Mayo Clinic Cardiology: Concise Textbook

Citation preview

Mayo Clinic Cardiology Concise Textbook THIRD EDITION

Mayo Clinic Cardiology Concise Textbook THIRD EDITION

Editors

Joseph G. Murphy, MD Margaret A. Lloyd, MD Associate Editors

Gregory W. Barsness, MD Arshad Jahangir, MD Garvan C. Kane, MD Lyle J. Olson, MD MAYO CLINIC SCIENTIFIC PRESS AND INFORMA HEALTHCARE USA, INC.

ISBN 0-8493-9057-5 The triple-shield Mayo logo and the words MAYO, MAYO CLINIC, and MAYO CLINIC SCIENTIFIC PRESS are marks of Mayo Foundation for Medical Education and Research. ©2007 by Mayo Foundation for Medical Education and Research. All rights reserved. This book is protected by copyright. No part of it may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means—electronic, mechanical, photocopying, recording, or otherwise—without the prior written consent of the copyright holder, except for brief quotations embodied in critical articles and reviews. Inquiries should be addressed to Scientific Publications, Plummer 10, Mayo Clinic, 200 First Street SW, Rochester, MN 55905. For order inquiries, contact Taylor & Francis Group, 6000 Broken Sound Parkway NW, Suite #300, Boca Raton, FL 33487. www.taylorandfrancis.com Catalog record is available from the Library of Congress. Care has been taken to confirm the accuracy of the information presented and to describe generally accepted practices. However, the authors, editors, and publisher are not responsible for errors or omissions or for any consequences from application of the information in this book and make no warranty, express or implied, with respect to the contents of the publication. This book should not be relied on apart from the advice of a qualified health care provider. The authors, editors, and publisher have exerted efforts to ensure that drug selection and dosage set forth in this text are in accordance with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new or infrequently employed drug. Some drugs and medical devices presented in this publication have Food and Drug Administration (FDA) clearance for limited use in restricted research settings. It is the responsibility of the health care providers to ascertain the FDA status of each drug or device planned for use in their clinical practice. Printed in Canada 10 9 8 7 6 5 4 3 2

DEDICATION This book is dedicated to my parents, my wife Marian, without whose support and encouragement this textbook would not have been possible, and my children Owen, Sinéad, and Aidan, as well as Tornados, Spartans, Pink Panthers, and Tommies everywhere. Joseph G. Murphy, MD

For my parents, who taught me to love books. Booksellers everywhere, thank you as well. Margaret A. Lloyd, MD

v

FOREWORD

It is a distinct honor and pleasure to write this foreword for the third edition of Mayo Clinic Cardiology: Concise Textbook. I have had the pleasure of working on the staff of Mayo Clinic’s Division of Cardiovascular Diseases for the past 25 years. Although the diagnosis and treatment of cardiovascular diseases and the day-to-day practice of medicine have changed greatly during that time, the Mayo Clinic tradition of clinical excellence in cardiovascular disease has not. The unique strength of the Division is its breadth of clinical expertise across the areas of acute coronary care, electrophysiology, intervention, adult congenital heart disease, valvular heart disease, vascular disease, heart failure, and others. This expertise covers both common conditions in the practice of cardiovascular disease and those that are very uncommon, even in major tertiary referral centers. The breadth of that expertise is reflected in the range of topics covered in this book. The common conditions include STsegment elevation myocardial infarction, for which Mayo Clinic conducted one of the first clinical trials comparing thrombolytic therapy with acute angioplasty, and chronic mitral insufficiency, to which Mayo Clinic investigators have made multiple major contributions to both diagnosis and the timing and benefit of mitral valve repair. The uncommon conditions include adult congenital heart disease, hypertrophic cardiomyopathy, and pericardial disease, on which the size of our practice has permitted a few of my colleagues to focus their expertise. This book began as an outgrowth of the syllabus for the Mayo Cardiovascular Review Course for Cardiology Boards and Recertification. This highly successful course attracts an annual attendance of more than 700, including cardiology fellows preparing for their initial boards, practicing cardiologists preparing for recertification, and experienced clinicians who simply want to ensure that they are up-to-date on the latest cardiovascular science and care. Readers from any one of these broad categories will find this book very useful. Both the education of cardiology fellows and the practice of cardiovascular medicine are increasingly subject to time constraints. Our fellows complain that 3 or 4 years is simply inadequate to master the rapidly expanding scope of cardiovascular science and practice. Practicing physicians find that their working day grows ever longer, leaving less time for continuing medical education. The strength of this book is its concise presentation of the existing state of cardiovascular practice, as emphasized by its subtitle. There is a growing crisis in the health care system, focused on rapid increases in health care costs and evidence of suboptimal quality. The practice of cardiovascular medicine will be under increasing pressure to shift from the more-care-is-better paradigm that dominated in the past to a focus on improving quality and efficiency. The Dartmouth Atlas of Health Care identified the Medicare referral region centered on Rochester, Minnesota, as a “high-quality, low-cost” region. The principles underlying that efficiency are evident throughout this text. It is hoped that it will assist the reader in his or her personal quest to improve the quality of cardiovascular care in clinical practice.

Raymond J. Gibbons, MD Consultant, Division of Cardiovascular Diseases Mayo Clinic Arthur M. & Gladys D. Gray Professor of Medicine Mayo Clinic College of Medicine Rochester, Minnesota

vii

PREFACE The cover art of the “iceberg” heart is meant to symbolize the significant extent of occult cardiovascular disease in our society and the ruthless “icy” nature of cardiovascular death that curses the sea of humanity. It has been a great honor to oversee the publication of this, the third, edition of Mayo Clinic Cardiology: Concise Textbook (formerly titled Mayo Clinic Cardiology Review). Large textbooks are never the work of one or two individuals but rather the product of a team of dedicated professionals, as has been the case for this book. This textbook from a single institution was written by a diverse faculty of more than 100 cardiologists from more than 17 countries. This textbook is primarily a teaching and learning textbook of cardiology rather than a reference textbook. In response to welcome feedback from readers of our two previous textbook editions, we have maintained a relatively large typeface to make the textbook easily readable and have avoided the temptation to reduce the font size to increase content. Newer electronic search modalities have made textbook references less timely and we have deleted most chapter references and all multiple-choice questions to save space. This textbook is designed to present the field of cardiology in a reader-friendly format that can be read in about 12 months. Many small cardiology textbooks are bare-bones compilations of facts that do not explain the fundamental concepts of cardiovascular disease, and many large cardiology textbooks are voluminous and describe cardiology in great detail. Mayo Clinic Cardiology: Concise Textbook is designed to be a bridge between these approaches. We sought to present a solid framework of ideas with sufficient depth to make the matter interesting yet concise, aimed specifically toward fellows in training or practicing clinicians wanting to update their knowledge. The book contains 1,400 figures, 483 of which are color photographs to supplement the text. Teaching points and clinical pearls have been added to make the textbook come alive and challenge the reader. The concept for this textbook originated from the first syllabus for the Mayo Cardiovascular Review Course, a function the textbook continues to fulfill. The impetus to produce this textbook owes much to the encouragement of Rick Nishimura, MD, and Steve Ommen, MD, the directors of the Mayo Cardiovascular Review Course now in its 11th year. This third edition is a complete revision of all previous chapters of the textbook and has been expanded at the suggestion of cardiology fellows to now include 40 new chapters, including newer aspects of electrophysiology, interventional cardiology, noninvasive imaging, and randomized clinical trials. The text is intended primarily for cardiology fellows studying for cardiology board certification and practicing cardiologists studying for board recertification. It will also be useful for physicians studying for examinations of the Royal Colleges of Physicians, anesthesiologists, critical care physicians, internists and general physicians with a special interest in cardiology, and coronary care and critical care nurses. We thank all our colleagues in the Mayo Clinic Division of Cardiovascular Diseases at Rochester, Arizona, and Jacksonville who generously contributed to this work. We also thank William D. Edwards, MD, for permission to use slides from the Mayo Clinic cardiology pathologic image database. LeAnn Stee and Randall J. Fritz, DVM, at Mayo Clinic, contributed enormously through their editorial guidance. Sandy Beberman at Informa Healthcare patiently guided this project through countless tribulations. We thank both Mayo Clinic and the Informa Healthcare production teams: at Mayo—Roberta Schwartz (production editor), Sharon Wadleigh (scientific publications specialist), Jane Craig and Virginia Dunt (editorial assistants), Kenna Atherton and John Hedlund (proofreaders), Karen Barrie (art director), Jonathan Goebel (graphic designer) and Charlene Wibben (Continuing Medical Education); at Informa Healthcare—Suzanne Lassandro (project editor), and Rick Beardsley (production and manufacturing). We specifically acknowledge colleagues from outside North America who contributed many ideas to this book and who translated previous editions of the book into several foreign languages. We have included a short SI conversion table for common laboratory values to aid their reading of the book. We would appreciate comments from our readers about how we might improve this textbook or, specifically, about any errors that you find. Joseph G. Murphy, MD Consultant, Division of Cardiovascular Diseases, and Chair, Section of Scientific Publications, Mayo Clinic Professor of Medicine Mayo Clinic College of Medicine Rochester, Minnesota [email protected]

Margaret A. Lloyd, MD Consultant, Division of Cardiovascular Diseases, Mayo Clinic Assistant Professor of Medicine Mayo Clinic College of Medicine Rochester, Minnesota

ix

SI UNITS AND ALTERNATIVE SCIENTIFIC NAMES SI UNITS Cholesterol (Total Cholesterol, LDL Cholesterol, HDL Cholesterol) 200 mg/dL = 5.2 mmol/L 160 mg/dL = 4.2 mmol/L 130 mg/dL = 3.4 mmol/L 100 mg/dL = 2.6 mmol/L 70 mg/dL = 1.8 mmol/L 40 mg/dL = 1.0 mmol/L

Triglycerides 100 mg/dL = 1.1 mmol/L 200 mg/dL = 2.2 mmol/L

Glucose 100 mg/dL = 5.5 mmol/L 200 mg/dL = 11.0 mmol/L

Creatinine 1 mg/dL = 88.4 μmol/L 2 mg/dL = 177 μmol/L 3 mg/dL = 265 μmol/L

ALTERNATIVE SCIENTIFIC NAMES Epinephrine = adrenaline Norepinephrine = noradrenaline Isoproterenol = isoprenaline

xi

CONTRIBUTORS Jerome F. Breen, MD Consultant, Department of Radiology* Assistant Professor of Radiology† John F. Bresnahan, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Frank V. Brozovich, MD, PhD Senior Associate Consultant, Division of Cardiovascular Diseases and Department of Physiology and Biomedical Engineering* Professor of Medicine and of Physiology† T. Jared Bunch, MD Fellow in Cardiovascular Diseases and Assistant Professor of Medicine† John C. Burnett, Jr, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine and of Physiology† Mark J. Callahan, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Yong-Mei Cha, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Krishnaswamy Chandrasekaran, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Panithaya Chareonthaitawee, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Frank C. Chen, MD Fellow in Cardiovascular Diseases† Horng H. Chen, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Stuart D. Christenson, MD Senior Associate Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Alfredo L. Clavell, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Heidi M. Connolly, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine†

Michael J. Ackerman, MD, PhD Consultant, Divisions of Cardiovascular Diseases and Pediatric Cardiology and Department of Molecular Pharmacology and Experimental Therapeutics* Associate Professor of Medicine, Pediatrics, and Pharmacology† Thomas G. Allison, PhD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Naser M. Ammash, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Nandan S. Anavekar, MB, BCh Chief Medical Resident and Instructor in Medicine† Christopher P. Appleton, MD Consultant, Division of Cardiovascular Diseases‡ Professor of Medicine† Samuel J. Asirvatham, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† John W. Askew III, MD Fellow in Nuclear Cardiology† Luciano Babuin, MD Research Collaborator, Mayo School of Graduate Medical Education† Gregory W. Barsness, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Malcolm R. Bell, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Patricia J. M. Best, MD Senior Associate Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Joseph L. Blackshear, MD Consultant, Division of Cardiovascular Diseases§ Professor of Medicine† David J. Bradley, MD, PhD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Peter A. Brady, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† *Mayo Clinic, Rochester, Minnesota. †Mayo Clinic College of Medicine, Rochester, Minnesota. ‡Mayo Clinic, Scottsdale, Arizona. §Mayo Clinic, Jacksonville, Florida.

xiii

Sharonne N. Hayes, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Anthony A. Hilliard, MD Fellow in Cardiovascular Diseases† Michael J. Hogan, MD, MBA Consultant, Division of Regional and International Medicine‡ Assistant Professor of Medicine† David R. Holmes, Jr, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Allan S. Jaffe, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Arshad Jahangir, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Traci L. Jurrens, MD Fellow in Cardiovascular Diseases† Ravi Kanagala, MD Senior Associate Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Garvan C. Kane, MD Fellow in Cardiovascular Diseases and Instructor in Medicine† Birgit Kantor, MD, PhD Senior Associate Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Tomas Kara, MD, PhD Research Fellow in Hypertension and Assistant Professor of Medicine† Bijoy K. Khandheria, MD Chair, Division of Cardiovascular Diseases‡ Professor of Medicine† Stephen L. Kopecky, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Iftikhar J. Kullo, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Sudhir S. Kushwaha, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† André C. Lapeyre III, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Hon-Chi Lee, MD, PhD Consultant, Division of Cardiovascular Diseases* Professor of Medicine†

Leslie T. Cooper, Jr, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Richard C. Daly, MD Consultant, Division of Cardiovascular Surgery* Associate Professor of Surgery† Brooks S. Edwards, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Robert P. Frantz, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Paul A. Friedman, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Robert L. Frye, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Apoor S. Gami, MD Fellow in Cardiovascular Diseases and Assistant Professor of Medicine† Gerald T. Gau, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Thomas C. Gerber, MD, PhD Consultant, Division of Cardiovascular Diseases and Department of Radiology§ Associate Professor of Medicine and of Radiology† Bernard J. Gersh, MB, ChB, DPhil Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Jason M. Golbin, DO Fellow in Thoracic Diseases and Critical Care Medicine† Martha A. Grogan, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Richard J. Gumina, MD Senior Associate Consultant, Division of Cardiovascular Diseases* Stephen C. Hammill, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† David L. Hayes, MD Chair, Division of Cardiovascular Diseases* Professor of Medicine†

*Mayo Clinic, Rochester, Minnesota. †Mayo Clinic College of Medicine, Rochester, Minnesota. ‡Mayo Clinic, Scottsdale, Arizona. §Mayo Clinic, Jacksonville, Florida.

xiv

Rick A. Nishimura, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Jae K. Oh, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Lyle J. Olson, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Steve Ommen, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Oyere K. Onuma, BS Research Trainee, Division of Cardiovascular Diseases* Thomas A. Orszulak, MD Consultant, Division of Cardiovascular Surgery* Professor of Surgery† Michael J. Osborn, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Narith N. Ou, PharmD Pharmacist* Lance J. Oyen, PharmD Pharmacist* Assistant Professor of Pharmacy† Douglas L. Packer, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† John G. Park, MD Consultant, Division of Pulmonary and Critical Care Medicine* Assistant Professor of Medicine† Robin Patel, MD Consultant, Division of Infectious Diseases* Associate Professor of Microbiology and Professor of Medicine† Patricia A. Pellikka, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Sabrina D. Phillips, MD Senior Associate Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Co-burn J. Porter, MD Consultant, Division of Pediatric Cardiology* Professor of Pediatrics† Udaya B. S. Prakash, MD Consultant, Division of Pulmonary and Critical Care Medicine* Professor of Medicine,†

Amir Lerman, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Margaret A. Lloyd, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Francisco Lopez-Jimenez, MD, MS Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Verghese Mathew, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Robert D. McBane, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Marian T. McEvoy, MD Consultant, Division of Dermatology* Associate Professor of Dermatology† Michael D. McGoon, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Shaji C. Menon, MD Fellow in Pediatric Cardiology† Fletcher A. Miller, Jr, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Todd D. Miller, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Wayne L. Miller, MD, PhD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Andrew G. Moore, MD Consultant, Division of Cardiovascular Diseases* Instructor in Medicine† Thomas M. Munger, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Joseph G. Murphy, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Ajay Nehra, MD Consultant, Department of Urology* Professor of Urology†

*Mayo Clinic, Rochester, Minnesota. †Mayo Clinic College of Medicine, Rochester, Minnesota. ‡Mayo Clinic, Scottsdale, Arizona. §Mayo Clinic, Jacksonville, Florida.

xv

James M. Steckelberg, MD Chair, Division of Infectious Diseases* Professor of Medicine† Thoralf M. Sundt III, MD Consultant, Division of Cardiovascular Surgery* Professor of Surgery† Imran S. Syed, MD Fellow in Cardiovascular Diseases† Deepak R. Talreja, MD Fellow in Cardiovascular Diseases and Instructor in Medicine† Zelalem Temesgen, MD Consultant, Division of Infectious Diseases* Associate Professor of Medicine† Andre Terzic, MD Consultant, Department of Molecular Pharmacology* Professor of Medicine and of Pharmacology† Randal J. Thomas, MD, MS Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Henry H. Ting, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Cindy W. Tom, MD Research Fellow in Cardiovascular Diseases† Laurence C. Torsher, MD Consultant, Division of Anesthesia* Assistant Professor of Anesthesiology† Teresa S. M. Tsang, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Eric M. Walser, MD Senior Associate Consultant, Department of Radiology§ Professor of Radiology† Carole A. Warnes, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Paul W. Wennberg, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Robert Wolk, MD, PhD Research Collaborator in Cardiovascular Diseases* R. Scott Wright, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Waldemar E. Wysokinski, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Leonid V. Zingman, MD Research Associate, Division of Cardiovascular Diseases* Assistant Professor of Medicine and Instructor in Pharmacology†

Abhiram Prasad, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Sarinya Puwanant, MD Research Fellow in Cardiovascular Diseases† Robert F. Rea, MD Consultant, Division of Cardiovascular Diseases* Associate Professor of Medicine† Margaret M. Redfield, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Guy S. Reeder, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Charanjit S. Rihal, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Richard J. Rodeheffer, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Brian P. Shapiro, MD Fellow in Cardiovascular Diseases† Win-Kuang Shen, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Raymond C. Shields, MD Consultant, Division of Cardiovascular Diseases* Instructor in Medicine† Clarence Shub, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Justo Sierra Johnson, MD, MS Research Fellow in Cardiovascular Diseases† Robert D. Simari, MD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Lawrence J. Sinak, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine† Virend K. Somers, MD, PhD Consultant, Division of Cardiovascular Diseases* Professor of Medicine† Peter C. Spittell, MD Consultant, Division of Cardiovascular Diseases* Assistant Professor of Medicine†

*Mayo Clinic, Rochester, Minnesota. †Mayo Clinic College of Medicine, Rochester, Minnesota. ‡Mayo Clinic, Scottsdale, Arizona. §Mayo Clinic, Jacksonville, Florida.

xvi

TABLE OF CONTENTS

FUNDAMENTALS OF CARDIOVASCULAR DISEASE

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

1. Cardiovascular Examination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 Clarence Shub, MD

2. Applied Anatomy of the Heart and Great Vessels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27 Joseph G. Murphy, MD, R. Scott Wright, MD

3. Evidence-Based Medicine and Statistics in Cardiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55 Apoor S. Gami, MD, Charanjit S. Rihal, MD

4. Noncardiac Surgery in Patients With Heart Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .61 Traci L. Jurrens, MD, Clarence Shub, MD

5. Essential Molecular Biology of Cardiovascular Diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .73 Cindy W. Tom, MD, Robert D. Simari, MD

6. Medical Ethics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95 John G. Park, MD

7. Restrictions on Drivers and Aircraft Pilots With Cardiac Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . .103 Stephen L. Kopecky, MD

NONINVASIVE IMAGING

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115

8. Principles of Echocardiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .117 Teresa S. M. Tsang, MD

9. Stress Echocardiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .143 Patricia A. Pellikka, MD

10. Transesophageal Echocardiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .151 Sarinya Puwanant, MD, Lawrence J. Sinak, MD, Krishnaswamy Chandrasekaran, MD

11. Nuclear Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .161 John W. Askew III, MD, Todd D. Miller, MD

12. Positron Emission Tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .173 Panithaya Chareonthaitawee, MD

13. Cardiovascular Computed Tomography and Magnetic Resonance Imaging . . . . . . . . . . . . . . . . . . .185 Thomas C. Gerber, MD, PhD, Eric M. Walser, MD

14. Cardiac Radiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .205 Jerome F. Breen, MD, Mark J. Callahan, MD

15. Atlas of Radiographs of Congenital Heart Defects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .223 Sabrina D. Phillips, MD, Joseph G. Murphy, MD 16. Cardiopulmonary Exercise Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .231 Thomas G. Allison, PhD 17. Stress Test Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .241 Stuart D. Christenson MD

xvii

ELECTROPHYSIOLOGY

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .247

18. Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities . . . . . . . . . . . . . . . . . . .249 Stephen C. Hammill, MD

19. Cardiac Cellular Electrophysiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .295 Hon-Chi Lee, MD, PhD

20. Normal and Abnormal Cardiac Electrophysiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .309 Douglas L. Packer, MD

21. Indications for Electrophysiologic Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .321 Michael J. Osborn, MD

22. Cardiac Channelopathies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .335 T. Jared Bunch, MD, Michael J. Ackerman, MD, PhD

23. Pediatric Arrhythmias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .345 Co-burn J. Porter, MD

24. Atrial Fibrillation: Pathogenesis, Diagnosis, and Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .351 Paul A. Friedman, MD

25. Atrial Fibrillation: Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .363 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.

David J. Bradley, MD, PhD Atrial Flutter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .371 Yong-Mei Cha, MD Supraventricular Tachycardia: Diagnosis and Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .379 Samuel J. Asirvatham, MD Ventricular Tachycardia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .389 Thomas M. Munger, MD Arrhythmias in Congenital Heart Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .405 Peter A. Brady, MD Arrhythmias During Pregnancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .415 Peter A. Brady, MD Heritable Cardiomyopathies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .425 Shaji C. Menon, MD, Steve R. Ommen, MD, Michael J. Ackerman, MD, PhD Syncope: Diagnosis and Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .443 Win-Kuang Shen, MD Pacemakers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .455 David L. Hayes, MD, Margaret A. Lloyd, MD Cardiac Resynchronization Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .467 David L. Hayes, MD Technical Aspects of Implantable Cardioverter-Defibrillators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .473 Robert F. Rea, MD

36. Implantable Cardioverter-Defibrillator Trials and Prevention of Sudden Cardiac Death . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .481 Margaret A. Lloyd, MD, Bernard J. Gersh, MB, ChB, DPhil

37. Sudden Cardiac Death . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .493 Ravi Kanagala, MD

xviii

38. Heart Disease in Athletes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .507 Stephen C. Hammill, MD

39. Atlas of Electrophysiology Tracings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .509 Douglas L. Packer, MD

VALVULAR HEART DISEASE

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .521

40. Valvular Stenosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .523 Rick A. Nishimura, MD

41. Valvular Regurgitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .535 Rick A. Nishimura, MD

42. Rheumatic Heart Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .549 Andrew G. Moore, MD

43. Carcinoid and Drug-Related Heart Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .555 Heidi M. Connolly, MD, Patricia A. Pellikka, MD

44. Prosthetic Heart Valves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .563 Martha A. Grogan, MD, Fletcher A. Miller, Jr, MD 45. Surgery for Cardiac Valve Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .575 Thomas A. Orszulak, MD

AORTA AND PERIPHERAL VASCULAR DISEASE

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .583

46. Peripheral Vascular Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .585 Peter C. Spittell, MD

47. Cerebrovascular Disease and Carotid Stenting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .595 48. 49. 50. 51. 52. 53. 54.

Peter C. Spittell, MD, David R. Holmes, Jr, MD The Aorta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .601 Peter C. Spittell, MD Renovascular Disease and Renal Artery Stenting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .613 Verghese Mathew, MD Pathophysiology of Arterial Thrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .625 Robert D. McBane, MD, Waldemar E. Wysokinski, MD Treatment and Prevention of Arterial Thrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .635 Robert D. McBane, MD, Waldemar E. Wysokinski, MD Venous and Lymphatic Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .655 Raymond C. Shields, MD Vasculitis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .663 Paul W. Wennberg, MD Marfan Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .673 Naser M. Ammash, MD, Heidi M. Connolly, MD

xix

CORONARY ARTERY DISEASE RISK FACTORS

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .685

55. Coronary Heart Disease Epidemiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .687 Thomas G. Allison, PhD

56. Metabolic Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .695 Thomas G. Allison, PhD

57. Pathogenesis of Atherosclerosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .699 Joseph L. Blackshear, MD, Birgit Kantor, MD, PhD

58. Dyslipidemia and Classical Factors for Atherosclerosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .715 59. 60. 61. 62. 63. 64.

Francisco Lopez-Jimenez, MD, MS, Justo Sierra Johnson, MD, MS, Virend K. Somers, MD, PhD, Gerald T. Gau, MD Novel Risk Markers for Atherosclerosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .725 Iftikhar J. Kullo, MD Diabetes Mellitus and Coronary Artery Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .735 Robert L. Frye, MD, David R. Holmes, Jr, MD Hypertension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .741 Michael J. Hogan, MD, MBA Heart Disease in Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .751 Patricia J. M. Best, MD, Sharonne N. Hayes, MD Heart Disease in the Elderly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .761 Imran S. Syed, MD, Joseph G. Murphy, MD, R. Scott Wright, MD Erectile Dysfunction and Heart Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .767 Bijoy K. Khandheria, MD, Ajay Nehra, MD

MYOCARDIAL INFARCTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .771

65. Cardiac Biomarkers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .773 66. 67. 68. 69. 70.

Brian P. Shapiro, MD, Luciano Babuin, MD, Allan S. Jaffe, MD Acute Coronary Syndromes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .781 Anthony A. Hilliard, MD, Stephen L. Kopecky, MD Chronic Stable Angina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .795 Frank C. Chen, MD, Frank V. Brozovich, MD, PhD Right Ventricular Infarction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .807 Richard J. Gumina, MD, R. Scott Wright, MD, Joseph G. Murphy, MD Adjunctive Therapy in Acute Myocardial Infarction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .813 R. Scott Wright, MD, Imran S. Syed, MD, Joseph G. Murphy, MD Complications of Acute Myocardial Infarction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .827 Joseph G. Murphy, MD, John F. Bresnahan, MD, Margaret A. Lloyd, MD, Guy S. Reeder, MD

71. Reperfusion Strategy for ST-Elevation Myocardial Infarction: Fibrinolysis Versus Percutaneous Coronary Intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .843 Henry H. Ting, MD

72. Fibrinolytic Trials in Acute Myocardial Infarction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .855 Patricia J. M. Best, MD, Bernard J. Gersh, MB, ChB, DPhil, Joseph G. Murphy, MD xx

73. Risk Stratification After Myocardial Infarction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .869 Randal J. Thomas, MD, MS

74. Cardiac Rehabilitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .875 Thomas G. Allison, PhD

75. Coronary Artery Bypass Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .883 Thoralf M. Sundt III, MD

DISEASES OF THE HEART, PERICARDIUM, AND PULMONARY CIRCULATION

. . . . . . . . . . . . . . . .891

76. Pericardial Diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .893 Jae K. Oh, MD

77. Pulmonary Embolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .913 Jason M. Golbin, DO, Udaya B. S. Prakash, MD

78. Pulmonary Hypertension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .929 Michael D. McGoon, MD

79. Pregnancy and the Heart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .951 Heidi M. Connolly, MD

80. Adult Congenital Heart Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .965 Carole A. Warnes, MD

81. HIV Infection and the Heart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .977 82. 83. 84. 85. 86. 87. 88.

Joseph G. Murphy, MD, Zelalem Temesgen, MD Infective Endocarditis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .983 Robin Patel, MD, Joseph G. Murphy, MD, James M. Steckelberg, MD Systemic Disease and the Heart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1017 Marian T. McEvoy, MD, Joseph G. Murphy, MD Cardiac Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1027 Joseph G. Murphy, MD, R. Scott Wright, MD Sleep Apnea and Cardiac Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1035 Tomas Kara, MD, PhD, Robert Wolk, MD, PhD, Virend K. Somers, MD, PhD Cardiovascular Trauma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1043 Joseph G. Murphy, MD, R. Scott Wright, MD Acute Brain Injury and the Heart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1049 Nandan S. Anavekar, MB, BCh, Sarinya Puwanant, MD, Krishnaswamy Chandrasekaran, MD Noncardiac Anesthesia in Patients With Cardiovascular Disease . . . . . . . . . . . . . . . . . . . . . . . . . . .1057 Laurence C. Torsher, MD

CARDIOMYOPATHY AND HEART FAILURE

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1069

89. Cardiovascular Reflexes and Hormones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1071 Alfredo L. Clavell, MD, John C. Burnett, Jr, MD 90. Systolic Heart Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1077 Wayne L. Miller, MD, PhD, Lyle J. Olson, MD 91. Diastolic Heart Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1087 Christopher P. Appleton, MD xxi

92. Heart Failure: Diagnosis and Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1101 Richard J. Rodeheffer, MD, Margaret M. Redfield, MD

93. Pharmacologic Therapy of Systolic Ventricular Dysfunction and Heart Failure . . . . . . . . . . . . . . .1113 Richard J. Rodeheffer, MD, Margaret M. Redfield, MD

94. Myocarditis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1125 Leslie T. Cooper, Jr, MD, Oyere K. Onuma, BS

95. Dilated Cardiomyopathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1139 Horng H. Chen, MD

96. Restrictive Cardiomyopathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1145 Sudhir S. Kushwaha, MD

97. Hypertrophic Cardiomyopathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1153 Steve Ommen, MD

98. Right Ventricular Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1167 Robert P. Frantz, MD

99. Congestive Heart Failure: Surgical Therapy and Permanent Mechanical Support . . . . . . . . . . . . .1173 Richard C. Daly, MD, Brooks S. Edwards, MD

100. Cardiac Transplantation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1179 Brooks S. Edwards, MD, Richard C. Daly, MD

CARDIAC PHARMACOLOGY

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1187

101. Principles of Pharmacokinetics and Pharmacodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1189 Arshad Jahangir, MD, Leonid V. Zingman, MD, Andre Terzic, MD, PhD

102. Antiarrhythmic Drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1205 Peter A. Brady, MD

103. Modulators of the Renin-Angiotensin System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1223 Garvan C. Kane, MD, Peter A. Brady, MD

104. Principles of Diuretic Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1231 Garvan C. Kane, MD, Joseph G. Murphy, MD

105. Digoxin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1233 Arshad Jahangir, MD

106. Principles of Inotropic Drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1245 107. 108. 109. 110. 111.

Garvan C. Kane, MD, Joseph G. Murphy, MD, Arshad Jahangir, MD Nitrate Therapies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1249 Garvan C. Kane, MD, Peter A. Brady, MD Calcium Channel Blockers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1253 Arshad Jahangir, MD β-Adrenoceptor Blockers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1265 Arshad Jahangir, MD Antiplatelet Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1283 Garvan C. Kane, MD, Yong-Mei Cha, MD, Joseph G. Murphy, MD Cardiac Drug Adverse Effects and Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1291 Narith N. Ou, PharmD, Lance J. Oyen, PharmD, Arshad Jahangir, MD

xxii

112. Lipid-Lowering Medications and Lipid-Lowering Clinical Trials . . . . . . . . . . . . . . . . . . . . . . . . . . .1309 Joseph G. Murphy, MD, R. Scott Wright, MD

INVASIVE AND INTERVENTIONAL CARDIOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1325 113. Endothelial Dysfunction and Cardiovascular Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1327 Brian P. Shapiro, MD, Amir Lerman, MD

114. Coronary Artery Physiology and Intracoronary Ultrasonography . . . . . . . . . . . . . . . . . . . . . . . . . .1343 Abhiram Prasad, MD

115. Coronary Anatomy and Angiographic Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1357 André C. Lapeyre III, MD

116. Principles of Interventional Cardiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1369 Gregory W. Barsness, MD, Joseph G. Murphy, MD

117. High-Risk Percutaneous Coronary Interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1381 Gregory W. Barsness, MD

118. Invasive Hemodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1393 Rick A. Nishimura, MD

119. Contrast-Induced Nephropathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1407 Patricia J. M. Best, MD, Charanjit S. Rihal, MD

120. Diagnostic Coronary Angiography and Ventriculography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1413 Joseph G. Murphy, MD

121. Catheter Closure of Intracardiac Shunts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1441 Guy S. Reeder, MD

122. Atlas of Hemodynamic Tracings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1451 Deepak R. Talreja, MD, Rick A. Nishimura, MD, Joseph G. Murphy, MD

123. Endomyocardial Biopsy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1481 Joseph G. Murphy, MD, Robert P. Frantz, MD, Leslie T. Cooper, Jr, MD

124. Coronary Stents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1489 Joseph G. Murphy, MD, Gregory W. Barsness, MD

125. Cardiac Emergencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1503 Arshad Jahangir, MD, Joseph G. Murphy, MD 126. Cardiogenic Shock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1515 Malcolm R. Bell, MD

APPENDIX

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1525

Preparing for Cardiology Examinations Joseph G. Murphy, MD, Margaret A. Lloyd, MD

CREDIT LINES

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1541

INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1553

xxiii

SECTION

I

Fundamentals of Cardiovascular Disease

Transected Aorta: Motor Vehicle Accident

1 CARDIOVASCULAR EXAMINATION Clarence Shub, MD

GENERAL APPEARANCE

wall. During mid and late systole, the left ventricle (LV) is diminishing in volume and the apical impulse moves away from the chest wall.Thus, outward precordial apical motion occurring in late systole is abnormal. Remember that point of maximal impulse is not synonymous with apical impulse.

The physical examination, including the general appearance of the patient, is an extremely important component of cardiology examinations. Almost every question has physical examination findings that provide critical clues to the answer in the stem of the question. Important clues to a cardiac diagnosis can be obtained from inspection of the patient (Table 1).

Palpation of the Apex Constrictive pericarditis or tricuspid regurgitation produces a subtle systolic precordial retraction. The apical impulse of LV enlargement is usually widened or diffuse (>3 cm in diameter), can be palpated in two interspaces, and is displaced leftward. A subtle presystolic ventricular rapid filling wave (A wave)— frequently associated with LV hypertrophy—may be better visualized than palpated by observing the motion of the stethoscope applied lightly on the chest wall, with appropriate timing during simultaneous auscultation. Likewise, a palpable A wave can be detected in this manner. The apical impulse of LV hypertrophy without dilatation is sustained and localized but should not be displaced. Causes of a palpable A wave (presystolic impulse) include the following: 1. Aortic stenosis 2. Hypertrophic obstructive cardiomyopathy 3. Systemic hypertension

BLOOD PRESSURE Blood pressure should always be determined in both arms and in the legs if there is any suspicion of coarctation of the aorta. A difference in systolic blood pressure between both arms of more than 10 mm Hg is abnormal (Table 2).

ABNORMALITIES ON PALPATION OF THE PRECORDIUM The patient should be examined in both the supine and the left lateral decubitus position. Examining the apical impulse by the posterior approach with the patient in the sitting position may at times be the best method to appreciate subtle abnormalities of precordial motion. The normal apical impulse occurs during early systole with an outward motion imparted to the chest 3

4

Section I Fundamentals of Cardiovascular Disease

Table 1. Clinical Clues to Specific Cardiac Abnormalities Detectable From the General Examination Condition Marfan syndrome Acromegaly

Turner syndrome

Pickwickian syndrome Friedreich ataxia

Duchenne type muscular dystrophy Ankylosing spondylitis Jaundice

Sickle cell anemia Lentigines (LEOPARD syndrome*)

Hereditary hemorrhagic telangiectasia (Osler-Weber-Rendu disease) Pheochromocytoma

Lupus

Sarcoidosis Tuberous sclerosis Myxedema

Right-to-left intracardiac shunt

Appearance Tall Long extremities Large stature Coarse facial features “Spade” hands Web neck Hypertelorism Short stature Severe obesity Somnolence Lurching gait Hammertoe Pes cavus Pseudohypertrophy of calves Straight back syndrome Stiff (“poker”) spine Yellow skin or sclera

Cutaneous ulcers Painful “crises” Brown skin macules that do not increase with sunlight Small capillary hemangiomas on face or mouth, with or without cyanosis Pale, diaphoretic skin Neurofibromatosis—café-aulait spots Butterfly rash on face Raynaud phenomenon—hands Livedo reticularis Cutaneous nodules Erythema nodosum Angiofibromas (face; adenoma sebaceum) Coarse, dry skin Thinning of lateral eyebrows Hoarseness of voice Cyanosis and clubbing of distal extremities Differential cyanosis and clubbing

Associated cardiac abnormalities Aortic root dilatation Mitral valve prolapse Cardiac hypertrophy

Aortic coarctation Pulmonary stenosis Pulmonary hypertension Hypertrophic cardiomyopathy

Cardiomyopathy Aortic regurgitation Heart block (rare) Right-sided congestive heart failure Prosthetic valve dysfunction (hemolysis) Pulmonary hypertension Secondary cardiomyopathy Hypertrophic obstructive cardiomyopathy Pulmonary stenosis Pulmonary arteriovenous fistula

Catecholamine-induced secondary dilated cardiomyopathy Verrucous endocarditis Myocarditis Pericarditis Secondary cardiomyopathy Heart block Rhabdomyoma Pericardial effusion Left ventricular dysfunction Any of the lesions that cause Eisenmenger syndrome Reversed shunt through patent ductus arteriosus

Chapter 1

Cardiovascular Examination

5

Table 1. (continued) Condition Holt-Oram syndrome Down syndrome

Scleroderma

Rheumatoid arthritis Thoracic bony abnormality Carcinoid syndrome

Appearance

Associated cardiac abnormalities

Rudimentary or absent thumb Mental retardation Simian crease of palm Characteristic facies Tight, shiny skin of fingers with contraction Characteristic taut mouth and facies Typical hand deformity Subcutaneous nodules Pectus excavatum Straight back syndrome Reddish cyanosis of face Periodic flushing

Atrial septal defect Endocardial cushion defect

Pulmonary hypertension Myocardial, pericardial, or endocardial disease Myocardial, pericardial, or endocardial disease (often subclinical) Pseudocardiomegaly Mitral valve prolapse Right-sided cardiac valve stenosis or regurgitation

*LEOPARD syndrome: lentigines, electrocardiographic changes, ocular hypertelorism, pulmonary stenosis, abnormal genitalia, retardation of growth, deafness.

■

■

■

The apical impulse of LV hypertrophy without dilatation is sustained and localized. It should not be displaced but may be accompanied by a palpable presystolic outward movement, the A wave. Outward precordial apical motion occurring in late systole is abnormal. Multiple abnormal outward precordial movements may occur: presystolic, systolic, or late systolic rebound and an A wave in late diastole.

Palpation of the Lower Sternal Area Precordial motion in the lower sternal area usually reflects right ventricular (RV) motion. RV hypertrophy due to systolic overload (such as in pulmonary stenosis) causes a sustained outward lift. Diastolic overload (such as in atrial septal defect [ASD]) causes a vigorous nonsustained motion. In severe mitral regurgitation, the left atrium expands in systole but is limited in its posterior motion by the spine. The RV may then be pushed forward, and the parasternal region is “lifted” indirectly. Significant overlap of sites of maximal pulsation occurs in LV and RV overload states. For example, in RV overload, the abnormal impulse can overlap with

the LV in the apical sternal region (between the apex and the left lower sternal border). An LV apical aneurysm may produce a delayed outward motion and cause a “rocking” motion. Palpation of the Left Upper Sternal Area Abnormal pulsations at the left upper sternal border (pulmonic area) can be due to a dilated pulmonary artery (e.g., poststenotic dilatation in pulmonary valve stenosis, idiopathic dilatation of the pulmonary artery, or increased pulmonary flow related to ASD or pul-

Table 2. Causes of Blood Pressure Discrepancy Between Arms or Between Arms and Legs Arterial occlusion or stenosis of any cause Dissecting aortic aneurysm Coarctation of the aorta Patent ductus arteriosus Supravalvular aortic stenosis Thoracic outlet syndrome

6

Section I Fundamentals of Cardiovascular Disease

monary hypertension). Pulsations of increased blood flow are dynamic and quick, whereas pulsations due to pressure overload cause a sustained impulse. ■

If the apical impulse is not palpable and the patient is hemodynamically unstable, consider cardiac tamponade as the first diagnosis.

Palpation of the Right Upper Sternal Area Abnormal pulsations at the right upper sternal border (aortic area) should suggest an aortic aneurysm. An enlarged left lobe of the liver associated with severe tricuspid regurgitation may be appreciated in the epigastrium, and the epigastric site may be the location of the maximal cardiac impulse in patients with emphysema or an enlarged RV. ■

■

■

■

RV hypertrophy due to systolic overload causes a sustained outward lift. Diastolic overload (as in ASD) causes a vigorous nonsustained motion. In severe mitral regurgitation, the left atrium expands in systole but is limited in its posterior motion by the spine. The RV may then be pushed forward, and the parasternal region is “lifted” indirectly. Significant overlap of sites of maximal pulsation occurs in LV and RV overload states. Pulsations of increased blood flow are dynamic and quick, whereas pulsations due to pressure overload cause a sustained impulse.

Normal jugular venous pressure decreases with inspiration and increases with expiration. Veins that fill at inspiration (Kussmaul sign), however, are a clue to constrictive pericarditis, pulmonary embolism, or RV infarction (Table 5). ■

Jugular veins that fill at inspiration (Kussmaul sign) are a clue to constrictive pericarditis, pulmonary embolism, or RV infarction.

“Hepatojugular” (Abdominojugular) Reflux Sign The neck veins distend with steady (>10 seconds) upper abdominal compression while the patient continues to breathe normally without straining. Straining may cause a false-positive “hepatojugular” reflux sign. The neck veins may collapse or remain distended. Jugular venous pressure that remains increased and then falls abruptly (≥4 cm H2O) indicates an abnormal response. It may occur in LV failure with secondary pulmonary hypertension. In patients with chronic congestive heart failure, a positive hepatojugular reflux sign (with or without increased jugular venous pressure), a third heart sound (S3), and radiographic pulmonary vascular redistribution are independent predictors of increased pulmonary capillary wedge pressure. The

JUGULAR VEINS Abnormal waveforms in the jugular veins reflect abnormal hemodynamics of the right side of the heart. In the presence of normal sinus rhythm, there are two positive or outward moving waves (a and v) and two visible negative or inward moving waves (x and y) (Fig. 1).The x descent is sometimes referred to as the systolic collapse. Ordinarily, the c wave is not readily visible. The a wave can be identified by simultaneous auscultation of the heart and inspection of the jugular veins. The a wave occurs at about the time of the first heart sound (S1). The x descent follows. The v wave, a slower, more undulating wave, occurs near the second heart sound (S2). The y descent follows. The a wave is normally larger than the v wave, and the x descent is more marked than the y descent (Tables 3 and 4).

Fig. 1. Normal jugular venous pulse. The jugular v wave is built up during systole, and its height reflects the rate of filling and the elasticity of the right atrium. Between the bottom of the y descent (y trough) and the beginning of the a wave is the period of relatively slow filling of the “atrioventricle” or diastasis period. The wave built up during diastasis is the h wave. The h wave height also reflects the stiffness of the right atrium. S1, first heart sound; S2, second heart sound.

Chapter 1 Cardiovascular Examination

Table 3. Timing of Jugular Venous Pulse Waves

Table 4. Abnormal Jugular Venous Pulse Waves

a wave—precedes the carotid arterial pulse and is simultaneous with S4, just before S1 x descent—between S1 and S2 v wave—just after S2 y descent—after the v wave in early diastole

Increased a wave 1. Tricuspid stenosis 2. Decreased right ventricular compliance due to right ventricular hypertrophy in severe pulmonary hypertension Pulmonary stenosis Pulmonary vascular disease 3. Severe left ventricular hypertrophy due to pressure by the hypertrophied septum on right ventricular filling (Bernheim effect) Hypertrophic obstructive cardiomyopathy Rapid x descent Cardiac tamponade Increased v wave Tricuspid regurgitation Atrial septal defect Rapid y descent (Friedreich sign) Constrictive pericarditis

abdominojugular maneuver can also be useful for eliciting venous pulsations if they are difficult to visualize. ■

■

A positive “hepatojugular” (abdominojugular) reflux sign may be found in LV failure with secondary pulmonary hypertension. If the jugular veins are engorged but not pulsatile, consider superior vena caval obstruction.

7

ARTERIAL PULSE Abnormalities of the Carotid Pulse Hyperdynamic Carotid Pulse A vigorous, hyperdynamic carotid pulse is consistent with aortic regurgitation. It may also occur in other states of high cardiac output or be caused by the wide pulse pressure associated with atherosclerosis, especially in the elderly. Dicrotic and Bisferiens Pulses A dicrotic carotid pulse occurs in myocardial failure, especially in association with hypotension, decreased cardiac output, and increased peripheral resistance. Dicrotic and bisferious are the Greek and Latin terms, respectively, for twice beating, but in cardiology they are not equivalent. The second impulse occurs in early diastole with the dicrotic pulse and in late systole with the bisferiens pulse. The bisferiens pulse usually occurs in combined aortic regurgitation and aortic stenosis, but occasionally it occurs in pure aortic regurgitation. Aortic Stenosis Pulsus parvus (soft or weak) classically occurs in aortic stenosis but can also result from severe stenosis of any cardiac valve or can occur with low cardiac output of

any cause. Severe aortic stenosis also produces a slowly increasing delayed pulse (pulsus tardus). Because of the effects of aging on the carotid arteries, the typical findings of pulsus parvus and pulsus tardus may be less apparent or absent in the elderly, even with severe degrees of aortic stenosis. Hypertrophic Obstructive Cardiomyopathy In hypertrophic obstructive cardiomyopathy, the ventricular obstruction begins in mid systole, increases as

Table 5. Differentiation of Internal Jugular Vein Pulse and Carotid Pulse Jugular vein pulse Double peak when in sinus rhythm Obliterated by gentle pressure Changes with position and inspiration

Carotid pulse Single peak Unaffected by gentle pressure Unaffected by position or inspiration

8

Section I Fundamentals of Cardiovascular Disease

contraction proceeds, and decreases in late systole. The initial carotid impulse is brisk. The pulse may be bifid as well (Table 6). Inequality of the carotid pulses can be due to carotid atherosclerosis, especially in elderly patients. In a young patient, consider supravalvular aortic stenosis. (The right side then should have the stronger pulse.) Aortic dissection and thoracic outlet syndrome may also produce inequality of arterial pulses. A pulsating cervical mass, usually on the right, may be caused by atherosclerotic “buckling” of the right common carotid artery and give the false impression of a carotid aneurysm. Transmitted Murmurs Transmitted murmurs of aortic origin, most often due to aortic stenosis (less often due to coarctation, patent ductus arteriosus, pulmonary stenosis, and ventricular septal defect), decrease in intensity as the stethoscope ascends the neck, whereas a carotid bruit is usually louder higher in the neck and decreases in intensity as the stethoscope is inched proximally toward the chest. Both conditions may coexist, especially in elderly patients. An abrupt change in the acoustic characteristics (pitch) of the bruit as the stethoscope is inched upward may be a clue to the presence of combined lesions. Pulsus Paradoxus Paradoxical pulse is an exaggeration of the normal (≤10 mm) inspiratory decline in arterial pressure. It occurs classically in cardiac tamponade but occasionally with other restrictive cardiac abnormalities, severe conges-

Table 6. Causes of a Double-Impulse Carotid Arterial Pulse Dicrotic pulse (systolic + diastolic impulse) Cardiomyopathy Left ventricular failure Bisferiens pulse (two systolic impulses) Aortic regurgitation Combined aortic valve stenosis and regurgitation (dominant regurgitation) Bifid pulse (two systolic impulses with intervening pulse collapse) Hypertrophic cardiomyopathy

tive heart failure, pulmonary embolism, or chronic obstructive pulmonary disease (Table 7). Pulsus Alternans Pulsus alternans (alternation of stronger and weaker beats) rarely occurs in healthy subjects and then is transient after a premature ventricular contraction. It usually is associated with severe myocardial failure and is frequently accompanied by an S3, both of which impart an ominous prognosis. Pulsus alternans may be affected by alterations in venous return and may disappear as congestive heart failure progresses. Electrical alternans (alternating variation in the height of the QRS complex) is unrelated to pulsus alternans (Table 8). ■

■

■

A dicrotic carotid pulse occurs in myocardial failure, often in association with hypotension, decreased cardiac output, and increased peripheral resistance. Pulsus parvus (soft or weak) classically occurs in aortic stenosis but can also result from severe stenosis of any cardiac valve or can occur with severely low cardiac output of any cause. Because of the effects of aging on the carotid arteries, the typical findings of pulsus parvus and

Table 7. Causes of Pulsus Paradoxus Constrictive pericarditis Pericardial tamponade Severe emphysema Severe asthma Severe heart failure Pulmonary embolism Morbid obesity

Table 8. Pulsus and Electrical Alternans Pulsus alternans Severe heart failure Electrical alternans Pericardial tamponade Large pericardial effusions

Chapter 1 Cardiovascular Examination

■

■

■

■

pulsus tardus may be less apparent or absent in the elderly, even with severe degrees of aortic stenosis. Inequality of the carotid pulses can be due to carotid atherosclerosis, especially in elderly patients. In a young patient, consider supravalvular aortic stenosis. (The right side then should have the stronger pulse.) Transmitted murmurs of aortic origin, most often due to aortic stenosis (less often due to coarctation, patent ductus arteriosus, pulmonary stenosis, or ventricular septal defect), decrease in intensity as the stethoscope ascends the neck, whereas a carotid bruit is usually louder higher in the neck and decreases in intensity as the stethoscope is inched proximally toward the chest. Paradoxical pulse occurs classically in cardiac tamponade but occasionally with other restrictive cardiac abnormalities, severe congestive heart failure, pulmonary embolism, or chronic obstructive pulmonary disease. Pulsus alternans usually is associated with severe myocardial failure and is frequently accompanied by an S3, both of which impart an ominous prognosis.

Abnormalities of the Femoral Pulse In hypertension, simultaneous palpation of radial and femoral pulses may reveal a delay or relative weakening of the femoral pulses, suggesting aortic coarctation. The finding of a femoral (or carotid) bruit in an adult suggests diffuse atherosclerosis. Fibromuscular dysplasia is less common and occurs in younger patients.

Mitral Valve Disease Mitral stenosis produces a loud S1 if the valve is pliable. When the valve becomes calcified and immobile, the intensity of S1 decreases. The S1 may also be soft in severe aortic regurgitation (related to early closure of the mitral valve) caused by LV filling from the aorta. The Rate of Increase of Systolic Pressure Within the LV A loud S1 can be produced by hypercontractile states, such as fever, exercise, thyrotoxicosis, and pheochromocytoma. Conversely, a soft S1 can occur in LV failure. If S1 seems louder at the lower left sternal border than at the apex (implying a loud T1), suspect ASD or tricuspid stenosis. Atrial fibrillation produces a variable S1 intensity. (The intensity is inversely related to the previous RR cycle length; a longer cycle length produces a softer S1.) A variable S1 intensity during a wide complex, regular tachycardia suggests atrioventricular dissociation and ventricular tachycardia. The marked delay of T1 in Ebstein anomaly is related to the late billowing effect of the deformed (sail-like) anterior leaflet of the tricuspid valve as it closes in systole.Table 9 lists causes of an abnormal S1. ■

If S1 seems to be louder at the base than at the apex, suspect an ejection sound masquerading as S1. If the S1 is louder at the lower left sternal border than at the apex (implying a loud T1), suspect ASD or tricuspid stenosis.

HEART SOUNDS Table 9. Abnormalities of S1 and Their Causes

First Heart Sound Only the mitral (M1) and tricuspid (T1) components of S1 are normally audible. M1 occurs before T1 and is the loudest component. Wide splitting of S 1 occurs with right bundle branch block and Ebstein anomaly. Factors Influencing the Intensity of S1 PR Interval The PR interval varies inversely with the loudness of S1—with a long PR interval, the S1 is soft; conversely, with a short PR interval, the S1 is loud.

9

Loud S1 Short PR interval Mitral stenosis Left atrial myxoma Hypercontractile states Soft S1 Long PR interval Depressed left ventricular function Early closure of mitral valve in acute severe aortic incompetence Ruptured mitral valve leaflet or chordae Left bundle branch block

10

■

■

Section I Fundamentals of Cardiovascular Disease

A variable S1 intensity during a wide complex, regular tachycardia suggests atrioventricular dissociation and ventricular tachycardia. The marked delay of T1 in Ebstein anomaly is related to the late billowing effect of the deformed (saillike) anterior leaflet of the tricuspid valve as it closes in systole.

change. The pulmonary click is best heard along the upper left sternal border, but if it is loud enough or if the RV is markedly dilated, it may be heard throughout the precordium. The aortic click radiates to the aortic area and the apex and does not change with respiration.The causes of ejection clicks are listed in Table 10. ■

Systolic Ejection Clicks (or Sounds) The ejection click (sound) follows S1 closely and can be confused with a widely split S1 or, occasionally, with an early nonejection click. Clicks can originate from the left or right side of the heart. The three possible mechanisms for production of the clicks are as follows: 1. Intrinsic abnormality of the aortic or pulmonary valve, such as congenital bicuspid aortic valve 2. Pulsatile distention of a dilated great artery, as occurs in increased flow states such as truncus arteriosus (aortic click) or ASD (pulmonary click) or in idiopathic dilatation of the pulmonary artery 3. Increased pressure in the great vessel, such as in aortic or pulmonary hypertension Because an aortic click is not usually heard with uncomplicated coarctation, its presence should suggest associated bicuspid aortic valve. In the latter condition, the click diminishes in intensity, becomes “buried” in the systolic murmur, and ultimately disappears as the valve becomes heavily calcified and immobile later in the course of the disease. Although a click implies cusp mobility, its presence does not necessarily exclude severe stenosis. A click would be expected to be absent in subvalvular stenosis. The timing of the pulmonary click in relationship to S1 (reflecting the isovolumic contraction period of the RV) is associated with hemodynamic severity in valvular pulmonary stenosis. With higher systolic gradient and lower pulmonary artery systolic pressure, the isovolumic contraction period shortens and thus the earlier the click occurs in relationship to S1. A pulmonary click can occur in idiopathic dilatation of the pulmonary artery, and this condition may be a masquerader of ASD, especially in young adults. The pulmonary click due to valvular pulmonary stenosis is the only right-sided heart sound that decreases with inspiration. Most other right-sided auscultatory events either increase in intensity with inspiration (most commonly) or show minimal

■

■

■

■

The presence, absence, or loudness of the ejection click does not correlate with the degree of valvular stenosis. An aortic click is not heard with uncomplicated coarctation; its presence should suggest associated bicuspid aortic valve. A click is absent in subvalvular or supravalvular aortic stenosis or hypertrophic obstructive cardiomyopathy. A pulmonary click can occur in idiopathic dilatation of the pulmonary artery, a condition that may mimic ASD, especially in young adults. The pulmonary click is best heard along the upper left sternal border. The aortic click radiates to the aortic area and the apex and does not change with respiration.

Mid-to-Late Nonejection Clicks (Systolic Clicks) Nonejection clicks are most commonly due to mitral valve prolapse. Rarely, nonejection clicks can be caused by papillary muscle dysfunction, rheumatic mitral valve disease, or hypertrophic obstructive cardiomyopathy.

Table 10. Causes of Ejection Clicks Aortic click Congenital valvular aortic stenosis Congenital bicuspid aortic valve Truncus arteriosus Aortic incompetence Aortic root dilatation or aneurysm Pulmonary click Pulmonary valve stenosis Atrial septal defect Chronic pulmonary hypertension Tetralogy of Fallot with pulmonary valve stenosis (absent if there is only infundibular stenosis) Idiopathic dilated pulmonary artery

Chapter 1 Cardiovascular Examination Other rare causes of nonejection clicks (that can masquerade as mitral prolapse) include ventricular or atrial septal aneurysms, ventricular free wall aneurysms, and ventricular and atrial mobile tumors, such as myxoma. A nonejection click not due to mitral valve prolapse does not have the typical responses to bedside maneuvers found with mitral valve prolapse, as outlined below. Mitral Valve Prolapse Maneuvers that decrease LV volume, such as standing or the Valsalva maneuver, move the click earlier in the cardiac cycle. Conversely, maneuvers that increase LV volume, such as assuming the supine position and elevating the legs, move the click later in the cardiac cycle. With a decrease in LV volume, a systolic murmur, if present, would become longer. Interventions that increase systemic blood pressure make the murmur louder.

■

■

11

Miscellaneous causes of nonejection clicks (that can masquerade as mitral prolapse) include ventricular or atrial septal aneurysms, ventricular free wall aneurysms, and ventricular and atrial mobile tumors, such as myxoma. Maneuvers that decrease LV volume, such as standing or the Valsalva maneuver, move the click earlier in the cardiac cycle. Conversely, maneuvers that increase LV volume, such as assuming the supine position and elevating the legs, move the click later in the cardiac cycle.

Second Heart Sound S2 is often best heard along the upper and middle left sternal border. Splitting of S2 (Fig. 2) is best heard during normal breathing with the subject in the sitting position.

SPLITTING?

NORMAL

Normal

NARROW FIXED

WIDE FIXED

REVERSED

Pulm HT

RBBB PS ASD

PDA (L to R shunt) LBBB AS IHD

VSD & L to R shunt Pseudo: S2 + OS S2 + S3 S2 + pericardial knock S2 + tumor plop

A2 vs P2?

A2 > P2

P2 > A2

Normal

Pulm HT (any) cause AS

Fig. 2. Branching logic tree for second heart sound (S2) splitting. A2, aortic closure sound; AS, aortic stenosis; ASD, atrial septal defect; HT, hypertension; IHD, ischemic heart disease; LBBB, left bundle branch block; L to R, left-toright; OS, opening snap; P2, pulmonic closure sound; PDA, patent ductus arteriosus; PS, pulmonary stenosis; Pulm HT, pulmonary hypertension; RBBB, right bundle branch block; S3, third heart sound; VSD, ventricular septal defect.

12

Section I Fundamentals of Cardiovascular Disease Determinants of S2 include the following: 1. Ventricular activation (bundle branch block delays closure of the ventricle’s respective semilunar valve) 2. Ejection time 3. Valve gradient (increased gradient with low pressure in the great vessel delays closure) 4. Elastic recoil of the great artery (decreased elastic recoil delays closure, such as in idiopathic dilatation of the pulmonary artery)

Splitting of S2 Wide but physiologic splitting of S2 (Fig. 3) may be due to the following: 1. Delayed electrical activation of the RV,such as in right bundle branch block or premature ventricular contraction originating in the LV (which conducts with a right bundle branch block pattern) 2. Delay of RV contraction,such as in increased RV stroke volume and RV failure 3. Pulmonary stenosis (prolonged ejection time) In ASD, there is only minimal respiratory variation in S2 splitting. This is referred to as fixed splitting. Fixed splitting should be verified with the patient in the sitting or standing position because healthy subjects occasionally appear to have fixed splitting in the supine position. When the degree of splitting is unusually wide, especially when the pulmonary component of the second heart sound (P2) is diminished, suspect concomitant pulmonary stenosis. Indeed, this condition is the cause of the most widely split S2 that can be recorded. Wide, fixed splitting, although considered typical of ASD, occurs in only 70% of patients with ASD. However, persistent expiratory splitting is audible in most. Normal respiratory variation of the S2 occurs in up to 8% of patients with ASD. With Eisenmenger physiology, the left-to-right shunting decreases and the degree of splitting narrows. A pulmonary systolic ejection murmur (increased flow) is common in patients with ASD, and with a significant left-to-right shunt, a diastolic tricuspid flow murmur can be heard as well. As with aortic stenosis, as pulmonary stenosis increases in severity, P2 decreases in intensity, and ultimately S2 becomes single. The wide splitting of S2 in mitral regurgitation and ventricular septal defect is related to early aortic valve closure (in ventricular septal defect, P2 is delayed

Fig. 3. Diagrammatic representation of normal and abnormal patterns in the respiratory variation of the second heart sound. The heights of the bars are proportional to the sound intensity. A, aortic component; AS, aortic stenosis; ASD, atrial septal defect; Exp., expiration; Insp., inspiration; MI, mitral incompetence; P, pulmonary component; PS, pulmonary stenosis; VSD, ventricular septal defect.

as well), which, in turn, is due to decreased LV ejection time, but the loud pansystolic regurgitant murmur often obscures the wide splitting of S2 so that the S2 appears to be single. Partial anomalous pulmonary venous connection may occur alone or in combination with ASD (most often of the sinus venosus type). Wide splitting of S2 occurs in both conditions, but it usually shows normal respiratory variation in isolated partial anomalous pulmonary venous connection.

Chapter 1 Cardiovascular Examination Pulmonary hypertension may cause wide splitting of S2, although the intensity of P2 is usually increased and widely transmitted throughout the precordium. ■

■

■

■

Fixed splitting should be verified with the patient in the sitting or standing position because healthy subjects occasionally appear to have fixed splitting in the supine position. Wide, fixed splitting, although considered typical of ASD, occurs in only 70% of patients with ASD. Wide splitting of S2 occurs in both partial anomalous pulmonary venous connection and ASD, but it usually shows normal respiratory variation in isolated partial anomalous pulmonary venous connection. Pulmonary hypertension may cause wide splitting of S2, although the intensity of P2 is usually increased and widely transmitted throughout the precordium.

Paradoxical (Reversed) Splitting of S2 Paradoxical splitting of S2 is usually caused by conditions that delay aortic closure. Examples include the following: 1. Electrical delay of LV contraction, such as left bundle branch block (most commonly) 2. Mechanical delay of LV ejection, such as aortic stenosis and hypertrophic obstructive cardiomyopathy 3. Severe LV systolic failure of any cause 4. Patent ductus arteriosus,aortic regurgitation,and systemic hypertension are other rare causes of paradoxic splitting Paradoxical splitting of S2 (that is, with normal QRS duration) may be an important bedside clue to significant LV dysfunction. In severe aortic stenosis, the paradoxical splitting is only rarely recognized because the late systolic ejection murmur obscures S2. However, when paradoxical splitting of S2 is found in association with aortic stenosis, usually in young adults (assuming left bundle branch block is absent), severe aortic obstruction is suggested. Similarly, paradoxical splitting in hypertrophic obstructive cardiomyopathy implies a significant resting LV outflow tract gradient. Transient paradoxical splitting of S2 can occur with myocardial ischemia, such as during an episode of angina, either alone or in combination with an apical systolic murmur of mitral regurgitation (papillary muscle dysfunction) or prominent fourth heart sound (S4).

■

■

13

When paradoxical splitting of S2 is found in association with aortic stenosis, usually in young adults (assuming left bundle branch block is absent), severe aortic obstruction is suggested. Similarly, paradoxical splitting in hypertrophic obstructive cardiomyopathy implies a significant resting LV outflow tract gradient. Transient paradoxical splitting of S2 can occur with myocardial ischemia, such as during an episode of angina, either alone or in combination with an apical systolic murmur of mitral regurgitation (papillary muscle dysfunction) or a prominent S4.

Intensity of S2 Loud S2 Ordinarily, the intensity of the aortic component of the second heart sound (A2) exceeds that of the P2. In adults, a P2 that is louder than A2, especially if P2 is transmitted to the apex, implies either pulmonary hypertension or marked RV dilatation, such that the RV now occupies the apical zone.The latter may occur in ASD (approximately 50% of patients). Hearing two components of the S2 at the apex is abnormal in adults, because ordinarily only A2 is heard at the apex. Thus, when both components of S2 are heard at the apex in adults, suspect ASD or pulmonary hypertension. Soft S2 Decreased intensity of A2 or P2, which may cause a single S2, reflects stiffening and decreased mobility of the aortic or pulmonary valve (aortic stenosis or pulmonary stenosis, respectively). A single S2 may also be heard in older patients and the following cases: 1. With only one functioning semilunar valve,such as in persistent truncus arteriosus,pulmonary atresia, or tetralogy of Fallot 2. When one component of S2 is enveloped in a long systolic murmur, such as in ventricular septal defect 3. With abnormal relationships of great vessels,such as in transposition of the great arteries ■

When both components of S2 are heard at the apex in adults, implying an increased pulmonary component of S2, suspect ASD or pulmonary hypertension.

14

Section I Fundamentals of Cardiovascular Disease

Opening Snap A high-pitched snapping sound related to mitral or tricuspid valve opening, when present, is abnormal and is referred to as an opening snap (OS). This may arise from either a doming stenotic mitral valve or tricuspid valve, more commonly the former. The intensity of an OS correlates with valve mobility. Rarely, an OS occurs in the absence of atrioventricular valve stenosis in conditions associated with increased flow through the valve, such as significant mitral regurgitation. In mitral stenosis, the presence of an OS, often accompanied by a loud S1, implies a pliable mitral valve. The OS is often well transmitted to the left sternal border and even to the aortic area. In mitral stenosis, the absence of an OS implies the following: 1. Severe valvular immobility and calcification (note that an OS can still be heard in some of these cases) 2. Mitral regurgitation is the predominant lesion

An LV S3, which implies that rapid LV filling can occur, is rare in pure mitral stenosis. Also, an RV S3 can occur in mitral stenosis with severe secondary pulmonary hypertension and RV failure. An RV S3 is found along the left sternal border and increases with inspiration. A tumor “plop” due to an atrial myxoma has the same early diastolic timing as an OS and can be confused with it. ■

■

■ ■

Significant mitral stenosis may be present in the absence of an OS if the mitral valve leaflets are fixed and immobile.

S2-OS Interval The S2–mitral OS interval reflects the isovolumic relaxation period of the LV. With increased severity of mitral stenosis and greater increase in left atrial pressures, the S2-OS interval becomes shorter and may be confused with a split S2. The S2-OS interval should not vary with respiration. The S2-OS interval widens on standing, whereas the split S2 either does not change or narrows. Mild mitral stenosis is associated with an S2-OS interval of more than 90 ms, and severe mitral stenosis with an interval of less than 70 ms. However, the S2-OS interval is an unreliable predictor of the severity of mitral stenosis. Other factors that increase left atrial pressures, such as mitral regurgitation or LV failure, can also affect this interval. When the S2-OS interval is more than 110 to 120 ms, the OS may be confused with an LV S3. In comparison, the LV S3 is usually low-pitched and is localized to the apex. A tricuspid valve OS caused by tricuspid stenosis can be recognized by its location along the left sternal border and its increase with inspiration. In normal sinus rhythm, a prominent A wave can be seen in the jugular venous pulse, along with slowing of the Y descent.

In mitral stenosis, the presence of an OS, often accompanied by a loud S1, implies a pliable mitral valve that is not heavily calcified. (In such cases, the patient may be a candidate for mitral commissurotomy or balloon valvuloplasty rather than mitral valve replacement.) In general, mild mitral stenosis is associated with an S2-OS interval >90 ms, and severe mitral stenosis with an interval 100 pg/mL

+ −

Criterion standard (cardiologists’ review) + − a=650 c=220 b d a+b=722 a+b+c+d=1,538

Values for b and d are calculated by use of simple algebra. The following are derived using the formulas in Table 2: Sensitivity = 650/722 = 90% Specificity = 596/816 = 73% PPV = 650/870 = 75% NPV = 596/668 = 89% +LR = 0.9/0.27 = 3.3 −LR = 0.1/0.73 = 0.14 See text for abbreviations.

4 NONCARDIAC SURGERY IN PATIENTS WITH HEART DISEASE Traci L. Jurrens, MD Clarence Shub, MD

The American College of Cardiology and the American Heart Association (ACC/AHA) have published guidelines for the perioperative evaluation and management of patients with heart disease who are to have noncardiac operations. The guidelines recommend a conservative approach. Expensive testing, invasive strategies, and revascularization are rarely, if ever, warranted just to “get the patient through an operation.” Rather, the indications for extensive perioperative testing or revascularization are generally similar to those in a nonoperative setting. ■

■

Patients without clinical evidence of heart disease are at low risk (about 0.15%) of perioperative myocardial infarction.

The mortality rate in association with perioperative myocardial infarction is significantly higher than that with an infarct unrelated to an operation. Previously, the risk of perioperative myocardial infarction was less well recognized and the antemortem diagnosis was more difficult. Increased awareness of the problem, better patient selection, improved anesthetic and operative techniques, improved perioperative monitoring and management, and improved diagnostic techniques (including the newer biomarkers such as serum troponin) have all contributed to a significant reduction in mortality from perioperative myocardial infarction. The risk of perioperative reinfarction is increased in the first 6 months after an index myocardial infarction. This risk decreases with increasing time between the index infarction and the planned operation. After percutaneous revascularization, patients have a highrisk period of 6 weeks and then an intermediate-risk period of 3 to 6 months. Ideally, noncardiac surgery is delayed at least 3 months. Sometimes clinical circumstances may warrant proceeding with surgery sooner than recommended (e.g., rapidly spreading tumors,

Testing or revascularization is not indicated just to “get the patient through an operation.”

EFFECT OF CORONARY ARTERY DISEASE The risk of a perioperative myocardial infarction in patients without clinical evidence of heart disease is approximately 0.15%. In patients with clinical heart disease, the risk of a perioperative myocardial infarction can be stratified according to the cardiovascular profile of the patient (major, intermediate, minor, or no clinical predictors of increased risk) and according to the cardiac stress of the operation (high, medium, or low stress or risk). 61

62

Section I Fundamentals of Cardiovascular Disease

impending aortic aneurysm rupture, major fractures, and infections requiring drainage). According to the ACC/AHA guidelines, an elective surgical procedure can be performed before 6 months has elapsed as long as the patient undergoes postinfarction risk stratification. Absence of postinfarction ischemia, a negative postinfarction stress test, and complete myocardial revascularization after infarction suggest a reduced risk of reinfarction with an elective operation. The ACC/AHA guidelines do suggest that it is prudent to wait at least 4 to 6 weeks after infarction before proceeding with an elective operation.

Risk factors in patients undergoing noncardiac operations include 1) type of operation (intrathoracic and intra-abdominal procedures have a higher risk than

limb operations); 2) presence and severity of coronary artery disease, especially if unstable (heart failure or unstable angina); 3) status of left ventricular function (ejection fraction); 4) age of patient; 5) severe valvular heart disease, especially aortic stenosis; 6) serious cardiac arrhythmias; 7) associated medical conditions (e.g., chronic obstructive pulmonary disease, hypoxemia, diabetes mellitus, and renal insufficiency); and 8) overall functional status. Clinical risk stratification tools can identify patients at high risk of perioperative ischemic events and guide appropriate perioperative medical strategies. The Revised Cardiac Risk Index identifies six independent predictors of major cardiac complications: high-risk surgery, history of ischemic heart disease, history of congestive heart failure, history of cerebrovascular disease, preoperative treatment with insulin, and preoperative creatinine level greater than 2.0 mg/dL (Table 1). One can easily calculate the risk of major cardiac complications by using the number of predictors (Table 2). Perioperative risks can be stratified further into major, intermediate, and low (minor) risks (Table 3). Active conditions are more important than dormant ones, and the degree of abnormality is also important. The presence of major risk predictors warrants further evaluation and (usually) treatment that may delay or cancel the elective operation. The urgency of a noncardiac operation may dictate patient management. Thus, a patient with a recent myocardial infarction and an acute abdominal crisis generally requires laparotomy without delay. The presence of intermediate predictors of increased perioperative risk warrants careful clinical assessment and, when appropriate, use of additional cardiac testing. Minor predictors have relatively less clinical importance.

Table 1. Predictors in the Revised Cardiac Risk Index

Table 2. Rates of Major Cardiac Complications in the Revised Cardiac Risk Index

1. 2. 3. 4. 5. 6.

No. of predictors

Cardiac risk, %

0 1 2 ≥3

0.4 0.9 7 11

■

■

Risk of perioperative reinfarction varies inversely with the time between the index infarction and the operation. Patients with a negative postinfarction stress test or complete postinfarction myocardial revascularization can proceed with an elective operation at 4 to 6 weeks after infarction.

The risk of perioperative reinfarction is not significantly different between patients who have had a Q-wave infarction and those who have had a non–Q-wave infarction. ■

Recent Q-wave and non–Q-wave myocardial infarctions are associated with the same risk of perioperative reinfarction.

PREOPERATIVE CARDIAC RISK INDEXES

High-risk surgery History of ischemic heart disease History of congestive heart failure History of cerebrovascular disease Preoperative treatment with insulin Preoperative creatinine >2.0 mg/dL

Chapter 4 Noncardiac Surgery in Patients With Heart Disease Risk stratification based on the type of operation planned is also important (Table 4). High-risk operations include 1) major intrathoracic procedures, 2) abdominal (intraperitoneal) operations, 3) aortic surgi-

Table 3. Clinical Predictors of Increased Perioperative Cardiovascular Risk (Myocardial Infarction, Congestive Heart Failure, and Death) Major risk Unstable coronary syndromes Recent myocardial infarction* with evidence of important ischemic risk by clinical symptoms or noninvasive study Unstable or severe† angina (Canadian class III or IV) Decompensated congestive heart failure Significant arrhythmias High-grade atrioventricular block Symptomatic ventricular arrhythmias in the presence of underlying heart disease Supraventricular arrhythmias with uncontrolled ventricular rate Severe valvular disease Intermediate risk Mild angina pectoris (Canadian class I or II) Prior myocardial infarction by history of pathologic Q waves Compensated or prior congestive heart failure Diabetes mellitus Minor risk Advanced age Abnormal ECG (left ventricular hypertrophy, left bundle branch block, ST-T abnormalities) Rhythm other than sinus (e.g., atrial fibrillation) Low functional capacity (e.g., inability to climb 1 flight of stairs with a bag of groceries) History of stroke Uncontrolled systemic hypertension ECG, electrocardiogram. *The American College of Cardiology National Database Library defines recent myocardial infarction as myocardial infarction occurring >7 days but ≤30 days previously. †May include “stable” angina in patients who are unusually sedentary.

63

cal procedures (e.g., aortic aneurysmectomy), and 4) peripheral vascular operations. High-risk operations have been associated with a higher incidence of postoperative congestive heart failure and a threefold greater incidence of myocardial infarction in comparison with other general surgical procedures. A major operation is often associated with large extravascular and intravascular fluid shifts or blood loss and postoperative hypoxemia. The magnitude and anticipated duration of the procedure are also important. Patients undergoing peripheral vascular operations are at high risk, primarily because of the increased incidence of associated coronary artery disease. Emergency major operations, especially in the elderly, are also considered high risk. Intermediate-risk operations include 1) carotid endarterectomy, 2) head and neck procedures, 3) orthopedic and prostate operations, and 4) less extensive intraperitoneal and intrathoracic procedures. Low-risk operations include 1) ophthalmologic procedures, 2) endoscopic surgery, 3) breast surgery, and 4) uncomplicated herniorrhaphy. ■

■

High-risk operations include major intrathoracic, abdominal (intraperitoneal), and aortic surgical procedures (e.g., aortic aneurysmectomy). Patients undergoing peripheral vascular operations are also at high risk, primarily because of the increased incidence of associated coronary artery disease.

NONVASCULAR VERSUS VASCULAR SURGERY Overall, perioperative cardiac event rates recently have decreased markedly, especially for patients undergoing nonvascular operations, partly because of improved patient selection, anesthetic techniques, and perioperative management. Most studies have focused on patients having vascular procedures, because they are at higher risk. Routine coronary angiography performed before a vascular operation has demonstrated that more than one-half of patients with clinically suspected coronary artery disease have severe multivessel or inoperable coronary artery disease. Even patients with peripheral vascular disease and no previous history of heart disease may have severe coronary artery disease, especially those with diabetes mellitus.

64

■

Section I Fundamentals of Cardiovascular Disease

Even patients with peripheral vascular disease and no previous history of heart disease may have severe coronary artery disease, especially those with diabetes mellitus.

Table 4. Cardiac Risk* Stratification for Noncardiac Surgical Procedures High (reported cardiac risk often >5%) Emergency major operations, particularly in the elderly Aortic and other major vascular Peripheral vascular Anticipated prolonged surgical procedures associated with large fluid shifts or blood loss Intermediate (reported cardiac risk generally 1-3 min after exertion 3 or 4 abnormal leads Low risk No ischemia or ischemia induced at high-level exercise (>7 METs or heart rate >130 bpm [0.1 mV Typical angina 1 or 2 abnormal leads Inadequate test Inability to reach adequate target workload or heart rate response for age without an ischemic response. For patients undergoing a noncardiac operation, ability to exercise to at least the intermediate-risk level without ischemia should be considered a low risk for perioperative ischemic events bpm, beats per minute; CAD, coronary artery disease; ECG, electrocardiographically; MET, metabolic equivalent. *Workload and heart rate estimates for risk severity require adjustment for patient age. Maximal target heart rates for 40- and 80-year-old subjects taking no cardioactive medication are 180 and 140 bpm, respectively.

66

Section I Fundamentals of Cardiovascular Disease

at a brisk pace (3.5-4 mph) are associated with energy costs of about 5 METs. Climbing a flight of stairs, scrubbing floors, and playing golf generally exceed 4 METs.

PREOPERATIVE FUNCTIONAL ASSESSMENT OF PATIENTS UNABLE TO EXERCISE In patients unable to exercise adequately, pharmacologic stress testing (intravenous dipyridamole [or adenosine] thallium [or sestamibi] imaging or dobutamine stress echocardiography) has been used as an alternative means of stress testing, and each type of pharmacologic stress testing demonstrates similar patterns of risk prediction. The stress testing data are most valu-

Table 6. Estimated Requirements for Various Activities 1 MET

4 METs

>10 METs

Can you take care of yourself? Eat, dress, or use the toilet? Walk indoors around the house? Walk a block or two on level ground at 2-3 mph or 3.2-4.8 km/h? Do light work around the house such as dusting or washing dishes? Climb a flight of stairs or walk up a hill? Walk on level ground at 4 mph or 6.4 km/h? Run a short distance? Do heavy work around the house like scrubbing floors or lifting or moving heavy furniture? Participate in moderate recreational activities such as golf, bowling, dancing, doubles tennis, or throwing a baseball or football? Participate in strenuous sports such as swimming, singles tennis, football, basketball, or skiing?

MET, metabolic equivalent.

able if the test results are negative (i.e., they have a high negative predictive value). Patients with normal scans are at low risk. The opposite is not true, however: positive stress tests have a low positive predictive value for perioperative events. The incorporation of clinical factors improves the specificity and predictive value of a positive scan. For example, patients with a thallium redistribution defect in patients with one or more clinical risk factors has been associated with a higher incidence of perioperative cardiac complications than in patients with a reversible thallium defect but without such clinical risk factors. Severe ischemia has greater predictive value than mild ischemia. In addition to thallium redistribution, ischemic electrocardiographic changes during the test are predictive of perioperative events. Dobutamine stress testing allows identification of the heart rate when ischemia first appears and calculation of the ischemic threshold, based on the expected age-related heart rate response. Ischemic thresholds less than 70% have been shown to increase perioperative risk.

CLINICAL APPROACH TO PREOPERATIVE ASSESSMENT AND MANAGEMENT Generally, the perioperative risk for nonvascular non–high-risk operations is low, limiting the predictive value of stress testing. Standard clinical evaluation should suffice in most of these low-risk patients. Functionally active patients without major- or intermediate-risk factors for whom there is no clinical suspicion of coronary artery disease do not need to undergo routine stress testing before a noncardiac operation, especially for a low-risk procedure. Patients with chronic stable angina who are active and able to perform activities of daily living (4-5 METs) probably can tolerate most types of low- or intermittent-risk noncardiac operations. This group of patients routinely would not require preoperative stress testing. For optimal use of resources, the ACC/AHA guidelines recommend that if a patient has had coronary revascularization within the past 5 years and if the clinical status has remained stable without recurrent symptoms or signs of ischemia, additional cardiac testing generally is not needed. For patients with ischemic heart disease who have not had coronary revascularization but who have undergone coronary evaluation in the past 2 years (assuming adequate testing and a

Chapter 4 Noncardiac Surgery in Patients With Heart Disease favorable outcome of testing), it usually is unnecessary to repeat testing unless there has been an acceleration of angina or new symptoms of ischemia have appeared during the interim. For patients at intermediate clinical risk, consideration of both the functional capacity and the level of operation-specific risk is necessary. Further noninvasive testing should be considered for patients with poor functional capacity or moderate functional capacity before a high-risk procedure is performed. Patients about to undergo a high-risk operation, especially if they have serious clinical risk factors, should be considered for preoperative stress testing. In selected patients with known coronary artery disease and significant (class III or IV) symptomatic limitation or accelerating angina, preoperative coronary angiography is often indicated, as it would be even if a noncardiac operation were not being contemplated. The CASS (Coronary Artery Surgery Study) registry showed that coronary revascularization before a noncardiac operation can decrease perioperative cardiac mortality to approximately 1% or less, compared with 2.4% for patients with similar coronary artery disease treated medically. There are no randomized trials showing benefit of coronary revascularization before noncardiac surgery. The potential risks (morbidity and mortality) of coronary artery bypass grafting itself, especially in patients older than 70 years, must also be considered before a noncardiac operation is performed. A recent Veterans Administration study demonstrated that perioperative outcomes are as good with perioperative medical treatment as with revascularization (percutaneous coronary intervention [PCI] or coronary artery bypass graft [CABG]) in carefully screened patients undergoing vascular surgery. This study excluded patients with left main coronary artery disease, severe systolic dysfunction, unstable angina, and severe aortic stenosis. A multicenter study of percutaneous transluminal coronary angioplasty (PTCA) has demonstrated an overall mortality of 1% (2.8% in the presence of triplevessel disease), a 4.3% incidence of nonfatal myocardial infarction, and a need for emergency CABG in 3.4% of patients. Coronary stenting (which is increasingly used during PCI) affects the timing of surgery. The strategy of performing coronary angiography and PTCA before a noncardiac operation to reduce the risk of a noncar-

67

diac procedure depends on individual circumstances and has not proved beneficial in controlled clinical trials. Prolonged antiplatelet therapy using clopidogrel after coronary stenting mandates delaying noncardiac surgery. The frequency of perioperative cardiac events in a patient who has had PTCA is highest immediately after the operative procedure and decreases thereafter with a second increase in cardiac events 90 days after PTCA because of neointimal hyperplasia. Although brief clopidogrel treatment can be used with bare metal stents (vs. drug-eluting stents), there is an increased risk of perioperative cardiac complications in the first 6 weeks after stenting procedures. The risk of converting a stable but flow-limiting coronary lesion into a less stable, nonflow-limiting lesion as a result of PCI should be taken into account. Stent thrombosis is most common in the first 2 weeks after stent placement but can occur later as well. The rate of stent thrombosis diminishes after endothelialization of the stent occurs (4-8 weeks). In patients who require noncardiac surgery after placement of bare metal stents, noncardiac surgery should be delayed at least 6 weeks after stent placement, at which time stents are generally endothelialized and antiplatelet therapy can be safely discontinued. There are no studies currently available on drug-eluting stents, but if they are used, delaying surgery for at least 6 months should be considered. The potential risk of stent thrombosis due to prematurely stopping clopidogrel has to be considered in the overall perioperative risk assessment. In most ambulatory patients who are active enough to perform adequate stress testing, the exercise electrocardiographic treadmill test is usually preferred because it provides an estimate of both functional capacity and ischemic response. In patients with an abnormal resting electrocardiogram (e.g., left ventricular hypertrophy, left bundle branch block, digitalis effect, and nonspecific ST-T abnormalities), a cardiac imaging exercise test such as exercise echocardiography or myocardial perfusion imaging should be considered. The choice of the specific test depends on various factors, especially local expertise with a specific technique. Interventional procedures rarely are needed just to lower the risk of a noncardiac operation, unless the intervention is thought to be indicated anyway (i.e., if the patient were not undergoing a noncardiac

68

Section I Fundamentals of Cardiovascular Disease

operation). Thus, the strategy of performing coronary revascularization just to avoid perioperative cardiac complications should be reserved for only a small subset of very high-risk patients. According to the ACC/AHA guidelines, class I indications for preoperative coronary angiography for patients with suspected or proven coronary artery disease include the following (Table 7): 1. High-risk results of noninvasive testing 2. Severe (class III or IV) angina unresponsive to medical therapy 3. Unstable angina 4. Nondiagnostic or equivocal noninvasive test results in a high-risk patient,for example,multiple clinical risk factors in a patient undergoing a highrisk operation (see above) Coronary angiography generally is not indicated in low-risk patients or in those who are asymptomatic after coronary revascularization and have good exercise capacity. ■

■

■

■

■

Generally, the perioperative risk for nonvascular non–high-risk operations is low, limiting the predictive value of cardiac stress testing. Standard clinical evaluation should suffice in most of these low-risk patients. Patients with chronic stable angina who are active and able to perform activities of daily living (4-5 METs) probably can tolerate the stress of most types of noncardiac operations. If a patient has undergone coronary revascularization within the past 5 years and if the clinical status has remained stable without recurrent symptoms or signs of ischemia, additional cardiac testing generally is not needed. For patients with ischemic heart disease who have not had coronary revascularization but who have undergone coronary evaluation in the past 2 years (assuming adequate testing and a favorable outcome of testing), it usually is unnecessary to repeat testing unless there has been an acceleration of angina or new symptoms of ischemia have appeared during the interim. In selected patients with known coronary artery disease and significant (class III or IV) symptomatic limitation or accelerating angina, preoperative coronary angiography is often indicated, as it would be even if a noncardiac operation were not being contemplated.

Table 7. Recommendations for Coronary Angiography in Perioperative Evaluation Class I: Patients with suspected or known CAD Evidence for high risk of adverse outcome based on noninvasive test results Angina unresponsive to adequate medical therapy Unstable angina, particularly when facing intermediate-risk or high-risk noncardiac surgery Equivocal noninvasive test results in patients at high clinical risk undergoing high-risk surgery Class IIa Multiple markers of intermediate clinical risk and planned vascular surgery (noninvasive testing should be considered first) Moderate to large region of ischemia on noninvasive testing but without high-risk features and without lower LVEF Nondiagnostic noninvasive test results in patients of intermediate clinical risk undergoing high-risk noncardiac surgery Urgent noncardiac surgery while convalescing from acute MI Class IIb Perioperative MI Medically stabilized class III or IV angina and planned low-risk or minor surgery Class III Low-risk noncardiac surgery with known CAD and no high-risk results on noninvasive testing Asymptomatic after coronary revascularization with excellent exercise capacity (≥7 METs) Mild stable angina with good left ventricular function and no high-risk noninvasive test results Noncandidate for coronary revascularization owing to concomitant medical illness, severe left ventricular dysfunction (e.g., LVEF 180 mm Hg and diastolic pressure >110 mm Hg) should be controlled before the operation is performed. Patients with hypertension whose blood pressure is controlled with medication usually tolerate anesthesia better than those with poorly controlled blood pressure. The decision to delay the

70

Section I Fundamentals of Cardiovascular Disease

operation to achieve improved blood pressure control should take into account the urgency of the operation. Significant perioperative hypertension occurs in approximately 25% of patients with hypertension, appears unrelated to preoperative control, and occurs frequently in patients undergoing abdominal aortic aneurysm repair and other peripheral vascular procedures, including carotid endarterectomy. If oral intake of antihypertensive medications must be interrupted, parenteral therapy may be needed perioperatively. Various antihypertensive medications can be used, including intravenous β-blockers, vasodilators, calcium channel blockers, and angiotensin-converting enzyme inhibitors. For patients taking clonidine orally, it may be helpful to switch to a long-acting clonidine cutaneous patch preoperatively to avoid “rebound hypertension” perioperatively. Preoperative myocardial ischemia occurs in up to 35% of elderly patients before hip fracture surgery. Early preoperative administration of epidural analgesia may lessen overall perioperative risks. Appropriate analgesia is especially important in this elderly group. ■

■

■

Treatment with β-blockers should remain uninterrupted as long as possible, especially in patients with coronary artery disease. Routine use of β-blockers should be used in high-risk patients undergoing surgery. Although calcium channel blockers and anesthetics have additive vasodilator and negative inotropic effects, most patients who take these agents can be anesthetized safely. Although mild or moderate hypertension usually does not warrant delaying the operation, severe hypertension should be controlled before the operation is performed.

ARRHYTHMIAS AND CONDUCTION DISTURBANCES Rapid postoperative atrial arrhythmias affect almost 1 million patients annually. In contrast, bradyarrhythmias or ventricular arrhythmias severe enough to require treatment affect less than 1% of patients undergoing noncardiac surgery. Clinical evaluation should seek to uncover any underlying heart or pulmonary disease, drug toxicity, and electrolyte or metabolic abnormality

that might be causing arrhythmias or conduction disturbances. Symptomatic or hemodynamically significant arrhythmias should be treated before the patient undergoes a noncardiac operation; the indications for treatment are similar to those in the nonoperative setting. It is important to correct even mild degrees of preoperative hypokalemia in patients taking digitalis. The respiratory alkalosis that usually occurs during general anesthesia may cause a decrease in extracellular potassium concentration and provoke arrhythmias. Asymptomatic conduction system disease such as bundle branch block, bifascicular block, or even trifascicular block does not predict high-grade or complete heart block during a noncardiac operation and does not by itself mandate prophylactic temporary pacing. Atrial tachyarrhythmia is a common complication after thoracic surgery and is associated with longer hospital stay. Calcium channel blockers and β-blockers reduce the risk of atrial tachyarrhythmias. ■

■

■

Symptomatic or hemodynamically significant arrhythmias should be treated before the patient undergoes a noncardiac operation. It is important to correct even mild degrees of preoperative hypokalemia in patients taking digitalis. Asymptomatic conduction system disease such as bundle branch block, bifascicular block, or even trifascicular block does not predict high-grade or complete heart block during a noncardiac operation and does not by itself mandate prophylactic temporary pacing.

APPROACH TO PATIENTS REQUIRING LONGTERM ORAL ANTICOAGULATION The issue of discontinuation of oral anticoagulation in the perioperative setting requires balancing the thromboembolic potential of the patient’s cardiovascular disease with the hemorrhagic risk of the operation. For most cardiovascular situations, including patients with bioprosthetic valves, the acute thromboembolic potential is low. The thromboembolic potential is high for patients with mechanical prosthetic valves, especially those in the tricuspid or mitral position, and for patients with recent embolic episodes from, for example, cardiomyopathies, atrial fibrillation, ventricular aneurysms, or acute infarctions. Unfortunately, there

Chapter 4 Noncardiac Surgery in Patients With Heart Disease are no large randomized trials studying the risk of thromboembolism versus the risk of hemorrhage in various conditions and types of operation. Several small studies have suggested that patients with low or moderate risk can have an international normalized ratio (INR) less than 2.0 for 5 to 7 days with relative safety. A reasonable approach to these patients would be to discontinue the use of warfarin several days in advance of the operation, which should be performed as soon as the INR is 1.5 or less. Oral anticoagulation is resumed as soon as possible postoperatively, and heparin is reserved for patients whose INR is less than 2.0 for 5 days or more. In patients at high risk of thromboembolic complications (Table 8), intravenous heparin coverage can be instituted until 6 hours before the operation and then resumed as soon as possible postoperatively. The use of heparin can be discontinued after the use of warfarin has been resumed and the INR is in the therapeutic range. Recent studies have shown that standardized periprocedural use of subcutaneous low-molecularweight heparin (LMWH) is associated with a low risk of thromboembolic and major bleeding complications. LMWH is an alternative to unfractionated heparin for bridging therapy. The use of LMWH is appealing because it can be given in an outpatient setting. In patients who have mechanical valves and require interruption of warfarin therapy for emergency noncardiac surgery, fresh frozen plasma is preferred over high-dose vitamin K. Five days of subtherapeutic INR for patients with low to moderate thromboembolic risk is probably reasonable. ■

■

Perioperative heparin coverage should be used in patients with a high thromboembolic risk. Perioperative use of subcutaneous LMWH is being used as an alternative to standard anticoagulation with unfractionated heparin.

EXAMINATION STRATEGY For the purposes of cardiology examinations, the most important information to remember about evaluating

71

Table 8. High-Risk Factors for Thromboembolism Older generation thrombogenic valves (BjörkShiley) Mechanical mitral valve replacement Mechanical aortic valve replacement with any risk factor or multiples of the following risk factors in the absence of a prosthetic heart valve: 1. Atrial fibrillation 2. Left ventricular dysfunction 3. Previous thromboembolism 4. Hypercoagulable condition

patients with heart disease before noncardiac operations includes the following: 1. The clinical indicators of high,intermediate,and low risk for a cardiac event in the perioperative period 2. The surgical procedures associated with high, intermediate, and low risk for precipitating perioperative myocardial infarction and cardiac complications 3. Patients with peripheral vascular disease are at high risk of a perioperative cardiac event,and most of these patients should have some type of stress test before an elective vascular operation 4. The indications for preoperative testing and revascularization are similar to those in the nonoperative setting and are not based on getting the patient through the planned operation Not all patients with peripheral vascular disease need preoperative stress tests (e.g., patients who had complete coronary revascularization 50% systemic from any cause

Medical approval*†

No

No

No

Yes

Yes No

AVNRT, atrioventricular nodal reentrant tachycardia; BP, blood pressure; CHD, coronary heart disease; ECG, electrocardiogram; EF, ejection fraction; ETT, exercise treadmill test; INR, international normalized ratio; LV, left ventricular; LVEDD, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; LVESD, left ventricular end-systolic diameter; MET, metabolic equivalent; NSVT, nonsustained ventricular tachycardia; PPM, permanent pacemaker; VT, ventricular tachycardia; WPW, Wolff-Parkinson-White. *For medical approval for a CMV license (“Yes”), all criteria must be present. For disapproval (“No”), only one criterion must be present. †Restrictions are national. Individual states may impose additional restrictions on private and commercial drivers.

examiners [AMEs]) around the world who are designated to provide medical applications, give forensic examinations, and issue FAA medical certificates to qualified airmen. There are three standards of medical fitness (examinations for classes I, II, and III), each designed for the type of flying in which a pilot participates. The most stringent standards are required for a class I medical certificate: An examination is required for airline captains of scheduled air carriers and is valid

for 6 calendar months. A class II medical certificate is required for cocaptains (first officers) and other professional pilots (e.g., agricultural spray pilots). A class III medical certificate is required for recreational pilots. The cardiovascular evaluation of a pilot with underlying cardiac conditions may be done by a cardiologist in conjunction with an AME, although ultimate certification involves review by the FAA at either the office of the applicable regional flight surgeon or the

114

Section I Fundamentals of Cardiovascular Disease

Aeromedical Certification Division of the FAA in Oklahoma City, Oklahoma. Of the 15 disqualifying medical conditions facing pilots, seven are related to cardiovascular disease: 1. Angina pectoris 2. Coronary heart disease that has been treated or,if untreated,has been symptomatic or clinically significant 3. Myocardial infarction 4. Cardiac valve replacement 5. Permanent cardiac pacemaker 6. Heart replacement 7. Disturbance of consciousness without satisfactory explanation of cause Of these, only persistent angina pectoris, having more than one heart valve replaced (except for the Ross procedure), and heart transplantation are absolutely disqualifying. If an implantable cardioverter-defibrillator (ICD) is implanted for cardiac arrest, for ventricular tachycardia or ventricular fibrillation, or for any of the standard criteria for ICD, the pilot is automatically disqualifed; however, all other indications (i.e., atrial fibrillation) are evaluated by the FAA on a case-by-case basis. Using a waiver system outlined in Federal Aviation Regulation 67.401, the FAA may allow a pilot to fly if medical stability can be established. This is also known as special issuance authorization. If the systolic blood pressure is greater than 155 mm Hg, or if the diastolic blood pressure is greater than 95 mm Hg, a pilot is grounded until the blood pressure is under control and the FAA is satisfied that no serious underlying cardiovascular disease is the cause. For consideration of special issuance, the FAA usually asks three questions that the evaluating cardiologist must answer:

1. Is the coronary artery risk factor controlled (e.g., lipid levels, fasting blood glucose level, and exercise tolerance)? 2. Is there reversible ischemia? 3. Does the patient require any medications? Are there any side effects of the medications? Pilots who have a myocardial infarction or undergo any invasive procedure such as stent insertion, valve replacement, or coronary artery bypass grafting require a mandatory 6-month stand-down time to establish condition stability before forensic testing may be performed for certification purposes. All pilots must undergo testing 6 months after a percutaneous coronary intervention; class I pilots must undergo coronary angiography to show patency, but class II and class III pilots may undergo noninvasive imaging to show absence of ischemia. Pilots who require insertion of a permanent pacemaker require 6 weeks of condition stability before forensic testing may ensue. Heart valve replacement cases are considered by the Aeromedical Certification Division of the FAA on a case-by-case basis.The FAA is developing specialized centers of excellence for forensic examinations involving pilots with complex cardiovascular and other potentially disqualifying medical conditions (Mayo Clinic is one of those centers). Currently, AMEs cannot approve certification of cardiovascular conditions that would be otherwise disqualifying without approval from the FAA directly. Therefore, they will typically defer the application (FAA form 8500-8) for authorization. Because the review process involves a decision from an external review panel of academic cardiologists, the pilot’s certification decision may require up to 6 months for final approval.

SECTION

II

Noninvasive Imaging

Cardiac Myxoma

8 PRINCIPLES OF ECHOCARDIOGRAPHY Teresa S. M. Tsang, MD

M-MODE AND TWO-DIMENSIONAL ECHOCARDIOGRAPHY

This chapter summarizes the central role of echocardiography in both the initial diagnosis and the quantification of the nature and severity of specific cardiovascular diseases. It is important to appreciate the relative strengths, weaknesses, and incremental value of information obtained by different echocardiographic methods. The American College of Cardiology, the American Heart Association, and the American Society of Echocardiography jointly published practice guidelines in 2003 for the use of echocardiography. These guidelines make recommendations about appropriate and inappropriate uses of echocardiography. The guidelines divide indications into the following categories: I, generally indicated; IIa, evidence is conflicting but in favor of usefulness; IIb, conflicting evidence with less well-established indications; and III, generally thought to be either not useful or contraindicated. The ensuing sections discuss some of the generally accepted indications and their technologic considerations; Table 1 lists important class III examples.

M-mode imaging, which dates from the early days of echocardiography, is still a useful part of a complete ultrasonographic examination and can be acquired using two-dimensional (2D) guidance. The typical views obtained with the transducer placed at the left parasternal region, sweeping from the ventricular level to the mitral valve level to the aortic valve level, are shown in Figure 1. Measurements of the left ventricular (LV) dimensions and wall thickness can be readily derived from M-mode recordings and are usually made according to the recommendations of the American Society of Echocardiography at end diastole (the onset of the QRS complex) and end systole (the point of maximal upward motion of the LV posterior wall endocardium). These measurements are made from leading edge to leading edge. LV ejection fraction can be readily calculated from measurements obtained by M-mode or 2D assessments (see “Assessment of Ventricular Function” section in this chapter). 2D imaging provides important structural and functional information on cardiac disease. The American Society of Echocardiography has recommended that cardiac imaging be performed in three orthogonal planes: long-axis (from the aortic root to the apex),

TRANSTHORACIC ECHOCARDIOGRAPHY Anatomical and functional assessment of cardiac chambers, valves, myocardium, pericardium, and the aorta are important aspects of the echocardiographic examination. 117

118

Section II Noninvasive Imaging

Table 1. ACC/AHA/ASE Guidelines for Clinical Applications of Echocardiography: Conditions for Which Echocardiography is Generally Not Recommended 1. Patients for whom the results of the study would have no effect on the diagnosis, clinical decision making, or management 2. Routine repetition of echocardiography in past users of anoretic drugs with prior normal studies or known trivial valvular abnormalities 3. Asymptomatic heart murmur thought to be functional by an experienced clinician 4. Routine reevaluation of asymptomatic, mild aortic stenosis in patients with stable physical signs and normal left ventricular systolic function 5. Routine reevaluation of asymptomatic, mild to moderate mitral stenosis in patients with stable signs 6. Routine reevaluation of asymptomatic, mild to moderate mitral or aortic regurgitation in patients with stable signs in the absence of chamber dilatation 7. Exclusion of mitral valve prolapse in patients without clinical symptoms or signs or family history 8. Routine repetition of echocardiography for patients with mitral valve prolapse who have no or mild regurgitation and no change of symptoms or signs 9. Routine reevaluation of patients who have uncomplicated endocarditis during antibiotic therapy when there are no changes in symptoms or signs 10. Transient fever without evidence of bacteremia 11. Routine reevaluation of valve replacement when there are no clinical symptoms or signs to suggest dysfunction or a failing prosthesis 12. Evaluation of chest pain when a clear-cut noncardiac cause is responsible 13. Assessment of prognosis more than 2 years after myocardial infarction 14. Patients who have been receiving long-term therapeutic anticoagulation and who do not have mitral valve disease or hypertrophic cardiomyopathy before cardioversion (unless there are other reasons for anticoagulation, e.g., prior embolus or thrombus known from previous transesophageal echocardiography) 15. Routine screening echocardiogram for participation in competitive sports if patients have a normal cardiovascular history, electrocardiogram, and physical examination 16. Suspected myocardial contusion in hemodynamically stable patients who have a normal electrocardiogram and no abnormal cardiothoracic physical findings or no mechanism of injury that suggests cardiovascular contusion ACC, American College of Cardiology; AHA, American Heart Association; ASE, American Society of Echocardiography.

short-axis (perpendicular to the long axis), and fourchamber (traversing both ventricles and atria through the mitral and tricuspid valves). Long-axis and short-axis refer to axes of the heart not the body. The three planes can be visualized with four standard transducer positions: parasternal, apical, subcostal, and suprasternal.The views obtained are depicted in Figure 2.

DOPPLER ECHOCARDIOGRAPHY Doppler echocardiography uses the Doppler effect, that is, the change in the frequency of sound waves as the sound source moves toward or away from the observer (Equation 1).

Equation 1. Frequency Shift (Δf) 2ftv cos θ Δf = ________ c Δf = Doppler frequency shift ft = transmitted frequency cos θ = (angle theta) angle between the vector of the moving object and the interrogating beam c = (constant), velocity of sound in tissue or water (1,560 m/s) v = velocity of the moving object

Chapter 8 Principles of Echocardiography

A

C

119

B

D

Fig. 1. A, An M-mode cursor is placed along different levels (1, ventricular level; 2, mitral valve level; 3, aortic valve level) of the heart, with parasternal long-axis two-dimensional echocardiographic guidance. B-D, Representative normal M-mode echocardiograms at the midventricular, mitral valve, and aortic valve levels, respectively. Arrows in B indicate end-diastolic (EDd) and end-systolic (ESd) dimensions of the left ventricle. C, The M-mode echocardiogram of the anterior mitral leaflet: A, peak of late opening with atrial systole; C, closure of the mitral valve; D, end systole before mitral valve opening; E, peak of early opening; F, mid-diastolic closure. The double-headed arrow in D indicates the dimension of the left atrium at end systole. Ao, aorta; AV, aortic valve; LA, left atrium; LV, left ventricle; PW, posterior wall; RVOT, right ventricular outflow tract; VS, ventricular septum.

120

Section II Noninvasive Imaging

A

B

Fig. 2. A, Drawings of the longitudinal views from the four standard transthoracic transducer positions. Shown are the parasternal long-axis view (1), parasternal right ventricular (RV) inflow view (2), apical four-chamber view (3), apical five-chamber view (4), apical two-chamber view (5), subcostal four-chamber view (6), subcostal long-axis (five-chamber) view (7), and suprasternal notch view (8). B, Drawings of short-axis views. These views are obtained by rotating the transducer 90° clockwise from the longitudinal position. Drawings 1-6 show parasternal short-axis views at different levels by angulating the transducer from a superior medial position (for the imaging of the aortic and pulmonary valves) to an inferolateral position, tilting toward the apex (from level 1 to level 6 short-axis views). Shown are shortaxis views of the right ventricular outflow (RVO) tract and pulmonary valve (1), aortic valve and left atrium (LA) (2), RVO tract (3), and short-axis views at the left ventricular (LV) basal (mitral valve [MV] level) (4), the LV midlevel (papillary muscle) (5), and the LV apical level (6). A good view to visualize the RVO tract is the subcostal short-axis view (7). Also shown is the suprasternal notch short-axis view of the aorta (Ao) (8). RPA, right pulmonary artery.

This Doppler frequency shift is detected and translated into a blood flow velocity (Equation 2) by the Doppler transducer and instrument.

Equation 2. Velocity (v), m/s Δf c v = ________ 2ft cos θ (Definitions as in Equation 1)

The velocity of blood can be used to determine gradients, intracardiac pressures, volumetric flow, and valve areas.

Pulsed wave Doppler echocardiography and continuous wave Doppler echocardiography are the two most commonly used spectral Doppler modalities (Table 2). Pulsed wave Doppler echocardiography is “site specific,” allowing the measurement of blood velocities at a particular region of interest. The disadvantage is aliasing of the signal when velocities reach one-half of the pulse repetition frequency, or the Nyquist limit (Fig. 3).This property limits the maximal velocity that can be measured with pulsed wave Doppler echocardiography. Continuous wave Doppler echocardiography measures all velocities in the path of the ultrasound beam, is not site specific, and is not limited by aliasing. The disadvantage is that although very high velocities

Chapter 8 Principles of Echocardiography

Table 2. Appropriate Utilization of Doppler Modalities Pulsed wave Flow volume Diastolic filling variables Pulmonary/hepatic vein flow Localizing site of flow disturbance

Continuous wave Valvular and other stenotic gradients Intracardiac pressure Mitral regurgitant velocities Pressure half-time measurements Intracavitary gradients

121

can be recorded, the specific anatomical site where the highest velocity is present cannot be accurately localized (but it can be inferred). Continuous wave Doppler echocardiography is typically used to measure highvelocity jets and gradients. Color flow imaging is computer-enhanced pulsed wave Doppler echocardiography that displays the velocity and directional information of blood flow. Red depicts blood flow toward the transducer and blue,away from the transducer. Color flow imaging, like pulsed wave Doppler imaging, has a Nyquist limit and displays aliasing. Color flow imaging is used to detect, localize, and semiquantitate abnormal flow, such as that resulting from valvular regurgitation, shunts, or intracavitary obstruction. Tissue Doppler imaging is now an integral part of routine electrocardiography and is most commonly used for assessment of mitral anular motion, which is part of a comprehensive diastolic function assessment (see “Diastolic Function Assessment”section).The velocity of the mitral anulus motion represents the velocity of changes in the LV long-axis dimensions. The diastolic velocity has been considered a measure of the intrinsic speed of myocardial relaxation. Early diastolic velocity is recorded at the septal or lateral mitral anulus using a pulse wave technique with a 1.5-mm sample volume.The ratio of peak early mitral diastolic LV filling velocity (E) to the mitral anulus velocity (e') by tissue Doppler imaging (i.e., E/e') provides an excellent assessment of LV diastolic filling pressures in sinus rhythm and in atrial fibrillation. Cutoff values differ among echocardiographic laboratories, depending on specificities and sensitivities chosen. In the laboratory at Mayo Clinic, E/e' is high if it is more than 15 and low or normal if it is less than 8. However, between 8 and 15, there is considerably variability in filling pressures.

CONTRAST ECHOCARDIOGRAPHY

Fig. 3. Pulsed wave and continuous wave Doppler spectra from a patient with aortic stenosis (AS) and aortic regurgitation (AR). The pulsed wave sample volume is in the left ventricular outflow tract (LVOT) and demonstrates aliasing and “wrapping around” the baseline of the high-velocity AR signal. The continuous wave signal displays the entire AS and AR signals.

Identification of intracardiac shunts is one of the most frequent indications for contrast echocardiography, and agitated saline solution remains the most commonly used contrast agent. Saline bubbles do not cross the pulmonary vascular bed, and this precludes opacification of left-sided chambers without an intracardiac shunt. In recent years, stabilized solutions of microbubbles have been developed that can traverse the pulmonary capillary bed in high concentration after intravenous

122

Section II Noninvasive Imaging

injection. These microbubble agents are capable of producing high-intensity signals not only within the LV but also within the myocardium following intravenous injection. Contrast agent facilitates the identification of the endomyocardial border and is most often used in stress echocardiography when visualization of the endocardium is essential for assessment of ischemia. Second harmonic imaging enhances the ultrasonic backscatter from contrast microbubbles (which resonate in an ultrasonic field) while decreasing the returning signal from myocardium (which does not resonate).

ASSESSMENT OF CHAMBER SIZE AND WALL THICKNESS The American Society of Echocardiography has recently published updated guidelines for cardiac chamber quantitation. Left Ventricular Size It is recommended that LV end-diastolic diameter (LVEDD), end-systolic diameter (LVESD), and wall thicknesses be measured at the level of the LV minor axis, approximately at the mitral valve leaflet tips (Fig. 1 A). These linear measurements can be made directly from 2D images (Fig. 1 A) or by using 2D-targeted Mmode echocardiography (Fig. 1 B). It is not always possible to align the M-mode cursor perpendicularly to the long axis of the ventricle, a requirement that is critical for measurement of a true minor-axis dimension. As an alternative, chamber dimension and wall thicknesses can be acquired from the parasternal short-axis view using direct 2D measurements or targeted Mmode echocardiography, provided that the M-mode cursor can be positioned perpendicularly to the septum and LV posterior wall. As a general guideline, the upper limit of a normal LVEDD is approximately 5.5 cm, but it varies according to body surface area. LV enlargement is an important finding, especially in patients with valvular regurgitant lesions, hypertension, cardiomyopathy, and LV remodeling after myocardial infarction.Thus, accurate measurement of LV diameter with serial echocardiographic monitoring is important for many clinical diagnoses. LV size is often interpreted as LV end-diastolic diameter.

Left Ventricular Wall Thickness LV wall thickness is routinely measured in a standard echocardiographic study. LV septal wall thickness (SWT) and posterior wall thickness (PWT) are measured at end-diastole (d) from 2D or Mmode recordings routinely. The measurements of the septal and posterior walls are obtained at the same level of the ventricle as the LV diameter (Fig. 1 B). LV mass can then be calculated from the following formula: LV mass = 0.8 × [1.04 (LVEDD + PWTd + SWTd)3 − (LVEDD)3 + 0.6 g

Left Atrial Size Traditionally, a single-dimension M-mode left atrial (LA) dimension has been used for assessment of LA size (Fig. 1 D). Recently, LA volume, indexed to body surface area, has been shown to be more accurate. The upper limit of the normal range is 28 mL/m2. The biplane area-length method (Fig. 4) and biplane Simpson summation of discs method (Fig. 5) have been considered valid for assessment of LA volume. At Mayo Clinic, LA volume by the biplane area-length method is routinely assessed. Right Ventricular Size, Right Atrial Size, and Right Ventricular Wall Thickness Right ventricular and right atrial size assessment is qualitatively described in most clinical laboratories. This is particularly important in patients with pulmonary hypertension, pulmonary diseases, and tricuspid or pulmonary valvular lesions. Abnormalities may also reflect the severity of left heart disease. Some guidelines with respect to assessment and interpretation of the right ventricular and right atrial sizes have been included in the most recent “Recommendations for Chamber Quantitation” report by the American Society of Echocardiography. Right ventricular free wall thickness, normally less than 0.5 cm, is measured using either M-mode or 2D imaging. Although right ventricular free wall thickness can be assessed from the apical and parasternal long-axis views, the subcostal view at the level of the tricuspid valve chordae tendineae, measured at the peak of the R wave, provides less variation and closely correlates with right ventricular peak systolic pressure.

Chapter 8 Principles of Echocardiography

Fig. 4. The four-chamber (4C) and two-chamber (2C) lengths should be within 5 mm of each other; otherwise, foreshortening should be considered. In the formula, L is the average of the two lengths. (Note: the smaller of the two lengths has been used for L, which is acceptable if the difference between the two is no more than 5 mm). The arrow in the electrocardiographic tracing indicates the stage of the cardiac cycle represented by the drawings. A, area; AL, area-length; L, length; LA, left atrial.

ASSESSMENT OF VENTRICULAR FUNCTION (SYSTOLIC, DIASTOLIC, GLOBAL, AND REGIONAL) Systolic Function Assessment LV global systolic function can be evaluated with several echocardiographic techniques.These include 2D volumes derived from two- and four-chamber areas and length measurements (area-length method) and the modified Simpson method (or summation of discs). Formulae for calculating LV ejection fraction from 2D volumes (Equation 3) or 2D-directed M-mode (Equation 4), and fractional shortening (Equation 5) are shown below.

123

Fig. 5. The length (L) is the dimension perpendicular to the discs, from the plane of the mitral anulus to the superior aspect of the left atrium (LA). The arrow in the electrocardiographic tracing indicates the stage of the cardiac cycle represented by the drawings. ai, area by integration along chord in four-chamber view; bi, area by integration along chord in two-chamber view.

Equation 3. Ejection Fraction (EF), % LVEDV − LVESV EF = ________________ × 100 LVEDV LVEDV = LV end-diastolic volume LVESV = LV end-systolic volume (This formula can be applied to any contracting cavity; volumes are measured by the modified Simpson method with online software.)

Equation 4. Ejection Fraction (EF), % LVED 2 − LVES 2 D D __________________ EF = × 100 LVED 2 D

LVEDD = LV end-diastolic diameter LVESD = LV end-systolic diameter

124

Section II Noninvasive Imaging

Equation 5. Fractional Shortening (FS), % LVEDD − LVESD FS = ________________ × 100 LVEDD LVEDD = LV end-diastolic diameter LVESD = LV end-systolic diameter Cardiac output can be derived from 2D volumes or from use of Doppler echocardiographic techniques (Equations 6-9). Equation 6. Stroke Volume (SV), mL SV = LVEDV − LVESV (Definitions as in Equation 3)

Equation 7. Stroke Volume (SV), mL SV = area × TVI Area = πr2 (i.e., the cross-sectional area [cm2] through which velocity is recorded) d 2 = 0.785d2 πr2 = π __ 2

( )

d = diameter r = radius TVI = time-velocity integral = stroke distance (cm), which is the distance over which blood travels in one cardiac cycle (the cycle velocity [cm/s] divided by time [s])

Equation 8. Cardiac Output (CO), L/min CO = Stroke volume × Heart rate

Equation 9. Cardiac Index (CI), L/min per m2 Cardiac output CI = ______________ Body surface area

Regional LV function is based on the 2D assessment of the contractility of 16 LV wall segments (six segments at the base and mid-ventricle and four at the apical level) (Fig. 6). A numerical score is given to each segment depending on contractility: 1, normal; 2, hypokinetic; 3, akinetic; 4, dyskinetic; and 5, aneurysm. A wall motion score index can then be derived (Equation 10). Studies have demonstrated adverse prognostic significance from high wall motion scores.

Equation 10. Wall Motion Score Index Sum of wall scores ÷ Number of segments visualized Scoring of segmental contraction 1 = normal 2 = hypokinetic 3 = akinetic 4 = dyskinetic 5 = aneurysm (Hyperdynamic walls are considered normal [i.e., score = 1].)

Diastolic Function Assessment Mitral inflow assessment is fundamental to the evaluation of diastolic function. Mitral E (early filling phase) and A (atrial contraction) velocities, deceleration time, and isovolumic relaxation time (IVRT) are measured (Fig. 7). In general, three abnormal patterns are recognized: impaired relaxation (grade 1 diastolic dysfunction), pseudonormal filling (grade 2 diastolic dysfunction), and restrictive filling (grade 3 [reversible] and grade 4 [irreversible] diastolic dysfunction) (Fig. 8). At Mayo Clinic, abnormal relaxation with elevated filling pressures (grade 1A) is distinguished from that without elevated filling pressures (grade 1). Mitral inflow patterns change depending on loading conditions. Therefore, other assessments are also necessary to provide a more comprehensive evaluation, especially to distinguish pseudonormal from normal. The Valsalva maneuver can be used to decrease preload and unmask the seemingly normal pattern of pseudonormal filling to reveal a pattern characteristic of relaxation abnormality.The pulmonary venous flow pattern, the tissue Doppler mitral anular velocity profile (Fig. 9), left atrial size, and color M-mode all contribute to the assessment of diastolic function and filling pressures,

Chapter 8 Principles of Echocardiography Parasternal short-axis

Apical 4-chamber

125

Parasternal long-axis

Basal

Mid

Apical

Apical 2-chamber

Apical long-axis

Fig. 6. Schema of the 16 left ventricular wall segments used to assess regional systolic function and wall motion score index. A, anterior; AL, anterolateral; Ao, aorta; AS, anteroseptal; I, inferior; IL, inferolateral; IS, inferoseptal; L, lateral; LA, left atrium; LV, left ventricle; P, posterior; PL, posterolateral; PS, posteroseptal; RA, right atrium; RV, right ventricle; S, septal.

allowing classification of diastolic function and LV filling pressures (Fig. 8). At Mayo Clinic, left atrial size is measured as left atrial volume.

HEMODYNAMIC ASSESSMENT The following is a list of commonly used echocardiographic hemodynamic variables and their clinical usefulness:

Relaxation abnormality

1. Pressure gradients (maximal instantaneous and mean)—valvular stenosis, prosthetic valve, left and right ventricular outflow tract obstruction, and coarctation of aorta 2. Intracardiac pressures—right ventricular, pulmonary artery, and LV systolic and end-diastolic pressures 3. Volumetric flow—stroke volume, cardiac output, regurgitant volume and fraction, and, less commonly,

Restrictive

Fig. 7. Schematic left ventricular (LV), aortic (Ao), and left atrial (LA) pressure tracings and corresponding mitral inflow Doppler spectrum. A, atrial contraction; DT, deceleration time; E, early filling phase; IVRT, isovolumic relaxation time.

126

Section II Noninvasive Imaging

Fig. 8. Diastolic function assessment using mitral inflow, pulmonary venous flow, tissue Doppler imaging, left atrial size, and color M-mode. LV, left ventricular.

Fig. 9. Patterns of mitral inflow and mitral anulus velocity from normal to restrictive physiology. The mitral anulus velocity was obtained from the septal side of the mitral anulus with tissue Doppler imaging. Each calibration mark in the recording of mitral anulus velocity represents 5 cm/s. Early diastolic anulus velocity (e') is greater than late diastolic anulus velocity (a') in a normal pattern. In all other patterns, e' is not greater than a'. In relaxation abnormality, e' and a' parallel early (E) and late (A) velocities of mitral inflow. However, when filling pressure is increased (pseudonormalization and restrictive physiology), e' remains decreased (i.e., persistent underlying relaxation abnormality) while mitral inflow E velocity increases. Hence, E/e' may be useful in estimating left ventricular filling pressure.

Chapter 8 Principles of Echocardiography shunt fraction (pulmonary stroke volume/systemic stroke volume [Qp/Qs]) 4. Valve areas—continuity equation and pressure half-time 5. Diastolic filling variables To make these measurements, it is essential to understand and use the modified Bernoulli equation (Equation 11 and Fig. 10), in which the decrease in pressure across a stenosis is equal to 4v2, and the concept of the time-velocity integral (TVI), or “stroke distance” (Fig. 11).

2. PA diastolic pressure = 4 (PR end-diastolic velocity)2 + RA pressure 3. LA pressure = systolic BP − 4 (MR systolic velocity)2 4. RV systolic pressure = systolic BP − 4 (VSD velocity)2 where BP = blood pressure, LA = left atrium, MR = mitral regurgitation, PA = pulmonary artery, PR = pulmonary regurgitation, RA = right atrium, RV = right ventricle, TR = tricuspid regurgitation, and VSD = ventricular septal defect. P2

P1

Equation 11. Gradient (ΔP), mm Hg ΔP = 4(v22 − v12) or ΔP = 4v2 P = pressure v2 = accelerated velocity across a stenosis v1 = velocity proximal to a stenosis Note: Normally v1 is much smaller than v2 and can usually be omitted.Therefore, the equation can be simplified to 4v2 v = velocity across any vessel, chamber, or valve

When comparing Doppler-derived gradients with those measured invasively, it is important to remember that the maximal instantaneous gradient measured by Doppler is not equal to the peak-to-peak gradient measured at catheterization (Fig. 12). The maximal instantaneous gradient is always higher than the “nonphysiologic” (i.e., nonsimultaneous) peak-to-peak gradient. Doppler- and catheter-derived mean pressure gradients are comparable. By using the modified Bernoulli equation (Equation 11) and the measured Doppler velocity of a regurgitant or restrictive flow jet, the pressure difference between the two chambers can be calculated. If the pressure in one of the chambers can be measured accurately or estimated noninvasively, the pressure in the other chamber can be derived as shown in the following examples: 1. RV or PA systolic pressure = 4 (TR systolic velocity)2 + RA pressure

127

2 1/ 2

p (v22

v12 ) +

p 1

dv ds + R (v) dt

4v 2

Fig. 10. Derivation of the modified Bernoulli equation, which measures the pressure difference (P1 - P2) across a restrictive orifice. In most clinical situations, the viscous friction and flow acceleration components are negligible and can be ignored. If the proximal velocity (v1) is very small compared with the distal velocity (v2), as in severe aortic stenosis, the proximal velocity term can be omitted, resulting in the simplified equation ΔP = 4v2.

TVI (cm)

X

Area (cm 2)

=

ke Stro me volu L) A (m

Fig. 11. The time-velocity integral (TVI) is the calculated area under the Doppler spectrum over time. It is also known as “stroke distance” because it represents the distance (cm) that blood travels with each stroke or beat. The stroke volume (mL) is the volume of the cylinder formed by the product of the cross-sectional area (cm2) of the blood vessel or orifice and the distance (TVI) that the blood moves in a specified time period (i.e., systole or diastole).

128

Section II Noninvasive Imaging

Maximal instantaneous gradient

Peak-to-peak gradient

Equation 12. Regurgitant Volume, mL Regurgitant volume = SVvalve − SVsystemic SV

= stroke volume

SVsystemic = systemic flow measured elsewhere in an unaffected area of the heart (area × TVI)

Doppler peak velocity

Fig. 12. Schema of left ventricular (LV) and aortic (Ao) pressure tracings and the corresponding Doppler velocity spectrum demonstrating the difference between peak-topeak and maximal instantaneous gradients. The mean gradient (hatched area) is the area under the curve of the Doppler spectrum and is closely correlated with the mean gradient measured invasively (stippled area).

Right atrial pressure can be estimated by any one or a combination of techniques, including clinical estimate of central venous pressure, nomograms derived from Doppler catheter correlation studies, and echocardiographic estimates based on right atrial and inferior vena caval size and inferior vena caval reactivity to inspiratory effort. In practice, if the right atrium and inferior vena cava appear normal, 5 mm Hg is used for right atrial pressure estimates. If the inferior vena cava is mildly dilated or has blunted inspiratory collapse, 10 to 14 mm Hg is assumed. If the inferior vena cava is plethoric, has little or no inspiratory motion, or the clinical examination findings are consistent with marked increase of central venous pressure, 20 mm Hg or more is added to the pressure difference measured by Doppler echocardiography. Regurgitant volume and fraction (Equations 12 and 13) and Qp/Qs (Equation 14) are obtained by comparing the flow through a nonregurgitant reference valve with flow through the affected valve or chamber.

SVvalve

= flow volume (area × TVI) across the regurgitant valve (forward plus regurgiant flow)

TVI

= time-velocity integral

Equation 13. Regurgitant Fraction, % SVvalve − SVsystemic Regurgitant fraction = _________________ SVvalve (Definitions as in Equation 12)

Equation 14. Pulmonary-to-Systemic Flow Ratio (Qp/Qs) Qp ____________________ AreaPV × TVIPV ___ = Qs AreaLVOT × TVILVOT Qp = pulmonary stroke volume (usually measured at pulmonary valve anulus [PV]) Qs = systemic stroke volume (usually measured at left ventricular outflow tract [LVOT]) TVI = time-velocity integral The continuity equation, which is based on the principle of conservation of mass (“what goes in must come out”), states that flow proximal and distal to an orifice must be equal in a closed system (Equation 15). Rearrangement of the continuity equation allows calculation of stenotic and regurgitant orifice areas by

Chapter 8 Principles of Echocardiography measuring three variables and solving for the fourth (Equation 16).

Equation 15. Continuity Equation Flowproximal = Flowdistal A1 × TVI1 = A2 × TVI2 A1 × TVI1 = proximal flow A2 × TVI2 = flow across valve A1 = reference area A2 = area of the stenotic valve (cm2) TVI = time-velocity integral

Equation 16. Valve Area, cm2 (Rearrangement of continuity equation [Equation 15]) TVI1 A2 = A1 × _____ TVI2 A1 = reference area A2 = area of the stenotic valve (cm2) TVI = time-velocity integral

Mitral valve area can be measured with the continuity equation (Equation 15) or the pressure half-time method (Equations 17 and 18).

Equation 17. Pressure Half-Time (PHT), ms PHT = DT × 0.29 PHT = time required for the peak gradient to decrease by one-half DT = deceleration time (time [ms] from the maximal velocity to zero velocity) 0.29 = an algebraic constant that converts velocity to gradient

129

Equation 18. Mitral Valve Area (MVA) by Half-Time Measurement, cm2 220 759 MVA = _____ or ____ PHT DT 220 and 759 = empirical time constants equating to an MVA of approximately 1 cm2 (Definitions as in Equation 17)

The proximal isovelocity surface area (PISA) method (Fig. 13) is used most frequently in the context of quantifying mitral and aortic regurgitation. This method represents a variation of the continuity equation and uses the property of flow convergence of fluid as it approaches a restrictive orifice. Blood forms multiple concentric “shells” or “hemispheres” of isovelocity. As the surface area decreases, the velocity increases. The velocity at a given distance from the orifice (vr) can be measured by altering the aliasing velocity of the color flow Doppler signal. The flow rate through the orifice can be calculated (Equation 19). The effective regurgitant orifice (ERO), also referred to as regurgitant orifice area (ROA), and regurgitant volume can be calculated using the continuity equation and the peak velocity and TVI of the continuous-wave mitral regurgitant signal (Equations 20 and 21). Variations of the PISA technique also allow calculation of flow rate and volume and orifice area of stenotic mitral valves, atrial and ventricular septal defects, and aortic coarctation.

Equation 19. Proximal Isovelocity Surface Area (PISA) Flow Rate, mL/s Flow = 2 π2 × vr Flow = instantaneous flow rate (mL/s) r = radial distance of isovelocity shell from orifice (cm) vr = flow velocity radius r (cm/s)

130

Section II Noninvasive Imaging

PISA flow = MR flow 2πR2 × PISA V = ERO × MR V PISA

R

MR

2πR2 × Alias V = ERO × MR V 2πR2 × Alias V ERO = ______________ MR V 6.28 R2 × Alias V = ________________ MR V

A

B

Fig. 13. A, Diagram of proximal isovelocity surface area (PISA) (arrows) of mitral regurgitation. As blood flow converges toward the mitral regurgitant orifice, blood-flow velocity increases gradually and forms multiple isovelocity hemispheric shells. The flow rate calculated at the surface of the hemisphere is equal to the flow rate going through the mitral regurgitant orifice. Ao, aorta; LA, left atrium; LV, left ventricle. B, Calculation and derivation of effective regurgitant orifice (ERO) area of mitral regurgitation (MR) with the PISA method. R, PISA radius; V, velocity.

Equation 20. Effective Regurgitant Orifice (ERO) (cm2) for Quantifying Mitral Regurgitation Flow (mL/s) ERO = ___________ vMR (cm/s) vMR = peak velocity of continuous-wave mitral regurgitant signal

3. Mitral stenosis—transvalvular gradient, pressure half-time, and mitral valve area (Fig. 16) 4. Mitral regurgitation—regurgitant volume, fraction, and systolic flow reversals in pulmonary veins 5. Pulmonary artery pressure—tricuspid regurgitant velocity 6. Hypertrophic cardiomyopathy—left ventricular outflow tract gradient 7. Tricuspid regurgitation—systolic flow reversals in hepatic veins and marked dilated inferior vena cava and hepatic veins

Equation 21. Regurgitant Volume (mL) for Quantifying Mitral Regurgitation Regurgitant volume = ERO (cm2) × TVIMR (cm) ERO = effective regurgitant orifice TVIMR = time-velocity integral of continuouswave mitral regurgitant signal

For cardiology examinations, be able to identify the Doppler signals and assess the hemodynamic significance of the following conditions: 1. Aortic stenosis—transvalvular velocity, gradient, and aortic valve area by the continuity equation (Fig. 14) 2. Aortic regurgitation—pressure half-time and diastolic flow reversals in aorta (Fig. 15)

EVALUATION OF SPECIFIC DISORDERS Aortic Stenosis 1. M-mode/2D echocardiography—valve morphology (unicuspid, bicuspid, or tricuspid) and calcification. 2. Doppler echocardiography—peak aortic velocity, TVI, mean gradient (Fig. 14), and aortic valve area by the continuity equation (Equation 16). Severe aortic stenosis is usually present if the peak aortic velocity is 4.5 m/s or greater, the mean pressure gradient is 50 mm Hg or greater, the valve area is less than 0.80 cm2, or the left ventricular outflow tract–aortic valve TVI ratio is 0.25 or less. A small calculated aortic valve area associated with a low gradient and a

Chapter 8 Principles of Echocardiography

131

Fig. 14. Doppler signal obtained from the apical window in a patient with severe, symptomatic calcific aortic stenosis. LVOT vel = 1 m/s; LVOT TVI = 20 cm; LVOT diameter = 2.0 cm. By the continuity equation, the aortic valve area = 0.47 cm2. AV vel = 5 m/s; AV TVI = 135 cm; mean gradient across the aortic valve = 54 mm Hg. AV, aortic valve; LVOT, left ventricular outflow tract; TVI, time-velocity integral; vel, velocity.

A

B

C

D

Fig. 15. A, Holodiastolic reversal flow (arrows) in the descending aorta indicates severe aortic regurgitation. Similar diastolic reversal can be seen in a descending thoracic aneurysm or shunt into the aorta during diastole (as in Blalock-Taussig shunt). The sample volume usually is located just distal to the takeoff of the left subclavian artery. PA, pulmonary artery. B, Two-dimensional color flow imaging of the descending thoracic aorta during diastole. The orange-red flow in the descending aorta during diastole indicates flow toward the transducer, that is, reversal flow due to severe aortic regurgitation. Ao, aorta. C, Color M-mode from the descending thoracic aorta shows holodiastolic reversal flow (arrows). D, Pulsed-wave Doppler recording of abdominal aorta showing diastolic flow reversal (arrows) in severe aortic regurgitation.

132

Section II Noninvasive Imaging

Fig. 16. Continuous wave Doppler signal from a patient with severe mitral stenosis. Mean gradient is 16 mm Hg. Pressure half-time (t1⁄2) is 210 ms. Mitral valve area by pressure half-time method is 1.0 cm2.

low cardiac output state requires careful evaluation to differentiate decreased LV systolic function due to truly severe aortic stenosis from milder aortic stenosis and the presence of unrelated myocardial dysfunction. Dobutamine echocardiography has been used to increase contractility and to increase cardiac output to differentiate anatomical from “relative” aortic stenosis. The major pitfall in assessment of aortic stenosis is underestimation of the gradient and overestimation of the valve area when the highest velocity Doppler signal is not obtained because of technical or anatomical factors. When there is a discrepancy between clinical assessment and calculated valve area by transthoracic study, transesophageal echocardiography (TEE) may be required for more sensitive assessment of the valve morphology and degree of stenosis, and planimetry of the valve area can also be performed. Mitral Stenosis 1. M-mode/2D echocardiography—valve morphology, doming or “hockey stick” (long-axis view)

(Fig. 17),“fish mouth”(short-axis view), leaflet and subvalvular thickening, calcification and mobility (Abascal echocardiographic score), commissural anatomy, and left atrial size. 2. Doppler echocardiography—mean gradient, mitral valve area by continuity equation, pressure halftime, and planimetry methods (Equations 16-18); pulmonary artery pressure; and degree of mitral regurgitation. All three methods of echocardiographic assessment of mitral valve area correlate well with invasive measures, but each has unique features that render it more or less accurate in a given patient (Table 3). Therefore, all three methods should be performed to achieve an integrated approach to the severity of mitral stenosis. A high transvalvular gradient with normal pressure half-time may reflect severe mitral regurgitation rather than mitral stenosis. Severe mitral stenosis is usually present if the mitral valve area is 1.0 cm2 or less, the mean resting pressure gradient is 10 mm Hg or greater, or the pressure half-time is 220 ms or longer. Exercise Doppler echocardiography can be very useful to assess stress-induced changes in gradient, mitral valve area, and pulmonary artery pressures. TEE is essential before percutaneous mitral balloon valvuloplasty and can help define further the presence or absence of commissural fusion and calcification. The presence of heavy calcification at both commissures, significant subvalvular disease, and

Fig. 17. Rheumatic mitral stenosis with left atrial (LA) enlargement and obvious doming of the anterior mitral leaflet. LV, left ventricle; MV, mitral valve; RA, right atrium; RV, right ventricle.

Chapter 8 Principles of Echocardiography

133

Table 3. Limitations and Pitfalls in Assessing Mitral Valve Area Two-dimensional (2D) planimetry Dependent on 2D image quality, gain-setting, and ability to visualize the minimal orifice area Less accurate when extensive calcification is present Difficult after commissurotomy because of irregular orifice Doppler pressure half-time Tachycardia Nonlinear pressure decay Significant or acute aortic regurgitation increases the rate of increase in left ventricular pressure and shortens pressure half-time (mitral valve area overestimated) Immediately after percutaneous mitral valvuloplasty when hemodynamics are not stable (mitral valve area overestimated) Continuity equation Cumbersome to perform, multiple measurements are subject to error Mitral valve area underestimated when significant mitral regurgitation is present

marked leaflet thickening and immobility predict suboptimal results for valvuloplasty. Left atrial thrombus must be excluded to avoid embolic complications. Aortic Regurgitation 1. M-mode/2D echocardiography—valve morphology, LV size and function, premature mitral valve closure, diastolic opening of the aortic valve (severe aortic regurgitation), fluttering of the mitral valve, and etiology: Marfan syndrome, bicuspid aortic valve, endocarditis, and dissection. 2. Color flow imaging—ratio of jet width or area to LV outflow tract width or area (mild, 60%). 3. Pulsed wave Doppler echocardiography—holodiastolic flow reversals in the descending or abdominal aorta are indicative of significant regurgitation. 4. Continuous wave Doppler echocardiography— pressure half-time (mild, ≥400 ms; severe, ≤250 ms). High LV end-diastolic pressure can shorten pressure half-time, causing overestimation of the severity of regurgitation. 5. Quantitative methods—Regurgitant fraction (mild 55%) or regurgitant volume ≥60 mL; effective regurgitant orifice (mild, 1.0

Mild disease Moderate disease Severe disease

0.8-0.9 0.5-0.8 < 0.5

Post-exercise ABI No change or increase > 0.5 > 0.2 < 0.2

ABI, ankle:brachial systolic pressure index. *After treadmill exercise (1-2 mph, 10% grade, 5 minutes or symptom-limited) or active pedal plantar-flexion (50 repetitions or symptom-limited).

of chronic back pain or previous lumbosacral spinal surgery. The diagnosis of lumbar spinal stenosis can be confirmed by characteristic findings on computed tomography (CT) or magnetic resonance imaging (MRI) of the lumbar spine and electromyography (EMG) (Table 2), together with normal or minimally abnormal ABIs before and after exercise.

PSEUDOCLAUDICATION Pseudoclaudication is caused by lumbar spinal stenosis and is the condition most commonly confused with intermittent claudication. Pseudoclaudication is usually described as a “paresthetic” discomfort that occurs with standing and walking (variable distances). Symptoms almost always are bilateral and relieved by sitting and/or leaning forward. The patient often has a history

■

585

The discomfort of intermittent claudication is always exercise-induced.

586

Section V Aorta and Peripheral Vascular Disease

Table 2. Differential Diagnosis of True Claudication and Pseudoclaudication Feature

■

Claudication Pseudoclaudication

Onset

Walking

Character Bilateral Walking distance Cause Relief

Cramp, ache +/Fairly constant

Standing and walking “Paresthetic” + More variable

Atherosclerosis Spinal stenosis Standing still Sitting down, leaning forward

Pseudoclaudication is usually described as a “paresthetic” discomfort that occurs with standing as well as walking.

complications of coronary artery disease. Diabetes mellitus in combination with femoral artery disease results in a further decrease in overall survival and an increased incidence of major amputation. Patients with diabetes mellitus, particularly type II diabetes mellitus, have a distinct pattern and distribution of atherosclerosis in the lower extremity arteries.Compared to nondiabetics, there is less involvement of the aortoiliac segment, equal occurrence in the femoropopliteal segments, and more extensive disease in the infrapopliteal segments (tibial and peroneal arteries) (Fig. 1). Occlusive arterial disease in other locations (e.g., the subclavian artery) is also important, especially in patients being considered for a coronary artery bypass graft and in those who have recurrent angina after having a left internal mammary artery-left anterior descending coronary artery bypass. Stenting of subclavian stenoses may improve myocardial perfusion in patients with internal mammary artery grafts and flow-limiting proximal subclavian stenoses (i.e coronary steal). ■

NATURAL HISTORY OF PERIPHERAL VASCULAR DISEASE Peripheral arterial occlusive disease is associated with considerable mortality because of its association with coronary and carotid atherosclerosis. The 5-year mortality rate in patients with intermittent claudication is 29%, and the overall amputation rate over 5 years is 4%. More than half of patients have stable or improved symptoms over this same period. Continued use of tobacco results in a 10-fold increase in the risk for major amputation and a more than 2-fold increase in mortality. The effect of diabetes mellitus on patients with intermittent claudication deserves special mention, as it accounts for the majority of amputations in a community (12-fold increased risk of below-knee amputation and a cumulative risk of major amputation exceeding 11% over 25 years). Other clinical features that predict an increased risk of limb loss in lower extremity arterial occlusive disease include ischemic rest pain, ischemic ulceration, and gangrene. The location of occlusive arterial disease also affects overall prognosis. Patients with aortoiliac disease have a lower 5-year survival rate (73%) than those with predominantly femoral artery disease (80%). The increased mortality is attributable primarily to

■

The 5-year mortality rate in patients with intermittent claudication is 29%, predominantly due to associated coronary atherosclerosis. Tobacco use, diabetes mellitus, and critical limb

Fig. 1. Angiogram demonstrating the characteristic infrapopliteal location of arterial disease in a diabetic patient.

Chapter 46 Peripheral Vascular Disease

587

ischemia (ischemic rest pain, ulceration, and gangrene) are associated with an increased risk of limb loss in patients with intermittent claudication.

DIAGNOSIS OF PERIPHERAL VASCULAR DISEASE Peripheral angiography is not needed to diagnose intermittent claudication. The diagnosis is made clinically and confirmed by noninvasive testing (ABI before and after exercise). A ratio of the ankle systolic pressure to the brachial systolic pressure (ABI) at rest provides a measurement of disease severity in patients with lower extremity arterial occlusive disease: normal 1.0-1.4, mild disease 0.8-0.9, moderate disease 0.5-0.8, severe disease 70% stenosis), making precise identification of the source of embolism difficult.Treatment in symptomatic patients with no other identifiable source of embolism is surgical resection of the involved aorta, if the patient’s general medical condition permits. Oral anticoagulation treatment for 3 months on the presumption that the friable components will have organized, followed by antiplatelet therapy, is an alternative. ■

Many patients with cerebral ischemic events and protruding and mobile atheromas of the thoracic aorta have coexistent carotid artery disease.

SPONTANEOUS DISSECTION OF CEPHALIC ARTERIES Spontaneous dissection of the cervical cephalic arteries is uncommon but important for two reasons: 1) the clinical presentation is characteristic—either hemicrania with oculosympathetic paresis (Horner’s syndrome) or hemicrania with delayed focal cerebral ischemic symptoms and 2) the prognosis is good for recovery, and recurrences are rare.

Fig. 2. Transesophageal echocardiography in a

patient with recurrent left hemispheric transient ischemic attacks. Transverse view of the distal transverse aortic arch demonstrates large mobile thrombus (short arrow) adjacent to origin of the left common carotid artery (LCA).

48 THE AORTA Peter C. Spittell, MD

AORTIC ATHEROEMBOLISM Microemboli or macroemboli from atherosclerotic plaque and thrombus in the aorta are important causes of cerebral and systemic embolization. Cerebral atheroembolism suggests the source of embolic material is intracardiac or is in the ascending aorta and/or transverse aortic arch. Lower extremity atheroembolism is caused most commonly by abdominal aortic aneurysm or diffuse atherosclerotic disease. Unilateral blue toes suggest that the embolic source is distal to the aortic bifurcation. Atheroembolism is characterized by livedo reticularis, blue toes, palpable pulses, hypertension, renal insufficiency, increased erythrocyte sedimentation rate, and eosinophilia (transient) (Fig. 1). Atheroembolism can occur spontaneously or be due to medication (warfarin or thrombolytic therapy), or to angiographic or surgical procedures. Thoracic aortic atherosclerotic plaque is most accurately assessed with TEE. Plaque thickness more than 4 mm or mobile thrombus (of any size) are associated with an increased risk of embolism (Fig. 2). Severe aortic atherosclerosis is present in approximately 27% of patients with previous embolic events and is also a strong predictor of coronary artery disease. Antiplatelet agents and a statin medication should be used in all patients with aortic embolic events unless there are absolute contraindications. Warfarin therapy

Fig. 1. Livedo reticularis over both patellae and multiple blue toes in a patient with atheroembolism from an abdominal aortic aneurysm.

601

602

Section V Aorta and Peripheral Vascular Disease

Fig. 3. Gross specimen showing isolated intimal tear of ascending aorta in giant cell aortitis. Fig. 2. Transesophageal echocadiography demonstrates advanced immobile atherosclerosis in the descending thoracic aorta.

may be beneficial for reducing subsequent embolic events, but may also exacerbate embolism in some patients, and further randomized trials are required. The treatment of choice is to identify the source of embolism and, if possible, to surgically resect it. ■

■

Emboli from atherosclerotic plaque and thrombus in the thoracic aorta are important causes of stroke and peripheral emboli. Atheroembolism is characterized by livedo reticularis, blue toes, palpable pulses, hypertension, renal insufficiency, increased erythrocyte sedimentation rate, and eosinophilia (transient).

THORACIC AORTIC ANEURYSMS Thoracic aortic aneurysms are caused most commonly by atherosclerosis, but they also occur in patients with systemic hypertension, Marfan syndrome, bicuspid aortic valve, giant cell arteritis (Fig. 3 and 4) (cranial and Takayasu disease [Fig. 5]), and infections (syphilis), and as a result of trauma. Most thoracic aortic aneurysms are asymptomatic and are discovered incidentally on chest radiography. Computed tomography (CT), magnetic resonance imaging (MRI), and transesophageal echocardiography (TEE) are all accurate noninvasive techniques for imaging thoracic aortic aneurysms (Fig. 6). Indications for surgical resection

include the presence of symptoms attributable to the aneurysm, an aneurysm rapidly enlarging under observation (particularly if the patient has hypertension), posttraumatic aneurysm, pseudoaneurysm, and an aneurysm 6 cm or greater in diameter (5.5-6 cm in low-risk patients). In patients with Marfan syndrome, surgery is usually indicated when the ascending aortic diameter is between 4.5 and 5 cm.

ABDOMINAL AORTIC ANEURYSM Abdominal aortic aneurysm can be diagnosed reliably with ultrasonography, CT, or MRI (Fig. 7 and 8). Angiography is not required unless the renal or peripheral arterial circulation needs to be visualized to plan treatment. Because physical examination in the detection of AAA lacks sensitivity, screening tests are indicated in high-risk subsets of patients. It has been shown that early detection of abdominal aortic aneurysm can reduce mortality, furthermore, a single screening ultrasound of men >65 years of age can identify the majority of abdominal aortic aneurysms. The United States Preventive Services Task Force recommends a one-time screening ultrasound in men age 65 to 75 years who have ever smoked. Screening of siblings and first-degree relatives of patients with aneurysm generally begins at age 50 years. In a goodrisk patient, elective surgical treatment of an abdominal aortic aneurysm should be considered for aneurysms greater than 54 cm in diameter. Elective surgical repair is definitely indicated when the aneurysm diameter is

Chapter 48 The Aorta

603

Fig. 4. Histologic specimen showing isolated intimal tear of ascending aorta in giant cell aortitis.

between 5.5 cm and 6.0 cm in good-risk patients. In patients with significant comorbid conditions (pulmonary, cardiac, renal, or liver disease), surgical therapy is individualized. In persons with a large and/or symptomatic abdominal aortic aneurysm whose comorbid

condition makes them poor surgical candidates, exclusion of the aneurysm by an endovascular approach (placement of an intraluminal stent-anchored polyethylene terephthalate fiber [Dacron] prosthetic graft via retrograde transfemoral cannulation under local anesthesia) has given encouraging results. Percutaneous abdominal aortic aneurysm repair is a safe and effective treatment compared with open surgical repair for infrarenal abdominal aortic aneurysms with appropriate anatomy. Results are comparable to surgical repair with regard to mortality and may have improved shortterm and long-term morbidity rates. All patients treated with endovascular repair need continued lifelong follow-up with tomographic imaging. ■

Fig. 5. Takayasu arteritis in descending thoracic aorta.

Elective surgical repair is definitely indicated when aneurysm diameter is greater than 5.0 cm in goodrisk patients.

An inflammatory abdominal aortic aneurysm is suggested by the triad of back pain, weight loss, and increased erythrocyte sedimentation rate. Obstructive uropathy may occur with ureteral involvement. The findings on CT are diagnostic (Fig. 9).The treatment is surgical resection. Surgery is also indicated for abdominal aortic

604

Section V Aorta and Peripheral Vascular Disease

A

B

C

Fig. 6. Magnetic resonance imaging/angiography in a patient with asymptomatic thoracic aortic aneurysm. Images in the transverse, A, and longitudinal, B, planes demonstrate a large aneurysm of the ascending aorta (7.8 cm) and moderate dilatation (4.5 cm) of the descending thoracic aorta. Moderate aortic regurgitation is also demonstrated, C.

AORTIC DISSECTION

aneurysms that are symptomatic, traumatic or infectious in origin, or are rapidly expanding (>0.5 cm/year). ■

■

An inflammatory abdominal aortic aneurysm is suggested by the triad of back pain, weight loss, and increased erythrocyte sedimentation rate. Surgery is indicated for abdominal aortic aneurysms that are greater than 5 cm in diameter, symptomatic, traumatic, or infectious in origin or are rapidly expanding (>0.5 cm/year).

Etiology The most common predisposing factors for aortic dissection are advanced age, male gender, hypertension, Marfan syndrome, and congenital abnormalities of the aortic valve (bicuspid or unicuspid valve). When aortic dissection complicates pregnancy, it usually occurs in the third trimester. Iatrogenic aortic dissection, as a result of cardiac surgery or invasive angiographic procedures, can also occur. Classification Aortic dissection involving the ascending aorta is designated as type I or type II (proximal, type A), and dissection confined to the descending thoracic aorta is designated as type III (distal, type B) (Fig. 10 and 11).

Chapter 48 The Aorta

Fig. 7. Computed tomography with intravenous contrast demonstrating a large aneurysm of the infrarenal abdominal aorta. There is a small amount of laminated thrombus within the aneurysm and dense peripheral calcification.

605

Fig. 9. Computed tomography scan of abdomen of patient with inflammatory abdominal aortic aneurysm. Note the high attenuation change surrounding the aorta, representing inflammatory change in periaortic retroperitoneal tissue.

Clinical Features The acute onset of severe pain (often migratory) in the anterior chest, back, or abdomen is the most suggestive clinical finding (sensitivity of 90% and specificity of 84%). Additional findings include hypertension (49% of patients), an aortic diastolic murmur (28% of patients), pulse deficits or blood pressure differential (31% of patients), and neurologic changes (17% of

Type A (proximal)

Fig. 8. Aneurysms of abdominal aorta and iliac artery.

Type III

{

Type II

{

Type I

Type B (distal)

Fig. 10. Classification of aortic dissection.

606

Section V Aorta and Peripheral Vascular Disease

Fig. 11. Type III aortic dissection involving the abdominal aorta.

patients). Syncope in association with aortic dissection occurs when there is rupture into the pericardial space, producing cardiac tamponade. Congestive heart failure is due most commonly to severe aortic regurgitation. Acute myocardial infarction (most commonly inferior infarction due to right coronary artery ostial dissection) and pericarditis are additional cardiac presentations. Clues to type I aortic dissection include substernal pain, aortic valve incompetence, decreased pulse or blood pressure in the right arm, decreased right carotid pulse, pericardial friction rub, syncope, ischemic electrocardiographic changes, and Marfan syndrome. Clues to type III aortic dissection include interscapular pain, hypertension, and left pleural effusion. ■

In a patient with a catastrophic presentation, systemic hypertension, and unexplained physical findings of vascular origin—especially in the presence of chest or back pain and an aortic murmur—aortic dissection should always be included in the differential diagnosis, and an appropriate screening test should be performed emergently.

Laboratory Tests Chest radiography may reveal widening of the mediastinum and supracardiac aortic shadow, deviation of the trachea to the right, a discrepancy in diameter between the ascending and descending aorta, and pleural effusion

(Fig. 12). Normal findings on chest radiography do not exclude aortic dissection. Diagnosis Definitive diagnosis of aortic dissection can be made using any of the following imaging modalities: echocardiography, CT, MRI, and aortography. Echocardiography The combination of transthoracic echocardiography (TTE) and TEE can be used to identify an intimal flap, communication between the true and false lumina, a dilated aortic root (>4.2 cm), thrombus formation, widening of the aortic walls, aortic regurgitation, and pericardial effusion/tamponade. Multiplane techniques have markedly improved the accuracy of TEE (Fig. 13). Advantages of TEE include portability, safety, accuracy, rapid diagnosis, use in patients with hemodynamic instability, and use intraoperatively. CT CT can accurately detect the intimal flap, identify two lumina, demonstrate displaced calcification, pericardial effusion, pleural effusion, and abdominal aorta involvement, and provide accurate aortic diameters (Fig. 14). The disadvantages include nonportability (limiting its use in patients with hemodynamic instability) and the need for intravenous contrast agents.

Chapter 48 The Aorta

A

607

B

Fig. 12. Chest radiographs of patient before (A) and after (B) aortic dissection. Note widening of superior mediastinum after aortic dissection (arrow).

MRI MRI is as accurate as CT in the diagnosis of aortic dissection, although MRI, with its inherent multiplanar imaging capability, can be used with or without contrast enhancement. Demonstration of the intimal flap, entry/exit sites, thrombus formation, aortic regurgitation, pericardial effusion, pleural effusion, and abdominal aorta and branch vessel involvement is possible (Fig. 15). Disadvantages of MRI include cost and nonportability.

Aortography Aortography can accurately diagnose aortic dissection by showing the intimal flap, opacification of the false lumen, and deformity of the true lumen. Also, associated aortic regurgitation and coronary artery anatomy can be visualized. The disadvantages include invasive risks, exposure to intravenous contrast agents, delay in diagnosis, and nonportability. The choice of test (TTE, TEE, CT, MRI, or aor-

Fig. 13. Transesophageal echocardiographic view of ascending aorta in the longitudinal plane demonstrating an intimal flap originating in right coronary sinus.

Fig. 14. Computed tomography with intravenous contrast demonstrating dilatation of descending thoracic aorta and an intimal flap. Note the relatively equal opacification of true and false lumina.

608

Section V Aorta and Peripheral Vascular Disease

Fig. 15. Magnetic resonance angiography demonstrating dissection of mid and distal abdominal aorta in patient with a remote history of sudden deceleration injury.

tography) in a patient with suspected acute aortic dissection depends on which is most readily available and the hemodynamic stability of the patient. Currently, our test of choice in suspected acute aortic dissection is the combination of TTE and TEE. The initial management of suspected acute aortic dissection is shown in Figure 16. The most common cause of death in aortic dissection is rupture into the pericardial space, with cardiac tamponade. Echocardiographically guided pericardiocentesis is associated with an increased risk of aortic rupture and death. Cardiac tamponade due to aortic dissection is a surgical emergency, and generally peri-

Fig. 16. Initial management of suspected acute aortic dissection. Angio, angiography; CT, computed tomography; CXR, chest radiography; ECG, electrocardiography; TEE, transesophageal echocardiography; TTE, transthoracic echocardiography.

cardial fluid should be removed only in the operating room after cardiopulmonary bypass has been instituted. Other causes of death include acute congestive heart failure due to severe aortic regurgitation, rupture through the aortic adventitia, rupture into the left pleural space, and occlusion of vital arteries. Factors that propagate dissection include impulse pulsatile flow and increased mean arterial pressure. ■

Cardiac tamponade due to aortic dissection is a surgical emergency, and pericardial fluid should be removed only in the operating room after cardiopulmonary bypass has been instituted.

Chapter 48 The Aorta Treatment Pharmacologic therapy should be instituted as soon as the diagnosis of aortic dissection is suspected (see below). Emergent surgery is indicated for types I and II (proximal, type A) aortic dissection. Coronary angiography prior to surgery is not indicated as it does not improve survival. Pharmacologic therapy in the coronary care unit is the preferred initial treatment for type III (distal, type B) aortic dissection, with delayed surgical therapy (2 to 3 weeks) for selected patients whose general medical condition permits operation. The initial pharmacologic therapy is outlined in Table 1. When long-term pharmacologic therapy is used for type III aortic dissection, indications for surgery include development of saccular aneurysm, increasing aortic diameter, or symptoms related to chronic dissection.

Table 1. Initial Pharmacologic Therapy for Aortic Dissection Hypertensive patients Sodium nitroprusside intravenously, 2.5 μg/kg per minute with Propranolol intravenously, 1 mg every 4 to 6 hours (The goal is to have systolic blood pressure 3.5 supports the diagnosis of significant RAS. The calculated resistive index is one variable that has often been used to determine the likelihood of benefit with renal revascularization: a resistive index of >0.8 suggests renal parenchymal disease which would not be expected to improve with revascularization. Additionally, the literature suggests the kidneys 30 mm Hg. Cardiac catheterization confirms increased pulmonary artery and right atrial pressures. Pulmonary capillary pressure is normal. The treatment of HIV-associated pulmonary hypertension is similar to that of patients with primary pulmonary arterial hypertension: options include pulmonary vasodilators (epoprostenol), calcium channel blockers, oral anticoagulation therapy, diuretics, and sildenafil. Responses have been variable. The effect of antiretroviral therapy on the course of HIV-related pulmonary hypertension is controversial. Some studies have shown a benefit, others have failed to do so.

Worsening of the clinical course with antiretroviral therapy has also been reported. The prognosis of HIV-related pulmonary hypertension is poor with a median interval from the diagnosis of pulmonary hypertension to death of 6 months.

ENDOCARDIAL DISEASE A common incidental finding at autopsy in patients dying of AIDS is thrombotic nonbacterial marantic endocarditis, a condition in which sterile valvular vegetations occur without an infectious cause. Systemic embolization is an uncommon clinical presentation of marantic endocarditis: valve destruction or clinical valve dysfunction is rare. Infective endocarditis in HIV-infected patients is uncommon and occurs almost exclusively in intravenous drug users. In a retrospective review of infective endocarditis in HIV-infected patients between 1979 and 1999 at a tertiary care hospital, only 8 out of 599 cases of infective endocarditis were diagnosed in non intravenous drug users. In another review, infective endocarditis was responsible for 5-20% of hospital admissions and for 5-10% of total deaths in intravenous drug using patients with HIV infection. The clinical presentation was similar to what has been observed in HIV-negative patients.The clinical outcome of the patients appeared to depend more on the affected valve and the causative organism rather than the HIV serostatus of the patient. In intravenous drug users, the most common valve involved is the tricuspid valve and the most common causative organism, Staphylococcus aureus. The microbiologic spectrum in non-intravenous drug use-related infective endocarditis in HIV-infected patients is wide and includes unusual organisms such as Salmonella, Aspergillus, Cryptococcus, and Candida species in addition to the usual bacteria causing endocarditis in the general population.

CARDIAC TUMORS Kaposi's sarcoma as seen in patients with AIDS, can involve the myocardium and pericardium and classically presents with pericardial effusion or less commonly cardiac tamponade. Primary cardiac lymphoma is a rare malignancy associated with AIDS that presents with heart failure or ventricular arrhythmias secondary to

Chapter 81 HIV Infection and the Heart

981

diffuse infiltration of the ventricular wall or less commonly mechanical obstruction to valve function due to localized nodules or intracavitary masses. Surgery, chemotherapy and radiation are generally palliative (Figs. 3 and 4).

CORONARY ARTERY DISEASE With continued use of HAART and longer survival of HIV-infected patients, a number of metabolic complications of HIV infection and its treatment have been observed.These include dyslipidemias, insulin resistance, hyperglycemia, and body composition changes (lipodystrophy, lipoatrophy).The appreciation of these metabolic disorders associated with antiretroviral therapy has led to a growing concern about a possible increased risk for cardiovascular disease. Results from studies that have attempted to analyze this risk have not always been consistent but the weight of evidence suggests that there is a link between antiretroviral therapy, particularly the use of protease inhibitors, and an increased risk for coronary artery disease. In an analysis of data from a cohort of 5,672 outpatients with HIV-1 at nine US HIV clinics, the use of protease inhibitors (PI) was associated with an increased risk of myocardial infarction. In a study of 19,795 HIV-infected French men receiving a PI-based regimen, morbidity ratios were greater with longer exposure to PIs. Similarly, receipt of a PIbased regimen was associated with myocardial infarction after adjustment for age in the Frankfurt HIV Cohort.The D/A/D study, a large international collaborative observational study, evaluated the incidence of myocardial infarction (MI) in 23,400 HIV-infected

Fig. 3. Primary cardiac lymphoma.

Fig. 4. Lymphoma cells in the myocardium.

patients from 11 cohorts in Europe, Australia, and the United States. The risk for MI rose progressively with the number of years on combination antiretroviral therapy (adjusted relative risk [RR] 1.16/year of exposure [95% CI 1.09 to 1.23]). Increased PI exposure was associated with an increased risk of MI, which is partly explained by dyslipidemia. Age, male sex, a past history of cardiovascular disease, smoking, elevated total cholesterol level at baseline and the presence of diabetes mellitus at baseline were also independent predictors for MI. Contrary to these results, data from a large cohort of HIV-infected patients (n = 36,766) in the Veterans Affairs Health System did not indicate increases in myocardial infarction related hospitalizations, despite the substantial use of PIs. The risk of HAART-related cardiovascular disease is outweighed by the benefits of antiretroviral therapy and should not be a reason to withhold therapy. While consideration of dyslipidemia and cardiovascular risk is appropriate in the construction of an antiretroviral regimen, virologic suppression remains the overriding goal of antiretroviral therapy. Current guidelines recommend that HIV-infected adults undergo evaluation and treatment on the basis of the National Cholesterol Education Program Adult Treatment Panel (NCEP ATP) guidelines for dyslipidemia. Nonpharmacologic interventions such as dietary counseling, exercise, and modification of other risk factors (e.g. smoking) should generally be attempted first before instituting drug therapies. Pharmacologic therapy of dyslipidemias usually includes treatment with statins with or without fenofibrates. Protease inhibitors are in general sub-

982

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

strates as well as inhibitors of cytochrome P450. The primary route of metabolism for most statins is also cytochrome P450. Thus a significant potential for drug interaction exists. Simvastatin and lovastatin should not be used in patients taking PIs or the nonnucleoside

reverse transcription inhibitor (NNRTI) delavirdine. Atorvastatin, fluvastatin, pravastatin, and rosuvastatin appear to be safe for use with PIs.Fibrates are metabolized by glucuronidation and thus do not present a significant potential for drug interaction.

82 INFECTIVE ENDOCARDITIS Robin Patel, MD Joseph G. Murphy, MD James M. Steckelberg, MD

NATIVE VALVE INFECTIVE ENDOCARDITIS

Men are affected 2.5 times more commonly than women by infective endocarditis. Chronic rheumatic valvular disease has been supplanted by mitral valve prolapse with mitral regurgitation and degenerative aortic valve disease as the leading cardiac conditions predisposing to bacterial endocarditis in adults. Nosocomial endocarditis associated with therapeutic interventions (intravenous catheters, hyperalimentation lines, pacemakers, dialysis shunts, etc.) is increasing in frequency. A high proportion of cases of right-sided endocarditis occur in intravenous drug users.

Epidemiology There are about 15,000 new cases of infective endocarditis annually in the U.S. The annual number of cases of endocarditis has remained relatively constant, but the spectrum of underlying cardiac conditions and the etiologic organisms have changed over time.The median age of patients with infective endocarditis has increased, and currently, about one-half of all patients are older than 60 years, with the median age of those with enterococcal endocarditis even higher. Infective endocarditis is rare in children and is usually associated with underlying structural congenital heart disease, surgical repair of congenital heart disease, or nosocomial catheter-related bacteremia, especially in infants. Complex congenital heart disease and unrepaired ventricular septal defect are the most common underlying structural cardiac lesions in children. ■

Predisposing Heart Lesions for Endocarditis The heart valve most commonly involved in infective endocarditis is the mitral valve (Fig. 1), followed by the aortic valve. Isolated aortic valve endocarditis is more common in men than women and often is associated with congenitally bicuspid aortic valve.The mitral valve is infected in more than 85% of cases of infective endocarditis that follow damage by rheumatic fever. The tricuspid valve is typically involved in intravenous drug users, while the pulmonary valve is rarely infected. Both right- and left-sided endocarditis may occur simultaneously (Fig. 1).

Complex congenital heart disease and unrepaired ventricular septal defect are the most common underlying cardiac lesions predisposing to infective endocarditis in children.

We would like to acknowledge Walter R. Wilson, MD, for the generous provision of patient photographs in Figure 1 and Figures 3-14.

983

984

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation transplantation, including Corynebacterium, Candida species, and Aspergillus flavus endocarditis in liver transplant recipients, cytomegalovirus and Staphylococcus epidermidis endocarditis in heart transplant recipients, and Staphylococcus aureus and Candida albicans endocarditis in heart-lung transplant recipients. Other conditions associated with an increased incidence of infective endocarditis include poor dental hygiene, long-term hemodialysis, diabetes mellitus, and infection with the human immunodeficiency virus (HIV).

Fig. 1. Infective endocarditis involving the mitral valve.

■

The heart valve most commonly affected by endocarditis is the mitral valve, followed by the aortic, tricuspid, and pulmonary valves.

High Risk Congenital heart lesions include patent ductus arteriosus, ventricular septal defect, coarctation of the aorta, bicuspid aortic valve, tetralogy of Fallot, and, rarely, pulmonic stenosis. Surgical closure of ventricular septal defect decreases the risk of infective endocarditis provided no residual shunt is present. Endocarditis is extremely rare in secundum atrial septal defects. Patients with hypertrophic cardiomyopathy (HCM) are at increased risk for infective endocarditis, especially those with hemodynamically severe forms of the disease (high peak systolic pressure gradient and markedly symptomatic patients). Either the mitral or aortic valve, or both valves, may be infected. ■

■

Endocarditis is extremely rare in secundum atrial septal defects. Patients with HOCM and high gradients are at increased risk for endocarditis.

Infective endocarditis is associated with mitral valve prolapse, especially in patients with marked valvular redundancy, valve leaflet thickening, or significant mitral regurgitation. Unusual organisms may cause endocarditis after immunosuppressed patients following in solid organ

■

Mitral valve prolapse with regurgitation and degenerative aortic valve disease are the most common predisposing lesions for infective endocarditis in adults.

Pathogenesis The development of infective endocarditis (Fig. 2) requires that the valve surface must first be damaged to uncover a suitable site for bacterial attachment and colonization. Typically, this damage is caused by blood flow turbulence, which leads to the deposition of platelets and fibrin and the formation of “nonbacterial thrombotic endocarditis.” Hemodynamic factors contribute to the localization of these lesions downstream from a regurgitant flow, characteristically on the atrial surface of the mitral valve and on the ventricular surface of the aortic valve. Lesions with high degrees of turbulence (small ventricular septal defects with jet lesions, valvular stenoses) valve regurgitation readily create conditions that lead to bacterial colonization, whereas defects with large surface areas (large ventricular septal defects), low flow (ostium secundum atrial septal defects), or attenuation of turbulence (congestive heart failure) are rarely implicated in infective endocarditis. ■

Infective endocarditis characteristically occurs on the atrial surface of the mitral valve and on the ventricular surface of the aortic valve.

After the formation of the nonbacterial thrombotic media, bacteria colonize the lesion. Transient bacteremia typically occurs when a mucosal surface heavily colonized with bacteria is traumatized, as with dental procedures and with gastrointestinal, urologic, and gynecologic procedures. The degree of bacteremia is proportional to the trauma produced by the procedure

Chapter 82 Infective Endocarditis Valve or vascular endothelium Trauma Turbulence Metabolic changes

985

Mucous membranes or other colonized tissue Trauma Intravenous drug use Intravascular lines

Endothelial damage Platelet, fibrin deposition (biofilm) Bacteremia Nonbacterial thrombotic endocarditis Adherence Colonization Bacterial division Fibrin deposition Platelet aggregation Protection from neutrophils Protection from antibiotics Mature infected vegetation

Valve ring abscess

Valve incompetence

Valve perforation

Fig. 2. Pathogenesis of infective endocarditis.

and the number of organisms inhabiting the area. Most of the cases of bacterial endocarditis cannot be attributed to an identifiable invasive procedure. Certain species and strains of bacteria appear to have a selective advantage in adhering to platelets, fibrin, or valvular endothelium and/or formation of biofilm. Biofilm is particularly important in the context of foreign bodies (e.g., prosthetic heart valves, cardiovascular devices). After colonization of the valve occurs and a critical mass of adherent bacteria develops, the vegetation enlarges by additional deposition of platelets and fibrin and continued proliferation of bacteria. ■

Less than one-half of the cases of bacterial endocarditis can be attributed to an identifiable invasive procedure.

antibodies may also occur in infective endocarditis and contribute to musculoskeletal manifestations, low-grade fever, and pleuritic pain. Circulating immune complexes are found with increased frequency in connection with a long duration of illness, extravalvular manifestations, hypocomplementemia, and rightsided infective endocarditis. Concentrations decrease with successful therapy. Patients with infective endocarditis and circulating immune complexes may develop a diffuse glomerulonephritis. Immune complexes and complement are deposited subepithelially along the glomerular basement membrane to form a “lumpy-bumpy” pattern. Some of the peripheral manifestations of infective endocarditis, such as Osler nodes, may also result from the deposition of circulating immune complexes. ■

Infective endocarditis stimulates both humoral and cellular immunity. Rheumatoid factor develops in about one-half of the patients with infective endocarditis of more than 6 weeks’ duration. Antinuclear

The vascular endothelium is damaged most often by turbulent blood flow; platelets and fibrin are deposited; the nonbacterial thrombotic endocarditis lesion is seeded during a bacteremic episode, and a mature vegetation develops.

986

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Pathologic Changes Pathologic changes of infective endocarditis are shown in Table 1. The classic vegetation is located along the line of closure of the valve leaflet. Vegetations may be single or multiple, vary from a few millimeters to several centimeters in size. With treatment, healing occurs by fibrosis and, occasionally, calcification. In acute endocarditis, the vegetation is larger, softer, and more friable. Infection may lead to perforation of the valve leaflet or rupture of the chordae tendineae, interventricular septum, or papillary muscle. Endocarditis, especially S.

aureus endocarditis, may produce valve-ring abscesses, with fistula formation into the myocardium or pericardial sac. Aneurysms of the valve leaflet or sinus of Valsalva are also seen. Myocardial abscesses are associated with S. aureus endocarditis,high fever,rapid onset of congestive heart failure, and conduction disturbances. Embolic phenomena are common in infective endocarditis; major embolic episodes occur in at least one-third of patients. Most emboli occur before diagnosis, the incidence of emboli falls once effective antimicrobial therapy begins. Embolic phenomena most frequently involve

Table 1. Pathologic Findings in Infective Endocarditis Location

Manifestation

Central nervous system

Cerebral emboli Cerebral infarction Arteritis Abscess Mycotic aneurysm Intracerebral or subarachnoid hemorrhage Cerebritis Meningitis Splenic infarct Splenic abscess Splenic enlargement

Spleen

Lung

Skin

Pulmonary emboli Pleural effusion Empyema Petechiae Osler nodes

Janeway lesions

Eye

Roth spots

Comment

Pathologic findings include hyperplasia of lymphoid follicles, proliferation of reticuloendothelial cells, scattered focal necrosis Associated with right-sided infective endocarditis

Consist of arteriolar intimal proliferation with extension to venules and capillaries, which may be accompanied by thrombosis and necrosis; diffuse perivascular infiltrate consisting of neutrophils and monocytes surrounds dermal vessels; immune complexes may be seen in dermal vessels Consist of bacteria, neutrophilic infiltration, necrosis, and subcutaneous hemorrhage; secondary to septic emboli; subcutaneous abscesses on histologic examination Consist of lymphocytes surrounded by edema and hemorrhage in nerve fiber layer of retina

Chapter 82 Infective Endocarditis

987

the cerebral, renal, splenic, or coronary circulation. When large emboli occlude major vessels, consider fungal endocarditis, marantic endocarditis, or an intracardiac myxoma. ■

■

Circulating immune complexes have been found in high titer in virtually all patients with infective endocarditis. S. aureus is frequently associated with valve-ring abscesses and myocardial abscess.

Renal Complications of Infective Endocarditis Abscess, infarction, or glomerulonephritis may be found in the kidney in infective endocarditis (Fig. 3 and 4). Glomerulonephritis may be a focal, local, or segmental process characterized by endothelial and mesangial proliferation, hemorrhage, neutrophilic infiltration, fibrinoid necrosis, crescent formation, and healing by fibrosis. Diffuse glomerulonephritis, consisting of generalized cellular hyperplasia in all glomerular tufts, may also be seen. Less commonly, membranoproliferative glomerulonephritis, characterized by marked mesangial proliferation and by splitting of the glomerular basement membrane, may be found. Fig. 4. Microscopic appearance of kidney in S. aureus endocarditis.

■

Fig. 3. Gross appearance of kidney in S. aureus endocarditis.

Glomerulonephritis may be a focal, local, or segmental process.

Mycotic Aneurysms Mycotic aneurysms may develop during active infective endocarditis, rarely months to years after successful therapy. They may arise by 1) direct bacterial invasion of the arterial wall, with subsequent abscess formation or rupture; 2) septic or bland embolic occlusion of the vasa vasorum; or 3) immune complex deposition, with resultant injury to the arterial wall. Mycotic aneurysms tend to occur at bifurcation points and are found in the cerebral vessels (especially the peripheral branches of the middle cerebral artery), but they also occur in the abdominal aorta, sinus of Valsalva, ligated patent ductus arteriosus, and in splenic, coronary, pulmonary, and superior mesenteric arteries. Importantly, mycotic aneurysms often are silent clinically until rupture occurs.

988

■

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Central nervous system mycotic aneurysms are usually silent clinically until rupture occurs.

Clinical Manifestations Clinical manifestations of infective endocarditis are listed in Table 2. Fever is the most common manifestation but is not present in elderly patients who have blunted febrile responses. Prolonged fever is associated with S. aureus, gram-negative bacilli, fungi, culturenegative endocarditis, embolization of major vessels, myocardial abscess, tissue infarction, pulmonary emboli, drug fever, and nosocomial infection. In patients with prolonged fever, abdominal computed tomography (CT) may be helpful to rule out splenic abscess, and transesophageal echocardiography, to exclude valve-ring abscess.

Table 2. Clinical Manifestations of Infective Endocarditis Symptom Fever Chills Weakness Dyspnea Sweats Anorexia Weight loss Malaise Cough Skin lesions Stroke Nausea Vomiting Headache Myalgia Arthralgia Edema Chest pain Abdominal pain Delirium Coma Hemoptysis Back pain

Physical finding Fever Heart murmur Changing murmur New murmur Embolic phenomena Skin manifestations--Osler nodes, petechiae, Janeway lesions Splenomegaly Septic complications-pneumonia, meningitis, etc. Mycotic aneurysms Clubbing Retinal lesions Signs of renal failure Arthritis--reactive or infectious Mucosal petechiae (e.g., palate, conjunctiva) Splinter hemorrhages

Heart murmurs occur in most patients but may be absent with right-sided endocarditis. The classic changing murmur or new murmur is uncommon but may be seen with acute staphylococcal disease. More than 90% of patients who demonstrate a new regurgitant murmur will develop congestive heart failure (the leading cause of death in infective endocarditis). Pericarditis is rare, and when present, it is usually accompanied by myocardial abscess formation as a complication of S. aureus infection. ■ ■

Prolonged fever is associated with S. aureus. More than 90% of patients who demonstrate a new regurgitant murmur will develop congestive heart failure.

Peripheral Lesions in Endocarditis Osler nodes are small, painful nodular lesions usually found on the pads of the fingers or toes and occasionally on the thenar eminence (Fig. 5). They range in size from 2 to 15 mm and are frequently multiple. They disappear in a matter of hours to days. Janeway lesions are hemorrhagic, macular, painless plaques with a predilection for the palms or soles (Fig. 6). They persist for several days and are thought to be embolic in origin. Roth spots are oval pale retinal lesions surrounded by hemorrhage (Fig. 7). They are usually near the optic disk. They are not specific for infective endocarditis. Musculoskeletal manifestations of infective endocarditis include proximal arthralgias, lower extremity mono- or oligoarticular arthritis, low back pain, and diffuse myalgias. Splinter hemorrhages and palatal or conjunc-

Fig. 5. Osler node.

Chapter 82 Infective Endocarditis

989

Fig. 8. Splinter hemorrhage.

Fig. 6. Janeway lesions.

tival petechiae may occur but are not specific for infective endocarditis (Fig. 7-10). ■

■

Osler nodes are small, painful nodular lesions usually found on the pads of the fingers or toes. Janeway lesions are hemorrhagic, macular, painless plaques with a predilection for the palms or soles.

Embolic Events Major embolic episodes are an important complication of infective endocarditis. Splenic artery emboli with infarction may result in left upper quadrant abdominal pain (with radiation to the left shoulder), splenic or pleural rubs, or left pleural effusion. Renal infarctions may be associated with microscopic or gross hematuria. Retinal artery emboli are rare and may be manifested by a complete, sudden loss of vision. Pulmonary emboli arising from right-sided endocarditis are a common feature in intravenous drug users. Coronary artery emboli usually arise from the aortic valve and may result in septic myocarditis with arrhythmias or myocardial infarction. Major vessel emboli (femoral, brachial, popliteal, or radial arteries) are more frequent in fungal endocarditis. Major cerebral emboli may result in hemiplegia, sensory loss, ataxia, aphasia, or alteration in mental status. ■

■

Fig. 7. Roth spots.

Splenic artery emboli with infarction may result in left upper quadrant abdominal pain (with radiation to the left shoulder). Renal infarctions may be associated with microscopic or gross hematuria.

Neurologic Complications Neurologic manifestations occur in one-third of patients with infective endocarditis. Up to 50% of these patients present with neurologic signs and symptoms as heralding features of their illness. Mycotic aneurysms

990

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Fig. 10. Conjuctival petechiae.

Fig. 9. Palatal petechiae.

of the cerebral circulation occur in 2% to 10% of patients (Fig. 11). Typically, these aneurysms are single, small, and peripheral and may lead to devastating subarachnoid hemorrhage. Other neurologic features include seizures, severe headache, visual changes, choreoathetoid movements, mononeuropathy, cranial nerve palsies, and toxic encephalopathy. Most central nervous system complications of infective endocarditis precede the diagnosis of infective endocarditis and initiation of effective therapy (Fig. 12). ■

prosthetic valve replacement. Acute infections account for the majority of hospital admissions among injection drug addicts, and infective endocarditis is found in ~10% of these episodes. Cocaine use, visualization of vegetations with echocardiography, and presence of embolic phenomena are among the most reliable indicators of infective endocarditis in febrile injection drug users.In this group of patients,two-thirds have no clinical evidence of preexisting underlying heart disease, and there is a predilection for infection of the tricuspid valve. The predominance of right-sided endocarditis in injection drug addicts is presumed to be due to the injection of both drug and adjunctive compounds used

Neurologic manifestations occur in one-third of patients with infective endocarditis.

Infective Endocarditis in Injection Drug Addicts The risk for the development of endocarditis among injection drug addicts is as high as 5% per patient per year and may be higher in patients who previously had

Fig. 11. Mycotic aneurysm (arrows).

Chapter 82 Infective Endocarditis to dilute the active agent, resulting in endocardial damage of the tricuspid (and pulmonic) valves. Tricuspid valve infection may result in pleuritic chest pain, cough, and hemoptysis, with chest radiographic findings of infiltrates and effusions. Signs of tricuspid insufficiency are present in a small proportion of these cases.The microbiology of endocarditis associated with injection drug addicts differs from that of nonaddicts, with a higher prevalence of gram-negative, fungal, and staphylococcal organisms identified. The course of acute S. aureus endocarditis in injection drug addicts tends to be less severe than in nonaddicts. ■

■

Changing murmurs or new murmurs are important findings in endocarditis but are uncommon. Cutaneous manifestations of endocarditis include petechiae, Osler nodes, and Janeway lesions.

Laboratory Findings in Infective Endocarditis Anemia is seen in ~75% of patients and typically is a mild normochromic normocytic anemia. Leukocytosis is seen in about one-third of patients and typically is mild; leukopenia is occasionally noted, especially in conjunction with splenomegaly. Large mononuclear cells (histiocytes) may be noted in peripheral blood. Positive rheumatoid factor, hypocomplementemia, a false-positive VDRL test, and a false-positive Lyme serologic test may occur in endocarditis. Circulating immune com-

plexes as well as mixed-type cryoglobulins may also be detected. The results of urinalysis may reveal proteinuria, microscopic hematuria, red blood cell casts, gross hematuria, pyuria, white blood cell casts, or bacteriuria. ■

A false-positive VDRL test and a false-positive Lyme serologic test may occur in endocarditis.

Blood Culture Blood culture is the most important laboratory test for diagnosing infective endocarditis. The bacteremia of infective endocarditis is typically continuous and low grade. When bacteremia is present, the first two blood cultures will yield the etiologic agent in more than 90% of cases. At least three blood culture sets should be obtained during the first 24 hours. More cultures may be necessary if the patient has received an antimicrobial agent in the preceding 2 weeks. Nutritionally variant streptococci require supplementation of the culture media. Some unusual organisms, such as Brucella species, Nocardia species, and members of the HACEK group (see below), are slow-growing, sometimes requiring that cultures be held for extended incubation (4 weeks). Special culture techniques or media may be required for some organisms (e.g., Legionella species, mycobacteria). Tissue cell culture may be useful for isolating obligate intracellular bacteria (e.g., Coxiella burnetii, Chlamydia species) from blood, vegetations, or valvular tissues. Blood culture results are negative in more than 50% of cases of fungal endocarditis. When embolization to major vessels occurs, embolectomy should be performed and the material examined with stains and culture for fungi. Serologic studies are useful for the diagnosis of Q fever, murine typhus, bartonellosis, brucellosis, legionellosis, and psittacosis. Gram stain of resected valve specimens may be helpful in the diagnosis of infective endocarditis. ■

■

■

Fig. 12. Brain abscess caused by Aspergillus species in a patient with Aspergillus infective endocarditis. (From same patient as in Fig. 14.)

991

■

Blood culture is the most important test for diagnosing infective endocarditis. Blood cultures are negative in more than 50% of cases of fungal endocarditis. Endocarditis is often associated with anemia and an increased erythrocyte sedimentation rate. At least three blood culture sets should be obtained during the first 24 hours.

992

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Polymerase Chain Reaction The polymerase chain reaction (PCR) can be used to detect organisms which grow poorly (or not at all) using conventional culture techniques. It can be applied to blood, excised vegetations, or systemic emboli.There are two basic approaches to PCR for the diagnosis of infective endocarditis. The first detects a genetic target unique to a specific genus or species of microorganism. The second broadly detects nucleic acid common to a large group of organisms based on a conserved genetic target. A common target used for broad range detection of bacteria using PCR is 16S ribosomal DNA which is present in all bacteria (but not in humans). Broad range amplification of 16S ribosomal DNA can be followed by sequencing of amplified DNA to identify the source bacterium. Several caveats about PCR should be noted. PCR is a very sensitive technique and can be associated with false-positive results. A positive PCR result must always be interpreted within the clinical context. Even several months after therapy for infective endocarditis, PCR results may still be positive. Since PCR does not provide information about antimicrobial susceptibility, routine culture and susceptibility testing remain important in the microbiologic diagnosis of infective endocarditis. Echocardiography Echocardiography is central in both the diagnosis and management of infective endocarditis. Specific echocardiographic findings are part of the Duke major diagnostic criteria. The objectives of the echocardiographic evaluation in a patient with suspected infective endocarditis are summarized in Table 3. There may be one or more vegetations involving one or more valves. Infrequent locations for vegetations (e.g., left atrial or ventricular wall in the path of a regurgitant jet) should also be assessed. When a vegetation is detected, size, location, and connections with other structures and associated local complications should be assessed.The characteristic finding noted in patients with endocarditis is shaggy dense irregular echoes distributed uniformly on one or more leaflets. The average size of vegetations is similar on the aortic and mitral valves. Tricuspid valve vegetations are typically significantly larger and pulmonic valve vegetations usually smaller. Large vegetations on the mitral valve, especially on the anterior leaflet, are associated with a higher risk of

embolism than vegetations of similar size elsewhere. Overall, the presence or absence of a vegetation and its size does not accurately predict future embolic events. Also, vegetation size has no definite relationship to the incidence of heart failure, the risk of death during the acute phase of infective endocarditis, or the final outcome. There is no size or location threshold that accurately predicts increased mortality associated with embolization in such a way as to justify surgery for the prevention of embolization. An increase in the size of vegetations detected by echocardiography during the course of therapy may identify a subgroup of patients with a higher rate of complications. Persistence of vegetations, as determined by echocardiography, is common after successful medical treatment of infective endocarditis and is not closely associated with late complications. Conversely, sudden disappearance of a vegetation may imply fragmentation and embolization (Fig. 13). Transthoracic echocardiography has excellent specificity (98%) but low sensitivity (30%-40%) for detection of small vegetations. Transthoracic echocardiography is inadequate in up to 20% of adult patients because of obesity, chronic obstructive pulmonary disease, endotracheal intubation, or chest-wall deformities. Transthoracic echocardiography may be used in the initial evaluation of patients with suspected infective endocarditis involving native valves but is generally much less useful in prosthetic valve endocarditis

Table 3. Objectives of Echocardiography in a Patient With Suspected Infective Endocarditis Detection of presence, location, and size of vegetations Evaluation of functional deficiencies of involved valves (e.g., valvular regurgitation) Identification of anatomy of infected valves and other possible companion diseases Assessment of consequences of valvular dysfunction (e.g., left ventricular function, pulmonary hypertension) Detection of other intracardiac complications (e.g., myocardial abscess)

Chapter 82 Infective Endocarditis

Fig. 13. Gram stain of S. aureus vegetation from a patient with infective endocarditis.

because of acoustic shadowing of the valve. If the clinical probability of native valve infective endocarditis is less than 4%, a negative transthoracic echocardiography is cost effective and clinically satisfactory in excluding infective endocarditis. In patients in whom the clinical probability of infective endocarditis is 4% or greater, negative transthoracic echocardiography should be followed by transesophageal echocardiography. Patients who should undergo urgent transesophageal echocardiography include those with gram-positive coccal bacteremia, those with catheter-associated S. aureus bacteremia, and those admitted with fever or bacteremia in the setting of injection drug use. Transesophageal echocardiography does not significantly improve the diagnostic accuracy of transthoracic echocardiography in the detection of vegetations associated with rightsided endocarditis in injection drug abusers, but is useful for detecting paravalvar abscess formation and for diagnosing unusual forms of right-sided endocarditis such as pulmonary valve involvement or infection of the eustachian valve. Transesophageal echocardiography is more sensitive than transthoracic echocardiography in detecting intracardiac vegetations, perivalvular abscesses, vegetations associated with prosthetic valves and vegetations smaller than 5 mm. Transesophageal echocardiography with color-flow Doppler techniques can be used to demonstrate the distinctive flow patterns of a fistula, pseudoaneurysm, or unruptured abscess cavity and is more sensitive than transthoracic echocardiography for identifying valvular perforation.

993

Negative findings on transesophageal echocardiography decrease the likelihood of endocarditis but do not exclude the diagnosis. Causes of such negative findings include an incomplete study, vegetation size less than 2 mm, prosthetic valve endocarditis, calcified valves, and vegetation fragmentation and embolization prior to the study. For patients with suspected right-sided infective endocarditis who are injection drug abusers, a transthoracic echocardiogram should first be performed. If there is a moderate or high suspicion of infective endocarditis, and initial transthoracic echocardiography is negative, this study should be repeated after an interval of about one week. Transesophageal echocardiography is recommended for patients in whom quality images are not obtained with transthoracic echocardiography. The development of a new high-grade atrioventricular block or bundle branch block seen on electrocardiography is highly specific as a predictor of perivalvular abscess. The degree of mitral valve preclosure in patients with aortic insufficiency, as determined by echocardiography, correlates with increased left ventricular end-diastolic pressure and the severity of hemodynamic compromise. CT and/or MRI of the head is indicated in all patients with endocarditis and neurologic symptoms. Infarction, hemorrhage, or abscess usually can be differentiated by these techniques. Cerebral angiography should be considered for selected patients with neurologic symptoms not readily explained with other imaging to exclude intracranial mycotic aneurysm. ■

Transesophageal echocardiography is more sensitive than transthoracic echocardiography in the detection of intracardiac vegetations and perivalvular abscess.

Diagnostic Criteria The Duke diagnostic criteria for infective endocarditis are listed in Table 4A, and definitions of the diagnostic terms are given in Table 4B. Microbiology Streptococcal Endocarditis Streptococci are the most common causative agents of native valve infective endocarditis in nonintravenous drug users (Table 5). Of these, viridans group streptococci are the most common subgroup and frequently

994

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Table 4A.Duke Criteria for Diagnosis of Infective Endocarditis Definite infective endocarditis Pathologic criteria 1. Microorganism demonstrated by culture or histologic examination* of a vegetation, a vegetation that has embolized, or an intracardiac abscess specimen; or 2. Pathologic lesions; vegetation or intracardiac abscess confirmed by histologic examination showing active endocarditis Clinical criteria 1. 2 major criteria; or 2. 1 major criteria and 3 minor criteria; or 3. 5 minor criteria Possible infective endocarditis 1. 1 major criterion and 1 minor criterion; or 2. 3 minor criteria Rejected 1. Firm alternate diagnosis explaining evidence of infective endocarditis; or 2. Resolution of infective endocarditis syndrome with antimicrobial therapy for ≤4 days; or 3. No pathologic evidence of infective endocarditis at surgery or autopsy, with antimicrobial therapy for ≤4 days; or 4. Does not meet criteria for possible infective endocarditis, as above *Some authors suggest that a microorganism detected by molecular-based techniques or by Gram stain, may fulfill a pathologic criterion.

are of oral origin. The cure rate of streptococcal endocarditis exceeds 90%, although complications occur in more than 30% of cases. An association of Streptococcus gallolyticus (formerly Streptococcus bovis) bacteremia with carcinoma of the colon and other lesions of the gastrointestinal tract has been shown; colonoscopy and/or barium enema should be performed when this organism is isolated from blood cultures. Streptococcus pneumoniae is a rare cause of infective endocarditis; however, when present, it typically has a fulminant course and often is associated with perivalvular abscess formation and pericarditis. In S. pneumoniae infective endocarditis, the aortic valve is typically involved and many such patients have a history of alcohol abuse. Concurrent meningitis is present in ~70% of patients. Infective endocarditis due to Abiotrophia defectiva and Granulicatella adiacens (formerly nutritionally variant streptococci) is typically indolent in onset and associated with previous heart disease. Therapy is difficult because of systemic embolization and frequent relapse.

Group B streptococcus (Streptococcus agalactiae) may cause infective endocarditis. Risk factors for group B streptococcal infective endocarditis in adults include diabetes mellitus, carcinoma, alcoholism, liver failure, elective abortion, and injection drug abuse. Group B streptococcal endocarditis has been associated with villous adenomas of the colon. The mortality of this type of infection approaches 50%. A similar clinical picture with a destructive process, left-sided predominance, frequent complications, and high mortality has been observed with group A or G streptococcus. Streptococcus anginosus is a rare cause of infective endocarditis, but it is notable because it has a predilection for suppurative complications involving the brain and liver; perinephric, myocardial, and other abscesses; cholangitis; peritonitis; pericarditis; and empyema more characteristic of S. aureus infections. ■

Viridans group streptococci are the most common causative agents of native valve infective endocarditis in noninjection drug abusers.

Chapter 82 Infective Endocarditis

995

Table 4B.Definition of Terms Used in the Diagnostic Criteria Major criteria Blood culture positive for infective endocarditis 1. Typical microorganisms consistent with infective endocarditis from 2 separate blood cultures: Viridans group Streptococcus species; *Streptococcus gallolyticus (formerly Streptococcus bovis), HACEK group, Staphylococcus aureus; or community-acquired Enterococcus species, in the absence of a primary focus or 2. Microorganism consistent with infective endocarditis from persistently positive cultures, defined as follows: At least 2 positive cultures of blood samples drawn >12 h apart; or All of 3 or a majority of ≥4 separate cultures of blood (with first and last sample drawn at least 1 h apart) 3. Single positive blood culture for Coxiella burnetii or antiphase I IgG antibody titer >1:800 Evidence of endocardial involvement Echocardiogram positive for infective endocarditis (transesophageal echocardiography recommended in patients with prosthetic valves, rated at least “possible infective endocarditis” by clinical criteria, or complicated infective endocarditis (paravalvular abscess); transthoracic echocardiography as first test in other patients), defined as follows: 1. Oscillating intracardiac mass on valve or supporting structures, in the path of regurgitant jets, or on implanted material in the absence of an alternative anatomic explanation; or 2. Abscess; or 3. New partial dehiscence of prosthetic valve New valvular regurgitation (worsening or changing of pre-existing murmur not sufficient) Minor criteria Predisposition, predisposing heart condition or injection drug use Fever, temperature >38°C Vascular phenomena, major arterial emboli, septic pulmonary infarct, mycotic aneurysm, intracranial hemorrhage, conjunctival hemorrhage, and Janeway’s lesions Immunologic phenomena: glomerulonephritis, Osler nodes, Roth spots, and rheumatoid factor Microbiological evidence: positive blood culture but does not meet a major criterion as noted above† or serological evidence of active infection consistent with infective endocarditis‡ *Including Abiotrophia defectiva, Granulicatella adiacens. †Excludes single positive cultures for coagulase negative staphylococci and organisms that do not cause endocarditis. ‡Serologic test result positive for Brucella species, Chlamydia species, Legionella species, Bartonella species.

■

■ ■

The cure rate of nonenterococcal streptococcal endocarditis is more than 90%. Viridans group streptococci are often of oral origin. S. gallolyticus (formerly S. bovis) may be associated with colon lesions.

Enterococcal Endocarditis Enterococcal endocarditis typically affects older men after genitourinary tract manipulation or younger

women after an obstetric procedure. More than 40% of patients with enterococcal endocarditis have no previously recognized underlying heart disease, although more than 95% develop a heart murmur during the course of the illness. Classic peripheral manifestations are uncommon. Factors that suggest that a patient with enterococcal bacteremia may have infective endocarditis include no identifiable extracardiac focus of infection and preexistent valvular heart disease or heart murmur.

996

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Staphylococcal Endocarditis Staphylococci are the second most common cause of infective endocarditis. Of the staphylococci, S. aureus is the most common cause of native valve infective endocarditis (Fig. 13) and may attack normal heart valves in addition to diseased ones. The course of S. aureus infective endocarditis is typically fulminant when it involves the mitral or aortic valve, with widespread metastatic infection and a 40% chance of death. Myocardial abscesses, purulent pericarditis, valve-ring abscesses, and peripheral foci of suppuration (lung, brain, spleen, kidney, etc.) are common with S. aureus infective endocarditis. Approximately one-third of patients with S. aureus endocarditis experience neurologic manifestations, with two-thirds or more of this group presenting with neurologic symptoms before initiation of antimicrobial therapy. The most frequent neurologic presentation is hemiparesis. In injection drug addicts, S. aureus is the most frequent cause of infective endocarditis, but the disease tends to be less severe than that in nonaddicted patients. Children with endocarditis due to S. aureus are more likely than those with infections due to viridans group streptococci to have prolonged fever, complications and to require surgery. Infective endocarditis caused by methicillinresistant S. aureus is increasingly common, especially in injection drug addicts. Coagulase-negative staphylococci are an important cause of prosthetic valve endocarditis.

rate approaching 80%. Salmonella species are associated with valvular perforation or destruction (or both), atrial thrombi, myocarditis, and pericarditis. Several cases of infective endocarditis due to Serratia marcescens have been noted in injection drug abusers. Typically, this infection has involved the aortic and mitral valves, with large vegetations and near-total occlusion of the valve orifice in the absence of significant underlying valvular destruction. Pseudomonas species infective endocarditis occurs in injection drug addicts and usually affects normal valves. Major embolic phenomena, inability to sterilize valves, neurologic complications, ring and anular abscesses, splenic abscesses, bacteremic relapses, and rapidly progressive congestive heart failure are common.Pseudomonas aeruginosa endocarditis has been associated with the use of pentazocine and tripelennamine. Neisseria gonorrhoeae occasionally causes infective endocarditis and typically follows an indolent course, with aortic valve involvement, large vegetations, associated valve-ring abscesses, congestive heart failure, and nephritis. A high frequency of complement component deficiencies has been noted in patients with gonococcal endocarditis. ■

■ ■

■

■

Staphylococci are an important cause of infective endocarditis. In injection drug addicts, S. aureus is the most frequent cause of infective endocarditis. S. aureus endocarditis has frequent intra- and extracardiac complications.

Gram-Negative Endocarditis Gram-negative bacilli may also cause infective endocarditis. Typically, the gram-negative bacilli involved are fastidious organisms such as those belonging to the HACEK group (see below), although occasionally the Enterobacteriaciae may be involved. Persons addicted to narcotics, prosthetic valve recipients, elderly individuals, and patients with cirrhosis appear to be at increased risk for gram-negative bacillary endocarditis. Congestive heart failure is common in this group of patients, and the prognosis is poor, with the mortality

Persons addicted to narcotics, prosthetic valve recipients, and patients with cirrhosis appear to be at increased risk for gram-negative bacillary endocarditis. P. aeruginosa endocarditis has been associated with the use of pentazocine and tripelennamine.

HACEK Endocarditis Members of the HACEK group of organisms include Haemophilus species, Actinobacillusactinomycetemcomitans, Cardiobacterium hominis, E. corrodens, and Kingella species. Infective endocarditis due to the HACEK group of organisms (normal inhabitants of the human oropharynx) has been reported in patients who have dental infections and a history of dental procedures and in injection drug abusers who have “cleaned” the injection site with saliva. HACEK endocarditis characterized by a lengthy (i.e., 2 weeks to 6 months) course before diagnosis, large friable vegetations, frequent emboli, and the development of congestive heart failure, with eventual valve replacement. The HACEK group of organisms are fastidious and may require weeks for primary isolation.

Chapter 82 Infective Endocarditis Other Agents The microbiology of infective endocarditis in injection drug abusers is distinct from that in noninjection drug abusers (Table 5). Fungal Endocarditis Fungal endocarditis his increasing in occurance because of the increased number of immunocompromised patients and drug users, the extensive use of broad spectrum antimicrobial agents, and the use of indwelling central venous catheters or hyperalimentation. Candida parapsilosis and Candida tropicalis predominate in injection drug addicts, and C. albicans and Aspergillus species cause most cases of fungal infective endocarditis in noninjection drug addicts. Fungal endocarditis carries a poor prognosis because of large bulky vegetations, tendency for fungal invasion of the myocardium, widespread systemic septic emboli, and poor penetration of antifungal agents into vegetations (Fig. 14-16). Surgical intervention is almost always required. In patients with Aspergillus endocarditis, most blood cultures will be negative, in contrast to the continuous bacteremia in bacterial endocarditis. Other peripheral lesions (e.g., emboli, cutaneous lesions, oropharyngeal lesions, sputum, and bronchoalveolar lavage fluid) should be exam-

ined and cultured for fungi in suspected settings. Valvular vegetations are not always seen on echocardiography, and clinical manifestations of endocarditis are not always present in Aspergillus endocarditis. The case fatality rate associated with fungal endocarditis is more than 80% for molds and more than 40% for yeasts. ■

In patients with Aspergillus infective endocarditis, most blood cultures will be negative.

Culture-Negative Endocarditis Culture-negative endocarditis accounts for a small proportion of cases (10 mm in the context of persisting fever as an indication for surgery) *Criteria also apply to repair mitral and aortic allograft or autograft valves.

few weeks after valve replacement which then decreases to a stable low incidence rate during subsequent months to years. The risk of infection for mechanical and bioprosthetic valves is similar and there is no difference in the risk of endocarditis between mitral or aortic prostheses. Prosthetic valve endocarditis has been classified arbitrarily as “early” the first 60 days after implantation and “late” after 60 days. Classically, it was thought that “early” cases were acquired at the time of implantation and “late” cases thereafter. Subsequently, it was shown that many cases of prosthetic valve endocarditis that occur during the first year after surgery are acquired at the time of implantation. Some investigators have recommended that the time limit for “early” disease be extended to 6 months or even 1 year. The mortality associated with prosthetic valve endocarditis is 30% to 80% in the “early” form and 20% to 40% in “late” postsurgical endocarditis, with a worse prognosis associated with a new or changing murmur, new or worsening heart failure, persistent fever despite appropriate antimicrobics, new conduction myocardial

abscess, renal insufficiency, S. aureus as the causative agent, and neurologic complications. ■

■

Prosthetic valve endocarditis has been reported to occur in up to 10% of patients during the lifetime of the prosthesis. Prosthetic valve endocarditis has been classified arbitrarily as “early” when it occurs within the first 60 days after implantation and “late”when it occurs after 60 days.

Pathogenesis of Prosthetic Valve Endocarditis Early S. epidermidis prosthetic valve endocarditis is thought to result from valve contamination during the perioperative period.This may occur at the time of surgery or in the immediate postoperative period when the prosthetic valve and sewing ring are not yet endothelialized and are susceptible to microbial adherence. Nosocomial bacteremia is an important risk factor for prosthetic valve endocarditis. Another potential (but uncommon) source of infection is contamination of the prosthesis before implantation (e.g., contamination of

1006

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

glutaraldehyde-fixed porcine prosthetic valves with Mycobacterium chelonei). The pathogenesis of late prosthetic valve endocarditis is similar to that of native valve endocarditis, with microorganisms from a transient bacteremia localizing on a prosthesis or area of damaged endothelium. Valve-Ring Abscess Valve-ring abscess is a serious complication of prosthetic valve endocarditis and is seen with both mechanical and bioprosthetic valves. Valve-ring abscesses occur where infection involves the sutures used to secure the sewing ring to the perianular tissue; this may result in dehiscence of the valve. The clinical finding of a new perivalvular leak in a patient with prosthetic valve endocarditis is presumptive evidence of a valve-ring abscess. Extension of the abscess beyond the valve ring may result in myocardial abscess formation, septal perforation, or purulent pericarditis. In addition to sewing-ring abscesses, prosthetic valve endocarditis of the bioprosthesis may cause leaflet destruction, with resulting valvular incompetence. Large vegetations occasionally obstruct blood flow and lead to functional valvular stenosis or a combination of stenosis and insufficiency.This complication appears to be more common in mitral prosthetic valve endocarditis than in aortic disease. Bioprosthetic valve endocarditis may involve only the valve cusps,the sewing ring,or both. ■

■

Early prosthetic valve endocarditis results from valve contamination during the perioperative period. Late prosthetic valve endocarditis results more often from transient bacteremia.

Echocardiography in Prosthetic Valve Endocarditis Transthoracic echocardiography is less accurate in the diagnosis of prosthetic valve endocarditis than in native valve endocarditis, because the echoes generated by the prosthesis may mask subtle abnormalities such as small vegetations. Transesophageal echocardiography is more sensitive than transthoracic echocardiography for the detection of vegetations, periprosthetic tissue destruction with prosthetic dehiscence, myocardial abscesses, fistulas, pseudoaneurysms, and perivalvular abscesses. Since infective endocarditis involving mechanical prostheses usually starts at the prosthetic ring, the

search of vegetations in such patients must focus on the prosthetic ring. In contrast, infective endocarditis of biological prostheses involves both the ring and the valvular leaflets. For patients with negative transesophageal echocardiography and an intermediate probability of infective endocarditis, repeat transesophageal echocardiography is recommended after a week, especially if an aortic prosthesis could be involved. ■

■

Transesophageal echocardiography is recommended for diagnosis of prosthetic valve endocarditis. CT and/or MRI of the head is indicated in patients with prosthetic valve endocarditis and neurologic symptoms.

Diagnostic Criteria The currently accepted diagnostic criteria for infective endocarditis are outlined in Table 4. Microbiology Among cases of prosthetic valve endocarditis, coagulasenegative staphylococci are the dominant cause of endocarditis occurring in the first postoperative year. The organisms causing prosthetic valve endocarditis more than 12 months after valve implantation are similar to those associated with native valve endocarditis (except for nosocomial and drug abuse-associated infective endocarditis). During this late postoperative period, the predominant causes of infection are streptococci, coagulase-negative staphylococci, enterococci, S. aureus, and members of the HACEK group of organisms. A broad range of bacteria have also caused sporadic cases of prosthetic valve endocarditis. Corynebacterium species cause prosthetic valve endocarditis that occurs within the first 6 postoperative months and are notable because of their relative resistance to many antimicrobial agents (other than vancomycin) and their fastidious growth requirements. Fungi not only account for a number of cases of prosthetic valve endocarditis but are associated with high case fatality rates. Candida species followed by Aspergillus species are the two most common fungi that cause prosthetic valve endocarditis. Fungal vegetations formed on prosthetic valves are bulky and may partially occlude the orifice or embolize and occlude mediumsized arteries. Notably, patients with prosthetic heart valves who develop nosocomial candidemia are at risk

Chapter 82 Infective Endocarditis for either having or developing candidal prosthetic valve endocarditis months or years later. Late-onset candidemia and lack of an identifiable portal of entry should heighten concern about candidal prosthetic valve endocarditis in such patients. ■

Coagulase-negative staphylococci are the most common cause of prosthetic valve endocarditis in the first postoperative year.

Treatment Antimicrobial therapy is based on laboratory identification of the etiologic microorganism and in vitro susceptibility testing. Bactericidal antimicrobials are necessary. Recommended antimicrobial regimens are listed in Table 7. Many isolates of coagulase-negative staphylococci isolated from patients with prosthetic valve endocarditis are resistant to oxacillin. For methicillinresistant staphylococcal infection on prosthetic valves, treatment with a combination of vancomycin, rifampin, and gentamicin is recommended. Fungal prosthetic valve endocarditis usually requires combined medical and surgical therapy. For Candida endocarditis, high doses of amphotericin B given intravenously in combination with oral flucytosine are often used. Alternative considerations include caspofungin or combined fluconazole and oral flucytosine. For culture-negative prosthetic valve endocarditis, treatment must be individualized. When prosthetic valve endocarditis is considered but the level of clinical suspicion is low, 3 or 4 blood specimens should be obtained by separate venipunctures for culture and the patient observed. If valve replacement surgery is imminent, it is reasonable to initiate empiric antimicrobial therapy with vancomycin and gentamicin. The diagnosis of prosthetic valve endocarditis can usually be confirmed or excluded at the time of valve replacement. ■

For methicillin-resistant staphylococcal infection on prosthetic valves, treatment with a combination of vancomycin, rifampin, and gentamicin is recommended.

After initiation of antimicrobial therapy, blood should be cultured daily for the first few days and weekly thereafter until the completion of therapy. Usually, blood cultures will be sterile within 3 to 5 days

1007

after appropriate antimicrobial therapy is initiated. After completion of therapy, blood should be cultured weekly for 1 month. A relapse necessitates reinstitution of antimicrobial therapy, retesting of the microorganism for antimicrobial susceptibility, and consideration of valve replacement. Indications for cardiac surgery in patients with prosthetic valve endocarditis are listed in Table 10. Moderate to severe congestive heart failure associated with prosthesis dysfunction is a common indication for surgery. Few patients with prosthetic valve endocarditisinduced heart failure are alive 6 months after medical treatment, whereas combined surgical and medical treatment has resulted in survival rates of up to 64%. Patients with culture-negative endocarditis who continue to experience fever during empiric antibiotic therapy are candidates for surgical intervention. Surgery may allow a definitive microbiologic diagnosis and development of specific, effective antimicrobial therapy. Also, some of these patients will be found to have fungal endocarditis or unrecognized invasive infection that warrants surgery. ■

Medical therapy alone is appropriate for some patients with prosthetic valve endocarditis.

The timing of cardiac surgery in patients with prosthetic valve endocarditis must be individualized. The hemodynamic status of the patient is the most important consideration in determining the timing of operation. As in patients with native valve endocarditis, the likelihood of those with prosthetic valve endocarditis surviving valve replacement is inversely related to the severity of the patient’s heart failure at the time of operation. Thus, although in theory it may be desirable to control infection with antimicrobial therapy preoperatively, this must not be attempted at the expense of progressive destruction of perivalvular tissue and further deterioration in the patient’s hemodynamic status. Longer periods of antimicrobial therapy preoperatively do not correlate with inability to recover bacteria from intraoperative cultures or with a more favorable outcome. Renal dysfunction preoperatively is one of the most important predictors of both increased operative mortality and overall long-term poor prognosis. Renal failure is often associated with advanced decompensated heart failure and low cardiac output.

1008

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Table 10. Recommendations for Surgery for Prosthetic Valve Endocarditis* Indication Early prosthetic valve endocarditis (first 2 months or less after surgery) Heart failure with prosthetic valve dysfunction Fungal endocarditis Staphylococcal endocarditis not responding to antimicrobial therapy Evidence of paravalvular leak, anular or aortic abscess, sinus or aortic true or false aneurysm, fistula formation, or new-onset conduction disturbances Infection with gram-negative organisms with a poor response to antimicrobials Persistent bacteremia after prolonged course (7 to 10 days) of appropriate antimicrobial therapy without noncardiac causes for bacteremia Recurrent peripheral embolus despite therapy *Criteria exclude repaired mitral valves or aortic allograft or autograft valves.

■

The hemodynamic status of the patient is the most important consideration in determining the timing of operation.

In selected patients with prosthetic valve endocarditis, the results of treatment with antimicrobial agents alone are comparable to the results of combined surgical and medical therapy; for these patients, medical therapy is recommended. Included in this subgroup are patients with late-onset prosthetic valve endocarditis (12 months or more postoperatively) who are infected with less virulent organisms (viridans group streptococci, enterococci, and fastidious gram-negative coccobacilli) and who do not develop complicated endocarditis. Careful anticoagulation therapy is recommended for patients with mechanical prosthetic valve endocarditis involving prostheses that usually would warrant maintenance of anticoagulation. Anticoagulation should be reversed temporarily, however, if a patient

experiences a hemorrhagic central nervous system event. Anticoagulation is not recommended for bioprosthetic valve endocarditis that under usual circumstances do not require anticoagulation therapy.

PROPHYLAXIS OF INFECTIVE ENDOCARDITIS Prophylaxis for infective endocarditis is advised for patients who have an underlying cardiac condition that places them at increased risk for endocarditis and who are undergoing a procedure that carries a risk of transient bacteremia due to an organism that causes endocarditis. Endocarditis prophylaxis is recommended in high-risk and, in most cases, moderate-risk patients; it is not required in low-risk patients (Table 11). Mitral valve prolapse is common and represents a spectrum of abnormalities. A clinical approach to determine the need for prophylaxis in persons with suspected mitral valve prolapse, as recommended by the American Heart Association, is shown in Figure 17. When normal valves prolapse without leaking, as in patients with one or more systolic clicks but no murmurs and no Doppler-demonstrated mitral regurgitation, the risk of endocarditis is not increased above that of the normal population. Therefore, antimicrobial prophylaxis against endocarditis is not necessary. This is because it is not the abnormal valve motion but the jet of mitral insufficiency that creates the shear force and flow abnormalities that increase the likelihood of bacterial adherence on the valve during bacteremia. However, patients with prolapsing and leaking mitral valves, evidenced by audible clicks and murmurs of mitral regurgitation or by Doppler-demonstrated mitral insufficiency, should receive prophylactic treatment with antimicrobials. Similarly, patients with myxomatous mitral valve degeneration with regurgitation valve leaflet thickening should receive endocarditis prophylaxis. Because older age and male sex have been shown to be risk factors for the development of endocarditis, men older than 45 years with mitral valve prolapse, even without a consistent systolic murmur, may warrant prophylaxis even in the absence of resting regurgitation. The American Heart Association has identified common procedures for which prophylaxis is recommended or not recommended according to the perceived degree of risk (Table 12). The recommended antimicrobial regimens for infective endocarditis

Chapter 82 Infective Endocarditis

1009

Table 11. Cardiac Conditions Associated With Endocarditis* Endocarditis Prophylaxis Recommended High-risk category Prosthetic cardiac valves, including bioprosthetic and homograft valves Previous bacterial endocarditis Complex cyanotic congenital heart disease (e.g., single ventricle states, transposition of the great arteries, tetralogy of Fallot) Surgically constructed systemic pulmonary shunts or conduits Moderate-risk category Most other congenital cardiac malformations (other than above and below) Acquired valvar dysfunction (e.g., rheumatic heart disease) Hypertrophic cardiomyopathy Mitral valve prolapse with valvar regurgitation and/or thickened leaflets Endocarditis Prophylaxis Not Recommended Negligible-risk category (no greater risk than the general population) Isolated secundum atrial septal defect Surgical repair of atrial septal defect, ventricular septal defect, or patent ductus arteriosus (without residua beyond 6 mo) Previous coronary artery bypass graft surgery Mitral valve prolapse without valvar regurgitation Physiologic, functional, or innocent heart murmurs Previous Kawasaki disease without valvar dysfunction Previous rheumatic fever without valvar dysfunction Cardiac pacemakers (intravascular and epicardial) and implanted defibrillators *Please consult the original table for references to the sources of the data.

prophylaxis are outlined in Tables 13 and 14. Before elective valve replacement, the dental health of every patient should be evaluated and any necessary dental work completed under close observation and with appropriate antibiotic coverage. The number of organisms in the mouth and gingival crevices can be decreased temporarily by local irrigation with an antiseptic solution such as iodinated glycerol. Some dental experts recommend routine use of this measure before dental extractions. Occasionally, a patient may be taking an antimicrobial agent when going to see a physician or dentist. If the patient is taking an antimicrobial agent normally used for endocarditis prophylaxis, it is prudent to select a drug from a different class rather than to increase the dose of the current antibiotic. In particular, antimicrobial regimens used to prevent the recurrence of acute

rheumatic fever are inadequate for the prevention of bacterial endocarditis. Persons who take an oral penicillin for secondary prevention of rheumatic fever or for other purposes may have viridans group streptococci in their oral cavities that are relatively resistant to penicillin, amoxicillin, or ampicillin. In such cases, clindamycin, azithromycin, or clarithromycin should be selected for endocarditis prophylaxis. Because of possible cross-resistance with cephalosporins, this class of antibiotic should be avoided. If possible, one should delay the procedure until at least 9 to 14 days after completion of the antimicrobial agent to allow the usual oral flora to be reestablished. ■

Patients receiving a low dose of penicillin for rheumatic fever prophylaxis should not receive penicillin for endocarditis prophylaxis.

1010

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Patient with suspected mitral valve prolapse

Murmur of mitral regurgitation

Presence or absence of mitral regurgitation not determined or not known

Refer for evaluation

Murmur and/or echocardiographic/ Doppler demonstration of mitral regurgitation Prophylaxis

No confirmation available Immediate need for procedure No regurgitation or echocardiographic findings, if performed

Prophylaxis

No prophylaxis

Fig. 17. Clinical approach to determination of need for prophylaxis in patients with suspected mitral valve prolapse.

INFECTIONS OF CARDIOVASCULAR DEVICES Symptoms and signs of cardiovascular device-related infections depend on the location of the infected part(s) of the device. When infected intravascular or endovascular portions of a device are present, clinical manifestations resemble those of infective endocarditis or endarteritis. Fever is often present, as are embolic events, which involve either the pulmonary or systemic vasculature (depending on the site of the infected device). Virulent pathogens, such as P. aeruginosa and S. aureus may be associated with sepsis. Subacute and chronic presentations are found in association with less virulent organisms. As with endocarditis, immunemediated events (e.g., immune complex-mediated nephritis, vasculitis) may be present. Cardiovascular device-related infections may present as bacteremia with fever in the absence of other clinical findings. Alternatively, findings may include local pain, erythema, induration, warmth and/or purulent drainage at the exit site of the device (e.g., percutaneous drivelines). Cellulitis or abscess formation may be present in association with subcutaneously implanted devices. Pseudoaneurysms may develop in cases of infections of vascular graft anastomoses; occlusion of a graft may lead to ischemia or necrosis.

The microbiology of cardiovascular device-related infections involves predominantly S. aureus and coagulase-negative staphylococci. The incidence of cardiovascular device-related infections depends on the device; details are provided in Table 15. Endothelialization is viewed as a healing response following device implantation in humans, and is considered important in the prevention of subsequent infection. The exact length of time required for endothelialization of implanted cardiovascular devices is not known, but is likely in the range of 1 to 3 months. Pacemaker and Implantable Cardioverter-Defibrillator Infections Pacemaker infections most commonly occur in the pocket in which the generator is placed. Involvement of the electrode leads occurs less frequently. In pacemaker or implantable cardioverter-defibrillator infective endocarditis, vegetations may form on the tricuspid valve and/or anywhere along the course of the electrode, including in the endocardium of the right ventricle or atrium. Pulmonary emboli or empyema may be present. The pathogenesis of generator box infections is thought to relate to contamination of the device by skin flora at the time of implantation. Most

Chapter 82 Infective Endocarditis

1011

Table 12. Recommendations for Prophylaxis During Various Procedures That May Cause Bacteremia* Endocarditis Prophylaxis Recommended Other Procedures Dental extractions Respiratory tract Periodontal procedures including surgery, scaling Tonsilectomy and/or adenoidectomy and root planing, probing, and recall maintenance Surgical operations that involve respiratory Dental implant placement and reimplantation of mucosa avulsed teeth Bronchoscopy with a rigid bronchoscope Endodontic (root canal) instrumentation or surgery Gastrointestinal tract‡ only beyond the apex Sclerotherapy for esophageal varices Subgingival placement of antibiotic fibers or strips Esophageal stricture dilation Initial placement of orthodontic bands but not Endoscopic retrograde cholangiography with brackets biliary obstruction Intraligamentary local anesthetic injections Biliary tract surgery Prophylactic cleaning of teeth or implants where Surgical operations that involve intestinal mucosa bleeding is anticipated Genitourinary tract Prostatic surgery Cystoscopy Urethral dilation

Dental Procedures†

Endocarditis Prophylaxis Not Recommended Dental Procedures Gastrointestinal tract § Restorative dentistry (operative and prosthodontic) Transesophageal echocardiography¶ // with or without retraction cord Endoscopy with or without gastrointestinal Local anesthetic injections (nonintraligamentary) biopsy// Intracanal endodontic treatment; post placement Genitourinary tract and buildup Vaginal hysterectomy¶ Placement of rubber dams Vaginal delivery¶ Postoperative suture removal Cesarean section¶ Placement of removable prosthodontic or orthoIn uninfected tissue: dontic appliances Urethral catheterization Taking of oral impressions Uterine dilatation and curettage Fluoride treatments Therapeutic abortion Taking of oral radiographs Sterilization procedures Orthodontic appliance adjustment Insertion or removal of intrauterine devices Shedding of primary teeth Other Other Procedures Cardiac catheterization, including balloon angioRespiratory tract plasty Endotracheal intubation Implanted cardiac pacemakers, implanted deBronchoscopy with a flexible bronchosope, fibrillators, and coronary stents with or without biopsy¶ Incision or biopsy of surgically scrubbed skin Tympanostomy tube insertion Circumcision *Please consult the original table for references to the sources of the data. †Prophylaxis is recommended for patients with high- and moderate-risk cardiac conditions. ‡Prophylaxis is recommended for high-risk patients; optional for medium-risk patients. §This includes restoration of decayed teeth (filling cavities) and replacement of missing teeth. //Clinical judgment may indicate antibiotic use in selected circumstances that may create significant bleeding. ¶Prophylaxis is optional for high-risk patients.

1012

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Table 13. Prophylactic Regimens for Dental, Oral, Respiratory Tract, or Esophageal Procedures* Situation

Agent

Standard general prophylaxis

Amoxicillin

Unable to take oral medications

Ampicillin

Allergic to penicillin

Clindamycin or Cefalexin‡ or cefadroxil‡ or Azithromycin or clarithromycin Clindamycin or Cefazolin‡

Allergic to penicillin and unable to take oral medications

Regimen† Adults: 2.0 g; children: 50 mg/kg orally 1 hr before procedure Adults: 2.0 g intramuscularly (IM) or intravenously (IV); children: 50 mg/kg IM or IV within 30 min before procedure Adults: 600 mg; children: 20 mg/kg orally 1 hr before procedure Adults: 2.0 g; children: 50 mg/kg orally 1 hr before procedure Adults: 500 mg; children: 15 mg/kg orally 1 hr before procedure Adults: 600 mg; children: 20 mg/kg IV within 30 min before procedure Adults: 1.0 g; children: 25 mg/kg IM or IV within 30 min before procedure

*Please consult the original table for references to the sources of the data. †Total children’s dose should not exceed adult dose. ‡Cephalosporins should not be used in persons with immediate-type hypersensitivity reaction (urticaria, angioedema, or anaphylaxis) to penicillins.

such infections are present soon after pacemaker implantation but may not become clinically evident for two years or longer. Wound infection or erosion of the box through the overlying skin may also lead to microbial contamination and subsequent infection. Microorganisms from the pocket can spread along the electrode to the endocardium and electrode tip. Hematogenous seeding of the endovascular electrode during transient bacteremia may also occur. Pacemaker or implantable cardioverter-defibrillator endocarditis most commonly occurs as a result of pocket infection; the most common pathogens of pacemaker and implantable cardioverter-defibrillator endocarditis are skin flora, including staphylococci and corynebacteria. Hematogenous seeding from a distant source of infection may account for late-onset infection due to S. aureus or other organisms (e.g., viridans group streptococci, enterococci, gram-negative bacilli, anaerobes, fungi, nontuberculous mycobacteria).

Risk factors for infection include the presence of diabetes mellitus, steroid use, underlying malignancy, overlying dermatologic disorders (especially pustular disorders), hematoma formation within the pocket, urgent placement or frequent replacement of the generator, and inexperience of the implantation team. The diagnosis should be entertained in patients with pacemakers or implantable cardioverter-defibrillators and unexplained fever. The diagnosis is typically confirmed by positive blood cultures and an echocardiogram that demonstrates vegetations on a pacemaker or implantable cardioverter-defibrillator lead. Transesophageal echocardiography is more sensitive than transthoracic echocardiography. The blood, the pacemaker pocket, and any other wound site should be cultured. A definitive diagnosis of pacemaker infection depends on isolation of the etiologic microorganism from the pacemaker pocket or the blood. Ideally, all of the hardware should be removed in

Chapter 82 Infective Endocarditis

1013

Table 14. Prophylactic Regimens for Genitourinary and Gastrointestinal (Excluding Esophageal) Procedures* Situation

Agent†

High-risk patients

Ampicillin plus gentamicin

High-risk patients allergic to ampicillin/amoxicillin

Vancomycin plus gentamicin

Moderate-risk patients

Amoxicillin or ampicillin

Moderate-risk patients allergic to ampicillin/amoxicillin

Vancomycin

Regimen‡ Adults: ampicillin 2.0 g intramuscularly (IM) or intravenously (IV) plus gentamicin 1.5 mg/kg (not to exceed 120 mg) within 30 min of starting the procedure; 6 hr later, ampicillin 1 g IM/IV or amoxicillin 1 g orally Children: ampicillin 50 mg/kg IM or IV (not to exceed 2.0 g) plus gentamicin 1.5 mg/kg within 30 min of starting the procedure; 6 hr later, ampicillin 25 mg/kg IM/IV or amoxicillin 25 mg/kg orally Adults: vancomycin 1.0 g IV over 1-2 hr plus gentamicin 1.5 mg/kg IV/IM (not to exceed 120 mg); complete injection/infusion within 30 min of starting the procedure Children: vancomycin 20 mg/kg IV over 1-2 hr plus gentamicin 1.5 mg/kg IV/IM; complete injection/infusion within 30 min of starting the procedure Adults: amoxicillin 2.0 g orally 1 hr before procedure, or ampicillin 2.0 g IM/IV within 30 min of starting the procedure Children: amoxicillin 50 mg/kg orally 1 hr before procedure, or ampicillin 50 mg/kg IM/IV within 30 min of starting the procedure Adults: vancomycin 1.0 g IV over 1-2 hr; complete infusion within 30 min of starting the procedure Children: vancomycin 20 mg/kg IV over 1-2 hr; complete infusion within 30 min of starting the procedure

*Please consult the original table for references to the sources of the data. †Total children’s dose should not exceed adult dose. ‡No second dose of vancomycin or gentamicin is recommended.

both generator box and electrode infections. Infection relapse has been strongly associated with failure to remove all of the hardware. The mortality of patients with pacemaker or implantable cardioverter-defibrillator endocarditis treated with antimicrobial agents alone is significantly higher than the mortality of such patients treated with a combination of antimicrobial agents and

electrode removal. In patients with S. aureus bacteremia and a pacemaker or implantable cardioverter-defibrillator, removal of the device is recommended in the following situations: 1) if there is clinical or echocardiographic evidence of device infection; 2) if there is no other source of S. aureus bacteremia identified; 3) if there is relapsing S. aureus bacteremia after a course of appropriate antimi-

1014

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

crobial therapy. Removal of pacemaker wires may be challenging because of neoendothelialization and fibrocollagenous sheath formation along the electrode. Options for removal of such leads include the use of a locking stylet introduced onto the lead and affixed close to the distal end of the electrode to apply traction directly to the tip, the use of a telescoping sheath that can be advanced over the lead to disrupt fibrous attachments of the lead to vein or cardiac tissue and to free the lead by countertraction, the use of a laser sheath to photoablate the fibrous attachments, or the use of minimally invasive video-assisted pacemaker removal under thoracoscopic guidance. The need for reimplantation should be reassessed, as 13% to 52% of patients may no longer require pacemaker support following pace-

maker removal. Device reimplantation should be at a new site when the patient is no longer bacteremic. ■

■

Staphylococci are the most common cause of pacemaker infections. When possible, all the hardware should be removed in pacemaker infections.

Left Ventricular Assist Device Infections Infection is a frequent complication of left ventricular assist device use; the risk increases with the duration of use, and most commonly occurs in patients in whom the device has been in place for at least two weeks. Infection has been reported to complicate 25% to 70% of left ventricular assist device placements.

Table 15. Nonvalvular Cardiovascular Device-Related Infections Type of device

Incidence of infection, %

Intracardiac Pacemakers (temporary and permanent) Defibrillators Left ventricular assist devices Total artificial hearts Ventriculoatrial shunts Pledgets Patent ductus arteriosus occlusion devices (investigational in the United States: plugs, double umbrellas, buttons, discs, embolization coils) Atrial septal defect and ventricular septal defect closure devices (Bard clamshell occluders, discs, buttons, double umbrellas) Conduits Patches Arterial Peripheral vascular stents Vascular grafts, including hemodialysis Intra-aortic balloon pumps Angioplasty/angiography-related bacteremias Coronary artery stents Patches Venous Vena caval filters *Closure device use ≤1.9%.

0.13-19.9 0.00-3.2 25-70 2.4-9.4 Rare Rare Rare Rare Pare Rare 1.0-6 ≤5-26 5 episodes of apnea or hypopnea per hour 1035

1036

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Fig. 1. Schematic of patent upper airways during sleep (left) and collapsed upper airways (right) causing obstructive sleep apnea.

of sleep, OSA is relatively common, affecting 24% and 9% of middle-aged men and women, respectively. Sleep apnea, particularly OSA, constitutes a major public health problem and is closely associated with obesity. Emerging evidence strongly links OSA to cardiovascular morbidity.

OBSTRUCTIVE SLEEP APNEA Acute Responses Repetitive apneas can induce nocturnal oxygen desaturation to levels as low as 40-50%. This hypoxemia, together with CO2 retention, activates the peripheral and central chemoreflexes, resulting in sympathetic activation with vasoconstriction and surges in blood pressure. Blood pressure can reach levels as high as 240/130 mm Hg (Fig. 2), at a time when there is severe simultaneous hypoxic and neurohumoral stress on homeostatic mechanisms. Also implicated in the acute effects of OSA are the hemodynamic and cardiac structural consequences of inspiration against a closed airway, also known as the Mueller maneuver. The intrathoracic negative pressure generated during obstructive apneas, which can reach up to −60 mm Hg, may elicit consequences such as

increased ventricular load and elevated left ventricular transmural pressure, thereby increasing cardiac wall stress. These hemodynamic effects may be particularly important in OSA patients with coexisting heart disease. Hypoxemia may also activate the diving reflex, which is able to simultaneously elicit both sympathetic vasoconstriction to multiple vascular beds (with the exception of the heart and the brain) and vagal activation to the heart. Thus, in perhaps ~10% of sleep apnea patients, the nocturnal apneas may be associated with severe bradyarrhythmias including AV block and occasionally sinus arrest. Severe sleep apnea may also acutely trigger the release of several vasoactive substances, including catecholamines, atrial natriuretic peptide, and endothelin (Fig. 3). Sustained Effects of OSA Persisting Into Daytime Wakefulness There is considerable evidence that the acute pathophysiologic mechanisms activated by repetitive apneas may carry over into daytime disease processes. Even during normoxic wakefulness, patients with OSA have higher sympathetic activity, faster heart rates, diminished heart rate variability and increased blood pressure variability. High sympathetic drive may in part be due to

Chapter 85 Sleep Apnea and Cardiac Disease

1037

Fig. 2. Neural and circulatory changes in obstructive sleep apnea. Recordings of sympathetic nerve activity, breathing, and intra-arterial blood pressure in the same individual when awake, with OSA during rapid eye movement (REM) sleep, and with elimination of OSA episodes by continuous positive airway pressure (CPAP) therapy during REM sleep. Sympathetic nerve activity is very high during wakefulness, but increases even further secondary to obstructive apnea during REM sleep. Blood pressure increases from 130/65 mm Hg when the patient is awake to 256/110 mm Hg at the end of the apneic episode. Elimination of apneic episodes by CPAP therapy results in decreased sympathetic activity and prevents blood pressure surges during REM sleep.

tonic chemoreflex activation, since chemoreflex deactivation by 100% O2 lowers sympathetic drive, slows heart rate, and lowers blood pressure in OSA patients. There is also evidence for a chronic systemic inflammatory state, as indicated by higher levels of Creactive protein and of serum amyloid A (Fig. 4). In addition, there is evidence of increased leukocyte activation and binding to endothelial cells. Inflammatory and other mechanisms that may damage the endothelium can contribute to resistance vessel endothelial dysfunction in OSA. OSA and Cardiovascular Disease Conditions Hypertension While many reports have shown an association between both pulmonary and systemic hypertension with OSA, studies from the Wisconsin Sleep Cohort have provided the first prospective evidence of a dose response association between the severity of OSA and the likelihood for development of systemic hypertension

4 years later in subjects who were normotensive at baseline (Fig. 5). This association was independent of other known risk factors such as baseline blood pressure, body mass and habitus, age, gender and alcohol and cigarette use. No single causal factor responsible for the occurrence of incident hypertension in OSA has been identified. The mechanisms are probably multifactorial and include activation of the sympathetic nervous system, endothelial dysfunction (including that caused by systemic inflammation), and increased endothelin, all of which would potentiate vasoconstriction. Treatment of obstructive sleep apnea may induce decreases in blood pressure not only at night but also during the daytime. Sleep apnea should be especially considered in patients with resistant hypertension, particularly if they are also obese. Sleep apnea should also be suspected in patients who are “nondippers,” i.e. in those in whom blood pressure does not fall appropriately during the night. The most recent JNC Guidelines have listed sleep apnea as an important identifiable cause of hypertension.

1038

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Fig. 3. Intermediary mechanisms associated with obstructive sleep apnea that potentially contribute to risk of cardiovascular disease. Abnormalities associated with obstructive sleep apnea may be intermediary mechanisms that contribute to the initiation and progression of cardiac and vascular pathology. These mechanisms may interact with each other, thus potentiating their pathophysiological implications.

Atherosclerosis and Coronary Artery Disease In the presence of coexisting coronary artery disease, OSA may trigger acute nocturnal cardiac ischemia with ST-segment depression. Such nocturnal exacerbation of ischemic heart disease by OSA may be a result of oxygen desaturation, high sympathetic activity, increased cardiac oxygen demand (due to tachycardia

and increased systemic vascular resistance), as well as a prothrombotic state. Whether these mechanisms may also lead to coronary plaque rupture and an acute coronary event is not known. Nevertheless, from the clinical perspective, OSA should be considered in the differential diagnosis of patients with evidence of cardiac ischemic events triggered at nighttime. Indeed, it has been

Fig. 4. Box plot showing plasma CRP in OSA patients (n=22) and controls (n=20). Middle horizontal line inside box indicates median. Bottom and top of the box are 25th and 75th percentiles, respectively.

Fig. 5. Adjusted odds ratios for the presence of incident hypertension at 4-year follow-up according to the apnea-hypopnea index (AHI) at baseline. Data are shown as odds ratio (line bars indicate lower and upper 95% confidence intervals). P for trend=0.002.

Chapter 85 Sleep Apnea and Cardiac Disease proposed that OSA may be a poor prognostic factor in patients with coronary artery disease. 5-Year mortality in those with OSA was 38% compared to 9% mortality in patients without OSA. There is epidemiological evidence that OSA may be etiologically linked to the development of atherosclerosis. There is a high prevalence of OSA in patients with coronary artery disease, and several case-control or prospective studies implicate OSA as an independent predictor of coronary artery disease. Observational studies suggest that treatment of OSA in patients with stable coronary artery disease by continuous positive airway pressure may decrease the incidence of new cardiovascular events. The exact mechanisms of any atherogenic effects of OSA have not been established. However, recent reports linking OSA to inflammation lend support to the possibility that systemic or local inflammation may play a direct role in atherogenesis in OSA subjects. The role of oxidative stress in the vascular pathophysiology of OSA remains controversial. Finally, OSAinduced hypertension may contribute to endothelial damage and thereby to atherosclerosis. Atrial Fibrillation and Bradyarrhythmias In an unselected population of atrial fibrillation patients presenting for cardioversion, the risk for OSA (assessed using the Berlin questionnaire) is approximately 50%, as compared to a 30% risk in a general cardiology clinic population. Patients with OSA undergoing coronary artery bypass grafting have an increased likelihood of post-op atrial fibrillation. Furthermore, in atrial fibrillation patients undergoing cardioversion, the presence of untreated OSA was associated with a twofold greater risk of one year recurrence of atrial fibrillation (82%) as compared to the recurrence risk in those OSA patients receiving appropriate therapy (42%) (Fig. 6). Several pathophysiological mechanisms may contribute to the association between OSA and atrial fibrillation. OSA-related hypoxia, atrial stretch, sympathetic activation, acute blood pressure surges, and increased C-reactive protein may all reasonably be expected to predispose to increased risk for atrial fibrillation. Other arrhythmias frequently associated with OSA are sinus arrest, sinoatrial block, and atrioventricular block, all of which may lead to ventricular asystole. The mechanism of these bradyarrhythmias is usually a reflex

1039

Fig. 6. Recurrence of atrial fibrillation at 12 months comparing patients who did not have sleep studies (controls) with treated OSA patients and with untreated (including noncompliant) OSA patients (mean±SD).

increase in cardiac vagal tone triggered by a combination of apnea and hypoxemia. It occurs in the presence of peripheral vasoconstriction (due to increased sympathetic outflow to resistance vessels), and represents the response to activation of the diving reflex by simultaneous hypoxemia and apnea. Arryhthmias and/or cardiac ischemia may contribute to the increased likelihood of sudden cardiac death during the nighttime hours in patients with OSA. In contrast to patients without OSA, who are more likely to experience sudden cardiac death in the morning hours between 6 am and noon, about 50% of sudden cardiac deaths in OSA patients occur at night, between 10 pm and 6 am. Stroke The association between OSA and atrial fibrillation may in part explain the increased stroke risk in OSA. Atherosclerosis, vascular inflammation, hypertension, and procoagulant effects of OSA are also important risk factors for stroke. Furthermore, the decreased cardiac output and increased intracranial pressure during acute OSA episodes, with a consequent reduction in cerebral blood flow, may also play an important role. However, the nature of the association between stroke and OSA remains unclear. It is known that patients with stroke have a high prevalence of OSA, exceeding 40%. What is uncertain is whether OSA is a causal factor in the

1040

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

occurrence of stroke, or whether sleep apnea is a consequence of stroke and results from stroke-induced impairment of respiratory and muscle tone control. Also, whether patients with transient ischemic attacks who also have OSA have an increased risk for stroke compared to those without OSA remains controversial. Nevertheless, recent prospective observational data show the presence of severe OSA to be an important predictor of risk for new stroke. What is also clinically important is that OSA in stroke survivors could further compromise physical and cognitive functions, and have detrimental effects on prognosis.The diagnosis of sleep apnea in stroke patients should be pursued and, if indicated, appropriate therapy should be instituted. Heart Failure The acute hemodynamic and cardiac structural effects of OSA, together with adrenergic activation, systemic inflammation, increased endothelin, endothelial dysfunction, hypertension, and ischemic heart disease, are some of several mechanisms that may underlie the association between OSA and chronic heart failure, as well as between OSA and acute exacerbations of heart failure. Heart failure, by virtue of soft tissue edema, may predispose to OSA. While it is not clear whether the prevalence of OSA in heart failure is greater than the prevalence of OSA in the general population, OSA should be suspected in heart failure patients who are obese. Treatment of OSA may significantly improve clinical status. The very high risk for development of heart failure over 20-year follow-up noted in the Framingham population may in part be explained by occult sleep apnea in obese subjects. Diagnosis and Treatment of OSA OSA should always be considered in patients with risk factors for OSA (especially obesity, age, and male gender) as well as in those with symptoms suggestive of a sleep disorder, including daytime sleepiness, fatigue, snoring and witnessed cessation of breathing during sleep. It should also be suspected in patients with specific clinical features, such as resistant hypertension (especially the “non-dipping pattern”), resistant heart failure with frequent nocturnal exacerbations, sleep-time cardiac ischemia, recurrent atrial fibrillation, stroke, etc. The risk for OSA can be further assessed by overnight oximetry (severe and repetitive oxygen desaturations

being suggestive of OSA) and by using questionnaires (e.g. the Berlin questionnaire), which contain scoring systems indicating the probability of OSA. However, the definitive diagnosis can be made only by polysomnography, which is usually performed in a specialized sleep center. The severity of OSA will often be attenuated by behavioral and lifestyle modifications, such as weight loss, avoidance of sedatives and alcohol, and avoidance of sleeping on the back. In selected patients surgical procedures designed to increase the diameter of the upper airway (such as uvulopalatopharyngoplasty and laser assisted uvuloplasty) can be used, but beneficial effects on OSA may be transient and unpredictable. Other treatment options include dental applicances which serve to minimize posterior displacement of the mandible during sleep, and may be helpful in less severe OSA. The treatment of choice in OSA remains continuous positive airway pressure (CPAP), which should be used in patients with moderate to severe OSA, and perhaps also in those with mild OSA. CPAP therapy is associated not only with a decrease in OSA severity and OSA-related symptoms (leading to a marked improvement in the quality of life) but it can also favorably affect the risk and the course of OSA-related cardiovascular disease. Specifically, in OSA patients with hypertension, effective CPAP treatment has been shown to significantly reduce daytime and nocturnal blood pressure (Fig. 7). Nocturnal ST-segment depression and nocturnal angina might be significantly reduced after CPAP therapy. Observational data suggest that CPAP treatment of patients with severe OSA is accompanied by a reduction in acute cardiovascular events. CPAP has also been shown to reduce the recurrence of atrial fibrillation in patients undergoing cardioversion. Treatment of OSA in heart failure has been associated with improvements in ejection fraction, lower blood pressure and slower heart rate. However, whether treatment of OSA results in any mortality benefit in heart failure or in any other OSA-related cardiovascular conditions remains unknown. An important and as yet unresolved question is how to treat those OSA patients who are not compliant with CPAP therapy or in whom CPAP is not effective in alleviating OSA severity. In patients with life threatening OSA, who cannot tolerate CPAP, tracheostomy

Chapter 85 Sleep Apnea and Cardiac Disease

Fig. 7. Changes in mean (MAP), systolic, and diastolic blood pressure with effective (closed bars) and subtherapeutic (open bars) continuous positive airway pressure (CPAP). *Significant difference.

should be considered. Better understanding of the biology of OSA may help in developing new pharmacologic and nonpharmacologic therapies acting on specific pathophysiological pathways involved in the development of OSA and/or OSA-associated cardiovascular disease.

CENTRAL SLEEP APNEA Acute and Chronic Effects Many acute and sustained effects of CSA are similar to those of OSA as described earlier. These effects are related primarily to chemoreceptor activation by hypoxia, and lead to elevated sympathetic drive and an increase in circulating catecholamines. Similarly, frequent arousals and sleep fragmentation due to apneas can result in sleep deprivation and daytime fatigue and somnolence. However, in contrast to OSA, CSA is not accompanied by airway occlusion and strenuous respiratory efforts, and thus hemodynamic effects related to changes in intrathoracic and cardiac transmural pressures are modest, and hypoxemia is usually less marked. The Association Between CSA and Cardiovascular Disease The main clinical association of CSA is with chronic heart failure (CHF), although it can also occur in healthy individuals during exposure to hypobaric

1041

hypoxia at altitudes. In the setting of CHF, CSA occurs in the form of Cheyne-Stokes respiration, characterized by repetitive episodes of apnea/hypopnea and hyperpnea (i.e. hyperventilation) manifested as crescendodecrescendo (a waxing-waning pattern) changes in tidal volume (periodic breathing). Although the pathophysiology of CSA in CHF is not fully understood, its etiology is probably multifactorial and includes instability in the respiratory rhythm control (such as heightened chemoreceptor drive and fluctuating arterial CO2 levels), upper airway narrowing, pulmonary venous congestion (resulting in reflex afferent stimulation of pulmonary mechanoreceptors), and prolonged circulation time, which leads to a delay in sensing changes in arterial blood gases by chemoreceptors. It occurs predominantly during stage I and II of non-REM sleep, when ventilation is regulated by the levels of arterial CO2. The prevalence of CSA in CHF has been estimated to be between 33 to 70 percent of patients with stable systolic heart failure, although it depends on various factors including heart failure etiology, gender (being more common in men), age, ejection fraction and hemodynamic status. What is important is that CSA is frequent not only in patients with advanced CHF but also in those with asymptomatic left ventricular dysfunction. The clinical significance of CSA in CHF is twofold. First, CSA is associated with the severity of CHF. The nature of this reciprocal association between CSA and CHF severity is complex and not fully understood. On the one hand, CHF patients with CSA are characterized by lower exercise capacity and ejection fraction, increased left ventricular volumes, elevated pulmonary capillary wedge pressure, and a higher prevalence of cardiac arrhythmias, indicating that the presence of CSA may be merely an index of CHF severity. On the other hand, it is conceivable that CSA may lead to the progression of CHF through several mechanisms, such as neuroendocrine effects (elevated catecholamines), hypoxia, blood pressure and heart rate fluctuations, cardiac arrhythmias etc. In fact, even in patients with asymptomatic left ventricular dysfunction, CSA is accompanied by impaired cardiac autonomic control and increased cardiac arrhythmias, suggesting that perhaps CSA may precede the development of overt heart failure. Secondly, irrespective of the mechanisms involved,

1042

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

the presence of CSA is an important risk factor and impairs prognosis in CHF—an effect that is independent of other known risk factors, such as left ventricular ejection fraction, hemodynamic parameters, or peak oxygen consumption. For example, in clinically stable patients with CHF, CSA has been shown to be an independent predictor of cardiac death and transplantation during 2 year follow-up. Treatment The primary goal of treating CHF patients with CSA is optimization of pharmacological heart failure therapy. Decreased CHF severity and hemodynamic improvement of CHF is often associated with a significant decrease in CSA. More aggressive treatment of CSA may be indicated in case of persistent CSA and refractory CHF. Theophylline or nocturnal oxygen supplementation or acetazolamide have been suggested to decrease the severity of CSA, but their long-term effects on outcome are not known. Pilot studies suggest that CPAP therapy may improve transplant-free survival, but the recent CANPAP trial showed no mortality benefit from CPAP treatment of CSA in CHF patients. However, the efficacy of CPAP in reducing CSA was limited, with suboptimal patient compliance, which may help explain the absence of benefit.

A new potentially promising therapeutic strategy is cardiac resynchronization therapy, which has been suggested to have beneficial hemodynamic effects in CHF as well as decrease the severity of CSA.

CONCLUSIONS Both OSA and CSA are associated with cardiovascular disease. While a causal relationship between sleep apnea and cardiovascular disease is not definitively proven, the coexistence of sleep apnea with cardiovascular disease exacerbates symptoms and may accelerate progression. The diagnosis of sleep apnea should always be considered in cases of refractory heart failure, resistant hypertension, transient ischemic attacks, and nighttime cardiac ischemia or arrhythmias, especially in persons with risk factors for sleep apnea. Treating sleep apnea may improve CV disease management. The diagnosis and treatment of sleep apnea is important even in the absence of any clinically overt cardiovascular disease, because sleep apnea might be conducive to their development. Randomized trials examining the cardiovascular effects of treatment of sleep apnea are lacking. Whether treatment of sleep apnea truly prevents cardiovascular events and has a mortality benefit remains unknown.

86 CARDIOVASCULAR TRAUMA Joseph G. Murphy, MD R. Scott Wright, MD

Cardiovascular trauma is a significant cause of death particularly in young men in our society. Cardiovascular trauma is classified into penetrating injuries, blunt, nonpenetrating trauma, and medical injuries to the heart sustained at the time of an invasive cardiovascular procedure, medical device implantation, or cardiopulmonary resuscitation. Penetrating cardiac injury is due primarily to knife or gunshot injuries, whereas blunt cardiac injury is usually due to automobile or motorcycle accidents or industrial incidents. Iatrogenic cardiac trauma also may occur as a result of cardiopulmonary resuscitation, endomyocardial biopsy, or the use of intravascular catheters, including Swan-Ganz catheters (Table 1). ■

■

■

Males between the ages of 15 and 35 are the most common victims of cardiac trauma. Nonpenetrating cardiac trauma usually results from automobile or industrial injuries, whereas penetrating cardiac injuries usually result from knife or gunshot wounds. 50% of patients with traumatic penetrating cardiac injury die rapidly, usually before hospitalization.

PENETRATING CARDIAC INJURY Penetrating cardiac trauma most commonly affects the right ventricle, followed in order by the left ventricle, right atrium, and left atrium. Cardiac injury may occur from direct injury or indirectly from rib fractures that puncture the cardiac chambers. The principal consequences of perforating cardiac injury are cardiac tamponade and exsanguinating hemorrhage, both of which lead to death rapidly if not treated on an emergency basis. Whether cardiac tamponade develops will depend on the chamber penetrated, the size of the penetration, and whether the pericardium is also lacerated. The left ventricle is usually capable of sealing a small hole because of the thickness of the surrounding muscle, whereas a perforation of the right atrium or right ventricle usually leads to rapid hemopericardium. If the pericardium also is opened by the initial injury, tamponade usually will be prevented and the bleeding will present as a hemothorax. Occasionally, the pericardial tear also may act as a flap valve and prevent blood drainage into the pleural space and lead to tamponade.The signs and treatment of pericardial tamponade are discussed in the chapter on pericardial disease (Fig. 1). 1043

1044

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Table 1. Traumatic Cardiac Lesions I.

II.

III. IV.

V. VI.

■

Pericardial A. Hemorrhagic pericarditis B. Pericardial laceration C. Tamponade D. Purulent pericarditis, due to associated rupture of esophagus G. Constrictive pericarditis H. Intrapericardial diaphragmatic hernia Myocardial A. Contusion B. Ischemic infarction, secondary to traumatic occlusion of coronary artery C. Myocardial hematoma D. Myocardial laceration E. Myocardial rupture F. Aneurysm G. Pseudoaneurysm H. Diffuse calcification (“myocarditis ossificans”) Endocardial A. Thrombus Valvular A. Atrioventricular valves (chordal rupture, papillary muscle rupture, torn leaflet) B. Semilunar valves (avulsion of cusp, avulsion of commissure, torn cusp, intimal tear in adjacent aorta with cusp displacement, sinus of Valsalva aneurysm with cusp displacement) Coronary artery (laceration, arteriovenous fistula) Aorta and pulmonary artery A. Rupture B. Aneurysm

Cardiac injury may occur from direct injury or indirectly from rib fractures that puncture the cardiac chambers.

BLUNT CARDIAC INJURY Nonpenetrating cardiac trauma may result from a direct force on the chest wall or indirectly from pressure on the abdomen displacing a large volume of blood suddenly into the heart. Both forms of injury are

Fig. 1. Motor vehicle accident resulting in cardiac trauma and fatal hemopericardium.

frequent in automobile accidents.Nonpenetrating cardiac trauma may result in myocardial contusions; chamber or vessel lacerations; rupture of chordae tendineae, papillary muscles, or cardiac valves; pericarditis; pericardial lacerations; and, rarely, the late development of constrictive pericarditis.

PERICARDIAL INJURY Pericardial injury may result from blunt or pentrating chest injuries and may lead to rapid death due to cardiac tamponade or slowly progressive pericardial inflammation and fibrosis leading to late pericardial constriction. Delayed pericardial tamponade and localized pericardial tamponade are variants of pericardial injury that are often difficult to diagnose.

MYOCARDIAL CONTUSION Myocardial contusion is primarily a pathologic diagnosis, and there are no definitive clinical methods to establish this diagnosis. Although late complications from blunt cardiac trauma may occur in a small number of patients, in the majority of patients with cardiac trauma who present with stable vital signs, a short period of electrocardiographic monitoring (about 24 hours) is usually sufficient to determine whether arrhythmia or heart failure will develop. Sudden cardiac death due to

Chapter 86 Cardiovascular Trauma

1045

ventricular fibrillation may occur with low-energy impact to the chest wall (such as from a baseball impact) if the blow coincides with a narrow window of time during cardiac repolarization. Damage to the coronary artery may result in occlusion, laceration, or, more rarely, fistula formation. Rupture of the intraventricular septum, myocardial aneurysm, and pseudoaneurysm formation also have been reported. ■

Damage to the coronary artery may result in occlusion, laceration, or, more rarely, fistula formation.

MEDICAL CARDIAC INJURY CPR frequently results in nondisplaced rib fractures and may also lead to rupture of the left ventricle or right ventricle if performed too vigorously, especially in the setting of a recent myocardial infarction when softening of the myocardium has occurred. Other complications include rupture of the papillary muscles that support the tricuspid valve, resulting in severe tricuspid regurgitation, and, more rarely, rupture of the aorta. Penetrating cardiac injuries caused by medical trauma occur during endomyocardial biopsy of the right ventricle or after perforation with a temporary pacemaker wire. Dissection of the aorta or coronary ostia may occur during coronary angiography. In rare cases, coronary angioplasty has resulted in coronary artery rupture and tamponade. Intra-aortic balloon counterpulsation also may cause aortic dissection, but it is more likely to cause thromboembolic complications as a result of pulsation against the atheromatous plaques in the aorta. Indwelling venous catheters may migrate and perforate the pulmonary arteries, and improper use of balloon-tipped pulmonary artery catheters may lead to branch pulmonary artery rupture and intrapulmonary hemorrhage (Fig. 2).

DIAGNOSIS OF CARDIAC INJURY Cardiac trauma must always be suspected in the setting of blunt or penetrating trauma to the chest or abdomen. A rapid assessment of the patient (airways, breathing, circulation [ABC]), neck veins, and extremities, looking for clues to tamponade or hemorrhagic shock, is mandatory. Cardiac contusion is frequently unrecognized. Myocardial contusion may lead to regional wall

Fig. 2. Catheter perforation of right atrium with tamponade.

motion abnormalities and associated hemorrhagic infiltrate and myocyte necrosis on histologic examination. Acute heart failure and ventricular arrhythmias may occur, but they usually resolve within hours or days. Cardiac injury can occur in the absence of sternal or rib fractures or other significant chest injuries. In cases of blunt cardiac injury, the electrocardiogram may show 1) nonspecific ST-T wave changes, 2) electrocardiographic changes of acute pericarditis, or 3) pathologic Q waves. An increase in the troponin level confirms the presence of cardiac injury. Echocardiography is the imaging method of choice for identification of cardiac injury. The findings include pericardial contusion, pericardial tamponade, regional wall motion abnormalities,chamber enlargement, valvular incompetence, and the presence of intracardiac shunts. An adequate transthoracic echocardiographic examination is not possible in up to 30% of trauma victims, and transesophageal echocardiography may be needed. Transesophageal echocardiography may not be possible in patients with cervical, maxillary, or mandibular injuries (Fig. 3). ■

■

Cardiac injury can occur in the absence of sternal or rib fractures or other significant chest injuries. Echocardiography is the imaging method of choice for identification of cardiac injury.

TREATMENT OF CARDIAC INJURY Emergency pericardiocentesis for cardiac tamponade

1046

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

DAMAGE TO INTRACARDIAC STRUCTURES

Fig. 3. Traumatic right ventricular contusion and tricuspic valve disruption; modified right ventricular inflow view at 0°multiplane imaging. The right ventricle (RV) is markedly enlarged and was nearly akinetic on realtime examination. The anterior tricuspid leaflet (large arrow) is unsupported and flail; both the anterior and the septal (small arrow) leaflets are dwarfed by profound tricuspid annular dilatation, which is responsible for an expansive gap between the noncoapting leaflets. RA, right atrium.

may be life-saving if the patient is hemodynamically unstable. Echocardiography-guided pericardiocentesis is preferred if immediately available. Emergency thoracotomy is the treatment of choice for severe hemorrhage due to cardiac trauma. Lesser degrees of cardiac contusion may be managed conservatively. For hypotensive patients who have multiple-injury trauma and do not respond to fluids, consider cardiac tamponade or ventricular hypokinesia—both conditions are easily diagnosed with echocardiography. Inotropic agents and intra-aortic balloon counterpulsation (provided there is no aortic injury) may be beneficial in patients with ventricular hypokinesia due to myocardial contusion. Traumatic rupture of the atrial or ventricular septum or major valve injury generally requires surgical repair. Traumatic cardiac rupture due to blunt trauma occurs most commonly in the right or left ventricle. Late cardiac rupture may occur from contusion complicated by intramyocardial hemorrhage, necrosis, and softening. Emergency operation is the treatment of choice for patients with cardiac rupture. Patients who present with pseudoaneurysm formation should have surgical repair because future rupture is unpredictable.

The aortic valve is the most frequently damaged valve in nonpenetrating chest injuries. Patients with underlying valvular heart disease are considered to be at a higher risk than those without preexisting disease. Aortic or mitral incompetence due to valve leaflet tears usually presents early and worsens with time. Tricuspid valve injury is unique in that it may be recognized only years after the original injury. Aortic incompetence may result from a combination of damage to the aortic wall and damage to the valve leaflets, and it may improve when perivalvular edema and hemorrhage subside. Sudden traumatic obstruction to left ventricular outflow during systole may result in papillary muscle or mitral valve rupture. The risk of cardiac valve injury is dependent on the time at which the injury occurs during the cardiac cycle. Injury during systole damages the mitral valve, whereas injury during diastole damages the aortic valve. Severe abdominal injury, even in the absence of chest trauma, may result in tricuspid valve or right ventricular papillary muscle rupture. Definitive treatment for significant valve injury is valve replacement or repair. Injury to the coronary arteries from blunt or penetrating trauma may lead to coronary occlusion and myocardial infarction. Left ventricular pseudoaneurysm or aneurysm formation may result from coronary trauma. Atrioventricular fistula formation is a rare complication of penetrating trauma and most commonly affects the right coronary artery. The fistula may extend from the right coronary artery into the coronary sinus, the great cardiac vein, the right ventricle, or the right atrium (Fig. 4). ■

■

The risk of cardiac valve injury is dependent on the time at which the injury occurs during the cardiac cycle. Atrioventricular fistula formation is a rare complication of penetrating trauma and most commonly affects the right coronary artery.

INJURY TO THE AORTA AND GREAT VESSELS Rupture of the aorta is the most frequent nonpenetrating injury to the great vessels and occurs in about 8,000 cases annually in the United States. Traumatic rupture of the aorta occurs after rapid deceleration injuries such as high falls or automobile accidents. Traumatic aortic

Chapter 86 Cardiovascular Trauma

1047

Fig. 4. Traumatic rupture of right coronary artery. (x5.)

rupture is present in about 20% of patients who die of injuries from motor vehicle accidents. The site of the aortic tear is usually the junction of the aortic arch and the descending aorta at a point where the descending aorta is fixed to the spine by the intercostal arteries just distal to the origin of the left subclavian artery. This injury is usually fatal in 80% to 90% of patients, but survival has been reported with emergency cardiac operation. Partial rupture of the aorta is associated with increased arterial pressure in the upper extremities, decreased arterial pressure in the lower extremities, and evidence of widening of the superior mediastinum on chest radiography or computed tomography. Pseudoaneurysm of the aorta tends to expand and rupture, but it also may contain thrombus that embolizes to distant sites. Fistulas may form to adjoining structures. Transesophageal echocardiography, computed tomography, and aortic angiography are the imaging methods of choice in cases of suspected aortic injuries (Fig. 5). The commonest angiographic findings are an intimal flap and a pseudoaneurysm. With aggressive surgical intervention, about 80% of patients who reach a hospital will survive; without surgery, 2% to 5% of patients will develop a chronic pseudoaneurysm (Figs. 6 and 7). ■

The site of the aortic tear is usually the junction of the aortic arch and the descending aorta.

ELECTRICAL INJURY TO THE HEART Electrical injury to the heart may occur from a lightning strike (about 100 deaths annually in the United States), industrial, home electrical accident, or from an

Fig. 5. Traumatic rupture of the descending thoracic aorta after a motor vehicle accident; transverse transesophageal echocardiographic plane. A large rent in the aorta (Ao) is clearly visualized (arrow) communicating with an adjacent para-aortic space (arrowheads); there is also hematoma formation (H).

Fig. 6. Motor vehicle accident with aortic transection just distal to the ligamentum arteriosum.

electronic stun gun (Taser). Ventricular fibrillation is the usual cause of death. Drug intoxication with cocaine or phencyclidine (PEP) may increase the lethality of an electrical injury by increasing catecholamines and predisposing to fibrillation.

1048

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Fig. 7. Wide mediastinum (on chest radiograph) in acute aortic dissection with rupture and hemopericardium.

87 ACUTE BRAIN INJURY AND THE HEART Nandan S. Anavekar, MB, BCh Sarinya Puwanant, MD Krishnaswamy Chandrasekaran, MD

Neurological and cardiovascular functions are closely intertwined: cerebral perfusion is dependent on cardiac performance while much of cardiac function is regulated by higher brain centers. Dysfunction in either organ system can precipitate dysfunction in the other. Acute brain injury may result from vascular injury (stroke, subarachnoid or parenchyma hemorrhage), trauma (closed head injury), and inflammation (encephalitis, meningitis). Cardiac sequelae may manifest as fluctuations in blood pressure and hemodynamics, electrocardiographic changes, arrhythmias, and elevations in cardiac biomarkers.

CARDIAC INNERVATION The autonomic nervous system (Fig. 1) strongly influences the electrical and mechanical activities of the heart; it is composed of two broad categories of efferent pathways, namely the parasympathetic and sympathetic nervous system. These efferent pathways are modulated by higher brain centers which constitute a functional unit referred to as the central autonomic network. Neurons in the cerebral cortex, basal forebrain hypothalamus, midbrain, pons, and medulla participate in autonomic control.The central autonomic network integrates visceral, humoral, and environmental information

to produce coordinated autonomic, neuroendocrine, and behavioral responses to external or internal stimuli. Parasympathetic Innervation of the Heart The heart is innervated by both arms of the autonomic nervous system. The parasympathetic preganglionic neurons originate in the nucleus ambiguus of the medulla and synapse with intracardiac ganglia, via the vagus nerve. From these ganglia, short postganglionic parasympathetic neurons innervate the myocardial tissue. The parasympathetic innervation of the heart is particularly abundant in the sinus node and atrioventricular conduction system. Parasympathetic innervation of the heart is mediated entirely via the vagus nerve.The right vagus nerve innervates the sinoatrial node and when stimulated excessively predisposes to sinus node bradyarrhythmias. The left vagus nerve innervates the atrioventricular node and when hyperstimulated predisposes the heart to atrioventricular (AV) blocks. The parasympathetic postganglionic neurons release acetylcholine which activates M2 muscarinic receptors in the heart, the effects of which result in slowing of the heart rate, reduced contractile forces of the atrial cardiac muscle, and slowing of conduction velocity through the atrioventricular node (AV node). Vagal stimulation has no effect on ventricular muscle function. Excessive 1049

1050

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Fig. 1. Demonstrates the innervation of the heart.

Chapter 87 Acute Brain Injury and the Heart vagal tone during emotional stress, which is usually a parasympathetic overcompensation to strong sympathetic activation during stress, can cause syncope because of a sudden drop in blood pressure and heart rate. Patients with bulimia and anorexia or spinal cord injury may have high vagal activity which is associated with the cardiac arrythmias often seen in these patients. Sympathetic Innervation of the Heart The sympathetic innervation of the heart originates from the intermediolateral column of the thoracic spinal cord and synapses with the sympathetic postganglionic neurons in the superior, middle and inferior cervical ganglia. The postganglionic sympathetic neurons innervate the sinoatrial and atrioventricular nodes, the conduction system and myocardial fibers, most prominently in the ventricles. The postganglionic sympathetic nerves release norepinephrine, which primarily through β-adrenergic receptor stimulation increases heart rate, conduction velocity through the atrioventricular node (AV node) and contractility of the heart. The sympathetic outflow to the peripheral circulation can produce vasoconstriction via activation of alpha adrenergic receptors. Such activation can increase the metabolic requirements of the heart which can manifest clinically as myocardial ischemia. The autonomic outflow to the heart and peripheral vasculature fluctuates on a moment to moment basis. It is regulated by a variety of reflexes, which are initiated by arterial baroreceptors and chemoreceptors as well as a variety of intracardiac receptors.The autonomic innervation of the heart can also be influenced by pathologic events occurring in the central nervous system which alter the balance between parasympathetic and sympathetic outflow. These alterations can ultimately lead to disturbances in cardiac function and hemodynamics.Thus cardiac innervation serves as the common link between acute brain injury and its cardiac complications.

CARDIOVASCULAR COMPLICATIONS OF CEREBROVASCULAR ACCIDENT Stroke Patients who suffer a stroke are predisposed to cardiac disturbances. It is often difficult to ascertain in an indi-

1051

vidual patient whether the cerebrovascular event was initiated by a primary cardiac disturbance or the converse namely that the stroke caused a cardiac disturbance since the prevalence of cardiac morbidity in stroke patients is high. ECG Abnormalities in Stroke ECG abnormalities are present in up to 90% of patients presenting with acute stroke. Typical ECG demonstrates large upright T-waves and prolonged QT intervals (Fig. 2). Such ECG changes are also seen in the setting of subarachnoid hemorrhage, transient ischemic attacks, and nonvascular cerebral lesions. The ECG changes have been postulated to arise from subendocardial ischemia as a result of increased centrally mediated catecholamine release in the setting of hypothalamic hypoperfusion. Repolarization Abnormality ECG repolarization abnormality manifesting as QT prolongation is seen in about 38% of stroke patients. QT prolongation increases the vulnerable period of the cardiac cycle for arrhythmias and sudden death. QT prolongation in the absence of hypokalemia in stroke patients identifies high risk patients vulnerable to arrhythmia related sudden death. QT prolongation is more common after right middle cerebral artery stroke. ST Segment Abnormality Nonspecific ST segment changes are seen in over 20% of patients presenting with acute stroke and commonly include ST segment depression, a feature more frequently seen with left middle cerebral artery strokes. Dynamic ST segment changes may also be indicative of true myocardial ischemia: ST segment changes secondary to stroke are generally transient and paradoxically improve with brain death. Q waves New Q waves occur in up to 10% of patients with acute stroke. The Q waves may be a transient feature of the ECG, or may proceed through the typical evolutionary changes seen in myocardial infarction. Q waves seen in this setting do not reflect myocardial ischemic damage. U Waves The presence of U waves is a common finding in the

1052

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Fig. 2. Electrocardiography in a patient with acute ischemic stroke demonstrates classic tall upright T wave (arrows) in leads V3-5 and prolonged corrected QT interval of 471 msec. Note occasional ventricular extrasystole (open arrow) is also seen.

setting of acute stroke.These are usually unrelated to any electrolyte abnormality and may be found alone or in concert with T wave changes or prolonged QT intervals.

Restoration of normal autonomic tone can take up to 6 months after the acute cerebrovascular event: during this period, there is an increased risk of sudden cardiac death.

Arrhythmias Cardiac arrhythmias frequently follow an acute stroke, even when not present on the admission ECG. All types of dysrhythmia can be seen including ventricular extrasystoles, atrial extrasystoles, supraventricular tachycardia, and ventricular tachycardia. Only ventricular arrhythmia is associated with increased mortality in stroke patients. Many cardiac arrhythmias occur in patients with normal cardiac function, suggesting a neurogenic etiology. Furthermore, the type and location of a stroke predicts the type of arrhythmia seen. The pathophysiologic mechanism of these dysrhythmias is postulated to be alterations in the central autonomic outflow to the heart. Bradycardia and vasodepressor effects are more common with injury to the right insular region whereas tachycardia and hypertension are more common with injury to the left insular region.

Cardiac Biomarkers Elevation in cardiac biomarkers including troponin T, creatine kinase, and myoglobin may be seen in acute stroke. The magnitude of rise is usually small. Often troponin will only become elevated in stroke patients when there is existing coronary artery disease, and then it is associated with left ventricular dysfunction and poor prognosis. Increased autonomic sympathetic discharge increases myocardial oxygen demand and may result in both myocardial stunning and elevation in cardiac biomarkers. Classical myocardial infarction occurs in up to 6% of patients with acute stroke and is a therapeutic dilemma as many myocardial infarct therapies increase the risk of intracerebral bleeding. Neurogenic LV Dysfunction “Neurogenic stunned myocardium” is a clinical term

Chapter 87 Acute Brain Injury and the Heart that describes neurologically mediated cardiac injury that is reversible in nature and clinically manifest by ventricular dysfunction with or with out hemodynamic instability. Commonly it is accompanied by ECG changes, arrhythmias, and cardiac biomarker release. This phenomenon is unrelated to any underlying coronary artery disease and is frequently seen in stroke patients, more commonly with left insular stroke. Pathologic findings include a characteristic pattern of petechial subendocardial hemorrhage and “contraction band necrosis.” Excessive autonomic sympathetic discharge and catecholamine release is thought to be the pathogenic mechanism. Catecholamine blockade is protective against neurogenic cardiac injury.Tako-tsubo syndrome (apical balloon cardiomyopathy ), a condition generally seen in elderly women that closely mimics the clinical presentation of acute anterior wall myocardial infarction is postulated to result from stress associated catecholamine released. It is associated with a characteristic pattern of abnormal ventricular wall motion, with hypokinesis of the cardiac apex and mid ventricle

A

1053

with relative sparing of the cardiac base (Fig. 3). Prognosis for recovery is excellent. Subarachnoid Hemorrhage Subarachnoid hemorrhage is associated with significant patient morbidity and mortality and accounts for 10% of all strokes. It usually results from rupture of a saccular intracerebral aneurysm, but other causes include trauma, arteriovenous malformation and illicit drug use including cocaine and amphetamine use. Almost all the changes observed in the ECG, cardiac biomarkers and LV dysfunction can be seen in SAH, however, there are certain characteristics typically seen in SAH. Acute ECG changes are noted in greater than 50% of patients with subarachnoid hemorrhage classically described as deep T wave inversions or ‘‘cerebral T waves’’ (Fig. 4 and 5). ECG changes may be seen up to 2 weeks after the acute bleed. Prolongation of the QT interval and QT dispersion are seen in >70% of patients presenting with subarachnoid hemorrhage, a finding that predisposes to ventricular arrhythmia and sudden

B

Fig. 3. Left ventriculography in a patient with apical ballooning syndrome. End diastolic (A) and end systolic (B) frames demonstrate akinesis of mid, and dyskinesis of apical regions of the left ventricle with hyperdynamic contraction of basal left ventricle. The appearance is similar to “ampulla” or “tako-tsubo”—a vessel used for catching octopus in Japan. This phenomenon can be seen in all acute brain injury situations including inflammation. However, it is more common in SAH and any ischemic stroke involving thalamus, and brainstem. In SAH the apex and base are spared and only mid ventricle demonstrates systolic ballooning (dyskinesis).

1054

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Fig. 4. Electrocardiography in a patient with a subarachnoid hemorrhage demonstrates ST elevation and T wave inversion (arrows) in leads V2-V3 suspicious for ST-elevation anteroseptal infarction. However, absence of reciprocal changes and marked prolongation of QT interval are common in SAH and help diffferentiate from ST elevation myocardial infarction (STEMI).

Fig. 5. Follow-up electrocardiography from the same patient shown in Figure 2, with subarachnoid hemorrhage, demonstrates persistence of ST elevation and changes in the T waves. Note the T waves are large and are symmetrically inverted (arrows) in leads V1-V5 with markedly prolonged corrected QT interval.

Chapter 87 Acute Brain Injury and the Heart death. A prolonged QTc interval of >440 msec identifies severe head trauma patients at risk for ventricular arrhythmias and sudden death. Elevations in cardiac enzymes are common after subarachnoid hemorrhage, the putative mechanism being excessive sympathetic discharge. The greater the troponin rise, the worse the clinical outcome. Neurogenic left ventricular dysfunction is much more common in subarachnoid hemorrhage than in ischemic stroke with an incidence of about 10%. The pathophysiology of left ventricular dysfunction is similar to that seen in stroke, namely cardiac myocyte injury from catecholamine surge due to increased sympathetic discharge. In some patients with subarachnoid hemorrhage the apex and the base of the left ventricle contract normally while the mid ventricle demonstrates akinesis, a condition Japanese authors have coined “panic myocardium.”

CARDIAC COMPLICATIONS OF HEAD TRAUMA Closed head trauma leads to 175,000 deaths and 500,000 hospitalizations per year with a peak incidence in men 15 to 24 years of age. Head trauma is associated with significant cardiac complications that can negatively impact clinical outcomes including cardiac rhythm and conduction disturbances. EKG findings include diffuse tall upright or deep inverted T waves, prolonged QT intervals, ST segment depression or elevation, and U waves. Dysrhythmias in head-injured patients often resolve when intracranial pressure is reduced. QT prolongation and associated fatal arrhythmias are more commonly associated with intracerebral hemorrhage with raised intracranial pressure. Reduction of ischemic cerebral injury is dependent on preservation of adequate cardiac output,and in the setting of concomitant cardiac dysfunction, neurologic outcome is also worsened. Brain Death and the Heart Patients with brain death following head injury, ischemic stroke or subarachnoid hemorrhage are potential organ donors for cardiac transplantation. Cardiac dysfunction from neurogenic mechanisms generally will recover with time as excess autonomic sympathetic activity subsides which may take from 72

1055

hours to 1 week. Myocardial recovery is dependent upon the maintenance of adequate mean arterial pressure to ensure coronary perfusion. β-Blockers can protect the myocardium from the toxic effects of catecholamines and the use of glucocorticoids has also been shown to reduce cardiac dysfunction after brain death. Head Injury and Anticoagulation Head injury in patients who are on anticoagulation adversely effects survival by increasing the incidence of intracranial hemorrhage. An international normalized ratio (INR) greater than 3.3 has been shown to be associated with an increased incidence of intracerebral hemorrhage following a head trauma and an adverse outcome.

CARDIAC MANIFESTATION IN EPILEPSY Cardiovascular manifestations of epileptic seizures are common and symptoms and signs can occur in the preictal, ictal or post-ictal period.The typical cardiovascular effects of alterations in autonomic function include changes in heart rate, blood pressure and ECG changes. ECG manifestations of epilepsy include STdepression, ST elevation, and T-wave inversion. ECG changes can occur in the absence of changes in cardiac rhythm, in the setting of a seizure. Sinus tachycardia is seen in greater than 60% of patients. Sinus tachycardia accompanied with peripheral vasoconstriction and an increase in blood pressure is associated with hypothalamic lesions. Bradyarrhythmias occur in less than 5% of seizures and include sinus bradycardia, sinus arrest, AV block and prolonged asystole. Other cardiac arrhythmias seen in the setting of seizures include acceleration and deceleration of heart rate, enhanced sinus arrhythmia, atrial premature beats, sinus pauses, AV block, nodal escape, paroxysmal supraventricular tachycardia and ventricular ectopy. Electroconvulsive Therapy Electroconvulsive therapy (ECT) is an artificially induced seizure often associated with a significant catecholamine elevation and autonomic discharge particularly parasympathetic discharge. ECG changes occur fairly commonly during treatment with ECT and are similar to those changes seen in primary seizure disorders. There has been one prospective study analyzing the effects of

1056

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

ECT on left ventricular systolic function which showed that ECT associated left ventricular systolic dysfunction is a common but transient early phenomenon. Multiple ECT treatments do not have a cumulative effect on ventricular function, and tolerance to shocks appears to develop after multiple treatments.

CARDIAC COMPLICATIONS OF ENCEPHALOMYELITIS Encephalomyelitis is an inflammatory disorder of the central nervous system,dorsal root ganglia,and autonomic

nerves due to an infectious or noninfectious etiology, the latter often associated with vasculitis or paraneoplastic syndromes. Encephalomyelitis may be associated with cardiac manifestations owing to profound disturbances in autonomic function that are common in this condition. These findings include hypertension, tachycardia and high plasma catecholamine levels, consistent with a hypersympathetic state. The plethora of cardiac manifestations in brainstem encephalitis is explicable by the density of autonomic fibers that traverse the brainstem and the location of the primary vasomotor area.

88 NONCARDIAC ANESTHESIA IN PATIENTS WITH CARDIOVASCULAR DISEASE Laurence C. Torsher, MD

The perioperative period stresses the cardiovascular system due to hemodynamic and neuroendocrine physiologic changes induced by the trauma attendant to the surgical interventions, fluid compartment shifts, blood loss, as well as anesthetic medications.

Preoperative evaluation, in addition to giving information about operative risk, should also provide information that will affect perioperative management decisions, e.g. decision to forgo or modify proposed surgical procedure, delay a procedure to optimize the

PREOPERATIVE PREPARATION

Table 1. ASA Physical Status Classification System*

Screening and Mitigation of Perioperative Cardiovascular Risk Preoperative patient assessment has the role of stratifying patients into risk groups such that physicians and patients can make informed decisions about the risks and benefits of proposed surgery. Preoperative assessment should also provide information to the perioperative caregivers that will allow them to optimize care of the patient. Risk scoring systems like the Goldman and Detsky systems allow caregivers to identify patients at high or low operative risk but do not provide information to optimize patients care.The American Society of Anesthesiologists’ classification (Table 1) is an extremely simple global risk stratification scheme that is based solely on functional status. In spite of its simplicity it has proven to be as sensitive as many more sophisticated schemes.

ASA-I: Healthy patient with no systemic disease ASA-II: Mild systemic disease, no functional limitations ASA-III: Moderate to severe systemic disease, some functional limitations ASA-IV: Severe systemic disease, incapacitating, and a constant threat to life ASA-V: Moribund patient, not expected to survive >24 hours without surgery ASA-VI: Brain-dead patient undergoing organ harvest E: Added when the case is emergent *The American Society of Anesthesiologists physical

status classification, developed in 1941, is used for risk stratification in outcome studies.

1057

1058

Section VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

patient’s medical conditions, choice of intra- and postoperative monitoring methods, modification of perioperative medical therapy, disposition of patient postoperatively (floor, ICU, outpatient) and even location of care (outpatient surgicenter, small community hospital, tertiary care center). With respect to coronary artery disease, emphasis should be placed on identifying the extent of the disease and left ventricular performance.The impact of the disease and its impact on outcome have been formalized in the 2002 ACC/AHA guideline entitled “Perioperative Cardiovascular Evaluation for Noncardiac Surgery.” By identifying clinical predictors, patients are classified as major, intermediate, or minor risk (Table 2).The surgical procedure being proposed is classified as high, intermediate, or low risk (Table 3).Then a stepwise approach as outlined in Figure 1 is utilized to determine whether to

immediately proceed to surgery, or proceed with further physiologic testing. The recurring theme throughout the guidelines is an assessment of functional status (as a clinical reflection of left ventricular performance) as reflected by exercise tolerance. Patients with minor or no clinical predictors with moderate or excellent exercise capability (>5METs) may proceed directly to the operating room without further work-up. Aortic stenosis is the most important valvular heart lesion that needs identification before surgery because of the associated incidence of sudden death as well as the well documented ineffectiveness of cardiac massage should cardiac arrest occur. Surgical aortic valve replacement may be appropriate for some patients with severe aortic stenosis while patients with moderate aortic stenosis can generally be managed medically with avoidance of tachycardia, maintaining of vascular

Table 2. Clinical Predictors of Increased Perioperative Cardiovascular Risk (Myocardial Infarction, Heart Failure, Death)* Major Unstable coronary syndromes Acute or recent myocardial infarction† with evidence of important ischemic risk by clinical symptoms or noninvasive study Unstable or severe‡ angina (Canadian Cardiovascular Society class III or IV)§ Decompensated heart failure Significant arrhythmias such as High-grade atrioventricular block Symptomatic ventricular arrhythmias in the presence of underlying heart disease Supraventricular arrhythmias with uncontrolled ventricular rate Severe valvular disease

Intermediate Mild angina pectoris (Canadian Cardiovascular Society class I or II) Prior myocardial infarction by history of pathological Q-waves Compensated or prior heart failure Diabetes mellitus (particularly insulin-dependent) Renal insufficiency Minor Advanced age Abnormal electrocardiogram (left ventricular hypertrophy, left bundle branch block, ST-T abnormalities) Rhythm other than sinus (e.g., atrial fibrillation) Low functional capacity (e.g., inability to climb one flight of stairs with a bag of groceries) History of stroke Uncontrolled systemic hypertension

*In conjunction with exercise tolerance and surgical risk factors, will determine degree of preoperative cardiac investigations as well as need for β-blockade. †The American College of Cardiology National Database Library defines recent myocardial infarction as greater than 7 days but less than or equal to 1 month (30 days); acute MI is within 7 days. ‡May include “stable” angina in patients who are unusually sedentary. §Campeau L. Grading of angina pectoris. Circulation. 1976;54:522-3.

Chapter 88 Noncardiac Anesthesia

1059

Table 3. Cardiac Event Risk* Stratification for Noncardiac Surgical Procedures† High (Reported cardiac risk often >5%) Emergent major operations, particularly in the elderly Aortic and other major vascular surgery Peripheral vascular surgery Anticipated prolonged surgical procedures associated with large fluid shifts and/or blood loss

Intermediate (Reported cardiac risk generally 3.0 while an indwelling catheter is in place, warfarin should be held to facilitate removal of the catheter. 5. Nonsteroidal anti-inflammatory drugs by themselves pose no additional bleeding risk for neuraxial placement. Patients should stop taking ticlopidine for 14 days and clopidogrel for 7 days prior to spinal needle placement.Normal platelet function returns 24-48 hours after abciximab cessation and 4-8 hours after eptifibatide or tirofiban discontinuation. One should avoid spinal/epidural needle placement within that time. Product inserts are more conservative and suggest that these agents should not be used within 4 weeks after surgery. Other anticoagulants used in addition to these agents will increase the risk of bleeding. When indwelling catheters are used as part of a postoperative analgesia regimen, a mechanism should be in place to flag and bring to special attention any new orders for anticoagulation or antiplatelet therapy in patients with indwelling catheters. Sedation or Monitored Anesthesia Care Intuitively one would expect that patients having only moderate sedation or sedation with local anesthesia would be at very low risk of cardiac events: confidential reporting of anesthetic complications in Australia

have shown that overall there is a similar risk of patient death with sedation compared with other anesthesia techniques. This may be a result of decreased vigilance by caregivers because of the misperception that sedation is very low risk, inadequate local anesthesia blockade of procedural site resulting in escalating dosing of sedation, different staffing models with less skilled caregivers providing care or the proceduralist attempting to simultaneously take responsibility for sedation care. Monitoring The role of pulmonary artery catheter placement for perioperative patient management during and after major surgery is now less clear than once thought. Numerous clinical trials conducted both within the operating room as well as the intensive care environment have shown no patient survival benefit. Transesophageal echocardiographic (TEE) monitoring during surgery and close clinical observation after surgery seems to be as successful as invasive pulmonary artery catheter monitoring. Transesophageal echocardiography (TEE) has been used by anesthesiologists for many years during cardiac surgery cases and its role is now expanding beyond the cardiac surgery to monitoring cardiac anatomy and function during major noncardiac surgery, e.g., liver transplantation to evaluate both RV and LV function and filling and to screen for patent foramen ovale and air emboli in patients with unusual anesthesia positioning, e.g., sitting for neurosurgical patients or some orthopedic cases. Other Maintaining perioperative normothermia decreases surgical infections, as well as cardiac complications. Presumably the detrimental physiologic mechanism is shivering and increased systemic vascular resistance (SVR) noted in the hypothermic patient. Transfusion thresholds for the surgical patient continue to be debated. Transfusing patients with heart failure to a hemoglobin >12.5 g/dL improves outcomes. Patients with preexisting cardiovascular disease have an increased incidence of complications with hemoglobin levels below 10 g/dL. It is reasonable to extrapolate from these data that a hemoglobin level >10 g/dL is optimal for surgical patients with cardiac disease.

Chapter 88 Noncardiac Anesthesia

POSTOPERATIVE CARE Disposition Postoperatively patients at significant risk of myocardial ischemia, cardiac rhythm disturbances, or major bleeding should be monitored in an intensive care setting (ICU). Achievement of optimal pain control and monitoring of marginal respiratory status with a propensity for secondary cardiac stress may also be indications for ICU admission. Postoperative Analgesia Poor postoperative pain control can lead to increased cardiac and pulmonary events. Aggressive pain control can decrease these complications. Pain control may be achieved with careful titration of narcotics, often with a patient controlled analgesia (PCA) pump, although this still requires careful choice and alteration of dosing by caregivers. Indwelling epidural catheters, running narcotic with or without dilute local anesthetic, can be very effective for lower extremity, abdominal or thoracic pain. Anticoagulant considerations as outlined earlier in this chapter as well as increased sensitivity to respiratory depression from concomitant use of additional narcotics or sedatives require thoughtful analgesia management. Peripheral nerve catheters infusing local anesthetic, in which an indwelling catheter is laid next to a peripheral nerve covering the surgical site, e.g. femoral nerve for knee surgery, allow analgesia with few systemic side effects. Early data suggested that epidural analgesia

1067

decreased the rate of cardiac complications with noncardiac surgery but later, carefully designed studies now suggest that aggressive pain control, whether epidural or narcotic, is the key rather than anesthesia type.

RESOURCES Auerbach A, Goldman L. Assessing and reducing the cardiac risk of noncardiac surgery. Circulation. 2006;113:1361-76. Eagle KA, Berger PB, Calkins H, et al, American College of Cardiology; American Heart Association. ACC/AHA guideline update for perioperative cardiovascular evaluation for noncardiac surgery: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). J Am Coll Cardiol. 2002;39:542-53. Erratum in: J Am Coll Cardiol. 2006;47:2356. Fleisher LA, Beckman JA, Brown KA, et al. ACC/ AHA 2006 guideline update on perioperative cardiovascular evaluation for noncardiac surgery: focused update on perioperative beta-blocker therapy. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2002 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). J Am Coll Cardiol. 2006;47:2343-55.

SECTION

IX

Cardiomyopathy and Heart Failure

Aspergillus Myocarditis

89 CARDIOVASCULAR REFLEXES AND HORMONES Alfredo L. Clavell, MD John C. Burnett, Jr, MD

Optimal regulation of the circulation is dependent on an integration of cardiovascular reflexes with local and circulating humoral factors that regulate myocardial contractility, vascular tone, and intravascular volume (intravascular volume is regulated primarily through renal sodium excretion). Under physiologic conditions, cardiovascular reflexes function in short-term cardiovascular control, whereas humoral mechanisms function as more long-term modulators of cardiovascular homeostasis.

during cardiac systole; their rate of discharge is directly related to the force of myocardial contraction and to cardiac filling pressure. Afferent signals from both arterial and cardiopulmonary receptors go to the nucleus solitarius in the brain stem. The principal functions of these receptors are twofold: 1. To inhibit efferent sympathetic neural outflow to the heart and circulation,resulting in decreases in arterial blood pressure and systemic vascular resistance. 2. To augment efferent parasympathetic neural outflow to the heart, resulting in sinus node slowing and prolongation of atrioventricular conduction.

CARDIOVASCULAR REFLEXES Two principal cardiovascular reflex arcs are involved in the regulation of blood pressure: 1. Arterial baroreceptors are located in the carotid sinus and aortic arch and respond with increasing neural discharge in response to stretch caused by increases in arterial blood pressure. 2. Cardiopulmonary baroreceptors are located in the ventricular myocardia and also in the atria and venoatrial junctions. Normal Cardiac Function The arterial and cardiopulmonary reflexes discharge

During reductions in arterial pressure and cardiac filling pressures under physiologic conditions, the inhibitory discharge of these receptors declines. Efferent sympathetic neural outflow increases, resulting in an increase in systemic vascular resistance, and efferent parasympathetic outflow decreases, resulting in tachycardia. Conversely, during increases in arterial blood pressure and cardiac filling pressures, the inhibitory discharge of these receptors is enhanced. 1071

1072

Section IX Cardiomyopathy and Heart Failure

Efferent sympathetic neural outflow decreases, resulting in a decrease in systemic vascular resistance,and parasympathetic outflow increases, resulting in bradycardia. Congestive Heart Failure In chronic congestive heart failure (CHF), a chronic reduction in arterial filling results in a decrease in inhibitory signaling to the cardiovascular reflex center, causing a significant increase in systemic vascular resistance. Despite high cardiac filling pressures due to ventricular dysfunction, an attenuation in the inhibitory action of the cardiopulmonary baroreceptors occurs. The dysfunction of cardiovascular reflexes in CHF results in enhanced adrenergic activity with systemic vasoconstriction. Additionally, sympathetic activation may have secondary actions and lead to activation of local and neurohumoral systems (such as the reninangiotensin system) and to avid sodium retention due to increased sodium resorption by the kidney. ■

Arterial baroreflexes are located in the carotid sinus and aortic arch and respond to increases in arterial

Fig. 1. Natriuretic peptide system.

■

blood pressure. Dysfunction of cardiovascular reflexes in congestive heart failure results in enhanced adrenergic activity with systemic vasoconstriction.

LOCAL AND CIRCULATING HUMORAL SYSTEMS Vasodilatory, Natriuretic, and Antimitogenic Systems Natriuretic Peptides The natriuretic peptide system encompasses a family of cardiovascular peptides: atrial (ANP) and brain (BNP) natriuretic peptides are of cardiac myocyte origin, whereas C-type natriuretic peptide (CNP) is of endothelial cell origin. These peptides are released in response to both acute and chronic atrial stretch (ANP and BNP) and in response to numerous other humoral stimuli (CNP). They have important actions on the heart, functioning through autocrine and paracrine mechanisms, and on other organ systems such as the kidney, adrenal gland, and the vascular wall (Fig. 1).

Chapter 89 Cardiovascular Reflexes and Hormones Important biologic actions include modulation of myocardial function and structure, natriuresis, inhibition of the renin-angiotensin-aldosterone system, vasodilatation, and an antimitogenic effect on vascular smooth muscle cells. CNP is devoid of natriuretic actions but is a powerful vasodilatory and antimitogenic peptide. The biologic actions of this important cardiovascular humoral system are via activation of specific particulate guanylate cyclase receptors, which function via the second messenger cyclic guanosine monophosphate. Importantly, the activity of this system is modulated by two pathways responsible for clearance and degradation of the natriuretic peptides, including neutral endopeptidase and a unique receptor-based clearance mechanism (Fig. 2). In chronic CHF, ANP and BNP circulating levels are increased. They have functional significance in the overall regulation of the cardiovascular system in CHF because their inhibition with unique receptor antagonists results in a rapid deterioration in experimental animal models of heart failure, as manifested by rapid activation of the renin-angiotensin-aldosterone system

Fig. 2. Natriuretic peptide hormone binding and clearance.

1073

together with vasoconstriction and sodium retention (Table 1). The increase of the levels of the natriuretic peptides in heart failure has significance for both prognosis and diagnosis of early asymptomatic left ventricular dysfunction. In particular, BNP has been recognized as a marker for left ventricular dysfunction and hypertrophy. Because of this functional importance, therapeutic strategies have emerged to potentiate the endogenous natriuretic peptides through inhibition of their degradation by neutral endopeptidase and by exogenous administration of natriuretic peptides via the intravenous and subcutaneous routes. However, the use of a peptidase inhibitor in a recent clinical trial had disappointing results. ■

In chronic CHF, ANP and BNP levels are increased. Other causes of elevated BNP are listed in Table 2.

Endothelium-Derived Relaxing Factor (Nitric Oxide) In addition to the natriuretic peptides, an endothelial

1074

Section IX Cardiomyopathy and Heart Failure

Table 1. Neurohumoral Mechanisms in Congestive Heart Failure Vasodilatory, natriuretic, Vasoconstrictive, and antimitogenic antinatriuretic, and factors mitogenic factors Natriuretic peptides Kallikrein, kinins Prostaglandin Dopamine Endothelium-derived relaxing factor—nitric

Renin-angiotensinaldosterone system Sympathetic nervous system Vasopressin Thromboxane

oxide Adrenomedullin

Endothelin Cytokines

cell-derived relaxing factor, nitric oxide (NO), also functions via activation of cyclic guanosine monophosphate through stimulation of soluble guanylate cyclase. The functional role of this endogenous factor is to cause vasodilatation and natriuresis and to inhibit vascular proliferation. Indeed, inhibition of endogenous NO by unique inhibitors results in systemic, renal, and pulmonary vasoconstriction and sodium retention. Longterm inhibition of the endogenous NO system results in hypertension and ventricular and vascular remodeling. Nitric oxide synthetases (NOS) are responsible for NO production; several isoenzymes have been identified. At the level of the endothelium the production and function of NO appears to be impaired in CHF. Other factors such as cytokines, free radicals and changes in cellular calcium handling contribute to the apparent dysfunction of the NO system in CHF. However, studies conflict with regard to NO activity in CHF. Some studies suggest that NO activity is enhanced in human and experimental animal heart failure, because inhibition of its generation in heart failure results in further ventricular dysfunction and systemic vasoconstriction. ■

■

Nitric oxide (NO) functions via activation of cyclic guanosine monophosphate through stimulation of soluble guanylate cyclase. The clinical significance of NO activity in CHF is unclear.

Table 2. Other Causes of Elevated BNP LVH Myocarditis Cardiac allograft rejection Kawasaki disease Primary pulmonary hypertension Renal failure Ascitic cirrhosis Cushing’s disease Primary hyperaldosteronism Advanced age

Vasoconstrictor, Antinatriuretic, and Mitogenic Systems Endocrine mechanisms exist to modulate vascular tone, growth of cardiac myocytes and vascular smooth muscle, and sodium excretion by the kidney. The sympathetic, renin-angiotensin-aldosterone, and endothelin systems emerge as three important vasoconstrictor, antinatriuretic, and mitogenic systems that control cardiovascular homeostasis and play a role in the pathophysiology of CHF. Sympathetic Nervous System Plasma catecholamines (norepinephrine and epinephrine) are the circulating humoral counterparts of the sympathetic nervous system. Norepinephrine is released locally from sympathetic nerve endings adjacent to myocardium and modulates myocardial contractility. The adrenal medulla also releases both catecholamines in response to diverse stimuli and amplifies the cardiovascular response to sympathetic nervous system activation.The myocardium is rich in β receptors, which are the target of these cardiovascular hormones. In chronic CHF there is activation of the sympathetic nervous system as a response to the reduction in myocardial contractility and cardiac output. Although the resultant vasoconstriction and increase in myocardial contractility are essential for maintaining blood pressure, eventually this response becomes deleterious and contributes to a further decline in myocardial function. In the presence of chronically increased serum norepinephrine levels, there is down-regulation of myocardial β receptors, perhaps as a protective mechanism.

Chapter 89 Cardiovascular Reflexes and Hormones Circulating levels of norepinephrine correlate with patient mortality in CHF. β-Adrenergic blockade is an important strategy in the therapeutic neurohumoral modulation of CHF. Recent studies demonstrate a paradoxic increase in left ventricular function with β-blockers, improved clinical symptoms, and better prognosis in heart failure, regardless of the cause of the CHF and in addition to angiotensinconverting enzyme inhibition. Additionally, studies suggest that non-selective β-blockade is superior to selective β−1 blockade in the management of CHF. In fact, the mortality benefit of β-blockade appears to be superior to the benefit observed with angiotensin converting enzyme inhibition. ■

■

In CHF there is chronic activation of norepinephrine and down-regulation of myocardial β receptors. Treatment with β-blockade results in improved left ventricular function, clinical symptoms, and prognosis in patients with CHF.

Renin-Angiotensin-Aldosterone System Angiotensin II is one of the most potent vasoconstrictor and mitogenic peptides that is produced both systemically and locally in the heart, lung, kidney, and vascular endothelium as a result of the abundant presence of angiotensin-converting enzyme (Table 3). Angiotensin II, which functions via specific receptor subtypes, also is responsible for stimulation of norepinephrine release and sympathetic activation. Metabolism and growth in myocyte and non-myocyte cells also are altered by circulating and locally generated angiotensin II, which increases cellular proliferation and impairs myocyte contractile activity. Additionally, aldosterone produced by the adrenal gland is activated by angiotensin II and has effects on non-myocytes in addition to its sodiumretaining action in the kidney. Most recently, studies have suggested that aldosterone may be responsible for cardiac fibrosis via specific receptors within the heart. These two important hormones, angiotensin II and aldosterone, have emerged as the targets for pharmacologic inhibition in the treatment of CHF; in severe human CHF, inhibition of angiotensin II generation has resulted in improvement in mortality and morbidity. However, escape from angiotensin-converting enzyme inhibition is noted chronically, and newer strategies relying on angiotensin

1075

II receptor and aldosterone antagonists have proved useful in the management of refractory heart failure. ■

Angiotensin II, which functions via specific receptor subtypes, is responsible for stimulation of norepinephrine release and sympathetic activation.

Endothelin System Endothelin is a 21-amino acid peptide that is produced by the endothelium. Much like angiotensin II, an endothelin-converting enzyme cleaves large endothelin into its biologically active form. Its biological actions are mediated through several ET-receptors. Although its role in physiology continues to be elucidated, most likely, as with angiotensin II, it serves to maintain vascular tone and arterial blood pressure. In CHF, it functions as a compensatory mechanism to mediate vasoconstriction and possibly augment inotropic function. Myocardial responsiveness to endothelin also may be preserved in late heart failure when the myocardium has become refractory to other endogenous agonists. As with angiotensin II, endothelin has growthpromoting and mitogenic potential and, therefore, may contribute to cardiac and vascular remodeling. Endothelin stimulates renin and aldosterone release and augments activation of cardiac fibroblasts.

Table 3. Angiotensin II: Sites and Actions Targets Heart Kidney Adrenal body Brain Sympathetic nervous system Vascular smooth muscle

Actions Positive inotropism, hypertrophy Renin release, mesangial contraction, sodium resorption Aldosterone release Vasopressin release, thirst, increased sympathetic outflow Norepinephrine release

Vasoconstriction, hypertrophy

1076

Section IX Cardiomyopathy and Heart Failure

Endothelin also has potent renal vasoconstricting and sodium-retaining actions in CHF. Studies also have suggested that an increase in plasma endothelin may have adverse prognostic implications in CHF. However, chronic endothelin receptor blockade resulted in no benefit in a recent randomized study in CHF patients. Selective ET-A receptor blockade has

proven efficacious in the management of pulmonary hypertension. ■

■

An increase of plasma endothelin may have prognostic implications in CHF. Endothelin receptor blockade is emerging as a new strategy in the management of pulmonary hypertension.

90 SYSTOLIC HEART FUNCTION Wayne L. Miller, MD, PhD Lyle J. Olson, MD

CELLULAR ASPECTS OF LV CONTRACTION

■

Microanatomy The myocardium is composed of cardiac myocytes enveloped in a dense extracellular matrix of collagen, the main structural protein of the heart. Cardiac myocytes account for 70-75% of the myocardium by cell volume but only 25-30% by cell number. Cardiac myocytes contain myofibrils that are composed of longitudinally repeating sarcomeres separated by Z bands (thickened and invaginated portions of the surface membrane called the sarcolemma). The sarcomeres occupy about 50% of the mass of cardiac myocytes. Thin filaments composed of actin are attached to each Z line and interdigitate with the thick filaments composed of myosin molecules. The thick and thin myofilaments slide past one another in a “ratchet-type” mechanism to generate force and shorten the myocyte. The myofilaments maintain a fixed length throughout contraction. Mitochondria compose about 20% of the cell volume and are the organelles in which adenosine triphosphate (ATP) is generated and located in close proximity to the myofibrils, as well as just beneath the sarcolemma. Platelike folds, or cristae, project inward from the surface membrane of the mitochondria and contain the respiratory enzymes necessary for energy production (Fig. 1).

Excitation and Contraction Coupling The coupling of cardiac excitation (electrical event) and contraction (mechanical event) are fundamentally molecular in character. The sarcolemma is a thin phospholipid membrane which functions to maintain electrical polarization. The phospholipid bilayer acts as an ionic barrier and maintains relative high intracellular potassium [K+], and low intracellular sodium [Na+] and calcium [Ca+] concentrations (Fig. 2). Near the Z lines are wide invaginations of the sarcolemma, the T (tubule) system, which branch through the cell. Closely coupled to but not continuous with the T system is the sarcoplasmic reticulum, a complex network of anastomosing membrane-limited intracellular tubules that surround each myofibril and play a critical role in the excitation-contraction coupling of the heart muscle. Troponin (which is composed of troponin C, I, and T) and tropomyosin are regulatory proteins found in the thin filaments. In the absence of troponin and tropomyosin, the contractile proteins actin and myosin are activated, requiring only the presence of Mg2+ and ATP. The regulatory proteins, when present prevent cross-bridge formation between myosin and actin. 1077

Contractile sarcomeres occupy about 50% of the mass of cardiac myocytes.

1078

Section IX Cardiomyopathy and Heart Failure

Fig. 1. The major shifts of calcium ions during myocyte excitation-contraction coupling and relaxation. The dots represent calcium ions, and the positive and negative signs indicate electrical charge across membrane partitions.

Fig. 2. The regulation of excitation-contraction coupling. The sarcolemma and sarcoplasmic reticulum modulate cytoplasmic calcium availability, and the troponin-tropomyosin complex regulates responsiveness to cytoplasmic calcium. AC, adenylate cyclase; ADP, adenosine phosphate; AMP, adenosine monophosphate; ATP, adenosine triphosphate; ATPase, adenosine triphosphatase; cAMP, cyclic adenosine monophosphate; GI, guanine nucleotide-binding regulatory protein that inhibits adenylate cyclase; GS, guanine nucleotide-binding regulatory protein that stimulates adenylate cyclase.

Chapter 90 Systolic Heart Function When Ca 2+ binds to troponin C, the binding of troponin I to actin is inhibited, which in turn causes a conformational change in tropomyosin, such that tropomyosin instead of inhibiting, now enhances crossbridge formation.Thus, Ca2+ blocks an inhibitor of the interaction between actin and myosin.The key element in the initiation of contraction is the release of sarcoplasmic [Ca2+]. Depolarization of the sarcolemma caused by the upstroke of the action potential opens the ion channels that carry the inward Ca2+ current, which in turn triggers a release of the large stores of calcium in the sarcoplasmic reticulum. With cellular depolarization, the myoplasmic [Ca2+] rises and is bound to troponin. Once each cross-bridge sliding action is completed, the myosin head releases its ATP breakdown products, binds another ATP molecule, and detaches from the actin site. The myosin head then returns to its original configuration and the cycle is repeated. Relaxation is brought about by the active (ATPrequiring) reuptake of the calcium into the sarcoplasmic reticulum.Thus,calcium is essential to the excitationcontraction coupling, and when the calcium concentration decreases to a critical point, contraction ceases. ■

■

■

Troponin and tropomyosin are regulatory proteins found in the thin filaments. The key element in the initiation of cardiac contraction is the release of sarcoplasmic [Ca2+]. The transmembrane calcium current does not directly cause cardiac contraction but promotes release of sarcoplasmic Ca2+.

Mechanics of Contraction The motion of the LV during contraction can be summarized in the mnemonic TARTT. During systole, the LV Translates (moves from side-to-side), Accordions (moves with the base and apex attempting to approximate each other), Rotates (about the LV “long axis”), Tilts (perpendicular to the long axis), and Thickens (Fig. 3). Myocardial fibers are arranged in a spiral fashion around the central LV cavity. The subendocardial and subepicardial fibers run largely parallel to the long axis of the cavity, and the mid-wall fibers are mostly perpendicular to the long axis (i.e. circumferential). During ventricular ejection, these fibers shorten and thicken,

1079

and as the LV cavity decreases circumferentially and longitudinally, the inner surface decreases more than the external surface (as dictated by geometry). Because muscle mass remains constant, an increase in wall thickness must occur. During isovolumic LV contraction, the chordae tendineae become tense, the mitral valve closes, and the ellipsoid LV becomes more spherical. During LV ejection with the opening of the aortic valve, the longitudinal axis shortens by only about 10%, whereas the short-axis diameter shortens by about 25%, thus accounting for 80% to 90% of the normal stroke volume. Isovolumic contraction refers to the interval (about 50 ms) between the onset of ventricular systole and the opening of the semilunar (aortic and pulmonic) valves. The LV pressure must exceed that in the aorta during diastole for the valves to open. There is a small increase of pressure in the aorta just before the semilunar valves

Translation

Accordian Rotation Tilt Thicken Fig. 3. Mechanisms of contraction and motion of the heart.

1080

Section IX Cardiomyopathy and Heart Failure

open causing the incisura or dicrotic notch. Ventricular ejection is that phase of ventricular systole (about 350 to 400 ms in duration at a normal heart rate) in which blood is ejected through the aortic valve. The first phase (about 100 ms) is rapid, and then ejection slows toward the end of ventricular systole. The increase in ventricular pressure is more marked in the rapid phase. ■

Myocardial fibers are arranged in a spiral fashion around the central LV cavity.

DETERMINANTS OF CONTRACTION OF THE INTACT LV The mechanical determinants of cardiac function are preload, afterload, contractility and heart rate. When the intact heart is compared with isolated muscle, heart volume and pressure are analogous to muscle length and tension. Starling’s Law of the Heart is a fundamental property of heart muscle in which the force of contraction at any given tension depends on the initial muscle fiber length.This, in turn, depends on the ultrastructural disposition of thick and thin myofilaments within the sarcomeres. It was in the classic isolated heart and muscle strip experiments that the concepts of preload, afterload, and contractility first became clinically useful terms. Alterations in preload, operating through changes in end-diastolic fiber length, are important determinants of the performance of the intact ventricle and provide the basis for the length-function curves of the intact ventricle. The ability to augment preload provides a functional reserve to the heart in situations of acute stress or exercise. Preload is thus an important factor in maintaining LV systolic performance in many disease states (Fig. 4). Afterload Afterload in the intact LV is the tension (force or wall stress) acting on the fibers of the LV after the onset of shortening.This is primarily the arterial pressure and is a major determinant of stroke volume. An abrupt increase in the impedance to LV ejection, when preload is constant, causes a decrease in fiber shortening and LV stroke volume. The LV becomes smaller during normal ejection and its walls thicken. Thus, despite an increase in aortic pressure during LV ejection, the

afterload or wall stress decreases during ejection. In this situation, there is an inverse relationship between afterload (systolic pressure or wall stress) and stroke volume, extent of wall shortening, and velocity of shortening. In the normal individual, stroke volume can be maintained despite a modest increase in arterial pressure by augmenting LV end-diastolic pressure and volume; that is, the increment in afterload is met by an increase in preload. However, in the diseased heart with little preload reserve (such as heart failure), the LV stroke volume would decrease. Also, even in the normal heart, when there is relative hypovolemia (such as with sepsis or hemorrhage), the preload cannot increase sufficiently and an increase in afterload will reduce the stroke volume (Fig. 5).Table 1 shows LV loading in disease states.

Left Ventricular Preload: • "Stretch" in isolated muscle preparations • End-diastolic wall stress in intact heart • Common to use LVEDV with substitution of LVEDP frequently in clinical situations

Contributing factors: Total blood volume Atrial contribution to LV filling

Body position

Intrathoracic pressure

Stretching of myocardium

Pumping action of skeletal muscle

Intrapericardial pressure

Venous tone

LV performance

LVEDP LVEDV

Fig. 4. Factors affecting myocardial stretch and left ventricular (LV) preload. LVEDP, left ventricular enddiastolic pressure; LVEDV, left ventricular end-diastolic volume.

Chapter 90 Systolic Heart Function Contractility The term “contractility” has been used synonymously with “inotropic state.” It is difficult to define in a quantitative sense because there is no clear-cut single measurement of contractility that provides a numeric value that can be assigned to a given heart. However, when loading (preload and afterload) conditions remain constant, an improvement in contractility augments cardiac performance, whereas a depression in contractility lowers cardiac performance. Inotropic influences generally act through altered Ca2+ availability to the myofilaments or through an alteration of myofilament Ca2+ sensitivity. Additional factors that directly or indirectly affect contractility are sympathetic neural activity and circulating catecholamines (Fig. 6). The LV pressure-volume relationship is a convenient assessment framework used to understand the responses of LV contraction to alterations in preload, afterload, and contractility. In its simplest sense, the average circumferential wall stress (σ, force per unit of cross-sectional area of wall) in the intact heart is the product of intraventricular pressure (P) and the internal radius of curvature of the chamber (a) divided by the thickness of the muscle walls (h x 2). Laplace’s law for a spherical chamber is σ = Pa/2h

1081

Preload Defining preload for the intact LV as the ventricular end-diastolic wall stress provides a direct analogy to the preload of the isolated muscle strip, which in turn determines the resting length of the sarcomeres. Increases in preload augment the stroke volume, as well as the extent and velocity of wall shortening. At a constant preload, there is an inverse relation between systolic wall stress and stroke volume. ■

The ability to augment preload provides a functional reserve to the heart in situations of acute stress or exercise.

Heart Rate Increasing the frequency of contraction does not produce a shift of the ventricular performance curve relating LV end-diastolic pressure and stroke work, but it does increase stroke power at any given level of filling pressure. Thus, increasing the heart rate will improve myocardial contractility, because the systolic fraction of the cardiac cycle is increased. The positive inotropic effect resulting from an increase in heart rate is more prominent in the depressed heart than in the normal heart. In the normal heart, an artificial increase in heart rate (such as via a pacemaker) will not increase cardiac output, because venous return to the heart is reflexly

Table 1. Left Ventricular Loading in Disease States Condition

Preload

Afterload

Contractility

Sepsis Dehydration Heart failure Cardiogenic shock

↓ ↓ ↑ ↑

→↓ → → →

→↓ → ↓ ↓

RV infarct

↓

→

→

Acute mitral regurgitation

↑

→

→

Aortic stenosis Systemic hypertension

→ →

↑ ↑

→ →

RV, right ventricle.

Therapy Fluids and antibiotics Fluids Diuretics Inotropes Intra-aortic balloon pump Increase intravenous fluids to maintain high RV filling pressure Intra-aortic balloon pump Surgery, if severe Surgery, if severe Antihypertensive medications

1082

Section IX Cardiomyopathy and Heart Failure

Left Ventricular Afterload: • Tension, force, or wall stress acting on the fibers of the LV after the onset of shortening • Although LV systolic pressure and systemic vascular resistance can affect afterload, and thus cardiac output, neither is its equivalent

Left Ventricular Contractility: • Synonymous with "inotropic state," but there is no single measurement of its value • Generally mediated through altered Ca2+ availability or via alteration of myofilament Ca2+ sensitivity • Contractility by definition is independent of loading conditions

Contributing factors:

Contributing factors:

LV pressure

PVR

Circulating Force-frequency Loss of relation catecholamines myocardium Sympathetic nerve traffic

Afterload

Contractility

Pharmacologic depressants

Digitalis and other "inotropes"

LV geometry

Shortening

A B

C A

Normal

B

Anoxia Hypercapnia Acidosis

Ventricular Performance

A = basal B = mild pressor C = severe pressor

C A B

C

Mild dysfunction Severe dysfunction

Afterload

Fig. 5. Factors affecting left ventricular (LV) afterload. PVR, peripheral vascular resistance.

LVEDP

Fig. 6. Determinants of left ventricular contractility. LVEDP, left ventricular end-diastolic pressure.

and metabolically stabilized. However, if the diastolic volume of the LV is increased by increasing venous return, as during exercise, then tachycardia plays a major role in increasing cardiac output. This assumes, however, that not only the speed of contraction but also the speed of relaxation is increased. This effect requires preservation of both systolic and diastolic function. Of course, when the heart rate is too fast, the short duration of diastole can impede ventricular filling, and a decrease in cardiac output can be observed, generally with rates greater than 180 beats/min (Table 2). Myocardial Infarction Infarction of 30% or more of the LV mass results in a decrease in LV ejection fraction. Initially, the cardiac output will be depressed and, in circumstances of considerable damage to the LV, function may deteriorate

further, leading to hemodynamic compromise and death. However, in most circumstances, when adequate reserve is present, the cardiac stroke volume is augmented by increases in ventricular preload within hours of the infarction.This change is generally accomplished by an increase in LV end-diastolic pressure and reflects a direct consequence to Starling’s law of the heart. Increases in afterload may also accompany these changes and thus may offset the increases in stroke volume brought about by increased preload. There are limits to preload reserve, and further increases in cardiac output must then be brought about by increased heart rate. This situation is also observed in patients with dilated cardiomyopathy and congestive heart failure. In such circumstances, use of agents to reduce afterload

Chapter 90 Systolic Heart Function

Table 2. Effect of Heart Rate on Left Ventricular Systolic Function Heart rate: Positive inotropic effect. Note there is little effect by ventricular pacing on stroke volume in normal patients because of reflex stabilization of venous return In the overloaded LV (e.g., CHF), increases in heart rate augment stroke volume (to a point) In exercise, with increased venous return, increases in heart rate are the major contributors to increased cardiac output; in normals, the speeds of LV contraction and relaxation are increased, facilitating accommodation of the increased venous return (up to 180-220 beats/min, but much lower rates in CHF) CHF, congestive heart failure; LV, left ventricle.

may be beneficial in augmenting LV stroke volume (Table 3). ■

Myocardial infarction of 30% or more of the LV mass results in depression of the LV ejection fraction.

Left Ventricular Hypertrophy Left ventricular hypertrophy may occur in conditions

1083

of pressure and volume overload. Acquired disorders associated with a pathologic increase in LV preload include chronic aortic and mitral valvular regurgitation, dilated cardiomyopathy, and, often, myocardial infarction. Disorders associated with a pathologic increase in LV afterload include severe aortic stenosis and chronic systemic hypertension. Although the changes in the sarcomeres are different for these two overload states, in both situations the overall mass of the LV is increased. Nevertheless, the hypertrophic response, which is an important adaptive process that enables the heart to compensate for overloading, is a complex process that is both beneficial and detrimental to LV performance.The hypertrophied cells are not necessarily normal, and abnormalities in inotropic response and vascular reactivity have been shown. LV hypertrophy increases myocardial oxygen demand and, along with changes in ventricular loading (primarily afterload) and heart rate, is a major contributor to increased myocardial oxygen consumption (Table 4).

PHYSIOLOGIC MEASURES OF LV SYSTOLIC FUNCTION Ejection Fraction The most commonly available measure of LV systolic performance or contraction is the ejection fraction (EF). This is simply a ratio of the LV stroke volume

Table 3. Effect of Loss of Myocardium on Left Ventricular Systolic Function

Table 4. Effect of Left Ventricular Hypertrophy on Left Ventricular Systolic Function

Loss of myocardium: Infarction (> 30% of LV mass), fibrosis, infiltration, “myopathies” all reduce LV systolic performance Generally, preload reserve (Starling’s law) can assist in augmentation of stroke volume However, in some circumstances, reflex and intrinsic regulatory humoral factors may “pathologically” increase SVR (increase afterload); when preload reserve is exhausted, there is an afterload-preload “mismatch”

Left ventricular hypertrophy: Common in both chronic “pressure” and “volume” overloading; also common in dilated cardiomyopathy and after myocardial infarction (if > 20% of LV) May assist in “normalizing” LV wall stress, but is a major component of myocardial oxygen demand and can be associated with reduced myocardial flow reserve (?mechanism of “angina” in aortic stenosis) and altered inotropic responsiveness

LV, left ventricle; SVR, systemic vascular resistance.

LV, left ventricle; SVR, systemic vascular resistance.

1084

Section IX Cardiomyopathy and Heart Failure

(SV) to the LV end-diastolic volume. It can be determined with various imaging methods (Table 5). Many of these methods rely on an assumption that the LV shape can be approximated by an ellipsoid. However, the geometry of the LV can be distorted by various diseases, and thus the accuracy of any measure of EF is dependent on the completeness of the measurements. For instance, use of simple formulas derived from two-dimensional echocardiographic measures of end-systolic and enddiastolic dimensions and estimates of LV long-axis length can be erroneous if there are regional LV wall abnormalities of contraction, such as after infarction. In such instances, accurate measures of EF can be derived with radionuclide ventriculography (which does not require assumptions of ventricular shape) or directly via quantitation of LV end-diastolic volume (EDV) and end-systolic volume (ESV) with magnetic resonance imaging or electron beam computed tomography.Thus: EDV - ESV SV Ejection Fraction (EF) = ___________ = _____ EDV EDV ■

The most commonly applied and clinically available measure of systolic performance or contraction is the ejection fraction.

Table 5. Clinical Methods of Measuring Left Ventricular Systolic Function 1. Ejection fraction (EF) can be determined with available imaging tools; be cautious of methods that rely on assumptions of LV geometry; acute increase in preload or decrease in afterload will increase EF, and vice versa 2. The velocity of circumferential fractional shortening (VCF) is a better index of contractility than the actual amount of shortening; is relatively insensitive to acute changes in preload; difficult to calculate clinically 3. PER: peak LV systolic emptying rate; loaddependent index of systolic function; use angio, RNA, cine-CT angio, angiography; cine-CT, cine-computed tomography; RNA, radionuclide angiography.

Maximal Elastance Another method for measurement of left ventricular contractility is the concept of Emax (maximal elastance). This is based on the observation that there is a linear relationship between pressure and volume at end-systole. Stated another way, all end-systolic pressurevolume intercepts form a straight line on the pressurevolume curve for a given degree of contractility. The slope of this line is called the Emax. With an increase in contractility, there is an increase in the slope of the Emax; with a decrease in contractility, there is a decrease in the slope of the Emax. Calculation of Emax requires construction of pressure-volume curves and manipulation of either preload or afterload (Fig. 7).

MYOCARDIAL RELAXATION Myocardial relaxation is the process by which the myocardium returns to its initial length and tension relationship. Cardiac relaxation is an energy-dependent process that consumes high-energy phosphates. At the cellular level, abnormalities of calcium reuptake may account for LV diastolic abnormalities and impaired relaxation. Relaxation also depends on systolic and diastolic loads and the passive elastic characteristics of the ventricle. Relaxation may be simplistically regarded as occurring during the isovolumic relaxation period and part of the rapid filling period. If the ventricle is able to fully and quickly complete relaxation, the ventricle will rapidly expand and a large portion of blood flows in from the left atrium to the LV after mitral valve opening. However, if there is a delay in the rate and duration of relaxation, the ventricle will continue to expand slowly even after mitral valve opening. Thus, there will be a decrease in the rate of early rapid filling. Ventricular Compliance In mid and late diastole, pressure and volume increase, and the passive diastolic properties of the ventricle, namely, chamber stiffness (or its inverse, chamber compliance), can be assessed. LV compliance is the change in volume per unit pressure as the LV fills with blood from the left atrium.Thus, a decrease in compliance will result in less blood entering the LV for a given increase in pressure. Myocardial fibrosis from any cause can be

Chapter 90 Systolic Heart Function

1085

LV pressure-volume curve 300

240

LV pressure, mm Hg

Increased contractility

End-systolic and isovolumic pressure-volume curve (Emax)

Normal

Heart failure

180

120 Diastolic pressure-volume curve

15

SV 0

25

50

75

100

125

LV volume Fig. 7. Maximal elastance (Emax) is a sensitive measure of left ventricular (LV) function and is derived from LV pressure-volume loops.

expected to in-crease ventricular stiffness because collagen fibers are very rigid and virtually nondistensible at normal pressures. The term “myocardial stiffness” is used to differentiate changes in the stiffness properties of each unit of muscle as opposed to overall chamber stiffness. Thickening of ventricular walls from any cause (for

example, LV hypertrophy) tends to increase both myocardial and chamber stiffness. An increased volume/mass ratio is often associated with increased chamber stiffness, whereas in other cases increased chamber stiffness may occur in the presence of a normal volume/mass ratio, implying increased myocardial stiffness.

91 DIASTOLIC HEART FUNCTION Christopher P. Appleton, MD

Normal left ventricular (LV) diastolic function can be defined as the ability of the ventricle to fill to a normal end-diastolic volume, during both rest and exercise, with a mean left atrial (LA) pressure that does not exceed 12 mm Hg. Because the process of LV relaxation is more energy dependent than contraction, abnormalities of LV diastolic function occur earlier than systolic dysfunction in virtually all cardiac diseases. They increase in frequency with aging, so that about 50% of patients over 70 years old with symptoms of heart failure (HF) have a normal LV ejection fraction, or “diastolic heart failure” (DHF) as their primary cardiac problem. Studies show the symptoms of diastolic and systolic heart failure (SHF) are clinically indistinguishable. Patients with DHF are either unable to adequately distend their slowly relaxing and stiffened left ventricles, or can do so only with elevated filling pressures. This results in symptoms due to pulmonary congestion, atrial arrhythmias or reduced exercise capacity due to a inability to increase LV stroke volume at faster heart rates. Recognition of patients with DHF is important because they have a prognosis nearly as poor as SHF, and even asymptomatic patients with diastolic dysfunction are at increased risk for adverse cardiovascular events. In addition, in patients with

SHF, the degree of diastolic dysfunction is a powerful predictor of mortality. Reliable, noninvasive ways to diagnose diastolic function at its earliest stages continue to be pursued and potential therapies for LV diastolic function are being studied. ■

■

■

■

The symptoms of HF from LV systolic and diastolic dysfunction are indistinguishable. Recognition of patients with a normal LVEF who have asymptomatic LV diastolic dysfunction is important because of their increased risk for future adverse cardiovascular events. Once symptomatic DHF is present, the prognosis is nearly as poor as that of symptomatic SHF. In patients with SHF the degree of diastolic abnormality is a powerful predictor of survival.

EPIDEMIOLOGY In the United States over 500,000 new cases of HF are diagnosed yearly. Epidemiological studies show risk factors for new onset HF to be advancing age, coronary artery disease, hypertension, LV hypertrophy and diabetes mellitus. In patients 50%) or “diastolic heart failure” (DHF) as their primary cardiac abnormality (Fig. 1). In this more elderly group female gender and hypertension predominate.These patients complain of dyspnea on exertion and reduced exercise capacity. They may present acutely with marked hypertension and pulmonary edema. They are also at increased risk for new onset atrial fibrillation or stroke. Although the mortality associated with symptomatic LV DHF had previously been thought to be about one third of that seen in SHF, many studies now indicate that cardiovascular (CV) events and mortality are nearly equal for the two types of HF (Fig. 2). It has also been shown that the burden of unrecognized and asymptomatic LV diastolic dysfunction is common in the general community, and increases the risk for cardiovascular events. With our aging population increasing numbers of patients with DHF will be seen, and yet compared to SHF there are few clinical studies on therapy to guide clinical practice. ■

In patients >70 years old who have HF symptoms, new onset atrial fibrillation or stroke, DHF should

Fig. 1. Age distribution of new onset HF in 216 patients from Olmsted County, Minnesota in 1991. Patients with diastolic HF tend to be older (>70 years) than those with a reduced LVEF.

■

■

■

always be suspected, especially if the LV ejection fraction is normal. In the elderly the CV event rate associated with DHF appears to be higher than previously thought, and approaches that of patients with SHF. Since asymptomatic LV diastolic dysfunction is common and associated with increased future cardiovascular risk, reliable cost effective methods for identifying these patients are needed. The best therapies for DHF are unknown at the present time.

Fig. 2. Survival of patients with new onset HF in Olmsted County, Minnesota in 1991. Survival is markedly reduced regardless of LV ejection fraction (LVEF) compared to that expected for an age matched group.

Chapter 91 Diastolic Heart Function

HEMODYNAMIC PHASES OF DIASTOLE Diastole is divided into four phases: 1) isovolumic relaxation, 2) early LV filling, 3) diastasis, and 4) filling at atrial contraction. As shown in Figure 3, the determinants of LV diastolic performance vary in their importance and interaction during these different phases. Isovolumic relaxation begins with aortic valve closure, and continues until LV pressure falls below left atrial (LA) pressure. Early diastolic LV filling begins with mitral valve opening and ends when the rising ventricular pressure equals or exceeds the LA pressure. If the diastolic filling period is relatively long, a period of diastasis follows where LA and LV pressures are nearly equal and little additional LV filling is occurring. Finally, atrial contraction reestablishes a transmitral pressure gradient and a variable amount of blood is transferred from atrium to ventricle in late diastole. ■

Diastole is divided into four phases: 1) isovolumic relaxation, 2) early LV filling, 3) diastasis, and 4) filling at atrial contraction.

Although dividing diastole into phases aides description and quantitation of LV diastolic properties, in reality such a separation is artificial in that the factors that influence each phase usually influence all others, especially in disease states. This interaction of diastolic properties, the multiple other factors which influence these properties (systolic function, pericardial restraint, coronary artery turgor, etc.) and the overlap of their

Isovolumic relaxation

1089

effects on the different phases of diastole, has contributed to the difficulty and understanding in studying LV diastolic function.

LV DIASTOLIC PROPERTIES LV Relaxation The process of LV contraction and relaxation is dependent upon two biologic systems, cellular Ca2+ extruding pumps and exchangers, and myofilament (actin-myosin) interaction (Fig. 4). In mammalian hearts contraction occurs after cellular depolarization results in the passive release of large stores of Ca2+ from the sarcoplasmic reticulum (SR), and subsequent activation of the Ca2+/troponin/actin/myosin cascade. In contrast to the process of Ca2+ release, the reuptake of cytosolic Ca2+ back into the SR is an active, energy (ATP) and loaddependent process accomplished by a powerful SR Ca2+ (SERCA) pump. The energy dependence of Ca2+ resequestration explains why diastolic properties are altered before contraction becomes abnormal. Under normal circumstances the rate of Ca2+ reuptake is rapid so that LV relaxation occurs largely by “elastic recoil”from energy stored in compressible interstitial elements during systolic compression. Delay in deactivation of the contractile proteins interferes with this process, occurs by several different mechanisms, and is a consistent finding early in the course of all cardiac diseases. Electrical dyssynchrony, increased

Early filling

Diastasis

Myocardial relaxation

Passive filling characteristics of LV Characteristics of left atrium, pulmonary veins, and mitral valve Heart rate

Fig. 3. Determinants of LV diastolic performance and their relation to phases of diastole.

Atrial systole

1090

Section IX Cardiomyopathy and Heart Failure

Fig. 4. Schematic diagram of calcium dependent excitation-contraction coupling and relaxation in the heart (see text for discussion). AMP, adenosine monophosphate; ATP, adenosine triphosphate; cAMP, cyclic AMP; BR, β-adrenergic; P, degree of phosphorylation; Tm, tropomyosin; TnC, troponin C; TnI, troponin I; TnT, troponin T.

mechanical loading (hypertension, aortic stenosis), arterial endothelial dysfunction, ischemia, and negative inotropic drugs are some of the ways LV relaxation can be slowed or “impaired,” decreasing the rate of elastic recoil and time available for ventricular filling. Genetic alterations in the SERCA handling of Ca2+ are induced by LV hypertrophy and thyroid hormone. ■

In contrast to LV contraction, LV relaxation is an energy dependent process and more susceptible to disruption by disease states such as electrical dyssynchrony, increased mechanical loading (hypertension, aortic stenosis), and coronary artery disease.

The cellular processes and molecular biology that influence myocyte Ca2+ handling, LV relaxation, and their alteration in cardiac diseases is becoming clearer (Fig. 4).The SERCA pump has an endogenous regulator

phospholamban (PLB), which is stimulated by phosphorylation with a cAMP dependent protein kinase. β1-adrenergic phosphorylation of PLB by drugs speeds relaxation. SERCA pump dysfunction can result in a reduced rate of Ca2+ uptake, which may limit the rate of LV relaxation, or result in incomplete crossbridge dissociation and increased chamber stiffness. Other factors that influence LV relaxation include shifts in myofilament Ca2+ sensitivity, mechanical stretch or load on the ventricle and displacement and load dependence of the myofilaments themselves, whose effects differ depending on which part of systole they are applied. ■

■

The sarcoplasmic reticulum Ca2+ (SERCA) pump, helps control the rate of LV relaxation, and has an endogenous regulator, phospholamban. LV contraction and relaxation is also be affected by

Chapter 91 Diastolic Heart Function mechanisms which alter the availability of Ca2+ to the contractile proteins, the sensitivity of the proteins to Ca2+, or the mechanical stretch and load on the ventricle. In the intact mammalian heart both experimental and clinical studies suggest the normal left ventricle contracts to a volume below its equilibrium volume, compressing elastic cardiac elements. This creates early diastole restoring forces that produce elastic recoil, a “suction” effect that lowers LV minimum pressure and increases early LV diastolic filling. In the normal canine heart approximately 20% of filling occurs while LV pressure is falling. Rapid LV relaxation helps maximize the beneficial effect of these restoring forces and together they result in a normal pattern of LV filling that occurs predominantly in early diastole, rather than at atrial contraction. Slower LV relaxation antagonizes the “suction” effect of normal restoring forces on LV filling and results in a delayed mitral valve opening, lower early transmitral gradient and a shift to an LV filling pattern which has a greater proportion of filling at atrial contraction. This filling pattern is less favorable, especially during faster heart rates, because a shorter diastolic filling time may not allow the ventricle to relax and fill to an optimal end-diastolic volume without elevating mean LA pressure and causing pulmonary congestion and dyspnea.

increased filling pressures will be required to maintain a normal LV end-diastolic volume, stroke volume and cardiac output. Left ventricular myocardial compliance is related to the cardiac interstitial elements; the collagen-elastic struts and network that help connect and provide support for the cardiac myocytes. Normally these supporting structural interstitial elements compose less than 5% of cardiac mass. With increase in heart size due to aerobic athletic training, collagen increases in proportion to myocardial mass. However, in the presence of pressure overload induced LV hypertrophy, LV ischemia, or dilated cardiomyopathy, release of angiotensin II and aldosterone occurs and stimulates a disproportionate increase in interstitial elements. An increase in diastolic myocardial stiffness occurs when the collagen concentration increases two to threefold without a similar increase in myocyte volume. Increased resting passive stiffness in DHF may also be due to abnormal phosphorylation of sarcomeric proteins. Viscoelastic properties, or loss of potential energy due to frictional forces associated with LV elastic recoil, are small enough to be ignored except when marked cardiac fibrosis is present. ■

■ ■

Normal ventricular contraction compresses elastic elements whose recoil in early diastole helps augment early diastolic myocardial filling.

LV PASSIVE DIASTOLIC PROPERTIES After LV relaxation is complete, the remainder of LV filling is influenced by more “passive” LV characteristics. These are composed of inherent cardiac elements such as collagen fibers and sarcomeric proteins that affect myocyte and myocardial compliance, and external elements such as the pericardium and pulmonary airway pressure that affect LV chamber compliance. Together the sum effect of all components is described by the exponential diastolic LV pressure-volume (P-V) relationship. This describes the ability of a relaxed or “passive” left ventricle to distend with increasing volume. A decrease in LV chamber compliance means that

1091

Myocardial compliance is affected by inherent cardiac elements such as sarcomeric proteins and cardiac interstitial elements, especially collagen fibers. LV chamber compliance is determined by both myocardial compliance and external elements that affect LV compliance such as the pericardium.

Under normal circumstances about 40% of resting diastolic pressure is due to extrinsic cardiac forces, mostly from the pericardium. However, the effect of pericardial restraint in limiting cardiac filling becomes clinical significant only during maximal exercise, or when there is acute cardiac dilatation or pericardial disease present. Left ventricular chamber compliance decreases with increasing LV volume, in part due to the increased stretch on the elastic interstitial elements. Similarly, only with a marked increase in pulmonary airway pressures, as seen in asthmatics or with positive pressure ventilation, are intracardiac pressures affected sufficiently to inhibit LV filling or decrease cardiac output. ■

The effect of pericardial restraint in limiting cardiac

1092

Section IX Cardiomyopathy and Heart Failure

filling becomes clinically significant only during maximal exercise or when acute cardiac dilatation or pericardial disease is present.

MEASURING LV DIASTOLIC PROPERTIES LV Relaxation The rate of LV relaxation can be measured from highfidelity (micromanometer) pressure recordings taken during LV isovolumic relaxation (Fig. 5). LV pressure change is usually exponential between maximal -dP/dt (occurring approximately at aortic valve closure) and the time of mitral valve opening. The pressure decrease can be described by the relationship: P(t) = Po.e-t/T where Po is LV pressure at maximal -dP/dt (the point at which the rate of LV pressure decline is maximal), t is the time after onset of relaxation, and T is the time constant of isovolumic relaxation (τ). This time constant represents the time for the LV pressure to decrease to l/e of its initial value, τ typically being 30-40 milliseconds in humans, with lower values representing faster relaxation. Relaxation is believed to be “complete” after three to four time constants (120-150 ms in humans), which corresponds in time to shortly after peak early diastolic filling in normals. In mammalian hearts, τ appears proportional to heart rate, being as short as 10 milliseconds in rats, in which the normal resting heart rate is 350 beats/min. ■

■

LV relaxation is quantitated by analyzing the exponential decrease in LV pressure during isovolumic relaxation. Tau (τ) is a quantitative measure of LV relaxation. Lower values represent faster relaxation.

Despite its usefulness, several limitations of quantitating LV relaxation with a simple exponential model are recognized. In patients with hypertrophic cardiomyopathy or markedly asynchronous relaxation, LV pressure decline may significantly deviate from an exponential relation.The simple model also assumes that LV pressure decays to zero pressure, which does not take into account the effect of either LV elastic recoil or external

Fig. 5. Calculation of time constant of myocardial relaxation (tau, τ). LV pressure is plotted on y-axis, and time is plotted on x-axis. Pressure is measured by highfidelity, manometer-tipped catheters. Pressure from time of aortic valve closure (upper dot) to mitral valve opening (lower dot) is fited to a monoexpoential equation. Time constant of relaxation (T) is obtained from equation, as shown. E, natural logarithm; p, pressure; t, time; T, τ.

forces. Studies in normal animals suggest that if LV filling did not occur, LV minimum pressure would be negative. Although the addition of an intercept term to the original equation provides for a “floating”or non-zero LV asymptote pressure, the method to best quantify LV relaxation under different circumstances remains uncertain.

LV PASSIVE PROPERTIES Developing a stress-strain relationship quantitates LV myocardial compliance, or the ability of the muscle to distend. This requires applying a force to a known mass of myocardium while simultaneously measuring its deformation. Because of the many assumptions about LV geometry, and the inability to exclude the effects of external and “active” (relaxation, viscoelastic properties) LV forces, it has been impractical to measure in vivo. Therefore, the sum effect of myocardial compliance,

Chapter 91 Diastolic Heart Function chamber compliance, and external forces is studied by constructing LV pressure-volume (P-V) relationships obtained during diastasis (Fig. 6). Operating chamber compliance, or its reciprocal term “chamber stiffness,” is defined as the slope of a tangent to the P-V relationship at a specified point.The steeper the slope of the tangent, the less compliant or “stiffer” the LV chamber (Fig. 6 A). Because chamber stiffness depends on which point of the P-V curve is used for assessment, several methods have been proposed to normalize this value so that it can be compared after interventions, in serial studies, or between ventricles of different sizes. Although no consensus exists on this point, operating chamber compliance is most commonly measured at LV end-diastolic pressure. ■

LV chamber compliance is evaluated by analyzing the diastolic portion of pressure-volume (P-V) relationships.

Shifts and shape changes in the LV P-V relationship have different implications. Because the P-V relationship is exponential, the same incremental increase in volume results in a greater pressure increase as the ventricle progressively distends (Fig. 6 B). If the shape of the P-V curve does not change, a shift leftward indicates

1093

decreased chamber compliance (same slope tangent at smaller LV volume), and a shift rightward indicates increased compliance (Fig. 6 C). In reality, when measured in sequential studies, curve shifts to the right and left are usually accompanied by alterations in the shape of the P-V curve, and so the incremental change in pressure for a given volume is also changing. The requirements for accurate construction of LV P-V relationships are formidable. The points should be taken after LV relaxation is complete (during diastasis), so that only passive properties are in effect. High-fidelity LV pressure recordings should be used and transmural LV pressure (LV pressure-intrapleural pressure) should be calculated to avoid inaccuracies caused by respiratoryinduced changes in intrapleural pressure.To characterize the P-V relationship over its entire clinical range, LV volume must be varied by rapid changes in preload and afterload without markedly affecting heart rate or LV contraction and relaxation. Accurate calculation of LV volume itself is difficult by current angiographic and echocardiographic methods. These requirements have proved impractical for most clinical studies. As a result, most research involving chamber compliance, or sequential changes in LV chamber compliance has been performed in experimental studies.

Fig. 6. Left ventricular (LV) pressure-volume relationships. Slope of tangent at different pressures (a and b, panel A) represents chamber compliance at different end-diastolic volumes and pressures in the same ventricle. The steeper the slopes the greater the incremental pressure change for the same amount of volume increase (panel B). A leftward shift of the PV relation represents a stiffer chamber, while a rightward shift, a more compliant chamber compared to normal (panel C). P, pressure; V, volume.

1094

Section IX Cardiomyopathy and Heart Failure

LV DIASTOLIC PROPERTIES IN CARDIAC DISEASE STATES It is estimated that as many as 30 million Americans have high blood pressure (>140/80 mm Hg), with only 30% being adequately treated. Therefore, the most common cardiac abnormality encountered in clinical practice is impaired LV relaxation due to hypertension with or without LV hypertrophy. A minority of these patients shows clinical evidence of reduced LV chamber compliance and increased LA pressures. Patients with ischemic heart disease are similar, with the majority showing mostly LV relaxation abnormalities, and only about 10% having altered compliance that raises mean LA pressure to abnormal values. Patients with hypertrophic cardiomyopathy also have abnormal LV relaxation (sometimes severe), but a higher proportion of individuals also demonstrate a decrease in chamber compliance. ■

Shifting of the normal LV P-V relationship to the right indicates a more compliant ventricle and is a compensatory mechanism in patients with dilated cardiomyopathies, even though filling pressures are often elevated due to the increased end-diastolic volume.

A leftward shift of the P-V relationship occurs in patients with DHF due to increased myocardial stiffness. A similar leftward shift is seen in restrictive cardiomyopathies and constrictive pericarditis due to a thickened, noncompliant pericardium.

Patients with isolated mitral or aortic regurgitation most often show little change in LV relaxation or chamber compliance. When the left ventricle is enlarged this indicates a remodeling of the ventricle to an eccentric type hypertrophy whose P-V relation has shifted to the right keeping filling pressures normal despite the increase in cardiac volume. A marked decrease in cardiac compliance in MR or AR indicates severe acute regurgitation or additional ventricular disease. ■

The most common cause of LV diastolic dysfunction when the ejection fraction is normal is ventricular hypertrophy due to hypertension.

Patients with dilated cardiomyopathies typically have both impaired LV relaxation and reduced chamber compliance. Interestingly, the LV pressure-volume relation is shifted right (more compliant) in most cases, but this is offset by the increase in LV volume so that filling pressures are often elevated. More rarely (15 mg/100 mL forearm tissue/min) COX, cyclooxygenase; eNOS, endothelial NO synthase; L-NMMA, NG-monomethyl-L-arginine; NO, nitric oxide.

Table 4. Practical Setup for Brachial Artery Ultrasound Quiet, temperature-controlled examination room Fasting (including caffeine) ≥8 h No smoking (active and passive) No exercise, no night work No mental stress Stop vasoactive medication ≥4 x drug half-life In case of long-term follow-up: note changes in BP, cholesterol level, and weight; be aware of stage of menstrual cycle

angiotensin II receptor blockers also appear to be beneficial. Further studies are needed to develop novel therapeutic agents that restore vascular endothelial homeostasis.

1338

Section XI Invasive and Interventional Cardiology

Fig. 8. Mechanical and hormonal stimuli cause flow-mediated dilation (FMD) of conduit vessels (a). FMD can be assessed in the brachial artery using high-resolution ultrasound (b) at baseline, after occlusion cuff deflation and following the administration of sublingual nitroglycerin.

Chapter 113 Endothelial Dysfunction and Cardiovascular (CV) Disease

1339

Table 5. Markers of Endothelial Dysfunction and Vascular Inflammation Molecule Adhesion molecules ICAM-1 VCAM-1 E-selectin P-selectin sCD40L

Cytokines Interleukin 6

Main sources ECs, circulating leukocytes ECs, VSMC Activated ECs ECs, (Weibel Palade bodies), platelets (α granules) Activated platelets, T lymphocytes, ECs, VSMCs, macrophages, mast cells

ECs, macrophages, fibroblast, T cells

Interleukin 18

Macrophages

TNFα

Macrophages

hs-CRP

Liver

8-iso-PFG2α

Nonenzymatic lipid peroxidation

ET-1

Endothelium, VSMC, others

Metalloproteinases

Macrophages, others

Biological function Promotes leukocyte adherence and migration Promotes leukocyte adherence Promotes leukocyte tethering and rolling EC/platelet-leukocyte interaction Promotes expression of adhesion molecules, tissue factor and metalloproteinases, release of chemokines, generation of reactive oxygen species, B-cell proliferation, generation of memory B cells and antibody class switching; inhibits B-cell apoptosis. Induces acute phase proteins (e.g. CRP), increases antibody production, promotes generation of cytotoxic lymphocytes. Promotes expression of adhesion molecules and metalloproteinases and inflammatory cell recruitment. Promotes expression of adhesion molcules, inflammatory cell recruitment and activation of T and B cells. Promotes upregulation of adhesion molecules, chemoattractants, chemokines and angiotensin type I receptor, generation of reactive oxygen species; reduces endothelial progenitor cell survival and differentiation. Promotes platelet activation, expression of tissue factor and uptake of oxidized LDL by macrophages. Vasoconstrictor; upregulates adhesion molecule expression; promotes VSMC proliferation; modulates effect of numerous compounds. Physiologically involved in tissue remodeling; when upregulated favor plaque instability.

ECs, endothelial cells; hs-CRP, high sensitivity C-reactive protein; ICAM-1, intercellular adhesion molecule-1; 8-iso-PgF2α, 8-iso-prostaglandin F2α; LDL, low-density lipoprotein; sCD40L, soluble CD40 ligand; TNFα, tumor necrosis factor α; VCAM-1, vascular cell adhesion molecule-1; VSMCs, vascular smooth muscle cells.

1340

Section XI Invasive and Interventional Cardiology

Table 6. Effect of Interventions on Endothelial Function and CVD Effect on endothelial Intervention

function

Effect on CVD events

Lipid-lowering therapy Smoking cessation Exercise ACE inhibitors Angiotensin receptor blockers N-3 fatty acids Glycemic control in diabetes mellitus Hormone replacement therapy Vitamin E Combination antioxidants

+ + + + + + + ± ± −

+ + + + + + + − − −

L-arginine

+ + + + + + + + + + + +

? ? ? ? ? ? ? ? ? ? ? ?

Dietary flavonoids Vitamin C Folate Tetrahydrobiopterin Specific metal ion chelation therapy Protein kinase C inhibition Cyclooxygenase-2 inhibition Thromboxane A2 inhibition Troglitazone treatment in diabetes Xanthine oxidase inhibition Tumor necrosis factor inhibition

+, weight of evidence indicates an improvement; −, weight of evidence indicates no effect or worsening; ?, there are insufficient data at the present time. ACE, angiotensin converting enzyme; CVD, cardiovascular disease.

Chapter 113 Endothelial Dysfunction and Cardiovascular (CV) Disease

1341

Table 7. Potential Strategies for Treating Oxidant Stress Agent Vitamin C

Mechanisms of action Free radical scavenger ↓ Inactivation of NO

Vitamin E

↑ ↓ ↓ ↓

ACE inhibitors

↓ LDL oxidation ↓ ACE-oxidant production

eNOS activity Monocyte adhesion LDL oxidation Monocyte adhesion

↓ Inflammatory mediator levels

ARBs

↓ NO bioavailability ↓ NO bioavailability

Statins

↓ LDL oxidation

Metformin

↓ ↓ ↓ ↓ ↓

LDL levels eNOS expression p22phax expression Inflammatory mediators Lipid peroxidation

TDZ

↑ SOD and glutathione levels Promotes less oxidizable LDL conformation

Folate

↓ Inflammatory mediator levels ↓ p47phax expression ↓ eNOS- and XOD-mediated O2 production ↓ eNOS transcription

Estrogen

Summary of clinical trials Improves endothelial dysfunction in NIDDM Further investigation needed to determine if improves cardiovascular mortality

Conflicting long term trials regarding cardiovascular mortality benefits Improves endothelial dysfunction as well as mortality from cardiovascular death independent of antihypertensive effect Tissue-specific may be superior to humorally active ACEIs Conflicting evidence regarding ability to ameliorate endothelial function Improves endothelial dysfunction as well as mortality from cardiovascular death independent of lipid-lowering effect

Improves endothelial dysfunction in NIDDM patients independent of hypoglycemic effect Improves endothelial dysfunction in prediabetic and diabetic states while reducing oxidant levels independent of hypoglycemic or antihypertensive effects

Combination therapy with B12 improves endothelial responses Improves endothelial function, yet not effective in secondary prevention of cardiovascular events

↓ LDL oxidation DM II, diabetes type II; TDZ, thiazolidinediones; SOD, superoxide dismutase; eNOS, endothelial nitric oxide synthase; XOD, xanthine oxidase; ↑, increase; ↓, decrease.

114 CORONARY ARTERY PHYSIOLOGY AND INTRACORONARY ULTRASONOGRAPHY Abhiram Prasad, MD

NORMAL PHYSIOLOGY

■

Myocardial Oxygen Consumption The principal function of the coronary arteries is to provide oxygen and nutrients to the myocardium. • Myocardial oxygen consumption (MVO2) is equal to the product of coronary blood flow and the arteriovenous oxygen gradient across the coronary vascular bed, that is, arterial oxygen content minus coronary sinus oxygen content. In the resting state, myocardial oxygen extraction is near maximum and coronary sinus oxygen saturations are typically 30% or less (or PO2 < 20 mm Hg). Because myocardial oxygen extraction is already • near maximum, MVO2 can increase only by increasing • coronary blood flow. MVO2 is dependent on coronary • blood flow, and changes in MVO2 closely parallel changes in coronary blood flow. Important determi• nants of MVO2 are heart rate, inotropic state (contrac• tility), and intramyocardial wall stress. MVO2 can be approximated clinically by the product of systolic blood pressure and heart rate (called the “rate-pressure product”). The rate-pressure product is an estimate of • MVO2 (and, thus, coronary blood flow) and is frequently used during exercise testing.

■

■

In the resting state, myocardial oxygen extraction is near maximal. • Important determinants of MVO2 are heart rate, inotropic state (contractility), and intramyocardial wall stress.

Coronary Blood Flow Regulation During rest, normal coronary blood flow is approximately 60 to 90 mL/min per 100 g of myocardium. Its regulation in humans is complex and involves metabolic, autonomic, and mechanical factors. The most important metabolic factors include adenosine, prostaglandins, and endothelial-derived factors (e.g. the vasodilator nitric oxide, and the vasoconstrictor endothelin). Myocardial oxygen and carbon dioxide tensions, ATP-sensitive potassium channels (K-ATP channels) also have a role in regulating coronary blood flow. Of the agents released from myocardial cells, adenosine is probably the most important. Adenosine is produced from the breakdown of high-energy phosphates (adenosine triphosphate [ATP]) which cannot be regenerated during ischemia due to the low oxygen tension. The breakdown product, adenosine monophosphate (AMP) accumulates and is converted to adenosine. The contribution of the autonomic nervous system to the control of coronary blood flow is likely

•

MVO2 can be approximated clinically by the product of systolic blood pressure and heart rate. 1343

1344

Section XI Invasive and Interventional Cardiology

modest. Changes in coronary blood flow with either sympathetic or parasympathetic stimulation are due predominantly to the accompanying changes in loading conditions and contractility. Mechanical factors have a major effect on coronary blood flow. During myocardial contraction, intramyocardial pressure increases, causing compression of small vessels and a reduction, or “throttling,” of coronary blood flow. The result is a predominant diastolic blood flow pattern (Fig. 1). Approximately 60% of coronary blood flow occurs during diastole in the left coronary artery. The situation is opposite in the proximal right coronary artery, where there is much less vessel compression during low-pressure right ventricular contraction, with the result that there is much less reduction in blood flow during systole. Blood flow in the proximal right coronary artery during systole is nearly equal to that during diastole. However, in the distal right coronary artery (beyond the right ventricular marginal branches), coronary blood flow predominantly perfuses the inferior left ventricle, and diastolic flow again predominates.

S

D

S

The myocardial compressive effects are greater in the subendocardial layer than in the subepicardial layer, thus making the subendocardium at increased risk for ischemia. During maximal vasodilatation, myocardial perfusion is regulated primarily by coronary perfusion pressure and myocardial compressive effects. When coronary blood flow is reduced, as from an epicardial coronary artery stenosis, the subendocardial layer is the first region of the myocardium to become ischemic. Subendocardial ischemia can be detected clinically with ST-segment depression on an electrocardiogram. Although flow may be adequate at rest, subendocardial ischemia may occur with exercise or stress.This effect can be particularly pronounced in hypertrophied left ventricles, even with normal coronary arteries. Diastolic Pressure-Time Index Coronary blood flow is closely correlated with the diastolic pressure-time index, which is the product of the average difference between aortic and left ventricular cavity pressure and the duration of diastole (i.e., it is the area between diastolic aortic pressure and left ventricular

S

D

S

CFR = 2.6 (adenosine 18 µg)

Fig. 1. Intracoronary Doppler velocities from the left anterior descending coronary artery showing predominant diastolic flow. S, onset of systole; D, onset of diastole. Heart rate and aortic pressure are shown. A, Flow during basal conditions and, B, flow after microvessel vasodilatation with adenosine. Coronary flow reserve (CFR) is the ratio of maximal diastolic flow to basal diastolic flow in the coronary vessel.

Chapter 114 Coronary Artery Physiology and Intracoronary Ultrasonography pressure). The diastolic pressure-time index can be altered by changes in aortic diastolic pressure, left ventricular diastolic pressure, and length of diastole. Coronary blood flow is decreased by systemic hypotension (by decreasing aortic diastolic pressure), increased left ventricular end-diastolic pressure (by increasing left ventricular diastolic pressure), and tachycardia (by shortening diastole). Coronary blood flow can be augmented by increased systemic pressure, decreased left ventricular end-diastolic pressure, and slowing of the heart rate. Intra-aortic balloon pumping can augment coronary blood flow by increasing aortic diastolic pressure. Myocardial Sinusoids The coronary circulation drains primarily through the coronary sinus and cardiac veins (Fig. 2). A small portion of the venous return drains into the thebesian veins and myocardial sinusoids, which empty directly into the chambers of the left side of the heart. A small right-to-left shunt occurs at this level, even in normal subjects. ■

■

■

Approximately 60% of coronary blood flow occurs during diastole in the left coronary artery. Blood flow in the proximal right coronary artery during systole is nearly equal to that during diastole. Coronary blood flow is decreased by systemic hypotension, increased left ventricular end-diastolic pressure, and tachycardia.

Extracardiac branches

Autoregulation of Coronary Blood Flow During resting conditions, coronary blood flow is maintained at a fairly constant level over a range of aortic pressures by the process of autoregulation (Fig. 3). As aortic pressure decreases, coronary blood flow is maintained by dilatation of the resistance vessels. The resistance vessels, or arterioles, are small vessels proximal to the capillaries and are below the resolution of coronary angiography. The converse occurs with an increase in aortic pressure. Therefore, during normal resting conditions, coronary blood flow is pressureindependent. At either extreme, however, autoregulation is overcome and coronary blood flow becomes pressure-dependent. At low perfusion pressures, the resistance vessels are dilated maximally and any additional decrease in pressure results in a linear decrease in blood flow. At pressures less than 70 mm Hg, the pressure-flow relationship becomes linear, with blood flow decreasing in direct proportion to the decrease in perfusion pressure. At very high perfusion pressures, vasoconstriction is maximal and an additional increase in pressure results in a linear increase in blood flow. At extremely low perfusion pressures (approximately 20 mm Hg), blood flow ceases altogether. This effect is called the “vascular waterfall phenomenon,” which is caused by the compressive effects of extravascular intramyocardial pressure. The pressure at which flow ceases is called the “critical closure pressure,” or “the critical flow pressure.”

Arterioluminal vessels

Heart chambers

Veins

Aorta

Coronary arteries

Arterioles

Coronary sinus

Capillaries

Thebesian veins Arteriosinusoidal vessels

Fig. 2. Diagram of the coronary circulation.

1345

Myocardial sinusoids

Heart chambers

Heart chambers

Right atrium

1346

Section XI Invasive and Interventional Cardiology 12-36 μg for the right coronary arteries, respectively). Papaverine has a longer half-life and can prolong the QT interval, rarely resulting in life-threatening arrhythmias. ■

■

Fig. 3. Autoregulation demonstrated by changes in perfusion pressure. ■

■

During normal resting conditions, coronary blood flow is pressure-independent. At pressures less than 70 mm Hg, blood flow decreases in direct proportion to the decrease in perfusion pressure.

Coronary Flow Reserve With physical or mental stress, the metabolic demands of the myocardium increase and coronary blood flow • must increase to increase MVO2. Coronary blood flow increases through dilatation of resistance vessels. When the resistance vessels are dilated maximally, coronary blood flow cannot be increased further without an increase in aortic pressure. The vessels proximal to the resistance vessels (i.e., the epicardial and prearteriolar vessels) offer only minimal resistance to coronary blood flow. The ratio of maximal blood flow to resting (or basal) blood flow is termed the “coronary flow reserve” (CFR) (Fig. 4 and Table 1):

Coronary blood flow increases through dilatation of resistance vessels. Maximal vasodilatation is produced with vasodilators such as adenosine.

Endothelial Function The endothelium comprises the single layer of cells between the vascular smooth muscle and the blood and circulating components. It is the largest “organ” in the body, with approximately one trillion cells, a total surface area equivalent to six tennis courts, and a total weight greater than that of the liver. Although the endothelium functions as a semipermeable membrane, its role in coronary artery physiology is more complex (Table 2). A normally functioning endothelium is essential for maintaining normal coronary blood flow. The increase in coronary blood flow with both physical and mental stress is modulated largely by endothelialdependent changes in vasomotor tone. Endothelial dysfunction is believed to be one of the earliest stages in pathogenesis of atherosclerosis. The endothelium continuously produces substances to modulate vascular tone, including nitric oxide, prostacyclin, and endothelial-derived contracting factors such as endothelin. The relaxing factor produced by the endothelium and originally called “endothelium-derived relaxing factor,” or “EDRF,” was

Maximal Coronary Blood Flow CFR = ________________________ Resting Coronary Blood Flow Coronary flow reserve, also called the “absolute flow reserve,” is a measure of the ability to augment blood flow with stress. It can be measured easily with intracoronary Doppler techniques. Maximal vasodilatation is produced with vasodilators such as adenosine. Adenosine has been the easiest to use because of its short half-life, ability to promote maximal vasodilatation, and safety profile. Bradycardia and complete heart block can occur, particularly with injections into the right coronary artery, but are rare at the recommended doses (intracoronary bolus of 24-60 μg for the left and

Fig. 4. Coronary flow reserve.

Chapter 114 Coronary Artery Physiology and Intracoronary Ultrasonography

1347

Table 1. Comparison of Three Types of Coronary Flow Reserve

Definition

Independent of driving pressure Easily applicable in humans Applicable to 3-vessel disease Unequivocal reference value Abnormal value

Absolute flow reserve

Relative flow reserve

Fractional flow reserve

Ratio of hyperemic to resting flow

Ratio of hyperemic flow in the stenotic region to hyperemic flow in a normal region

Ratio of hyperemic flow in the stenotic region to hyperemic flow in that same region if no lesion is present

− + + − 75% in three or more coronary arteries of significant caliber), particularly when associated with left ventricular dysfunction

1369

1370

Section XI Invasive and Interventional Cardiology

left main or proximal LAD stenoses or ischemic left ventricular dysfunction). Paradoxically, coronary revascularization does not significantly reduce the risk of late myocardial infarction in patients with coronary artery disease. The survival advantage conferred by coronary revascularization diminishes over time due to saphenous vein graft occlusion (50%-60% saphenous graft patency at 10 years) and the development of new stenoses in other coronary vessels. In approximate terms the late cardiac mortality after CABG is about 1 percent per year in patients with mammary grafts and 2.0 percent per year in patients with only saphenous vein grafts—a reflection of the superior patency of internal mammary artery grafts, particularly a LIMA graft to the LAD (>95% patency at 10 years) and RIMA graft patency of about 80% at 10 years. Internal mammary graft failure is generally related to a surgical technical problem (poor distal anastomosis or graft kinking) or intense competitive flow between the native coronary artery and mammary graft rather than late atherosclerosis, as is the case with saphenous vein graft occlusions. A major extrapolation of early CABG versus PTCA revascularization trials to current PCI practice are the multiple pharmacological and technical PCI innovations, particularly drug-eluting stents that have dramatically lowered the coronary restenosis rate following PCI. In early PTCA versus CABG trials, the principal differentiating advantage of CABG over PTCA was the much lower rate of late target vessel revascularization associated with CABG, due primarily to the high coronary restenosis rate in the PTCAtreated patients. Drug-eluting stents have reduced late restenosis to less than 10%, making PCI the preferred initial approach to coronary revascularization for most patients with obstructive coronary artery disease. Indications for Percutaneous Coronary Revascularization (PCI) The leading indication for PCI, independent of whether the patient has anginal symptoms or not, is a large territory of myocardial ischemia due to one or two culprit lesions, deemed technically suitable for PCI in a patient without diabetes mellitus. Patients with and without diabetes mellitus who have one or more coronary lesions that result in a moderate area of

myocardial ischemia should also be considered for PCI. Percutaneous coronary revascularization is indicated for the treatment of moderate to severe anginal symptoms not controlled with medical therapy in patients with anatomic features that place them at increased procedural risk, including vein graft lesions, multivessel disease in patients with diabetes mellitus, and patients with significant left ventricular dysfunction (Tables 2 and 3).

SINGLE-VESSEL CORONARY REVASCULARIZATION Percutaneous coronary revascularization is a good treatment strategy for most symptomatic patients with

Table 2. Indications for PCI in Patients with Chronic Angina or Asymptomatic Myocardial Ischemia Class 1 Nondiabetic patients with one or more hemodynamically significant coronary lesions in one or more coronary arteries suitable for PCI with high likelihood of success and low procedural risk; vessels should subtend a large area of viable myocardium Class IIa Same as class I except the myocardial area at risk is of moderate size or the patient has treated diabetes Class IIb Three or more coronary arteries suitable for PCI with high likelihood of success and low risk; vessels subtend at least moderate area of viable myocardium with evidence of myocardial ischemia Class III Small area of myocardium at risk, absence of ischemia, low likelihood of PCI success, absence of symptoms of ischemia, increased PCI risk, left main stenosis 50%) is associated with significant patient mortality (~30% two-year mortality with medical management) probably because of the very large area of left ventricular myocardium at risk (~70% of the LV). There is an even higher risk in patients with stenosis ≥75 percent or left ventricular failure. Coronary artery bypass graft surgery is the treatment of choice for patients with left main disease. Surgical revascularization has been demonstrated to improve both symptoms and survival compared to medical therapy alone. Left main PCI is generally limited to patients who decline CABG, are inoperable or at very high risk with CABG, or have “protected” left

main disease with a patent CABG graft to either the left anterior descending or circumflex artery from prior CABG surgery. Indeterminate Left Main Coronary Artery Disease Patients with indeterminate left main coronary disease at coronary angiography should have intracoronary ultrasonography (ICUS) performed to guide revascularization. In a Mayo study of 121 patients, the lower range of normal left main minimum luminal area (MLA) was 7.5 mm2. Surgical revascularization was performed on patients below this cutoff value and medium-term follow-up at a mean 3.3 years showed no significant difference in major adverse cardiac events (target vessel revascularization, acute myocardial infarction, and death) between patients with an MLA 10% to 25%) Score 15, risk 35% (range >25%)

Chapter 116 Principles of Interventional Cardiology

1377

Fig. 2. The Mayo Risk Model for procedural major adverse events.

ATHERECTOMY Atherectomy is a catheter-based technique in which the atheromatous plaque is removed with interventional cutting devices. Three atherectomy devices are in clinical use. Directional Coronary Atherectomy Directional coronary atherectomy (DCA) was the first new interventional device approved by the FDA after PTCA. Directional atherectomy consists of a tiny cylinder atop a catheter which passes over a guidewire down the coronary artery. Conceptually the device is attractive in that it allows atheromatous plaque to prolapse through a window in the cylinder head and to be cut using a contained tiny circular cutting blade, trapping the excised tissue within the cylinder head, which is then removed from the patient completely. The downside is the considerable arterial wall injury and resultant stimulus to neointimal proliferation that result. Atherectomy has been studied in several early large randomized trials (CAVEAT-1, CAVEAT-2,

CCAT, and BOAT) and some more recent trials (DESIRE, AMIGO, ADAPTS, SOLD, ABACAS, OARS). CAVEAT-1 Trial The CAVEAT-1 Trial (Coronary Angioplasty Versus Excisional Angioplasty Trial-1) compared DCA with PTCA in de novo native coronary stenoses. DCA was associated with a better initial angiographic result, but an approximate doubling of the procedural complication rate (5% [PTCA] to 11% [DCA] for the combined end point of death, myocardial infarction, or CABG. This excess mortality associated with DCA was still present 1 year postprocedure. CAVEAT-2 Trial The CAVEAT-2 Trial (Coronary Angioplasty Versus Excisional Angioplasty Trial-2) and CCAT (Canadian Coronary Atherectomy Trial) found no significant benefit for DCA over PTCA in saphenous vein graft or proximal LAD lesions, respectively.

1378

Section XI Invasive and Interventional Cardiology

BOAT Trial The Balloon Versus Optimal Atherectomy Trial (BOAT) randomized 89 patients to PTCA or DCA. The major complication rate was low with DCA (no death, 2% Q-wave MI, 1% emergency CABG, and 1.4% perforation), with an angiographic restenosis rate of 31% at 7 months compared to 40% in the PTCA arm. At one year, there was no significant difference between PTCA and DCA in major clinical events. The later atherectomy trials (DESIRE, AMIGO, ADAPTS, SOLD, ABACAS, OARS) showed some improvement in complications in comparison with older atherectomy trails but insufficient benefit to justify routine atherectomy prior to stenting.

(calcium and atheromatous plaque), with relative sparing of the normal elastic arterial wall constituents (differential cutting phenomenon). In theory, the abraded arterial wall elements are pulverized to particles less than 10 μm in diameter that should pass through the microcirculation unhindered. In practice, microembolization may lead to no reflow with myocardial infarction or slow reflow (TIMI 2 flow) with regional wall hypokinesis. ■

■ ■

■

■

■

■

Directional coronary atherectomy showed no statistically significant sustained benefit compared with PTCA in randomized trials performed in either native-vessel or saphenous vein graft disease. Directional coronary atherectomy,while generally associated with a better initial angiographic results, does not reduce the restenosis rate compared to PTCA and was associated with more myocardial infarcts at 30 days and major adverse cardiac events at one year. DCA currently is occasionally used by some experienced operators in highly selected cases in large coronary vessels (≥3.0 mm) with severely eccentric atheromatous plaques particularly proximal located bifurcation lesions or in-stent restenosis lesions prior to further stenting. In the rare cases where DCA is used, it should be used in conjunction with intravascular ultrasound to determine plaque volume and location, which guide DCA use. No randomized trials to date have directly compared directional coronary atherectomy with drug eluting stents.

Rotational and Cutting Balloon Atherectomy Rotational atherectomy utilizes a diamond-studded burr spinning at very high speed (160,000-180,000 revolutions per minute) to selectively abrade the calcified, atheromatous plaque. It is effective in opening many arterial lesions not amenable to conventional PTCA techniques, such as heavily calcified lesions and diffusely diseased vessels, by selectively abrading the relatively more rigid tissues within the arterial wall

Rotational atherectomy is contraindicated in the presence of intraluminal thrombus or in lesions likely to contain thrombus, such as ulcerated lesions associated with unstable angina or vein graft lesions Rotational atherectomy is contraindicated in patients with a preexisting coronary dissection, such as after PTCA with complications.

Rotablator is frequently used is situations where PTCA has failed to open the coronary stenosis due to an inability to either cross or dilate the lesions with the PTCA balloon, although it should not be used if there is any angiographic evidence of dissection. Eccentric lesions can be treated, provided great care is used to ensure the guidewire is coaxial with the vessel lumen (minimize guidewire bias), thus minimizing the risk of vessel perforation. ■

■

■

The luminal area produced by rotational atherectomy is usually larger greater than the burr size used. Rotational atherectomy does not decrease coronary restenosis rates compared to PTCA but the debulking effect on complex calcified lesions makes it a valuable adjunctive tool. Rotational atherectomy is generally followed by stent implantation.

COBRA Trial The COBRA Trial (Comparison of Balloon Angioplasty Versus Rotational Atherectomy) compared PTCA versus rotational atherectomy in just over 500 patients with complex coronary lesions and found no differences in the short- or long-term clinical or angiographic results between these PCI techniques. The trial was performed without stents and the restenosis rate was very high by current standards at approximately 50% in both groups.

Chapter 116 Principles of Interventional Cardiology

■

Rotational atherectomy is associated with unique complications, including slow coronary flow and no coronary reflow phenomenon, arterial wall dissection and perforation, guidewire fracture, burr stall, and, rarely, burr detachment.

Cutting balloon atherectomy is a technique useful in bulky, fibrotic, and ostial or bifurcation lesions. The cutting balloon, handled and delivered in a manner similar to standard balloon catheters, has small cutting microtomes adherent to the surface to “score”

1379

lesions as a stand-alone procedure or, more commonly, to facilitate passage of additional devices such as a stent. ■

■

■

Rotational atherectomy is valuable for heavily calcified lesions or long, diffusely diseased vessels. Rotational atherectomy provides no restenosis benefit over balloon angioplasty alone. Regional wall hypokinesis after rotational atherectomy use may be due to the slow reflow phenomenon due to microemboli.

117 HIGH-RISK PERCUTANEOUS CORONARY INTERVENTIONS Gregory W. Barsness, MD

PERCUTANEOUS CORONARY INTERVENTIONS (PCI) IN SAPHENOUS VEIN GRAFTS ■

Coronary artery saphenous vein bypass grafts have limited long-term longevity. In the first month after coronary revascularization, about 10 percent of vein graft bypass conduits fail due to graft thrombosis secondary to low blood flow often associated with poor distal vessel run-off or competitive blood flow between native coronary artery and graft or due to technical surgical problems such as graft kinking or poor surgical anastomosis. Saphenous vein grafts to the left anterior descending coronary artery have better long-term patency compared to other coronary vessels, as do grafts to native coronary arteries greater than 2.0 mm in diameter.

Sapheous vein graft patency at one week after surgery 90% 1 year after surgery 80% 10 years after surgery 50%

Late saphenous vein occlusion pathologically is characterized by friable, necrotic “gruel” which is a combination of intimal hyperplasia, lipid deposition, cholesterol crystals, foam cells, thrombus and atherosclerotic plaque. Late saphenous vein occlusion increases with years after bypass surgery; poor lipid control

including both low HDL concentration, high LDL concentration and high triglyceride levels; left ventricular dysfunction; male sex; and active smoking. Vein graft failure may lead to recurrent ischemic symptoms or may be asymptomatic in many patients. Repeat bypass surgery is complicated; a periprocedural myocardial infarction rate is 3-5 times that of the first operation, and the operative mortality with second and third bypass procedures may approach 20% in some patients with poor left ventricular function. In addition, repeat coronary bypass grafting is associated with less complete anginal relief and reduced long-term graft patency rates compared to the initial surgery. Advanced age and additional comorbidities contribute to the added-risk nature of repeat bypass surgery. For patients with medically refractory ischemic symptoms related to graft failure, current AHA/ACC guidelines support repeat intervention. Implicit in these guidelines is consideration of a percutaneous intervention as an initial revascularization strategy, particularly in patients with non-LAD territory ischemia or a patent IMA graft (Tables 1 and 2). Percutaneous vein graft interventions entail greater risk and universally poorer long-term outcome than native vessel interventions (Fig. 1). Much of the frustration regarding vein graft intervention relates to the underlying pathophysiology of the vein graft lesion, 1381

1382

Section XI Invasive and Interventional Cardiology

Table 1. AHA/ACC Guidelines for Coronary Angiography in Patients with Post-CABG Ischemia Class I (conditions with evidence or general agreement supporting the usefulness and effectiveness of angiography) None Class IIa (conditions with evidence/opinion in favor of the usefulness of angiography) Recurrent symptoms and ischemia within 12 months of coronary artery bypass grafting High-risk features by noninvasive testing (LVEF 80% indicates a high output and a saturation 1.5 is considered a significant shunt. Total Pulmonary Flow Qp/Qs = __________________ Total Systemic Flow

Table 3. Oximetry for Shunts Step-up, O2% saturated Location SVC-IVC/RA RA/RV RV/PA

Maximum

Mean

>11 >10 >5

>7 >5 >5

Shunt detection, Qp/Qs 1.5-1.9 1.3-1.5 1.3

IVC, inferior vena cava; PA, pulmonary artery; RA, right atrium; RV, right ventricle, SVC, superior vena cava.

Fig. 5. Pulmonary recirculation, or left-to-right shunt. Curve I, Time-indicator concentration relationship from a typical systemic arterial sample. Total pulmonary flow is proportional to shaded area A. Curve II, Generated by subtracting, point-by-point, from curve I the portion of the curve bounding area A. Shunt flow is then proportional to the shaded area As. An approximate formula uses: PC, peak indicator concentration of initial deflection; PCs, peak indicator concentration of early recirculation from the shunt; BT, buildup time for PC; and Ts, time from initial appearance of dye until PCs.

Chapter 118 Invasive Hemodynamics dye curve has a secondary bump. Because it may be difficult to determine whether an extra bump is present, double-sampling dye curves have become standard for diagnosing left-to-right shunts. A complete dye curve measurement should consist of injecting dye first into the pulmonary trunk and simultaneously sampling in the ascending aorta and right ventricle. In the absence of a shunt, sampling in the ascending aorta should produce a normal dye curve. Dye should not appear in the right ventricle until after the blood has fully recirculated through the body. In the presence of a left-to-right shunt, dye appears early in the right ventricle, concomitant with the appearance of dye in the ascending aorta. The magnitude of the shunt can be calculated by comparing the area underneath the two curves, using a forward triangle method. After an intracardiac shunt has been diagnosed, further evaluation by double-sampling dye curves can provide information about shunt localization. This should be done by injecting dye into the right pulmonary artery and sampling the left pulmonary artery and ascending aorta. Next, the dye should be injected into the left pulmonary artery, with sampling in the right pulmonary artery and ascending aorta. In the presence of anomalous pulmonary venous drainage, the dye will appear early after it is injected into one pulmonary artery but not the other. In the presence of an intracardiac shunt, the dye will appear early after it is injected into either pulmonary artery (Fig. 5). After the question of a partial anomalous pulmonary venous drainage has been answered, injections should be made into the main pulmonary artery, with simultaneous sampling in the ascending aorta and in the right side of the heart. The second sampling site should be made in the right ventricle, then in the right atrium, superior vena cava, and inferior vena cava. The early appearance of dye in the right ventricle alone indicates a shunt at the ventricular level. The appearance of dye in the atrium and ventricle indicates a shunt at the atrial level, and the early appearance of dye in the superior vena cava or inferior vena cava indicates the presence of an anomalous pulmonary venous connection with these venous sites. Evaluation of Arterial Desaturation Whenever arterial desaturation (arterial saturation 50 mm Hg). However, it will not provide an accurate transaortic gradient in cases of low-output states or irregular rhythms, as with atrial fibrillation or multiple ectopic beats. The optimal method for obtaining an aortic valve gradient is simultaneously to use two different catheters—one in the left ventricle and one in the aorta—to measure a mean aortic valve gradient. This can be accomplished by a transseptal approach or by using two different arterial accesses. A simultaneous femoral pressure from the sidearm of a sheath has been used to obtain an aortic valve gradient. However, the discrepancy between the femoral artery pressure and the ascending aorta pressure may be significant because of transmission delay and compliance of the peripheral arterial tree, and this may lead to overestimation or underestimation of the mean gradient (Fig. 10). The aortic valve gradient depends on flow and severity of obstruction. An equation for aortic valve area has been described by Gorlin et al. that incorporates pressure and flow for measurement of the severity of stenosis. For this measurement, mean aortic valve gradient, systolic ejection period, heart rate, and cardiac output are used. Flow Area = ____________ 44.3 × C × √ΔP 1,000 × Cardiac Output (L/min) Aortic Flow = _______________________________ Heart Rate (beats/min) × SEP (s/beat) C = 10 where SEP = systolic ejection period. A modified Hakke equation can be used for aortic stenosis. Cardiac Output Area = _____________ √ΔP There are limitations to calculation of aortic valve area. Measurement of flow through the aortic valve needs to be accurate. In a patient with severe concomitant aortic regurgitation, neither the Fick method nor the thermodilution method can be used, because they will underestimate aortic flow. Aortic valve areas are inaccurate at low and high heart rates, at low and high cardiac outputs, and in the presence of irregular rhythms.

Chapter 118 Invasive Hemodynamics

1405

Fig. 10. Catheterization pressures from a patient with aortic stenosis. Left, The simultaneous left ventricular (LV) and femoral artery (FA) pressures are shown. This demonstrates that there is a delay in the femoral artery pressure curve and an “overshoot” from peripheral amplification. Therefore, the true aortic valve gradient cannot be determined. Right, Simultaneous pressures in left ventricle (LV) and ascending aorta (Ao). This is the methodology to determine the true transaortic gradient. PA, pulmonary artery; LA, left atrium.

REGURGITANT VALVULAR LESIONS Regurgitant valvular lesions often are assessed in a cardiac catheterization laboratory by injecting contrast into a cardiac chamber or great vessel and visually estimating the amount of contrast that leaks backward into a more proximal chamber. This approach is only semiquantitative, and it has many limitations. Regurgitant fractions have been used in some cardiac catheterization laboratories, but they are cumbersome and have many sources of error. Doppler echocardiographic techniques provide a more quantitative assessment of the severity of a regurgitant lesion (i.e., volumetric regurgitant fractions, proximal isovelocity surface area) but are operator-dependent and should be done only in experienced laboratories. Injections of Contrast Media Left ventriculography has been the standard for semiquantitation of mitral regurgitation. It consists of injecting 45 to 50 mL of contrast medium at a rate of

12 to 14 mL/s into the left ventricle during cineangiography and examining the density, timing, and appearance of the contrast medium in the left atrium. This requires that a large-bore catheter with side holes be well positioned in the left ventricle. Sellars criteria have been established for a semiquantitative estimate of the degree of mitral regurgitation (Table 4). Left ventriculography has many well-known limitations. The degree to which the contrast medium opacifies the left atrium depends on many factors, including the size and compliance of the left atrium, the size of the left ventricle, the function of the left ventricle, the amount of contrast medium injected, and the rate of injection. Also, catheter entrapment of the mitral apparatus or movement of the catheter into the left atrium can produce erroneous results. Premature beats caused by the catheter or jet of contrast medium also result in erroneous interpretations. The same concept and grading system are used for determining the severity of aortic regurgitation by aor-

1406

Section XI Invasive and Interventional Cardiology

Table 4. Sellars Criteria for Estimating Degree of Mitral Regurgitation Grade 1+ 2+

3+

4+

Criterion Contrast medium does not completely fill left atrium Contrast medium completely opacifies left atrium but does not reach intensity of that in left ventricle Contrast medium completely opacifies left atrium and reaches intensity of that in left ventricle after 4 or 5 beats Contrast medium completely opacifies left atrium and reaches intensity of that in left ventricle within first 2 or 3 beats

ratories. However, there may be questions regarding calculation of these parameters on examinations. For mitral regurgitation, the regurgitant fraction (RF) is the percentage of the total amount of blood ejected by the left ventricle which goes back into the left atrium. The RF is the regurgitant volume (RV) divided by the total amount of blood the ventricle ejects in one beat (total volume [TV]). TV is derived from the left ventriculogram by subtracting the endsystolic volume (ESV) from the end-diastolic volume (EDV).The forward flow volume (FFV) is the amount of blood the ventricle ejects out the aortic valve and is equal to the systemic flow. Thus, this FFV is obtained from the Fick equation. The regurgitant volume is the TV - FFV. TV = EDV − ESV (from left ventriculography) Cardic Output FFV = _____________________________ Heart Rate (from the Fick equation)

tic root angiography. At least 50 to 60 mL of contrast medium is injected into the aortic root at a rate of 20 mL/s, and the severity of aortic regurgitation is evaluated by the amount of contrast medium visualized in the left ventricle. As with left ventriculography, the visual estimate of the degree of aortic regurgitation depends on several factors, including the position of the catheter, the amount of contrast medium injected, rate of injection, the size of the aortic root, and the size and function of the left ventricle. Regurgitant Fractions Regurgitant fractions and regurgitant volume can be calculated in cardiac catheterization laboratories. Although this method was once considered the standard by early investigators, is prone to error due to the sometimes inaccurate measurement of left ventricular volume; thus, it is not routinely used in many clinical labo-

RV = TV − FFV RF = RV/TV ■

Regurgitant Volume Regurgitant Fraction = _____________________ Total Ventricular Volume

The major limitation of this technique is the inability to obtain accurate measurements of angiographic stroke volumes. Various methods have been proposed for making such measurements, including monoplane vs. biplane approaches, planimetrically determined area vs. videodensitometry measurements, and various geometric assumptions of left ventricular size. Each of these methods has inherent limitations. A similar approach can be used for patients with aortic regurgitation.

119 CONTRAST-INDUCED NEPHROPATHY Patricia J. M. Best, MD Charanjit S. Rihal, MD

DEFINITION The most commonly used definition of contrastinduced nephropathy is a rise in the serum creatinine of 0.5 mg/dL. Occasionally, a 1 mg/dL rise in the serum creatinine or a 25% increase in the serum creatinine has been used. More recently, with the improved awareness of the need for estimating glomerular filtration rate (GFR) for its greater accuracy in estimating renal function over the serum creatinine, studies have started to use a 25% decrease in the GFR as the definition of contrast-induced nephropathy.

TIMING Alterations in renal function usually begin 24-48 hours after contrast exposure. The serum creatinine typically peaks at 3-7 days and usually normalizes by 7-10 days. Because the creatinine often does normalize, the importance of contrast-induced nephropathy has been under appreciated.

EPIDEMIOLOGY Careful evaluation is needed to determine if contrast-

induced nephropathy occurred, otherwise it can easily be undiagnosed. The importance of contrast-induced nephropathy is the potential for progression of renal dysfunction including the need for dialysis, and the association with increased mortality. The significance of contrast-induced nephropathy has been underestimated, yet contrast-induced nephropathy accounts for 12% of in-hospital renal failure. This makes it the third most common cause of in-hospital acute renal failure and it exceeds aminoglycosides in its nephrotoxic potential. Importantly, in the cardiac catheterization laboratory this is one of the most common complications, seen in over 3% of cases. It has been associated with increased in-hospital mortality that is over 11 times higher than those who do not develop contrast-induced nephropathy and it is one of the most powerful predictors for mortality after coronary angiography. In one study, the incidence of contrast-induced nephropathy requiring dialysis after percutaneous coronary revascularization was 7 per 1,000 patients, but these patients had an in-hospital mortality of 27.5%. Contrast-induced nephropathy is also associated with long-term mortality, and the importance of this risk is continued over 4 years after percutaneous coronary intervention (Fig. 1). 1407

1408

Section XI Invasive and Interventional Cardiology

Freedom from Death

Survival probability

1.0

No ARF

0.8

ARF

0.6

0.4

0.2

0.0 0

1

2

3

4

Years from procedure

Fig. 1. Kaplan-Meier survival analysis for freedom from death among patients surviving to hospital dismissal. Patients are stratified by presence or absence of contrast-induced nephropathy (ARF), defined as increase in serum Cr ≥0.5 mg/dL from baseline. The survival curve is adjusted for other significant predictors of late survival.

RISK FACTORS FOR CONTRAST-INDUCED NEPHROPATHY Multiple risk factors for contrast-induced nephropathy have been identified. Consistent strong risk factors are baseline renal function, diabetes mellitus, and age. The incidence of contrast-induced nephropathy based on baseline creatinine and diabetic status from the Mayo Clinic Catheterization Laboratory has been described and is helpful in determining the risk of the procedure (Table 1). Other risk factors which have been identified include female sex, volume depletion, need for an intraaortic balloon pump, acute coronary syndromes, congestive heart failure, low cardiac output states, and hypotension.

MECHANISMS OF CONTRAST-INDUCED NEPHROPATHY The specific cause for contrast-induced nephropathy is not known, but it is characterized by acute tubular necrosis. Several potential mechanisms exist. Contrast agents cause vasoconstriction and decrease renal blood flow in the renal medulla. When this occurs it creates a relative hypoxia to the kidney which is undergoing osmotic diuresis from the contrast agent which increases the metabolic demand to the medulla. Pathologically, necrosis may be observed in the thick ascending limb of the loop of Henle. Contrast agents also have a direct toxic effect which is associated with increased renal interstitial inflammation, cellular necrosis, and alterations in cellular enzymes. There are immune mechanisms which have been implicated including activation of the complement system, but these are the least understood mechanisms. Many prevention strategies have tried to target these mechanisms, primarily by preventing renal vasoconstriction and decreasing inflammation and oxidative stress.

RISK SCORES FOR PREDICTING CONTRASTINDUCED NEPHROPATHY Two recent risk scores have been formulated to help predict the development of contrast-induced nephropathy and risk-stratify patients. These risk scores may be very useful to determine when patients should undergo staged procedures for percutaneous coronary intervention and to help identify patients which would be at the highest risk for and in greatest need of preventive therapy. Furthermore, these scores are useful in discussions of the risk-benefit ratio, particularly in patients at the highest risk for contrast-induced nephropathy (Fig. 2,Tables 2 and 3).

Table 1. The Observed Incidence of Contrast-Induced Nephropathy in the Mayo Clinic Catheterization Laboratory Based on Serum Creatinine and Diabetic Status Creatinine, mg/dL 0-1.1 1.2-1.9 2.0-2.9 ≥3.0

Risk, all patients, % 2.4 2.5 22.4 30.6

Risk, diabetic patients, % 3.7 4.5 22.4 33.9

Risk, nondiabetic patients, % 2.0 1.9 22.3 27.4

Chapter 119 Contrast-Induced Nephropathy

Risk Factors

Integer Score

Hypotension

5

IABP

5

CHF

5

Risk Score

Age >75 years

4

?5

7.5%

0.04%

Anemia

3 6 to 10

14.0%

0.12%

Diabetes

3

11 to 16

26.1%

1.09%

?16

57.3%

12.6%

Contrast media volume

Riskof Risk of CIN Dialysis

Calculate

1 for each 100 cc3

Serum creatinine >1.5 mg/dl

1409

4

OR

2 for 40-60 4 for 20-40 6 for