in epilepsy into four categories: (1) those arising from the longcontinued effects of the disease on the body; (2) deaths after a rapid succession of ἀts; (3) sudden deaths in a ἀt; and (4) accidents due to ἀts. Category 3 was further deἀned into three subsets: (1) asphyxia from the spasms; (2) mechanical suffocation; and (3) sudden loss of nervous power, due most probably to the state of the heart or its nerves. In 1902, Spratling (1902) reported that approximately 4% of deaths that occurred over a number of years in a large population of institutionalized epileptic individuals could be classiἀed as sudden, unexpected, and attributable to no cause on postmortem examination. Numerous recent articles have summarized sudden, unexpected, unexplained deaths in persons with epilepsy once demonstrable causes of status epilepticus, accidents, drowning,

Teaching of Therapeutics and Clinical Pharmacology of Antiepileptic Drugs

805

drug overdoses, and intercurrent disease, etc., have been eliminated (Jay and Leestma 1981; Leestma et al. 1989). The phenomenon of sudden, unexpected, unexplained death in epileptic persons probably accounts for more than 10% of deaths in persons with epilepsy (Jay and Leestma 1981; Leestma et al. 1985; Leestma 1990a). The annual incidence of sudden death among individuals with epilepsy is at least 1 in every 500 to 1000, and perhaps higher (Jay and Leestma 1981). Annegers and Blakley (1990) reported the annual risk of sudden death to be 1 in 738, but noted that the annual risk for those older than 30 years of age was 1 in 371. They also noted that the risk varied by etiology of epilepsy: the annual risk was only 1 in 1000 in persons with idiopathic epilepsy compared with 1 in 100 for those with remote symptomatic epilepsy (i.e., symptoms caused, e.g., by a previous injury to the head as opposed to idiopathic epilepsy, in which there is no speciἀc structural or metabolic cause of the primary generalized epileptogenic activity).

49.5.2â•…Cases of Sudden Death in Persons with Epilepsy

Case 1 Terrence et al. (1981) described a sudden death occurring in a 19-year-old white male susceptible to grand-mal seizures since the age of 5 years. Before his death he was taking phenytoin, 100 mg, and phenobarbital, 30 mg, each three times a day. His seizures were infrequent and he was reported to go for days without taking his medications. On the evening before his death, the victim drank two small glasses of beer and returned home at 2:00 a .m. At 10:00 a .m., he was awakened to go to work; 15 min later he was found on the floor, deeply cyanotic, and in the midst of a grand-mal seizure. A paramedic team reported cardiopulmonary arrest and were unable to successfully resuscitate him. On autopsy, only pulmonary edema was noted. Plasma phenytoin and phenobarbital levels were subtherapeutic (5.8 and 3.2 μg/ml, respectively) and blood alcohol content was zero. Case 2 On a more personal level, one of the authors (PLS) treated a hospitalized woman who had just been stabilized at a therapeutic level of her usual anticonvulsant medication. As the neurologic team made rounds, the patient was told she was doing well and would be discharged later that day. While standing in the hall, they heard a noise in her room and found that she was in asystole. Resuscitation was unsuccessful.

49.5.3â•… Cases of Near Sudden Death Associated with Seizures The often difficult differentiation between seizures caused by cerebral epileptiform discharges and those resulting from cardiac arrhythmias or asystole, that is, syncopal seizures, is illustrated by the following two cases from one of the author’s (PLS) clinical experience.

806 Sudden Death in Epilepsy: Forensic and Clinical Issues

Case 3 A 45-year-old man was admitted to a hospital because of the onset of daily episodes of loss of consciousness without any evidence of seizures. While in the hospital, he had another such episode. An electrocardiogram (ECG) showed multiple premature ventricular beats that resolved once he regained consciousness. He was scheduled for a cardiologic workup, including invasive electrophysiologic studies. Further questioning of the patient revealed that he experienced the sensation of being far away from reality and had a strange taste in his mouth before each loss of consciousness. A prolonged electroencephalogram (EEG) was then obtained with video and ECG monitoring. During this study, he again experienced the sensation of dissociation from reality and had a bad taste in his mouth. The EEG at that time recorded left temporal spikes that proceeded to a bilateral, bitemporal electrographic seizure with onset of tachycardia with frequent premature ventricular beats. He was started on carbamazepine, achieving a serum level of 8 μg/ml and has had no more losses of consciousness. Case 4 The patient, a 14-year-old girl, had a history of generalized tonic–clonic seizures that started at the age of 10 years. She was placed on valproic acid, and a blood level of 55.0 μg/ml was achieved. A neurologic consultant was asked to see the patient because of recurrent seizures. The patient related that most of the events occurred when she saw her urologist for interval cystoscopy (she had a congenital urinary tract malformation) or when she was subjected to phlebotomy. The patient and her parents agreed that an EEG with ECG monitoring would be performed, during which the patient would undergo a phlebotomy. At the moment of skin penetration, the patient complained of nausea, followed by bradycardia progressing to asystole. After 13 s of asystole on ECG, the EEG became silent and a 15-s duration tonic–clonic seizure was observed. After a total of 40 s, the heart spontaneously started with a tachycardia of 120 beats/min. The patient awakened and was immediately alert and oriented. She was given an atropine inhaler to be used before any painful procedures and the valproic acid was discontinued. No more seizures have occurred in the 3-year period subsequent to the withdrawal of valproic acid. Case 3 illustrates that seemingly cardiogenic syncope can be the result of complex partial seizures. The history of symptoms compatible with a temporal lobe aura resulted in speciἀc studies being obtained that established the pathophysiologic mechanism of the cardiac arrhythmia. Case 4, in turn, demonstrates that reproducing the offending circumstances under appropriate monitoring conditions is necessary to establish a primary cardiovascular mechanism for “typical” tonic–clonic seizures. Further discussion of the spectrum of relationships between seizures and the cardiovascular system can be found in the work of Marshall et al. (1983) and Schraeder and Lathers (1989). Schraeder et al. (1983) reported a case involving a young man undergoing treatment with anticonvulsants who had tonic–clonic seizures. The seizures were the

Teaching of Therapeutics and Clinical Pharmacology of Antiepileptic Drugs

807

result of asystole induced by the mental image of pain. This report and cases 3 and 4 described above emphasize the potential for complex interactions between brain and heart in situations in which the patient has bona ἀde clinical seizure activity, albeit cardiogenic or epileptogenic.

McLeod and Jewitt (1978) used 24-h continuous ECG monitoring and noted that a variety of arrhythmias appeared in asymptomatic patients. Schott et al. (1977) using the same monitoring system found that some patients classiἀed as having suspected idiopathic epilepsy and no cardiac symptoms (only one had a transient focal abnormality in the EEG) had serious cardiac arrhythmias that were detected with the continuous ECG monitoring. Treatment with antiarrhythmic agents or the installation of a pacemaker eliminated the seizures and led to the conclusion that arrhythmias associated with seizure may be misdiagnosed as epilepsy more often than is generally recognized. Howell and Blumhardt (1990) suggest that the diagnostic possibilities include anoxia-induced seizures secondary to the cardiac arrhythmia or perhaps an overlap in the symptoms of hindbrain ischemia and temporal lobe seizures. Ictal EEG recordings are required to separate these two mechanisms. 49.5.4â•… Epidemiology of Sudden Death in Persons with Epilepsy Characteristics of some individuals in a population of patients with epilepsy who died suddenly have been compiled by Leestma et al. (1984, 1985, 1989). The data are based on 500 cases in Chicago, Illinois, during a period of 10 years. The greatest number of cases occurred in the 31- to 40-year-old age group (Figure 49.2). As depicted in Table 49.1, the

60

31-40

40

0

51–60

>60

10

10–20

20

41–50

21–30

30

L2 > L3, then there is an α,β ∈ (0,1) for combining the lotteries such that αL1 + (1 – α)L3 > L2 and L2 > βL1 + (1 – β)L3 • Independence axiom, in which for all L1, L2, L3 from the set of alternatives, and any α ∈ [0,1], then L1 ⩾ L2 if and only if αL1 + (1 – α)L3 ⩾ αL2 + (1 – α)L3 The difficulty with expected utility theory before Von Neumann and Morgenstern was the lack of a theory to explain the decision maker’s preferences (utilities) over the outcomes in a consistent and quantitatively meaningful way. For example, if the preference for one side-effect-free treatment over another is 5 “utils,” then how many utils is the preference for choosing a sports car over a sedan? Von Neumann and Morgenstern showed, counterintuitively, that a decision maker’s preferences are understood and modeled consistently if they are preferences over the lotteries (e.g., the probabilities), not the outcomes. Ultimately, preferences (utilities) over the outcomes, U(xi), can be inferred by gaining understanding of the decision maker’s preferences over the lotteries in a given decision context. Utility functions are not derived from ἀrst principles but are functions chosen empirically based on the preferences elicited from the decision makers themselves or are simply approximated using functions of the appropriate shape (Keeney 1992). Generally, there are three classes of utility function that capture the decision maker’s risk tolerance, including utility curves for • Risk averse decision makers of the general concave functional form, –e–αx • Risk neutral decision makers U(x) = x • Risk seeking decision makers, of the general convex form, e–αx A simple interpretation of risk tolerance in medical decision making is that the risk averse patient tightly holds a known outcome with known side effects that engages in a treatment and uncertain lottery to signiἀcantly improve the outcome but at risk of worse side effects. The relatively risk-seeking patient would be expected to give up the known outcome for even a small chance at signiἀcant improvements in the face of potentially worse side effects. Although health care providers have always had a major role in guiding patients through difficult decisions often having complicated trade-offs, increasingly accessible tools of decision analysis can serve as visual aids to help the patients comprehend differences that various trade-offs have in affecting the outcomes of their decisions. Health care providers have a role guiding patients directly but also in ensuring that the self-guided decision tools on internet health sites present objective information as input to the patient’s decision making. There are many other elements for modeling utilities of decision makers that are beyond this discussion. Except for a few modiἀcations and some related theories such as ambiguity as opposed to risk avoidance in stating preferences, expected utility theory has guided much of decision analysis for decades. However, perhaps one of the most important

894 Sudden Death in Epilepsy: Forensic and Clinical Issues

descriptive theories to further explain the decision maker’s judgments leading to preferences over the probabilities is prospect theory, in which it was shown that decision makers might be risk seeking at low probabilities of occurrence and risk averse at high probabilities (Kahneman and Tversky 1979; Tversky and Kahneman 1992). This work followed and was conἀrmed by many studies on judgments of probabilities and risk perception that show human difficulties in judging the probabilities of low-probability events and underestimate the probabilities of high probability events. Judgment flaws and the influence of decision making behaviors that run counter to classic, rational decision theory are discussed for general audiences in Predictably Irrational. The Hidden Forces that Shape Our Decisions (Ariely 2008). Outcomes or consequences in the discussion thus far have been univariate in nature. In reality, a disease treatment decision is unlikely to have only one objective, because the treatment comes with costs—both in terms of the charge for the treatment and in possible health consequences—and uncertain levels of beneἀcial outcome (efficacy). For instance, the decision maker’s objective might include maximizing the chance for life, minimizing the ἀnancial costs, and minimizing treatment-related discomfort. Clearly, there might be three or more objectives in this multiobjective decision (Figure 55.2). In reality, the decision maker must specify value weights or “scaling constants” (ki) for trading-off the objectives.* For example, a decision with two objectives and corresponding utilities, u1(x1) and u2(x2), has an overall utility, u(x1,x2), expressed as the linear combination (Keeney 1992),

u(x1,x2) = k1u1(x1) + k2u2(x2) + k3u1(x1)u2(x2).

If it is shown that additive independence of the utilities u1 and u2 exists, then the third term in the additive equation is omitted. There are several approaches commonly used to discover hidden objectives revealed by ἀnding incomplete independence of the component utilities. Generally, trade-off analysis, in which indifferences (equal utilities) are calculated among at least two terms, can reveal the presence of an underlying objective (Keeney 1992; Keeney and Raiffa 1993; Clemen 1996; Haimes 2009).

55.5â•…Uncertainties Outcomes in medical decision making are seldom known with certainty (Man-Son-Hing et al. 2000; Shakespeare et al. 2001). Uncertain states of nature (or chances in lotteries) include the probability of invasive disease revealed only after a decision to treat using surgery (Vickers and Elkin 2006) or the chance that a patient might not tolerate the chosen drug treatment. Choosing one treatment over another is an example of expected utility decision in which the probabilities over the consequences are different for the two treatments. Moreover, these are multiobjective decisions (Keeney and Raiffa 1993) for which objectives might include maximizing chances for a cure, minimizing chances of side effects, and minimizing time in inpatient care. Speciἀc measures for the attainment of * The k i are referred to a scaling constants to convey that they rescale sometimes differing independent objective scales to 0 → 1 for convenience and consistency of the analysis. The k i can more appropriately be referred to as “value weights” in a multiobjective value analysis, v(x1 ,x 2), where the individual terms are value functions as opposed to utility functions.

Decision Analysis and Risk Management

895

objectives (a.k.a. “attributes”) can be deἀned some of the attribute values can be informed using predictive models from population studies. For example, “cure” might be deἀned as “disease-free at 5 years after treatment” and scaled as a dichotomous variable in the set, [0,1]. Uncertainties in decision analysis are frequently modeled as discrete values. This simpliἀes the mathematical optimization problem for the quantitative analysis of the decision; however, in cases relying on population studies, it can overlook useful information. On the other hand, the value of including imperfect information studies has shown that, in some cases, modeling with symmetrical distributions can cancel the effect of including uncertainty (Morgan and Henrion 1990). Partly a result of difficulty in optimizing the decision under the continuous probability, some analyses use probability represented by discrete values for, e.g., the mid-range of a portion of the continuous distributions (Haimes 2009). Contemporary simulation approaches, such as Markov chain Monte Carlo methods, have greatly simpliἀed using more complex relationships. Thus, continuous distributions are appearing more commonly in decision models.

55.6â•… Benefits and Quality-of-Life Objectives Evaluation of the quality of life after treatment as an outcome of a treatment decision is useful for both population-based and individual decision analyses (Neumann et al. 2000; Hammitt 2002). Health policy analysts seek an understanding of how individuals value health as input for cost–beneἀt analyses (CBA) in health policy decisions. If valuation can be accomplished, then traditional CBA supports a simple decision rule: ἀnding that a net beneἀt for a health care program or treatment alternative exceeds its net costs means that the program should be adopted (or a treatment should be chosen). In the cost–beneἀt model, valuation of treatment alternatives for decision making requires a monetary value of life or, in other words, the economic production potential of “human capital.” The valuation exercise often generates controversy about the appropriate value of a life and also whether human preferences for health outcomes can be adequately modeled for the decision analysis (Hirth et al. 2000; Neumann et al. 2000, 2009; Hammitt 2002). Cost-effectiveness analysis circumvents some of the challenges of valuation in CBA, making it a preferred method to CBA for economic evaluation of health care policies (Neumann et al. 2000). Cost-effective analysis compares net costs for the net health beneἀts achieved for a speciἀc treatment compared to a speciἀc alternative treatment. The cost-per-effect ratio can be used to compare intermediate health states in terms that are relevant to speciἀc diseases and treatments, making it also useful for communicating health care policies or recommendations to clinicians in a given discipline (Mortimer and Segal 2008; Cohen and Neumann 2008). On the other hand, comparisons across diseases are difficult to accomplish using CEA because the intermediate outcomes of treatment for a speciἀc disease, e.g., cancer, can differ substantially from those in, e.g., epilepsy treatment. Furthermore, the relief from difficulties assigning monetary values to life in CBA comes at a sacriἀce of simplicity in a decision rule: there is no generally applicable decision rule in CEA. Although the direct valuation of life can be avoided in CEA, a measure of the relative beneἀt to at least intervals of lifetime after treatment is needed to close the beneἀts loop of a decision analysis. Policy makers and researchers sometimes use observational studies (Mortimer and Segal 2008) and willingness-to-pay methods (Hirth et al. 2000) to infer the

896 Sudden Death in Epilepsy: Forensic and Clinical Issues

dollar beneἀt per effective change in health status after a given treatment to derive a decision rules for deἀne problems. To model the relative beneἀt of treatment alternatives and the patient’s preferences over the various health outcomes as needed for CEA, the quality-adjusted life-year (QALY) has become a commonly used measure in health care evaluation studies (Neumann et al. 2000; Hirth et al. 2000; Brixner et al. 2009). In its simplest form, the patient’s preferences for different levels of quality of life generally assumes her utility function is zero for death and 1 for perfect health. A state of health quality that is judged to be a “fate worse than death” is sometimes given a negative quality weight by analysts. QALY analysis assumes that an individual moves through states of health over a lifetime and each of those states can be weighted with respect to the relative quality relative to death and perfect health. Although health status can improve or decline from one interval to the next, the summation over the remaining lifespan after treatment and without treatment is an integral measure of the beneἀt for the treatment of the disease. QALY is calculated simply as n

QALY =

∑q T , i i

i =1



where the quality of life during interval i is weighted as qi for the interval duration, Ti. For population-based analyses, the individual values can be summed across the population. As is the case for measures of preference or utility, QALYs need to be utility independent (Keeney 1992), risk neutral, and have constant proportional values for trade-off analysis. The latter concept means that a person would be willing to trade some fraction of their lifespan for an improvement in quality from one level to a preferred level fraction of lifespan would depend only on the differences in the level of QALYs and not the expected lifespan at the beginning of the period (Neumann et al. 2000). Risk neutrality means that the utilities are directly proportional to longevity for a ἀxed level of quality. However, if assumptions of utility independence (e.g., utility for the length of life and the quality of life) and constant proportionality hold, then QALYs can be modeled using general riskadjusted utility models.

55.7â•… Weighting the Quality of Life There is yet to emerge a single model for estimating weights (q) for QALY calculations. Direct questioning of patients is difficult for a variety of reasons, including the notion that patients who are ill over-weight the value of not having an existing condition. Sometimes psychometric tests are used to quantify relative health and mental status; however, these tests do not address the patient’s preferences for various health states—the fundamental core of a decision analysis. Additionally, prediction of QALYs from statistical analysis of descriptive health outcomes in, e.g., clinical trials remains challenging and suggests that algorithms need to ἀt speciἀc disease conditions (Mortimer and Segal 2008). Methods for scaling individual preferences for a health-related quality level in QALY analysis included direct elicitation using the standard gamble, time trade-off, visual analogue scaling, and person trade-off methods (Hammitt 2002). The standard gamble follows decision–theoretic modeling useful in everyday decisions (Hammond et al. 1999), in which

Decision Analysis and Risk Management

897

the respondent speciἀes the smallest chance of survival he or she could accept in a lottery where the alternative outcome is immediate death. This approach ἀts many life and death medical decisions but might be more difficult to apply in instances in which death is not a probable outcome for either lottery. For those instances, a time trade-off method might be preferable. All procedures are further complicated by the fact that the quality of life possibly changes from one period to the next (Figure 55.3). Time trade-off methods call for the patient to state the number of years in perfect health (q = 1) he or she considers equivalent to a speciἀc duration of chronic health condition. For example if the patient is indifferent between 10 years with a mild condition and 5 years of perfect health, then q = 5/10 = 0.5. The QALYs for either condition for which the patient is indifferent are 5 QALYs. The person trade-off method externalizes the preference framework by asking the respondent to consider the value of improving the health of groups in different health states (Hammitt 2002). Like the previous two methods, an indifference point is sought from which to estimate the QALY. For example, if the individual were asked to choose between extending the life of 100 healthy individuals for a year and extending the life of partially paralyzed individuals for a year, at what number (n) of partially paralyzed individuals would a decision maker be indifferent about the choice? The health-related quality level for living with partial paralysis would be 100/n. Another common approach for directly eliciting preferences for health outcomes is the visually adjusted scales. Visual methods call for asking the respondent to place various health outcomes along a linear scale from death to perfect health (Hammitt 2000; Neumann et al. 2000). Alternatively, preferences can be elicited using probability-based visualizations, such as those used in a comparison probability (gamble) trade-off techniques by Feldman-Stewart and colleagues (2007). Finally, willingness to pay (WTP) is an alternative approach to QALY for valuing health status in various intervals and health states after an event (Hirth et al. 2000; Hammitt 2002). WTP is often preferred to QALY in environmental risk analysis (Hammitt 2002). The essential key difference between WTP and QALY is that WTP is based on the amount of wealth (income) that an individual is willing to forego or pay to avoid a particular level of health decrement or, alternatively, “willing to pay” to improve the health statues to a Health related quality level (q)

1.0

0

0

Time after treatment

DC

DR

Figure 55.3╇ Hypothetical expected quality of life after disease treatment. Treatment alternative (dashed line) leads to longer life (DR) than “no treatment” option (solid line). Quality of life immediately after treatment is worse for treatment than control; however, the summed quality of life over periods—the QALYs—shows that treatment not only extends life but also increases the average.

898 Sudden Death in Epilepsy: Forensic and Clinical Issues

particular state of health in the intervals of interest. Although both QALY and WTP seek to support a clear beneἀts analysis and to avoid the controversy of directly valuing life, once incremental values are placed on either the WTP or QALYs, the value of a statistical life can be inferred from the data (Hammitt 2002; Alberini 2005).

55.8â•… Example Decision Tree Decisions can be analyzed to the degree of skills and interests of the decision maker and using tools that range from pencil and paper to sophisticated software. Various approaches at simplifying complex decisions and presenting the results visually have been published and some examples are given in Table 55.1 and Figure 55.4. The decision tree is often used because it can be drawn to illustrate all elements of the decision and provide a framework for calculations. This example will use a decision tree (Figure 55.5) and the discussion and comparison of odds ratio approaches in Lathers et al. (2003). Figure 55.5 shows a possible decision tree for the choice among antiepileptic drugs (AEDs), including both positive and negative consequences. The decision is modeled as the choice (square node) among alternative AEDs followed by the chance (circular nodes) of improvement, given the drug selection. The side effect of “sedation” is modeled as a subsequent chance node having probabilities of sedation that are also drug dependent. The numbers at the ends of the nodes are the outcomes or consequences. Figure 55.5 depicts a simplifying assumption that sedation is not a factor for the decision maker given a failure to improve after initiating treatments that do not achieve improvements. Although the onset of side effects can begin with the beginning of drug treatment, the depicted model assumes that the ineffective (and its side effects) treatment would cease in favor of an alternative. Moreover, the simpliἀed model in Figure 55.5 does not capture the fact that treatment regimen can be altered deliberately with duration of treatment, i.e., while titrating Table 55.1â•… Typical Decision Analysis and Visualizations for Treatment Decisions Authors

Method

Type of Presentation

Otoul et al. 2005

Two-way plots: odds ratios for withdrawal vs. QR responder

Visual

Vickers and Elkin 2006

“Decision curve.” Variation on receiver operator characteristics calculated as the net beneἀt against the probability of disease among patients Classic meta-analysis calculating odds ratios and conἀdence interval estimates Risk–beneἀt curves: adverse events and efficacy probability contours Pairwise comparisons of odds ratios Odds ratios in four directions Graphical decision analysis for studying perception

Visual and table analysis

Song et al. 2003 Shakespeare et al. 2001 Caldwell et al. 2005 Lathers et al. 2003 Feldman-Stewart 2007

Visual plots of odds ratios ± conἀdence intervals Visual—contours Table Visual—plot Visual qualitative presentation

Includes Uncertainties in Analysis? Yes; classic conἀdence interval estimates uncertainties Discrete and distributions for the disease uncertainty Yes; classic conἀdence intervals Yes—discrete in contour levels Yes No No

Decision Analysis and Risk Management Relative OR 1.0 0.8 0.5

899

Drug A Drug B Drug C

0.3 Placebo

0.0

Drug

Relative sample size

Figure 55.4╇ Four-way plot of a decision problem comparing three alternative drug treatments for epileptic seizures. The largest relative odds ratio for each two-way drug efficacy comparison is used as denominator for the “Relative OR” axis. Drug treatment improvement rate appears on right horizontal axis, placebo rate for the same study on left horizontal axis, and relative sample size on lower vertical axis. The larger the area under similar placebo rates, the better the comparison. (From Lathers, C. M. et al., J Clin Pharmacol, 43, 491–503, 2003. With permission.)

the patient’s tolerance for side effects. Although beyond the scope of this chapter, linked or multistage decisions, in which the outcome of one decision is an input to the next, can be represented visually and in basic calculations. For example, Hammond et al. (1999) discussed the features of linked decisions for everyday decision making. The calculations that the decision tree represents follow from the expected value formulae shown in Figure 55.1. Tracing backward on the topiramate branch, the expected value of the sedation lottery, given that improvement has occurred from the choice of topiramate, is the sum of the probabilities of the possible outcomes times the values of their respective outcomes or consequences. As discussed above, the outcomes may include a vector of beneἀcial and deleterious consequences, such as physical health, emotional wellbeing, loss/maintenance of income, and other factors. The consequences must be in the same value units, such as dollars, to complete the expected value calculations. Given that the branches of the example tree are identical in structure, the expected value (EV) of a branch for drug alternative a is given by:

{



}

 psed ,a w °T (1 + Qepi )/ 2 − ∆Qsed  − Ca − Csed ,a +      EVa = pI,a *   (1 − psed ,a )* wT (1 + Qepi )/ 2 

where pI,a = probability of improvement, given treatment with drug a psed,a = probability of sedation, given drug a T = duration in years (or remaining lifetime, if applicable) Qepi = quality loss per year due to disease (epilepsy)

900 Sudden Death in Epilepsy: Forensic and Clinical Issues Sedation Improvement

45%

Topiramate net outcome No sedation

Topiramate

55% 0 Sedation

Improvement

27%

7%

Lamotrigine net outcome No sedation

93%

Lamotrigine treatment –135 Neutral

AED

80%

Topiramate treatment –177 Neutral

Lamotrigine

20%

73%

AED choice 19% –2500 Phenobarbital net outcome Sedation

Improvement

58%

No sedation Phenobarbital

81% 5000

Phenobarbital treatment –100 Neutral

42%

Figure 55.5╇ Typical decision tree for the choice among three AEDs including sedation side effect. The example from Lathers et al. (2003) was used to develop the decision tree; thus, probabilities expressed as percentages along the branches are average response rates from Tables VI and VII in Lathers, C. M., et al. (2003). See text for assumptions used in developing the decision tree, here shown using Precision Tree Ⓡ software for ExcelⓇ.

ΔQsed = quality decrement (QALYs lost) due to sedation Ca = cost of treatment for drug alternative a Csed = cost of treating sedation w = value weight of a unit of quality adjusted life year The expected beneἀt of one drug treatment alternative compared to a second is the difference between the expected values of the branches, e.g., EVtopiramate – EVphenobarbital. For example, if the values for the equation are substituted from Tables VI and VII in Lathers et al. (2003) and we arbitrarily assign w = $75,000 per year; Qepi, = 0.3 per year; ΔQsed = 0.3; Ca€ = $177, $135, and $100 per month for topiramate, lamotrigine, and phenobarbital, respectively (Lathers et al. 2003); and T = 5 years, then the beneἀt of topiramate is about 24% less and the beneἀt of lamotrigine about 53% less than that for phenobarbital.

Decision Analysis and Risk Management

901

Of course, this example considers only seizure improvement, costs, and sedation rates. Including other side effects would change and possibly reverse the outcome. Once the branches of the decision tree have been calculated, derived measures in addition to net beneἀt are useful when comparing alternative treatments. In particular, the expected value of perfect information for a decision is useful in indentifying optimal strategies for obtaining additional information. This is the difference between the highest expected net beneἀt in the decision and the expected value of the decision, given the uncertainties. For the static, simple model shown in Figure 55.5, the expected value of perfect information for each branch is simply the difference between the inner terms in the branch equation above. However, the expected value of perfect information is maximally useful when the uncertainties of the probabilities and possibly other terms in the equation are uncertainty distributions. Then, the relative values of reducing uncertainty for each term can be estimated from the distributions (Clemen 1996; Ades et al. 2004; Willan 2007). The sources of the uncertainties are the clinical trial statistics (e.g., the improvement rate or probability ± error) and judgments about the appropriate form of the uncertainty distribution. Often, beta distributions in the range 0 to 1 are used for modeling uncertainty in probabilities along the branches (Ades et al. 2004). Finally, contemporary software for decision modeling affords decision analyses the efficient tools for incorporating uncertainty distributions in the calculations. In a Bayesian approach to modeling and analyzing the decision, the starting parameters, such as the branch probabilities, can be estimated from prior knowledge—the outcomes of clinical trials or public health follow-on studies such as those used by Lathers and colleagues (2003). Because the number of subjects in a given trial affect the relative uncertainty in the parameters, simulations of the decision using distributions of the decision variables will reflect the relative information or “strength” of the data supporting the overall calculation. Although computationally and mathematically more complex, the solutions of comparative therapies using full uncertainty analysis yields richer information for decision support than the back of the envelope methods. Again, the ultimate choice of a computation method for the decision analysis depends on the purpose of the decision (public health or individual?) and the needs of the decision maker for either rigorous or informal decision support.

55.9â•… A Structured and Focused Way of Making Difficult Choices This chapter presents some of basic elements of decision analysis. There are numerous advances in the practice of decision analysis that are used to scale decision analysis from personal decisions (Howard et al. 1999; Glasziou 1995) to high-impact health care policy (Devlin and Parkin 2004; Brixner et al. 2009) or even environmentally contentious decisions (Linkov et al. 2006). Across the diverse applications of decision analysis, the basic principles of linking objectives, chances, consequences (or outcomes), trade-offs, and risk tolerance in a structured thinking process is central to creating robust decision support and to selecting a single treatment from among many. Ultimately, there might be as many ways to approach analyzing the decision problem as there are personal, organizational, or societal values and preferences for beneἀcial health outcomes for balanced costs. The structured or “focused thinking” processes described by many (Keeney 1992; Clemen 1996; Linkov et al. 2006; Haimes 2009) provide ways to include the preferences, concerns,

902 Sudden Death in Epilepsy: Forensic and Clinical Issues

and values of the decision makers with the quantitative observations from clinical trials of safety and efficacy for optimum choices of treatment. This chapter emphasizes the importance of decision analysis in resolving uncertainties, risk and value trade-offs, and dealing with linked decisions over an extended period of treatment. Difficult decisions, depending on many variables and uncertainties that might affect outcomes, are a continuing challenge for health care providers treating persons risk for SUDEP. The optimal course of treatment of patients at risk for SUDEP necessitates establishing a balance among efficacies, side effects, and the possible occurrence of SUDEP (Lathers et al. 2003). Decision risk analysis as it applies to risk management is a process that can beneἀt in determining modiἀcation of risk factors for SUDEP.

References Ades, A. E., G. Lu, and K. Claxton. 2004. Expected values of sample information calculations in medical decision making. Med Decis Making 24: 207–226. Alberini, A. 2005. What is a life worth? Robustness of VLS values from contingent valuation surveys. Risk Anal 25: 783–800. Ariely, D. 2008. Predictably Irrational. The Hidden Forces that Shape Our Decisions. New York, NY: Harper-Collins. Brixner, D. I., A. P. Holtorf, P. J. Neumann, D. C. Malone, and J. B. Watkins. 2009. Standardizing quality assessment of observational studies for decision making in health care. J Manag Care Pharm 15: 275–283. Caldwell, D. M., A. E. Ades, and J. P. T. Higgins. 2005. Simultaneous comparison of multiple treatments: Combining direct and indirect evidence. BMJ 331: 897–900. Claycamp, H. G. 2007. Perspective on quality risk management of pharmaceutical quality. Drug Inf J 41: 353–367. Claycamp, H. G. 2006. Rapid beneἀt-risk assessments: No escape from expert judgments in risk management. Risk Anal 26: 135–146. Clemen, R. T. 1996. Making Hard Decisions. An Introduction to Decision Analysis. 2nd ed. Paciἀc Grove, CA: Duxbury Press. Cohen, J. T., and P. J. Neumann. 2008. Using decision analysis to better evaluate pediatric clinical guidelines. Health Affairs 27: 1467–1475. Devlin, N., and D. Parkin. 2004. Does NICE have a cost-effectiveness threshold and what other factors influence its decisions? A binary choice analysis. Health Econ 13: 4437–4452. Feldman-Stewart, D., M. D. Brundage, and V. Zotov. 2007. Further insight into the perception of quantitative information: Judgments of gist in treatment decisions. Med Decis Making 27: 34–43. Food and Drug Administration (FDA). 2006. Guidance for Industry Q9 Quality Risk Management. CenÂ� ter for Drug Evaluation and Research (CDER) and Center for Biologics Evaluation and Research (CBER). http://www.fda.gov/downloads/drugs/GuidanceComplianceRegulatoryInformation/ Guidances/ucm073511.pdf (Accessed February 1, 2009). Funtowicz, S. O., and J. R. Ravetz. 1992. Risk management as a postnormal science? Risk Anal 12:€95–97. Glasziou, P. P. 1995. An evidence based approach to individualizing treatment. BMJ 311: 1356–1359. Haimes, Y. Y. 2009. Risk Modeling, Assessment, and Management. New York, NY: John Wiley & Sons, Inc. Hammond, J. H., R. L. Keeney, and H. Raiffa. 1999. Smart Choices. New York, NY: Broadway Books. Hammitt, J. K. 2002. QALYs versus WTP. Risk Anal 22: 985–1001.

Decision Analysis and Risk Management

903

Harrell, Jr., F. E., K. L. Lee, and D. B. Mark. 1996. Multivariable prognostic models: Issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med 15: 361–387. Hirth, R. A., M. E. Chernew, E. Miller, A. M. Fendrick, and W. G. Weissert. 2000. Willingness to pay for a quality-adjusted life year: In search of a standard. Med Decis Making 20: 332–342. Hocine, M. N., P. Tubert-Bittner, T. Moreau, M. Chavance, E. Varon, and D. Guillemot. 2007. Relativerisk ratio was a useful measure of differential association in cohort and case-series studies. J Clin Epidemiol 60: 361–365. Kahneman, D., and A. Tversky. 1979. Prospect theory: An analysis of decisions under risk. EconÂ� ometrica 47: 263–291. Keeney, R. 1992. Value-Focused Thinking. Cambridge, MA: Harvard University Press. Keeney, R. 2004. Making better decision makers. Decis Anal 1: 193–204. Keeney, R. L., and H. Raiffa. 1993. Decisions with Multiple Objectives: Preferences and Value Tradeoffs. Cambridge, UK: Cambridge University Press. Lathers, C. M., P. L. Schraeder, and H. G. Claycamp. 2003. Clinical pharmacology of topiramate versus lamotrigine versus phenobarbital: Comparison of efficacy and side effects using odds ratios. J Clin Pharmacol 43: 491–503. Linkov I., F. K. Satterstrom, G. Kiker, C. Batchelor, T. Bridges, and E. Ferguson. 2006. Comparative risk assessment to multi-criteria decision analysis and adaptive management: Recent developments and applications. Environ Int 32: 1072–1093. Man-Son-Hing, M., A. Laupacis, A. M. O’Connor, D. Colyle, R. Berquist, and F. McAlister. 2000. Patient preference-based treatments thresholds and recommendation: A comparison of decision-analytic modeling with the probability-tradeoff technique. Med Decis Making 20: 394–402. Morgan, M. G., and M. Henrion. 1990. Uncertainty. A Guide to Dealing with Uncertainty in Quantitative Risk and Policy Analysis. Cambridge, UK: Cambridge University Press. Mortimer, D., and L. Segal. 2008. Comparing the incomparable? A systematic review of competing techniques for converting descriptive measures of health status in QALY-weights. Med Decis Making 28: 66–89. Neumann, P. J., S. J. Goldie, and M. C. Weinstein. 2000. Preference-based measure in economic evaluation in health care. Annu Rev Public Health 21: 587–611. Neumann, P. J., P. D. Jacobson, and J. Palmer. 2009. Measuring the value of public health systems: The disconnect between health economists and public health practitioners. Am J Pub Health 98: 2173–2180. National Research Council. 1996. Understanding Risk: Informing Decisions in a Democratic Society, eds. P. C. Stern and H. V. Fineberg. Washington, DC: National Academy Press. Otoul, C., C. Arrigo, K. van Rijckevorsel, and J. A. French. 2005. Meta-analysis and indirect comparisons of levetiracetam with other second-generation antiepileptic drugs in partial epilepsy. Clin Neuropharmacol 28: 72–78. Paté-Cornell, M. E., and R. L. Dillon. 2006. The respective roles of risk and decision analyses in decision support. Decision Anal 3: 220–232. Shakespeare, T. P., V. J. Gebski, J. Veness, and J. Simes. 2001. Improving interpretation of clinical studies by use of conἀdence levels, clinical signiἀcance curves, and risk–beneἀt contours. Lancet 357: 1349–1353. Slutsky, J. R., and C. M. Clancy. 2009. AHRQ’s Effective Health Care Program: Why comparative effectiveness matters. Am J Med Qual 24: 67–70. Slovic, P. 2000. The Perception of Risk. London, UK: Earthscan Publications. Song, F., D. G. Altman, A. M. Glenny, and J. J. Deeks. 2003. Validity of indirect comparison for estimating efficacy of competing interventions: Empirical evidence from published meta-analyses. Br Med J 326: 472–477. Thompson, K. M. 2002. Variability and uncertainty meet risk management and risk communication. Risk Anal 22: 647–654.

904 Sudden Death in Epilepsy: Forensic and Clinical Issues Tversky, A., and D. Kahneman. 1974. Judgment under uncertainty: Heuristics and biases. Science 165: 1232–1238. Tversky, A., and D. Kahneman. 1992. Advances in prospect theory: Cumulative representation of uncertainty. J Risk Uncertainty 5: 297–323. van Dijk, M. R., E. W. Steyerberg, and J. D. F. Habbema. 2008. A decision-analytic approach to deἀne poor prognosis patients: A case study for non-seminomatous germ cell cancer patients. BMC Med Inform Decis Mak 8: 1–9. Willan, A. R. 2007. Clinical decision making and the expected value of information. Clin Trials 4: 279–285. Whitney, S. N., M. Holmes-Rovner, and C. H. Brody et al. 2008. Beyond shared decision making: An expanded typology of medical decisions. Med Decis Making 28: 699–705. Vickers, A. J., and E. B. Elkin. 2006. Decision curve analysis: A novel method for evaluating prediction models. Med Decis Making 26: 565–574. Von Neumann, J., and O. Morgenstern. 1947. Theory of Games and Economic Behavior, 2nd ed. Princeton, NJ: Princeton University Press. Wilson, E. C. F., S. J. Peacock, and D. Ruta. 2008. Priority setting in practice: What is the best way€to€compare costs and beneἀts? Health Economics. 17 Jun 2008 Publ online http://www3╉ .interscience.wiley.com/journal/5749/home.

Epilepsy Surgery and the Prevention of SUDEP Ryan S. Hays Michael R. Sperling

56

Contents 56.1 Introduction 56.2 Methods 56.2.1 Selected Studies 56.2.2 Study Design 56.2.3 Calculating Mortality 56.2.4 Duration of Follow-Up 56.3 Results 56.3.1 Effect of Postoperative Seizure Control on Mortality 56.3.2 Effect of Laterality of Surgery on Mortality 56.3.3 Effect of Gender of Surgical Patients on Mortality 56.3.4 Effect of Pathologic Diagnosis on Mortality 56.3.5 Effect of Cerebral Lobe on Mortality 56.3.6 Comparison of Resection/Subpial Transection with Callosal Section 56.4 Conclusion References

905 906 906 907 907 907 908 908 910 911 911 911 911 912 912

56.1â•…Introduction Epilepsy can occasionally be a fatal condition. People with epilepsy have mortality rates that are two to three times higher than the general population (Cockerell et al. 1994; Hauser et al. 1980; Lhatoo et al. 2001; Nilsson et al. 1997), and the cause of the excess mortality is likely related in part to the underlying etiology of epilepsy and in part to the seizures themselves. Sudden unexplained death in epilepsy (SUDEP) is a leading diagnosis, especially in younger individuals, with the risk of sudden death perhaps as much as 24 times higher than that of the general population (Ficker et al. 1998; Walczak et al. 2001). While the exact pathophysiologic mechanisms of SUDEP remain elusive, a number of risk factors have been identiἀed as being associated with an elevated risk of SUDEP (Tomson et al. 2008). These include developmental delay or mental retardation, subtherapeutic anticonvulsant levels, treatment with more than two antiepileptic medications, and uncontrolled seizures, particularly generalized tonic–clonic seizures (Hughes 2009; Strauss et al. 2003; Walczak et al. 2001). The potential pathophysiologic mechanisms of SUDEP, as well as the risk factors, are reviewed in detail elsewhere in this volume (Lathers et al. 2010;€ Walczak 2010, this book). Ultimately, for death to occur, there must be an acute disruption of autonomic function, with cardiac arrhythmias and respiratory arrest or€ hypoventilation likely to be the ἀnal common direct pathway for SUDEP. Given the 905

906 Sudden Death in Epilepsy: Forensic and Clinical Issues

strong direct association between seizures and SUDEP, one might theorize that preventing seizures might reduce the risk of SUDEP and other seizure-related causes of mortality. Consequently, some have studied how improving seizure control might affect mortality. There is some evidence that controlling seizures with medications can lower the risk of SUDEP (Nilsson et al. 1999; Strauss et al. 2003). Epilepsy surgery provides another means of controlling seizures in some patients. At least 35% of people with epilepsy do not fully respond to medical therapy (Kwan and Brodie 2000), and surgery often alleviates seizures when medications fail to control seizures (Engel 1996; Engel et al. 2003; Sperling et al. 1996; Téllez-Zenteno et al. 2005; Wiebe et al. 2001; Wieser et al. 2003). No prospectively randomized studies have been performed to examine the effect of surgery on mortality. Indeed, only one prospective randomized study of epilepsy surgery has ever been performed (Weibe et al. 2001). This was conducted in Canada, where the typical wait for surgery was approximately one year, and patients could be randomized to either remain in the surgery queue while continuing medical therapy for another year, or advance to the head of the line and immediately have surgery. This study found that the patients treated surgically had a signiἀcantly higher rate of seizure freedom compared to patients who were awaiting surgery additionally, patients treated surgically had a signiἀcantly higher quality of life. The report contains one other interesting observation. In the group randomized to late surgery, that is, continued medical therapy, one death occurred. In the patients randomized to immediate surgery, no one died. Only 80 patients were included in this trial; thus, a statistical analysis cannot be done. Nonetheless, it is intriguing that the single death occurred in the medically treated patients, while none were observed in surgical patients. Only a few nonrandomized studies have examined the effect of the surgical treatment on long-term mortality in epilepsy, and these suggest that seizure control is associated with reduction in mortality. This chapter will review this literature regarding the impact of epilepsy surgery on mortality.

56.2╅ Methods 56.2.1╅ Selected Studies A literature review of studies published in English that examined the long-term mortality of surgically treated patients with epilepsy yielded six key studies (Hennessy et al. 1999;€Nilsson et al. 2003; Salanova et al. 2002; Sperling et al. 1999, 2005; Stavem and Guldvog 2005). All of the studies rely on retrospective analyses of nonrandomized subjects. With the exception of one study that reported no impact of surgery on mortality, all other studies reported a positive effect of epilepsy surgery on the mortality rate and the incidence of SUDEP. The sole negative study compared patients treated surgically and medically in Norway from€1948 through 1988, many of whom had surgery before the advent of modern evaluation techniques (Stavem and Guldvog 2005). Most of their surgically treated patients continued to experience frequent seizures after surgery, and surgical morbidity was quite high. Consequently, the absence of a difference in death rates between medically and surgically treated patients likely reflects these underlying facts. The other reports include three studies performed at academic institutions in the United States, one multicenter report from Sweden utilizing the nationwide epilepsy registry, and one study from a tertiary hospital in London. One

Epilepsy Surgery and the Prevention of SUDEP

907

of the American studies (Sperling et al. 2005) updates an earlier report by the same group (Sperling et al. 1999), and the more recent study will be discussed in this chapter. 56.2.2â•… Study Design The nonrandomized studies tell a largely consistent story. The study by Nilsson and colleagues (2003) compared mortality rates in epilepsy patients with that of equally intractable patients who did not have surgery, either because they were deemed not to be surgical candidates, or were unwilling to have surgery, or died before surgery was offered. The other studies followed surgically treated patients and examined mortality in relation to postoperative seizure control. Both study designs are limited by necessity. The ἀrst design does not offer ideal control subjects for comparison with the surgical patients, since the composition of the medical group differs somewhat from the surgically treated group. The second design does not prospectively account for the possibility that patients who ultimately fail to respond to surgery differ in some fundamental way from those who respond. This shortcoming might be partly overcome by comparing mortality in patients who become seizure free with expected mortality in the general population (Sperling et al. 1999); however, this approach still lacks the robustness of a prospective, randomized controlled trial. However, the extant studies probably represent the best that can be done at present since it is not ethical to randomize some patients with intractable epilepsy to an unnecessary delay in operation. 56.2.3â•… Calculating Mortality In calculating the rate of death, two different measures were used. The standardized mortality ratio compares the mortality rate in a subject population with that of the general population. It is calculated by dividing the number of observed deaths in a particular cohort by the number of expected deaths in an age, gender, and perhaps racially matched general population. The standardized mortality ratio is helpful in comparing the all-cause mortality of patients with epilepsy or subgroups thereof with that of the general population. The second method relates the total number of deaths of a speciἀc cohort during a deἀned time period (usually per 1000 person-years of follow-up), thus allowing the mortality incidence to be tabulated. An expected number of deaths per thousand person-years can be determined for an age, gender, and ethnically matched general population. The death rate per thousand person-years of the speciἀc cohort can then be compared with the death rate in the general population, and an annualized risk of death estimated. 56.2.4â•…Duration of Follow-Up The studies varied in duration of follow-up and in frequency of assessment during that time. Most patients were followed on an annual basis postoperatively, and studies had a mean follow-up of 5–9 years. The Nilsson et al. (2003) study was limited by following each patient for 2 years after surgery and then relied on medical records of only selected patients for further detail. This design has the risk of underreporting the number of patients with recurrent postoperative seizures. Indeed, in that study, some of the patients who had been reported as being seizure free at the ἀxed 2-year interval were discovered to have had seizure relapses when the individual medical records were subsequently reviewed. The issue

908 Sudden Death in Epilepsy: Forensic and Clinical Issues

of follow-up duration is a complex one. New seizure recurrences happen with each passing year after surgery. Recurrence rates are low after the ἀrst few postoperative years, only 2–4% per year (Sperling et al. 2008). Since the bulk of postoperative seizures occur in the early years after surgery, one can still reasonably estimate mortality risk if seizure recurrence in the ἀrst few years after surgery is known. Moreover, use of survival analysis helps account for disparities in duration of follow-up. This issue is even more complicated, since the added risk of dying might mainly exist around the time that seizures occur. Analyses might subdivide time after surgery for each patient into seizure-free and non–seizure-free epochs to better estimate risk of mortality imposed by seizures. This is methodologically fraught, however, since methods for deἀning such epochs would be arbitrary. Moreover, all studies are limited by accuracy of subject reporting. Patients may either intentionally or unintentionally underreport seizures to their physicians. They may hide seizure recurrence to gain driving privileges, may be loathe to disappoint their physicians, or may be unaware of their seizures. Patients may also forget whether seizures have occurred if follow-up contact is infrequent. In these circumstances, the bias is to underreport seizures, and patients might inappropriately be registered as seizure free. On the other hand, other patients may report continued seizures after surgery when they may not be experiencing epileptic seizures but rather may have non-epileptic psychogenic seizures or other non-epileptic phenomena instead. While it is the authors’ experience that overreporting is probably less common than under-reporting seizures, no studies have been performed to evaluate the frequency of these confounders.

56.3â•…Results 56.3.1â•… Effect of Postoperative Seizure Control on Mortality Table 56.1 summarizes the pertinent ἀndings from the aforementioned studies. Even after controlling for any mortality directly related to the surgery itself, each study found that mortality rates were lower in the surgically treated patients. This effect was most striking in patients who stopped having seizures after surgery (Figure 56.1). While Nilsson et al. (2003) demonstrated no signiἀcant difference in the rate of SUDEP between surgical patients who remained seizure free versus surgical patients who had recurrent seizures, they demonstrated a signiἀcant difference in the standardized mortality ratio between surgical and medically treated patients. Surgical patients had a standardized mortality ratio of 4.9 with a SUDEP incidence of 2.4 per 1000 person-years, whereas medically treated patients had a standardized mortality ratio of 7.9 and SUDEP incidence of 6.3 per 1000 person-years. Salanova et al. (2002) demonstrated that surgical patients who became seizure free had a signiἀcantly lower mortality (standardized mortality ratio = 1.7, approaching that of the general population), while surgical patients with persistent seizures had a much higher mortality (standardized mortality ratio = 7.4). Finally, Sperling et al. (2005) demonstrated that the excess mortality associated with refractory epilepsy was eliminated when epilepsy surgery successfully abolished seizures. Patients with recurrent postoperative seizures had a mortality rate of 11.4 per 1000 person-years, and those without recurrent seizures had a rate of 0.85 per 1000 personyears (p = 0.001), which is the same as that of the general population. These studies all demonstrate a consistent trend. Patients who stop having seizures tend to have lower mortality than patients who continue to experience seizures, and mortality

Epilepsy Surgery and the Prevention of SUDEP

909

Table 56.1â•… Results of Selected Studies Study Years studied Surgery type

Number of surgical patients

Number of nonsurgical patients Surgical outcome, Engel class Percent of mortality due to SUDEP or possible SUDEP SUDEP incidence, surgical cohort

SMR, all-cause death, surgical cohort SUDEP incidence, nonsurgical cohort SMR, nonsurgical cohort

Sperling et al. 2005

Nilsson et al. 2003

1986–2000 89.4% Resection and/or subpial transection; 10.6% CC 583 (Includes 393 from the analysis published in 1999) N/A

1990–1998 84.2% Resection; 11.5% CC

1984–1999 100% Resection

Hennessy et al. 1999 1975–1995 100% Resection

215

299

212

N/A

N/A

44% Class I (50% in patients with focal surgery) 53%

49% Class I

69% Class I

40% Class I A/B; 24% class I C/D or class II

43%

55%

20%

6.3/1000 (based on 10 cases of€SUDEP per€1580 person-years)a

2.4/1000 person-years

2.2/1000 person-years (reported as 1/455 person-years)

Whole surgical cohort = 3.56 Persistent sz = 5.75 Sz-free = 0.45 N/A

Whole surgical cohort = 4.9 Persistent sz = 4.3 Sz-free = 3.8 6.3/1000 Person-years 7.9

4.0/1000 personyears (based on 6 unexplained or seizure-related deaths per 1514 person-years) Whole surgical cohort = 5.27 Persistent sz = 7.9

N/A

651 Operations in 596 patients

Salanova et al. 2002

Sz-free€ = 1.7 N/A N/A

Whole surgical cohort€= 4.5 Persistent sz = not reported Sz-free = not reported N/A N/A

Note: CC, corpus callosotomy; N/A, not applicable; sz, seizure; SMR, standardized mortality ratio. Class I: seizure free (no seizures with loss of awareness); class I A, B, C, and D are subtypes of seizure freedom (A, no seizures after surgery; B, simple partial seizures only; C, some postoperative seizures but none for 2 years; D, generalized seizures after drug withdrawal only); class II: rare seizures. a All cases of SUDEP occurred in patients with persistent seizures after surgery.

rates in seizure-free patients approach that of the general population. This latter ἀnding, although the studies are neither controlled nor randomized, most strongly suggests that epilepsy surgery reduces the excess mortality imposed by epilepsy. Untreated, these patients should have had a much higher death rate, and it seems reasonable to hypothesize that deaths were prevented by successful surgical treatment. Equally striking is the fact that patients with persistent seizures after surgery die at a rate similar to that observed in medically refractory patients treated at epilepsy centers (Nashef et al. 1995; Tomson 2000).

910 Sudden Death in Epilepsy: Forensic and Clinical Issues 1.00

Cumulative proportion surviving

0.95

+

0.90

(

+

(

(

((

()

(

+ + + ++++

0.85

+

0.80 0.75 0.70 0.65 0.60 0.55 0.50

0

2

4

6

8

10

12

14

16

Years

Figure 56.1╇ Survival as a function of postoperative seizure recurrence after surgery for the entire cohort. Top line shows survival in patients who did not recur (n = 258, 1 death); lower line shows survival in patients who had recurrent seizures (n = 325, 18 deaths). Survival was significantly better in patients who did not experience recurrent seizures after surgery. (From Sperling, M. R., et al., Epilepsia 46, 49–53, 2005. With permission.)

This was observed even when there was a substantial reduction in seizure frequency postoperatively; some of the patients who died had only a few seizures per year (Hennessy et al. 1999; Sperling et al. 1999, 2005). Hence, any reduction in mortality following surgery appears to require elimination of seizures. Whether the patient with rare,€sporadic seizures after surgery has excess mortality only around the time of seizure recurrence is unknown. This poses methodological problems when assessing the effect of surgery on mortality. Since most patients have at least one seizure after epilepsy surgery if they€live for many years (McIntosh et al. 2004; Sperling et al. 2008), a partial rather than complete reduction in the excess mortality from epilepsy might be anticipated. 56.3.2â•… Effect of Laterality of Surgery on Mortality While none of the studies found that the side of resection influenced mortality, two studÂ� ies€noted an increase in mortality and SUDEP in a speciἀc subgroup of patients that had right temporal lobe resections with a pathological determination of mesial temporal sclerosis or gliosis. Hennessy and colleagues (1999) reported that patients with right-sided mesial temporal sclerosis had a postoperative mortality of 1 per 53 person-years (standardized mortality ratio = 32.0), whereas those with left-sided mesial temporal sclerosis had a signiἀcantly lower mortality of 1 per 310 person-years (standardized mortality ratio = 3.3). Nilsson et al. (2003) also observed that patients with pathologically conἀrmed gliosis who underwent right temporal lobe resection had a 2.5-fold increase in the standardized mortality ratio compared to the left, and a SUDEP incidence of 8.3 per 1000 person-years from the€right€temporal€subgroup, compared to no SUDEP occurrences from the left temÂ� poral subgroup€with€pathologically conἀrmed gliosis. This ἀnding was not reproduced in the€studies€by€Salanova and colleagues (2002) or Sperling and colleagues (2005). Surgery

Epilepsy Surgery and the Prevention of SUDEP

911

rendered€fewer€right-sided surgical patients in the Hennessy et al. (1999) study seizure free; thus, the increased mortality may have been related to seizure control. Moreover, the number of deaths in both series was small. At present, there appears to be insufficient data to conclude that right-sided epilepsy poses greater risk of death and further study is needed. 56.3.3â•… Effect of Gender of Surgical Patients on Mortality In the study by Sperling et al. (2005), gender did not influence death rates when assessed in person-years. However, the standardized mortality ratio has been reported to be higher in women than men (Nilsson et al. 2003; Sperling et al. 2005). This elevation in the standardized mortality ratio is likely an artifact of the method by which the value is calculated. In the general population, women have lower mortality rates than men; thus, the denominator used in calculating the standardized mortality ratio is considerably smaller for women than men. Consequently, while the absolute mortality rates are similar in men and women, the higher standardized mortality ratio in women is an artifact of the methodology. 56.3.4â•… Effect of Pathologic Diagnosis on Mortality Patients with progressive lesions or malignant tumors as the etiology of the seizures were generally excluded from the reviewed studies. Only one study (Hennessy et al. 1999) found that almost all of the epilepsy-related deaths occurred in surgical patients with either hippocampal sclerosis or nonspeciἀc ἀndings. This ἀnding was not conἀrmed in the Salanova et al. (2002) study, and the other studies did not assess the relationship between pathology and mortality. Further study is needed, therefore, to determine how underlying pathology affects mortality in surgically treated patients with epilepsy. 56.3.5â•… Effect of Cerebral Lobe on Mortality Only two of the studies included epileptic lesions beyond the temporal lobe (Nilsson et al. 2003; Sperling et al. 2005). Neither found a difference in mortality rates between patients with temporal lobe resections and those with extra-temporal resections. However, studies with more patients are needed to conἀrm this ἀnding since the relatively small number of deaths could mask a type 2 error and not discover an effect of lobe on mortality. A multicenter French trial has been proposed to compare the mortality of patients with a purely temporal epileptogenic zone to that of patients with a “temporal plus” epileptogenic zone (Ryvlin et al. 2005). 56.3.6â•… Comparison of Resection/Subpial Transection with Callosal Section One study (Sperling et al. 2005) reported that patients who underwent an anterior or complete corpus callosotomy had a higher late mortality rate than patients who had surgery for a focal epilepsy (p = 0.001). Most of these patients had symptomatic generalized epilepsy and, presumably, had signiἀcant underlying neurological impairment. This, itself, is a risk€factor for excess mortality (Forsgren et al. 1996). Comparing survival among patients with recurrent seizures after either callosal section or focal surgery, a nonsigniἀcant trend was observed for increased mortality in the patients after callosal section (22.6 per 1000 person-years, p = 0.06).

912 Sudden Death in Epilepsy: Forensic and Clinical Issues

56.4â•… Conclusion In the modern era, epilepsy surgery has a low morbidity and eliminates seizures in many patients with medically refractory seizures. The studies reviewed above, while with limitations, suggest that successful epilepsy surgery reduces the excess mortality in epilepsy, and the mortality rates in seizure-free patients can approach that of the general population. Data from medically treated patients conἀrm that even rare seizures carry a signiἀcant risk of mortality. Half of the patients who die suddenly have infrequent seizures, and nearly one-third have fewer than three seizures per year (Hirsch and Martin 1971; Leestma et al. 1989). These ἀndings are replicated in the surgical studies. Since it is unlikely that large-scale randomized studies will ever be performed to address the effects of epilepsy surgery on mortality, future investigations will need to study speciἀc questions that can be answered in surgical populations. Does limbic epilepsy convey a greater mortality risk than extra-limbic epilepsy? Since surgical patients are at high risk for SUDEP, studies might be conducted to attempt to elucidate the pathology of SUDEP. Perhaps, detailed autonomic testing might be performed before surgery so that more can be learned should SUDEP occur. Changes in autonomic function might be assessed before and after surgery as well, to explore effects of surgery on various autonomic parameters. Electrocardiographic analyses thus far of patients who subsequently died of SUDEP have been provocative but limited in scope (Nei et al. 2004; Nei 2009). For the present, it seems appropriate to counsel patients with uncontrolled epilepsy that surgery may lessen their chance of dying and that the risks of recurrent seizures may exceed the risks of surgery.

References Cockerell, O. C., A. L. Johnson, J. W. Sander, Y. M. Hart, D. M. Goodridge, and S. D. Shorvon. 1994. Mortality from epilepsy: Results from a prospective population-based study. Lancet 344 (8927): 918–921. Engel, Jr., J. 1996. Surgery for seizures. N Engl J Med 334 (10): 647–652. Engel, Jr., J., S. Wiebe, and J. French et al. 2003. Quality Standards Subcommittee of the American Academy of Neurology; American Epilepsy Society; American Association of Neurological Surgeons. Practice parameter: temporal lobe and localized neocortical resections for epilepsy: Report of the Quality Standards Subcommittee of the American Academy of Neurology, in association with the American Epilepsy Society and the American Association of Neurological Surgeons. Neurology 60: 538–547. Ficker, D. M., E. L. So, W. K. Shen et al. 1998. Population-based study of the incidence of sudden unexplained death in epilepsy. Neurology 51 (5): 1270–1274. Forsgren, L., S. O. Edvinson, L. Nystrom, and H. K. Blomquist. 1996. Influence of epilepsy on mortality in mental retardation: An epidemiological study. Epilepsia 37: 956–963. Hauser, W. A., J. F. Annegers, and L. R. Elveback. 1980. Mortality in patients with epilepsy. Epilepsia 21 (4): 399–412. Hennessy, M. J., Y. Langan, R. D. Elwes, C. D. Binnie, C. E. Polkey, and L. Nashef. 1999. A study of mortality after temporal lobe epilepsy surgery. Neurology 53 (6): 1276–1283. Hirsch, C. S., and D. L. Martin. 1971. Unexpected death in young epileptics. Neurology 21 (7): 682–690. Hughes, J. R. 2009. A review of sudden unexpected death in epilepsy: Prediction of patients at risk. Epilepsy Behav 14 (2): 280–287. Kwan, P., and M. Brodie. 2000. Early identiἀcation of refractory epilepsy. N Engl J Med 342 (5): 314–319.

Epilepsy Surgery and the Prevention of SUDEP

913

Leestma, J. E., T. Walczak, J. R. Hughes, M. B. Kalelkar, and S. S. Teas. 1989. A prospective study on sudden unexpected death in epilepsy. Ann Neurol 26: 195–203. Lhatoo, S. D., A. L. Johnson, D. M. Goodridge, B. K. MacDonald, J. W. Sander, and S. D. Shorvon. 2001. Mortality in epilepsy in the ἀrst 11 to 14 years after diagnosis: Multivariate analysis of a long-term, prospective, population-based cohort. Ann Neurol 49: 336–344. McIntosh, A. M., R. M. Kalnins, L. A. Mitchell, G. C. Fabinyi, R. S. Briellmann, and S. F. Berkovic. 2004. Temporal lobectomy: Long-term seizure outcome, late recurrence and risks for seizure recurrence. Brain 27 (Pt 9): 2018–2030. Nashef, L., D. R. Fish, J. W. A. S. Sander, and S. D. Shorvon. 1995 Incidence of sudden unexpected death in an adult outpatient cohort with epilepsy at a tertiary referral centre. J Neurol Neurosurg Psychiatry 58: 462–464. Nei, M. 2009. Cardiac effects of seizures. Epilepsy Curr 9 (4): 91–95. Nei, M., R. T. Ho, B. W. Abou-Khalil et al. 2004. EEG and ECG in sudden unexplained death in epilepsy. Epilepsia 45: 338–345. Nilsson, L., A. Ahlbom, B. Y. Farahmand, and T. Tomson. 2003. Mortality in a population-based cohort of epilepsy surgery patients. Epilepsia 44 (4): 575–581. Nilsson, L., B. Y. Farahmand, P. G. Persson, I. Thiblin, and T. Tomson. 1999. Risk factors for sudden unexpected death in epilepsy: A case–control study. Lancet 353 (9156): 888–893. Nilsson, L., T. Tomson, B. Y. Farahmand, V. Diwan, and P. G. Persson. 1997. Cause-speciἀc mortality in epilepsy: A cohort study of more than 9000 patients once hospitalized for epilepsy. Epilepsia 38 (10): 1062–1068. Ryvlin, P., A. Montavont, and P. Kahane. 2005. The impact of epilepsy surgery on mortality. Epileptic Disord 7 (S1): S39–S46. Salanova, V., O. Markand, and R. Worth. 2002. Temporal lobe epilepsy surgery: Outcome, complications, and late mortality rate in 215 patients. Epilepsia 43 (2): 170–174. Sperling, M. R., M. Nei, A. Zangaladze et al. 2008. Prognosis after late relapse following epilepsy surgery. Epilepsy Res 78 (1): 77–81. Sperling, M. R., A. Harris, M. Nei, J. D. Liporace, and M. J. O’Connor. 2005. Mortality after epilepsy surgery. Epilepsia 46 (Suppl 11): 49–53. Sperling, M. R., H. Feldman, J. Kinman, J. D. Liporace, and M. J. O’Connor. 1999. Seizure control and mortality in epilepsy. Ann Neurol 46 (1): 45–50. Sperling, M. R., M. J. O’Connor, A. J. Saykin, and C. Plummer. 1996. Temporal lobectomy for refractory epilepsy. JAMA 276 (6): 470–475. Stavem, K., and B. Guldvog. 2005. Long-term survival after epilepsy surgery compared with matched epilepsy controls and the general population. Epilepsy Res 63 (1): 67–75. Strauss, D. J., S. M. Day, R. M. Shavelle, and Y. W. Wu. 2003. Remote symptomatic epilepsy: Does seizure severity increase mortality? Neurology 60: 395–399. Téllez-Zenteno, J. F., R. Dhar, and S. Wiebe. 2005. Long-term seizure outcomes following epilepsy surgery: A systematic review and meta-analysis. Brain 128 (5): 1188–1198. Tomson, T. 2000. Mortality in epilepsy. J Neurol 247 (1): 15–21. Tomson, T., L. Nashef, and P. Ryvlin. 2008. Sudden unexpected death in epilepsy: Current knowledge and future directions. Lancet Neurol 7: 1021–1031. Walczak, T. S., I. E. Leppik, M. D’Amelio et al. 2001. Incidence and risk factors in sudden unexpected death in epilepsy. A prospective cohort study. Neurology 56: 519–525. Wiebe, S., W. T. Blume, J. P. Girvin, and M. Eliasziw. 2001. Effectiveness and Efficiency of Surgery for Temporal Lobe Epilepsy Study Group. A randomized, controlled trial of surgery for temporallobe epilepsy. N Engl J Med 345 (5): 311–318. Wieser, H. G., M. Ortega, A. Friedman, and Y. Yonekawa. 2003. Long-term seizure outcomes following amygdalohippocampectomy. J Neurosurg 98 (4): 751–763.

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP

57

Jane Hanna Rosemary Panelli

Contents 57.1 Introduction 57.2 SUDEP—Out of the Shadows 57.3 United Kingdom National Sentinel Clinical Audit of Epilepsy-Related Death (2002) 57.4 Fatal Accident Inquiry 57.5 Clinical Guidelines (2003, 2004) 57.6 Scottish Public Services Ombudsman—Case 200700075 (2009) 57.7 SUDEP—Assessment and Communication of Risk 57.7.1 The Evidence 57.7.2 Current Practice 57.8 Ethical, Legal, and Practical Arguments in Favor of Disclosure 57.8.1 Autonomy, Truthfulness, and Trust 57.8.2 Potential for Reduction in Fear 57.8.3 Adherence, Self-Management, and Shared Decisions 57.8.4 Disclosing a Material Risk—The Legal Right to Know 57.8.5 Reducing Trauma of Those Bereaved through SUDEP 57.9 Ethical, Legal, and Practical Arguments against Disclosure 57.9.1 Harms of Communication 57.9.2 Not All Patients Want to Share in Discussion and Decision Making 57.9.3 Denying the Patient’s Right Not to Know 57.10 SUDEP Information and Education—When and How? 57.10.1 Tell if Asked 57.10.2 High-Risk Patients Only 57.10.3 If a Patient Does Not Adhere to Treatment 57.10.4 Routine Discussion 57.10.5 Risk Communication 57.11 SUDEP and Bereavement 57.12 SUDEP Support Services 57.13 Summary References

915

916 916 917 917 918 918 919 920 920 921 921 922 922 923 924 924 924 925 926 926 926 926 927 928 929 929 930 931 932

916 Sudden Death in Epilepsy: Forensic and Clinical Issues

57.1â•…Introduction Sudden unexpected death in epilepsy (SUDEP) is slowly emerging from the shadows of neglect and medical uncertainty. As awareness of SUDEP spreads and research increases, risk factors are slowly being identiἀed although the causes remain elusive. Responses to SUDEP have included guidelines emphasizing quality medical care in epilepsy, not only to tackle SUDEP but also to enhance the well-being of all epilepsy patients. Few deny the importance of such a goal; however, it is apparent that when it comes to assessment and communication of SUDEP risks, there are divergent views on what deἀnes best practice. Ethical, legal, and practical arguments associated with this issue are all points of debate. Nevertheless, patients are increasingly included in the conversation about SUDEP and much remains to be learned about optimal timing and the method of this interaction. SUDEP support services have played a central role in the promotion of SUDEP awareness and action on behalf of people with epilepsy, and their friends and families. Activities include personal support and counseling for those directly affected by SUDEP, production of resource materials, training of health professionals, and facilitation of research, either directly or through fundraising. SUDEP advocacy groups have also provided speciἀc information and support to bereaved families around the time of the post mortem, thereby assisting the family while also facilitating the work of the forensic agency. This chapter discusses some of the developments in the SUDEP story over the past 15 years. The debate concerning legal and ethical aspects of the assessment and communication of SUDEP risk is explored, and options for the timing and method of disclosure are presented. The long-lasting effect of SUDEP on the bereaved is discussed, and the chapter concludes with a brief review of the role played by SUDEP support services.

57.2â•… SUDEP—Out of the Shadows SUDEP was well recognized in the early 20th century (Spratling 1904), along with identiἀed risks and recommendations for prevention; however, by the 1960s, a myth had gained hold in the medical literature that epilepsy was not fatal. “Patients with epilepsy had moved from asylums into the community and there was much less opportunity for observation. Risks from epilepsy were minimized then denied; that seizures could not be fatal became ‘common knowledge’ despite evidence to the contrary” (Nashef 1995). However, in the early 1990s, families bereaved by epilepsy began to ask questions of the scientiἀc community. In an era when modern medicine had mastered the techniques of organ transplants and in vitro fertilization, they began to ask why a young person in apparent good health, apart from epilepsy, would die unexpectedly with no explanation. In the United Kingdom the lack of medical knowledge and social support led bereaved families to join together for mutual support and to seek answers to the questions raised by SUDEP. In collaboration with the few clinicians who were interested in the issue, these motivated few pushed SUDEP onto the international scene as a serious topic for research and clinical practice. In 1996, the newly founded charity, Epilepsy Bereaved, convened the ἀrst ever international workshop on SUDEP. Those present identiἀed the need for research on all key aspects of SUDEP (Nashef et al. 1997). In addition, an information review concluded that SUDEP information was not easily accessible to patients at this time (Preston 1997). Today, the nature of the scientiἀc uncertainty on SUDEP has changed. While there are no

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP 917

identiἀed causes, research has led to advances in understanding the risk factors associated with SUDEP. In addition, community action groups for SUDEP have multiplied.

57.3â•…United Kingdom National Sentinel Clinical Audit of Epilepsy-Related Death (2002) Between September 1999 and August 2000, a nationwide audit of epilepsy-related deaths was conducted in the United Kingdom. This National Sentinel Clinical Audit of EpilepsyRelated Death (Hanna et al. 2002) aimed to establish whether deἀciencies in the standard of clinical management, or in the overall health care package, could have contributed to the deaths. The audit found signiἀcant problems of access to epilepsy services and reviews as well as medication issues, concluding that there was signiἀcant potential for avoidance of death if these risks were addressed. Others concluded that if a similar audit were to be repeated in the United States, the results would probably be the same because of problems of access to services (Pedley and Hauser 2002). The audit motivated policy makers in the United Kingdom to develop government action plans on epilepsy (Department of Health 2003; Welsh Assembly Government 2009) and to fund the development of patient information on risk, involving patients who valued the inclusion of information on SUDEP.

57.4â•… Fatal Accident Inquiry In the same year as the audit of epilepsy-related deaths, a Fatal Accident Inquiry was held into the sudden unexpected death of a young woman, aged 17 years (Taylor 2002). This type of judicial inquiry is held in Scotland by a judicial office holder called a Sheriff where a death concerns the public interest. The purpose is to make recommendations to prevent future deaths. The deceased’s mother had died from epilepsy in 1988, and in 1991 the deceased presented with seizures. The family was reassured that the deceased suffered from benign focal seizures of childhood and she was discharged by the specialist to general practice on antiepileptic drugs. From 1991, she had four to ἀve seizures yearly, varying in frequency and severity, but there was no annual review or re-referral for specialist care. The court found a catalogue of failures to look after the deceased in a proper manner. These included a failure on the part of specialists to alert the general practitioner as to what circumstances required re-referral; a failure on the part of the general practice to prescribe appropriate levels of medication; a failure to reâ•‚refer the deceased when the seizures did not stop after 2€years; a failure to re-refer the deceased when the intensity, form, and duration of her seizures altered as the deceased matured; and a failure by the medical team to discuss with the deceased’s family the diagnosis, the attendant risks, and how these risks might be properly managed. The sheriff stated that given the association between seizures and SUDEP and the potential for control, that it was a “short step” to the view that if the€deceased had been€referred for review she might not have died. He determined that the family ought to have been informed that the deceased was suffering from epilepsy, the risks of SUDEP explained, and a discussion held on how her condition might be managed. The most important recommendation was considered to be the need for a personal care plan. The sheriff suggested that all the key issues would have been addressed if a care plan,

918 Sudden Death in Epilepsy: Forensic and Clinical Issues

“. . . shared or otherwise” had been produced, and “. . . it might have saved her life” (Taylor 2002). The sheriff was clear that for the purposes of the public inquiry, it was not necessary for there to be any scientiἀc certainty. Any legal judgment is determined by balance of probabilities, and in this case the concern was whether preventative measures “might” have saved a life. The sheriff was clear that information on SUDEP could be relevant to how proactive a family might be in probing decisions about treatment as well as informing discussions on how risks might be reduced. The sheriff accepted that the question of informing about SUDEP must be left to the discretion of the medical profession to form a view. In particular, he accepted there might be people of “an extreme disposition” where discussion might cause harm. Nevertheless, he said, “I do, however, accept that in the vast majority of cases there should be such a discussion” (Taylor 2002).

57.5â•… Clinical Guidelines (2003, 2004) Clinical guidelines in Scotland and those produced by the National Institute of Health and Clinical Excellence (NICE) for England and Wales now recommend discussion of SUDEP as part of general epilepsy information (Scottish Intercollegiate Guidelines Network 2003; Stokes et al. 2004). The NICE guideline sets out information to show why preventing seizures is important. It is based on the view that SUDEP risk can be minimized by optimizing seizure control and by families being aware of the potential consequences of nocturnal seizures. It also recommends tailored information on the individual’s risk of SUDEP as part of any counseling checklist for people with epilepsy, and their families and/or care givers.

57.6â•… Scottish Public Services Ombudsman—Case 200700075 (2009) In 2009, 7 years after the Scottish Fatal Accident Inquiry, ongoing concerns with epilepsy care in the United Kingdom were highlighted by another public investigation, carried out by the Scottish Public Services Ombudsman. The Ombudsman is an individual appointed by the Queen on the nomination of the Scottish Parliament (www.spso.org.uk/about-us). The investigation into the death of an 18-year-old woman upheld the complaint of the deceased’s family that there was a failure in patient-centered care following a withholding of information on SUDEP (Scottish Public Services Ombudsman 2009). Although the recommendations of the ombudsman are not legally binding, they are normally followed as their purpose is not only to provide justice for the individual but also to improve the delivery of public services. The deceased was due to leave home for the university when she was diagnosed with tonic–clonic seizures and prescribed antiepileptic drugs following a nocturnal seizure. On a second review some 5 months later, no further seizures were reported, and the medication was reduced because of concerns about side effects. The family had reason to be concerned that she was not adhering to treatment following her move to the university. The deceased died suddenly in her sleep shortly after, and toxicology testing at postmortem showed no evidence of antiepileptic drugs. A complaint was brought by the family that the deceased was not provided with sufἀcient information on her treatment and the risk of SUDEP to inform her decisions in self-

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP 919

management of her condition. The complainant argued that information about SUDEP should have been given to the deceased, unless the clinician had good reason not to. The clinician argued that to give information could cause undue distress and there was nothing that the deceased could do to prevent her death. Therefore SUDEP information should only be given when there was good reason to do this. The ombudsman accepted that there was signiἀcant divergence in view between clinicians on this question and described these as the “proactive” approach supporting the complainant’s view and the “reactive” approach supporting the clinician’s view. Because of this divergence the ombudsman was not able to form any conclusion on the issue of modiἀcation of speciἀc SUDEP risks or on the issue of patient harm from disclosure of risk, and has urged that both are the subject of research as a matter of priority. Despite this the ombudsman upheld the complaint on the grounds that there was no evidence in the record to suggest the clinician had made an individualized decision to withhold information on SUDEP, and that this amounted to a failure to provide individualized patient care by the clinician and the board of the hospital. The ombudsman noted that the intention of the Fatal Accident Inquiry (Taylor 2002), the ethos of the health service, general professional guidance from the regulatory body, and the plain reading of clinical guidelines on the epilepsies pointed in the direction of information giving as the norm. She recognized that while deviation may be appropriate in some cases, it should be recorded. The ombudsman found that the signiἀcance of the particular guideline was not determined simply by the level of evidence supporting it, and in this case the decision to withhold the information should be recorded. Further ἀndings were a failure to provide written information regarding SUDEP and other aspects of the medical condition. The ombudsman noted the lack of research into the effect of SUDEP awareness on bereaved families and made mention of reports from support groups indicating that SUDEP information helps families to better deal with the tragedy. She commented that in the case under investigation, the family could not help but wonder: . . . how they might have altered the outcome if they had only known this information. Mrs C accepted that Miss C may still have died even if they had known about SUDEP, but because there is the suggestion of the chance she might not have, they are not able to ἀnd any peace. (Scottish Public Services Ombudsman 2009)

The Health Board acted on the complaint in respect of the provision of written epilepsy information to patients following diagnosis and the appointment of an epilepsy specialist nurse, but has not accepted the ombudsman’s ἀnding on the failure to evidence an individualized decision to withhold information on SUDEP. The ombudsman will be drawing the attention of the Scottish Government to the disagreement and has referred her report to the national guidance body in Scotland.

57.7â•… SUDEP—Assessment and Communication of Risk The report of the Scottish Public Services Ombudsman (2009) highlighted an ongoing lack of consensus on how SUDEP should be addressed in epilepsy management. Despite the increase of research data, the development of some clinical guidelines that include SUDEP (Scottish Intercollegiate Guidelines Network 2003; Stokes et al. 2004), and adjustments to practice that have occurred worldwide, inconsistencies in care persist. Factors influencing

920 Sudden Death in Epilepsy: Forensic and Clinical Issues

opinion include the perceived avoidability of SUDEP; the estimation of risk for individuals; and the ethical, legal, or practical issues that are related to patient disclosure. This section aims to summarize the evidence for prevention of SUDEP before addressing current practice and the main arguments in favor and against communication of risk. 57.7.1â•… The Evidence Although there is recognition that more research is needed on risk factors, there is now a strong body of evidence supporting the association between SUDEP and seizures, especially generalized tonic–clonic seizures. Despite some contradictory voices, a body of medical opinion has been building over a number of years that some SUDEP deaths are potentially avoidable (Faught et al. 2008; Hanna et al. 2002; Hitiris 2007; Hughes€2009;€Monte et al. 2007; Opeskin and Berkovic 2003; So 2006; So et al. 2009; Surges et al. 2009; Tomson et al. 2008). Recent reviews (Hughes 2009; Monte et al. 2007; Tomson et al. 2008) have concluded that there is evidence of varying degrees for risk factors that include seizures (especially generalized tonic–clonic seizures), young age, early onset of epilepsy, absence of treatment or nonadherence to antiepileptic drugs, polytherapy, the prone position, being in the bedroom, sleeping, and being male. The effect of nocturnal supervision was examined in only one case–control study but was found protective (Langan et al. 2005). It has been suggested that the precautionary approach of the sudden infant death syndrome (SIDS) risk reduction campaign is a suitable model for SUDEP (So et al. 2009). SIDS is a rare event with no one cause identiἀed, yet during the past two decades many countries around the world have launched successful public campaigns informing about its risk factors (Moon et al. 2007). A precautionary approach for SUDEP would support prevention strategies based on optimization of seizure control, effective antiepileptic drug treatment, supervision in appropriate cases, and information to patients and care givers. This precautionary approach is supported by a reasonable body of policy makers (Department of Health 2003; Stokes et al. 2004; Welsh Assembly Government 2009). Most recently a report of the American Epilepsy Society and the Epilepsy Foundation Joint Task Force on SUDEP supported the view that certain risk factors associated with SUDEP “may” be modiἀable— namely, uncontrolled seizures, subtherapeutic drug levels, and the number of antiepileptic drugs used. The report supported optimization of seizure control as a treatment goal. It also noted the need to raise awareness of SUDEP in both medical and lay communities and to establish frameworks for the guidance of future research (So et al. 2009). 57.7.2â•… Current Practice A survey of clinical practice asked neurologists in the United Kingdom if they told patients about SUDEP (Morton et al. 2006). The neurologists who replied reported a wide variation in practice. While 30.3% of respondents reported that they discussed SUDEP with the majority or all of their patients, 68.7% discussed with very few or none of their patients. Interestingly clinicians with a special interest in epilepsy were signiἀcantly more likely to discuss SUDEP. The paper does not quantify at any point what was meant by the terms “few” or “many” used in the questionnaire, and statistics given cannot be considered anything other than a general indicator (Scottish Public Services Ombudsman 2009). Those who support routine discussion of epilepsy-related death suggest that informing of the small risk of fatality associated with seizures sits appropriately with an early discussion of

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP 921

risk (Cook 2005; Tomson et al. 2008). Such a discussion would not necessarily require the inclusion of the term SUDEP. Specialist nurses have been shown to be more likely than clinicians to discuss SUDEP. A recent survey in the United Kingdom found that 56% of specialist nurses discussed SUDEP with the majority of their patients. Most combined SUDEP with discussion of general and speciἀc risks of epilepsy (Lewis et al. 2008). No specialist nurse supported nondisclosure.

57.8â•… Ethical, Legal, and Practical Arguments in Favor of Disclosure The main arguments for disclosure are that it supports patient autonomy, informs choices on adherence and self-management, reduces fear by managing natural anxieties, and avoids the harm of false assurance. Research repeatedly tells us that people with epilepsy feel they receive insufficient information about their condition (Couldridge et al. 2001). A recent qualitative study found that patients felt a pressing need for information, including “. . . more concrete answers on SUDEP,” to discuss treatment options and medication in detail (Prinjha et al. 2005). This suggests that clinicians need to check and regularly revisit with patients how much information they would like about their condition. The perspective of care givers is also a consideration. Research shows that patient decision making involves an extended social context (Fraenkel and McGraw 2007); however, in any event, care givers may need accurate information in their own right. Families affected by SUDEP described a variety of roles associated with care such as keeping records of seizures and responding to seizures. Many considered that they needed more information on SUDEP as it pertains to these roles (Kennelly and Riesel 2002). One study found parents expected this (Gayatri et al. 2010). 57.8.1â•… Autonomy, Truthfulness, and Trust The ethical principle of patient autonomy involves the patient’s right to know about his or her own medical condition and prognosis. The American Epilepsy Society and the Epilepsy Foundation Joint Task Force on SUDEP considered discussion of SUDEP as consistent with the ethical principle of patient autonomy and also consistent with the need to accept that some persons with epilepsy have increased risks of morbidity and death (So et al. 2009). Examples of professional guidance based on the autonomy principle fully support the patient’s right to information about their condition. The amount of information given on a condition should depend on patient wishes, and need, for information. This should be determined by discussion with a patient and not be based on assumptions of what patients require (General Medical Council 2008). Provision of information is also important to the relationship of trust between doctor and patient. Where information on a risk is withheld and that risk materializes, it is natural that those affected by the consequences will seek to understand why information was not shared. They may experience the harm of false assurance. There is some limited evidence on the negative impact of false assurance in the context of medical screening, including public conἀdence and legal action (Petticrew et al. 2001); however, the impact of most seizures being presented as benign (apart from accidents and status epilepticus) has not been researched. Relatives bereaved through SUDEP frequently report, however, that their grief is exacerbated because epilepsy was presented in this way and it proved to be false

922 Sudden Death in Epilepsy: Forensic and Clinical Issues

(Kennelly and Riesel 2002). Further, the relationship of trust between the bereaved and the medical professional is more likely to be maintained where information withholding is patient-centered and withstands scrutiny. 57.8.2â•… Potential for Reduction in Fear A common assumption is that discussing SUDEP will create anxieties. In fact the only research on epilepsy and death anxiety comprises studies looking at the death anxiety associated generally with epilepsy. A study of 373 epilepsy patients found that approxÂ� imately two-thirds harbored fears of death from their next seizure (Mittan 1986). In a more€recent cross-sectional study of 92 patients having epilepsy for a minimum of 5 years, 56.52% of patients had either moderate or high death anxiety (Otoom et al. 2007). Not surprisingly, death anxiety was likely to be higher in patients with generalized epilepsy. Otoom et al. (2007) emphasize the importance of counseling patients to reduce anxiety. Anecdotal reports from clinicians who regularly talk about SUDEP mention the potential for reduction of fear by putting fears into perspective. Indeed, there is some evidence that this may be so. A case–control study on the beneἀts of a weekend educational program (SEE, known then as the Sepulveda Epilepsy Education program but more recently as Seizures and Epilepsy Education), which included discussion of risk of mortality, found a subsequent decrease in anxiety (Helgeson et al. 1990). The experience of this program was that lack of discussion of mortality led to adverse suppression of natural anxiety. Fear was also associated with overprotection and overcontrol of the person with epilepsy (Mittan 2005). The program was therefore concerned to ensure that patients and their care givers had a more realistic appreciation of the risk. 57.8.3â•… Adherence, Self-Management, and Shared Decisions Nonadherence to prescribed medication regimens in epilepsy is associated with a more than threefold increased risk of mortality (Faught et al. 2008). A recent review of all SUDEP studies since 1975 found 18 studies where subtherapeutic drug levels were signiἀcant and only 1 study that claimed that this factor was not signiἀcant (Hughes 2009). Research across chronic conditions has consistently found that lack of adherence is associated with patient doubts about their need for medication and concerns about side effects (Horne 2006). Patients with asthma, for example, were signiἀcantly more likely to endorse the need for regular medication if they shared the “medical view” of asthma as an “acute or chronic condition” with potentially serious consequences. If, on the other hand, patients saw their condition as chronic only, they were more likely to doubt the need for regular medication. The implications for asthma treatment were that it is not sufficient to advise a patient to take medication, but that a clear rationale is needed. This issue has not been researched in epilepsy. However, in a recent study, epilepsy specialist nurses perceived a positive association (Lewis et al. 2008). Other issues of self-management include reporting of seizures, avoiding triggers to seizures, lifestyle choices, and supervision. The risk of death following a seizure might be signiἀcant to some patients in balancing the risks and beneἀts of such behaviors. For example, a study of general practice found that 40% of patients who had anonymously reported seizures to a questionnaire, but not to their general practitioner, held a driving license. This suggests that patients perceive the beneἀts of concealment to outweigh what they understand as the risks (Dalrymple and Appleby 2000). Two deaths were reported

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP 923

by bereaved families in people choosing not to visit a doctor because of potential loss of license (Kennelly and Riesel 2002). The severity of any potential harm is recognized as a key influence on health behavior and decision making (Weinstein 1999). If patients and care givers do not appreciate that avoidance of fatality is one motivation for treatment, it raises the question of whether this lack of information is signiἀcant to patient decisions about adherence and self-management. Chronic illness literature supports and promotes key themes of self-management and shared decision making (Department of Health 2001a; Lorig and Holman 2003; National Health Priority Action Council 2006; Whitney et al. 2008). Patient participation fulἀlls the ethical principle of patient autonomy, and when clinicians deny patients a choice that should be theirs, it causes harm (Katz 1984). In some countries supporting this approach, detailed professional guidance has been developed to mark the shift toward patients and doctors making decisions together and reflects exchange of information between patients and doctors as central to decision making (Nunes et al. 2009). This should include information on diagnosis, prognosis, and uncertainties, plus information on the potential beneἀts, risks, and burdens of treatment and of not treating (General Medical Council 2008). 57.8.4â•…Disclosing a Material Risk—The Legal Right to Know A legal duty to discuss a risk prima facie arises where there is a signiἀcant risk that would affect the judgment of the reasonable patient (Bolitho v City and Hackney Health Authority 1998; Pearce v United Bristol Healthcare NHS Trust 1999). Risks that are only remote possibilities can be regarded as material if the severity of the risk materializing is very serious. In the context of a SUDEP death, it could be argued that there is a material risk to be addressed where discussion might influence behavior, improve compliance, be relevant to issues of supervision, or where optimum treatment is not in place (Beran 2006). The courts in many countries will test the medical evidence offered by parties in litigation to reach their own conclusions on the magnitude and severity of risk, and the ease by which the risk might be avoided (Bolitho v City and Hackney Health Authority 1998; Rogers v Whitaker 1992; Videto v Kennedy 1981). The existence of guidelines, while not predictive of negligence, would be a relevant consideration in examining the rationality of withholding information. It has been argued that a claim would fail in a SUDEP case because of the need to prove causation between the actions of the physician and the death (Beran et al. 2004). However, the more recent authority of Chester v Afshar (2005) has extended the law of causation to include the scenario where risk is inherent in a condition and not caused by medical intervention. This would be a helpful legal authority in a SUDEP case. Although there has been no litigation into this question, a judicial determination was made in the Fatal Accident Inquiry (Taylor 2002). The inquiry concluded that patients should be told unless there is good reason not to. The sheriff did not specify precisely what such reasons would be, leaving this to the exercise of reasonable discretion by the medical practitioner. I stop short of recommending that the risk of sudden and unexplained death from epilepsy .€.€. should be discussed with every epilepsy sufferer or their family. . . . Discretion must be left with the medical profession to form a view. There may well be families who are of an extreme disposition or perhaps who have learning difficulties, where such a discussion would do more harm than good. I do, however, accept that in the vast majority of cases there should be such a discussion. (Taylor 2002)

924 Sudden Death in Epilepsy: Forensic and Clinical Issues

It is worth noting, however, that no actions for negligence have yet been reported in cases involving SUDEP deaths, and instead systems for investigation in the public interest have been activated by some families affected by SUDEP. The two Scottish cases detailed above both led to changes in service provision. Following the 2002 Fatal Accident Inquiry, an epilepsy clinic to review patients with epilepsy was established by the general practice and the 2009 Ombudsman’s Report makes reference to the creation of a new epilepsy specialist nurse post. 57.8.5â•…Reducing Trauma of Those Bereaved through SUDEP When SUDEP occurs and there has been no prior disclosure of risk, the justiἀcation for withholding information on grounds of causing anxiety or the lack of direct evidence of prevention is unlikely to withstand scrutiny by those who are suddenly bereaved. A relationship of trust with the particular medical team and with the broader medical profession may inevitably be placed under some strain at this time. A qualitative study involving 78 people affected by SUDEP reported that many found the lack of SUDEP information exacerbated their feelings of grief (Kennelly and Riesel 2002). As noted above, the aggravation of grief for families that lacked information before a SUDEP death was a point of discussion for the Scottish Public Services Ombudsman (2009) during a recent inquiry.

57.9â•… Ethical, Legal, and Practical Arguments against Disclosure The main arguments against disclosure are (1) that telling patients will cause anxiety for no beneἀt; (2) that there is no research on patient wishes in this area and that patients may not want to hear about SUDEP; and (3) that a clinician may be successfully sued for denying a patient “the right not to know” the risks associated with a condition. 57.9.1â•…Harms of Communication A therapeutic privilege exists in medical care to withhold information where it would cause harm to the patient. This concept is recognized by medical law as a defense to a legal action based on failure to disclose a material risk. Therapeutic privilege is normally conἀned to a psychiatric setting. What is clear from professional guidance and from legal precedents is that therapeutic privilege does not mean unfettered discretion to decide one way or another, but instead is conditional on a process of rational decision making. The General Medical Council of the United Kingdom (2008), for example, states that such information should not be withheld, and that it is necessary for making decisions unless the clinician believes the patient would be caused serious harm beyond being upset or refusing treatment. Any decision to withhold information should be recorded, justiἀed, and reviewed. The concern about the harm of raising patient anxiety is raised regularly in the medical literature on SUDEP as the reason for not providing information. Temporary anxiety in a patient is not normally viewed by the courts or ethical bodies as sufficient to constitute sensible medical grounds for withholding information (Deriche v Easling Hospital NHS Trust 2003). The only related research on this subject is a survey of neurologists in the United Kingdom where doctors were asked to report on how patients reacted to being told about SUDEP (Morton et al. 2006). About one-third of doctors did not respond to this

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP 925

question, suggesting perhaps the difficulty of assessment. Some reported that telling such information to patients was difficult. One-third of respondents thought that information about SUDEP caused anxiety, but they were not asked how they assessed this, whether they followed up to check if anxieties were lasting, and whether the patient was offered or responded well to further information and support. Interestingly, doctors who discussed SUDEP with the majority of their patients were less likely to report negative reactions, suggesting, as the author recognized, a practice effect. The lack of evidence on patient harm from discussion of SUDEP was highlighted in the recent Scottish investigation. The Scottish Public Services Ombudsman (2009) declared that “much of the evidence in this area is . . . at best anecdotal and any reliance on an assumption about patients’ reactions must be tempered by the lack of actual hard evidence.” Outside of the epilepsy ἀeld, there is some evidence in the literature from which to draw, including a systematic review and a meta-analysis that provide some general indication of the relationship between bad news and anxiety. The evidence suggests that anxiety is a common and adaptive initial response to “at risk” notiἀcation, but it usually dissipates within a month (Marteau 2008; Shaw et al. 1999). Psychological theories of self-regulation describe the ways that humans maintain equilibrium while responding to threat (Taylor 1991). Clearly, fear unchecked can become negative and restrictive, but natural anxiety should not be assumed as harmful. 57.9.2â•…Not All Patients Want to Share in Discussion and Decision Making It is understood that patients vary in personality and coping styles, and therefore in their attitude to information and how they use it to navigate health issues (Andrewes et al. 1999; Politi et al. 2007). Some may not want to be involved in aspects of discussion or decision making with their health care providers. In the SUDEP debate, those who argue for the right not to know highlight differing patient-information-seeking behavior such as active searching (seekers), conscious blocking (avoiders), and a combination of styles (weavers) (Morton et al. 2006; Pinder 1990), expressing concern that where broad requirements to discuss SUDEP with all patients are mandatory, preference is therefore given to the needs of seekers above all others. But this need not be so. Where guidelines do exist, there are options for variation. Patients’ wishes are clearly highlighted in ethical literature on information giving, and clinical guidelines are always recognized as subject to clinical discretion where there is good reason not to follow a guideline. The General Medical Council (2008) advises doctors to provide patients with appropriate information, which should include an explanation of any risks to which they may attach signiἀcance. Doctors must not make assumptions about a patient’s understanding of risk or the importance they attach to different outcomes. They should discuss these issues with their patient. A clinician who probed patient wishes on information provision would have good reason not to inform, if the patients indicated that they did not want to have full information including risks of seizures; if the issue was regularly revisited; and if there was no material reason to override this, such as nonadherence to treatment. Pinder (1990) investigated information provision using Parkinson’s disease as the case study. She found that clinical practice on information giving was rarely determined by an accurate assessment of what the patient wanted, but was largely dictated by clinician assumptions as to what patients did or did not want to know. Clinicians were broadly divided into three groups of information-giving style: closed, open, or changing. Pinder concluded that

926 Sudden Death in Epilepsy: Forensic and Clinical Issues

identifying patient wishes necessitates proactive engagement by a clinician to discover what the patient wants to know, and ought to know, about their condition and its treatment. However, the medical literature suggests that many clinicians have taken speciἀc positions on when to discuss SUDEP, such as only if a patient asks, at the time of prescribing antiepileptic drugs, or only in cases with recognized risk factors (Morton et al. 2006). 57.9.3â•…Denying the Patient’s Right Not to Know Some authors contributing to the SUDEP debate have put forward the argument for a patient’s “right not to know” (Beran 2006; Black 2005), even suggesting that in some circumstances clinicians could be sued if they tell patients about SUDEP (Beran et al. 2004). While therapeutic privilege provides a defense to the legal right to know where there is good reason that discussing SUDEP with a patient will cause harm, the concept of a legal right not to know, although recently mooted in the context of routine HIV testing and genetics, is not one that has rooted yet in any established legal concept or legal authority.

57.10â•… SUDEP Information and Education—When and How? Where the decision is made to inform patients about SUDEP, the timing and method of communication continue to be points of debate. 57.10.1â•… Tell if Asked The literature on this question supports disclosure should a patient ask a question relevant to SUDEP (Beran et al. 2004; Black 2005; Brodie and Holmes 2008; Morton et al. 2006; So et al. 2009). However, this approach clearly places the onus of asking upon patients who may not know which questions to frame. Consequently, disclosure is likely to be restricted to those individuals who are already well informed and conἀdent enough to seek information. 57.10.2â•… High-Risk Patients Only It is generally agreed that SUDEP information should be provided to high-risk groups, such as those with poorly controlled seizures, allowing them the opportunity to participate in risk assessment (Brodie and Holmes 2008; Tomson et al. 2008; So 2006). This may include screening for cardiac interventions and discussion of options for supervision (Finsterer and Stollberger 2009; Tomson et al. 2008, 2009). Where patients have epilepsy and learning difficulties, professional guidelines recommend that health professionals be aware of the higher mortality risks and discuss these with the individuals affected, and their families and/or care givers (Stokes et al. 2004). For people with newly diagnosed or mild epilepsy, risk may be lower, but SUDEP deaths do occur in this group as noted by the Scottish Public Services Ombudsman (2009). Nevertheless the assumption has prevailed that SUDEP is a risk for only those with intractable epilepsy (Tomson et al. 2008), which, to some extent, ignores “. . . the larger group of individuals with better but not fully controlled epilepsy who have a lower, but nevertheless real, risk of SUDEP” (Tomson et al. 2008). In the clinical audit of epilepsy-related death in

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP 927

the United Kingdom, a number of SUDEP cases were recorded in patients having only rare seizures. Looking at 158 adult patients who had attended specialist care, 5% were known to have had less than one seizure per year, 3% were people noted to have had a single seizure or the ἀrst in many years, and 4% were people considered to be seizure free. Twenty-ἀve percent of the deaths were in people whose records did not record the frequency of seizures, and 22% were in people where the records were not clear. Six percent of the patients who died were not treated with antiepileptic drugs. The audit also identiἀed some 45 people who died during the year who were never referred to secondary care or who died waiting for referral (Hanna et al. 2002). 57.10.3â•…If a Patient Does Not Adhere to Treatment There is general agreement that SUDEP discussion should take place with the nonadherent patient (Black 2005; Morton et al. 2006; So et al. 2009). The challenge is timely identiἀcation of nonadherence. Research in other chronic conditions has shown that it is often difἀcult for a doctor to know who will comply with treatments as prescribed and why (Horne 2006; Wertheimer and Santella 2003). Infrequent reviews limit monitoring. Further, during the consultation, patients often overestimate adherence so as not to disappoint (Horne 2006). Waiting until the patient is known to be nonadherent may be too late, or it may be impossible to change behaviors. The report of the Scottish Public Services Ombudsman (2009) is illustrative as the family reported that the deceased was not taking medications as prescribed, whereas the clinical record noted concerns by the patient about side effects, leading to a reduction in medication, but not problems with adherence. The issue of adherence is not straightforward (Horne 2006; Wertheimer and Santella 2003). A recent guideline for the United Kingdom suggests that medicine taking is “. . . a complex human behavior . . .â•›,” and that unwanted and unused medicines “. . . reflect inadequate communication between professionals and patients . . .â•›” (Nunes et al. 2009). The complexities of adherence are constantly underlined for those of us working with SUDEP. For example we regularly see a troubling scenario, aspects of which are reflected in the stories of David, Celine, and Peter presented in Sudden Unexpected Death in Epilepsy: A Global Conversation (2005). A young person dies, and in discussions with family and friends a picture emerges of a vibrant individual who did not want to have epilepsy. Seizures were embarrassing and a nuisance, but they believed coping well meant not to fuss, and friends were told not to worry. It appears that their doctors may have considered them to be wellcontrolled, adherent, informed patients. However, these patients and their families did not realize epilepsy could be fatal. Frequently, the bereaved families believe that the young adults would have handled their epilepsy just a little differently if they had known .€.€.€. and of course no one can answer this question. We know that issues of career, driving, and the adverse effects of medication are all questions for young people. They may choose to selfmanage in their own way, based on the understanding they have been given. This implies that some coping styles, while apparently successful, if uninformed, may in fact lead to greater vulnerability. Whether an individual patient is perceived to be of high or low risk, behavior cannot be predicted. Private decisions take place outside of the consulting room. Life changes occur, and people with epilepsy will make decisions based on the framework of knowledge they have been given, with some choices leading to fatal outcomes. For example, when a young woman unexpectedly becomes pregnant, she may decide to stop her epilepsy

928 Sudden Death in Epilepsy: Forensic and Clinical Issues

medication without consulting any doctor. National investigations of maternal deaths consistently raise concerns that many pregnant women with epilepsy are concerned about the side effects of medication, and there are a steady number of women who die each year who appear to have stopped medication, with and without the knowledge of their medical team. The importance of preconception counseling for women with epilepsy of child-bearing age is a key recommendation of the report into all maternal deaths released in the United Kingdom (Lewis 2007). Since not all pregnancies are planned, it is imperative that such information must be given at the ἀrst opportunity, with frank two-way communication underpinning true physician–patient concordance (Horne 2006) and hopefully engendering informed patient adherence. 57.10.4â•…Routine Discussion The National Guidelines for England and Wales support discussion about SUDEP following diagnosis, tailored to the individual’s relative risk (Stokes et al. 2004). Routine discussion is also recommended by a growing body of expert opinion (Brodie and Holmes 2008; So et al. 2009; Tomson et al. 2008). With evidence that maintenance of a stable antiepileptic drug regimen might reduce risk, it would be timely for discussion about risks and beneἀts of treatment to include the small risk of fatality from a seizure (Cook 2005; Faught et al. 2008; Hughes 2009; Tomson et al. 2008). In the words of a person with epilepsy, “for many, the decision to take medication is a huge one. If they are not aware of the dangers of seizures as well as the side-effects of medication I do not feel that their decisions are truly informed” (Kearton 2005). Some patients may prefer information in stages, while others may desire as much information as possible on the condition. Strategies for identifying individual preferences may be the use of counseling checklists with SUDEP included, or the provision of preconsultation written epilepsy information (Stokes et al. 2004). Where investigation and diagnosis are delayed following a seizure, although it may be considered inappropriate to mention SUDEP (So et al. 2009), it has been recommended that essential information include how to recognize seizures, ἀrst aid, and the importance of reporting further attacks. Some information on epilepsy routinely imparted to patients may create major disincentives for engagement with treatment. The concern over possible loss of a driving license, or unpleasant side effects, for example, may inhibit reporting of seizures or attendance at appointments. It is logical to fully inform patients about risks soon after diagnosis when there is possibly the greatest potential to communicate the imperative of aiming for seizure freedom through appropriate treatment and lifestyle choices. As time passes, if few seizures are experienced with no apparent harmful effects, there is the potential for patients to become blasé about seizures and more concerned about the daily inconvenience of treatment and lifestyle adaptation. Focusing on comprehensive epilepsy education as the framework for SUDEP discussions is a positive recommendation (So et al. 2009), although models and evaluation of such programs are limited. The Sepulveda Epilepsy Education project (now known as Seizures and Epilepsy Education) was one early interesting example, in that fear was seen as a key issue to tackle even before SUDEP became a topic of public discussion (Helgeson et al. 1990). However, development of such programs is lacking. Interestingly the SUDEP discussion has injected some urgency into discussion on how epilepsy education generally should take place, something which is long overdue (Prinjha et al. 2005).

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP 929

It is understood that individuals vary in their approach to health information. Some seek it to cope well while others cope by avoidance (Andrewes et al. 1999; Pinder 1990; Politi et al. 2007). Clearly health information giving must aim to minimize anxiety and maximize the effectiveness of communication. This already occurs in epilepsy where risk communication is a core aspect of management. Rather than excluding SUDEP on the premise that this will avoid patient anxiety, why not include the information sensitively as part of the existing education? There may be no need to use the term SUDEP if the clinician can convey the substance of the risk. For example, the rare risk of dying from a seizure, not only through drowning and accidents but also following a seizure, can be mentioned (Cook 2005; Tomson et al. 2008). Additional information can be provided if patients request it. In some cases, gentle questioning of patients as to their fears may indicate whether questions lie unanswered. A patient’s attitude to information and involvement is likely to vary over time so that revisiting information topics is important (Pinder 1990). For those whose risk is low, a frank explanation of what is known may be reassuring (Mittan 2005; So et al. 2009). Individualization of information does not mean ἀltering or excluding information. SUDEP is in the public domain. An internet search will bring many hits, and not all offer rational, accurate material. The use of the internet grows daily (Escoffery et al. 2008; Fox 2007; Lu et al. 2005), and patients who discover SUDEP themselves may feel disappointed or even angry that their physician did not mention it. Trust may be damaged, and attempts to put the issue in context may no longer be as well received by the patient (Cook 2005). Where physicians routinely offer advice on SUDEP, they report few patient concerns, which suggests that with practice, providing such information to patients is successful (Morton et al. 2006). 57.10.5â•…Risk Communication If a message is to be meaningful to the recipient, it must be in a language that can be understood; however, in health care, this is not always so. For example, research literature on lay understanding of any health risks shows that the public has difficulty with odds and percentages (Weinstein 1999). Weinstein (1999) argues that for successful communication, the evidence points to the need for a basic set of core information rather than statistics. This includes the identity and severity of the potential harm, the likelihood of harm under various circumstances, and the possibility and difficulty of reducing that harm. One note of caution in risk communication is that with nearly all hazards, people at risk apply an optimistic bias (Weinstein 1999). False assurance can occur if risks are communicated in such a reassuring way that they are interpreted as meaning no risk at all. Research into epilepsy risk communication is urgently required, and there is a burgeoning ἀle of literature of risk communication from which to draw (McComas 2006).

57.11╅ SUDEP and Bereavement Communication of SUDEP risk rightly begins with patient well-being in mind. Neverthe� less,€where SUDEP occurs in families that are ignorant of the phenomenon, it multiplies the distress to the bereaved. The shock and sadness extends beyond the family, to friends, work colleagues, and even the associated health professionals (Cook 2005; Kennelly and Riesel

930 Sudden Death in Epilepsy: Forensic and Clinical Issues

2002). Bereavement affects people uniquely, but there is evidence that sudden bereavement can complicate the grieving process because people are unprepared. Early intervention can reduce associated morbidity (Raphael 1977; Yates et al. 1990). Little research has been carried out regarding SUDEP bereavement. However, studies into the impact of SIDS are informative. These studies found acute distress and long-lasting damage, particularly relationship conflict, difficulties with surviving children, and anxiety about future children becoming victims (Woodward et al. 1985). In both SUDEP and SIDS, the death is wholly unexpected, the cause is unknown, and families face the bewildering interventions of police, a postmortem examination, ongoing inquiry, and uncertainty. Some families experiencing SUDEP have other family members with epilepsy (Kennelly and Riesel 2002). Like SIDS families, they can fear a recurrence. Unlike SIDS, which is well publicized, SUDEP may create an additional burden on the grieving family in explaining the death to the police, family, friends, and the local community who are ignorant of SUDEP (Sudden Unexpected Death in Epilepsy: A Global Conversation 2005). The lack of research on the extended impact of SUDEP leaves a serious gap in our understanding and should be rectiἀed.

57.12â•… SUDEP Support Services Community-based SUDEP support services have been slow to appear internationally, with the lack of SUDEP awareness being a barrier to their development. The ἀrst SUDEP support service, Epilepsy Bereaved, was established in the United Kingdom in 1996 (www .sudep.org). The service, not dissimilar to SIDS support programs (Woodward et al. 1985), today offers information, listening, and support through a dedicated help line, group meetings, and memorial services. Resources are available for families and health professionals. Families receive a regular magazine and can become active in the work of the organization if this helps their grieving process. Families describe the service as a “lifeline.” In addition, Epilepsy Bereaved is funding and working with researchers in the United Kingdom, as well as raising public awareness of SUDEP and campaigning for more effective epilepsy and coroner’s services. Epilepsy guidelines in the United Kingdom recommend that health professionals contact families to offer their condolences, invite them to discuss the death, and offer referral to bereavement counseling and a SUDEP support group (Stokes et al. 2004). With respect to communications with doctors following an epilepsy death, a qualitative study found that bereaved people reported positively when the medical team had offered condolences, support, and information, and negatively when contact was not offered, delayed, or was defensive in nature (Kennelly and Riesel 2002). Families were grateful for the offer regardless of whether it was taken up. Internationally, few services exist to meet the needs of families affected by SUDEP, but activity is increasing with international networking seen as a key to effective use of experience and resources. The Irish Epilepsy Association (www.epilepsy.ie/) has worked with Epilepsy Bereaved to organize meetings for families. In Australia, epilepsy agencies support and link SUDEP families, organize a biennial memorial service, and develop resource€material. A SUDEP-related research fund has also been recently established (www.epilepsyaustralia╉ .net/). In Canada, SUDEP Aware holds meetings twice a year for bereaved families to connect and€share experiences, offers referral to counseling, and is a contact point for North American

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP 931

families to network with other families (www.sudepaware.org/). In the United States, bereaved families have supported the development of a SUDEP-targeted research program through CURE (www.cureepilepsy.org/home.asp) and have also participated in the American Epilepsy Society and the Epilepsy Foundation Joint Task Force on SUDEP, and the National Institute of Health SUDEP symposium in 2008. Support agencies also have an important role to play in relation to forensic investigation of SUDEP. Through a collaborative approach, the agency can assist the work of the forensic institute by offering extended contact and counseling to families and friends of the bereaved. Where families are referred to a support agency immediately after a death, the rationale for a postmortem examination can be explained, and support can be offered by others who have themselves experienced the fear and trauma of facing such an ordeal. Newly bereaved families can be assisted to understand the longer-term beneἀt that may result through family participation in a thorough investigation (Lathers and Schraeder 2009). In the United Kingdom, Epilepsy Bereaved has worked with the Royal College of Pathologists, leading to national guidelines on the investigation of epilepsy-related deaths. In Australia, collaboration between the Epilepsy Foundation of Victoria and the Victorian Institute of Forensic Medicine has helped to raise awareness of SUDEP in health professionals and the community, and promote research including death scene investigation.

57.13â•… Summary An examination of the evolution of epilepsy care over the past 15 years reveals that SUDEP advocacy groups have been working collaboratively and effectively with health professionals to address questions previously unanswered by the scientiἀc community, and to facilitate signiἀcant improvements in epilepsy care. Their questions have prompted reÂ�examination of services when others could not despite years of effort (Hanna et al. 2002). As a lowerprevalence condition, epilepsy has struggled to attract attention and resources (Brown 1995; Chronic Disease Directors 2003). However, the SUDEP issue has harnessed community participation and leadership, which has proved to be a signiἀcant, positive force for change. The discussion of epilepsy has been elevated to the highest levels (All Party Parliamentary Group on Epilepsy 2007; Department of Health 2001b), and guidelines for improved care have emerged (Scottish Intercollegiate Guidelines Network 2003; Stokes et €l. 2004). Ongoing research is essential to evaluate progress and guide future improvements. Nevertheless, in the wake of the SUDEP debate, it would appear that epilepsy might ἀnally be coming out of the shadows. This chapter has presented the arguments for and against informing patients and care givers about SUDEP, and takes the position that for the vast majority of patients the arguments for informing are convincing. While the reduction of SUDEP may be partly influenced by quality medical management, where risk reduction is the core strategy, patient education and participation is integral to success. More research is urgently needed on the most effective way to do this. It is sometimes suggested that we do not hear the voices of epilepsy patients in the SUDEP discussion. While it might be true that their opinions are not often reflected in this debate through formal research publications, nevertheless in some countries many people with epilepsy are now well aware of SUDEP. They have heard of deaths in the media, lost friends to SUDEP, or learned about it through their epilepsy support agencies. Consequently,

932 Sudden Death in Epilepsy: Forensic and Clinical Issues

people with epilepsy are frequently supporters of SUDEP organizations. Our observations are that many people with epilepsy feel that it is right for them to know, and that, to some degree, it is considered empowering to be trusted with the information, which enables them to make informed choices about their epilepsy management. We would like to conclude our discussion with the words of one such young woman with epilepsy. I had uncontrolled seizures for many years never realizing that people could die from a seizure. . . . I had frequent seizures but I never considered this to be a serious condition—just something I had to acceptâ•›. . . Some of the things I discovered about epilepsy and its treatment were not easy to hear, and when a young man I knewâ•›. . . subsequently died as a result of a seizure, the issue of death really hit home. However, I don’t live in fear of death. Now that I know about epilepsy I make choices to take care of myself. I strongly believe people have the right to know about their condition. (Kearton 2005)

References All Party Parliamentary Group on Epilepsy. 2007. Wasted Money Wasted Lives: The Human and Economic Cost of Epilepsy in England. London: All Party Parliamentary Group on Epilepsy. Andrewes, D., K. Camp, C. Kilpatrick, and M. Cook. 1999. The assessment and treatment of concerns and anxiety in patients undergoing presurgical monitoring for epilepsy. Epilepsia 40 (11): 1535–1542. Beran, R. 2006. SUDEP—to discuss or not to discuss: That is the question. Lancet Neurology 5 (6): 464–465. Beran, R. G., S. Weber, R. Sungaran, and N. Venn. 2004. Review of the legal obligations of the doctor to discuss Sudden Unexplained Death in Epilepsy (SUDEP)—A cohort controlled comparative cross-matched study in an outpatient epilepsy clinic. Seizure 13 (7): 523–528. Black, A. 2005. SUDEP Beran whether to tell and when? Med Law 24 (1): 41–49. Bolitho v City and Hackney Health Authority. 1998. AC 232. Brodie, M. J., and G. L. Holmes. 2008. Should all patients be told about sudden unexpected death in epilepsy (SUDEP)? Pros and cons. Epilepsia 49 (S9): 99–101. Brown, S. W. 1995. What resources? Addressing the needs of the epilepsy community. Seizure 4 (3): 207–210. Chester v Afshar. 2005. 1 AC 134. Chronic Disease Directors. 2003. The role of public health in addressing lower prevalence chronic conâ•›ditions: The example of epilepsy. Atlanta, GA: Centers for Disease Control and Prevention. Cook, M. 2005. Reflecting on my clinical experience. In Sudden Unexpected Death in Epilepsy: A Global Conversation, ed. D. Chapman, B. Moss, R. Panelli, and R. Pollard. Camberwell: Epilepsy Australia & Epilepsy Bereaved. Couldridge, L., S. Kendall, and A. March. 2001. A systematic overview—A decade of research. The information and counselling needs of people with epilepsy. Seizure 10 (8): 605–614. Dalrymple, J, and J. Appleby. 2000. Cross sectional study of reporting of epileptic seizures to general practitioners. BMJ 320 (7227): 94–97. Department of Health. 2001a. The Expert Patient: A New Approach to Chronic Disease Management for the 21st Century. London: Department of Health. Department of Health. 2001b. Annual Report of the Chief Medical Officer: Epilepsy—Death in the Shadows: On the state of the public health. London: Department of Health. Department of Health. 2003. Improving Services for People with Epilepsy: Department of Health Action Plan in Response to the National Clinical Audit of Epilepsy-Related Death. London: Department of Health.

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP 933 Deriche v Easling Hospital NHS Trust. 2003 EWHC 3104 (QB). Escoffery, C., C. Dilorio, K. Yeager, F. McCarty, E. Robinson, E. Reisinger, T. Henry, and A. Koganti. 2008. Use of computers and the internet for health information by patients with epilepsy. Epilepsy Behav 12 (1): 109–114. Faught, R. E., M. S. Duh, J. R. Weiner, A. Guerin, and M. C. Cunnington. 2008. Nonadherence to antiepileptic drugs and increased mortality: Findings from the RANSOM study. Neurology 71 (20): 1572–1578. Finsterer, J., and C. Stollberger. 2009. Cardiopulmonary surveillance to prevent SUDEP. Lancet Neurol 8 (2): 131–132. Fox, S. 2007. E-patients with a chronic disability or chronic disease. Washington: Pew internet and American life project. http://www.pewinternet.org/~/media//Files/Reports/2007/EPatients_ Chronic_Conditions_2007.pdf. Fraenkel, L., and S. McGraw. 2007. Participation in medical decision making: The patients’ perspective. Med Decis Making 27 (5): 533–538. General Medical Council. 2008. Consent: Patients and Doctors Making Decisions Together. UK: General Medical Council. Gaitatzis, A., A. L. Johnson, D. W. Chadwick, S. D. Shorvon, and J. W. Sander. 2004. Life expectancy in people with newly diagnosed epilepsy. Brain 127 (Pt 11): 2427–2432. Gayatri, N. A., M. C. Morrall, V. Jain, P. Kashyape, K. Pysden and C. Ferrie. 2010. Parental and physician beliefs regarding the provision and content of written sudden unexpected death in epilepsy (SUDEP) information. Epilepsia 51 (5): 777–782. Hanna, N. J., M. Black, and J. W. S. Sander et al. 2002. The National Sentinel Clinical Audit of EpilepsyRelated Death: Epilepsy—Death in the Shadows. Norwich: The Stationary Office. Helgeson, D. C., R. Mittan, S. Y. Tan, and S. Chayasirisobhon. 1990. Sepulveda Epilepsy Education: The efficacy of a psychoeducational treatment program in treating medical and psychosocial aspects of epilepsy. Epilepsia 31 (1): 75–82. Hitiris, N., S. Suratman, K. Kelly, L. J. Stephen, G. J. Sills, and M. J. Brodie. 2007. Sudden unexpected death in epilepsy: A search for risk factors. Epilepsy Behav 10 (1): 138–141. Horne, R. 2006. Compliance, adherences, and concordance: Implications for asthma treatment. Chest 130 (1 Suppl.): 65S–72S. Hughes, J. R. 2009. A review of sudden unexpected death in epilepsy: Prediction of patients at risk. Epilepsy Behav 14 (2): 280–287. Katz, J. 1984. The Silent World of Doctor and Patient. New York, NY: Free Press. Kearton, M. 2005. Living with the risks. In Sudden Unexpected Death in Epilepsy: A Global ConÂ� versation, ed. D. Chapman, B. Moss, R. Panelli and R. Pollard. Camberwell: Epilepsy Australia & Epilepsy Bereaved. Kennelly, C., and J. Riesel. 2002. Sudden Death and Epilepsy—The Views and Experiences of Bereaved Relatives and Carers. UK: Epilepsy Bereaved. Langan, Y., L. Nashef, and J. W. Sander. 2005. Case–control study of SUDEP. Neurology 64 (7): 1131–1133. Lathers, C. M., and P. L. Schraeder. 2009. Verbal autopsies and SUDEP. Epilepsy Behav 14 (4): 573–576. Lewis, G. 2007. The conἀdential enquiry into maternal and child health (CEMACH). Saving mothers’ lives: Reviewing maternal deaths to make motherhood safer—2003–2005. The Seventh Report on Confidential Enquires into Maternal Deaths in the United Kingdom. London: CEMACH. Lewis, S., S. Higgins, and M. Goodwin. 2008. Informing patients about sudden unexpected death in epilepsy: A survey of specialist nurses. Br J Neurosci Nurs 4 (11): 30–34. Lorig, K. R., and H. R. Holman. 2003. Self-management education: History, deἀnition, outcomes, and mechanisms. Ann Behav Med 26 (1): 1–7. Lu, C., E. C. Wirrell, and M. Blackman. 2005. Where do families of children with epilepsy obtain their information? J Child Neurol 20 (11): 905–910. Marteau, T. 2008. Detection is not as harmful as it might seem. BMJ 336: 973–974.

934 Sudden Death in Epilepsy: Forensic and Clinical Issues McComas, K. A. 2006. Deἀning moments in risk communication: 1996–2005. J Health Commun 11 (1): 75–91. Mittan, R. 1986. Fear of seizures. In Psychopathology in Epilepsy: Social Dimensions, ed. S. Whitman and B. P. Hermann. New York, NY: Oxford University Press. Mittan, R. 2005. Managing fear. In Sudden Unexpected Death in Epilepsy: A Global Conversation, eds. D. Chapman, B. Moss, R. Panelli and R. Pollard. Camberwell: Epilepsy Australia & Epilepsy Bereaved. Monte, C. P., J. B. Arends, I. Y. Tan, A. P. Aldenkamp, M. Limburg, and M. C. de Krom. 2007. Sudden unexpected death in epilepsy patients: Risk factors. A systemic review. Seizure 16 (1): 1–7. Moon, R. Y., R. S. Horne, and F. R. Hauck. 2007. Sudden infant death syndrome. Lancet 370 (9598): 1578–1587. Morton, B., A. Richardson, and S. Duncan. 2006. Sudden unexpected death in epilepsy (SUDEP): Don’t ask don’t tell? J Neurol Neurosurg Psychiatry 77 (2): 199–202. Nashef, L. 1995. Sudden unexpected death in epilepsy: Incidence, circumstances and mechanisms. MD thesis, University of Bristol. Nashef, L., J. F. Annegers, and S. Brown. 1997. Introduction and overview. Epilepsia 38 (S11): S1–S2. National Health Priority Action Council. 2006. National Chronic Disease Strategy. Canberra: Australian Government Department of Health and Ageing. Nunes, V., J. Neilson, N. O’Flynn et al. 2009. Clinical Guidelines and Evidence Review for Medicines Adherence: Involving Patients in Decision about Prescribed Medicine and Supporting Adherence. London: National Collaborating Centre for Primary Care and Royal College of General Practitioners. Opeskin, K., and S. F. Berkovic. 2003. Risk factors for sudden unexpected death in epilepsy: A controlled prospective study based on coroners cases. Seizure 12 (7): 456–464. Otoom, S., A. Al-Jishi, A. Montgomery, M. Ghwanmeh, and A. Atoum. 2007. Death anxiety in patients with epilepsy. Seizure 16 (2): 142–146. Pearce v United Healthcare NHS Trust. 1999. PIQR P53. Pedley, T. A., and W. A. Hauser. 2002. Sudden death in epilepsy: A wake-up call for management. Lancet 359 (9320): 1790–1791. Petticrew, M., A. Sowden, and D. Lister-Sharp. 2001. False-negative results in screening programs: Medical, psychological and other implications. Int J Technol Assess Health Care 17 (2): 164–170. Pinder, R. 1990. What to expect: Information and the management of uncertainty in Parkinson’s disease. Disabil Soc 5 (1): 77–92. Politi, M. C., P. J. K. Han, and F. Nananda. 2007. Communicating the uncertainty of harms and beneἀts of medical interventions. Med Decis Making 27 (5): 681–695. Preston, J. 1997. Information of sudden deaths from epilepsy. Epilepsia 38 (S11): S72–S74. Prinjha, S., A. Chapple, A. Herxheimer, and A. McPherson. 2005. Many people with epilepsy want to know more: A qualitative study. Fam Pract 22 (4): 435–441. Raphael, B. 1977. Preventative intervention with the recently bereaved. Arch Gen Psychiatry 34 (12): 1450–1454. Riddle, M. C. 1980. A strategy for chronic disease. Lancet 2 (8197): 734–736. Rogers v Whitake. 1992. 175 CLR 479. Scottish Intercollegiate Guidelines Network. 2003. Diagnosis and Management of Epilepsy in Adults. Edinburgh: Scottish Intercollegiate Guidelines Network. Scottish Public Services Ombudsman. 2009. Case 200700075. Shaw, C., K. Abrams, and T. M. Marteau. 1999. Psychological impact of predicting individuals’ risks of illness: A systematic review. Soc Sci Med 49 (12): 1571–1598. So, E. L. 2006. Demystifying sudden unexplained death in epilepsy—Are we close? Epilepsia 47 (S1): 87–92.

Challenges in Overcoming Ethical, Legal, and Communication Barriers in SUDEP 935 So, E. L., J. Bainbridge, J. R. Buchhalter et al. 2009. Report of the American Epilepsy Society and the Epilepsy Foundation joint task force on sudden unexplained death in epilepsy. Epilepsia 50 (4): 917–922. Spratling, W. P. 1904. Prognosis. In Epilepsy and Its Treatment. Philadelphia, PA: W. B. Saunders. Stokes, T., E. J. Shaw, A. Juarez-Garcia, J. Camosso-Steἀnovic, and R. Baker. 2004. Clinical guidelines and evidence review for the epilepsies: Diagnosis and management in adults and children in primary and secondary care, CG20 full guideline. London: Royal College of General Practitioners. In Sudden Unexpected Death in Epilepsy: A Global Conversation. 2005. eds. D. Chapman, B. Moss, R. Panelli and R. Pollard. Camberwell: Epilepsy Australia and Epilepsy Bereaved. Surges, R., R. D. Thijs, H. L. Tan, and J. W. Sander. 2009. Sudden unexpected death in epilepsy: Risk factors and potential pathomechanisms. Nature Reviews Neurology 5: 492–504. Taylor, J. A. 2002. Determination of Sheriff James Taylor, Sheriff of the Sherifdom of Glasgow and Strathkelvin at Glasgow. Inquiry held under fatal accidents and sudden death inquiry (Scotland) Act 1976 into the death of Colette Marie Findlay. Taylor, S. E. 1991. Asymmetrical effects of positive and negative events: The mobilization–minimization hypothesis. Psychol Bull 110 (1): 67–85. Tomson, T., L. Nashef, and P. Ryvlin. 2008. Sudden unexpected death in epilepsy: Current knowledge and future directions. Lancet Neurol 7 (11): 1021–1031. Tomson, T., L. Nashef, and P. Ryvlin. 2009. Cardiopulmonary surveillance to prevent SUDEP: Authors’ reply. Lancet Neurol 8 (2): 132–133. Videto v Kennedy. 1981. 125 DLR (3d) 127. Weinstein, N. 1999. What does it mean to understand a risk? Evaluating risk comprehension. J Natl Cancer Inst Monogr 25: 15–20. Welsh Assembly Government. 2009. Designed for people with chronic conditions: Service DevelÂ� opment Directive—Epilepsy. Cardiff: Welsh Assembly Government. Wertheimer, A. I., and T. M. Santella. 2003. Medication compliance research: Still so far to go. The Journal of Applied Research in Clinical and Experimental Therapeutics 3 (3): 254–261. Whitney, S. N., M. Holmes-Rovner, and H. Brody et al. 2008. Beyond shared decision making: An expanded typology of medical decisions. Med Decis Making 28 (5): 699–705. Woodward, S., A. Pope, W. J. Robson, and O. Hagan. 1985. Bereavement counselling after sudden infant death. Br Med J (Clin Res Ed) 290 (6465): 363–365. Yates, D. W., G. Ellison, and S. McGuiness. 1990. Care of the suddenly bereaved. BMJ 301 (6742): 29–31.

Bereavement and Sudden Unexpected Death in Epilepsy Lina Nashef Lene Sahlholdt

58

Contents 58.1 Introduction 58.2 Grief 58.3 Grief Therapy 58.4 Bereavement in SUDEP: How Is It Different? 58.5 Self-Help Groups and Epilepsy Charities 58.6 Information, Physicians, and SUDEP 58.7 Publications by Bereaved Relatives in SUDEP 58.7.1 A Global Conversation 58.7.2 SUDEP: The Views and Experiences of Bereaved Relatives and Carers (Epilepsy Bereaved 2002) 58.8 Bereaved Relatives in the Medical Literature: An Interview Study 58.9 Conclusion References

937 937 938 938 939 939 939 939 939 940 941 942

58.1â•…Introduction Being sudden and unexpected, often affecting the young, sudden unexpected death in epilepsy (SUDEP) leaves much sorrow, pain, turmoil, and anguish in its wake. Communication between the bereaved and health services in such circumstances can be very helpful but can also be demanding for all concerned and needs to be handled with sensitivity. Regrettably, there has been little focus in the SUDEP medical literature on bereaved families or partners. This chapter will by necessity be concise, its intention being to draw attention to this area and to the need for further research. It will also include suggestions for management in this situation. Before discussing bereavement following SUDEP, grief following bereavement in general will be briefly considered.

58.2â•… Grief Grief is the complex psychic pain in the bereaved related to loss and longing for the dead person (Bonanno et al. 2007). The loss is often felt to be unfair and senseless. Grief is often mixed with many other feelings such as guilt and anger toward the dead person and may also be accompanied by fear of the future. Relatives grieve and look for comfort and consolation. They ask themselves many questions, which reflect feelings of guilt, disbelief, and/or anger. Why him/her? Why our family? Could we have done something 937

938 Sudden Death in Epilepsy: Forensic and Clinical Issues

to prevent it? Could the doctors have done more? Anger or mistrust toward the medical team responsible for the treatment may be a part of these mixed feelings. The most severe grief is often said to be after the death of a child, a parent, or a spouse. Grieving is often long lasting, with feelings of loss and longing often revived on special occasions for many years. Grieving may be accompanied by sleep disorders, attention problems, and depression, and thus may have some resemblance to aspects of depression or posttraumatic stress disorder. Long-lasting grief is not necessarily dysfunctional. Ways of grieving depend on personality and culture. Wijngaards-de Meij et al. (2007) found that grief is related to the “attachment style.” An insecure or weak attachment may produce longer and stronger grief than if the attachment is conἀdent. Most people, including clinicians, often understand grief as a step-by-step loosening of attachment or of bonds to the dead person in which the goal is to break the bonds, the assumption being that this provides the basis for the potential to “move on” with life. This may include, for example, sending good-bye letters to the deceased. However, there is no evidence for this step-by-step understanding of the grief process, although it is “common knowledge” and widely assumed (Wortman and Silver 2001). Newer evidence-based descriptive models offer a more complex view whereby the grief reaction is seen as a mixture in terms of focus not only on the loss but also on regeneration and restoration. It involves looking back and forward in a wave-like motion, an oscillation between concentration on loss and restoration of life. Breaking the bonds is not necessary for moving on (Stroebe and Schut 1999).

58.3â•… Grief Therapy While grief therapy may be seen currently as a way of acknowledging grief, referring all who lose a dear person for grief therapy is pointless as it is neither necessary nor effective in uncomplicated grieving. It is only helpful when asked for, if there are other risk factors, or if the grief is complicated, for example, by depression, unless the grieving person is a child who lost a parent. Grief therapy may, on the contrary, be both necessary and helpful if grief is complicated and otherwise would result in the bereaved being unable to cope with daily life (Stroebe et al. 2005, 2007).

58.4â•… Bereavement in SUDEP: How Is It Different? While bereavement following SUDEP shares features common to bereavement in general, there are a number of circumstances that add additional dimensions to some aspects of the grief reaction: the unexpected and sudden nature of SUDEP, which occurs in young and, apart from the epilepsy, often otherwise well individuals; the lack of abnormality on autopsy to explain the death; the lack of recognition of risk by the medical profession, at least until recently; the ongoing general lack of acknowledgement of the risk; and the dayto-day choices someone with epilepsy and their relatives might make in managing their condition and in leading their lives. Death may be the culmination of years of living with epilepsy, a chronic condition requiring major adjustment, where the requirements of safety need to be constantly reconciled with the desire for independence.

Bereavement and Sudden Unexpected Death in Epilepsy

939

58.5â•… Self-Help Groups and Epilepsy Charities This situation has led bereaved relatives in many countries, starting with “Epilepsy Bereaved” in the United Kingdom in 1993 (http://www.sudep.org), to form highly motivated self-help groups. In this, they have demonstrated commitment and strength of purpose, arising from their grief and fuelled by their dissatisfaction with the status quo. Their aims have been primarily to support and inform other bereaved relatives, and also to redress wider deἀciencies by increasing awareness of SUDEP, encouraging research as well as best medical and self-management. In doing so, self-help groups have provided a badly needed resource and backup for other bereaved relatives, often, but perhaps not always, welcomed by the medical profession or other epilepsy charities.

58.6â•…Information, Physicians, and SUDEP Although relevant to this discussion, it is outside the scope of this short chapter to address the ongoing debate among health workers and bereaved relatives regarding the need for physicians and nurses to improve information provision on risks associated with epilepsy, and how this is best provided, the role of guidelines, or legal aspects (Hanna and Panelli 2010, Chapter 57 in this book). Much of the expressed focus of differences in opinion can be taken to reflect limited research to guide best practice in this area and the need to make a distinction between the views and interests of the bereaved and those of the late relative or of people with epilepsy in general. There is, however, another dimension as yet unexpressed, but which may nevertheless be necessary to consider. While dwarfed by the strength of the grief reaction in relatives, we should also acknowledge the grief reaction in a health worker whose patient dies unexpectedly “on his or her watch.”

58.7â•… Publications by Bereaved Relatives in SUDEP The views of bereaved relatives are referred to in the two publications below, which are discussed in more detail in Chapter 57 (Hanna and Panelli 2010). 58.7.1â•… A Global Conversation In the publication A Global Conversation (Chapman et al. 2005), views of bereaved relatives were expressed and included the following: (a) not being informed of the risk, (b) believing that the dead relatives were not informed, (c) experiencing a taboo in the medical world, and (d) wishing they were informed. 58.7.2â•…SUDEP: The Views and Experiences of Bereaved Relatives and Carers (Epilepsy Bereaved 2002) Kennelly and Riesel (2002) conducted a qualitative study in parallel with the United Kingdom National Sentinel Clinical Audit of Epilepsy Related Death. A summary was published by Epilepsy Bereaved, although not to our knowledge, in a peer-reviewed journal. In this study, six focus groups were held in the United Kingdom to identify issues

940 Sudden Death in Epilepsy: Forensic and Clinical Issues

that bereaved relatives and carers considered important. In addition, 78 semistructured telephone interviews with bereaved relatives were carried out to explore important issues in further detail. The summary report concludes with a number of very useful broad and speciἀc recommendations about epilepsy care and response to bereavement, many in line with our recommendations below.

58.8â•…Bereaved Relatives in the Medical Literature: An Interview Study As has already been mentioned, there has been little focus in the medical literature on bereaved relatives. In 1993/1994, one of the authors (LN), and a colleague (S. Garner), interviewed 34 sets of self-referred bereaved relatives, carers, or partners of people with epilepsy-related deaths (Nashef et al. 1998), of whom 26 fulἀlled the deἀnition for SUDEP. In€addition to detailed assessment of epilepsy diagnosis, control, treatment, and circumstances of death, the semistructured interviews also touched on attitudes and perception of relatives/carers or partners. The publication arising from this research focused on the quantitative aspects of the research, including epilepsy classiἀcation, treatment history, and triggers for and evidence of a terminal seizure. Of note is that the SUDEP cases identiἀed did not all have severe epilepsy. Of the 26 cases, 8 had fewer than 10 generalized tonic–clonic seizures before SUDEP, 10 between 10 and 100, and 7 greater than 100 (1 unknown with continuing partial seizures but no tonic–clonic seizure reported for over 20 years). Five had a history of status epilepticus or serial seizures. Eleven had generalized epilepsy (9€of whom were idiopathic generalized), 10 localization-related epilepsy, and 5 undetermined. Although there was no control group, it was felt that in some, a number of facÂ�tors€relating to lifestyle and medical treatment could have influenced seizure control and, potentially, outcome. In addition, the personal and qualitative dimension was extremely enlightening. Many of the bereaved had witnessed severe epileptic seizures and had often been reassured by the medical profession of their benign nature, falsely as it turned out, in€their case. Yet, intuitively, many “knew” seizures were potentially dangerous and lived in fear of possible disaster with constant anxiety for their loved ones. Had risks associated with seizures been discussed with some of these relatives, such information would not have come as a surprise. Conversely, for another, the possibility that epilepsy could “kill” had€not€been considered at all. Feelings of guilt and blame were also evident, although not always explicit. Blame often focused on the medical profession and health services, particularly in relation to lack of information being made available during life. However, there€were also other concerns, relating to care, support, and communication during life and, to the bereaved, after death. In relation to care after death, examples of issues raised included being denied the opportunity to explore the environment or circumstances of death, not being offered the opportunity to ask questions or go over medical details, not being linked up with support services, lack of contact after death by the general practiÂ�tioner or the specialist, learning about SUDEP for the ἀrst time from the press, disagreement with death certiἀcation (e.g., status epilepticus entered as cause for death without supporting evidence), and SUDEP denial. During life, concerns related to medical matters are the lack of support, being discharged from specialist services too soon, issues relating to medication changes, not being referred to specialist services, and especially not acknowledging or appreciating the potential seriousness of the condition. Some felt that,

Bereavement and Sudden Unexpected Death in Epilepsy

941

had it been discussed, this would have allowed them to express their concerns and could have led them to be more proactive in managing the epilepsy and in pushing for more medical help and support for their relative. When the above research was being carried out, more than 15 years ago, bereaved relatives felt very isolated in their loss. The deaths occurred entirely unexpectedly and were apparently unexplained on a background of a chronic condition labeled as “benign,” but which required much support from the family. Many stated that they had been categorically told, or led to believe, that epilepsy could not be fatal. Almost all stated in retrospect that they would have preferred to know of the possibility of premature death, however remote. Many would have preferred to have been given the opportunity to acknowledge epilepsy as a serious condition rather than have intuitive fears dismissed, although, as already stated, many still harbored such fears despite reassurances to the contrary. Many regretted not being given the opportunity of making informed decisions regarding lifestyle and treatment. Counseling was not usually offered. Contact with the general practitioner or specialist was helpful when it took place after the death but was sometimes limited or absent (Nashef et al. 1998). At the time, we recommended that information about mortality and response to seizures be included in leaflets on epilepsy, and that in the event of SUDEP, the general practitioner (family doctor) makes early contact with bereaved relatives and the specialist writes with an open offer of a meeting. We also suggested that information about SUDEP, selfhelp groups, other supporting agencies, and counseling services be given as appropriate.

58.9â•… Conclusion There is more awareness now of the entity of SUDEP, and although the intense sense of isolation that characterized the feelings of bereaved relatives before may no longer be the same, it is likely that much of the above still applies. This remains a very difficult and tragic situation, sometimes, if not often, inadequately handled. On the basis of this study, we favor better information provision on risks associated with epilepsy during life, this being necessary for informed choice and optimizing treatment, self–management, and lifestyle. In the event of an epilepsy-related death, especially if sudden and unexpected, grief is a normal reaction and support should be offered. We recommend that health workers inform other members of the team about any death, and that letters of condolences are sent or contact made at an early opportunity, with an offer of an appointment for partners or relatives, both older and younger members of the family. Sympathy, support, and the opportunity to go over any aspects of the medical history or treatment are likely to be welcomed by relatives even if some chose not to, or are unable to, take up the offer immediately or at any stage: it is important to be open and prepared to accept all the mixed emotions of grief, anger, or aggression that the bereaved may be experiencing, including questioning one’s authority. Referral for appropriate counseling services and to self-help groups should also be offered. Offering psychological treatment—grief therapy—is a modern way of acknowledging grief. It is not always necessary but may be welcome and helpful to some. Although the above recommendations are based on experience in the United Kingdom, they are likely to have wider application. Needless to say, more research exploring current experiences of bereaved relatives and optimal management of bereavement following SUDEP in different settings is needed.

942 Sudden Death in Epilepsy: Forensic and Clinical Issues

References Bonanno, G. A., Y. Neria, A. Mancini, K. G. Coifman, B. Litz, and B. Insel. 2007. Is there more to complicated grief than depression and posttraumatic stress disorder ? A test of incremental validity. J Abnorm Psychol 116 (2): 342–351. Chapman, D., B. Moss, R. Panelli, and R. Pollard. 2005. Sudden Unexpected Death in Epilepsy—A Global Conversation. Camberwell, Australia: Epilepsy Australia Ltd. Kennelly, C., and J. Riesel. 2002. Sudden death and epilepsy. The views and experiences of bereaved relatives and carers. Epilepsy Bereaved, Oxford, U. K. Nashef, L., S. Garner, J. W. Sander, D. R. Fish, and S. D. Shorvon. 1998. Circumstances of death in sudden death in epilepsy: Interviews of bereaved relatives. J Neurol Neurosurg Psychiatry 64: 349–352. Stroebe, M., and H. Schut. 1999. The dual process model of coping with bereavement: Rationale and description. Death Studies 23: 197–224. Stroebe, M., H. Schut, and W. Stroebe. 2005. Grief work, disclosure and counselling: Do they help the bereaved? Clin Psychol Rev. 25: 395–414. Stroebe, M., H. Schut, and W. Stroebe. 2007. Health outcomes of bereavement. Lancet 370 (9603): 1960–1973. Wijngaards-de Meij, L., M. Stroebe, and H. Schut. 2007. Neuroticism and attachment insecurity as predictors of bereavement outcome. J Res Pers 41: 498–505. Wortman, C. B., and R. C. Silver. 2001. The myths of coping with loss revisited. In Handbook of Bereavement Research, ed. M. S. Stroebe, R. O. Hansson, W. Stroebe, and H. Schut, 405–429. Baltimore, MD: United Book Press.

SUDEP A Clinical and Communicative€Conundrum

59

Paul L. Schraeder Claire M. Lathers The risk of sudden unexpected death is associated with multiple disease categories, with the group most commonly at risk being that with atherosclerotic heart diseases (Lathers, Schraeder, and Bungo 2008a, 2008b; Kuller, Lilienfeld, and Fisher 1967). Other cardiac entities such as rheumatic valvular disease, hypertensive heart disease, congenital heart disease, myocarditis, and genetically determined arrhythmogenic channelopathies also put patients at risk for sudden death. In the pulmonary disease category, the commonly occurring examples of causes of sudden death are embolism and asthma. Also, it has long been known that persons with psychiatric disorders have a risk of sudden death, but whether the risk is associated with the psychiatric disorder itself or with the use of psychotropic drugs, or both, is not clear (Wendkos 1979). Others at risk for sudden death are members of ethnic-related entities, such as Filipino men who are victims of sudden death while sleeping, the phenomenon being called bangungut, as well as some refugee populations from Southeast Asia living in the United States (Wannamaker 1990). Voodoo deaths are another category that may have ethnic implications (Samuels 1997). Among children, sudden infant death syndrome is a well-recognized phenomenon (Paterson 2010, this book, Chapter 5). Finally, sudden unexpected/unexplained sudden death in epilepsy (SUDEP), which is the subject of this book, is one of the most common causes of death in persons with epilepsy, which, despite increasing interest in unraveling the mystery, still remains an enigma (Lathers, Schraeder, and Bungo 2008b). As mentioned above, the largest group at risk for sudden unexpected death is that of cardiac disease. The overall incidence in this group is 1/738, with a higher incidence of 1/371 in the over age 30 group (Annegers and Blakley 1990). However, epilepsy is likewise associated with a high risk of unexpected death (Tomson et al. 2005). In comparison, having idiopathic epilepsy overall has a 1/1000 yearly incidence risk. However, when one factors in decreasing degrees of seizure control, it is evident that the risk increases, resulting in persons with poorly controlled seizures having an incidence risk as high as 1/100 (Annegers and Blakley 1990). This conclusion was conἀrmed by Nashef et al. (1996). Persons with epilepsy who were seizure free for more than 2 years had a 1/2500 patient-year risk of unexpected death. These investigators found that in patients with poorly controlled active epilepsy, the risk was greater than 1/200 patient-years, and concluded that improved seizure control reduces the risk of unexpected death. Although the risk factors associated with SUDEP, as determined by statistical analysis of population data, are discussed in varying detail throughout this book, it is also important to personalize these risk factors with the clinical histories of individuals who succumbed to SUDEP. Obviously, not all issues are addressed, but the compelling circumstances illustrated in the cases discussed in different parts of the book certainly cover a broad spectrum of circumstances, all of which resulted in the same unfortunate end point of death. 943

944 Sudden Death in Epilepsy: Forensic and Clinical Issues

The causes of unexpected deaths in persons with epilepsy can be broken down arbitrarily into four categories. (1) Epilepsy may be a direct cause of death as exempliἀed by a person dying during uncontrolled status epilepticus; (2) epilepsy as an underlying cause would include circumstances such as drowning during a seizure or dying as the result of a seizure-related motor vehicle accident; (3) deaths in which epilepsy is present, but occurs as a secondary entity associated with a progressively fatal brain disorder such as a malignant tumor or intracerebral hemorrhage; and (4) the mysterious entity of sudden unexplained/ unexpected death in persons with epilepsy in which there is no evidence of any associated cause of death other than the history of epilepsy (Leestma 1990b). SUDEP is deἀned as “sudden, unexpected, witnessed or unwitnessed, nontraumatic and nondrowning death in patients with epilepsy, with or without evidence for a seizure and excluding documented status epilepticus, in which postmortem examination does not reveal a toxicologic or anatomic cause for death” (Nashef 1997). For many years, SUDEP was in many ways a risk associated with having epilepsy that was ignored and denied. One of the original long-term population-based research projects to investigate SUDEP was conducted by Leestma (Leestma 1990a, 1990b; Leestma et al. 1984). This study utilized the Cook County coroner’s office and was one of the ἀrst and the largest study that focused on thoroughly investigating all deaths in persons with a history of epilepsy in the greater Chicago region. The data from this project indicated that SUDEP accounted for at least 10% of epilepsy-related deaths, with an average age at time of death of 30+ years. SUDEP was found to be more common in males, with a male/ female ratio of 3:1. While higher seizure frequency was also a factor, in that over one-half of the victims had 3 to 10 seizures per year, two-ἀfths had as few as one to two seizures per year. The majority of SUDEP events occurred at home during sleep. In instances when a rescue team managed to arrive shortly after the onset of the fatal event, electrocardiograms showed ventricular ἀbrillation. More than 97% of the deaths (36/37) occurred at home, with most (82%) in the prone position. Of the four witnessed deaths, three victims had generalized tonic seizures and, interestingly, one had no observed clinical seizure at the time of death. Postmortem examination in the Cook County population found that in 50% of cases, none of the prescribed antiepileptic drugs were found in blood samples, a ἀnding that was consistent with noncompliance in the SUDEP group. Of interest was the observation that the hearts, livers, and lungs had heavier weights than expected. The cause of cardiomegaly is not known, and it is uncertain how it may relate to the mechanism of SUDEP. Histological examination found, on routine screening, that the heart had no obvious pathology to explain a cause of death. The livers had only passive congestion that was compatible with acute cardiac failure. The lungs revealed both pulmonary edema and congestion, with some cases showing increased protein content in the edema fluid, ἀndings that may be compatible with neurogenic pulmonary edema. Brain pathology was varied but consisted of no lesions that could be used to explain an acute cause of death. Static nonprogressive entities include sequellae of old trauma, stable hydrocephalus, old subdural hematomas, cortical malformations, etc., that were found in the brains were in all likelihood the underlying causes of the epilepsy and themselves were not the cause of death (Leestma et al. 1984, 1989). Comparable autopsy data were found in a population-based study in Denver, CO (Earnest et al. 1992). Subtherapeutic antiepileptic drug levels were evident in 92% of autopsies. Pulmonary congestion occurred in 86% of the lungs, 32% of the brains, and in 32% of the abdominal viscera. Static brain pathology that could be the

SUDEP

945

etiology of epilepsy, but not the direct cause of death, was found in 31% of the brains that were available for study. One of the more interesting histological ἀndings was the observation of interstitial myocardial ἀbrosis in 11% of the hearts, an observation that was conἀrmed by later studies of the hearts of SUDEP victims (Natelson et al. 1998). Based on the above and subsequent data (Nashef et al. 1996; Nashef and Shorvon 1997; Nashef et al. 1996; Nashef 1997), several associated risk factors have come into focus. These are being young (mean age, 32 years), being male (50% greater risk than female), sleeping (most deaths occur during sleep), generalized tonic–clonic epilepsy (usually with poor control, but not always), antiepileptic drug polypharmacy (with frequent dose and drug changes), and underlying static brain pathology that is often associated with mental retardation syndromes. Two other possible risk factors should also be discussed. The ἀrst of these, stress, is known to be a risk factor for sudden death in persons with a history of coronary artery diseases (Ong 2008). Examples of stressful circumstances associated with an increased risk of unexpected sudden death are threat of or actual loss of a person close to the individual, loss of status or self-esteem, and anticipation of or relief from sudden danger. Investigation of stress as a risk factor for SUDEP inexplicably has been mostly ignored. Having epilepsy itself is stressful. For example, there are associated fears of losing control in potentially embarrassing circumstances, social stigmata associated with public ignorance about epilepsy, as well as concerns about losing employment and one’s driver’s license (Lathers and Schraeder 2006). Persons with epilepsy appear to be more socially isolated, less likely to marry, more likely to be under- or unemployed, and are more likely to suffer from anxiety, depression, and poor self-esteem (Ring 1997). The second possible risk factor is the concurrent presence of inherited potentially arrhythmogenic channelopathies in persons with epilepsy. For example, an abnormal sodium channel gene SCN5A is found in Brugada syndrome, which consists of ST elevation, incomplete complete right bundle branch block, mutifocal ventricular tachycardia, and, most disturbingly, a usually normal baseline electrocardiogram (EKG) ἀnding. The same abnormal gene is also associated with the long QT syndrome, resulting in a torsade de pointe ventricular tachycardia. Both of these clinical entities are associated with a family history of cardiac disease (Lathers and Schraeder 2009) (Lathers, Schraeder, and Bungo 2008a). We do not know if the concurrent presence of epilepsy and a potentially arrhythmogenic inherited channelopathy increase the chance of SUDEP. Future studies need to be undertaken to screen family members of SUDEP victims to determine if such a link exists, for if it does, then a valid argument would exist for screening the highest risk group of persons with a history of epilepsy, namely those whose seizures are not under control. Needless to say, we do not yet understand how these two factors will ἀt in with those that are already established and those yet to be discovered. It is likely that SUDEP is associated with interactions of multiple risk factors (Lathers, Schraeder, and Bungo 2008a). While the mechanism of SUDEP is not known, the observation of clinical autonomic dysfunction in association with seizures mitigates in favor of a role for autonomically mediated acute disturbances having a role in explaining the occurrence of SUDEP. For example, Hilz et al. (2002) found cardiac single photon emission computed tomography evidence of decreased norepinethrine uptake in cardiac sympathetic nerves during seizures, and Opherk et al. (2002) described increased ictal heart rate, ST depression, and T wave inversion in association with seizures. However, in contradistinction to the observation of increased heart rate, Rugg-Gunn et al. (2004), using prolonged implanted EKG

946 Sudden Death in Epilepsy: Forensic and Clinical Issues

monitoring, observed ictal bradycardia, with 4/16 subjects requiring pacemaker implantation. Other instances of ictal bradycardia are well documented (Rocamora et al. 2003). Observations of autonomic changes consequent to cortical stimulation in persons with epilepsy have been extant for over half a century, allowing for longstanding speculation about the mechanism(s) of SUDEP. Penἀeld and Jasper (1954), who undertook systematic stimulation of multiple cortical areas in persons undergoing surgery for epilepsy while awake, under local anesthesia, documented a variety of autonomic responses to cortical stimulation. These included the following: observation of salivation with suprasylvian stimulation; gastrointestinal activation with nausea, abdominal discomfort, boborygmi, and a desire to defecate during insula stimulation; apnea with maintained voluntary ability to breath during anterior and inferior cingulate stimulation; irresistible apnea, i.e., inability to overcome apnea with voluntary effort, during rolandic or uncal region stimulation; and abdominal sensations, increased heart rate, flushing, and papillary changes during stimulation of the supplementary motor area. In animals’ stimulus of the amygdala,€Bonvallet and Bobo (1972) found neuronal groupings that indicated that cardiac and respiratory control were intertwined. These investigators demonstrated, for the ἀrst time, that within the amygdala, stimulation of one grouping of neurons resulted in bradycardia and respiratory deactivation, while stimulation of a different, nearby group resulted in tachycardia and respiratory activation. The complex interconnections between various cortical and subcortical structures, including the amygdala, hypothalamus, and cardiovascular/respiratory control centers in the brainstem, would give credence to the possibility of seizure-related potentially fatal cardiovascular and/or respiratory disruption contributing to the risk of SUDEP (Stollberger and Finsterer 2004). Ictal tachycardia is commonly observed during generalized tonic–clonic seizures. Marshall et al. (1983) found frequent episodes of tachycardia and premature ventricular contractions in association with complex partial seizures. Nei et al. (2004) reported increased ictal heart rates in persons who subsequently were victims of SUDEP, and suggested that potential SUDEP victims manifested increased autonomic activity with repolarization and rhythm abnormalities, especially during sleep. Ictal-related bradyarrhythmias seem to be associated most often with insular, cingulate, pyriform cortex, or amygdala activation. Of interest is the observation of associated episodes of apnea (Penἀeld and Jasper 1954) (Leung, Kwan, and Elger 2006). The association of cerebral discharges and disturbances of the normal balance between cardiac autonomic neural discharges, as illustrated by the lockstep phenomenon (Lathers, Schraeder, and Weiner 1987) (Lathers and Schraeder, 2010b, Chapter 28, this book), provides evidence that there is a direct relationship between ictal, and even interictal discharges, and disturbance of cardiac sympathetic and parasympathetic neural discharges that are involved in controlling normal cardiac conduction and rhythm. While the potential for adverse autonomic effects were alluded to above and has been little investigated since the study of Lathers and Schraeder (1990), Mameli et al. (1988) found in animal experiments that interictal discharges resulted in sinus arrest, supraventricular extra systoles, bradycardia, junctional rhythms, and associated drop in blood pressure. They also observed a very short latency between epileptiform discharges and the cardiovascular changes, implying that these were indeed neurogenic, not neurohumoral, events, and that the changes persisted beyond the duration of the interictal events. Respiratory factors should also be considered. It is well known that victims of SUDEP have increased lung weights compared to control (Leestma 1990c; Leestma et al. 1984), ἀndings that support the possible mechanistic role of seizure-induced or related neurogenic

SUDEP

947

pulmonary edema in SUDEP. While these human postmortem pulmonary ἀndings are supported by their occurrence in animal models of epilepsy, there is also evidence that seizure-related central apnea can occur concurrently or independently of the acute pulmonary changes (Johnston et al. 1995, 1997). The occurrence of central apnea in these series of experiments supports the data of Penἀeld and Jasper (1954) and Bonvallet and Bobo (1972), as discussed earlier in this chapter (vide supra). However, determination of whether any seizure-related observations are concomitant occurrences, potential risk factors, or potential mechanisms requires more clinical and animal-based investigation. Investigation of the potential role of psychological factors in association with risk of SUDEP has been for the most part ignored. This general lack of interest is in marked contrast to the longstanding acknowledgment of the association of psychological factors with risk for sudden death in persons with cardiac disease (Lathers and Schraeder 2006) (Lathers and Schraeder 2010a, Chapter 17, this book). The investigational scotoma related to the question of what role adverse psychological circumstances may have in SUDEP is difficult to explain. For example, it is well known that persons with epilepsy are more likely to be depressed than persons with other disorders (Mendez, Cummings, and Benson 1986). As anxiety and depression often occur concurrently, it is important to consider the reality of the negative societal issues that can be an ongoing stimulus for persons with epilepsy to be more likely to experience these symptoms than persons with other chronic disorders. The societal stigmata associated with having epilepsy impose a persisting impact on persons with epilepsy. Any neurologist committed to treating persons with epilepsy is well aware that these stigmata can result in discrimination in obtaining and keeping a job, the fear of embarrassment associated with having a seizure in public, and impaired mobility when not allowed to drive. Persons with epilepsy also face having associated psychological issues, such as depression (Mendez, Cummings, and Benson 1986), which may be aggravated by antiepileptic medication. In addition, another common source of anxiety in young parents is concern about the potential inheritability of epilepsy and the risks of maternal antiepileptic medication use during pregnancy. That psychological stress can have a potentially fatal adverse effect on some individuals is illustrated by the case of a 22-year-old male college student who came to the office because he had loss of consciousness associated with tonic–clonic seizures while attending church. At the time he lost consciousness, he was listening to a passage being read by his minister from Foxe’s Book of Martyrs, a book that described the sufferings of persons being persecuted and tortured because of their religious beliefs. The young man’s sister also described that he had lost consciousness in the past while listening to a friend describe to him the details of a recent hospitalization for major surgery. An electroencephalogram (EEG) was obtained, during which, unbeknownst to the patient, the minister arrived and recited the previously offending passage. At that point, the patient’s heart slowed, then ceased to beat. After 15 seconds of asystole, the EEG became isoelectric and generalized tonic–clonic motor activity occurred, after which the heart returned to a normal sinus rhythm and the EEG to alpha rhythm. Eventually, he had a pacemaker implanted and had no further events (Schraeder, Pontzer, and Engel 1983). Although this young man had been observed having generalized tonic–clonic seizure-like episodes, he did not have epilepsy. His sequence of events demonstrated how adverse emotional responses can adversely affect the autonomic regulation of cardiovascular homeostasis, resulting in potentially life-threatening changes. This case emphasizes the need for investigation of the relationship between stress and sudden death.

948 Sudden Death in Epilepsy: Forensic and Clinical Issues

Autonomic/homeostatic dysfunctions such as ictally related cardiac tachy- and bradyarrhythmias, pulmonary edema, and central apnea have each been long discussed as possible mechanisms of SUDEP. In the future, is seems quite possible that psychological stress may also need to addressed. Each potential mechanism has a following. It is possible to speculate that several combinations of factors that lead to SUDEP may be operant. These include (1) the occurrence of seizure-related central apnea plus neurogenic pulmonary edema leading to a respiratory death; (2) seizure-related hypoxia plus acute neurogenic pulmonary changes combined with acidosis resulting in a fatal arrhythmia; (3) the presence of an inherited channelopathy aggravated by an acute seizure-related autonomic dysfunction, resulting in a fatal arrhythmia; and (4) psychological stress adding to potentially fatal combinations of events, tipping the balance toward SUDEP in susceptible individuals. Future research needs to address a broad range of investigations. (1) At a clinical level, multicenter, prospective population-based autopsy studies will be the only method to determine the proἀle of persons at risk for SUDEP (Schraeder et al. 2006, 2009, 2010, Chapter 6, this book). Such studies should include detailed interviews of witnesses, family members, and health care providers (Lathers and Schraeder 2009), emphasizing a search for environmental, medical, and psychological risk-related data as well as detailed microscopic examination of the somatic and autonomic nervous system tissue and the heart (Lathers and Schraeder 2009). Additional investigation addressing the possible role of disturbed serotonin metabolism may also be important in the future (Paterson 2010). (2) There is also a need to determine whether a potential link exists between the subtle, genetic cardiac channelopathies and risk of SUDEP. Investigation should include detailed cardiac history, EKGs, and genetic screening for channelopathy genes in surviving family members. (3) Animal models need to be designed and used to investigate whether there is concomitance between seizure-related cardiovascular autonomic/respiratory dysfunction, induced stress, and sudden death. (4) Finally, there needs to be an investigation into possible preventive interventions such as sleeping position, prophylactic use of antiarrhythmic drugs, approaches to improving patient compliance with prescribed antiepileptic drugs, and stress management protocol for persons with epilepsy. As in most prevention efforts, while an absolute abolition of the occurrence of SUDEP would be desirable, it is unlikely. A measurable decrease in its risk of occurrence should be an achievable initial goal. The educational need around the issue of SUDEP should also to be addressed. A national survey of coroners and medical examiners was conducted to determine the extent of postmortem investigation in persons who died with a history of epilepsy and to ascertain how these officials approached the issue of SUDEP (Schraeder et al. 2006, 2009). This survey documented that while the overall quality of postmortem examination was sufἀciently thorough to determine the cause of death, in cases of epilepsy without a determinable cause, there seemed to be a dichotomous and paradoxical appreciation of SUDEP among those surveyed. The survey data found that while there was a general acknowledgment of the existence of SUDEP as a diagnosis, most of the officials acknowledged having reluctance to use it as a diagnosis in appropriate cases. This intellectual disconnect, even in medical examiners trained as forensic pathologists, impedes the collection of accurate national data on the prevalence of SUDEP. Relative to the education of patients with epilepsy and their families, in general, there is no consensus of how to approach the question of what patients and their families should know about the risk of SUDEP (Beran 2006). Clinical practice experience indicates that

SUDEP

949

there seems to be three approaches. The ἀrst is to say nothing, which is a variation of the “don’t ask, don’t tell” approach when dealing with uncomfortable issues; the second is to talk about SUDEP only if asked; the third is to tell those who are at increased risk, e.g., young males, those with poorly controlled generalized tonic–clonic seizures, persons tÂ�aking multiple antiepileptic drugs, especially when changes in medication is being contemplated, and those with localization related generalized tonic–clonic seizures; and ἀnally telling all patients. No one knows what the best approach is, but the elimination of the “don’t ask, don’t tell” approach should be the ἀrst effort. In conclusion, although the mechanism and prevention of SUDEP continue to be a clinical conundrum that will require future research to resolve, there is no justiἀcation for maintaining patients who are at risk for SUDEP and their families in a state of ignorance about its existence and the currently acknowledged risk factors.

References Annegers, J. F., and S. A. Blakley. 1990. Patterns of overall and unexplained death mortality among persons with epilepsy. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Beran, R. G. 2006. SUDEP—To discuss or not discuss: that is the question. Lancet Neurol 5 (6): 464–465. Bonvallet, M., and E. G. Bobo. 1972. Changes in phrenic activity and heart rate elicited by localized stimulation of amygdala and adjacent structures. Electroencephalogr Clin Neurophysiol 32 (1): 1–16. Earnest, M. P., G. E. Thomas, R. A. Eden, and K. F. Hossack. 1992. The sudden unexplained death syndrome in epilepsy: Demographic, clinical, and postmortem features. Epilepsia 33 (2): 310–316. Hilz, M. J., O. Devinsky, W. Doyle, A. Mauerer, and M. Dutsch. 2002. Decrease of sympathetic cardiovascular modulation after temporal lobe epilepsy surgery. Brain 125 (Pt 5): 985–995. Johnston, S. C., J. K. Horn, J. Valente, and R. P. Simon. 1995. The role of hypoventilation in a sheep model of epileptic sudden death. Ann Neurol 37 (4): 531–537. Johnston, S. C., R. Siedenberg, J. K. Min, E. H. Jerome, and K. D. Laxer. 1997. Central apnea and acute cardiac ischemia in a sheep model of epileptic sudden death. Ann Neurol 42 (4): 588–594. Kuller, L., A. Lilienfeld, and R. Fisher. 1967. Am epidemiological study of sudden an unexpected deaths in adults. Medicine 46: 341–361. Lathers, C. M., and P. L. Schraeder. 1990. Chapter 12. Synchronized cardiac neural discharge and epileptogenic activity, the lock-step phenomenon: Lack of correlation with cardiac arrhythmias. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9 (2): 236–242. Lathers, C. M., and P. L. Schraeder. 2009. Verbal autopsies. Epilepsy Behav 14: 573–576. Lathers, C. M., and P. L. Schraeder. 2010a. Stress and SUDEP (Chapter 17). In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. Leestma. Boca Raton, FL: CRC Press. Lathers, C. M., and P. L. Schraeder. 2010b. Animal model for sudden unexpected death in persons with epilepsy (Chapter 28). In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. Leestma. Boca Raton, FL: CRC Press. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008a. The mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12 (1): 3–24. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008b. Chapter 13. Sudden death: Neurocardiologic mystery. In Psychological Factors and Cardiovascular Disorders, ed. L. Sher. Hauppauge, NY: Nova Science.

950 Sudden Death in Epilepsy: Forensic and Clinical Issues Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67 (3): 247–259. Leestma, J. E. 1990a. Sudden unexpected death associated with seizures: A pathological review. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Leestma, J. E. 1990b. Natural history of epilepsy. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Leestma, J. E. 1990c. Sudden unexpected death associated with seizures: A pathological review (Chapter 5). In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Leestma, J. E., M. B. Kalelkar, S. S. Teas, G. W. Jay, and J. R. Hughes. 1984. Sudden unexpected death associated with seizures: Analysis of 66 cases. Epilepsia 25: 84–88. Leestma, J. E., T. Walczak, J. R. Hughes, M. B. Kalelkar, and S. S. Teas. 1989. A prospective study on sudden unexpected death in epilepsy. Ann Neurol 26 (2): 195–203. Leung, H., P. Kwan, and C. E. Elger. 2006. Finding the missing link between ictal bradyarrhythmia, ictal asystole, and sudden unexpected death in epilepsy. Epilepsy Behav 9 (1): 19–30. Mameli, P., O. Mameli, E. Tolu, G. Padua, D. Giraudi, M. A. Caria, and F. Melis. 1988. Neurogenic myocardial arrhythmias in experimental focal epilepsy. Epilepsia 29 (1): 74–82. Marshall, D. W., B. F. Westmoreland, and F. W. Sharbrough. 1983. Ictal tachycardia during temporal lobe seizures. Mayo Clin Proc 58 (7): 443–446. Mendez, M. F., J. L. Cummings, and D. F. Benson. 1986. Depression in epilepsy. Signiἀcance and phenomenology. Arch Neurol 43 (8): 766–770. Nashef, L. 1997. Sudden unexpected death in epilepsy: Terminology and deἀnitions. Epilepsia 38: S6–S8. Nashef, L., F. Walker, P. Allen, J. W. Sander, S. D. Shorvon, and D. R. Fish. 1996. Apnoea and bradycardia during epileptic seizures: Relation to sudden death in epilepsy. J Neurol Neurosurg Psychiatry 60 (3): 297–300. Nashef, L., and S. D. Shorvon. 1997. Mortality in epilepsy. Epilepsia 38 (10): 1059–1061. Natelson, B. H., R. V. Suarez, C. F. Terrence, and R. Turizo. 1998. Patients with epilepsy who die suddenly have cardiac disease. Arch Neurol 55 (6): 857–860. Nei, M., R. T. Ho, B. W. Abou-Khalil, F. W. Drislane, J. Liporace, A. Romeo, and M. R. Sperling. 2004. EEG and ECG in sudden unexplained death in epilepsy. Epilepsia 45 (4): 338–345. Ong, L. 2008. Negative affective disorders and arrhythmogenesis. In Psychological Factors and Cardiovascular Disorders, ed. L. Sher. Hauppauge, NY: Nova Science. Opherk, C., J. Coromilas, and L. J. Hirsch. 2002. Heart rate and EKG changes in 102 seizures: analysis of influencing factors. Epilepsy Res 52 (2): 117–127. Paterson, D. S. 2010. Medullary serotonergic abnormalities in sudden infant death syndrome: Implications in SUDEP. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. Leestma. Boca Raton, FL: CRC Press. Penἀeld, W., and H. Jasper. 1954. Summary of clinical analysis and seizure patterns. In Epilepsy and the Functional Anatomy of the Human Brain, ed. W. Penἀeld and H. Jasper. Boston, MA: Little Brown. Ring, A. R. 1997. Other psychiatric illnesses. In Epilepsy: A Comprehensive Textbook, ed. J. J. Engel and T. A. Pedley. Philadelphia, PA: Lippincott-Raven. Rocamora, R., M. Kurthen, L. Lickfett, J. Von Oertzen, and C. E. Elger. 2003. Cardiac asystole in epilepsy: Clinical and neurophysiologic features. Epilepsia 44 (2): 179–185. Rugg-Gunn, F. J., R. J. Simister, M. Squirrell, D. R. Holdright, and J. S. Duncan. 2004. Cardiac arrhythmias in focal epilepsy: A prospective long-term study. Lancet 364 (9452): 2212–2219. Samuels, M. 1997. Voodoo death revisited: The modern lessons of neurocardiology. The Neurologist 3: 293–304.

SUDEP

951

Schraeder, P. L., K. Delin, R. L. McClelland, and E. L. So. 2006. Coroner and medical examiner documentation of sudden unexplained deaths in epilepsy. Epilepsy Res 68 (2): 137–143. Schraeder, P. L., K. Delin, R. L. McClelland, and E. L. So. 2009. A nationwide survey of the extent of autopsy in sudden unexplained death in epilepsy. Am J Forensic Med Pathol 30 (2): 123–126. Schraeder, P. L., R. Pontzer, and T. R. Engel. 1983. A case of being scared to death. Arch Intern Med 143 (9): 1793–1794. Schraeder, P. L., E. L. So, and C. M. Lathers. 2010. Forensic case identiἀcation (Chapter 6). In Epilepsy in Sudden Death: Forensic and Clinical Issues. Boca Raton, FL: CRC Press. Stollberger, C., and J. Finsterer. 2004. Cardiorespiratory ἀndings in sudden unexplained/unexpected death in epilepsy (SUDEP). Epilepsy Res 59 (1): 51–60. Tomson, T., T. Walczak, M. Sillanpaa, and J. W. Sander. 2005. Sudden unexpected death in epilepsy: A review of incidence and risk factors. Epilepsia 46 (Suppl 11): 54–61. Wannamaker, B. B. 1990. Perspectives on death of persons with epilepsy. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder. New York: Marcel Dekker. Wendkos, M. H. 1979. Sudden Death and Psychiatric Illness. New York, NY: SP Medical and Scientiἀc Books.

Epilepsy and SUDEP Lessons Learned: Scientific and Clinical Experience

60

Claire M. Lathers Paul L. Schraeder

Contents 60.1 Proposed Areas of Inquiry about SUDEP [106] References

954 961

Most likely, different mechanisms and/or different combinations of mechanisms are responsible for death in persons with epilepsy. Contributing mechanisms at peripheral sites include, in part, a direct action on the heart, indirect and/or direct actions on the lungs, and actions on the adrenal glands. In the discussion of our papers published over the years, we listed the following mechanisms thought to be contributing to sudden unexpected death in epilepsy (SUDEP) (Table 60.1). Based on their ἀndings in the above animal models, Lathers and Schraeder emphasized the importance of continuous electrocardiography (EKG) and electroencephalography (EEG) recordings in identifying persons with epilepsy at risk for sudden death. Despite this three-decade effort, discussion at the recent (November 2008) National Institutes of Health (NIH) SUDEP Workshop (NINCDS 2008) revealed that many physicians still do not concurrently examine heart and brain electrical activities when evaluating persons at risk for sudden death. It will take a global effort of multidisciplinary research collaboration among physicians and basic scientists to evaluate the importance of simultaneously monitoring sophisticated EKG and EEG activity in persons thought to be at risk for sudden death. Integration of neurology, cardiology, and clinical pharmacology is needed for the best diagnostic and treatment interventions for persons at risk for SUDEP. Lathers and Schraeder (1990) edited the ἀrst book on SUDEP, Epilepsy and Sudden Death, which summarized the “state of the art” experimental and clinical information available about sudden death and epilepsy. Today it still provides a guide for how basic researchers and clinicians should approach detection, understanding, and prevention of SUDEP. J. Thomas Bigger Jr., MD, Professor Medicine and Pharmacology, College of Physicians and Surgeons at Columbia University, succinctly summarized in the Foreword eight areas of inquiry that needed to be addressed to answer questions about SUDEP. The following is published with permission (Lathers 2009).

953

954 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 60.1â•… Summary: Mechanisms/Sites Associated with SUDEP 1. Nonuniform autonomic postganglionic cardiac sympathetic neural discharge related to the nonuniform beta sympathetic receptor locations in the heart just before development of arrhythmias associated with interictal and ictal activity (Lathers and Schraeder 1982; Schraeder and Lathers 1983; Lathers, Schraeder, and Carnel 1984; Carnel, Schraeder, and Lathers 1985; Lathers et al. 1986; Lathers and Spivey 1990; Lathers, Levin, and Spivey 1986; Lathers, Spivey, and Levin 1988). 2. Autonomic imbalance between sympathetic and parasympathetic neural discharges and variability within each autonomic division are associated with varying degrees of epileptogenic activity and can be associated with ventricular ἀbrillation or asystole (Lathers and Schraeder 1982; Schraeder and Lathers 1983; Lathers, Schraeder, and Carnel 1984; Carnel, Schraeder, and Lathers 1985). 3. Autonomic parameter dysfunction of heart rate and blood pressure occurs before development of interictal discharges and continues with ictal discharges (Lathers and Schraeder 1982; Schraeder and Lathers 1983; Lathers, Schraeder, and Carnel 1984; Carnel, Schraeder, and Lathers 1985; Lathers, Schraeder, and Weiner 1987; Stauffer, Dodd-O, and Lathers 1989, 1990; Dodd-O and Lathers 1990; O’Rourke and Lathers 1990; Lathers and Schraeder 1990). 4. Autonomic cardiac neural discharges are intermittently synchronized 1:1 with the epileptogenic discharge, i.e., the lockstep phenomenon (LSP) (Lathers, Schraeder, and Weiner 1987; Stauffer, Dodd-O, and Lathers 1989, 1990; Dodd-O and Lathers 1990; O’Rourke and Lathers 1990; Lathers and Schraeder 1990). 5. Epileptogenic activity may alter autonomic central (Chadwick, Jenner, and Reynolds 1975; Oishi, Suenaga, and Fukuda 1979; Mason and Corcoran 1979) or peripheral autonomic release of catecholamines (Ceremuzynski, Staszewska-Barczak, and Herbaczynska-Cedro 1969; Kelliher, Widmer, and Roberts 1975). 6. Multiple areas of pulmonary punctuate hemorrhages and gross hemorrhage and edema were found in animals dying after epileptogenic-activity-induced asystole or ventricular ἀbrillation (Lathers and Schraeder 1982; Schraeder and Lathers 1983; Lathers, Schraeder, and Carnel 1984). 7. Tissue hypoxia, hypercarbia, and alterations in acid–base balance may contribute to the results in experimental epilepsy (Lathers and Schraeder 1982; Schraeder and Lathers 1983; Lathers, Schraeder, and Carnel 1984; Carnel, Schraeder, and Lathers 1985). 8. Changes in cardiac function alter cerebral blood flow and may produce central hypoxia, resulting to epileptogenic activity (Schott, McLeod, and Jewitt 1977; Schraeder, Pontzer, and Engel 1983; Schwartz and Lathers 1990). 9. Modulation of the major central nervous system presynaptic inhibitory neurotransmitter gamma aminobutyric acid release by prostaglandin E2 may be an explanation for epileptogenic–activity-related dysfunction of autonomic cardiac neural discharge leading to arrhythmias (Suter and Lathers 1984; Schwartz and Lathers 1990). Mechanisms for interference of GABA neurotransmission may lead to initiation of arrhythmias and/or epileptogenic activity (Schwartz 1988; Rastogi and Ticku 1986) and sudden death (Suter and Lathers 1984; Killam and Bain 1957). 10. Beta blockers exhibited anticonvulsant activity, whether administered via the intraosseous route or intravenously (Lathers, Schraeder, and Bungo 2008a; Jim et al. 1988; Lathers and Jim 1990). Source: Lathers, C. M., Epilepsy Behav, 15, 269–277, 2009. With permission.

60.1â•… Proposed Areas of Inquiry about SUDEP [106] Dr. Thomas Bigger has stated in the preface of Epilepsy and Sudden Death (Lathers and Schraeder 1990) that “Although many pieces of the epilepsy–sudden death puzzle are in place, the total picture is not clear. . . . We seem a long way from understanding and controlling the problem of sudden death in epilepsy. How shall we proceed toward those goals?” The following list summarizes his view of what research needed to be done to investigate the mechanisms and prevention of SUDEP.

Epilepsy and SUDEP

955

1. Additional epidemiologic studies to sharpen the focus on the high-risk group. 2. Studies with ambulatory EEG, EKG, and respiratory recordings in high-risk individuals to capture the events before and during sudden death. 3. The role of the sympathetic nervous system could be explored more intensively with functional and biochemical evaluation of cardiac sympathetic nervous activity. 4. The anatomic distribution of cardiac sympathetic nerves in persons with epilepsy vs. controls could be clariἀed by imaging studies using biological markers taken up by sympathetic nerve terminals in the heart. 5. Studies in relevant animal models will permit a forward leap in our capability for generating relevant new knowledge and greatly accelerate the rate of progress in clarifying the pathophysiology of SUDEP. 6. Hypotheses based on animal data about effective treatments will naturally follow and can be pursued using clinical studies in persons with epilepsy. 7. Studies of the factors that govern compliance in epileptic patients should proceed now because the results could be applied immediately to maximizing control of seizure disorders and thereby decreasing the risk of SUDEP. We are on the threshold of major advances in the problem of sudden death in epilepsy. The tools to advance our knowledge are at hand and we should energetically put them to use for the future beneἀt of patients with epilepsy. Knowledge gained from studies of sudden death in epileptics will very likely be useful for understanding sudden death in other situations as well. (Bigger 1990)

Only some of Dr. Bigger’s suggested areas of investigation have now been addressed. 1. Additional epidemiology studies have been conducted (Earnest et al. 1992; Ficker et al. 1998; Walczak et al. 2001; Moran et al. 2004; Rugg-Gunn et al. 2004). Also see reviews by Lathers et al. (Lathers, Schraeder, and Bungo 2008a, 2008b; Scorza, Arida, and Cavalheiro 2008; Lathers, Schraeder, and Bungo 2010; Bacon 1868; Lathers and Schraeder 1990). 2. Studies of ambulatory recordings to capture recordings before and during sudden death are still needed today. Recordings of EEG, EKG, and respiration must still be done. Simultaneous recordings of these variables, and arterial oxygen saturation, during sudden death would permit us to evaluate the role of apnea or hypoxia. Postictal central apnea appears to be one potential mechanism for SUDEP. A 55-s convulsive seizure occurred in a 20-year-old woman as she underwent videoEEG monitoring (So, Sam, and Lagerlund 2000). Persistent apnea then developed. Electrocardiogram-monitored rhythm was not altered for the ἀrst 10 s, then it gradually and progressively slowed and stopped 57 s later. Cardiorespiratory resuscitation was successful. No evidence of airway obstruction or pulmonary edema was noted. One previous cardiorespiratory arrest after a complex partial seizure without secondary generalization had been reported for this patient. So et al. (2000) note that although epileptic seizures may be associated with arrhythmogenic actions at the heart, in this patient the mechanism of marked central suppression of respiratory activity after seizures was clearly involved and almost resulted in sudden death. The timing of events such as seizures, respiratory and/or laryngospasm, and cardiac EKG changes varies in different patients. The physician

956 Sudden Death in Epilepsy: Forensic and Clinical Issues

must consider risk factors for a given patient to recommend protective procedures to minimize the chance of unwanted events that may result in SUDEP. The question must be asked as to whether a person ἀrst experienced seizures and respiratory events and then cardiac events or if the person experienced seizures and arrhythmia and then respiratory events. There is documented evidence in the literature to support both cardiac and respiratory events as initiating mechanisms of sudden death. Obviously, rapid reversal of these changes is essential and the “availability of resuscitation methods on the spot where the victim is located” certainly increases the likelihood that SUDEP will be prevented. In a recent study, Bateman et al. (2008) examined the incidence and severity of ictal hypoxemia in patients with localization-related epilepsy undergoing video-EEG telemetry. They measured seizure-associated oxygen desaturation and hypoventilation. Pulse oximetry revealed oxygen desaturations below 90% in one-third of all 304 seizure events. The degree of desaturation was signiἀcantly correlated with seizure duration and with electrographic evidence of seizure spread to the contralateral hemisphere. Central apneas or hypopneas occurred with 50% of all seizures. Ictal hypoxemia occurred often in these patients with localization-related epilepsy, and may be pronounced and prolonged, even if the seizures do not progress to generalized convulsions. End tidal carbon dioxide increase occurred in oxygen desaturation and supports the assumption that ictal oxygen desaturation is a consequence of hypoventilation. Both ictal hypoxemia and hypercapnia may be contributing risk factors to SUDEP occurrence. 3. The role of the sympathetic nervous system could be explored more intensively with functional and biochemical evaluation of cardiac sympathetic nervous activity. Lathers and Levin (2010) have examined the distribution of cardiac beta receptors and found a correlation with the release of norepinephrine at sympathetic nerve terminals in the heart concurrent with the production of arrhythmia. Innervation density is high in the subepicardium and the central conduction system. In diseased hearts, cardiac innervation density varies. This may lead to sudden cardiac death. After myocardial infarction, sympathetic denervation is followed by reinnervation within the heart, leading to unbalanced neural activation and lethal arrhythmia (Lathers et al. 1986; Lathers and Spivey 1990; Lathers, Levin, and Spivey 1986; Lathers, Spivey, and Levin 1988). Recently, Ieda et al. (2008), as in the earlier studies by Lathers et al. (Lathers et al. 1977; Lathers, Roberts, and Kelliher 1977; Lathers et al. 1978; Lathers, 1980b, 1981, 1982; Lathers and Schraeder 1982; Schraeder and Lathers 1983), have raised the question of whether regulation of cardiac nerves is a “new paradigm” in the management of sudden cardiac death since the heart is extensively innervated and its performance is regulated by the autonomic nervous system. In the case of diabetic sensory neuropathy, silent myocardial ischemia may occur, associated with loss of pain perception during myocardial ischemia, a major cause of sudden cardiac death in diabetes mellitus (Ieda et al. 2006). To date, molecular mechanisms underlying innervation density are not well understood. Ieda et al. (2006) have demonstrated that cardiac sympathetic innervation is determined by the balance of neural chemoattraction and cheomorepulsion, both of which occur in the heart. Nerve growth factor, a potent chemoattractant, is synthesized by cardiomyoctyes and is induced by endothelin-l upregulation in the heart. In contrast, Sema3a, a neural chemorepellent, is expressed strongly in

Epilepsy and SUDEP

957

the trabecular layer in early-stage embyros and, at a lower level after birth, cause epipcardial-to-endocardial transmural sympathetic innervation pattern. Cardiac nerve growth factor downregulation is a cause of diabetic neuropathy, and nerve growth factor supplementation rescues silent myocardial ischemia in diabetic neuropathy. Both Sema3a-deἀcient and Sema3a-overexpressing mice showed sudden death or lethal arrhythmias due to disruption of innervation patterning (Ieda et al. 2007). All of these regulatory mechanisms involved in neural development in the heart and their critical roles in cardiac performance need to be examined to determine relevance to methods to decrease the risk of SUDEP. 4. The issue of using imaging studies to examine the anatomic distribution of cardiac sympathetic nerves in epileptics vs. normals has been answered recently. A postmortem imaging study, using 123I-metaiodobenzylguanidine single photon emission computed tomography (SPECT), has been conducted of postganglionic cardiac sympathetic innervation in patients with chronic temporal lobe epilepsy (Druschky et al. 2001), ἀnding sympathetic dysfunction in the form of altered postganglionic cardiac sympathetic innervation in patients with chronic temporal lobe epilepsy. These observations suggested that the altered postganglionic cardiac sympathetic innervation may increase risk of cardiac abnormalities and/or SUDEP. The exact role of innervation in arrhythmogenesis and developmental and regulatory mechanisms determining density and pattern of cardiac sympathetic innervation are still unclear. This clinical study of Druschky et al. (2001), conducted in humans, conἀrms the results and conclusions of the animal studies conducted by Lathers et al. (Lathers et al. 1977; Lathers, Roberts, and Kelliher 1977; Lathers et al. 1978; Lathers 1980a, 1981, 1982) in which the relationship of postganglionic cardiac sympathetic neural discharge was associated with arrhythmias and/or sudden death. Kerling et al. (2009) stated that since tachyarrhythmias are common during epileptic seizures while bradyarrhythmias or asystoles occur less frequently, they evaluated cardiac postganglionic denervation in patients with epilepsy to evaluate ictal asystole. SPECT examined 123I-meta-iodobenzylguanidine as a marker of postganglionic cardiac norepinephrine uptake. The pronounced reduction in cardiac SPECT uptake in asystolic patients indicated postganglionic cardiac catecholamine disturbance. Impaired sympathetic cardiac innervation limits adjustment and modulation of heart rate and may increase the risk of asystolic events and, eventually, sudden unexpected death in persons with epilepsy. The data of Kerling et al. (2009) support the ἀndings of Lathers, Levin, and Spivey (Lathers et al. 1986; Lathers and Spivey 1990; Lathers, Levin, and Spivey 1986; Lathers, Spivey, and Tumer 1988) and those of Han and Moe (1964). 5. Studies in relevant animal models. Many different relevant animal models for SUDEP are still needed to understand the pathophysiology of sudden death in epileptics and to hypothesize about effective treatments (Bigger 1990). The importance of using many different animal models to study SUDEP to glean an insight into the various mechanisms of risks and their contribution to the initiation of the death event is discussed in detail by Lathers (2010). Schwartz et al. (1995) concluded that delayed enhancement of GABAergic neurotransmission directly at the site of vulnerability after an ischemic event protects the neurons from death. This ἀnding

958 Sudden Death in Epilepsy: Forensic and Clinical Issues

should be explored to further study the effect of diazepam on GABA-mediated effects that may prevent ischemia-induced neuronal death and ultimately prevent the worsening of central neuronal communication due to epileptogenic activity. This may eventually contribute a protective central nervous system effect to make an individual less likely to be at risk for SUDEP. There are many more basic science questions to be raised and answered. The study of Wang et al. (2006), discussed above, must be expanded to examine antiepileptic or other categories of drugs and their effect on the intermittent seizure-like ἀring of cardiac parasympathetic neurons that may cause neurogenic ictal bradyarrhythmias, cardiac asystole, or sudden death. So (2008) emphasized the signiἀcance of using audiogenic seizure mice to study postictal respiratory arrest. Postictal respiratory arrest was induced by serotonin receptor inhibition and prevented by selective serotonin reuptake inhibitor drugs. The role of serotonin in SUDEP must be examined in future animal studies (Tupal and Faingold 2006). The reader is referred to Dr. Paterson’s chapter in this book. He has presented a very thorough discussion of medullary serotonergic low levels found in sudden infant death syndrome and the implications for SUDEP. This ἀnding of low brainstem serotonin levels raises the possibility of doing similar functional magnetic resonance imaging studies in adults with epilepsy vs. controls. 6. Hypotheses about effective treatments will naturally follow and can be pursued using animal models, small pilot studies in epileptic patients, and, ἀnally, largescale clinical trials (Bigger 1990). We have not yet fully addressed this area of inquiry about SUDEP. There are several questions to be asked postmortem (Schraeder et al. 2006; Schraeder et al. 2009; Lathers and Schraeder 2009) about the patient who is on an antiepileptic drug but still becomes a SUDEP victim. A very important clinical pharmacology question that must be asked is whether the patient was on the correct antiepileptic drug to control his particular type/ mixture of seizures. A second question to be asked is whether the correct categories of drugs have been prescribed. There may be a role for using beta blocking agents. When evaluating the role of drugs as protectors of life, clinical pharmacologists (Lathers and Schraeder 2002) caution us to remember that use of all drugs is a risk/beneἀt ratio evaluation (Lathers and Schraeder 2002). Thus, the use of antiepileptic drugs may not 100% protect the patient against sudden death. Celiker at al. (2009) report clinical experience of patients with catecholaminergic polymorphic ventricular tachycardia and concluded medical treatment with propranolol and verapamil may decrease the incidence of arrhythmia. If a person is still refractory, implantation of intracardiac deἀbrillators should be considered. Houle et al. (2001) reported enhanced in vivo and in vitro contractile responses to beta2-adrenergic receptor stimulation in those susceptible to lethal arrhythmias. Billman et al. (1997, 2006) found that beta2-adrenergic receptor antagonist protected against ventricular ἀbrillation, and endurance exercise training attenuates cardiac beta2-adrenoceptor responsiveness and prevents ventricular ἀbrillation in animals. There seems to be a direct correlation between beta adrenergic receptor sensitivity and the level of arrhythmia and mortality following a coronary occlusion. The greater the sensitivity to beta adrenergic agonists, whether from hormones, exercise, or genetics, the greater is the level of arrhythmia and mortality (Houle, Altschuld, and Billman 2001; Billman et al. 1997, 2006; Du

Epilepsy and SUDEP



959

et al. 2000). Whether we can learn from these data when considering persons with epilepsy at risk for SUDEP remains to be seen. Data suggest that beta blockers exert a protective effect against seizure induction and/or the development of cardiac arrhythmias with interictal and ictal activity (Lathers, Schraeder, and Bungo 2008a, 2008b; Scorza, Arida, and Cavalheiro 2008; Lathers, Schraeder, and Bungo 2010; Jim et al. 1988; Lathers and Jim 1990; Lathers and Schraeder 1990; Lathers et al. 1990a, 1990b). The question must be asked if persons thought to be at risk for SUDEP should be placed on a beta blocker in addition to the prescribed anticonvulsant(s). Beta blockers are also used to reduce stress, and persons with epilepsy generally are stressed by the disease and associated problems (Lathers and Schraeder 2006). 7. Studies of the factors that govern compliance in epileptic patients should proceed now (Bigger 1990). We have not yet fully addressed this area of inquiry relative to prevention of€ SUDEP. There are several questions to be asked postmortem (Schraeder et al. 2006, 2009; Lathers and Schraeder 2009) about the patient who is on an antiepileptic drug but still becomes a SUDEP victim. To avoid SUDEP, one must inquire family and friends as to whether the person was compliant (Lathers and Schraeder 2002). In 2003, Lathers et al. (Lathers, Koehler, Wecht, and Schraeder 2003a, 2003b) found low or no levels of antiepileptic drugs postmortem in persons with epilepsy deemed to have died of SUDEP. These data suggested that compliance is a problem in some postmortem studies of victims of SUDEP. In 2009, Hughes (2009) deemed the most important SUDEP risk factor to be nonÂ�compliance with antiepileptic medication. Ryvlin et al. (2009) found the risk of SUDEP is increased in patients who have poor compliance and exhibit nocturnal seizures and generalized tonic–clonic seizures. Although there are some questions about the reliability of postmortem antiepileptic drug levels (Tomson et al. 1998), it seems reasonable to state that non-compliance is an issue if very low or no drug levels are found at autopsy. However, compliance is not the only question to ask about victims of SUDEP. One must also ask if the correct dose of the antiepileptic drug was being used for a given patient to control his/her seizures. All agree that maintenance of an optimal therapeutic drug level for a given individual is crucial to avoid SUDEP.

Scientists, clinicians, and granting authorities must not allow another 20 years to go by with little additional data gleaned relative to lowering the risk of SUDEP (Lathers 2009). Future steps have been suggested by The American Epilepsy Society and the Epilepsy Foundation Joint Task Force on SUDEP (So et al. 2009). During the interim, before the answers have been obtained, it is most important to provide prompt and optimal control of seizures, especially generalized convulsive seizures, to prevent the occurrence of SUDEP. Assessment of the current knowledge about SUDEP concludes: 1. A need for multidisciplinary workshops to reἀne current lines of investigation and identify additional areas of research for mechanisms underlying SUDEP. 2. To conduct a survey of patients, families, and caregivers to identify effective means of education that will enhance participation in SUDEP research. 3. A campaign to emphasize the need for complete autopsy examinations for patients with suspected SUDEP.

960 Sudden Death in Epilepsy: Forensic and Clinical Issues Table 60.2â•… Summary Lessons Learned—Epilepsy and SUDEP: Global Focus Needed 1. Mechanistic risk factors for SUDEP, obtained in animal studies in our laboratory, included cardiac arrhythÂ�Â�mias and/or death associated with changes in autonomic cardiac postganglionic sympathetic neural discharge, cardiac parasympathetic neural discharge, and respiratory changes including multiple areas of punctuate hemorrhages and large areas of gross pulmonary hemorrhage and pulmonary edema (Lathers and Schraeder 1982; Schraeder and Lathers 1983; Lathers, Schraeder, and Carnel 1984; Carnel, Schraeder, and Lathers 1985). Both central and peripheral sites are involved in the pathophysiology of SUDEP (Table 60.1). 2. Many different relevant animal models for SUDEP are still needed (Lathers 2010) to understand the pathophysiology of sudden death in epileptics, hypothesize about effective treatments, develop pilot studies in persons with epilepsy, and, ἀnally, conduct conἀrmatory large-scale clinical trials. 3. “Think outside of the box” when evaluating an established animal model with potential for modiἀcaÂ� tion(s) to be used to address questions about the mechanism(s) of SUDEP. 4. The ἀeld of pharmacology/clinical pharmacology has much to offer as we work to improve compliance and to develop new antiepileptic drugs and/or apply new categories of drugs, such as beta blockers or selective serotonin reuptake inhibitors, to prevent and resolve the mystery of SUDEP (Lathers, Schraeder, and Bungo 2008a, 2008b; Scorza, Arida, and Cavalheiro 2008; Lathers, Schraeder, and Bungo 2010; Jim et al. 1988; Lathers et al. 1989; Lathers, Jim, and Spivey 1989; Jim et al. 1989; Lathers and Jim 1990; Lathers and Schraeder 1990; Lathers and Schraeder 2002). 5. Team work is needed among different multidisciplinary professionals working in the clinical settings and/or within a laboratory, among laboratories within the United States, and in laboratories located around the world to solve the global mystery of SUDEP. 6. Ambulatory simultaneous EKG and EEG telemetry monitoring of patients thought to be at risk for sudden death will help identify the possibility of between cardiac predisposition to potentially arrhythÂ� mogenic and brain epileptogenic triggers/causes to be treated or prevented, in an attempt to decrease the risk for SUDEP. Respiratory function monitoring is also needed (Bigger 1990). 7. Academic fellowships and competitions for medical students and postdoctoral fellows/residents and faculty will attract medical and graduate students and faculty to work in the ἀeld of SUDEP. 8. Grant funding is essential to move the SUDEP knowledge base forward. Academic administrative leaders are not interested in faculty and investigators addressing a problem that is not well funded at the national/international levels. 9. Pharmaceutical industry leaders are not interested in addressing a problem if the market for a new product is not a large one. Incentives must be offered to encourage them to examine the problem of SUDEP and to develop new drugs that may or may not have a large market. If the true incidence of SUDEP is established to be much higher than previously thought to be, by the correct use of the term SUDEP on autopsy reports and/or the use of verbal autopsies postmortem, then the market for new antiepileptic and/or other new drugs in different categories to treat SUDEP will be larger than presently determined. The Food and Drug Administration category of orphan drug development should be considered if necessary. 10. Leadership foundations of vision, knowledge, and courage are essential to address the global mystery of SUDEP. Use of a leadership philosophy foundation that provides strength primarily for research and teaching programs from faculty members and students and secondarily on the administration that must provide the innovative vision and approaches, facilities, and monies to support the needs of the faculty and students. The interaction of teaching and research is essential. While a student is learning how to conduct research, he must simultaneously learn how to become a teacher himself. The components of teaching, including excellent communication and writing skills, coupled with patience, form the€foundÂ� ation for communicating the ἀndings of research to the academic and research communities and to fundÂ�ing agencies. Today’s medical and graduate programs will not be able to train each student for each and every advance that will develop in his selected profession, including a focus on SUDEP. Therefore, the program must help today’s student understand the basic approach to survival in a rapidly changing technological, academic, and political environment. The best prepared student for the challenges of tomorrow will be one trained to be flexible, possessing the basic knowledge and tools and courage to adapt to the different career twists and turns encountered as modern technology rushes to the forefront with numerous new techniques, understandings, and thoughts foreign to us at this time in our lives. (continued)

Epilepsy and SUDEP

961

Table 60.2â•… (Continued) Today’s medial and graduate leadership must have the vision to provide a fertile and proper environment for teachers to work with academic freedom to teach today’s students how to become the self-learning student of tomorrow. While addressing issues of SUDEP, teach our students of today to become selflearners and leaders in the ἀeld for tomorrow’s solutions (Lathers 1995a, 1995b, 1994). Source: Lathers, C. M., Epilepsy Behav, 15, 269–277, 2009. With permission.

4. To secure infrastructure grants to fund a consortium of centers that will conduct prospect clinical and basic research studies to identify preventable risk factors and mechanisms underlying SUDEP. Additional symposia must be held to encourage discussion and “thinking out of the box” to solve the problem of SUDEP (Table 60.2). Around the world, all must work to introduce excitement and an intellectual interest in young students, scientists, and clinicians, with the goal of encouraging them to focus on epilepsy and sudden death. Only by encouraging an understanding of the importance of solving the mystery of SUDEP for young investigators will we be able to ἀnd answers that solve the mystery of sudden death (Lathers, Schraeder, and Bungo 2008a, 2008b; Scorza, Arida, and Cavalheiro 2008; Lathers, Schraeder, Bungo 2010; Lathers and Schraeder 1990).

References Bacon, G. M. 1868. On the modes of death in epilepsy. Lancet 1: 555–556. Bateman, L. M., C. S. Li, and M. Seyal. 2008. Ictal hypoxemia in localization-related epilepsy: Analysis of incidence, severity and risk factors. Brain 131 (Pt 12): 3239–3245. Bigger, J. T. 1990. Foreword. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Billman, G. E., L. C. Castillo, J. Hensley, C. M. Hohl, and R. A. Altschuld. 1997. Beta2-adrenergic receptor antagonists protect against ventricular ἀbrillation: In vivo and in vitro evidence for enhanced sensitivity to beta2-adrenergic stimulation in animals susceptible to sudden death. Circulation 96 (6): 1914–1922. Billman, G. E., M. Kukielka, R. Kelley, M. Moustafa-Bayoumi, and R. A. Altschuld. 2006. Endurance exercise training attenuates cardiac beta2-adrenoceptor responsiveness and prevents ventricular ἀbrillation in animals susceptible to sudden death. Am J Physiol Heart Circ Physiol 290 (6): H2590–H2599. Carnel, S. B., P. L. Schraeder, and C. M. Lathers. 1985. Effect of phenobarbital pretreatment on cardiac neural discharge and pentylenetetrazol-induced epileptogenic activity in the cat. Pharmacology 30 (4): 225–240. Celiker, A., I. Erdogan, T. Karagoz, and S. Ozer. 2009. Clinical experiences of patients with catecholaminergic polymorphic ventricular tachycardia. Cardiol Young 19 (1): 45–52. Ceremuzynski, L., J. Staszewska-Barczak, and K. Herbaczynska-Cedro. 1969. Cardiac rhythm disturbances and the release of catecholamines after acute coronary occlusion in dogs. Cardiovasc Res 3 (2): 190–197. Chadwick, D., P. Jenner, and E. H. Reynolds. 1975. Amines, anticonvulsants, and epilepsy. Lancet 1 (7905): 473–476.

962 Sudden Death in Epilepsy: Forensic and Clinical Issues Dodd-O, J. H., and C. M. Lathers. 1990. Chapter 13. A characterization of the lockstep phenomenon in phenobarbital-pretreated cats. In Epilepsy and Sudden Death. New York, NY: Marcel Dekker. Druschky, A., M. J. Hilz, P. Hopp, G. Platsch, M. Radespiel-Troger, K. Druschky, T. Kuwert, H. Stefan, and B. Neundorfer. 2001. Interictal cardiac autonomic dysfunction in temporal lobe epilepsy demonstrated by [(123)I]metaiodobenzylguanidine-SPECT. Brain 124 (Pt 12): 2372–2382. Du, X. J., X. M. Gao, G. L. Jennings, A. M. Dart, and E. A. Woodcock. 2000. Preserved ventricular contractility in infarcted mouse heart overexpressing beta(2)-adrenergic receptors. Am J Physiol Heart Circ Physiol 279 (5): H2456–H2463. Earnest, M. P., G. E. Thomas, R. A. Eden, and K. F. Hossack. 1992. The sudden unexplained death syndrome in epilepsy: Demographic, clinical, and postmortem features. Epilepsia 33 (2): 310–316. Ficker, D. M., E. L. So, W. K. Shen, J. F. Annegers, P. C. O’Brien, G. D. Cascino, and P. G. Belau. 1998. Population-based study of the incidence of sudden unexplained death in epilepsy. Neurology 51 (5): 1270–1274. Han, J., and G. K. Moe. 1964. Nonuniform recovery of excitability in ventricular muscle. Circ Res 14: 44–60. Houle, M. S., R. A. Altschuld, and G. E. Billman. 2001. Enhanced in vivo and in vitro contractile responses to beta(2)-adrenergic receptor stimulation in dogs susceptible to lethal arrhythmias. J Appl Physiol 91 (4): 1627–1637. Hughes, J. R. 2009. A review of sudden unexpected death in epilepsy: Prediction of patients at risk. Epilepsy Behav 14 (2): 280–287. Ieda, M., H. Kanazawa, K. Kimura, F. Hattori, Y. Ieda, M. Taniguchi, J. K. Lee, et al. 2007. Sema3a maintains normal heart rhythm through sympathetic innervation patterning. Nat Med 13 (5): 604–612. Ieda, M., H. Kanazawa, Y. Ieda, K. Kimura, K. Matsumura, Y. Tomita, T. Yagi, et al. 2006. Nerve growth factor is critical for cardiac sensory innervation and rescues neuropathy in diabetic hearts. Circulation 114 (22): 2351–2363. Ieda, M., K. Kimura, H. Kanazawa, and K. Fukuda. 2008. Regulation of cardiac nerves: A new paradigm in the management of sudden cardiac death? Curr Med Chem 15 (17): 1731–1736. Jim, K. F., C. M. Lathers, V. L. Farris, L. F. Pratt, and W. H. Spivey. 1989. Suppression of pentylenetetrazol-elicited seizure activity by intraosseous lorazepam in pigs. Epilepsia 30 (4): 480–486. Jim, K. F., C. M. Lathers, W. H. Spivey, W. D. Matthews, C. Kahn, and K. Dolce. 1988. Suppression of pentylenetetrazol-elicited seizure activity by intraosseous propranolol in pigs. J Clin Pharmacol 28 (12): 1106–1111. Kelliher, G. J., C. Widmer, and J. Roberts. 1975. Influence of the adrenal medulla on cardiac rhythm disturbances following acute coronary artery occlusions. Recent Adv Stud Cardiac Struct Metab 10: 387–400. Kerling, F., M. Dutsch, R. Linke, T. Kuwert, H. Stefan, and M. J. Hilz. 2009. Relation between ictal asystole and cardiac sympathetic dysfunction shown by MIBG-SPECT. Acta Neurol Scand 120 (2): 123–129. Killam, K. F., and J. A. Bain. 1957. Convulsant hydrazides. I. In vitro and in vivo inhibition of vitamin B6 enzymes by convulsant hydrazides. J Pharmacol Exp Ther 119 (2): 255–262. Lathers, C. M. 1980a. Effect of timolol on autonomic neural discharge associated with ouabaininduced arrhythmia. Eur J Pharmacol 64 (2–3): 95–106. Lathers, C. M. 1980b. The effect of metoprolol on coronary occlusion-induced arrhythmia and autonomic neural discharge. Fed Proc 39: 771. Lathers, C. M. 1981. Induced disease. Myocardial infarction in dogs and cats. In Mammalian Models for Research on Aging, 224–228. Washington, DC: National Academy Press. Lathers, C. M. 1982. Lack of effect of methylprednisolone on cardiac neural discharge associated with coronary occlusion-induced arrhythmia and death. Eur J Pharmacol 85 (2): 233–238.

Epilepsy and SUDEP

963

Lathers, C. M. 2009. Epilepsy and sudden death: Personal reflections and call for global action. Epilepsy Behav 15 (3): 269–277. Lathers, C. M., G. J. Kelliher, J. Roberts, and A. B. Beasley. 1978. Nonuniform cardiac sympathetic nerve discharge: Mechanism for coronary occlusion and digitalis-induced arrhythmia. Circulation 57 (6): 1058–1065. Lathers, C. M., J. Roberts, and G. J. Kelliher. 1977. Correlation of ouabain-induced arrhythmia and nonuniformity in the histamine-evoked discharge of cardiac sympathetic nerves. J Pharmacol Exp Ther 203 (2): 467–479. Lathers, C. M., K. F. Jim, W. B. High, W. H. Spivey, W. D. Matthews, and T. Ho. 1989. An investigation of the pathological and physiological effects of intraosseous sodium bicarbonate in pigs. J Clin Pharmacol 29 (4): 354–359. Lathers, C. M., K. F. Jim, W. H. Spivey, C. Kahn, K. Dolce, and W. D. Matthews. 1990a. Chapter 24: Antiepileptic activity of beta-blocking agents. In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Lathers, C. M., K. F. Jim, and W. H. Spivey. 1989. A comparison of intraosseous and intravenous routes of administration for antiseizure agents. Epilepsia 30 (4): 472–479. Lathers, C. M., K. M. Keller, J. Roberts, and A. B. Beasley. 1977. Chapter 5. Role of the adrenergic nervous system in arrhythmia produced by acute coronary artery occlusion. In Pathophysiology and Therapeutics of Myocardial Ischemia, ed. A. M. Leffer, G. J. Kelliher and M. Rovetto. New York, NY: Spectrum. Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67 (3): 247–259. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008a. The mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12 (1): 3–24. Lathers, C. M., P. L. Schraeder, and S. B. Carnel. 1984. Neural mechanisms in cardiac arrhythmias associated with epileptogenic activity: The effect of phenobarbital in the cat. Life Sci 34 (20): 1919–1936. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23 (6): 633–647. Lathers, C. M., and P. L. Schraeder. 2002. Clinical pharmacology: Drugs as a beneἀt and/or risk in sudden unexpected death in epilepsy? J Clin Pharmacol 42 (2): 123–136. Lathers, C. M., and P. L. Schraeder. 2006. Stress and sudden death. Epilepsy Behav 9 (2): 236–242. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008b. Chapter 13. Sudden death: Neurocardiologic mystery. In Psychological Factors and Cardiovascular Disorders, ed. L. Sher. Hauppauge, NY: Nova Science. Lathers, C. M., and P. L. Schraeder, eds. 1990. Epilepsy and Sudden Death. New York, NY: Marcel Dekker. Lathers, C. M., R. M. Levin, and W. H. Spivey. 1986. Regional distribution of myocardial betaÂ�adrenoceptors in the cat. Eur J Pharmacol 130 (1–2): 111–117. Lathers, C. M., and R. M. Levin. 2010. Chapter 33: Animal model for sudden cardiac death:€Sympathetic innervation and myocardial beta-receptor densities. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. Leestma. Boca Raton, FL: CRC Press. Lathers, C. M., S. A. Koehler, C. H. Wecht, and P. L. Schraeder. 2003a. Forensic antiepileptic drug levels in 2001 autopsy cases of sudden, unexpected deaths in persons with epilepsy in Allegheny County Pennsylvania. Paper read at the FDA Science Forum, at Washington, DC. Lathers, C. M., S. A. Koehler, C. H. Wecht, and P. L. Schraeder. 2003b. Forensic antiepileptic drug levels in 2001 autopsy cases of sudden, unexpected deaths in persons with epilepsy in Allegheny County Pennsylvania. Paper read at the Annual Meeting of American College of Clinical Pharmacology, September, at Orlando, FL.

964 Sudden Death in Epilepsy: Forensic and Clinical Issues Lathers, C. M., W. H. Spivey, L. E. Suter, J. P. Lerner, N. Tumer, and R. M. Levin. 1986. The effect of acute and chronic administration of timolol on cardiac sympathetic neural discharge, arrhythmia, and beta adrenergic receptor density associated with coronary occlusion in the cat. Life Sci 39 (22): 2121–2141. Lathers, C. M., W. H. Spivey, and N. Tumer. 1988. The effect of timolol given ἀve minutes after coronary occlusion on plasma catecholamines. J Clin Pharmacol 28 (4): 289–299. Lathers, C. M., W. H. Spivey, R. M. Levin, and N. Tumer. 1990b. The effect of dilevalol on cardiac autonomic neural discharge, plasma catecholamines, and myocardial beta receptor density associated with coronary occlusion. J Clin Pharmacol 30 (3): 241–253. Lathers, C. M., W. H. Spivey, and R. M. Levin. 1988. The effect of chronic timolol in an animal model for myocardial infarction. J Clin Pharmacol 28 (8): 736–745. Lathers, C. M. 1994. Presidential Address. Paper read at the American College of Clinical Pharmacology (ACCP) Annual Meeting, December. Lathers, C. M. 1995a. Leadership needs nurturing environment. Times Union, 1995/10/22. Lathers, C. M. 1995b. Presidential Address, Albany College of Pharmacy. Paper read at the Profiles in Leadership Program, September 12, at Albany, NY. Lathers, C. M. 2010. Chapter 25: Sudden death: Animal models to study nervous system sites of action for disease and pharmacological intervention. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton, FL: CRC Press. Lathers, C. M., and P. L. Schraeder. 2009. Verbal autopsies. Epilepsy Behav 14: 573–576. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2010. Chapter 4. Unanswered questions: SUDEP studies needed. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton, FL: CRC Press. Mason, S. T., and M. E. Corcoran. 1979. Catecholamines and convulsions. Brain Res 170 (3): 497–507. Moran, N. F., K. Poole, G. Bell, J. Solomon, S. Kendall, M. McCarthy, D. McCormick, L. Nashef, J. Sander, and S. D. Shorvon. 2004. Epilepsy in the United Kingdom: Seizure frequency and severity, anti-epileptic drug utilization and impact on life in 1652 people with epilepsy. Seizure 13 (6): 425–433. National Institute of Neurological and Communicative Disorders and Stroke (NINCDS). 2008. Workshop on Sudden Unexpected Death in Epilepsy, November 12–14, at NINCDS, Rockville, MD. Oishi, R., N. Suenaga, and T. Fukuda. 1979. Possible involvement of brainstem norepinephrine in pentylenetetrazol convulsions in rats. Pharmacol Biochem Behav 10 (1): 57–61. O’Rourke, D. K., and C. M. Lathers. 1990. Chapter 15. Interspike interval histogram characterization of synchronized cardiac sympathetic neural discharge and epileptogenic activity in the electrocorticogram of the cat. In Epilepsy and Sudden Death., ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Rastogi, S. K., and M. K. Ticku. 1986. Anticonvulsant proἀle of drugs which facilitate GABAergic transmission on convulsions mediated by a GABAergic mechanism. Neuropharmacology 25 (2): 175–185. Rugg-Gunn, F. J., R. J. Simister, M. Squirrell, D. R. Holdright, and J. S. Duncan. 2004. Cardiac arrhythmias in focal epilepsy: A prospective long-term study. Lancet 364 (9452): 2212–2219. Ryvlin, P., T. Tomson, and A. Montavont. 2009. Excess mortality and sudden unexpected death in epilepsy. Presse Med 38 (6): 905–910. Schott, G. D., A. A. McLeod, and D. E. Jewitt. 1977. Cardiac arrhythmias that masquerade as epilepsy. Br Med J 1 (6074): 1454–1457. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32 (12): 1371–1382. Schraeder, P. L., K. Delin, R. L. McClelland, and E. L. So. 2006. Coroner and medical examiner documentation of sudden unexplained deaths in epilepsy. Epilepsy Res 68 (2): 137–143.

Epilepsy and SUDEP

965

Schraeder, P. L., K. Delin, R. L. McClelland, and E. L. So. 2009. A nationwide survey of the extent of autopsy in sudden unexplained death in epilepsy. Am J Forensic Med Pathol 30 (2): 123–126. Schraeder, P. L., R. Pontzer, and T. R. Engel. 1983. A case of being scared to death. Arch Intern Med 143 (9): 1793–1794. Schwartz, R. D. 1988. The GABAA receptor-gated ion channel: Biochemical and pharmacological studies of structure and function. Biochem Pharmacol 37 (18): 3369–3375. Schwartz, R. D., X. Yu, M. R. Katzman, D. M. Hayden-Hixson, and J. M. Perry. 1995. Diazepam, given postischemia, protects selectively vulnerable neurons in the rat hippocampus and striatum. J Neurosci 15 (1 Pt 2): 529–539. Schwartz, R. D., and C. M. Lathers. 1990. GABA neurotransmission, epileptogenic activity, and cardiac arrhythmias (Chapter 17). In Epilepsy and Sudden Death, ed. C. M. Lathers and P. L. Schraeder. New York, NY: Marcel Dekker. Scorza, F. A., R. M. Arida, and E. A. Cavalheiro. 2008. Preventive measures for sudden cardiac death in epilepsy beyond therapies. Epilepsy Behav 13 (1): 263– 264; author reply 265–269. So, E. L. 2008. What is known about the mechanisms underlying SUDEP? Epilepsia 49 (Suppl 9): 93–98. So, E. L., J. Bainbridge, J. R. Buchhalter, J. Donalty, E. J. Donner, A. Finucane, N. M. Graves, et al. 2009. Report of the American Epilepsy Society and the Epilepsy Foundation joint task force on sudden unexplained death in epilepsy. Epilepsia 50 (4): 917–922. So, E. L., M. C. Sam, and T. L. Lagerlund. 2000. Postictal central apnea as a cause of SUDEP: Evidence from near-SUDEP incident. Epilepsia 41 (11): 1494–1497. Stauffer, A. Z., J. Dodd-O, and C. M. Lathers. 1989. The relationship of the lock-step phenomenon and precipitous changes in mean arterial blood pressure. Electroencephalogr Clin Neurophysiol 72 (4): 340–345. Stauffer, A. Z., J. Dodd-O, and C. M. Lathers. 1990. Chapter 14. Relationship of the lockstep phenomenon and precipitous changes in blood pressure. In Epilepsy and Sudden Death. New York, NY: Marcel Dekker. Suter, L. E., and C. M. Lathers. 1984. Modulation of presynaptic gamma aminobutyric acid release by prostaglandin E2: Explanation for epileptogenic activity and dysfunction in autonomic cardiac neural discharge leading to arrhythmias? Med Hypotheses 15 (1): 15–30. Tomson, T., A. C. Skold, P. Holmgen, L. Nilsson, and B. Danielsson. 1998. Postmortem changes in blood concentrations of phenytoin and carbamazepine: An experimental study. Ther Drug Monit 20 (3): 309–312. Tupal, S., and C. L. Faingold. 2006. Evidence supporting a role of serotonin in modulation of sudden death induced by seizures in DBA/2 mice. Epilepsia 47 (1): 21–26. Walczak, T. S., I. E. Leppik, M. D’Amelio, J. Rarick, E. So, P. Ahman, K. Ruggles, G. D. Cascino, J. F. Annegers, and W. A. Hauser. 2001. Incidence and risk factors in sudden unexpected death in epilepsy: A prospective cohort study. Neurology 56 (4): 519–525. Wang, J., Y. Chen, K. Li, and L. Hou. 2006. Blockade of inhibitory neurotransmission evoked seizurelike ἀring of cardiac parasympathetic neurons in brainstem slices of newborn rats: Implications for sudden deaths in patients of epilepsy. Epilepsy Res 70 (2–3): 172–183.

SUDEP A Mystery Yet to Be Solved Claire M. Lathers Paul L. Schraeder

61

Contents 61.1 Animal Models for SUDEP 61.2 Clariἀcation: Role of Cardiac vs. Respiratory Mechanisms in Actual Death Events 61.3 Need for Multidisciplinary Clinical and Basic Science Collaboration References

967 968 969 969

Sudden unexpected death in epilepsy (SUDEP), including risk factors for and causes and prevention of, is a mystery yet to be solved. We are certainly further along than a few years ago. However, the following discussion briefly presents areas where work is needed.

61.1â•… Animal Models for SUDEP “Think out of the box” when evaluating any established animal model with a potential for modiἀcation(s) to study the mechanism(s) of SUDEP. Multiple relevant animal models are needed to understand the pathophysiology of SUDEP, hypothesize about effective treatments, develop pilot studies in persons with epilepsy, and conduct conἀrmatory largescale clinical trials (Lathers 2009). The importance of using many different animal models to study SUDEP to glean an insight into the various mechanisms of risks and their contribution to the initiation of the death event is discussed in detail by Lathers and colleagues (Lathers and Levin 2010; Lathers, Schraeder, and Bungo 2010, Chapter 1). Schwartz et al. (1995) concluded that delayed enhancement of GABAergic neurotransmission directly at the site of vulnerability after an ischemic event protects the neurons from death. This ἀnding should be explored, for example, to further study the effect of diazepam on gammaaminobutyric acid (GABA)–mediated effects that may prevent ischemia-induced neuronal death and ultimately prevent the worsening of central neuronal communication due to epileptogenic activity. This may eventually contribute a protective central nervous system effect to make an individual less likely to be at risk for SUDEP. There are many more basic science questions to be raised and answered. Wang et al. (2006) provided supporting data for the lockstep phenomenon ἀnding of Lathers and colleagues (Lathers and Schraeder 1982; Schraeder and Lathers 1983; Lathers and Schraeder 2010, Chapter 28). Blockade of GABAergic and glycinergic receptors in medulla slices of newborn rats evoked intermittent seizure-like ἀring of cardiac parasympathetic neurons, suggesting that seizure-like pattern of ἀring during an epileptic attack may cause neurogenic ictal bradyarrhythmias, cardiac asystole, or even sudden death in persons with epilepsy. The study of Wang et€al. 967

968 Sudden Death in Epilepsy: Forensic and Clinical Issues

(2006) must be expanded to examine antiepileptic or other categories of drugs and their effect on the intermittent seizure-like ἀring of cardiac parasympathetic neurons that may cause neurogenic ictal bradyarrhythmias, cardiac asystole, or sudden death. So (2008) emphasized the signiἀcance of using audiogenic seizure mice to study postictal respiratory arrest. Postictal respiratory arrest was induced by serotonin receptor inhibition and prevented by selective serotonin reuptake inhibitor drugs. The role of serotonin in SUDEP must be examined in future animal studies (Tupal and Faingold 2006; Faingold, Tupal, and Uteshev 2010; Paterson 2010).

61.2â•…Clarification: Role of Cardiac vs. Respiratory Mechanisms in Actual Death Events The ongoing discussion of the differential role of cardiac vs. respiratory mechanisms as operant in SUDEP has to some degree resulted in an either/or discussion (Simon et al. 1982; Simon 1993; Simon et al. 1988; Nashef et al. 1996; Nashef and Ryvlin 2009; Lathers, Schraeder, and Weiner 1987; Lathers, Schraeder, and Bungo 2008; Schraeder and Lathers 1983; Leestma et al. 1989; Leung, Kwan, and Elger 2006; Rocamora et al. 2003). The fallacy of such a dichotomous approach is that it diminishes the obvious likelihood that an interactional relationship may exist at several levels. We know that epilepsy-related disturbances of respiratory regulation and that central nervous system–mediated apnea can both occur (So, Sam, and Lagerlund 2000; Johnston et al. 1997; Langan, Nashef, and Sander 2000; Lee et al. 1999; Sethi and Chhabra 2008). In addition, neurogenic pulmonary edema is also seen in association with seizures and is a common ἀnding on autopsy of SUDEP victims (Leestma et al. 1989; Terrence, Rao, and Perper 1981; Sedy et al. 2007). We also know that seizures can be associated with the occurrence of tachyarrhythmias, bradyarrhythmias/asystole, and cardiac conduction changes (Aurlien et al. 2009; Brugada, Brugada, and Brugada 2003; Drake, Reider, and Kay 1993; Espinosa et al. 2009; Ieda et al. 2008; Nei et al. 2004; Opeskin, Thomas, and Berkovic 2000). Therefore, it is self-evident that the occurrence of seizure-related apnea/pulmonary edema/hypoxia/acidosis may well have the potential to set the stage for acute cardiac rhythm disturbance. Thus, there is the possibility of a combination of ingredients making up a stew of possible mechanisms for SUDEP. If such were to be the case, it would be difficult to say which single putative mechanism or combination of mechanisms predominated in any given victim of SUDEP. We also know that there are inherited disturbances of the sodium channel (Brugada, Brugada, and Brugada 2003; Aurlien et al. 2009; Kornick et al. 2003; Herreros 2010, Chapter 19; Lathers, Schraeder, and Bungo 2010, Chapter 1; Lathers, Schraeder, and Bungo 2010, Chapter 20) that predispose to potentially fatal arrhythmias. Persons with Brugada syndrome are known to have seizure-like activity during episodes of potentially fatal cardiac syncope (Sharma, Ho, and Kantharia 2010; Lathers, Schraeder, and Bungo 2010, Chapter 20), raising the distinct possibility that at least some SUDEP victims may have had an incorrect diagnosis of epilepsy. On the other hand, there has been no effort to investigate persons with epilepsy to determine if there is a subgroup that has a previously unrecognized cardiac predisposition, such a latent Brugada syndrome, to arrhythmia. Likewise, more work needs to be done on investigating the possibility of persons with epilepsy having any pattern or patterns of breathing disturbance, e.g., transient episodes of apnea, during sleep (Hughes and Sato 2010, Chapter 23).

SUDEP

969

The fact that the most cases of SUDEP occur during or associated with sleep implies that the sleep state may incorporate a predisposing factor or factors in association with epilepsy. Nasheff et al. (1998) have found that the sleeping position of SUDEP victims may be a risk for suffocation and advocate that persons with poorly controlled seizures who live in an institutional environment have ongoing monitoring during sleep. The overall approach to what interventions during sleep may be the most helpful preventive measures for the vast majority of potential victims of SUDEP who have relatively infrequent seizures and who live at home or alone is not known. As with sudden infant death syndrome, the possibility that a simple intervention such as sleeping supine rather than prone might be of measurable preventive beneἀt deserves investigation. As one can see from this brief discussion, there is an obvious need for interdisciplinary clinical research among epileptologists, cardiologists, pulmonologists, and sleep specialists. Unraveling the mystery of the mechanisms (the plural is used purposely) of SUDEP will not be an easy task. Developing simple preventive interventions that may at least diminish the occurrence of SUDEP could result from studies that screen persons with epilepsy and family members for potential cardiac sodium channel abnormalities, symptoms of sleep disturbance, and even sleeping position.

61.3â•…Need for Multidisciplinary Clinical and Basic Science Collaboration A worldwide network of professionals must focus on basic scientiἀc research programs as well as clinical and epidemiology studies. Team work among different multidisciplinary professionals in clinical settings and within and among laboratories should address the global issues of SUDEP. The ἀelds of neurology, pharmacology, clinical pharmacology, cardiology, pulmonology, and sleep have much to offer as we work to improve compliance, develop new antiepileptic drugs, and apply different categories of drugs to resolve the mystery of SUDEP. Ambulatory simultaneous electrocardiographic and electroencephalographic telemetry monitoring of patients at risk for sudden death will help identify cardiac vs. epileptogenic triggers to decrease risk of SUDEP.

References Aurlien, D., T. P. Leren, E. Tauboll, and L. Gjerstad. 2009. New SCN5A mutation in a SUDEP victim with idiopathic epilepsy. Seizure 18 (2): 158–160. Brugada, J., R. Brugada, and P. Brugada. 2003. Determinants of sudden cardiac death in individuals with the electrocardiographic pattern of Brugada syndrome and no previous cardiac arrest. Circulation 108 (25): 3092–3096. Drake, M. E., C. R. Reider, and A. Kay. 1993. Electrocardiography in epilepsy patients without cardiac symptoms. Seizure 2 (1): 63–65. Espinosa, P. S., J. W. Lee, U. B. Tedrow, E. B. Bromἀeld, and B. A. Dworetzky. 2009. Sudden unexpected near death in epilepsy: Malignant arrhythmia from a partial seizure. Neurology 72 (19): 1702–1703. Faingold, C. L., S. Tupal, Y. Mhaskar, and V. V. Uteshev. 2010. Chapter 41. DBA mice as models of sudden unexpected death in epilepsy. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. Bungo and J. E. Leestma. Boca Raton, FL: CRC Press.

970 Sudden Death in Epilepsy: Forensic and Clinical Issues Herreros, B. 2010. Cardiac channelopathies and sudden death. In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 19. Boca Raton, FL: CRC Press. Hughes, J. R., and S. Sato. 2010. Sudden death in epilepsy: Relationship to the sleep-wake circadian cycle and fractal physiology. In Sudden Death in Epilepsy: Relationship to the Sleep-Wake Circadian Cycle and Fractal Physiology, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 23. Boca Raton, FL: CRC Press. Ieda, M., K. Kimura, H. Kanazawa, and K. Fukuda. 2008. Regulation of cardiac nerves: A new paradigm in the management of sudden cardiac death? Curr Med Chem 15 (17): 1731–1736. Johnston, S. C., R. Siedenberg, J. K. Min, E. H. Jerome, and K. D. Laxer. 1997. Central apnea and acute cardiac ischemia in a sheep model of epileptic sudden death. Ann Neurol 42 (4): 588–594. Kornick, C. A., M. J. Kilborn, J. Santiago-Palma, G. Schulman, H. T. Thaler, D. L. Keefe, A. N. Katchman et al. 2003. QTc interval prolongation associated with intravenous methadone. Pain 105 (3): 499–506. Langan, Y., L. Nashef, and J. W. Sander. 2000. Sudden unexpected death in epilepsy: A series of witnessed deaths. J Neurol Neurosurg Psychiatry 68 (2): 211–213. Lathers, C. M. 2009. Epilepsy and sudden death: Personal reflections and call for global action. Epilepsy Behav 15 (3): 269–277. Lathers, C. M., P. L. Schraeder, and F. L. Weiner. 1987. Synchronization of cardiac autonomic neural discharge with epileptogenic activity: The lockstep phenomenon. Electroencephalogr Clin Neurophysiol 67 (3): 247–259. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2008. The mystery of sudden death: Mechanisms for risks. Epilepsy Behav 12 (1): 3–24. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2010. Sodium channel dysfunction: Common pathophysiological mechanism associated with sudden death ECG abnormalities in Brugada syndrome and some types of epilepsy. Case histories (Chapter 20). In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton, FL: CRC Press. Lathers, C. M., and P. L. Schraeder. 1982. Autonomic dysfunction in epilepsy: Characterization of autonomic cardiac neural discharge associated with pentylenetetrazol-induced epileptogenic activity. Epilepsia 23 (6): 633–647. Lathers, C. M., P. Schraeder et al. 2010. Animal model for sudden unexpected death in persons with epilepsy (Chapter 28). In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton, FL: CRC Press. Lathers, C. M., P. L. Schraeder, and M. W. Bungo. 2010. Neurocardiologic mechanistic risk factors in sudden unexpected death in epilepsy (Chapter 1). In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton, FL: CRC Press. Lathers, C. M., and R. M. Levin. 2010. Animal model for sudden cardiac death. Sympathetic innervation and myocardial beta receptor densities (Chapter 33). In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma, Chapter 1. Boca Raton, FL: CRC Press. Lee, H. W., S. B. Hong, W. S. Tae, D. W. Seo, and S. E. Kim. 1999. Partial seizures manifesting as apnea only in an adult. Epilepsia 40 (12): 1828–1831. Leestma, J. E., T. Walczak, J. R. Hughes, M. B. Kalelkar, and S. S. Teas. 1989. A prospective study on sudden unexpected death in epilepsy. Ann Neurol 26 (2): 195–203. Leung, H., P. Kwan, and C. E. Elger. 2006. Finding the missing link between ictal bradyarrhythmia, ictal asystole, and sudden unexpected death in epilepsy. Epilepsy Behav 9 (1): 19–30. Nashef, L., F. Walker, P. Allen, J. W. Sander, S. D. Shorvon, and D. R. Fish. 1996. Apnoea and bradycardia during epileptic seizures: Relation to sudden death in epilepsy. J Neurol Neurosurg Psychiatry 60 (3): 297–300.

SUDEP

971

Nashef, L., and P. Ryvlin. 2009. Sudden unexpected death in epilepsy (SUDEP): Update and reflections. Neurol Clin 27 (4): 1063–1074. Nashef, L., S. Garner, J. W. Sander, D. R. Fish, and S. D. Shorvon. 1998. Circumstances of death in sudden death in epilepsy: Interviews of bereaved relatives. J Neurol Neurosurg Psychiatry 64 (3): 349–352. Nei, M., R. T. Ho, B. W. Abou-Khalil, F. W. Drislane, J. Liporace, A. Romeo, and M. R. Sperling. 2004. EEG and ECG in sudden unexplained death in epilepsy. Epilepsia 45 (4): 338–345. Opeskin, K., A. Thomas, and S. F. Berkovic. 2000. Does cardiac conduction pathology contribute to sudden unexpected death in epilepsy? Epilepsy Res 40 (1): 17–24. Paterson, D. S. 2010. Medullary serotonergic abnormalities in sudden infant death syndrome: Implications in SUDEP (Chapter 5). In Sudden Death in Epilepsy: Forensic and Clinical Issues, ed. C. M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton, FL: CRC Press. Rocamora, R., M. Kurthen, L. Lickfett, J. Von Oertzen, and C. E. Elger. 2003. Cardiac asystole in epilepsy: Clinical and neurophysiologic features. Epilepsia 44 (2): 179–185. Schraeder, P. L., and C. M. Lathers. 1983. Cardiac neural discharge and epileptogenic activity in the cat: An animal model for unexplained death. Life Sci 32 (12): 1371–1382. Schwartz, R. D., X. Yu, M. R. Katzman, D. M. Hayden-Hixson, and J. M. Perry. 1995. Diazepam, given postischemia, protects selectively vulnerable neurons in the rat hippocampus and striatum. J Neurosci 15 (1 Pt 2): 529–539. Sedy, J., J. Zicha, J. Kunes, P. Jendelova, and E. Sykova. 2008. Mechanisms of neurogenic pulmonary edema development. Physiol Res 57 (4): 499–506. Sethi, D., and A. Chhabra. 2008. Seizure disorder leading to apnea and bradycardia in a 9-year-old child in immediate postoperative period. Paediatr Anaesth 18 (12): 1211–1212. Sharma, S., T. Ho, and B. K. Kantharia. 2010. Not seizure but syncope (Chapter 21). In Sudden Death in Epilepsy: Forensic and Clinical Issues,€ed C.€M. Lathers, P. L. Schraeder, M. W. Bungo, and J. E. Leestma. Boca Raton: CRC Press. Simon, R. P. 1993. Neurogenic pulmonary edema. Neurol Clin 11 (2): 309–323. Simon, R. P., B. Graham, L. L. Bayne, and T. M. Darragh. 1988. Effect of pulmonary vascular pressure on lung lymph flow following seizures. Chest 93 (2): 386–389. Simon, R. P., L. L. Bayne, R. F. Tranbaugh, and F. R. Lewis. 1982. Elevated pulmonary lymph flow and protein content during status epilepticus in sheep. J Appl Physiol 52 (1): 91–95. So, E. L. 2008. What is known about the mechanisms underlying SUDEP? Epilepsia 49 (Suppl 9): 93–98. So, E. L., M. C. Sam, and T. L. Lagerlund. 2000. Postictal central apnea as a cause of SUDEP: Evidence from near-SUDEP incident. Epilepsia 41 (11): 1494–1497. Terrence, C. F., G. R. Rao, and J. A. Perper. 1981. Neurogenic pulmonary edema in unexpected, unexplained death of epileptic patients. Ann Neurol 9 (5): 458–464. Tupal, S., and C. L. Faingold. 2006. Evidence supporting a role of serotonin in modulation of sudden death induced by seizures in DBA/2 mice. Epilepsia 47 (1): 21–26. Wang, J., Y. Chen, K. Li, and L. Hou. 2006. Blockade of inhibitory neurotransmission evoked seizurelike ἀring of cardiac parasympathetic neurons in brainstem slices of newborn rats: Implications for sudden deaths in patients of epilepsy. Epilepsy Res 70 (2–3): 172–183.

Forensic Evidence and Expert Witnesses Scientific Evidence: Getting It in and Keeping It Out

62

Thomas L. Bohan

Contents 62.1 62.2 62.3 62.4 62.5

Objectives Introduction Common Law vs. Statutory Law Tort Actions Forensic Evidence and Expert Witnesses 62.5.1 Frye vs. Daubert 62.5.2 Federal vs. State, State vs. State 62.5.3 Criminal vs. Civil 62.6 Conclusions and Some Practical Suggestions References

973 974 975 977 978 978 979 979 980 982

62.1â•…Objectives This chapter provides a primer of the American justice system* and then sets out within that framework the current rules (de facto as well as de jure) governing admission into evidence of expert testimony. It concludes with instructions of what to examine when one is facing a hearing to determine whether one will be allowed to testify as an expert witness on a medical subject matter. The rationale is to provide the reader of this book with not only a key to understanding trials he or she hears described second or third hand, but, more importantly, to prepare the reader for the eventuality that he or she will actually participate in a criminal or civil trial as an expert witness or consultant. It is in the latter capacity that he or she will be most focused on (1) helping introduce particular scientiἀc testimony into evidence at trial and (2) helping exclude that expert testimony that appears to be bogus. The U.S. justice system is easy to understand piecewise; it is its totality that can be quite overwhelming. Probably its most difficult aspect arises from the number of overlapping divisions to which it is subject, all of which must be taken into account in considering any particular trial. For openers, there is the usually-but-not-always-obvious distinction between criminal proceedings and civil proceedings. Having determined whether the trial is governed by criminal or civil rules of procedure is not enough. Within the United States,

* Although the major emphasis in this chapter is on U.S. trials, much of it will be applicable across all the countries sharing the common-law tradition (to be defined later in this chapter).

973

974 Sudden Death in Epilepsy: Forensic and Clinical Issues

there are more than 50 state* jurisdictions, with no assurance that any particular criminal or civil rule will be the same in any two of them. Although there is only one federal system, many rules regarding procedure and substance will vary from one to another of the 11 federal circuits.† Furthermore, when state civil actions are removed to federal court (under the latter’s “diversity” jurisdiction), the federal judge must adhere to state law. Nevertheless, there is at least the comfort that all federal courts must comply with Daubert and Kumho Tire (q.v.)‡ in determining whether particular expert testimony should be admitted into evidence. Of course, this is little comfort in the large scheme of things since essentially all trials occur in the state courts. Moreover, the postÂ�-Daubert, post-Kumho Tire track record in the federal courts reveal that some types of proffered expert testimony receive much greater scrutiny than others despite the universal reliability criterion imposed by Kumho Tire on all expert testimony. This irregular landscape will be explored in detail in the second half of this chapter. I will note here, however, that it is medical testimony that seems to be given practically a bye when it comes to the trial judges gatekeeper function, an ironic circumstance given that the seminal Daubert decision resulted in the exclusion of medical expert testimony.

62.2â•…Introduction For a good many years, disputes in the United States have by and large been settled through process of law rather than through dueling or lynching, a circumstance to reflect on when confronted with the whine that ours is a “litigious society.” Because of the widespread use of the legal system to settle disputes and to prosecute those accused of criminal behavior, there is probably no ἀeld of endeavor that does not at one time or another form the subject matter of courtroom proceedings. The practice of medicine is more likely than most ἀelds to be implicated in legal disputes, civil and criminal. It therefore behooves all who engage in, or hope to engage in, a medical career in the United States to familiarize themselves with the nature of civil and criminal trials in the courts of the United States. This chapter seeks to provide that familiarity, which should also serve well those who practice in Canada or, indeed, any English-speaking country. Civil litigation in English-speaking countries is dominated by what some call the “great common law tradition.” This includes the United States outside of Louisiana and Canada outside of Quebec (these exceptions having their legal systems rooted in the Napoleonic Code).§ Even in the United States and the other common-law countries, however, common law is only part of the story. Common law and statutory law work together to determine the operation of our legal system, especially that part of it involved in the resolution of civil disputes. The common law/statutory law distinction carries across all of the divisions * The District of Columbia, Puerto Rico, the U.S. Virgin Islands, etc., have state-like court systems. † The label “circuit” is a vestige from two centuries ago when these courts held trials in all of the federal districts within their circuits, to which they traveled throughout the year, accompanied by the attorneys who were to try the cases. These courts now sit in a fixed location, such as Boston for the First Circuit and New York City for the Second Circuit, and hear appeal from trials that have taken place in federal district courts within their territory. ‡ These cases will be discussed in this chapter. They are the governing decisions of the Supreme Court of the United States regarding expert testimony, issued in 1993 and 1999, respectively. § Further exceptions in each country include legal proceedings before Indian and First Nation courts, respectively.

Forensic Evidence and Expert Witnesses

975

in the American legal terrain, and therefore forms a good starting point for the present discussion.

62.3â•… Common Law vs. Statutory Law For those not schooled in jurisprudence and, perhaps more so for those who are schooled, common law can seem reminiscent of looking for guidance to scripture, the only difference being that, in the secular arena, the scripture is found not in a single tome but rather in tens of thousands of judicial decisions stretching back into ancient times. Supporting the appeal-to-scripture image is the practice of referring to appellate court decisions as being “handed down.” It is in the context of common law that one refers to “precedent.” To see what this means, consider a litigant in state court in Vermont. For the sake of deἀnitiveness, assume this litigant is a plaintiff in a civil suit, a party who brought a claim for damages against someone, the defendant in the action. In all legal actions except the most trivial, there will be a number of narrow issues* contested by the parties leading up to trial and even within the trial itself, issues that will require a speciἀc ruling from the court. In preparing to argue a particular issue, the plaintiff in Vermont looks for a precedent governing the issue. The goal in this quest will be to discover a governing precedent that is “on all fours”† with the issue in question and supports the plaintiff’s side of the argument regarding the issue. If such a precedent is found, the subsequent argument before the court reduces to simply directing the attention of the judge‡ to it. In a Vermont trial court, a governing precedent will be a decision rendered by the Vermont Supreme Court. Occasionally, a careless lawyer will make the serious mistake of citing a helpful decision by the governing appellate court, only to learn later (usually under the most embarrassing circumstances) that the decision had subsequently been overruled. One must not only ἀnd a helpful appellate decision, but also determine that it is still “good law.” Notice the use of the word “law” in this context. Despite strenuous complaints from one political faction and then another about the evil of “judges making law,” this is something that they have been doing for hundreds of years as an inherent and necessary component of legal systems based on common law. Returning to our Vermont plaintiff, it is reasonable to conclude that the hoped-for Vermont Supreme Court decision did not exist. One is rarely so lucky. Nevertheless, this will not be the end of the search, a well-argued decision from any respected appellate court supporting one’s side of an issue being of great use in arguing an issue, although not dispositive. Thus, the Vermont plaintiff looks to appellate-court rulings in other states, perhaps beginning with Vermont’s two companions in northern New England plus Massachusetts. Of course, once there is no hope of ἀnding a governing decision, there is no need to stick to state appellate courts. The federal appellate courts, that is, known as the Circuit Courts of Appeal, are a rich source of verbiage. One would expect the Vermont litigant to look ἀrst to decisions from the Court of Appeals for the Second Circuit, the circuit to which Vermont

* Contrary to creeping popular practice, the word “issue” here means “issue” and not “problem.” † “On all fours” is lawyer talk meaning that all relevant aspects of two situations are identical. ‡ When one speaks of issues to be decided, one is referring to interpretations of the law, a responsibility of the trial judge and not the jury.

976 Sudden Death in Epilepsy: Forensic and Clinical Issues

belongs. If there is no help there, the search widens, encompassing all the appellate courts, state and federal in the country. On rare occasions, the issue of interest may be so subtle or the underlying facts so unusual that no appellate court in the United States will have dealt with it. Obvious but worth stating is the fact that appellate decisions on an issue only exist if the issue has been appealed from a trial court. Only a small minority of trials give rise to appeals at all. The usual circumstance is that a party disappointed by the trial verdict appeals for a new trial based on his or her contention that the trial judge made an error in ruling on an issue raised at trial.* At any event, if there have been no appeals regarding the issue of interest to the Vermont plaintiff, he or she may then look to the small minority of trial decisions that have been published. Rather than argue empty-handed, a lawyer prefers to have something to point to, even if it is just a ruling by a trial judge. It has been observed that much legal work is driven by lawyers’ universal desire never to be the ἀrst person to argue a particular point of view on an issue. Indeed, if the issue is of particular importance, the party in Vermont state court may even look to judicial decisions and opinions from courts in other common-law countries. Although common law plays a small role in criminal matters, it dominates civil litigation. Even in that realm, however, statutory law (“black-letter law”) trumps common law wherever the two conflict. The origin of such conflict arises from legislatures from time to time enacting statutes to counter perceived hardships caused by common law. For example, under common law, you could not bring an action for wrongful death, regardless of how close the decedent was to you, how emotionally devastated you were by his or her death, and no matter how culpable the person who caused the death. Under common law, a claim for personal injury could only be brought by the person directly harmed. Since a dead person could not bring a claim, the cause of action died with the decedent. To remedy this harsh result, all the states have enacted wrongful-death statutes that permit recovery for the wrongful death of a family member, thereby overruling common law. While permitting the lawsuits for wrongful death, these statutes also limit the amount that can be recovered in such actions. Another example where statutes have been enacted to trump the rules of common-law actions lies in suits against the government. Such suits were forbidden across the board by the common-law prohibition of “claims against the sovereign.” It is only within the past 75 years in the United States that state and federal tort claims acts have been enacted to open the door somewhat. These tort claims acts narrowly deἀne the alleged governmental wrongs against a private person for which satisfaction can be sought. Only those speciἀed in the controlling act can form the basis for a lawsuit against the governmental body in question. As with wrongful-death statutes, tort claims acts place caps on the monetary recovery that can be obtained by plaintiffs prevailing in claims against the government. For example, the Maine tort claims act permits suits against the state for negligence in the course of highway repair but does not permit claims for bad highway design per se. Also, it limits successful plaintiffs to judgments not to exceed $300,000. From time to time, state legislatures will enact special legislation to waive the maximum recovery for a particular claim. * Most people have to go to law school before learning that appealing a decision to a higher court is not the same as a child appealing his mother’s opinion to his father, or vice versa. The appeals court does not retry the case but limits its consideration to specific errors by the trial judge alleged by the appellate party to have occurred and skewed the trial.

Forensic Evidence and Expert Witnesses

977

62.4â•… Tort Actions Thus far, one word that had not been deἀned is “tort.” A tort (from the same root as “torture”) is usually characterized as a private wrong; that is, a wrongful act—usually unintentional—that injures a private person. A crime is an intentional wrong against the public, although usually one member of the public bears the brunt of the injury to person or purse. An alleged crime may lead to criminal charges, culminating in a criminal trial in which the government is represented by a prosecutor and the accused is normally represented by a defense attorney. An alleged tort must be dealt with by the injured individual asserting a civil claim against the party alleged to have committed the tort. If a civil trial follows, the two individuals will normally be represented by a plaintiff’s attorney and a defense attorney, respectively. The criminal trial run to its conclusion will result in a verdict of “guilty” or “not guilty.”* A civil trial run to its conclusion will result in a “plaintiff’s verdict” or “defense verdict.”† Most civil actions involve the tort of negligence. The plaintiff asserts (and must prove) (1) that the defendant had a duty to the plaintiff, (2) that the defendant breached that duty through negligence, and (3) that the plaintiff was injured (personally or economically) as a result of that breach. Common law dominates the key components of these actions. Essentially no acts are deἀned as negligent per se, that is by statute. Therefore, the set of circumstances surrounding each alleged negligent act must be examined. Because these circumstances vary so widely from case to case, it is unlikely that the parties will be able to ἀnd a governing precedent. Although few cases go to trial without some issues being argued based on earlier appellate decisions, the ultimate issue of whether the defendant’s actions were negligent is not one of these.‡ The purpose of the trial is for the answer to this key question to be determined by the jury (the usual ἀnder of fact in the United States, but not in any other country that I am familiar with), based on the evidence laid before it. There are both common-law and statutory rules governing the nature of the evidence, factual and opinion (expert), that can be introduced, and they vary from jurisdiction to jurisdiction. For example, in negligence actions based on an alleged hazardous condition (slippery walkway, exposed high-voltage wires, etc.), it would seem reasonable that the plaintiff could introduce evidence that the defendant made post-injury repairs. However, many jurisdictions bar such evidence as being against public policy (because of its perceived tendency to discourage the defendant from remedying a hazard if he thinks it can be used against him). Some states did this through enacting statutes prohibiting the introduction of evidence of “subsequent repair,” whereas others did it through case law, decisions by their appellate courts. The result is the same.

* These verdicts establish the defendant’s status in the eyes of the law. They have no effect on whether the defendant was truly guilty or not guilty. Unfortunately, some judges lose track of this truth, as they fulminate at “failure to show remorse” and impose augmented sentences. † This is pretty basic material. However, I am moved to include it by the frequency with which I read newspaper stories announcing that a defendant in a civil suit has been found “guilty” or “cleared” of charges. ‡ For definitiveness, consider whether a merchant leaving the sidewalk bordering his shop icy for a period of 6 hours was thereby negligent. Does it depend on the prevailing temperature? The state of precipitation? Wind speed?

978 Sudden Death in Epilepsy: Forensic and Clinical Issues

62.5â•… Forensic Evidence and Expert Witnesses Although the previous paragraphs provide an essential background and context for a discussion of forensic evidence in the courtroom, it is nevertheless necessary to do some further parsing before getting into the meat of the matter. The matter is, of course, the rules governing the admission and exclusion of scientiἀc testimony at trial. 62.5.1â•… Frye vs. Daubert The modern era for scientiἀc evidence began in 1923. That was the year that an appellate court for the District of Columbia rendered a decision in a criminal case involving one Mr. Frye (Frye v. United States, 1923). Frye had contended that his defense in the court below had been wrongfully hampered by the trial judge’s refusal to allow into evidence testimony that he had “passed” a lie detector test that tended to support his claim of innocence. The appellate court’s decision affirming Frye’s conviction gave us the Frye standard: for scientiἀc testimony to be admitted into evidence, its profferer must establish that the testimony is based on a theory or technique that has general acceptance within the relevant scientiἀc community. Initially, Frye served as governing precedent only in the courts of the District of Columbia. However, over the subsequent 70 years, the majority of state high courts and federal circuit courts throughout the country adopted this standard (the “general acceptance” standard) as their own for both criminal and civil trials. Then, in 1993, following an appeal from a civil action in federal court, the Supreme Court of the United States ruled (Daubert v. Merrill-Dow Pharmaceuticals, Inc., 1993) that the general acceptance standard was too narrow and should not have been used to exclude the plaintiff’s expert testimony in the trial. However, in removing that particular barrier to admission, it replaced it with what it considered a more flexible one, namely that proffered scientiἀc testimony must be shown to be reliable if it was to be admitted into evidence. In ordering federal trial judges to be more vigilant gatekeepers when it came to the admission of scientiἀc evidence, the Supreme Court offered a number of suggestions to them as possible reliability criteria, questions to be posed. These suggested questions quickly became known as the Daubert factors. They can be summarized as follows. Has the technique or theory underlying the proffered testimony been tested and found to have a low error rate? Are there standards for the application of the underlying technique or theory? Has the underlying technique or theory been published in a peer-reviewed journal? Does the underlying technique or theory have general acceptance within the relevant scientiἀc community? Although the Daubert standard regarding reliability (although not the speciἀc suggestions) initially constituted federal common law, it subsequently became federal black-letter law when it was incorporated into Rule 702 of the Federal Rules of Evidence, the rule pertaining to expert testimony. Since the text of Daubert referred to scientiἀc evidence, there was a short-lived effort by civil and criminal attorneys, as well as prosecutors, to deἀne their expert testimony as nonscientiἀc, in the hope that this would relieve them the obligation to establish its reliability. Engineers were claiming that their work was nonscientiἀc, as were ἀre investigators,

Forensic Evidence and Expert Witnesses

979

medical experts, etc. It was rather a shameful episode, brought to an end by a subsequent Supreme Court decision. Kumho Tire (Kumho Tire v. Carmichael, 1999) holds that all expert testimony proffered under Federal Rule of Evidence 702, that is, all expert testimony offered in federal court,* must meet the reliability requirement of Daubert. This is consistent with the plenary coverage of FRE 702 as revised. 62.5.2â•… Federal vs. State, State vs. State Thus, for trials in any federal court, the profferer of any expert testimony must convince the trial judge that it is sufficiently reliable. The situation in the state trial courts is far more complicated. Although most state courts are governed by either the Frye or Daubert standard† as the result of rulings by the respective high courts, very few of them follow Kumho Tire. This is deἀnitely not a trivial distinction since, not being governed by the all-encompassing principle of the latter decision, the courts can narrowly deἀne the expert testimony that must pass a reliability test a la Daubert or Frye. Some states limit judicial scrutiny to testimony based on “cutting-edge science”; many distinguish between sciencebased testimony and experience-based testimony, requiring only the former to satisfy the reliability requirement. This distinction is obviously a lively concept when determining whether medical testimony is to be subject to preadmission reliability hearings (referred to as “Frye hearings” or “Daubert hearings” depending on the state). An extended discussion of the differences among the 50 states with respect to the admission of expert testimony can be found in Keierleber and Bohan (2005). 62.5.3â•… Criminal vs. Civil Although in neither state nor federal court should the admissibility of expert testimony be determined by whether the trial is civil or criminal, it seems to happen in actual practice. In particular, there appear to be more efforts to exclude testimony through a Frye or Daubert hearing or some other kind of pretrial hearing in civil trials than in criminal cases. One theory advanced to explain this asymmetry is that more resources are available to both sides in civil court than there are in criminal trials. There are also some handwaving arguments about Daubert having arisen from a civil trial rather than a criminal one being the reason its application is primarily limited to civil cases. It would be interesting to examine whether the Frye standard, having arisen in a criminal matter, was used predominantly in criminal trials. A more serious allegation, apparently backed up by facts, is that that expert testimony proffered by the prosecution in criminal cases in both state and federal trials seems to be exposed to a lower degree of scrutiny than does expert testimony proffered by the defense. Although economics may again be the explanation, criminal defendants generally not having the resources with which to mount a challenge that the prosecution does, the reaction to the few times that prosecution evidence has been excluded on Daubert grounds suggests that more is involved. In 2002, a federal district

* Interestingly, these do not include the courts of the District of Columbia, which still adhere to the Frye (general acceptance) standard. † There are some notable exceptions, especially Wisconsin, which purports to require only that the expert testimony be “relevant,” while explicitly eschewing any reliability tests. See, for example, Keierleber and Bohan (2005).

980 Sudden Death in Epilepsy: Forensic and Clinical Issues

trial judge in Pennsylvania took Daubert seriously when confronted with Federal Bureau of Investigation (FBI) ἀngerprint testimony purporting to link a latent partial print to a full-rolled print of the defendant. He opined that, contrary to everyone’s assumptions, there had never been any scientiἀc studies establishing the range over which the identiἀcation of latent prints was valid. “Adversarial testing in court is not .€.€. what the Supreme Court meant when it discussed testing as an admissibility factor.” (United States v. Llera Plaza, 2002.) He had no choice, he said, but to exclude testimony that the prints were made by the same person. A storm of protest from law enforcement and prosecutors’ offices erupted during the months between the original Daubert hearing and the judge’s recantation. The recantation was clothed by the judge announcing his belated realization that ἀngerprint identiἀcation was not a science. Rather, it was a “specialty” and therefore need not satisfy Daubert. Collaterally, he said, he now realized that the FBI ἀngerprint technicians never make mistakes when they match known and unknown prints. The comments he made when the 2004 Brandon Mayἀeld case* burst on the scene, showing his second opinion to be wishful thinking, remain unpublished.

62.6â•… Conclusions and Some Practical Suggestions For medical practitioners active in the legal system as advisors and as expert witnesses, it would appear that the good news and the bad news is the same. Judges are not scrutinizing medical expert testimony to the extent that this chapter’s analysis should suggest would be the case. On the one hand, this permits truly knowledgeable and objective medical experts to escape from the nonsensical rejections of their testimony that their counterparts in ἀelds such as engineering sometimes have to endure. On the other hand, it also permits truly awful medical testimony to be admitted into evidence in the face of solid arguments as to why it should not be. In the hope that trial judges are going to improve their ability to scrutinize medical testimony, I offer the following observations for medical practitioners who may be confronting Daubert and Frye hearings with increasing frequency, and who may more often be advising legal counsel on what questions to ask at such hearings for opposing experts. The key to maximizing one’s chance at succeeding at either function is to have a good idea of what is meant by “reliability” for the particular testimony in question. To take a simple example, viewers of CSI-type television probably have heard many times that a fractured hyoid bone in a cadaver indicates that the decedent was strangled. At a Daubert hearing for a pathologist opining a cause of death based on having observed such a fracture, legitimate questions would include a demand for peer-reviewed literature supporting this theory, including discussions of the frequency with which criminal strangulation and fractured hyoids occur and that of the occasions when a fractured hyoid occurs without criminal strangulation being involved. The questioning could legitimately explore how those frequencies were established; that is, how was it deἀnitely established in the studies giving rise to them as to whether criminal strangulation had taken place? If the basis for establishing for the sake of the studies that criminal strangulation had occurred was nothing more than confession by the suspect, it would be legitimate to attack that as a nonscientiἀc basis. * The FBI, using the full panoply of its “zero error” system, notoriously misidentified a latent print from the Madrid train bombing with an Oregon lawyer, Brandon Mayfield, a recent convert to Islam.

Forensic Evidence and Expert Witnesses

981

Indeed, any time that a pathology is deemed pathognomonic of a particular crime, questions of these types should be put to the proffered witness, and, in the absence of satisfactory (i.e., scientiἀcally valid) answers to them, that witness should not be allowed to testify. In the sense that there may be signiἀcant subsets of medical practitioners who accept the theory that a particular pathology is pathognomonic of a particular event, and yet the type of indicia of reliability just outlined does not exist, the Daubert standard is a much better one than the “general acceptance” Frye standard. For most of the years since Daubert was handed down by the U.S. Supreme Court, there have been databases maintained and available online that have reported cases dealing with expert testimony of all kinds. One present such source of information is the Daubert Tracker™. For those who wish to see how any particular issue has been treated in the past by one or another U.S. jurisdiction, or how a particular expert witness has fared at the hands of the courts, this is an invaluable source. One of its valuable features is that it is not limited to appellate decisions but also contains, although on a hit–and-miss basis, decisions issued by trial courts. I would go so far as to say that no one should attempt to write a paper on expert testimony, scientiἀc or otherwise, without consulting the Daubert Tracker. Although there is a fee for use, membership in some professional organizations such as the American Academy of Forensic Sciences provides a reduced rate. In closing, I turn to the overall theme of this book—sudden unexpected death in persons with epilepsy (SUDEP)—to provide an additional example of the legal context that medical specialists in the neurological sciences might encounter. Death while in custody occurs not infrequently in the United States, leading to the necessity of medical cause-ofdeath testimony at postmortem inquests. Often, disputes arise between the families of the decedent and the persons in charge of or overseeing the incarceration of the decedent. If the decedent had been a person with epilepsy, it can be expected that one argument of the latter group is that the death was just another unexpected death of a person with epilepsy. Adversarial litigation being what it is, it may also be expected that both sides of the argument will present medical testimony. It can also be expected that both sides will demand the exclusion of testimony from the expert medical witness for the other side. Depending on the state in question, these demands will lead to Daubert or Frye hearings, at which efforts will be made to show that the witness being questioned does not have a scientiἀc basis for his or her opinions regarding the likelihood or not of the prisoner having died a natural death. Both experts therefore determine in advance the scientiἀc basis for the opinions they are testifying to. Although at present, in many jurisdictions, they may get by simply by asserting their general expertise, their years in the ἀeld, their clinical experience, etc., this may not be the case. The expert attributing the prisoner’s death to SUDEP should be familiar with the known attributes such as age, general health, and so forth, of known cases of SUDEP. By the same token, the expert who is questioning the claim of natural death should be prepared to discuss the rate of SUDEP in persons possessing the characteristics of the decedent prisoner. If the probability is very small that a person with epilepsy will die during a speciἀc period of time as opposed to the probability that this will occur during the person’s lifetime—approximately 10%—the argument that SUDEP is a reasonable explanation for the prisoner’s death decreases accordingly. As can be seen from the above example and the earlier discussion, the legal system is an imperfect means of determining the truth. To the extent possible, scientiἀc issues should be resolved outside of the adversarial system. One excellent route to accomplish this resolution is the National Academy of Sciences, which, when requested and funded

982 Sudden Death in Epilepsy: Forensic and Clinical Issues

to study any topic will carry out its assignment with a committee broadly based in science and technology. One of many studies carried out by the National Academy of Sciences that has had a powerful impact on our legal system was its examination of forensic DNA. Another that is bound to have a far-reaching effect is the report it issued in early 2009: Forensic Science in the United States: A Path Forward (2009).

References Daubert v. Merrill-Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). Frye v. United States, 293 F. 1013 (D.C. Cir. 1923). Keierleber, J. A., and T. L. Bohan. 2005. Ten years after Daubert: The status of the states, J Forensic Sci€50 (5). Kumho Tire v. Carmichael, 526 U.S. 137 (1999). United States v. Llera Plaza, 188 F. Supp. 2d 549 (E.D. Pa. 2002). Strengthening Forensic Science in the United States: A Path Forward. 2009. Washington, DC: National Academies Press.