Cardiovascular Magnetic Resonance

Citation preview

Cardiovascular Magnetic Resonance

Cardiovascular Magnetic Resonance SECOND EDITION

Warren J. Manning, MD Professor of Medicine and Radiology Harvard Medical School Section Chief, Non-Invasive Cardiac Imaging Beth Israel Deaconess Medical Center Boston, Massachusetts

Dudley J. Pennell, MD, FRCP, FACC, FESC Professor of Cardiology National Heart and Lung Institute, Imperial College Director, Cardiovascular MR Unit Royal Brompton Hospital London, United Kingdom

Saunders / Elsevier Philadelphia, PA

An Imprint of Elsevier Inc. 1600 John F. Kennedy Blvd. Ste 1800 Philadelphia, PA 19103-2899 CARDIOVASCULAR MAGNETIC RESONANCE, Second Edition

ISBN 978-0-443-06686-3

Copyright # 2010, 2002 by Saunders, an imprint of Elsevier Inc. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. With respect to any drug or pharmaceutical products identified, readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of practitioners, relying on their own experience and knowledge of their patients, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein.

Library of Congress Cataloging-in-Publication Data Cardiovascular magnetic resonance / [edited by] Warren J. Manning, Dudley J. Pennell. – 2nd ed. p. ; cm. Rev. ed. of: Cardiovascular magnetic resonance / Warren J. Manning, Dudley J. Pennell. – 1st ed. 2002. Includes bibliographical references and index. ISBN 978-0-443-06686-3 1. Heart–Magnetic resonance imaging. I. Manning, Warren J. II. Pennell, Dudley J., 1958- III. Manning, Warren J. Cardiovascular magnetic resonance. [DNLM: 1. Cardiovascular Diseases–Diagnosis. 2. Magnetic Resonance Imaging–methods. 3. Diagnostic Techniques, Cardiovascular. WG 141.5.M2 C2664 2010] RC683.5.M35M364 2010 616.10 207548–dc22 2010003032

Acquisitions Editor: Rebecca Schmidt Gaertner Editorial Assistant: David Mack Design Direction: Ellen Zanolle

Printed in the United States of America. Last digit is the print number: 9

8 7 6 5 4

3 2 1

To the joys and inspirations of my life Susan Gail, Anya, Sara, Isaac, and Elie WJM To my parents Terence and Joan for ever being full of pride despite the vicissitudes of age, Elisabeth for always exceeding the singular prowess proffered at our wedding despite denial, and Indigo Lucy Li Ling for inspiring love and joy in boundless measure (how many times today Indi?). DJP

CONTRIBUTORS

Contributors Silvia Aguiar, MD Translational and Molecular Institute, Departments of Radiology and Medicine (Cardiology), Mount Sinai School of Medicine, New York, New York Chapter 25: Atherosclerotic Plaque Imaging: Aorta and Carotid Timothy S. E. Albert, MD Medical Director, Cardiovascular Diagnostic Center, Salinas Valley Memorial Healthcare System, Monterey, California; Assistant Consulting Professor of Medicine, Duke University Medical Center, Durham, North Carolina Chapter 2: Clinical Cardiovascular Magnetic Resonance Imaging Techniques Francisco Alpendurada, MD Consultant, CMR Unit, Royal Brompton Hospital, London, United Kingdom Chapter 39: Cardiac and Paracardiac Masses Andrew E. Arai, MD Senior Investigator, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland Chapter 18: Acute Myocardial Infarction: Cardiovascular Magnetic Resonance Detection and Characterization Robert S. Balaban, PhD Scientific Director, Division of Intramural Research, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland Chapter 1: Basic Principles of Cardiovascular Magnetic Resonance Jeroen J. Bax, MD Professor of Cardiology, Department of Cardiology, Leiden University Medical Center, Leiden, The Netherlands Chapter 8: Special Considerations for Cardiovascular Magnetic Resonance Safety, Electrocardiographic Setup, Monitoring, and Contraindications Nicholas G. Bellenger, MD Consultant Cardiologist, CMR Unit, Royal Brompton Hospital, London, United Kingdom Chapter 14: Assessment of Cardiac Function

David A. Bluemke, MD Professor, Radiology and Medicine, Johns Hopkins University School of Medicine, Baltimore; Director, Radiology and Imaging Sciences, National Institutes of Health Clinical Center; Senior Investigator, National Institute of Biomedical Imaging and Bioengineering, Bethesda, Maryland Chapter 35: Pulmonary Artery Cardiovascular Magnetic Resonance Rene´ M. Botnar, PhD Professor of Cardiovascular Imaging, Chair of Cardiovascular Imaging, Imaging Sciences Division, King’s College London, London, United Kingdom Chapter 21: Coronary Artery and Vein Imaging: Methods Chapter 26: Atherosclerotic Plaque Imaging: Coronaries Jens Bremerich, MD Head, Cardiothoracic Radiology, University Hospital Basel, Basel, Switzerland Chapter 30: Cardiovascular Magnetic Resonance in Complex Congenital Heart Disease Peter T. Buser, MD Professor of Cardiology, Chairman, Department of Cardiology; Head, Cardiac Imaging, University Hospital Basel, Basel, Switzerland Chapter 30: Cardiovascular Magnetic Resonance in Complex Congenital Heart Disease Peter Caravan, PhD Assistant Professor, Radiology, Harvard Medical School; Assistant in Chemistry, Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Boston, Massachusetts Chapter 6: Cardiovascular Magnetic Resonance Contrast Agents Jonathan Chan, MD Advanced Cardiovascular Imaging Fellow, Department of Medicine, Cardiovascular Division, Beth Israel Deaconess Medical Center, Boston, Massachusetts Chapter 22: Coronary Artery Imaging: Clinical Results Michael L. Chuang, MD Advanced Cardiovascular Imaging Fellow, Department of Medicine, Cardiovascular Division, Beth Israel Deaconess Medical Center, Boston, Massachusetts Chapter 11: Normal Cardiac Anatomy, Orientation, and Function

Cardiovascular Magnetic Resonance vii

CONTRIBUTORS

Albert de Roos, MD Professor of Radiology, Department of Radiology, Leiden University Medical Centre, Leiden, The Netherlands Chapter 29: Cardiovascular Magnetic Resonance of Simple Congenital Cardiovascular Defects Rohan Dharmakumar, MD Department of Radiology, Northwestern University, Chicago, Illinois Chapter 42: Magnetic Resonance Assessment of Myocardial Oxygenation Adam L. Dorfman, MD Assistant Professor, Department of Pediatrics and Communicable Diseases; Assistant Professor, Department of Radiology, University of Michigan Medical School, Ann Arbor, Michigan Chapter 31: Complex Congenital Heart Disease: Infant and Pediatric Patients Christopher K. Dyke, MD Consulting Associate, Duke University Medical Center, Durham, North Carolina; Adjunct Faculty, University of Washington School of Medicine, Seattle, Washington; Director, Alaska Heart Cardiovascular MRI and CT Centers, Alaska Heart Institute, Anchorage, Alaska Chapter 2: Clinical Cardiovascular Magnetic Resonance Imaging Techniques Robert R. Edelman, MD William B. Graham Chair, Department of Radiology, North Shore University Health System; Professor of Radiology, Feinberg School of Medicine, Northwestern University, Evanston, Illinois Chapter 34: Cardiovascular Magnetic Resonance Angiography: Carotids, Aorta, and Peripheral Vessels Michael D. Elliott, MD Director, Cardiovascular MR/CT, St. Vincent Heart Center, Indianapolis, Indiana Chapter 2: Clinical Cardiovascular Magnetic Resonance Imaging Techniques Zahi A. Fayad, PhD Professor, Department of Radiology and Medicine (Cardiology); Director, Translational Molecular Imaging Institute, Mount Sinai School of Medicine, New York, New York Chapter 25: Atherosclerotic Plaque Imaging: Aorta and Carotid David Firmin, MD Professor of Physics, Royal Brompton Hospital and Imperial College, London, United Kingdom Chapter 7: Blood Flow Velocity Assessment Chapter 10: Use of Navigator Echoes in Cardiovascular Magnetic Resonance and Factors Affecting Their Implementation

viii Cardiovascular Magnetic Resonance

Mark A. Fogel, MD Associate Professor of Pediatrics and Radiology, The University of Pennsylvania School of Medicine; Director of Cardiac Magnetic Resonance, The Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania Chapter 9: Special Considerations: Cardiovascular Magnetic Resonance in Infants and Children Herbert Frank, MD Professor in Internal Medicine; Director, Department of Internal Medicine, Landesklinikum Tulln and Medical University of Vienna, Vienna, Austria Chapter 39: Cardiac and Paracardiac Masses Matthias G. Friedrich, MD Associate Professor, Libin Cardiovascular Institute of Alberta, University of Calgary, Calgary, Alberta; Director, Stephenson CMR Centre, Foothills Medical Centre, Libin Cardiovascular Institute of Alberta, University of Calgary, Calgary, Alberta, Canada Chapter 38: Cardiomyopathies Tal Geva, MD Professor of Pediatrics, Harvard Medical School; Chief, Division of Noninvasive Imaging, Department of Cardiology, Children’s Hospital Boston, Boston, Massachusetts Chapter 31: Complex Congenital Heart Disease: Infant and Pediatric Patients James W. Goldfarb, PhD Assistant Professor of Biomedical Engineering, Stony Brook University, The Heart Center, Saint Francis Hospital, Roslyn, New York Chapter 34: Cardiovascular Magnetic Resonance Angiography: Carotids, Aorta, and Peripheral Vessels David Grand, MD Warren Alpert Medical School of Brown University, Department of Diagnostic Imaging, Rhode Island Hospital, Providence, Rhode Island Chapter 35: Pulmonary Artery Cardiovascular Magnetic Resonance John D. Grizzard, MD Associate Professor, Section Chief, Non invasive Cardiovascular Imaging, Department of Radiology, Virginia Commonwealth University, Richmond, Virginia Chapter 2: Clinical Cardiovascular Magnetic Resonance Imaging Techniques Craig A. Hamilton, PhD Associate Professor, Department of Biomedical Engineering, Wake Forest University School of Medicine, Winston Salem, North Carolina Chapter 15: Wall Motion Stress Cardiovascular Magnetic Resonance: Ischemia, Viability, and Prognosis

Robert M. Judd, PhD Associate Professor of Medicine and Radiology; Co Director, Duke Cardiovascular Magnetic Resonance Center, Duke University, Durham, North Carolina Chapter 2: Clinical Cardiovascular Magnetic Resonance Imaging Techniques

Sanjeet R. Hegde, MD, MRCPCH, MBBS Clinical Research Fellow, Evelina Children’s Hospital, Guy’s, St. Thomas NHS Foundation and Trust Division of Imaging Sciences, King’s College London, London, United Kingdom Chapter 44: Pediatric Interventional Cardiovascular Magnetic Resonance

Jennifer Keegan, PhD, MSc Principal Physicist and Honorary Senior Lecturer, Royal Brompton and Imperial College, London, United Kingdom Chapter 10: Use of Navigator Echoes in Cardiovascular Magnetic Resonance and Factors Affecting Their Implementation Chapter 23: Coronary Artery and Sinus Velocity and Flow

Charles B. Higgins, MD Professor of Radiology, University of California, San Francisco Medical School; UCSF Medical School, San Francisco, California Chapter 30: Cardiovascular Magnetic Resonance in Complex Congenital Heart Disease

Philip J. Kilner, MD Consultant CMR Unit Royal Brompton Hospital, London, United Kingdom Chapter 37: Valvular Heart Disease

Agnes E. Holland, MD Washington Radiology Associates, Washington, DC Chapter 34: Cardiovascular Magnetic Resonance Angiography: Carotids, Aorta, and Peripheral Vessels Peng Hu, PhD Cardiac MR Center, Beth Israel Deaconess Medical Center, Boston, Massachusetts Chapter 21: Coronary Artery and Vein Imaging: Methods W. Gregory Hundley, MD Professor, Departments of Biomedical Engineering, Internal Medicine, Molecular Medicine, and Radiology, Wake Forest University School of Medicine, Winston Salem, North Carolina Chapter 15: Wall Motion Stress Cardiovascular Magnetic Resonance: Ischemia, Viability, and Prognosis Fabien Hyafil, MD, PhD Assistant Professor, Departments of Nuclear Medicine and Cardiology, Inserm 698 “Cardiovascular Remodeling,” Bichat University Hospital, Assistance Publique Hopitaux de Paris, University Paris, Paris, France Chapter 25: Atherosclerotic Plaque Imaging: Aorta and Carotid Hu¨seyin Ince, MD Department of Internal Medicine I, Divisions of Cardiology, Pneumology, and Intensive Care, University Hospital, University of Rostock, Rostock, Germany Chapter 33: Thoracic Aortic Disease Michael Jerosch-Herold, PhD Associate Professor of Radiology, Harvard Medical School; Director of Cardiac Imaging Physics, Department of Radiology, Brigham and Women’s Hospital, Boston, Massachusetts Chapter 4: Myocardial Perfusion Imaging Theory

Hee-Won Kim, PhD Assistant Professor, Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, California Chapter 40: Cardiac Transplantation Raymond J. Kim, MD Associate Professor of Medicine and Radiology, Co Director, Duke Cardiovascular Magnetic Resonance Center, Duke University, Durham, North Carolina Chapter 2: Clinical Cardiovascular Magnetic Resonance Imaging Techniques Won Yong Kim, MD, PhD Associate Professor, Department of Cardiology and MR Center, Aarhus University Hospital, Aarhus, Denmark Chapter 26: Atherosclerotic Plaque Imaging: Coronaries Stephan Kische, MD Department of Internal Medicine I, Divisions of Cardiology, Rheumatology, and Intensive Care, University Hospital, University of Rostock, Rostock, Germany Chapter 33: Thoracic Aortic Disease Kraig V. Kissinger, RT(R)(MR) Senior Technologist, Cardiac MR Center, Beth Israel Deaconess Medical Center, Boston, Massachusetts Chapter 21: Coronary Artery and Vein Imaging: Methods Grigorios Korosoglou, MD University of Heidelberg, Department of Cardiology, Heidelberg, Germany Chapter 13: High Field Cardiovascular Magnetic Resonance Christopher M. Kramer, MD Professor of Medicine and Radiology; Director, Cardiovascular Imaging Center, University of Virginia Health System, Charlottesville, Virginia Chapter 19: Acute Myocardial Infarction: Ventricular Remodeling

Cardiovascular Magnetic Resonance ix

CONTRIBUTORS

Thomas H. Hauser, MD, MMSc, MPH Assistant Professor of Medicine, Harvard Medical School; Director of Nuclear Cardiology, Beth Israel Deaconess Medical Center, Boston, Massachusetts Chapter 22: Coronary Artery Imaging: Clinical Results Chapter 32: Pulmonary Vein Imaging

CONTRIBUTORS

Lucia J. M. Kroft, MD, PhD Academic Radiologist, Department of Radiology, Leiden University Medical Center, Leiden, The Netherlands Chapter 29: Cardiovascular Magnetic Resonance of Simple Congenital Cardiovascular Defects Robert J. Lederman, MD Senior Investigator, Translational Medicine Branch, Division of Intramural Research, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland Chapter 43: Interventional Cardiovascular Magnetic Resonance Debaio Li, PhD Professor of Radiology, Northwestern School of Medicine; Professor of Biomedical Engineering, Northwestern School of Engineering, Chicago, Illinois Chapter 42: Magnetic Resonance Assessment of Myocardial Oxygenation Alicia M. Maceira, MD ERESA Hospital General de Castellon, Castellon, Spain Chapter 14: Assessment of Cardiac Function Chapter 28: Cardiovascular Magnetic Resonance Assessment of Right Ventricular Anatomy and Function Heiko Mahrholdt, MD Division of Cardiology, Robert Bosch Krankenhaus, Stuttgart, Germany Chapter 20: Myocardial Viability David Maintz, MD Professor of Radiology, Department of Clinical Radiology, University of Munster, Munster, Germany Chapter 26: Atherosclerotic Plaque Imaging: Coronaries Warren J. Manning, MD Professor of Medicine and Radiology, Harvard Medical School; Section Chief, Non invasive Cardiac Imaging, Beth Israel Deaconess Medical Center, Boston, Massachusetts Chapter 11: Normal Cardiac Anatomy, Orientation, and Function Chapter 21: Coronary Artery and Vein Imaging: Methods Chapter 22: Coronary Artery Imaging: Clinical Results Chapter 26: Atherosclerotic Plaque Imaging: Coronaries Constantin B. Marcu, MD Cardiologist, Director of Advanced Cardiac Imaging, Hospital of Saint Raphael, Yale University, New Haven, Connecticut Chapter 24: Coronary Artery Bypass Graft Imaging and Assessment of Flow Raad H. Mohiaddin, MD Consultant CMR Unit Royal Brompton Hospital, London, United Kingdom Chapter 27: Assessment of the Biophysical Properties of the Arterial Wall Chapter 37: Valvular Heart Disease x Cardiovascular Magnetic Resonance

Eike Nagel, MD Professor of Clinical Cardiovascular Imaging; Chair of Cardiovascular Imaging, King’s College London, Division of Imaging Sciences, BHF Centre of Excellence, NIHR Biomedical Research Centre and Wellcome Trust, EPSRC Medical Engineering Centre, King’s College London, London, United Kingdom Chapter 17: Comparison of Perfusion and Wall Motion Cardiovascular Magnetic Resonance Imaging Stefan Neubauer, MD Professor of Cardiovascular Medicine, University of Oxford; Honorary Consultant Cardiologist, John Radcliffe Hospital, Oxford, United Kingdom Chapter 41: Cardiovascular Magnetic Resonance Spectroscopy Reza Nezafat, PhD Assistant Professor of Medicine, Harvard Medical School, Boston; Director, Translational Research, Cardiac MR Center, Beth Israel Deaconess Medical Center, Boston, Massachusetts Chapter 21: Coronary Artery and Vein Imaging: Methods Christoph A. Nienaber, MD Director, Department of Internal Medicine I, Divisions of Cardiology, Rheumatology, and Intensive Care, University Hospital, University of Rostock, Rostock, Germany Chapter 33: Thoracic Aortic Disease Thoralf Niendorf, PhD Professor for Experimental Ultra High Field MRI; Charite´ University Medicine; Director, Berlin Ultrahigh Field MR Faculty, Max Delbru¨ck Center for Molecular Medicine, Berlin, Germany Chapter 3: Advanced Cardiovascular Magnetic Resonance Imaging Techniques: Spiral, Radial, and Parallel Imaging Noriko Oyama, MD, PhD Assistant Professor, Department of Radiology, Hokkaido University Hospital, Sapporo, Japan Chapter 36: The Pericardium: Normal Anatomy and Spectrum of Disease Ingo Paetsch, MD Assistant Professor of Internal Medicine, Lecturer of Internal Medicine/Cardiology, Charite´ Medical School, Berlin; Director of Cardiovascular Magnetic Resonance Unit, German Heart Institute, Berlin, Germany Chapter 17: Comparison of Perfusion and Wall Motion Cardiovascular Magnetic Resonance Imaging Rajan A. G. Patel, MD Assistant Clinical Staff, Ochsner Clinic Foundation, New Orleans, Louisiana Chapter 19: Acute Myocardial Infarction: Ventricular Remodeling

Ronald M. Peshock, MD Professor of Radiology and Internal Medicine; Assistant Dean for Informatics, University of Texas Southwestern Medical Center at Dallas, Dallas, Texas Chapter 11: Normal Cardiac Anatomy, Orientation, and Function Dana C. Peters, PhD Assistant Professor, Department of Medicine, Harvard Medical School; Scientific Director of the Cardiac MR Center, Beth Israel Deaconess Medical Center, Boston, Massachusetts Chapter 1: Basic Principles of Cardiovascular Magnetic Resonance Chapter 32: Pulmonary Vein Imaging Sven Plein, MD, PhD Consultant Cardiologist, Wellcome Intermediate Clinical Fellow, Academic Unit of Cardiovascular Medicine, University of Leeds, Leeds General Infirmary, Leeds, United Kingdom Chapter 12: Comprehensive Cardiovascular Magnetic Resonance in Coronary Artery Disease Gerald M. Pohost, MD Professor, Keck School of Medicine; Professor, Viterbi School of Engineering, University of Southern California; Professor, School of Medicine, Loma Linda University, Loma Linda; Director, Cardiovascular Imaging, Westside Medical Imaging, Los Angeles, California Chapter 40: Cardiac Transplantation Andrew J. Powell, MD Associate Professor of Pediatrics, Harvard Medical School; Senior Associate in Cardiology, Department of Cardiology, Children’s Hospital Boston, Boston, Massachusetts Chapter 31: Complex Congenital Heart Disease: Infant and Pediatric Patients Chirapa Puntawangkoon, MD Research Fellow, Section on Cardiology, Department of Internal Medicine, Wake Forest University School of Medicine, Winston Salem, North Carolina Chapter 15: Wall Motion Stress Cardiovascular Magnetic Resonance: Ischemia, Viability, and Prognosis Amish N. Raval, MD Assistant Professor of Medicine, Division of Cardiovascular Medicine, Department of Medicine, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin Chapter 43: Interventional Cardiovascular Magnetic Resonance

Reza S. Razavi, MD, MBBs Professor of Paediatric Cardiovascular Science, Division of Imaging Sciences, King’s College London; Consultant Cardiologist, Guy’s and St. Thomas NHS Foundation Trust, London, United Kingdom Chapter 44: Pediatric Interventional Cardiovascular Magnetic Resonance Wolfgang G. Rehwald, PhD Senior Scientist, Cardiovascular MR Research and Development, Siemens Healthcare, Siemens Medical Solutions USA, Inc., Customer Solutions Group, Chicago, Illinois Chapter 2: Clinical Cardiovascular Magnetic Resonance Imaging Techniques Arno A. W. Roest, MD, PhD Fellow, Pediatric Cardiology, Department of Pediatric Cardiology, Leiden University Medical Center, Leiden, The Netherlands Chapter 29: Cardiovascular Magnetic Resonance of Simple Congenital Cardiovascular Defects James H. F. Rudd, PhD, MRCP HEFCE Senior Lecturer, Division of Cardiovascular Medicine, University of Cambridge; Honorary Consultant Cardiologist, Addenbrooke’s Hospital, Cambridge, United Kingdom Chapter 25: Atherosclerotic Plaque Imaging: Aorta and Carotid Juerg Schwitter, MD Associate Professor, University Hospital Lausanne; Director, CMR Center of the University Hospital Lausanne CHUV, Cardiology, University Hospital Lausanne, Lausanne, Switzerland Chapter 16: Stress Cardiovascular Magnetic Resonance: Myocardial Perfusion Udo P. Sechtem, MD Chairman, Department of Cardiology, Robert Bosch Krankenhaus, Stuttgart; Associate Professor of Medicine and Cardiology, University of Tu¨bingen, Tu¨bingen, Germany Chapter 20: Myocardial Viability Burkhard Sievers, MD Department of Medicine/Cardiology, Heart Center Dresden, University Hospital, Dresden, Germany Chapter 2: Clinical Cardiovascular Magnetic Resonance Imaging Techniques Daniel K. Sodickson, MD, PhD Vice Chair for Research, Department of Radiology; Director, Center for Biomedical Imaging; Associate Professor of Radiology, Physiology, and Neuroscience, New York University Langone Medical Center, New York, New York Chapter 3: Advanced Cardiovascular Magnetic Resonance Imaging Techniques: Spiral, Radial, and Parallel Imaging Cardiovascular Magnetic Resonance xi

CONTRIBUTORS

Dudley J. Pennell, MD Professor of Cardiology, National Heart and Lung Institute, Imperial College; Director, Cardiovascular MR Unit, Royal Brompton Hospital, London, United Kingdom Chapter 14: Assessment of Cardiac Function Chapter 23: Coronary Artery and Sinus Velocity and Flow Chapter 28: Cardiovascular Magnetic Resonance Assessment of Right Ventricular Anatomy and Function

CONTRIBUTORS

Elmar Spuentrup, MD Department of Radiology, University Hospital, University of Cologne, Cologne, Germany Chapter 26: Atherosclerotic Plaque Imaging: Coronaries Matthias Stuber, PhD Professor, University of Lausanne, Center for Biomedical Imaging (CIBM); Director, CIBM, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland Chapter 5: Cardiovascular Magnetic Resonance Tagging Assessment of Left Ventricular Diastolic Function Chapter 13: High Field Cardiovascular Magnetic Resonance Nicholas R. Teman, MD House Officer, Department of General Surgery, University of Michigan Health System, Ann Arbor, Michigan Chapter 40: Cardiac Transplantation Ernst E. van der Wall, MD Professor of Cardiology; Head, Department of Cardiology, Leiden University Medical Center, Leiden, The Netherlands Chapter 8: Special Considerations for Cardiovascular Magnetic Resonance: Safety, Electrocardiographic Setup, Monitoring, and Contraindications Albert C. van Rossum, MD, PhD Professor of Cardiology; Chairman of the Department of Cardiology, VU University Medical Center, Amsterdam, The Netherlands Chapter 24: Coronary Artery Bypass Graft Imaging and Assessment of Flow

xii Cardiovascular Magnetic Resonance

Anja Wagner, MD Department of Cardiology, Hahnemann University Hospital, Drexel University College of Medicine, Philadelphia, Pennsylvania Chapter 2: Clinical Cardiovascular Magnetic Resonance Imaging Techniques Thomas F. Walsh, MD Senior Assistant Resident, Department of Internal Medicine, Wake Forest University School of Medicine, Winston Salem, North Carolina Chapter 15: Wall Motion Stress Cardiovascular Magnetic Resonance: Ischemia, Viability, and Prognosis Norbert Wilke, MD Clinical Associate Professor of Medicine (Cardiology); Clinical Associate Professor of Radiology, Cardiovascular MR and CT Services, University of Florida, Gainesville, Florida Chapter 4: Myocardial Perfusion Imaging Theory Rolf Wyttenbach, MD, PhD Chairman, Department of Diagnostic and Interventional Radiology, Ospedale Regionale di Bellinzone e Valli (EOC), Bellinzona, Switzerland Chapter 30: Cardiovascular Magnetic Resonance in Complex Congenital Heart Disease Susan B. Yeon, MD, JD Assistant Professor of Medicine, Harvard Medical School, Cardiovascular Division, Beth Israel Deaconess, Boston, Massachusetts Chapter 26: Atherosclerotic Plaque Imaging: Coronaries Chapter 36: The Pericardium: Normal Anatomy and Spectrum of Disease

Since William Harvey’s discovery of the circulation, the assessment of cardiac structure and function has been a major goal of physicians responsible for the management of patients with heart disease. The first techniques were quite primitive palpation of the arterial pulse, observation and direct auscultation of the precordium followed by Rene´ Lae¨nnec’s development of the stethoscope, the first tool to amplify the senses. A paradigm shift occurred with the discovery of the x ray by Wilhelm Roentgen at the end of the 19th century, allowing the heart to be imaged for the first time. The 20th century can be considered the century of car diac imaging, beginning with angiography and arteriogra phy, both requiring the injection of contrast material through a catheter. The development and progressive refinement of noninvasive approaches echocardiography, radionuclide imaging, and computed tomography advanced cardiology enormously. Cardiovascular magnetic resonance imaging, however, may finally fulfill the dreams of cardiologists. The technique offers a veritable treasure trove of information about the heart and circulation. It not only provides superior displays of chamber structure but allows the identification of myocardial necrosis, infiltra tion, and fibrosis. Both valvular performance and ventri cular function can be measured. Magnetic resonance angiography allows assessment of vascular function and tis sue perfusion. Recently, both the walls and luminae of the major coronary arteries have been visualized. When

combined with the ability to determine myocardial viability and to measure cardiac high energy phosphate stores non invasively by magnetic resonance spectroscopy, these tech niques seem certain to improve enormously the care of patients with ischemic heart disease. It is no longer fanciful to think that coronary arteriography, now performed on more than 1.5 million patients in the United States each year, will become obsolete. Just as it became necessary for physicians entrusted with the care of cardiac patients to become familiar with cardiac roentgenography at the beginning of the 20th century, it now has become vital for their successors to do the same with cardiac magnetic resonance imaging. Warren J. Man ning and Dudley J. Pennell and their talented contributors to Cardiovascular Magnetic Resonance have taken a signifi cant step toward enabling clinicians to accomplish this. They deserve the appreciation of cardiovascular specialists of all types for producing this detailed, authoritative, yet eminently readable book. A century ago, great strides were being made in cardiac roentgenography and the future promise was enormous. The situation is similar with car diac magnetic resonance imaging today. This field is advancing at such a breathtaking speed that I can’t wait for the second edition. Eugene Braunwald, MD Boston, Massachusetts

Cardiovascular Magnetic Resonance xiii

FOREWORD TO THE FIRST EDITION

Foreword to the First Edition

PREFACE

Preface Cardiovascular magnetic resonance (CMR) is a medical imaging field that excites great interest because it combines superb image quality with new techniques for probing the cardiovascular system in novel ways. What surprises is the versatility of the technology: blood flow, angiography, assessment of atherosclerosis, myocardial perfusion, focal necrosis, oxygen saturation, technology, and chemical com position are among the measurements that are being refined for clinical use, in addition to the well known “gold standard” capabilities of CMR in defining anatomy and ventricular function. Such potential comes at a price, how ever, as this technology is not quickly learned, and high quality clinical practice needs experience. Professional didactic and clinical training is required for all newcomers to the field and to maintain cutting edge competency.

The aim of this book is to provide instruction in the current clinical practice of CMR, while also highlighting areas of clinical potential, which are presently in varying stages of development. If we succeed in drawing new investigators and clinicians to enter the field or illuminating new areas for those already involved, then we will have achieved our objective the healthy growth of competent and moti vated practitioners in CMR for the benefit of clinical sci ence, patient care, and the advancement of the field. The reader should be forewarned that CMR is constantly devel oping, and no text can include all of the most recent devel opments; however, the foundations provided will serve the reader for many years to come. The future of CMR is bright! Join us in the adventure! WJM, DJP

Cardiovascular Magnetic Resonance xv

ACKNOWLEDGMENTS

Acknowledgments It takes a town to create a book. Like any large endeavor, this text is a success because of the effort of many indivi duals to whom we owe thanks. These include the outstand ing contributions of our primary authors and their collaborators; Rebecca Schmidt Gaertner, our editor at Else vier Health Sciences; Linda R. Garber, our production edi tor; our office assistants, Iris Wasserman at the Beth Israel Deaconess Medical Center and Fei Wang at the Royal

Brompton Hospital; the multitude of CMR mentors, train ees, and colleagues who have stimulated and educated us over the past two decades; and most of all to our families, for allowing us the time to pursue this endeavor amongst the myriad of other activities that occupy our daily lives. To all, we express our thanks and high level of appre ciation and gratitude. WJM, DJP

Cardiovascular Magnetic Resonance xvii

2D 3D

two dimensional three dimensional

A2C A4C ACE-I ACS ADP AMI ARVC ASD AT 1R AT 2R ATP a.u. AUC AV AVM

apical two chamber apical four chamber angiotensin converting enzyme inhibitor acute coronary syndrome adenosine diphosphate acute myocardial infarction arrhythmogic right ventricular cardiomyopathy atrial septal defect angiotensin 1 receptor angiotensin 2 receptor adenosine triphosphate arbitrary units area under the curve atrioventricular arteriovenous malformation

BOLD

blood oxygen level dependent

C CABG CAD CE CE-CTA

carbon coronary artery bypass graft coronary artery disease contrast enhanced contrast enhanced computed tomography angiography CE-MRA contrast enhanced magnetic resonance angiography CFR coronary flow reserve CHD congenital heart disease CMP cardiomyopathy CMR cardiovascular magnetic resonance CMRS cardiovascular magnetic resonance spectroscopy CO cardiac output CNR contrast to noise ratio CP creatine phosphate CS circumferential shortening CSPAMM Complementary SPAtial Modulation of Magnetization CSI chemical shift imaging CT computed tomography DC DCM DCMR DORV DPG DSE DTPA

direct current dilated cardiomyopathy dobutamine stress cardiovascular magnetic resonance double outlet right ventricle diphosphoglycerate dobutamine stress echocardiography diethylenetriamine pentaacetic acid

Dy

dysprosium

ECF ECG ECM EDV EF EPI ESV

extracellular fluid electrocardiogram extracellular matrix end diastolic volume ejection fraction echo planar imaging end systolic volume

FFR FID FSE

fractional flow reserve free induction decay fast spin echo

Gd gadolinium Gd-DTPA gadolinium diethylenetriamine pentaacetic acid GRE gradient recalled echo H HCM HDL He HLA Hz

hydrogen/proton hypertrophic cardiomyopathy high density lipoprotein helium horizontal long axis Herz

ICD IMA IMH iNOS IR IV

implantable cardiac defibrillator internal mammary artery intramural hematoma inducible nitric oxide synthetase inversion recovery intravenous

LA LAD LCX LDL LGE LM LPA LV LVEF LVOT

left atrium/atrial left anterior descending coronary artery left circumflex coronary artery low density lipoprotein late gadolinium enhancement left main coronary artery left pulmonary artery left ventricle/ventricular left ventricular ejection fraction left ventricular outflow tract

MI MIP Mn MO MPA MPHRR MR MRA MRI MRS

myocardial infarction maximum intensity projection manganese microvascular obstruction main pulmonary artery maximum predicted heart rate response magnetic resonance magnetic resonance angiography magnetic resonance imaging magnetic resonance spectroscopy Cardiovascular Magnetic Resonance xxiii

COMMON ABBREVIATIONS USED IN THE TEXT

Common Abbreviations Used in the Text

COMMON ABBREVIATIONS USED IN THE TEXT

MTT MVO2

mean transit time myocardial oxygen consumption

NOS NSF NYHA

nitric oxide synthetase nephrogenic systemic fibrosis New York Heart Association

OCT

optical coherence tomography

P PA PCI PCr PDA PDE PDW PET PLA PSA PTCA PWV

phosphorus pulmonary artery percutaneous coronary intervention phosphocreatine patent ductus arteriosus phosphodiesters proton density weighted positron emission tomography parasternal long axis parasternal short axis percutaneous transluminal coronary intervention pulse wave velocity

QCA

quantitative cororonary angiography

RA RCA RCM RF ROC ROI RPA RPP RV RVEF RVOT

right atrium/atrial right coronary artery restrictive cardiomyopathy radiofrequency receiver operator characteristic region of interest right pulmonary artery rate pressure product right ventricle/ventricular right ventricular ejection fraction right ventricular outflow tract

xxiv Cardiovascular Magnetic Resonance

SENSE SMASH SNR SPAMM SPECT SPIO SSFP

SENSitivity Encoding SiMultaneous Acquisition of Spatial Harmonics signal to noise ratio SPAtial Modulation of Magnetization single photon emission tomography small particle iron oxide steady state free precession

T T1 T1w T2 T2w TD TE TEE TGA TI TMR TOF TR TTC TTE

Tesla longitudinal relaxation rate T1 weighted transverse relaxation rate T2 weighted delay time echo time transesophageal echocardiography transposition of the great arteries inversion time transmyocardial laser revascularization tetraology of Fallot repetition time 2,3,5 triphyltetrazolium chloride transthoracic echocardiography

USPIO

ultrasmall particle iron oxide

VA VENC VLA VNC VSD VSMC

ventriculoarterial velocity encoding range vertical long axis ventricular noncompaction ventricular septal defect vascular smooth muscle cells

Xe XFR

xenon X ray fluoroscopy

Basic Principles of Cardiovascular Magnetic Resonance Robert S. Balaban and Dana C. Peters

This introduction to the basic principles of cardiovascular magnetic resonance (CMR) describes the concepts of T1, T2, and T2* and image formation, and describes some common CMR pulse sequences and parameters. These basic ideas often require some time and rereading to fully appreciate, but are necessary to understand the fundamen tal properties that determine image properties. The human body is composed mostly of water and also a lot of fat. The water (H20) and fat contain many hydrogen atoms. Hydrogen atoms in turn are made up of a proton (1H, the hydrogen nucleus) and an electron. Hydrogen protons are in very high concentration in the body, roughly 100 molar. Because of this abundance, the nuclear magnetic resonance signal can be used to create a distribution map, or image, through magnetic resonance imaging (MRI). MRI depends on the detection of the intrinsic angular momentum, or spin, of protons that is a basic property of matter. Hydrogen protons have spin, and all nuclei with spin interact with magnetic fields. In the absence of a magnetic field, the hydrogen spins are randomly oriented (Fig. 1 1A). However, if placed in a large magnetic field (called B0), the water spins (i.e., the 1H nuclear magnets) partly align with this applied magnetic field, much like iron filings (see Fig. 1 1B), with larger magnetic fields caus ing greater alignment of the spins. In contrast to the iron filings analogy, however, the interaction of the angular momentum (spin) of the hydrogen protons with the B0 field results in a rotation of the spin angular momentum around the axis of the magnetic field (see Fig. 1 1C). This is why the hydrogen nuclei in MRI are also referred to as spins: they spin, or precess, around the B0 field. The frequency of precession (v) is an important fundamental property of the spins in a magnetic field and is defined by the Larmor equation: n ¼ g  B=ð2pÞ

(1)

where v is the precessional frequency in cycles/sec or Hertz (Hz), g is called the gyromagnetic ratio and is related to the mass and charge of the water proton, g/2p ¼ 42.58  106 Hz/Tesla (T; 1 Tesla ¼ 10,000 gauss) for 1H, and B is the applied magnetic field. The precessional frequency for water protons at 1.5 T (a common CMR field strength) is v ¼ 63.87  106 Hz, or roughly 64 MHz (about the frequency of an FM radio station and one of the reasons why the CMR environment must be shielded from FM radio waves).

Molarity of 1H can be estimated as approximately (2 moles hydrogen/mole H20) (1mole H20/18 g tissue). 1000 g/L (density of the body)  100 mole/L.

The Larmor equation (Equation 1) is the basis of CMR imag ing. To determine the location of different magnetic spins (or 1 H), the magnetic field, B, is made to vary linearly with posi tion using specialized coils of wire (called gradient coils) inside the magnet. This results in a precessional frequency (v) that is also a linear function of position in the scanner. By measuring the number of spins precessing at each frequency, a magnetic resonance (MR) image is created,1 as discussed later.

DETECTION OF THE MAGNETIC RESONANCE SIGNAL Alignment with the Main Magnetic Field How can the MRI signal from 1H spins be detected to image the sample? Remember, when a sample (i.e., a patient!) is placed in the main magnetic field (e.g., the bore of the 1.5 T magnet), the hydrogen spins align with that field (see Fig. 1 1B) and begin to precess at the Larmor frequency (see Fig. 1 1C). The direc tion of the main magnetic field B0 defines the Z axis direction, also called the longitudinal direction. Because the spins partially align with the B0 field, they create a net magnetic field along the Z axis (oriented parallel with B0). However, the spins pre cess, or rotate, around the Z axis in a random, incoherent fash ion, resulting in no net magnetization in the X Y plane (also called the transverse plane; see Fig. 1 1C). An MR signal is detected by placing a coil (loops of wire) near the sample to detect any spin precession in the X Y (transverse) plane. This coil is called a receiver coil because it receives signal. In CMR, the receiver coil is usually an array of four or more (as many as 32 to 128)2 coils placed around the chest. However, a receiver coil placed to detect the precession of spins in the X Y plane will not detect any signal because the spins are in dif ferent positions, or phase, thereby canceling each other (see Fig. 1 1C). Therefore, radiofrequency excitation is necessary.

Radiofrequency Excitation and Magnet Strength To create an MR signal, the water (1H) spins must be rotat ing in a coherent manner in the X Y plane. To accomplish this task, another (less powerful) magnetic field (called B1), Cardiovascular Magnetic Resonance 3

1 BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

CHAPTER 1

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

No net 1H magnetic field

Net 1H magnetic field

Random phase of individual spins with net 1H magnetic field

Z

Z

X X X Y Y Y B0 = 0

A

B0

B

B0

C

Figure 1 1 Orientation of nuclear spins. A, With no magnetic field, the spins are oriented randomly, producing no net magnetization. B, The spins orient with an applied magnetic field (B0), producing a net magnetic field aligned along B0. C, The magnetized spins rotate around the B0 field (Z axis) in a random way, resulting in no net magnetization in the X Y plane.

at 3.0 T3 has been demonstrated for coronary artery imag ing,4,5 late gadolinium enhancement, perfusion,6 and func tion.7 The excitation field (B1) has a strength of only a few gauss. In addition, the B1 field is only applied transiently (for milliseconds), long enough to temporarily deflect the spins into a plane perpendicular to B0. This effect on the net magnetic field is illustrated in Figure 1 2A. A B1 pulse that drives the magnetization completely into the transverse plane (X Y plane) is called a 90 pulse, referring to the angle the net 1H magnetization moves relative to the Z axis. However, in practice, the B1 pulse can be used to flip the magnetization by any angle from 0 to 180 . The process flips the spin magnetization from the longitudinal direction into the transverse plane. Once the B1 field is turned off,

with its strength oscillating at the Larmor frequency, is applied perpendicular to the main magnetic field (Fig. 1 2A). This process is called radiofrequency (RF) excita tion because it uses a B1 field rotating at a high (radio wave) frequency. The frequency of the perpendicular B1 field has to precisely match the Larmor frequency of the water 1H to rotate the spins in the presence of the stronger B0 field. This concept is analogous to exciting a piano string with a tuning fork. Only the string with the same resonant fre quency as the tuning fork will efficiently absorb the energy from the fork and resonate. The water (1H) spins will rotate around the B1 field into the X Y plane. The B0 field has a magnitude in the range of 1.0 to 7.0 T. For CMR, a B0 of 1.5 T is most commonly used; however, cardiac imaging

Now coherent spins rotate around Z (B0)

Z Final magnetization

B0

Z X

B0

Initial magnetization

Receiver coil

Initial magnetization Y

Pulsed B1 field oscillating at Larmor frequency 90° X

B1

Voltage

Final magnetization

A

Y RF excitation coil

Time

B

Free induction decay

Figure 1 2 Effects of a perpendicular B1 magnetic field. A, The B1 field oscillating at the Larmor frequency of the spins is absorbed by the spins and causes a rotation around the B1 field axis (Y axis). B, Once the B1 field is removed, the spins continue to rotate around the Z axis at the Larmor frequency, but the signal decays (in the X Y plane) as equilibrium is reestablished. The result is a coherent oscillating magnetic field that is detected with the coil as the free induction decay. The free induction decay is a decaying sinusoidal voltage signal with a frequency equal to the Larmor frequency of the spins. RF, radiofrequency. 4

Cardiovascular Magnetic Resonance

T2 Relaxation and Spin Phase The other form of relaxation, T2 relaxation, is that of the randomization of the phase of the spins. The phase indi cates the direction of the spin on the transverse X Y plane. The phase (f) of a spin in the transverse plane depends on its initial phase, f0, the precessional frequency, and the time, t, which it has spent in the transverse plane. f ¼ f0 þ 2pnt

Because precessional frequency depends on the magnetic field, which changes slightly with location, time, and even molecular environment, each spin on the transverse plane has a different frequency and thus phase, and this phase dif ference increases with time (Equation 3). Figure 1 3 defines the phase (f) of the spin magnetic field vector as the direction of the transverse magnetization relative to the X axis. The phase of a spin depends on its frequency and its history of frequencies. In Figure 1 3B, a phase dia gram is used to show the changing phase relationships for three spins in a sample that has different magnetic fields within it. The magnetic field strength increases from light to dark. The magnetic field strength always varies within the sample because of imperfections in the magnetic field and sometimes intentionally through application of mag netic field gradients. A phase diagram is a stroboscopic image because it is arbitrarily referenced to one spin’s vector, in this case, a spin in the center region for Figure 1 3B. This stroboscopic effect is analogous to taking a flash picture every time the rotating middle vector reaches the 0 position. Using this approach, the relative position of the spin vectors, or phase, can be easily seen. Immediately after a 90 pulse, all of the spins have the same phase; however, with time, the different spin frequencies cause their relative phase to

T1 Relaxation T1 relaxation, also known as spin lattice relaxation, is the release of energy to the environment, or lattice, that results in the reestablishment of the magnetization along the Z axis. Thus, T1 is the time constant by which the longitu dinal (or Z) magnetization relaxes to its equilibrium value, M0. After a 90 RF pulse, the Z magnetization (Mz[t]) is initially 0, but regrows with a relaxation time T1 to its equi librium value (M0). MzðtÞ ¼ M0 ð1  e

t=T1

Þ

(3)

(2)



where t is time after the 90 pulse. Each tissue type has a unique T1. While T1 measurements differ slightly, a recent report found that for myocardium, the T1 is 1100 msec

Low Field Center Field High Field

X-Y plane phase diagram Y 90°

L

L C

X

Y φ

φ 180°

X

C

H

0° Immediately after 90⬚ pulse, t = 0.

Z

A

Spins out of phase

270°

B

Time t later

Still later

Time

Figure 1 3 Phase diagram. A, A phase diagram is a view of the magnetization vectors projected into the X Y plane. B, The magnetic field of a sample varies as shown. Spins located in the low field (L), central field (C), and high field (H) regions are followed in time after a 90 pulse using the phase diagram. The phase diagram presents the phases of the spins relative to C, which is always shown with 0 phase. Note that the L vector lags behind the C vector, whereas the H vector is ahead because it has a higher phase velocity. This difference is caused by the higher frequency of the spins in the higher magnetic field. This process is called dephasing. Cardiovascular Magnetic Resonance 5

1 BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

and 1200 msec at 1.5 T and 3.0 T, respectively, and for blood, the T1 is 1600 msec at both 1.5 T and 3.0 T.8

the only field causing a rotation is B0. Now the net magne tization of the water proton spins is in the transverse plane and coherent for detection. As the spins rotate around the B0 field axis and return to the B0 orientation, an oscillating X Y magnetic field can be detected with the receiver coil poised as shown in Figure 1 2B. Figure 1 2B shows the signal detected by the receiver coil after placing the spins in the main magnetic field and then flipping them into the transverse (X Y) plane. The sig nal oscillates at the Larmor frequency, decays in amplitude with time, and is called a free induction decay (FID). The decay in net magnitude in the transverse plane is known as relaxation. The decay occurs because the spins, having absorbed energy from the transient B1 field, which is now turned “off,” now return to their original state in equilib rium with the B0 field, with spins partially aligned along the Z axis (see Fig. 1 1C). This occurs via two extremely important processes, called T1 and T2 relaxation. Nature abhors order and seeks to minimize the energy in the sys tem and return to equilibrium with the main magnetic field, B0, through T1 and T2 relaxation. T1 and T2 relaxa tion forms the foundation of MRI.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

decrease in the net transverse magnetization. Some of this signal decay can be reversed because some of it is caused by static “off resonance.” T2* is an important process in the heart because both the lung cavity and the deoxygen ated blood in the right side of the heart contribute to B0 inhomogeneity around the heart. The combined effect of all dephasing processes, including the molecular spin spin interactions (T2), and static off resonance, is called T2*. The myocardial T2 relaxation rate is approximately 50 msec at 1.5 T,9,10 whereas myocardial T2* is approximately 30 msec at 1.5 T.11 T2* is always less than T2 for all tissues because it includes the effects of T2 relaxation and off resonance effects. Later we will describe how to minimize or use these effects to make in vivo measurements.

change. The higher B0 field spins (dashed arrows), are advancing their phase faster than central B0 field spins (solid arrows), whereas the spins (dotted arrows) at low field are lagging behind both. By watching the time devel opment of this process, one can appreciate that the phase and the frequency are related (Equation 3). High frequency spins change phase rapidly, whereas slower frequency spins change phase more slowly. However, the initial starting points of each of these signals are independent of frequency. T2 relaxation can be understood as the time for the spins to dephase. After the initial excitation using a B1 pulse, all of the spins on the transverse plane have identical phase, but then they begin to dephase quickly. This dephas ing is described by a time constant, T2. The dephasing causes the bulk transverse magnetization to decay and approach 0 because of incoherent phase. Because the magnetization decays, the transverse signal, S(t), also decays in a simple exponential way, with a time constant of T2: SðtÞ ¼ S0 e

t=T2

Spin Echo Imaging In tissue, T2* causes very rapid magnetic relaxation of water protons, compared with T2 relaxation. Thus, the decay rate of the FID is actually a measure of T2*. Because T2* is a more rapid process, it limits the time that we can detect the MR signal. It is possible to circumvent and reverse some of the T2* dephasing because some of it is caused by fixed B0 field inhomogeneity. This reversal is accomplished using a B1 field generated echo, outlined in Figure 1 4. After the 90 pulse, dephasing occurs. Another B1 field is applied to rotate the magnetization 180 around the X axis. This 180 flip is achieved by applying the B1 field for twice as long or with twice the power as the 90 pulse. This 180 flip causes the spins to rotate into the opposite sector of the transverse plane. On this side of the plane, the spins now drift together (“rephase”), taking the same amount of time to rephase that they were allowed to dephase. This is analogous to two people standing back to back in a field and then walking apart (90 pulse) at some fixed rate (but different rates for each person). At some later time, they reverse (180 pulse) their direction and now walk toward each other, each at the same “indi vidual” pace. The time it takes them to reach each other

(4)

T2 relaxation is also called spin spin, or transverse, relaxa tion. It is called spin spin relaxation because the mechanism of the relaxation process is through the interaction of spins in the sample with each other at or below the Larmor fre quency, making this process dependent on the microscopic motions within the sample. To understand many of the issues in CMR, we need to further understand T2. In general, two distinct processes contribute to the dephasing of spins in the heart. The first mechanism is the true spin spin interaction (T2) that is unavoidable, irreversible, and dependent on the molecular interactions within the sample. The second process is called T2* (“T2 star”) decay and includes T2 relaxation and also relaxation as a result of the inhomogeneity of the main magnetic field (B0) within the sample. As illustrated in Figure 1 3B, if the B0 field is not homogeneous through out the sample, then the frequency of the spins in different regions will vary (Equation 1). These spins are called off resonance spins. This results in a randomization of the spin phases as they rotate at different frequencies, and a

Dephasing 90°

X

Rephasing

Spin echo

180°

X Y

t=0

X Y

X Y

t=τ (before 180°)

t=τ (after 180°)

Y

t = 2τ

Time Figure 1 4 Effects of a 180 B1 pulse after a 90 pulse. The spins are first excited by a 90 pulse, as in Figure 1 2. After some time in which dephasing occurs (as in Fig. 1 3B), a 180 pulse is applied. The 180 pulse, by flipping the spins 180 around the X axis, exchanges the phase position of the slower and faster spins, so that the faster spins are now lagging behind the slower spins. The faster spins begin to catch up, and after a time, t, equal to the time of dephasing, the spins are all aligned in the same direction (have the same phase). This process is called rephasing, or refocusing, to form a spin echo. 6

Cardiovascular Magnetic Resonance

Cardiovascular Magnetic Resonance Imaging To create a CMR image, the MR signal intensity from the sample (i.e., patient) must be determined in three dimen sions (X, Y, and Z). Thus, for a single image, it is necessary to collect information on X, Y, and Z positions and signal amplitude. The MR signal shown in Figure 1 2B has fre quency, amplitude, and phase. These can be used to deter mine some spatial or amplitude information. However, there is not enough information to create a two or three

MAGNETIC RESONANCE IMAGING BLOCK DIAGRAM Time

Slice select

Z

Phase encode

Y

Refocus gradient/RF

Frequency encode, readout

X

Figure 1 5 General imaging scheme for collecting a simple cardiovascular magnetic resonance image. RF, radiofrequency.

dimensional image from a single FID. This is a major limi tation of MRI that results in a relatively slow image acquisi tion rate compared with many other modalities. The simplest imaging experiment can be divided into four stages, as shown in the block diagram in Figure 1 5: (1) slice selection, (2) phase encoding, (3) refocusing echo (for spin echo sequences), and (4) frequency encoding (readout). Each stage is used to encode the MR data with information on the position and amplitude of the water proton (1H) signal. By convention, Z position information is encoded with the slice selection step, X position infor mation is determined in the frequency encoding/readout step, and Y position information is obtained with the phase encoding step. In practice, the frequency encoding (X), phase encoding (Y), and slice selection (Z) directions are rotated in any appropriate direction. MR is tomo graphic and can create an image of the body along any plane. Indeed, oblique imaging, slicing through a tissue at 45 or any other angle, is possible and common. This property is almost a unique attribute of MRI. For example, computed tomography can provide images in any plane only after reformatting, but always collects images in the axial plane.

Gradients Spatial encoding in MRI is performed using “gradients”: slice select gradients, phase encoding gradients, and fre quency encoding gradients. Gradient coils are special coils within the magnet that modify the main magnetic field (Bz), causing it to change slightly in space: Bz ðx; y; zÞ ¼ B0 þ Gx  x þ Gy  y þ Gz  z

Table 1-1 T1 and T2 Values at 1.5 T Tissue

T1 (msec)

T2 (msec)

Myocardium Arterial blood Fat Skeletal muscle Lung

1100 1600 260 880 820

50 250 110 45 140

(5)

where Gx, Gy, and Gz are called gradient strengths and x, y, and z are the spatial coordinates within the magnet. Using gradients, the main magnetic field, with which the spins align and around which they precess, varies with spatial position inside the scanner. Remember, this magnetic field points in the Z direction, and the gradients do not change this direction, but they do increase or decrease its strength spatially. These spatially varying magnetic fields are called Cardiovascular Magnetic Resonance 7

1 BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

will be the same amount of time as they walked apart. This is the nature of the echo that effectively recaptures all of the coherent transverse magnetization that had been destroyed by the B0 field inhomogeneity in the sample. Note that the echo has both a rephasing and a dephasing period. The spins continue to dephase after reaching the coherent echo as a result of ongoing field inhomogeneity, just as the people in the field would walk past each other and continue to “dephase.” A subsequent 180 B1 pulse could be used to refocus these spins again and again. This method of contin ually refocusing the magnetization with 180 pulses is called fast spin echo (or turbo spin echo), and is used in cardiac imag ing. The number of repeated 180 pulses used to refocus the spins is limited because a 180 pulse will not recover the T2 relaxation processes occurring on the molecular level. However, the magnetization dephases according to T2 relax ation, which is much slower than T2* relaxation. These basic relaxation processes, T1, T2, and T2*, are key in generating image contrast in MRI as well as deter mining the optimal image sequence for gathering informa tion on cardiovascular anatomy, function, and physiology. Table 1 1 provides values of T1 and T2 for some important tissues. What is important is that T1, T2, and T2* vary for different tissues, thereby providing “contrast” in a CMR image. For example, myocardial relaxation properties change with edema, which is present in different disease states.12,13 Overall, the T1 and T2 values are more prolonged with increasing water content, as with edema. Myocardial remodeling, or myocyte replacement with con nective tissue, also changes the water relaxation properties because the nature of the macromolecules in contact with water is critical for these relaxation processes. Finally, most exogenous, intravenously injected CMR contrast agents act by shortening either T114-16 or T2*17,18 of the water spins. By appropriately modifying the imaging sequences, these changes in relaxation properties caused by pathology or exogenous contrast agent can be highlighted in the MRI of the heart. How this is accomplished will be described below.

fðx; y; z; tÞ ¼ g  Bz ðx; y; zÞ  t ¼ g  ðB0 þ Gx  x þ Gy  y þ Gz  zÞ  t

(7)

Selective Excitation: Position in Z During the slice select stage, RF excitation is performed. One of the spatial coordinates (Z) is determined by only exciting (with the B1) Z magnetization in a selected slice of the sample. The slice selection process is illustrated in Figure 1 6. A linear magnetic field gradient, Gz, is applied to the sample along the slice direction (the Z axis). The magnetic field gradient causes the spins along this axis to have slightly different frequencies, as defined by the Lar mor equation: vðzÞ ¼ ðB0 þ Gz  zÞ  ðg=2pÞ

Gradient in Z

Excited slice

si

tio

n

+1000 –1000

po

Frequency or B field

–4 +4 Z position

X

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

gradients because they create a linearly spatially varying magnetic field (magnetic field gradient). Of course, because the magnetic field varies with position, the precessional fre quency n(x,y,z), and the phase f(x,y,z,t) of the spins also vary with position (Equations 1 and 3). g g n¼ Bz ðx; y; zÞ ¼ ðB0 þ Gx  x þ Gy  y þ Gz  zÞ (6) 2p 2p

Figure 1 6 Slice selection gradient. A slice selection gradient is applied on the Z axis, causing the B field to vary from less than to greater than B0. This leads to a linear distribution of frequencies as a function of position along the gradient (Z direction). A frequency selective (e.g.,  1000 Hz) B1 pulse is used to excite only spins within a frequency band, resulting in selection of a slice (e.g.,  4 mm) that is placed in the transverse (X Y) plane.

The B1 field is used to excite the spins, but only within a range of precessional frequencies. If only a selected band of frequencies is excited (e.g., 1000 to 1000 Hz), then a slice of spins with precessional frequencies in that range (e.g., an 8 mm thick slice) is put in the transverse plane, not affecting the rest of the sample. The “slice thickness” (here, 8 mm) can be freely chosen by adjusting the Z gradient. Typical slices range from thin (e.g., 0.5 mm) to thick (e.g., 10.0 cm) slabs. The slice position can also be freely chosen. Furthermore, if no Z gradient is applied during the slice selection, this is equivalent to an infinite slab thickness, so that all of the spins in the body will be tipped into the transverse plane. This is called a nonselective RF pulse, and is commonly used in CMR for preparation pulses (see Figs. 1 17 and 1 18). How can the B1 field be used to excite only spins preces sing within a range of frequencies? To excite a bandwidth of frequencies, the Z gradient is turned on, and the B1 strength is modulated in time: B1ðtÞ ¼ B10  sinðp  t=TÞ=ðp  t=TÞ

This shape, sin(t)/t, is called a “sinc” function, and is plotted in Figure 1 7. This B1(t) excites a well defined band of frequencies of width, Dn ¼ 1/T. This is shown in Figure 1 7. Here, a Fourier transform is used to show how the spins interpret the sin(t)/t pulse in terms of frequency. Modifications to the sinc pulse are possible to optimize the performance and slice definition of this approach.19,20 The slice selection process is summarized in the imaging “pulse sequence diagram” shown in Figure 1 8. A pulse sequence diagram presents all of the gradients, the B1 field excitation, and the data collection event on a timeline. In Figure 1 8, the slice selection gradient is first increased to a constant level by a ramp. The gradient is maintained while the frequency selective (i.e., slice selective) B1 field is applied; note its sinc shape. Subsequently, the slice gradi ent is ramped down and the rephasing gradient is applied to remove the phase introduced by the slice select gradient. After the slice select excitation, only a slice of the sample has been placed in the transverse plane and all of the spins are in phase. The rephasing gradient shown in Figure 1 8 is performed so that all of the spins are “in phase” after the slice selection process. During the 90 pulse, the spins will dephase, depend ing on where they are in the slice (similar to Fig. 1 3B), because a magnetic field gradient is applied. Thus, another magnetic field gradient of opposite magnitude and half the

FREQUENCY SELECTIVE B1 PULSE Sinc function

B1 amplitude FFT

Voltage T Time

8

Cardiovascular Magnetic Resonance

(8)

1/T Frequency

Figure 1 7 The slice selective B1 pulse. The shape of the slice selective B1 pulse in time is sin (t)/t. This is called a sinc function. It excites a distinct band of frequencies because the Fourier transform of the sinc B1 field is a square wave in frequency space. The excitation of a band of frequencies by the B1, used in conjunction with the slice selective gradient (Fig. 1 6), causes a slice of spins to be placed on the transverse plane. FFT, fast Fourier transform.

PULSE SEQUENCE SHOWING ONLY SLICE-SELECTION Sinc(t) RF excitation Flat top Ramp

Equal area Z slice select Rephasing gradient

Slice select gradient

Y phase encode

X readout gradient

Signal Time

direction: vðxÞ ¼ ðB0 þ Gx  xÞðg=2pÞ. This is illustrated in Figure 1 9 for a sphere. The sphere is placed in a main magnetic field, with a magnetic field gradient in the X posi tion. The dashed lines show how the frequency of spins is varying across the sphere. The frequency encoding data provide a signal from which the number of spins at each frequency (i.e., each location) can be determined, using a Fourier transform, which converts time oscillating data into its frequency (or location) components. Using a frequency encoding gradient, the MR signal provides a measure

1 B field

area of the slice selection gradient must be applied to reverse the dephasing caused by the slice selection process (shaded regions in Fig. 1 8). The idea of gradient “area” is very impor tant. The gradient “area” is the product of the gradient ampli tude, G, and the duration, Dt, and is proportional to how much phase the spins have accumulated (see Equation 7) because of the gradient. The gradient area to rephase the spins is half the slice select area, because the spins reach the trans verse plane exactly coincident with the peak of the RF sinc pulse. By applying the gradient in the opposite direction, the spins are forced back to the same phase as they had before application of the slice select gradient. This small gradient is called a rephasing gradient. In summary, to selectively excite a slice, a linear mag netic field gradient, Gz, is placed along the axis to be sliced. A sinc B1 field pulse is applied, and only the predefined slice within the sample will be placed onto the transverse plane for further modification to create an image.

0

–1

X-axis

Although the Z position is now known for the signal in the FID through the placing of a slice of the sample into the transverse plane, the X and Y positions remain unknown. Recall that to eliminate the effects of T2* on the signal, a spin echo is created by applying a 180 pulse after the 90 slice select pulse. This results in the formation of a spin echo at a time equal to the time interval between the 90 slice select and the 180 pulse (see Fig. 1 4). If a gradient Gx is applied at this time in the X direction (called the fre quency encoding direction in MRI), recall that the frequency of the spins will reflect the locations of the spins in the X

# of spins

Frequency Encoding: Position in X –1

0

1

Relative frequency Figure 1 9 Frequency encoding of position. A frequency encoding (or readout) gradient is applied on the X axis, causing the B field to vary from less than to greater than the main magnetic field. The signal from the spins during the application of this gradient is acquired. The signal provides a measurement of the number of spins precessing at each frequency, thereby indicating the number of spins at each X location. Cardiovascular Magnetic Resonance 9

1 BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Figure 1 8 Pulse sequence diagram. In a pulse sequence diagram, a timeline is shown for the B1 pulse, the gradients on each axis, and the signal received. This convention is used in all pulse sequence diagrams. Here, the slice selection process is shown, in which a Z gradient is used with a sinc B1 pulse. RF, radiofrequency.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

of the number of spins precessing at each frequency. Figure 1 9 plots the precessional frequency versus the number of spins at that frequency. This is equivalent to a plot of X position versus the number of spins at that posi tion. The position of these spins along the X axis is simply detected by determining the frequencies of the spins while a gradient is applied along this axis. Thus, the process is called frequency encoding, and the gradient played out dur ing this period is often termed the readout gradient, or the frequency encoding gradient. As a refinement, another X gradient, called a dephasing gradient, precedes the readout gradient. This dephaser’s area is half the readout gradient area and opposite in sign. The spins are dephased by the dephasing gradient, and then rephased by the readout gradient, causing the spins to be in phase at the center of the readout gradient (see Fig. 1 11). Without a dephasing gradient, the peak signal would occur at the time when the readout gradient was ramping up to the desired gradient amplitude; therefore, the peak signal would not be measured. When a dephaser precedes the readout gradient, the spins refocus to form an echo at the center of the readout gradient. This is what is called a gradient recalled echo (GRE) because the readout gradient is used to refocus the dephasing gradient applied before it. A very thoughtful reader may wonder why the gradients are always “ramped” up and down. The ramps reflect the reality that the magnetic field cannot be changed instantaneously; the change is constrained by the scanner hardware. With current hardware, the gradient coils can be “slewed” to reach their maximum value (4 to 6 gauss/ cm) within approximately 100 to 200 msec, but magnet hardware is always improving. The frequency encoding component is added to the pulse diagram (Fig. 1 10). First the dephasing gradient is applied with an area equal to half that of the readout gradi ent. The readout gradient is then ramped up in the oppo site direction and held constant for a time, and then the gradient is ramped back down to zero. This is the time dur ing which the signal is acquired (see Fig. 1 10). During this period, the signal is continuously sampled, providing Nx

Object with 3 spins

samples, where the spatial resolution (Dx) is related to Nx, by field of view (FOV)/Nx ¼ Dx. The FOV is the size of the sample (the patient) in the imaging field. Remember that this signal is localized in Z using slice selective RF excitation and localized in X using frequency encoding. The signal measured contains this X and Z information. However, it does not yet contain Y position information.

Phase Encoding: Position in Y Y spatial information is determined through a process called phase encoding. While the spins are still in the trans verse plane after the slice selection process, the spins can be phase encoded by transiently applying a magnetic field gradient along the axis chosen for phase encoding (Y axis by convention). This transient gradient imparts a phase proportional to its area, which also depends on the Y posi tion. However, many phase encoding gradients of 180° pulse Excitation Z slice select Y phase encode gradient X readout gradient

Equal area

Dephasing gradient

Echo Signal

Figure 1 10 Pulse sequence diagram, as in Figure 1 8, with the addition of the readout portion of the sequence.

No gradient

Gy Gy

Y Time

T

Time

T

C

C

Time

C x

B A

C Phase of spins

B

B A

B

A A

Figure 1 11 Phase encoding. A Y gradient is transiently applied for a time, T, using larger and larger gradient strengths (Gy). Three spins, A, B, and C, located at three different Y locations, respond differently to the phase encoding gradient. The gradient does not affect the spin at y 0 (spin B) because it still precesses at the Larmor frequency. However, the gradient causes spin A to precess more slowly and spin C to precess more quickly, resulting in phase differences proportional to the gradient strength and the time, t. These phase differences allow the locations of the spins to be determined. 10 Cardiovascular Magnetic Resonance

All of the steps are combined into a single acquisition scheme (Fig. 1 12). A separate echo is collected for each phase encoding step. The phase encoding step is most commonly applied immediately after the slice selection process, frequently at the same time that the dephasing gradient for readout occurs (as shown in Fig. 1 12). All of the frequency and phase encoded raw data are com bined to create k space (Fig. 1 13). The k space of Fig ure 1 13 represents the raw data collected at each phase encoding step. The phase and frequency encoding direc tions are labeled. MR imaging can be understood as

TR and TE in a Spin Echo Sequence TR 90º

180º

180º (NS)

90º

Excitation Slice Select Phase encoding

phase encoding strength changes

Readout Signal

TE

TE

Figure 1 12 Pulse sequence diagram with phase encoding. The phase encoding gradient is often applied just before frequency encoding. A cardiovascular magnetic resonance image is acquired by repetition of the pulse sequence multiple times, as shown here, where the 90 gradient, the 180 gradient, and all gradients are shown repeated. The repetition time (TR) and echo time (TE) are defined. During each repeated pulse sequence, the phase encoding strength changes as shown. NS, nonselective.

Raw data collected: k-space

Image

1

Phase encodings (Ky)

Figure 1 13 The frequency encoding data that are collected during each phase encoding step are known as the k space signal. The k space signal does not look like an image, but is transformed into an image with Fourier transform (FT).

Raw k-Space Data and the Fast Fourier Transform

FT

Y

0

–1 Frequency encodings (Kx)

X

Cardiovascular Magnetic Resonance 11

1 BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

progressively increasing areas are required to localize a spin in the Y direction, depending on the object size (called an FOV) and the desired spatial resolution (Dy). The number of phase encodings (Ny) is given by the relationship FOV/ Ny ¼ Dy. Recall that to determine the position of a spin within the sample, a gradient is applied and the resulting frequency reports its position. Frequency is simply a measure of how fast the phase is changing in time (Equation 3). Phase encoding is a method for measuring the locations of the spins by measuring their response to different strengths of Y gradient and deducing their frequency (and therefore location) from these measurements. Unlike frequency encoding, which provides the positions in X with a single measurement, phase encoding is performed slowly, over Ny measurements. In successive measurements, the ampli tude of the phase encoding gradient is progressively increased and the signal is measured. Each measurement is called a repetition, and the time to acquire a single mea surement is called the repetition time (TR). Figure 1 11 describes the phase encoding technique. Consider three spins (A, B, and C), located along Y at three positions. The application of a gradient, Gy, for a short time, t, results in a location dependent phase (see Equation 3) for the spins. The first phase encoding has zero ampli tude and imparts no phase. The signal measured is the bulk signal. In the next phase encoding step, a small gradi ent is used and the phase difference among the three spins is small. The net signal measured is less than before because the spins are slightly dephased. A larger phase encoding gradient causes even more phase dispersion and less signal. After the signal is collected after multiple phase encodings, the signals can be used to determine Y loca tions. It is important to realize that for each phase encod ing, an entire frequency encoding must be collected. When the readout signal is finally collected, each of the spins’ phases is influenced by its position in Y. In this way, X and Y spatial information is encoded together. Thus, this slow phase encoding process provides the last piece of information, Y axis position, needed to create the simplest MR image.

Pulse Sequences and Contrast The sequence depicted in Figure 1 12 is a spin echo sequence. Even in this simple sequence, several factors of its design will change the contrast of the MR image, based on T1 and T2 relaxation. Because a 180 refocusing pulse is used in this sequence, the total time that the spins stay in the transverse plane determines the amount of T2 relaxation that will occur (Equation 3). This time is called the time to echo, or echo time (TE) and is measured from the center of the slice select “sinc” pulse to the center of the refocused echo during readout or data acquisition (see Fig. 1 12). Generally, the longer the TE, the more T2 contrast or T2 weighting is generated in the image (Equation 4).

Another sequence timing parameter that influences the signal amplitude is based on T1 relaxation. For a 90 exci tation pulse used in spin echo imaging, one must wait approximately five times the T1 value of the tissue to per mit the spins to completely regrow to their original Z mag netization (longitudinal magnetization) before applying another pulse to collect the entire MR signal available. If a shorter time is used between slice selective pulses, the spins’ Z magnetization will not fully regrow between pulses, and this leads to a reduction in the MR signal. This reduction in signal is dependent on the T1 of the sample. The longer the T1, the greater the reduction in signal, because less of the magnetization can recover. The time between each slice excitation pulse is the critical factor in this sequence and is called the time to repetition, or repetition time (TR; see Fig. 1 12). As an example of the importance of TE, the MR water signal from normal (T2 80 msec) and acutely infarcted (T2 100 msec) regions of the myocardium is shown in Figure 1 14A. Note that the signal of both tissues decreases with increasing TE, but the infarcted tissue sig nal decays more slowly because of its longer T2. This results in a contrast, or increased difference, in signal between the normal and the infarcted tissue at prolonged TE. On inspection of the difference between the signals of the two tissue curves, a TE of approximately 50 msec could be selected to optimize the contrast between these two tissue types. Thus, simply by adjusting the imaging parameters, fundamental information on the heart struc ture can be obtained. Conversely, pathology may be obscured if the imaging parameters are not ideal. The effect of TR on signal amplitude is shown in Fig ure 1 14B for a spin echo sequence. The effect of TR is illustrated for normal myocardium with a T1 of 0.8 sec onds and for a chronic infarct with a T1 of 1.5 seconds. Note that the shorter the T1, the more rapidly the pulses

TE AND TR EFFECTS ON MYOCARDIAL SIGNAL AMPLITUDE 100.00

100.00 Normal

80.00

80.00 Infarct Relative signal

Relative signal

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

“traveling” through k space. Each frequency encoding acquires a full line of k space (kx). Each phase encoding step moves up along the phase encoding axis (ky) to acquire another frequency encoding line. Once fully sam pled, this k space is converted into an image using a two dimensional fast Fourier transform. Ideally, the raw k space data contain enough information so that after the two dimensional fast Fourier transform an image of infi nite spatial resolution results. In reality, there are limits on the amount of data that can be collected. The resolu tion of the MRI image is directly related to the number of phase encoded echoes collected (Dy ¼ FOV/Ny); there fore, infinite spatial resolution would require infinite time to collect. In CMR, time is limited because of respiratory and cardiac motions, so the trade off between the spatial resolution required to observe the structure of interest and scan time (i.e., Ny) is critically important. Schemes to increase the efficiency of collecting the phase encoded data are discussed later.

60.00 Normal 40.00 Difference

20.00

60.00

40.00

Difference

20.00

0.00

0.00 0

A

Infarct

0.02

0.04

0.06

0.08

TE (sec)

0.10

0.12

0.14

0

B

1

2

3

4

5

6

7

TR (sec)

Figure 1 14 A, Effect of echo time (TE) and (B) effect of repetition time (TR) on cardiovascular magnetic resonance signal amplitude for normal and infarcted myocardium. Because the T1 and T2 of infarcted myocardium differ from those of normal myocardium, contrast between these tissues can be created. 12 Cardiovascular Magnetic Resonance

Fast Spin Echo Imaging The relatively long TR required for full T1 relaxation is the major reason why spin echo methods are slow to collect the phase encoded information required to create an MR image. To circumvent this problem, one highly useful approach is to use multiple 180 refocusing pulses and to collect many echoes during each slice selective excita tion. This method, called fast spin echo (FSE), or turbo spin echo (TSE), is an important technique in CMR.21 A phase encoding gradient is applied between each 180 pulse, thereby providing multiple phase encoded steps from a single slice selective pulse. For CMR, typically, 16 to 64 echoes are collected for each slice selective pulse, thereby reducing the time to collect a spin echo image by 16 to 64 times, respectively. Because large blocks of time are required to collect all of these echoes, this method is usu ally restricted to relatively motion free phases of the car diac cycle, such as diastole. The inherently high signal to noise ratio (SNR) of these FSE approaches supports very high spatial resolution images of the heart. In addi tion, true T2 contrast can be generated by acquiring the central phase encoding data (which largely controls the image contrast) at a specific time within the echo train. This time is called the effective TE. The longer the time interval between the initial 90 and the acquisition of cen tral k space (i.e., the longer the effective TE), the more T2 weighting will occur. The FSE technique is very com monly used in CMR to visualize anatomy and measure

the sizes of cardiac chambers (see Fig. 1 13, which shows an FSE image).

Double Inversion Recovery (“Black-Blood”) Fast Spin Echo Fast spin echo imaging of the heart is usually performed with a preparation sequence that nulls flowing blood, so that only the ventricular and atrial myocardium are visible. This preparation is performed before FSE acquisition (Fig. 1 15), using a double inversion recovery sequence, also known as a black blood preparation. This preparation consists of two inversion (180 ) pulses that immediately follow each other. The first inversion is a nonselective inversion pulse that inverts all magnetization. The second inversion is a slice selective inversion pulse that reinverts a slice of interest, which is centered on the imaging slice, but is slightly larger. This preparation is timed so that at the time of acquisition, all blood magnetization has regrown from fully inverted to 0. Any blood that was rein verted because it was within the slice of interest has now flowed out of the imaging slice. Any blood that has flowed into the slice will contribute no signal because it’s nulled. T2 weighted black blood FSE has recently been introduced for identifying edematous tissue associated with acute infarct22 and for identifying the region at risk.23

Gradient Echo Imaging For imaging the beating heart and other highly dynamic applications, an imaging sequence with a very short TR is useful. For a short TR, lower flip angles must be used because a long TR is necessary for full relaxation from a 90 slice selective pulse. Furthermore, for a very short TR, the 180 refocusing pulse is eliminated. The signal on the transverse plane will then decay with T2* instead of T2, but this is acceptable if the TE is very short (TE < 3 msec). As the TR is shortened, the flip angle must be reduced to provide the optimal SNR per unit time. The flip

Figure 1 15 Pulse sequence diagram for fast spin echo imaging. TE, echo time.

Fast Spin Echo Sequence 908

1808

1808

Repeated n times

1808 Excitation Slice Select

Phase encode

1808

Readout

1808 Signal

TE

TE

TE

Black-blood preparation

Cardiovascular Magnetic Resonance 13

1 BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

can be applied and still maintain the MR signal. Also apparent in Figure 1 14B is that by varying the TR, the contrast or difference between tissues can be altered. Thus, the TR can be used to vary the image contrast based on T1. Note that a TR of 1 second would be optimal in generating the largest difference between normal and infarcted tissue. This is convenient because a typical TR for spin echo and fast spin echo (discussed later) imaging is 1 R R interval approximately 1 second applying the B1 pulse every heartbeat.

a ¼ cos 1 ðe

TR=T1

Þ

fast gradient recalled echo (GRE, turbo FLASH, or turbo TFE).24 GRE refers to the refocusing of the frequency encoding gradient, as described earlier. The fast GRE pulse sequence is schematically illustrated in Figure 1 16A. Note that to keep the phase information intact, rewinding gradi ents are applied after each acquisition equal to and oppo site the phase encoding gradient. Furthermore, the read out gradient is also rewound so that on all axes, the spins are in phase after each TR. Usually, an extra gradient is applied at the end of the TR on the X or Z axis, and is called a crusher, killer, or spoiler gradient (see Fig. 1 16), which dephases any remaining transverse magnetization so that it does not contribute signal during the next phase encode step. Although fast GRE images have T2* dephas ing, this does not usually reduce image quality because a very short TE is achievable (1 to 3 msec), unless a serious source of magnetic field inhomogeneity (e.g., a metallic implant) is present. Furthermore, the GRE sequence can be used to quantify myocardial T2* (by imaging at multiple TE times), which may help detect diseases with myocardial iron deposition, such as thalassemia,25,26 with good reproducibility.27

(9)

This low flip angle RF pulse is also called an alpha pulse and is suitable for short TRs. A short TR is needed for dynamic imaging. For example, collecting 32 phase encoded lines in k space, each with a TR of 5 msec, requires 160 msec (5 msec  32 phase encoded lines). This time is sufficiently short that 32 TRs can be acquired in the quiescent diastolic period. Thus, data can be col lected rather rapidly using a small excitation pulse rather than the 90 pulse. This imaging method, with a short TR, low flip angle (alpha pulse), and no 180 refocusing pulse, is known as

Gradient Recalled Echo TR

a

a

a

a

a

Excitation Slice Select

Equal area

Phase

Crusher

Readout

Signal

Repeated n times

TE

A

Segmented Inversion Recovery

TI ECG 1808 (NS) 1.0 0.0

Mz

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

angle providing maximum signal at steady state (i.e., after many slice selective pulses) is determined by a trade off. A larger flip angle provides more spins on the transverse plane (more signal), but less Z magnetization is preserved. Therefore, the following echoes have less signal. A smaller flip angle provides less signal on the transverse plane, but more Z magnetization is preserved for subsequent echoes. The maximum signal at steady state is achieved using the Ernst angle for a given TR and T1:

1808 (NS)

GRE

GRE

Fibrosis Myo

Blood

−0.0 1808 (NS)

B 14 Cardiovascular Magnetic Resonance

GRE Repeat n times to obtain single image

Figure 1 16 A, Pulse sequence diagrams for gradient recalled echo (GRE) imaging showing multiple repetition times (TR). B, Inversion recovery sequence, which uses the GRE sequence, acquiring multiple phase encoding lines in each R R interval. Before each acquisition, a nonselective 180 B1 pulse is used to generate contrast between tissues with different T1 values. The inversion time (TI) is the time between the 180 B1 pulse and data acquisition. The Z magnetizations for blood, myocardium (Myo), and infarct (labeled) are plotted versus time within the sequence. Because the T1 of infarcted myocardium in the equilibrium phase of a contrast injection is shorter than that of normal myocardium, contrast can be generated between these tissues by judicious choice of TI. ECG, electrocardiogram; NS, nonselective; TE, echo time.

MzðtÞ ¼ M0 ð1  e

Þ  MzSS e

t=T1

(10)

SS

The fast GRE pulse sequence is also used for contrast enhanced MR angiography, which is the imaging of arteries and veins during the first pass of an exogenous contrast agent.28,29 During the first pass of a gadolinium contrast agent, blood T1 is reduced to approximately 30 msec. The Ernst angle equation (Equation 9) identifies an optimal angle of approximately 30 for a TR of 5 msec. The acquisition of three dimensional fast GRE provides a high quality image of vessels (see Chapters 32 35). How ever, without contrast agent, the SNR of fast GRE imaging can be low because the time between TRs, during which the longitudinal magnetization can regrow, is short. This limitation is partly overcome using balanced steady state free precession (SSFP).

Inversion Recovery Fast Gradient Recalled Echo: Late Gadolinium Enhancement One important application for fast GRE is detection of myocardial scar or infarction30 through injection of a gadolinium contrast agent, which slowly accumulates in regions of scar. It is detected as a hyperenhanced (“bright”) signal when imaged 10 and 20 minutes after injection.31 The T1 of normal myocardium, blood, and scarred (infarcted) myocardium is roughly 380 msec, 300 msec, and 270 msec, respectively, at 20 minutes postinjection of 0.2 mmol/kg gadolinium (a typical dose and delay time).14,15,32 A T1 weighted sequence, called an inversion recovery sequence, is used to create contrast between these tissues, which have slightly different T1 values. Inversion recovery uses a non slice selective 180 pulse, and a delay before imaging is called the time to inversion, or inversion time (TI). Then the raw data are collected using a GRE segmented k space approach (see Fig. 1 16B).33 Many

Figure 1 17 Balanced steady state free precession imaging (SSFP). The sequence is identical to gradient recalled echo except that the flip angle for SSFP is high and its sign is alternated in each repetition time (TR). Also, the gradients in the SSFP sequence are fully rephased. TE, echo time.

t=T1

where Mz is the steady state Z magnetization. Figure 1 16B shows the Z magnetization for the inversion recovery sequence and the T1 typical of blood, myocardium, and infarct. An image acquired at a time when the magnetization from normal myocardium has regrown to zero (the optimum TI) provides contrast in which the scar appears bright and the myocardium dark (see Fig. 1 16B).

Balanced Steady-State Free Precession Directly related to the GRE sequence is the balanced SSFP method. Today, almost all CMR imaging of ventricular function at 1.5 T employs this method, and it also is used for anatomic localization, valve visualization, and coronary artery imaging.34 This very old technique35 was more recently revived36,37 because better scanner hardware allows for the short TR (e.g., 2 to 4 msec at 1.5 T) required for balanced SSFP. Balanced SSFP uses a pulse sequence identical to that of GRE (Fig. 1 17), except for two distinc tions. First, after each TR, the spins are rephased on each gradient axis (X, Y, and Z) to zero phase. No crushers, kill ers, or spoilers are used. This keeps the spins completely in phase (except for dephasing as a result of magnetic field inhomogeneities) so that the transverse magnetization can be reused in the next TR. For this reason, the sequence is called balanced. Second, a large flip angle (e.g., 60 ) is used, flipping spins around the positive and negative Y axis, in alternate TRs. This flip angle alternation scheme reuses the remaining transverse magnetization and mixes it with longitudinal magnetization, for increased signal in each TR. Balanced SSFP provides high SNR of blood and a unique T2/T1 weighting.38 As stated earlier, balanced SSFP is highly sensitive to off resonance, requires a short TR and a large flip angle, and is challenging at 3 T, where off resonance creates artifacts7 and RF heating concerns (specific absorption rate) limit the flip angle.38a

BALANCED SSFP +60°

–60°

+60°

–60°

+60°

Excitation Slice select Phase encode Readout TE Signal TR

Repeated n times

Cardiovascular Magnetic Resonance 15

1 BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

heartbeats are required to collect data for a full image. After a 180 pulse (called an inversion pulse), Mz(t) recovers with a relaxation time, T1:

Three-Dimensional Fast Gradient Echo: Magnetic Resonance Angiography

Analogous to reducing scan time by replacing spin echo acquisition by fast spin echo acquisition, the GRE sequence can be accelerated by acquiring a train of gradient recalled echoes after a single RF pulse. This approach is called multi echo, or echo planar, imaging (EPI), and was among the first CMR imaging methods described.39 For anatomy with tissues with long T2* (e.g., the head), EPI can result in a complete image collection with a single slice select pulse and a train of gradient echoes (Fig. 1 18). This is the basis for the functional MRI (fMRI) technique that is widely used in brain mapping. To reduce time between echoes, the readout gradient is rapidly reversed with every phase encode step, so that MR signals are collected during both the positive and negative lobes of the readout gradi ent. This approach is challenging in the heart, where T2* can be short11 compared with the brain. Because of the acquisition of many phase encode steps after one RF excitation, the EPI method results in poor SNR, image trajectory errors that can be corrected,40-42 and image distortions because of off resonance. However, these limitations can be overcome by collecting only a few

(e.g., 3 to 9) phase encode steps per slice select RF excitation (instead of all phase encode steps) and repeating this process until all of the data have been collected. With state of the art gradient hardware, nine echoes can be collected per slice exci tation with a TR of 10 msec. Thus, imaging times on the order of 80 msec can be achieved in low resolution images (64  128 image resolution over a 15  32 cm FOV) without the magnetization ever spending more than 10 msec in the trans verse plane.43 This method is sometimes used for myocardial perfusion imaging, which requires complete acquisition of multiple slices in every heartbeat. A schematic of an electrocar diogram gated segmented EPI sequence for myocardial perfu sion is shown in Figure 1 18. Then data are rapidly acquired using a multi shot EPI method. An image from several slices is acquired every heartbeat during the first passage of gadolin ium contrast agent through the heart. In this sequence, a non selective 90 pulse is used to saturate all of the spins to provide T1 weighting (Equation 2) before data acquisition. Spiral imaging is similar to EPI, with more image data obtained after a single RF pulse, compared with conven tional imaging, by continuously sampling k space in a spiral pattern (one interleaf is seen in Fig. 1 19A). One or more interleaves (spiral arms) are collected to fully sample k space. Spiral image quality is affected by off resonance arti facts and trajectory errors, but it can provide high SNR as

Multi-shot EPI

ECG 90⬚ (NS)

Tdelay ~ 100–200 msec

20⬚

20⬚

Excitation

20⬚

Slice Select Phase encode

Repeated n times for N dynamics

Readout Signal

SPIRAL TRAJECTORY

Figure 1 19 A, Spiral imaging acquires k space by sampling along a spiraling trajectory. One or more interleaves, rotated with respect to each other, are required to sample k space fully. This schematic shows one interleaf of an Archimedean spiral. B, Radial imaging acquires multiple projections at equally spaced angles to sample k space, analogous to computed tomography. The sampling density of the radial trajectory is greater at the center of k space, so the lower spatial frequencies are always well sampled.

Ky-axis

0.5

0

0

–0.5

–0.5 –0.5

A

Figure 1 18 Segmented echo planar imaging (also called multi echo imaging; EPI), in which a portion of the frequency encodings are collected after each B1 pulse. Here, a typical pulse sequence for myocardial perfusion is shown, in which eight frequency encodings are collected after each small B1 pulse (here with a 20 flip angle). The acquisition is repeated until all of the data are acquired for an image. Before the multi shot EPI sequence, a nonselective 90 B1 pulse is used to generate contrast between tissues with different T1 values. The delay time (Tdelay) is the time between the 90 B1 pulse and the data acquisition. ECG, electrocardiogram; NS, nonselective.

RADIAL TRAJECTORY

0.5

Ky-axis

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Echo Planar Imaging, Spiral, and Radial

0 Kx-axis

16 Cardiovascular Magnetic Resonance

0.5

–0.5

B

0 Kx-axis

0.5

imaging was the first imaging method,1 and it has regained some popularity because of trajectory improvements and the recent developments in reconstruction of undersampled MR data51,52 using sparsity constraints. In conclusion, these basic principles of CMR are the foundation of the many techniques of CMR imaging of nor mal and pathologic cardiac anatomy and function.

References 1. Lauterbur P. Image formation by induced local interactions: examples employing nuclear magnetic resonance. Nature. 1973;242:190–191. 2. Niendorf T, Hardy CJ, Giaquinto RO, Gross P, Cline HE, Zhu Y, Kenwood G, Cohen S, Grant AK, Joshi S, Rofsky NM, Sodickson DK. Toward single breath-hold whole-heart coverage coronary MRA using highly accelerated parallel imaging with a 32-channel MR system. Magn Reson Med. 2006;56(1):167–176. 3. Gharib AM, Elagha A, Pettigrew RI. Cardiac magnetic resonance at high field: promises and problems. Curr Probl Diagn Radiol. 2008;37(2):49–56. 4. Stuber M, Botnar RM, Fischer SE, Lamerichs R, Smink J, Harvey P, Manning WJ. Preliminary report on in vivo coronary MRA at 3 Tesla in humans. Magn Reson Med. 2002;48(3):425–429. 5. Nezafat R, Stuber M, Ouwerkerk R, Gharib AM, Desai MY, Pettigrew RI. B1-insensitive T2 preparation for improved coronary magnetic resonance angiography at 3 T. Magn Reson Med. 2006;55 (4):858–864. 6. Gebker R, Jahnke C, Paetsch I, Kelle S, Schnackenburg B, Fleck E, Nagel E. Diagnostic performance of myocardial perfusion MR at 3 T in patients with coronary artery disease. Radiology. 2008;247(1):57–63. 7. Schar M, Kozerke S, Fischer SE, Boesiger P. Cardiac SSFP imaging at 3 Tesla. Magn Reson Med. 2004;51(4):799–806. 8. Sharma P, Socolow J, Patel S, Pettigrew RI, Oshinski JN. Effect of GdDTPA-BMA on blood and myocardial T1 at 1.5T and 3T in humans. J Magn Reson Imaging. 2006;23(3):323–330. 9. Botnar RM, Stuber M, Danias PG, Kissinger KV, Manning WJ. Improved coronary artery definition with T2-weighted, free-breathing, three-dimensional coronary MRA. Circulation. 1999;99 (24):3139–3148. 10. Foltz WD, Yang Y, Graham JJ, Detsky JS, Dick AJ, Wright GA. T2 fluctuations in ischemic and post-ischemic viable porcine myocardium in vivo. J Cardiovasc Magn Reson. 2006;8(3):469–474. 11. Reeder SB, Faranesh AZ, Boxerman JL, McVeigh ER. In vivo measurement of T2* and field inhomogeneity maps in the human heart at 1.5 T. Magn Reson Med. 1998;39(6):988–998. 12. Croisille P, Revel D, Saeed M. Contrast agents and cardiac MR imaging of myocardial ischemia: from bench to bedside. Eur Radiol. 2006;16 (9):1951–1963. 13. Scholz TD, Martins JB, Skorton DJ. NMR relaxation times in acute myocardial infarction: relative influence of changes in tissue water and fat content. Magn Reson Med. 1992;23(1):89–95. 14. Goldfarb JW, Mathew ST, Reichek N. Quantitative breath-hold monitoring of myocardial gadolinium enhancement using inversion recovery TrueFISP. Magn Reson Med. 2005;53(2):367–371. 15. Klein C, Nekolla SG, Balbach T, Schnackenburg B, Nagel E, Fleck E, Schwaiger M. The influence of myocardial blood flow and volume of distribution on late Gd-DTPA kinetics in ischemic heart failure. J Magn Reson Imaging. 2004;20(4):588–593. 16. Parker DL, Goodrich KC, Alexander AL, Buswell HR, Blatter DD, Tsuruda JS. Optimized visualization of vessels in contrast enhanced intracranial MR angiography. Magn Reson Med. 1998;40(6):873–882. 17. Mathur-De Vre R, Lemort M. Invited review: biophysical properties and clinical applications of magnetic resonance imaging contrast agents. Br J Radiol. 1995;68(807):225–247. 18. Weissleder R, Elizondo G, Wittenberg J, Rabito CA, Bengele HH, Josephson L. Ultrasmall superparamagnetic iron oxide: characterization of a new class of contrast agents for MR imaging. Radiology. 1990;175(2):489–493. 19. Conolly S, Glover G, Nishimura D, Macovski A. A reduced power selective adiabatic spin-echo pulse sequence. Magn Reson Med. 1991;18(1):28–38. 20. Meyer CH, Pauly JM, Macovski A, Nishimura DG. Simultaneous spatial and spectral selective excitation. Magn Reson Med. 1990;15 (2):287–304.

21. Hennig J, Nauerth A, Friedburg H. RARE imaging: a fast imaging method for clinical MR. Magn Reson Med. 1986;3(6):823–833. 22. Abdel-Aty H, Zagrosek A, Schulz-Menger J, Taylor AJ, Messroghli D, Kumar A, Gross M, Dietz R, Friedrich MG. Delayed enhancement and T2-weighted cardiovascular magnetic resonance imaging differentiate acute from chronic myocardial infarction. Circulation. 2004;109 (20):2411–2416. 23. Aletras AH, Tilak GS, Natanzon A, Hsu LY, Gonzalez FM, Hoyt RF, Jr., Arai AE. Retrospective determination of the area at risk for reperfused acute myocardial infarction with T2-weighted cardiac magnetic resonance imaging: histopathological and displacement encoding with stimulated echoes (DENSE) functional validations. Circulation. 2006;113 (15):1865–1870. 24. Haase A, Matthaei D, Hanicke W, Frahm J. Dynamic digital subtraction imaging using fast low-angle shot MR movie sequence. Radiology. 1986;160(2):537–541. 25. Anderson LJ, Holden S, Davis B, Prescott E, Charrier CC, Bunce NH, Firmin DN, Wonke B, Porter J, Walker JM, Pennell DJ. Cardiovascular T2-star (T2*) magnetic resonance for the early diagnosis of myocardial iron overload. Eur Heart J. 2001;22(23):2171–2179. 26. Westwood M, Anderson LJ, Firmin DN, Gatehouse PD, Charrier CC, Wonke B, Pennell DJ. A single breath-hold multiecho T2* cardiovascular magnetic resonance technique for diagnosis of myocardial iron overload. J Magn Reson Imaging. 2003;18(1):33–39. 27. He T, Kirk P, Firmin DN, Lam WM, Chu WC, Au WY, Chan GC, Tan RS, Ng I, Biceroglu S, Aydinok Y, Fogel MA, Cohen AR, Pennell DJ. Multi-center transferability of a breath-hold T2 technique for myocardial iron assessment. J Cardiovasc Magn Reson. 2008;10(1):11. 28. Prince MR, Yucel EK, Kaufman JA, Harrison DC, Geller SC. Dynamic gadolinium-enhanced three-dimensional abdominal MR arteriography. J Magn Reson Imaging. 1993;3(6):877–881. 29. Zhang H, Maki JH, Prince MR. 3D contrast-enhanced MR angiography. J Magn Reson Imaging. 2007;25(1):13–25. 30. Kim RJ, Wu E, Rafael A, Chen EL, Parker MA, Simonetti O, Klocke FJ, Bonow RO, Judd RM. The use of contrast-enhanced magnetic resonance imaging to identify reversible myocardial dysfunction. N Engl J Med. 2000;343(20):1445–1453. 31. Simonetti OP, Kim RJ, Fieno DS, Hillenbrand HB, Wu E, Bundy JM, Finn JP, Judd RM. An improved MR imaging technique for the visualization of myocardial infarction. Radiology. 2001;218 (1):215–223. 32. Weinmann H, Laniado M, Mutzel W. Pharmacokinetics of GdDTPA/ Dimeglumine after intravenous injection into healthy volunteers. Physiol Chem Phys Med NMR. 1984;16:167–172. 33. Edelman RR, Wallner B, Singer A, et al. Segmented turboFLASH: method for breath-hold MR imaging of the liver with flexible contrast. Radiology. 1990;177(2):515–521. 34. Weber OM, Martin AJ, Higgins CB. Whole-heart steady-state free precession coronary artery magnetic resonance angiography. Magn Reson Med. 2003;50(6):1223–1228. 35. Oppelt A, Graumann R, Barfuss H, Fischer H, Hart W, Shajor W. FISP–a new fast MRI sequence. Electromedica. 1986;54:15–18. 36. Duerk JL, Lewin JS, Wendt M, Petersilge C. Remember true FISP? A high SNR, near 1-second imaging method for T2-like contrast in interventional MRI at .2 T. J Magn Reson Imaging. 1998;8(1):203–208. 37. Heid O. True FISP cardiac fluoroscopy. In: Fifth Annual Proceedings of International Society of Magnetic Resonance in Medicine. Vancouver, BC, Canada; 1997. 38. Scheffler K, Lehnhardt S. Principles and applications of balanced SSFP techniques. Eur Radiol. 2003;13(11):2409–2418. 38a. Scha¨r M, Kozerke S, Fischer SE, Boesiger P. Cardiac SSFP imaging at 3 Tesla. Magn Reson Med. 2004;51:799–806. Cardiovascular Magnetic Resonance 17

1 BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

a result of high SNR efficiency. Spiral imaging has been shown to have advantages in coronary artery imaging and real time imaging.44-46 Radial imaging is another trajectory that acquires k space data as radial spokes (Fig. 1 19B shows 16 radial spokes, or projections), analogous to computed tomography. It has been applied to the heart, especially using undersampling of k space for fast imaging.47–50 Radial

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

39. Mansfield P, Pykett I. Biological and medical imaging by NMR. J Magn Res. 1978;29:355–373. 40. Peters DC, Derbyshire JA, McVeigh ER. Centering the projection reconstruction trajectory: reducing gradient delay errors. Magn Reson Med. 2003;50(1):1–6. 41. Reeder SB, Atalar EA, Faranesh AZ, McVeigh ER. Referenceless interleaved echo-planar imaging. Magn Reson Med. 1999;41(1):87–94. 42. Duyn JH, Yang Y, Frank JA, van der Veen JW. Simple correction method for k-space trajectory deviations in MRI. J Magn Reson. 1998;132(1):150–153. 43. Epstein FH, Wolff SD, Arai AE. Segmented k-space fast imaging using an echo-train readout. Magn Reson Med. 1999;41(3): 609–613. 44. Nayak KS, Pauly JM, Yang PC, Hu BS, Meyer CH, Nishimura DG. Real-time interactive coronary MRA. Magn Reson Med. 2001;46 (3):430–435. 45. Yang PC, Meyer CH, Terashima M, Kaji S, McConnell MV, Macovski A, Pauly JM, Nishimura DG, Hu BS. Spiral magnetic resonance coronary angiography with rapid real-time localization. J Am Coll Cardiol. 2003;41(7):1134–1141.

18 Cardiovascular Magnetic Resonance

46. Hardy CJ, Zhao L, Zong X, Saranathan M, Yucel EK. Coronary MR angiography: respiratory motion correction with BACSPIN. J Magn Reson Imaging. 2003;17(2):170–176. 47. Larson AC, White RD, Laub G, McVeigh ER, Li D, Simonetti OP. Selfgated cardiac cine MRI. Magn Reson Med. 2004;51(1):93–102. 48. Peters DC, Ennis DB, McVeigh ER. High-resolution MRI of cardiac function with projection reconstruction and steady-state free precession. Magn Reson Med. 2002;48(1):82–88. 49. Bi X, Park J, Larson AC, Zhang Q, Simonetti O, Li D. Contrastenhanced 4D radial coronary artery imaging at 3.0 T within a single breath-hold. Magn Reson Med. 2005;54(2):470–475. 50. Barger AV, Grist TM, Block WF, Mistretta CA. Single breath-hold 3D contrast-enhanced method for assessment of cardiac function. Magn Reson Med. 2000;44(6):821–824. 51. Lustig M, Donoho D, Pauly JM. Sparse MRI: the application of compressed sensing for rapid MR imaging. Magn Reson Med. 2007;58(6): 1182–1195. 52. Block KT, Uecker M, Frahm J. Undersampled radial MRI with multiple coils: iterative image reconstruction using a total variation constraint. Magn Reson Med. 2007;57(6):1086–1098.

Clinical Cardiovascular Magnetic Resonance Imaging Techniques Wolfgang G. Rehwald, Anja Wagner, Timothy S. E. Albert, Burkhard Sievers, Christopher K. Dyke, Michael D. Elliott, John D. Grizzard, Raymond J. Kim, and Robert M. Judd

It has now been more than 20 years since the first magnetic resonance (MR) images of the human heart were described. Although few would deny that the quality of cardiovascular MR (CMR) images has dramatically improved since that time, it has nevertheless been only in the last 5 to 10 years that large numbers of dedicated CMR clinical services have opened in the United States. Although organizations such as the Society for Cardiovascular Magnetic Resonance (SCMR; www.scmr .org) are working toward scan standardization, there is often variability among CMR centers as to what actually comprises a “routine clinical CMR.” In this setting, a chapter that attempts to describe all of the possible permutations of pulse sequences, scan protocols, and interpretation methods can quickly become unwieldy. This chapter focuses on the spe cific clinical CMR imaging techniques that are used routinely at the Duke Cardiovascular Magnetic Resonance Center (DCMRC). More detailed sequences and details are found in the respective chapters addressing each setting or pathology. The dedicated CMR clinical service at the DCMRC first opened in July 2002. Since that time, the authors have experimented with and evaluated many approaches to CMR and have directly observed their effects on overall clin ical volume. Clinical volume at the DCMRC has grown steadily since the service first opened and currently includes more than 3000 CMR studies per year (Fig. 2 1). For the last several years, the distribution of CMR tests performed has remained relatively constant (Fig. 2 2), but may vary greatly depending on the expertise and patient population at other CMR centers.1 Nearly 50% of our clinical volume includes pharmacologic stress testing.2 The DCMRC definition of a CMR stress test can be described briefly as a group of three individual scans: 1. Left ventricular (LV) cines (to examine systolic function) 2. Adenosine stress/rest perfusion (to examine coronary flow reserve) 3. Late gadolinium enhancement (LGE; to examine viabil ity/infarction) An additional 25% of the clinical volume is referred for via bility testing. The DCMRC definition of a CMR viability test can be described briefly as a group of two individual scans: 1. LV cines (to examine systolic function) 2. LGE (to examine viability/infarction) It is immediately apparent that the only difference between a CMR viability test and a CMR stress test is that the former does not include perfusion imaging. Together, these two tests account for nearly three fourths of annual procedures.

Through the years, the DCMRC philosophy about how to perform CMR has evolved into the concept of a CMR “exam menu” that not only has allowed streamlining of the clinical service but also has made it considerably easier to teach CMR to cardiology fellows and level 2 trainees. The underly ing idea of the exam menu (Fig. 2 3) is that most, if not all, routine clinical CMR procedures can be performed simply by combining one or more items from a predetermined menu and performing these scan protocols/pulse sequences in a sequential and consistent manner, analogous to the per formance of a transthoracic echocardiogram or physical examination. Thus, both the CMR stress test and the CMR viability test are simply combinations of items from the exam menu. Virtually every study includes two or more of the items listed in Figure 2 3. Perhaps more importantly, only rarely is it necessary to perform a scan not listed on this menu. Because of issues related to the development of nephrogenic systemic fibrosis, all patients should have a renal function assessment before the administration of gado linium contrast (see Chapter 6). For those with moderate or severely depressed renal function (estimated glomerular fil tration rate, eGFR < 30 mL/min/1.73 m2), alternatives to gadolinium contrast should be sought. Accordingly, the remainder of this chapter focuses on each of the items on the exam menu, with an emphasis on providing practical information not typically provided in other CMR textbooks. The conclusion provides typical examples of how the items on the exam menu can be com bined to answer common diagnostic questions in cardio logy, such as the detection of coronary artery disease and the evaluation of aortic disease. The sections within this chapter are presented in order of increasing technical complexity, whereas items in the exam menu (see Fig. 2 3) are ordered from most to least frequently used in clinical practice. For consistency in this chapter, a “test” is defined as one of the components of the pie chart seen in Figure 2 2 and a “scan” is defined as one of the items on the exam menu seen in Figure 2 3.

SCOUTS (“SCAN” ¼ “SCOUT”) The goal of CMR scout scanning is to establish the short and long axis views of the heart and to confirm the optimal position of the anterior and posterior elements of the tho racic coil. Because of patient to patient anatomic variation, Cardiovascular Magnetic Resonance 19

2 CLINICAL CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES

CHAPTER 2

700 CMR procedures (quarterly)

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Figure 2 1 Clinical volume of the Duke Cardiovascular Magnetic Resonance Center since its inception in 2002.

800

600 500 400 300 200 100 0 2002- 2002- 2003- 2003- 2003- 2003- 2004- 2004- 2004- 2004- 2005- 2005- 2005Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3

Congenital evaluation (7%)

CMR EXAM MENU (“SCANS”)

CMR angiography (20%) Stress testing (47%)

1 −− Scout 2 −− Left ventricular/right ventricular cine 3 −− Stress–rest myocardial perfusion 4 −− Late gadolinium enhancement 5 −− Morphology 6 −− Angiography 7 −− Flow/velocity Figure 2 3 The Duke Cardiovascular Magnetic Resonance Center “Exam Menu.”

Cardiac viability testing (26%) Figure 2 2 Overall makeup of the clinical volume. CMR, cardiovascular magnetic resonance.

both the short and long axis cardiac views lie at arbitrary angles with respect to scanner coordinates and are there fore referred to as “double oblique” planes. The first step in acquiring these double oblique views is to acquire images along the axes of the scanner (i.e., axial, sagittal, and coronal planes) passing through the thorax. Figure 2 4 shows typical examples of scout images. Once images along the scanner axes have been acquired, they can be used to prescribe a single oblique, for example, perpendicular to the axial image of Figure 2 4 going through the LVapex and midmitral annulus while parallel to the inter ventricular septum (dashed yellow line). From this view, one can then prescribe a double oblique, for example, oriented on the true short axis of the LV. Thus, from the perspective of the scanner operator, the goal of scouting is simply to acquire a series of images similar to that of Figure 2 4 and including short and long axis images of the LV. The pulse sequence used to scout is based on steady state free precession (SSFP). The underlying concept of SSFP was described in the mid 1980s,3 but only since the late 1990s has CMR scanner hardware been capable of achieving the SSFP magnetization state. The SSFP pulse 20 Cardiovascular Magnetic Resonance

sequence timing diagram is characterized by an elegant simplicity in which there is symmetry around the data acquisition window on all three gradient axes (Fig. 2 5). The axis labeled “slice” selects the slice to be imaged. The waveforms on the “read” axis create the MR signal as an echo (see “signal” axis). During the positive portion of the “read” waveform, the MR signal is digitally sampled. The “phase” encoding axis imposes a different phase on each echo that allows spatial encoding of the second dimension called “y” in Figure 2 6. In practice, the pri mary advantages of SSFP imaging are rapid acquisition speed and resultant images that show very high signal to noise ratio (SNR) and contrast to noise ratio (CNR) in the blood pool and surrounding myocardium. Accord ingly, SSFP is used not only for scouts but also for cine CMR imaging. Acquiring any CMR image involves filling the raw data space, referred to as “k space.” Figure 2 6 shows the pro cess of filling k space as a series of “lines” starting at the top and proceeding to the bottom. Each line is actually one echo (see enlarged box) acquired during the read event, for example, of the sequence in Figure 2 5. The echo runs in direction x. Direction y is the phase encoding direction. For an SSFP pulse sequence running on a typical modern CMR scanner, the time needed to acquire each k space line is approximately 3 msec. Thus, for 100 k space lines (typical for scouts), the total image acquisition time

Coronal

Sagittal

Figure 2 4 Scout images. Typical examples of axial, sagittal, and coronal scouts.

is approximately 300 msec (100  3 msec). This is fast enough to effectively freeze heart motion, provided that the image data are acquired during diastole when the ven tricles are relatively quiescent. Acquiring the image data in diastole requires the scan ner hardware to be “triggered” to the cardiac cycle. This is typically achieved by accurately detecting the R wave of the electrocardiogram (ECG), a process that was previously difficult because of the ECG distortion caused by the magnetohydrodynamic effect of pulsatile blood in the aorta. Vector ECG and other solutions have largely mini mized this issue.4 Assuming a heart rate of 60 bpm (R R duration of 1000 msec), the first “event” the scanner hard ware must play out is a delay time of 500 msec (to get to diastole), followed by 300 msec of image data acquisition. Figure 2 7 shows cardiac gating for two successive scout images. Typically, 27 scout images are acquired to define the thoracic contents, including 9 parallel images in each of the axial, coronal, and sagittal imaging planes.

MORPHOLOGY (“SCAN” ¼ “MORPHOLOGY”) Morphology scanning is used less often than cine CMR, but the technical aspects of morphology, particularly as it relates to cardiac gating, are only modestly more complex than for scouting, so they are discussed next. The “goal” of morphology scanning is essentially to cre ate a series of parallel slices that “bread loaf” the thoracic cavity to examine vascular anatomy. Any orientation can be acquired, but frequently only axial planes (or axial in addition to coronal and sagittal planes). It is often desirable to use both “black blood” imaging (Fig. 2 8) and “bright blood” imaging (Fig. 2 9). The CMR pulse sequences used for these are half Fourier single shot fast spin echo (HASTE) and SSFP, respectively. Figure 2 10A shows the pulse sequence timing diagram of the fast spin echo sequence, and Figure 2 10B shows its predecessor, the spin Cardiovascular Magnetic Resonance 21

2 CLINICAL CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES

Axial

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Symmetry +α

–α TR

RF

Slice

Read

Phase

Signal

Time

Figure 2 5 Steady state free precession (SSFP) pulse sequence timing diagram. The SSFP consists of rapid gradient echoes whose magnetization is preserved across multiple k space lines, resulting in images with a high signal to noise ratio despite a short scan time. RF, radiofrequency; TR, repetition time.

One k-space line Acquiring each line takes ~3 msec (SSFP). Thus, 100 lines requires ~300 msec

y

x

Figure 2 6 Filling of raw data space in Cartesian fashion. The raw data space for CMR is often referred to as “k space” and must be filled line by line to generate an image. Each line contains one echo. Each echo is acquired after application of a different phase encoding step (direction y), and thus, each one contains different information. K space is filled using, for example, the pulse sequence of Figure 2 5. Then the raw data undergo a two dimensional Fourier transformation, resulting in an image of, for example, Figure 2 4. SSFP, steady state free precession.

echo sequence. HASTE is a special variant of the fast spin echo sequence in which only part of k space is sampled to reduce acquisition time. The timing diagram for SSFP was described earlier (see Fig. 2 5). Typically, three stacks of HASTE and SSFP images are acquired, one for each orthogonal scanner plane (axial, coronal, and sagittal). Once the images are acquired, the scanner operator can then step through each series of image slices and 22 Cardiovascular Magnetic Resonance

understand the patient’s vascular anatomy and plan further scan planes. Cardiac gating for SSFP morphology images is essentially the same as that used for scout imaging: acquire the entire image in a single heartbeat during diastole (Fig. 2 11). For example, a set of 40 SSFP images for morphology would require 40 successive heartbeats. Cardiac gating for HASTE is essentially the same as for SSFP except that the images are typically acquired every other heartbeat (e.g., 80 total heartbeats are required to complete 40 images). The reason for this relates to the mechanism responsible for making the blood appear black in HASTE images. Morphology imag ing typically takes 3 to 5 minutes and is acquired without any breath holding. For example, the acquisition of 40 slices with both SSFP and HASTE would take 40 þ 80 ¼ 120 heartbeats, or approximately 2 minutes. Black blood imaging is often advantageous for the exam ination of morphology in that it allows one to clearly distin guish the inner portion of the vessel wall from the blood. In essence, black blood HASTE can be achieved by carefully combining CMR physics with the physiology of rapidly moving blood. The underlying reason why the blood appears black in HASTE images is not reflected in the timing diagram of Figure 2 10. It is related to the prepara tory radiofrequency (RF) pulses. The basic concept is seen in Figures 2 12 and 2 13. Soon after the R wave, a brief (10 msec) nonselective 180 RF pulse is applied that causes all of the magnetization in the body to flip upside down (“inversion” pulse). Immediately afterward, a second slice selective RF pulse is applied that causes only the magneti zation within the to be imaged slice to flip back to where it started (“re inversion” pulse). The net effect of these two RF pulses is that all of the protons outside of the slice have an inverted magnetization, whereas all protons within the to be imaged slice do not. At this point, the scanner simply waits several hundred milliseconds (see Fig. 2 13), and during this period, the heart contracts and pushes blood out of the slice (orange in Fig. 2 12). This blood is then replaced with fresh blood (green in Fig. 2 12) previ ously located outside the slice. The net effect is that, when the actual image acquisition stage begins in diastole, the fresh blood now in the slice “remembers” that it was inverted shortly after the R wave. Its T1 recovery curve crosses zero (see Fig. 2 13), consequently produces no sig nal, and thus appears black. In summary, black blood HASTE is made possible by a clever combination of a “memory” effect related to MR imaging physics (T1 recov ery) as well as the physiology of moving blood. Such images are helpful in detecting abnormal large vessel anat omy, especially when combined with bright blood SSFP images at the same location.

CONTRACTILE FUNCTION (“SCAN” ¼ “CINE”) Contractile function is a fundamental part of the CMR examination. Cine CMR is used for global and regional LV and right ventricular (RV) wall motion assessment. It is highly accurate and reproducible for ventricular volume, ejection fraction, and mass measurements.5–7 With the use of SSFP sequences, cine CMR has become widely

R

R

R

ECG

Delay time

Delay time All k-space lines (e.g., 100 lines) Scout 1

Trigger

All k-space lines (e.g., 100 lines) Scout 2

Trigger

Left

Head

Posterior

Right

Feet

Anterior

Sagittal

Axial

Coronal

Figure 2 8 The “goal” of “black blood” half Fourier single shot fast spin echo imaging. Shown is a series of parallel images across the entire thoracic cavity that can then be inspected to assess the vascular anatomy. Usually, sagittal, axial, and coronal stacks are acquired.

accepted as the noninvasive gold standard for contractile function assessment. It has been used as an end point for the evaluation of LV remodeling8–10 and as a reference method for other imaging techniques.11–18 The “goal” of cine CMR is to capture a movie of the beat ing heart to visualize its contractile function. Figure 2 14

shows a representative example of a midventricular short axis slice during eight different time points within the car diac cycle. Typically, 20 to 30 cine frames are acquired with 30 to 50 msec temporal resolution. Images are acquired with SSFP (see Fig. 2 5). The advantage of SSFP over other pulse sequences, such as the gradient recalled echo (GRE) Cardiovascular Magnetic Resonance 23

2 CLINICAL CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES

Figure 2 7 Cardiac triggering for scout imaging consists of a delay after the electrocardiographic (ECG) R wave and then rapid acquisition of the entire image (all k space lines) in diastole. Imaging within a single heartbeat is made possible by the intrinsic speed of the steady state free precession pulse sequence.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Left

Head

Posterior

Right

Feet

Anterior

Sagittal

Axial

Coronal

Figure 2 9 The “goal” of “bright blood” steady state free precession imaging. The same image planes used for half Fourier single shot fast spin echo are acquired so that the two forms of images can be compared.

sequence, is its high SNR, fast speed, and excellent blood/ myocardium CNR that greatly facilitates the identification of the endocardial border. Typically, a stack of multiple closely spaced (contiguous or with 1 to 2 mm gaps) short axis slices (6 to 8 mm thick) is acquired to provide full coverage of the left and right ventricles. Short axis views, perpendicular to the long axis views, can be planned on long axis scout images, as described earlier. In addition, cine images can be obtained in multiple long axis views, such as the two chamber, three chamber, or four chamber orientations. A key consideration for cine imaging is that current CMR scanner hardware is not adequate to acquire cine images with high spatial resolution and high temporal res olution during a single cardiac cycle in real time. For exam ple, for a cine with 30 frames, each with 96 lines, a total of 30  96 ¼ 2718 lines must be acquired. Considering that the acquisition time for one line is approximately 3 msec, the acquisition time to create a complete dataset of cine images would be 2718  3 msec ¼ 8154 msec, consider ably longer than the typical R R interval. Imaging speed could be reduced by acquiring fewer phase encoding lines, but lower spatial resolution or lower temporal resolution would result. For cine imaging, each heartbeat is assumed 24 Cardiovascular Magnetic Resonance

to be “identical” to a subsequent heartbeat such that image quality can be improved by the use of k space segmentation. Segmented k space data acquisition allows one to collect only part of each movie frame over 8 to 10 consecutive car diac cycles and then to combine the data to form a cine loop. The underlying concept is similar to that used in gated single photon emission computed tomography. Figure 2 15 shows the scheme used for CMR in which only a fraction of k space for any given movie frame is acquired during any single heartbeat. This fraction is referred to as a “segment” (e.g., 10 k space lines per segment in Fig. 2 15) and typically is adjusted by the scanner operator such that an adequate number of movie frames (generally 20 to 30) fit within the patient’s R R interval. Accordingly, the full k space data for any one movie frame actually consist of multiple segments acquired during successive cardiac cycles. The data are then automatically reassembled during image reconstruction to form a movie showing a single cardiac cycle. For example, the scheme in Figure 2 15 acquires 30 segments of 10 lines each in every cardiac cycle. Assuming that the scanner operator has chosen to obtain 90 k space lines for each movie frame, the scan will acquire data across 90/10 ¼ 9 heartbeats. For example, the

180°

180°

180°

RF Slice

1st

2nd

3rd echo

Read Phase

Fast spin echo sequence

A 90°

180°

90°

180°

RF Slice Read Phase Signal TE/2

TE/2 TE TR

B

TE Spin echo sequence

Figure 2 10 Pulse sequence timing diagram for (A) a fast spin echo sequence and (B) conventional spin echo. In essence, the fast spin echo consists of a series of spin echoes. Additional postprocessing steps are required to obtain the image. The spin echo sequence (B) is the predecessor of the fast spin echo sequence (A). The blood appears black in the resulting images because of the preparatory pulse described in the text and in Figures 2 12 and 2 13. RF, radiofrequency; TE, echo time; TR, repetition time.

k-space slice 1

MYOCARDIAL PERFUSION AT STRESS AND REST (“SCAN” ¼ “PERFUSION”)

k-space slice 2 Heartbeat 1 Heartbeat 2

R

Slice 1

R

Slice 2

R

ECG Delay time Trigger

Image slice 1

Delay time

Image slice 2

Trigger

Figure 2 11 Electrocardiogram (ECG) triggering with each image acquired in a single diastole, similar to scout imaging. To skip systole, a delay time is inserted after the R wave is detected.

green segment is always acquired immediately after the R wave and is used to reconstruct the first movie frame, the orange segment is acquired next and becomes the sec ond movie frame, and so on. In practice, cine imaging is performed during breath holding and takes 10 to 12 sec onds for each cardiac slice acquisition (e.g., one short axis cine). Cine imaging can be performed at rest and with pharmacologic stress with low dose dobutamine (viabil ity) or graded high dose dobutamine (stress).19,20 For these higher heart rate scans, high temporal resolution is especially important.

Encouraged by a number of clinical studies,2,21–23 adeno sine stress/rest myocardial perfusion CMR has gained clini cal attention. Perfusion CMR accurately diagnoses coronary artery disease with high sensitivity and specificity. Rest per fusion CMR, in combination with LGE, is important for distinguishing true perfusion defects on stress images from artifacts. The subject of artifacts is discussed elsewhere.2 In the current protocol, the rest myocardial perfusion scan is routinely performed in all patients who receive gadolinium contrast. The “goal” of myocardial perfusion scanning is to create a movie showing the wash in of contrast media (typically gadolinium based) with the blood during its initial pass through the myocardium (“first pass perfusion”). The CMR pulse sequence that is most commonly used is a GRE sequence. Figure 2 16 shows an example of an adenosine stress/rest perfusion scan in a patient with a stress induced perfusion defect. The blue slice is shown at three representative time points: (1) before contrast arrival; (2) at the time of contrast arrival in the right ventricle; and (3) shortly after contrast arrival in the LV. During adeno sine stress, perfusion defects appear as dark regions (i.e., hypoenhancement) surrounded by bright contrast enhanced normally perfused myocardium. In the corresponding rest perfusion images, perfusion to the hypoen hanced areas may be relatively preserved and thus appear Cardiovascular Magnetic Resonance 25

2 CLINICAL CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES

90°

Systole

R-wave (end-diastole)

Diastole

Nonselective inversion Slice-selective “re-inversion” Imaged slice Figure 2 12 Physiology of “black blood” fast spin echo imaging. Immediately after electrocardiogram trigger, a nonselective inversion (green) is applied, followed instantly by a slice selective re inversion (orange). The prepared slice (orange) changes shape and position during cardiac contraction, but returns to its original geometry during diastole, when the blood signal is about zero (leading to black blood in the image) and the heart has little motion. Nonblack re inverted blood (orange) is expelled from the slice during systole. During diastole, an image is obtained from a slice (blue) that is thinner than and lies inside the prepared slice (orange). ECG, electrocardiogram.

• HASTE = Half-Fourier Single-Shot Turbo Spin Echo • Typically used with a dark blood pulse.

ECG

All lines acquired for one complete image.

Noninverted myocardium Selective re-inversion

Nonselective inversion

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

ECG

90180 180 180 180 180 180 180

Time needed to null blood Time d

te nver

i wing Inflo

d bloo

Figure 2 13 Cardiac triggering for “black blood” fast spin echo imaging. Image acquisition starts when the magnetic relaxation curve of the inverted blood in the cavity is passing through zero (i.e., when the blood is black). The center lines of k space are acquired first (“centric reordering”). These lines contain information about image brightness and contrast. Hence, they are acquired when the blood is black, at the beginning of the acquisition train.

26 Cardiovascular Magnetic Resonance

1

2

3

4 Late diastole

5

6

7

8

Figure 2 14 The “goal” of cine CMR. A typical midventricular series shows short axis cine steady state free precession images at the level of the papillary muscles. The cine series often contains 20 to 30 images. Eight are shown here because of space constraints.

Figure 2 15 Cardiac gating for cine imaging. Raw data lines are acquired in segments over the course of six heartbeats. The segments are then sorted into the k spaces of the 30 movie frames. Multiple lines form one segment, multiple segments form one k space, and Fourier transformation of one k space yields one movie frame. ECG, electrocardiogram.

1

2

3

4

Heartbeat 6

5

ECG Trigger Acquired in segments:

1 2

30

1 2

30

1 2

30

1 2

30

1 2

30

1 2

30

30 segments acquired each heartbeat

Each segment contains 10 lines

10 lines

Ordered after cardiac phase: 1

2

normal, indicating normal resting perfusion. The reduced coro nary flow reserve is caused by hemodynamically significant cor onary artery stenosis. Perfusion imaging data are acquired continuously (with each R R interval) rather than in the segmented manner previously described for cine CMR. Each single image is acquired within 100 to 150 msec. Because of R R interval

30

Cardiac phase

time constraints, perfusion imaging is typically performed throughout the entire cardiac cycle. All k space lines for each of four to five short axis slice locations are acquired during each cardiac cycle (Fig. 2 17) to characterize the first pass of the contrast agent throughout the entire LV myocardium. Because of time constraints, each anatomic level is acquired at a different phase of the cardiac cycle. Cardiovascular Magnetic Resonance 27

2 CLINICAL CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES

Early systole

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Stress perfusion

slice 1

slice 2

slice 3

slice 4 Time slice 1

slice 2

slice 3

Rest perfusion

slice 4

Figure 2 16 The “goal” of perfusion imaging. Stress (top) and rest (bottom) perfusion images are shown. In this example, four slices (red, green, blue, and yellow) are acquired during each R R interval. An apical slice (blue) is shown at different time points throughout image acquisition (3 of 50 time frames are shown: before contrast arrival, at the time of contrast arrival in the right ventricle, and shortly after contrast arrival in the left ventricle). Two stress perfusion defects appear as dark regions (hypoenhancement, arrows). They are surrounded by bright contrast enhanced normally perfused myocardium. The defects are reversible because they are not present (no hypoenhancement, arrows) during rest perfusion.

The need to acquire multiple images within a single cardiac cycle is the primary reason why image quality is dimin ished for perfusion imaging. The total number of slices that can be acquired within one cardiac cycle is limited by the R R interval of the patient. For example, for a heart rate of 80 bpm, the R R interval is approximately 750 msec, and five short axis slices, each acquired within 150 msec, could theoretically be acquired. For higher heart rates, the number of short axis slice locations would be reduced. In addition, the heart rate may increase somewhat during adenosine infusion, so the scanner operator must allow some extra time within the resting R R interval to account for this possibility. Altogether, perfusion imaging is typically per formed for 40 to 50 seconds. To enhance the differences in image intensity between poorly and normally perfused myocardium, a saturation pulse (90 pulse) is applied as preparation before each image acquisition (Fig. 2 18). This 90 pulse tips the longi tudinal magnetization into the transverse plane. The trans verse magnetization then is dephased (spoiled) by a spoiler gradient. The contrast agent (e.g., gadolinium diethyl tria mine pentaacetic acid [DTPA]) shortens T1, recovery of the longitudinal magnetization, with normal perfusion and higher contrast concentration (relatively shorter T1) 28 Cardiovascular Magnetic Resonance

than in regions with reduced perfusion and lower contrast concentration (relatively longer T1, but still shorter than T1 without contrast). Figure 2 18 shows the relaxation curves for both regions. The timing between the saturation pulse and image read out strongly influences the difference in image intensity (contrast) between normal and hypoper fused myocardium. The optimal time for image acquisition is shown at location “b” in Figure 2 18. In practice, how ever, the time between saturation pulse and the first slice acquisition is less than optimal (e.g., location “a”) to allow acquisition of multiple slices within each cardiac cycle. The basic concept of GRE timing is shown in Figure 2 19. The pulse sequence starts with a slice selective gradient on the slice encoded axis and an RF pulse. The RF pulse tips the lon gitudinal magnetization by the flip angle a into the transverse plane to make it available for reception by CMR coils. The flip angle is typically set to 20 to 30 , which is nearly optimal (“Ernst angle”) in the setting of the rapid train of RF pulses used in GRE imaging. The underlying principle of GRE is that the spins are dephased by a first gradient and then rephased by a second gradient with opposite polarity to form a gradient echo, which is then recorded by the CMR receiver coils. As soon as data collection has finished, spoiler gradients are applied to destroy the remaining transverse magnetization

Acquire multiple entire slices during each heartbeat R ECG Trigger

1 Heartbeat 1

2

3

4

5

Beat 2

Beat 3

Beat 4

Beat 5

Slice 1 Slice 2 Slice 3 Slice 4 All lines for each image (one slice and one time point) are acquired in a single shot.

Figure 2 18 Magnetization recovery. The recovery of the longitudinal magnetization is faster in myocardium with normal perfusion because of T1 shortening as a consequence of high contrast agent concentration. Relatively speaking, T1 is prolonged in myocardium with reduced perfusion. To maximize the difference in signal intensity between myocardium with normal and reduced perfusion, the optimal time for image acquisition is when the difference between both curves is largest (b).

Mz

Myocardium with normal perfusion has shorter T1 (recovers faster, looks brighter)

+M0

c b

a

Myocardium with relatively reduced perfusion has longer T1 (recovers slower, looks darker) Time

90° RF Gradient Spoil transverse magnetization

before the next RF excitation pulse is applied. This is impor tant for avoiding image artifacts that can result from mixing of longitudinal and transverse magnetization. For stress myocardial perfusion imaging, a dose of aden osine 140 mg/kg/min is typically used for at least 2 minutes. The adenosine causes coronary vasodilation mediated by the A2A receptor24 and has a very short half life of less than 10 seconds. Adenosine acts to maximally dilate the

Readout one entire image (when curves have maximum separation)

distal arteriolar bed. In the absence of epicardial stenoses, adenosine can dilate these blood vessels and increase coro nary blood flow four to fivefold.25 For epicardial coronary arteries with significant stenoses, however, these arterioles are already fully dilated (in the absence of adenosine) because of the autoregulation mechanism that aims to restore normal flow impeded by a stenosis. Thus, further vasodilation with adenosine administration does not occur. Cardiovascular Magnetic Resonance 29

2 CLINICAL CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES

Figure 2 17 Image acquisition in the perfusion scan. Images are acquired in a continuous stream (single shot) over 50 to 60 heartbeats. Depending on the duration of the R R interval and cardiovascular magnetic resonance scanner hardware, up to five slices can be acquired during each R R interval.

RF

Slice

Destroy transverse magnetization to avoid mixing of longitudinal and transverse magnetization

A A A Spoiling

Phase Phase encoder B

Read

B

B

ADC Dephasing Signal Dephasing

Time

Although symptoms such as chest discomfort can occur, serious side effects (e.g., bronchospasm or atrioventricular block) are uncommon. (See Chapters 16 and 17.) Figure 2 20A shows the timeline for a comprehensive CMR stress test. After scout and resting cine imaging, aden osine is infused for at least 2 minutes. This minimum in fusion duration is chosen based on physiologic studies in humans, showing that, on average, maximum coronary

0

Scout imaging

Cine imaging

1

8

15

17

1 min

≈ 15 min

1 min

Wait or do additional cine imaging or valvular evaluation

Rest perfusion

2 min

Stress perfusion

Contrast injection 1

The resulting heterogeneity in blood flow between regions supplied by coronary arteries with stenoses (little or no flow increase) and territories supplied by normal epicardial arteries (multifold increased flow) is observed as a differ ence in myocardial contrast agent concentration and thus image intensity. Regional myocardium supplied by the dis eased vessels results in a hypoenhancement pattern (see Fig. 2 16) compared with normally perfused myocardium.

Contrast injection 2

Refocusing

IMAGING

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Figure 2 19 Pulse sequence timing diagram for gradient recalled echo imaging. ADC, analog digital converter; RF, radiofrequency.

Rephaser to keep spins aligned

α

25

33

LGE imaging 40

45

Time [minutes]

A Adenosine Contrast

2 min

1 min

Saline flush Breath holding Image acquisition

B

1

2 3

4

5

6

Figure 2 20 Timeline for a comprehensive cardiovascular magnetic resonance stress test. Time points 1 to 6 are referenced in the text. The time scale is not linear. LGE, late gadolinium enhancement. 30 Cardiovascular Magnetic Resonance

contrast agent concentration, and therefore hyperenhance ment. Chronic infarcts consist of dense scar with an increased interstitial space between collagen fibers (relative to normal myocardium). This leads to increased distribution volume of the gadolinium DTPA and consequently increased con centration and resultant hyperenhancement. The “goal” of LGE imaging is to create images with high contrast between the hyperenhanced nonviable tissue and normal myocardium for a clear delineation of the regions (Fig. 2 21). This is currently best achieved by using a segmented inversion recovery GRE sequence.27–30,38,39 Figure 2 22 shows the gating for a segmented inversion recovery sequence. The acquisition of multiple k space lines in each cardiac cycle allows reductions in imaging times to the point where the entire image can be acquired during a single breath hold of 6 to 10 seconds. The images are acquired in mid diastole by using a trigger delay to minimize cardiac motion. The magnetization of the heart is prepared by a nonselective 180 inversion pulse to create T1 weighting. The inversion time delay between inversion pulse and data collection (more precisely, the center of k space) is chosen such that the magnetization of viable (nor mal) myocardium is near its zero crossing, meaning that these regions appear dark (Fig. 2 23). Nonviable myocar dium (acutely infarcted or scar), however, appears bright because of the shorter T1 (faster signal recovery curve in Fig. 2 23) after contrast administration. It is important to adjust the inversion time manually for each image to null normal myocardium to account for washout kinetics of the contrast agent.40 At some centers, free breathing 3D acquisitions have displaced 2D breath hold LGE due to the opportunity for superior spatial resolution.40a

VIABILITY AND INFARCTION (“SCAN” ¼ “LGE”) The clinical implications of viability imaging are steadily grow ing. Beside the assessment of acute and chronic myocardial infarction, viability imaging27–30 has been used for the pre diction of contractile improvement after revascularization,31 for measuring the response to beta blocker treatment,32 for the differentiation of ischemic versus nonischemic cardio myopathy,33 and for the diagnosis of various nonischemic cardiomyopathies.34–38 A large body of evidence shows that LGE can differenti ate between nonviable and viable tissue.27–30,38,39 Regions of irreversible injury exhibit high signal intensity (hyperen hancement) on T1 weighted images after administration of extracellular CMR contrast agent, such as gadolinium DTPA. The contrast agent significantly shortens local longi tudinal relaxation time, resulting in signal increase. The underlying mechanism of hyperenhancement continues to be the subject of debate. The most likely explanation is that in acutely infarcted regions, the ruptured myocyte membranes allow the extracellular contrast agent to diffuse passively into the intracellular space, resulting in increased tissue level

Figure 2 21 The “goal” of late gadolinium enhancement viability imaging. The image provides a clear delineation between nonviable (hyperenhanced) and viable (dark) myocardium in this short axis image. Cardiovascular Magnetic Resonance 31

2 CLINICAL CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES

blood flow is reached 1 minute after the start of adenosine infusion, and in virtually every patient, it is reached after 2 minutes.26 Then a CMR contrast agent, such as gadolin ium DTPA (0.04 0.075 mmol/kg), is rapidly administered (3.5 mL/sec), followed by a (50 mL) saline flush. The adenosine infusion is stopped as soon as image acquisition is completed. Figure 2 20B shows the specific steps performed during the stress perfusion scan in more detail: (1) Adenosine infusion is started. (2) Two minutes later, image acquisition commences. (3) Contrast agent is injected at a rate of 3.5 mL/sec while adenosine infusion continues (two different intravenous lines are therefore required to avoid administering an adenosine bolus and promoting heart block). The patient still breathes freely. (4) As soon as the contrast medium reaches the RV, the patient is asked to suspend respiration for approximately 15 seconds to characterize the initial wash in of contrast agent into the LV myocardium. (5) The remain ing image acquisition is done during shallow breathing. (6) Image acquisition ends approximately 50 to 60 seconds after it began. This description assumes that images are dis played on the scanner in real time to allow observation of contrast agent arrival in the RV. If the scanner does not pro vide for real time image display, breath hold (step 4) should be started five to six heartbeats after initiation of the contrast agent injection. Rest myocardial perfusion is then performed. However, before the rest perfusion scan, a 15 minute delay is required for the contrast agent to clear sufficiently from the blood pool. During this time, additional cine scans or flow imaging for valvular evaluation can be performed. For the rest perfusion scan, an additional dose of gadolin ium DTPA 0.05 to 0.075 mmol/kg is given. LGE imaging may then be performed 10 minutes after the completion of the second perfusion study. The total scan time for a comprehensive CMR stress test, including cines, stress/rest perfusion, and LGE, is approximately 45 minutes.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

R

Figure 2 22 Electrocardiogram (ECG) triggering of the inversion recovery segmented gradient recalled echo sequence.

R

R Heartbeat 2

Heartbeat 1 ECG Trigger TI

TI

Segment 1 180° inversion

Segment 2 180° inversion

k-space

Mz

Figure 2 23 Inversion recovery curve for viable and nonviable myocardium. RF, radiofrequency.

Nonviable (infarcted) myocardium has shorter T1 (recovers faster, looks brighter)

TI (normal) +M0 TI (infarct)

Time

180°

Viable (normal) myocardium has longer T1 (recovers slower, looks darker) –M0

TI = inversion time

RF

Gradient Spoil transverse magnetization Readout one segment (when normal myocardium is nulled)

FLOW/VELOCITY IMAGING (“SCAN” ¼ “FLOW/ VELOCITY”) Velocity encoded (VENC) CMR imaging of blood flow is usually performed to measure velocity in arteries, veins, valves, and shunts. The overall goal is to evaluate the sever ity of valve regurgitation/stenosis or an intracardiac shunt. Further, LV and RV stroke volume can be calculated by summing up the area under the flow/velocity curve of the ascending aorta and pulmonary artery, respectively. With VENC CMR imaging, a cine series of grayscale images reflecting flow during the different phases of the cardiac cycle is acquired. The gray level is proportional to the velocity of blood into or out of the measured plane. 32 Cardiovascular Magnetic Resonance

Both in plane and through plane flow can be assessed. Fig ure 2 24 shows such a series in the upper row referenced to the ECG. The corresponding anatomic cine frames are seen in the lower row. The VENC image shows bright image intensity where fast velocity is present and gray intensity where minimal flow is present (background and parts of the leaflets that do not open). Flow in the opposite direction would show up in levels darker than the back ground (black to dark gray). Analogous to cine imaging, each VENC image corre sponds to a cardiac phase, and ECG gating is required. The sequence is run in retrogated mode to cover the entire cardiac cycle. This is important for the determination of cardiac output, because skipping the last frame would miss the beginning of systole and hence deliver erroneous volumes.

ECG

Velocity-encoded images

Cine images at corresponding times Figure 2 24 The “goal” of flow (velocity encoded) imaging. A selection of velocity encoded images of the aortic valve exhibiting stenosis is shown in the middle. Images are registered with the electrocardiogram (ECG) on top, and corresponding cine frames on the bottom are included as anatomic reference.

The physics of VENC are complex.41,42 Simply put, in the presence of a magnetic field gradient, the spins in the image plane acquire a phase that is proportional to the area under the gradient plotted in a pulse sequence diagram, such as the one shown in Figure 2 25. For nonmoving

Figure 2 25 Pulse sequence timing diagram of the electrocardiogram triggered velocity encoded sequence. Gradients on the read axis are flow compensated. On the slice axis, two gradient waveforms are shown inside the dotted ellipses. One is insensitive to flow (“flow compensated block”) and one is sensitive toward flow (“flow sensitized block”). The sequence necessary to acquire one line of data is shown. This is an example for assessment of through plane flow, but in plane flow can be imaged as well.

spins, this phase can be rewound (the spins can be rephased) by a gradient of the same magnitude, but the opposite sign. If, however, the spins change position between the first gradient and the rewinding gradient, as is the case for spins in flowing blood, then they experience

ADC Time

RF

Read

Phase

Slice

Flow-compensated block

Flow-sensitive block

Cardiovascular Magnetic Resonance 33

2 CLINICAL CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES

R

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

a gradient during rewinding that is not only opposite in sign, but also different in magnitude. This incomplete or excessive rewinding causes a phase difference that is a lin ear function of velocity. Displaying this phase as image 1 represents velocity as (grayscale) image. Because the phase of spins is also dependent on many other factors (e.g., the properties of receiver coils), it is necessary to measure a base phase during flow insensitive gradient waveforms (see Fig. 2 25, left dotted oval). The flow sensitized waveform is shown in the right dotted oval. By subtracting the base phase from the flow phase and dis playing the difference on a grayscale, the images of Fig ure 2 24, upper row, are obtained. The gray level is proportional to velocity: black corresponds to a maximum backward flow, white corresponds to a maximum forward flow, and gray corresponds to no flow. The range from maximum backward flow to maximum forward flow can be quantified in centimeters per second and must be adjusted by the scanner operator to maximize sensitivity to differing physiologic blood flow velocities. The images allow both a coarse overview and a quantita tive assessment of abnormal flow patterns. Typically, flow is acquired in the axial plane at the level of the bifurcation of the pulmonary artery and in the near coronal plane in the proximal pulmonary artery. From these data, LV forward stroke volume, aortic regurgitation, RV forward stroke vol ume, and pulmonic regurgitation can be readily measured and the pulmonic:systemic (Qp:Qs) ratio can also be deter mined. Mitral and tricuspid regurgitation can then be cal culated as the difference between the respective stroke volume (from the short axis stack of cine images) and the forward stroke volume (see Chapter 37).

ANGIOGRAPHY (“SCAN” ¼ “ANGIOGRAPHY”) Vascular angiography is often performed independently of CMR stress or viability tests. However, this scan is frequently combined with VENC imaging. Occasionally, it is done in

conjunction with a more detailed cardiovascular examination (e.g., pulmonary vein imaging, aortic root angiography). This section discusses some practical issues with regard to ECG gated contrast enhanced MR angiography (CE MRA). This chapter does not discuss different angiography techniques, vessel wall or coronary imaging, or advanced forms of motion correction (“navigator echoes”). The “goal” of CE MRA is to create a three dimensional data set of a vessel or vascular bed of interest. Use of a T1 shortening paramagnetic contrast agent within the blood pool allows the generation of high signal within the vascular tree relative to surrounding tissue, thus creating a “luminogram.” Figure 2 26 shows two examples of CE MRA. The images were colored during routine postprocessing. Figure 2 26A shows the left ventricle, the aortic arch, the descending aorta, the truncus brachiocephalicus, and carotid and subclavian arteries. Figure 2 26B shows pulmonary arteries and veins. The timing of the image acquisition was specifically chosen such that the LV and aorta are not yet filled by contrast rich blood and thus are not visible. This “procedure” would com monly be done as part of a complete evaluation of a patient with vascular disease that would include multiple different scans from the exam menu, in addition to the CE MRA. Figure 2 27 shows the timeline of the procedure. The noncontrast scan is done for later subtraction from the contrast enhanced scan to eliminate background. Unlike for lower extremity MRA, ECG gating is critical for imaging of vascular structures that experience motion as a result of cardiac movement or pulsatile blood flow (e.g., aortic root). In these situations, it is important to acquire imaging data only during diastole (see Fig. 2 7) because motion, even during just a portion of the acquisi tion, will adversely affect image quality. In the case of highly dynamic structures, such as the ascending aorta, it is important to account for not only the systolic cycle length, but also the period in early diastole when passive relaxation of the aorta occurs because of the elastic recoil of the vessel. This short period of “diastolic vessel relaxa tion” leads to appreciable vascular motion that can blur the edges of an angiogram and compromise its quality. These considerations were taken into account for obtaining

Ao

LA RV

A

B

Figure 2 26 Three dimensional contrast enhanced magnetic resonance angiography maximum intensity projection of the aortic root and thoracic aorta. This image was electrocardiogram triggered for high resolution imaging of the aortic root in a young woman with a history of a bicuspid aortic valve and aortic root dilation. A, anterior view; B, posterior view. Ao, ascending aorta; LA, left atrium; RV, right ventricle. 34 Cardiovascular Magnetic Resonance

Contrast injection

CONTRAST OPTIONAL IMAGING IMAGING

Selected brightblood imaging

≈ 2 min

≈ 6 min

≈ 5 min

≈ 5 min

Scout imaging

HASTE (2–3 planes)

Noncontrast 3D angio (GRE)

3D CE-MRA (arterial +/– venous phase)

16

23

0

1

6

12 Time (min)

Figure 2 27 Timeline of a typical contrast enhanced magnetic resonance angiography (CE MRA). The time scale is not linear. In addition to scouts and fast spin echo imaging, selected “bright blood” (steady state free precession) images are obtained for analysis of extravascular structures and vascular structures that may not be in the angiographic field of view. A noncontrast three dimensional angiogram is performed before the CE MRA so that a subtraction image can be generated. A comprehensive study can be completed in 20 to 25 minutes. GRE, gradient recalled echo; HASTE, half Fourier single shot fast spin echo.

the images of Figure 2 26 because they would not be as crisp otherwise. For CE MRA, the most commonly used pulse sequence is a fast three dimensional spoiled GRE. This is a variant of the GRE technique that was discussed previously (see Fig. 2 19). The need for three dimensional data requires application of an additional gradient to encode the additional dimension. This gradient is played in the slice encoding direction. Com pared with two dimensional GRE, a much larger amount of data is acquired, leading to longer scan times. The use of advanced parallel image acquisition and faster gradient sets has decreased scan times significantly. With the use of mod ern hardware and sequences, high resolution ECG gated CE MRA can be completed in 10 to 20 seconds. The most common type of angiography used on the clinical service, CE MRA is easily incorporated into the clinical CMR examination without dramatically prolonging the duration of the study or compromising the quality of the other components.

CONCLUSION This chapter described an approach to routine clinical CMR that is based on selecting one or more scans from an exam menu and assembling these into a test. Figure 2 28 shows how one would assemble a test for a patient referred for the evaluation of coronary artery disease. Specifically, one would perform scouting, cine imaging, perfusion imaging at stress and rest, and finally late gadolinium enhancement imaging. The protocol of Figure 2 28 is identical to the CMR stress test2 discussed earlier. Importantly, and as previously noted, this test now accounts for approximately 50% of the 3000 annual CMR procedures performed at the DCMRC. Similarly,

Scout

LV cine

Perfusion

LGE

Figure 2 28 Combination of menu items for the evaluation of coronary artery disease. LGE, late gadolinium enhancement; LV, left ventricle.

Scout

Morphology

Angiography

Flow/ velocity

Figure 2 29 Combination of menu items for evaluation of the aorta.

Figure 2 29 summarizes how one would approach CMR scanning of a patient referred for the evaluation of aortic disease. In summary, although the definition of “routine clinical CMR” continues to evolve at the DCMRC and other institu tions,1 it is very useful to assemble a short list of predefined individual CMR scan protocols (see Fig. 2 3) from which one can create a “test package” specifically tailored to the diagnostic question. This approach substantially decreases the time needed to scan because the operator does not have to select from among the literally hundreds of buttons on the scanner console while the patient waits idly inside the magnet. Instead, the buttons are predefined for each menu item. This approach improves the throughput of the DCMRC clinical CMR service. Perhaps more importantly, however, this approach dramatically reduces the histori cally steep learning curve for new physicians interested in learning how to perform CMR. Cardiovascular Magnetic Resonance 35

2 CLINICAL CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES

≈ 3 min

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

References 1. Bruder O, Schneider S, Nothnagel, et al. Euro CMR (European Cardiovascular Magnetic Resonance) registry: results of the pilot phase. J Am Coll Cardiol. 2009;54:1457–1566. 2. Klem I, Heitner JF, Shah DJ, et al. Improved detection of coronary artery disease by stress perfusion cardiovascular magnetic resonance with the use of delayed enhancement infarction imaging. J Am Coll Cardiol. 2006;47:1630–1638. 3. Oppelt A, Graumann R, Barfuss H, et al. Eine neue schnelle pulssequenz fuer die kernspintomographie. Electromedica. 1986;54:15–18. 4. Fischer SE, Wickline SA, Lorenz CH. Novel real-time R-wave detection algorithm based on the vector cardiogram for accurate gated magnetic resonance acquisitions. Magn Reson Med. 1999;361–370. 5. Bellenger NG, Burgess MI, Ray SG, et al. Comparison of left ventricular ejection fraction and volumes in heart failure by echocardiography, radionuclide ventriculography and cardiovascular magnetic resonance: are they interchangeable? Eur Heart J. 2000;21:1387–1396. 6. Grothues F, Moon JC, Bellenger NG, et al. Interstudy reproducibility of right ventricular volumes, function, and mass with cardiovascular magnetic resonance. Am Heart J. 2004;147:218–223. 7. Grothues F, Smith GC, Moon JC, et al. Comparison of interstudy reproducibility of cardiovascular magnetic resonance with two-dimensional echocardiography in normal subjects and in patients with heart failure or left ventricular hypertrophy. Am J Cardiol. 2002;90:29–34. 8. Bellenger NG, Rajappan K, Rahman SL, et al. Effects of carvedilol on left ventricular remodelling in chronic stable heart failure: a cardiovascular magnetic resonance study. Heart. 2004;90:760–764. 9. Friedrich MG, Dahlof B, Sechtem U, et al. Reduction (TELMAR) as assessed by magnetic resonance imaging in patients with mild-to-moderate hypertension: a prospective, randomised, double-blind comparison of telmisartan with metoprolol over a period of six months rationale and study design. J Renin Angiotensin Aldosterone Syst. 2003;4:234–243. 10. Lamb HJ, Beyerbacht HP, de Roos A, et al. Left ventricular remodeling early after aortic valve replacement: differential effects on diastolic function in aortic valve stenosis and aortic regurgitation. J Am Coll Cardiol. 2002;40:2182–2188. 11. Ioannidis JP, Trikalinos TA, Danias PG. Electrocardiogram-gated single-photon emission computed tomography versus cardiac magnetic resonance imaging for the assessment of left ventricular volumes and ejection fraction: a meta-analysis. J Am Coll Cardiol. 2002; 39:2059–2068. 12. Kjaergaard J, Petersen CL, Kjaer A, et al. Evaluation of right ventricular volume and function by 2D and 3D echocardiography compared to MRI. Eur J Echocardiogr. 2006;7:430–438. 13. Persson E, Carlsson M, Palmer J, et al. Evaluation of left ventricular volumes and ejection fraction by automated gated myocardial SPECT versus cardiovascular magnetic resonance. Clin Physiol Funct Imaging. 2005;25:135–141. 14. Schaefer WM, Lipke CS, Standke D, et al. Quantification of left ventricular volumes and ejection fraction from gated 99mTc-MIBI SPECT: MRI validation and comparison of the Emory Cardiac Tool Box with QGS and 4D-MSPECT. J Nucl Med. 2005;46:1256–1263. 15. Thorley PJ, Plein S, Bloomer TN, et al. Comparison of 99mTc tetrofosmin gated SPECT measurements of left ventricular volumes and ejection fraction with MRI over a wide range of values. Nucl Med Commun. 2003;24:763–769. 16. Corsi C, Lang RM, Veronesi F, et al. Volumetric quantification of global and regional left ventricular function from real-time three-dimensional echocardiographic images. Circulation. 2005;112:1161–1170. 17. Dewey M, Muller M, Eddicks S, et al. Evaluation of global and regional left ventricular function with 16-slice computed tomography, biplane cineventriculography, and two-dimensional transthoracic echocardiography: comparison with magnetic resonance imaging. J Am Coll Cardiol. 2006;48:2034–2044. 18. Fischbach R, Juergens KU, Ozgun M, et al. Assessment of regional left ventricular function with multidetector-row computed tomography versus magnetic resonance imaging. Eur Radiol. 2007;17:1009–1017. 19. Potter DD, Araoz PA, McGee KP, et al. Low-dose dobutamine cardiac magnetic resonance imaging with myocardial strain analysis predicts myocardial recoverability after coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2008;135:1342–1347. 20. Jahnke C, Nagel E, Gebker R, et al. Prognostic value of cardiac magnetic resonance stress tests: adenosine stress perfusion and dobutamine stress wall motion imaging. Circulation. 2007;115:1769–1776.

36 Cardiovascular Magnetic Resonance

21. Giang TH, Nanz D, Coulden R, et al. Detection of coronary artery disease by magnetic resonance myocardial perfusion imaging with various contrast medium doses: first European multi-centre experience. Eur Heart J. 2004;25:1657–1665. 22. Paetsch I, Jahnke C, Wahl A, et al. Comparison of dobutamine stress magnetic resonance, adenosine stress magnetic resonance, and adenosine stress magnetic resonance perfusion. Circulation. 2004;110: 835–842. 23. Wolff SD, Schwitter J, Coulden R, et al. Myocardial first-pass perfusion magnetic resonance imaging: a multicenter dose-ranging study. Circulation. 2004;110:732–737. 24. Udelson JE, Heller GV, Wackers FJ, et al. Randomized, controlled dose-ranging study of the selective adenosine A2A receptor agonist binodenoson for pharmacological stress as an adjunct to myocardial perfusion imaging. Circulation. 2004;109:457–464. 25. Wilson RF, Wyche K, Christensen BV, et al. Effects of adenosine on human coronary arterial circulation. Circulation. 1990;82:1595–1606. 26. Rossen JD, Quillen JE, Lopez AG, et al. Comparison of coronary vasodilation with intravenous dipyridamole and adenosine. J Am Coll Cardiol. 1991;18:485–491. 27. Kim RJ, Fieno DS, Parrish TB, et al. Relationship of MRI delayed contrast enhancement to irreversible injury, infarct age, and contractile function. Circulation. 1999;100:1992–2002. 28. Wu E, Judd RM, Vargas JD, et al. Visualisation of presence, location, and transmural extent of healed Q-wave and non-Q-wave myocardial infarction. Lancet. 2001;357:21–28. 29. Fieno DS, Kim RJ, Chen EL, et al. Contrast-enhanced magnetic resonance imaging of myocardium at risk: distinction between reversible and irreversible injury throughout infarct healing. J Am Coll Cardiol. 2000;36:1985–1991. 30. Wagner A, Mahrholdt H, Holly TA, et al. Contrast-enhanced MRI and routine single photon emission computed tomography (SPECT) perfusion imaging for detection of subendocardial myocardial infarcts: an imaging study. Lancet. 2003;361:374–379. 31. Kim RJ, Wu E, Rafael A, et al. The use of contrast-enhanced magnetic resonance imaging to identify reversible myocardial dysfunction. N Engl J Med. 2000;343:1445–1453. 32. Bello D, Shah DJ, Farah GM, et al. Gadolinium cardiovascular magnetic resonance predicts reversible myocardial dysfunction and remodeling in patients with heart failure undergoing beta-blocker therapy. Circulation. 2003;108:1945–1953. 33. McCrohon JA, Moon JC, Prasad SK, et al. Differentiation of heart failure related to dilated cardiomyopathy and coronary artery disease using gadolinium-enhanced cardiovascular magnetic resonance. Circulation. 2003;108:54–59. 34. Mahrholdt H, Wagner A, Judd RM, et al. Delayed enhancement cardiovascular magnetic resonance assessment of non-ischaemic cardiomyopathies. Eur Heart J. 2005;26:1461–1474. 35. Mahrholdt H, Goedecke C, Wagner A, et al. Cardiovascular magnetic resonance assessment of human myocarditis: a comparison to histology and molecular pathology. Circulation. 2004;109:1250–1258. 36. Choudhury L, Mahrholdt H, Wagner A, et al. Myocardial scarring in asymptomatic or mildly symptomatic patients with hypertrophic cardiomyopathy. J Am Coll Cardiol. 2002;40:2156–2164. 37. Simonetti OP, Kim RJ, Fieno DS, et al. An improved MR imaging technique for the visualization of myocardial infarction. Radiology. 2001; 218:215–223. 38. Mahrholdt H, Klem I, Sechtem U. Cardiovascular MRI for detection of myocardial viability and ischaemia. Heart. 2007;93:122–129. 39. Weinsaft JW, Klem I, Judd RM. MRI for the assessment of myocardial viability. Magn Reson Imaging Clin N Am. 2007;15:505–525, v–vi. 40. Wagner A, Mahrholdt H, Thomson L, et al. Effects of time, dose, and inversion time for acute myocardial infarct size measurements based on magnetic resonance imaging-delayed contrast enhancement. J Am Coll Cardiol. 2006;47:2027–2033. 40a. Peters DC, Appelbaum E, Nezafat R, et al. Left ventricular infarct size, peri-infarct zone and papillary scar measurements: a comparison of high resolution 3D and conventional 2D late gadolinium enhancement cardiac MR. J Magn Reson Imaging. 2009;30:794–800. 41. Debatin JF, Ting RH, Wegmu¨ller H. Renal artery blood flow: quantitation with phase-contrast MR imaging with and without breath holding. Radiology. 1994;190:371–378. 42. Pelc LR, Pelc NJ, Rayhill SC, et al. Arterial and venous blood flow: noninvasive quantitation with MR imaging. Radiology. 1992;185: 809–812.

Advanced Cardiovascular Magnetic Resonance Imaging Techniques: Spiral, Radial, and Parallel Imaging Daniel K. Sodickson and Thoralf Niendorf

This chapter is concerned with the basic principles and car diovascular applications of spiral imaging, radial imaging, and parallel imaging. Arguably, all three approaches earn the moniker of “advanced cardiovascular magnetic reso nance (CMR) techniques” because of the nontraditional paths they take through the magnetic resonance (MR) data space. Traditionally, most routine clinical day to day MR data acquisitions have been aimed at traversing k space in a so called Cartesian pattern, with data points acquired at regular intervals on a two dimensional (2D) or three dimensional (3D) grid, and with one point and one line of data acquired at a time. Conceptually, at least, such an approach is straightforward to implement and to under stand. It mimics the regular gridded structure of pixels or voxels in the images that are ultimately produced, and the criteria for completion of image acquisition are clear: image acquisition stops when all of the points on the target grid are populated by data. However, the organs and the bodies that are imaged are continuous entities, and the data space representing them is at root continuous as well. Thus, noth ing prevents us from acquiring data along non Cartesian paths, so long as those data may be manipulated to yield reliable representations of image contents. Spiral imaging and radial imaging are two examples of non Cartesian acqui sition trajectories that have been studied extensively and used increasingly, particularly for rapid and real time CMR. Parallel magnetic resonance imaging (parallel MRI), mean while, challenges the notion that data acquisition must be a monolithic and sequential path through k space. Unlike traditional approaches that rely entirely on magnetic field gradients to move from one data point to another in MR acquisition, parallel MRI techniques use arrays of radiofre quency (RF) detector coils to supplement the spatial encod ing provided by gradients and to generate multiple data points at once rather than one after the other. The result is an acceleration of MR image acquisition beyond previous limits. For imaging of the cardiovascular system, imaging speed is of course at a premium, and parallel MRI tech niques have seen extensive use in cardiovascular applications. Another theme connecting spiral, radial, and parallel MRI is rapid imaging. Cardiovascular applications have been a significant motivating force for the development of

ever more rapid MR imaging techniques over the years, and nowhere are the challenges and the benefits of rapid MRI more apparent than in the field of CMR, where cardiac motion, respiratory motion, and blood flow all complicate imaging. Spiral and radial imaging sequences have proven to be powerful tools for rapid imaging, albeit for rather dif ferent reasons (the former because of efficient use of field gradients and the latter because of favorable undersampling behavior, as discussed in more detail later). Parallel MRI, on the other hand, is a general strategy that may be used to accelerate most existing imaging sequences, including spiral and radial sequences. In the sections that follow, the basic principles and typi cal cardiovascular applications of spiral imaging, radial ima ging, and parallel imaging will be surveyed. A concluding section explores future directions.

SPIRAL IMAGING Principles Spiral trajectories1,2 have been used extensively for real time imaging. A sample spiral trajectory is shown in Figure 3 1B, compared with a Cartesian trajectory shown in Figure 3 1A. Figure 3 2 summarizes the process of data acquisition and image reconstruction for spiral imaging sequences. The spiral path through k space is accom plished using gradients that oscillate in a coordinated fash ion along two in plane directions (see Fig. 3 2, left). The lack of sharp transitions in the oscillating gradient patterns allows efficient use even of gradients with limited switching rates, and high switching rates may be employed for still greater speed. Unfortunately, the irregular sampling pattern of a typical spiral trajectory precludes the use of a simple fast Fourier transform for image reconstruction. Various specialized reconstruction procedures have been proposed. Regridding algorithms3,4 that map the data to a Cartesian grid are most frequently used. After regridding, a fast Four ier transform may be performed to yield the reconstructed image (see Fig. 3 2, right). Although the regridding step adds somewhat to reconstruction time, with modern Cardiovascular Magnetic Resonance 37

3 ADVANCED CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES: SPIRAL, RADIAL, AND PARALLEL IMAGING

CHAPTER 3

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

B

A

C

Figure 3 1 Cartesian (A), spiral (B), and radial (C) data acquisition strategies. Paths through k space are shown in the top row, with arrows indicating the readout direction. Resulting k space data matrices after digitization of the cardiovascular magnetic resonance signal are seen in the bottom row.

hardware and software, it may be performed efficiently, such that low latency display is possible.5 The traditional distinction between phase encoding and frequency encoding directions is lost in spiral sequences, and this has some important consequences for the resulting images. First, errors can accumulate throughout the long spi ral readouts, as also occurs in Cartesian echo planar imaging readouts. However, in spiral imaging the effects of these errors tend to be distributed across the image, rather than being concentrated at well defined aliasing positions or visi ble as coherent spatial distortions or shifts, as is the case for echo planar imaging. For example, off resonance effects induced by field inhomogeneities, susceptibility variations, or chemical shift, coupled with a long readout time, may result in blurring of images obtained from spiral acquisi tions. An example of spiral blurring as a result of off reso nance effects is shown on the left of Figure 3 3A (with a crisper Cartesian image shown on the right). Such blurring can be reduced with: (1) a priori B0 field maps acquired in extra scans6,7; (2) estimates of the spatially varying off Acquisition

resonance frequency obtained from the spiral data itself8; or (3) frequency selective water excitation/fat suppression.9 A number of variants on spiral imaging sequences are used for various purposes. For real time implementations, data are typically acquired continuously during a single long spi ral readout, which promotes speed but also allows phase errors to accumulate. Multi shot spiral acquisitions, with interleaved spiral trajectories that follow separate RF pulses, may be used to control these errors, (see real time spiral images in Fig. 3 3B). A typical spiral trajectory begins in the center of k space and works its way symmetrically out ward, but reverse spiral trajectories with different relaxation weighting have also been explored.10,11 Three dimensional “stack of spiral” sequences, which apply traditional phase encoding along the direction perpendicular to the spiral, have also been described.12,13

Applications Spiral k space sampling saw its initial cardiovascular appli cation in the area of breath hold 2D coronary artery CMR, where the speed gain enabled (1
1
5) mm3 spatial res olution for a comparatively small number of slices.2 For this first application, interleaved spiral scanning was used without considering the motion of the coronaries. To follow the motion of the coronary artery in spiral coronary artery CMR, vessel tracking with prospective adjustment of the slice location as a function of the coronary position can be applied.14 To further improve the efficiency of spiral cor onary artery CMR techniques with submillimeter in plane spatial resolution as well as a high image signal to noise ratio (SNR), adaptive, subject tailored, automated tracking of the vessel motion over the cardiac cycle has been used.15 Alternatively, variable density spiral k space acquisitions were proposed to acquire coronary artery CMR at submil limeter spatial resolution with information for motion com pensation obtained directly from the coronary anatomy itself.16 This approach eliminates the need for cardiac Reconstruction

RF Regrid

Gx FFT

Gy

Gz

ADC

38 Cardiovascular Magnetic Resonance

Figure 3 2 Data acquisition and image reconstruction strategies for spiral imaging. A sample pulse sequence for a spiral acquisition is shown on the left, with various lines showing the timing of radiofrequency (RF) pulses, in plane gradients Gx and Gy, slice direction gradient Gz, and data reception (labeled ADC for analog to digital conversion). Note the smooth oscillation of in plane gradients. Reconstruction strategies for spiral datasets are shown on the right. The most common approach is to regrid the spiral data onto a Cartesian grid (using well established interpolation algorithms) and then to perform a fast Fourier transform (FFT) to generate a final image. It is also possible in principle to transform the spiral data directly into an image (dashed line), but this would increase computational burden significantly.

3 ADVANCED CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES: SPIRAL, RADIAL, AND PARALLEL IMAGING

Figure 3 3 Montage of cardiovascular magnetic resonance (CMR) applications of spiral imaging and comparisons with other trajectories. A, Short axis two dimensional (2D) steady state free precession (SSFP) cine images from spiral (left) and Cartesian (right) acquisitions. The spiral image on the left exhibits characteristic blurring as a result of B0 inhomogeneities as well as physiologic motion. For comparison, the Cartesian image on the right displays reduced image distortion and improved delineation of myocardial borders because of shortening of the echo train length. B, Long axis two dimensional cine images (left, systole; right, diastole) derived from CMR fluoroscopy using spiral k space trajectories (four interleaves) and a frame rate of 15 frames/sec (using sliding window reconstruction). A nominal in plane spatial resolution of (2.5
2.5) mm2 was used for data acquisition. C, Magnetic resonance angiogram (MRA) of the left coronary artery system obtained with a free breathing, navigated, three dimensional steady state free precession (SSFP) technique using spiral (left) and Cartesian (right) k space trajectories. The speed advantage of spiral imaging enabled an in plane spatial resolution of 0.77
0.77 mm2 (left), whereas the traditional Cartesian approach (right) yielded an in plane spatial resolution of 1.0
1.0 mm2. D, Visualization of the right coronary artery vessel wall using spiral (left) and radial (right) k space sampling schemes. Motion artifacts are reduced in radial k space sampling so that the vessel wall is better delineated in the case of radial k space sampling. (B, Images courtesy of Gabriele Krombach, RWTH Aachen University, Germany; C, images courtesy of Rene Botnar, PhD, Guy’s and St. Thomas’ Hospital, London, United Kingdom; D, images courtesy of Marcus Katoh, PhD, RWTH Aachen University, Germany.)

A

B

C

D

Cardiovascular Magnetic Resonance 39

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

triggering, breath holding, and navigator echoes, although extended acquisition times are needed to overcome the reduced SNR efficiency associated with the method. Free breathing 3D spiral coronary artery CMR affords submillime ter in plane spatial resolution (see Fig. 3 3C). Recent pulse sequence and hardware developments have served to further reduce the acquisition window length, enabling rapid 2D spiral acquisitions covering a large volume of the heart at a spatial resolution of 1
1
2 mm3.17 Double inversion prepared black blood spiral imaging in tandem with free breathing, navigator gated, cardiac triggered CMR facilitates coronary vessel wall imaging18 (see Fig. 3 3D). Assessment of global and regional cardiac function has also been performed with spiral imaging techniques. For example, electrocardiographic (ECG) gating of 2D steady state free precession (SSFP) spiral acquisitions have been used to achieve full R R coverage, high temporal resolution, and short scan times.19 However, overall image quality in these spiral studies was inferior to that derived from Carte sian sampling (see Fig. 3 3A). Variable density spiral trajec tories can be employed to further reduce motion artifacts in 2D cine CMR.20 Interactive spiral real time imaging methods, which eliminate the need for ECG gating, have gained importance for dynamic studies, such as assessment of cardiac func tion, guidance of interventional procedures, or flow mea surements. For example, the speed advantage of spiral imaging has been exploited for rapid left ventricular func tion assessment using free breathing cine CMR without car diac triggering, supported by real time reconstruction/ display and interactive section positioning.21 Other studies reported a temporal resolution of 120 msec (24 frames/sec using sliding window reconstruction and display) while accomplishing almost 1 mm in plane spatial resolution, together with high blood myocardium contrast, affording excellent visualization of blood pool, myocardial wall, and valve leaflet motion at 1.5 T and 3.0 T.22,23 Real time high temporal resolution spiral imaging reduces underesti mation of the peak velocity in flow velocity imaging as a result of averaging over a shorter period around the peak

Acquisition

amplitude.24 Alternatively, the speed of accelerated spiral imaging can be translated into an improved spatial resolu tion that decreases partial volume effects in flow velocity imaging that arise from partial occupancy of the voxels with static spins.24

RADIAL IMAGING Principles A variant of the radial trajectory was used to generate the very first MR images.25,26 Radial trajectories were subsequently reintroduced and further explored for their comparative insensitivity to motion.27 A sample radial acquisition trajectory is shown in Figure 3 1C, and Figure 3 4 shows a sample pulse sequence and image reconstruction. As is the case for spiral imaging, the tradi tional distinction between phase encoding and frequency encoding directions is abandoned in a radial acquisition. For radial trajectories, however, coordinated gradient switching along the two in plane directions occurs not dur ing but between readouts, and each readout occurs along a straight line, with some angular shift from the last readout. As is the case for spiral trajectories, radial image recon structions are more complex than for the Cartesian case, typically involving either regridding procedures or projec tion reconstruction algorithms (e.g., filtered back projec tion [FBP]) analogous to those used in X ray computed tomography (CT). The connections between MR data acquired in a radial trajectory and CT data acquired with a moving X ray gantry are so close, in fact, that radial MRI is sometimes referred to as projection reconstruction imaging. The motion insensitivity of radial trajectories results in part from their intrinsic oversampling of the center of k space, which is traversed by each angular readout, or pro jection. Because each copy of the k space center constitutes a low resolution representation of the imaged field of view, some degree of motion can be averaged out in an image

Reconstruction

RF Regrid

Gx

Gz

ADC

40 Cardiovascular Magnetic Resonance

FFT

PR

Gy

Figure 3 4 Data acquisition and reconstruction strategies for radial imaging. A sample pulse sequence for a radial acquisition is shown on the left, labeled as in Figure 3 2. In plane gradients are switched in a coordinated fashion to accomplish readouts along progressively rotated directions. Reconstruction strategies for radial datasets are shown on the right. Regridding and fast Fourier transform may be performed as for spiral data. Alternatively, projection reconstruction (PR) approaches similar to those used in X ray computed tomography (CT) may be used to generate an image. ADC, analog to digital conversion; RF, radiofrequency.

Figure 3 5 Undersampling behavior for Cartesian (A and C) and radial (B and D) imaging. A, A twofold undersampled Cartesian trajectory. Acquired k space lines are shown in solid black and omitted lines are shown in dashed gray. B, A twofold undersampled radial trajectory. C, The aliased image resulting from Fourier transformation of the undersampled Cartesian trajectory in (A). Note the coherent replication of structures along the undersampled direction, which clearly obscures important cardiac anatomy. D, The image resulting from regridding reconstruction of the undersampled radial trajectory in (B). Undersampling artifact in the form of radial streaking is visible, and cardiac structures are fully visible.

A

C

positions, producing characteristic radial streaking arti facts (see Fig. 3 5D). For undersampled radial trajectories in particular, it has been recognized that significant degrees of undersampling may be tolerated without obscuring important cardiac or vascular anatomy, and without compromising effective spa tial resolution.28,29 (The undersampled radial image at the right in Fig. 3 5D, for example, is preferable to the aliased Cartesian image in Fig. 3 5C.) As a result, smaller datasets may be acquired, restoring the speed and efficiency of radial trajectories. Undersampled radial (or undersampled projection reconstruction) techniques have been combined with real time SSFP CMR, for example, to yield high con trast between blood and myocardium.30–32 Further accel erations have been achieved using combinations of radial undersampling with multi echo readouts.32 The under sampled projection reconstruction principle has also been extended to three dimensions in at least two ways: either using a “stack of stars” approach with traditional phase encoding in the slice direction,28 or using a true 3D projection reconstruction approach with radial projections exten ding along all three directions in what has been called colloquially a koosh ball trajectory. The latter approach, termed vastly undersampled isotropic projection reconstruction (VIPR),33 has been used to achieve dramatic accelerations for angiographic applications in which the field of view is

B

D

Cardiovascular Magnetic Resonance 41

3 ADVANCED CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES: SPIRAL, RADIAL, AND PARALLEL IMAGING

reconstruction that effectively merges these representations. Somewhat paradoxically for a sequence that is often identi fied with rapid imaging, however, this same redundancy also represents an underlying inefficiency in data acquisi tion. In principle, the number of acquired data points needed to gather the outer portions of k space with a req uisite separation is generally larger than for Cartesian acquisitions. Nevertheless, radial trajectories have had a recent resurgence for rapid and real time imaging applica tions, in part because of their undersampling behavior.28 When Cartesian acquisitions are undersampled (i.e., when intermediate lines of k space data are omitted such that the spacing of lines exceeds the Nyquist limit [Fig. 3 5A]), the resulting images show well defined aliasing artifacts (see Fig. 3 5C). This aliasing results from the inability of the Fourier transform operation to separate certain regu larly spaced points in the image plane based only on the undersampled spatial frequency data. The regular pat tern of Cartesian undersampling results in coherent over lap of spatially separated image regions that can obscure important structures (see Fig. 3 5C). The appearance of undersampling artifacts is notably different in many non Cartesian trajectories, however. For example, the lack of a regular Cartesian grid structure to the acquired and omitted data in undersampled radial trajectories results in a spreading of aliasing artifact among multiple spatial

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

dominated by narrow vascular structures. In such applica tions, radial undersampling artifacts may be reduced to a low level by spreading throughout the imaged volume.

Applications Resilience to undersampling and motion artifacts renders radial trajectories effective candidates for the accurate assessment of global and regional cardiac wall motion. Radial 2D cine SSFP techniques afford image quality com parable to that of Cartesian k sampling techniques19,21 (Fig. 3 6A). In particular, real time assessment of cardiac function benefits from the reduced motion sensitivity of radial k space sampling. This leads to enhanced signal homogeneity in the ventricle while preserving blood myocardium contrast, which improves the accuracy of the cardiac functional assessment. The scan efficiency of radial imaging, together with real time sliding window recon struction, enables frame rates of up to 20 frames/sec for small matrix sizes (see Fig. 3 6B, left) and approximately 10 frames/sec for high spatial resolution imaging (see Fig. 3 6B, right). Real time radial cine CMR techniques eliminate the need for ECG gating but offer somewhat limited spatial and tempo ral resolution compared with conventional segmented radial techniques. Self gated radial acquisition strategies were intro duced to make up for this deficit. Radial self gated techniques extract the motion synchronization signal directly from the same MR signals used for image reconstruction.34 The self gat ing approach can also be extended to derive respiratory gating information directly from the raw imaging data, which enables free breathing segmented cine imaging using radial k space trajectories.34,35 Free breathing, ungated CMR fluoroscopy is another area that employs the scan efficiency of radial imaging to guide and monitor CMR interventions. Early explorations of real time imaging for interventions have included: (1) percutaneous intramyocardial application of contrast agents to track and supervise stem cell injections36; (2) safe automatic catheter tracking for real time CMR guided catheterization of the aorta, left ventricle, and carotid37; and (3) CMR guided coronary artery stent placement.38 Radial k space sampling techniques have increasingly been used for coronary artery CMR. As seen in Figure 3 6C, coronary artery CMR with a radial SSFP technique shows reduced motion artifacts and superior vessel sharpness com pared with the Cartesian approach.39 Scan efficiency for free breathing radial acquisitions of the heart can be improved through motion correction techniques40 or through the use of extended acquisition windows during the cardiac cycle (enabled by the motion insensitivity of radial scanning).41 The latter approach, in conjunction with intersegment motion correction using self guided epicardial fat tracking, holds the promise for rapid free breathing 3D coronary artery CMR with whole heart coverage.42 Four dimensional coro nary artery imaging has been realized using 3D stack of radial acquisitions across the entire cardiac cycle, with a phase sensitive SSFP pulse sequence for persistent fat suppression. The four dimensional approach allows multiple images at mid diastole to be averaged, thus enhancing SNR and vessel delineation.43 Inversion recovery preparation modules in conjunction with 3D radial k space sampling permit blood 42 Cardiovascular Magnetic Resonance

signal suppression in the coronaries and hence are suitable for vessel wall imaging at submillimeter in plane spatial resolution18 (see Fig. 3 3C). The benefits of radial k space sampling can be put to use in imaging large and small vessels. The diagnostic value of high isotropic spatial (1.25
1.25
1.25 mm3) and temporal (3 sec/frame) resolution VIPR imaging has been shown in time resolved contrast enhanced MR angiography (MRA) of the distal extremity.44 Further VIPR applications include high spatial resolution multi station MRA encom passing the abdomen, thigh, and calf.45 To extend the superior inferior coverage of peripheral MRA, 3D VIPR acquisitions can be combined with continuous table motion.46 Radial 3D SSFP imaging combined with a slab selective inversion prepulse affords flow targeted MRA without con trast medium application. This approach provides images well suited for vessel geometry assessment and reliable ste nosis detection in renal arteries (see Fig. 3 6D).47 A signifi cant improvement in motion artifact suppression, vessel sharpness, and detectable vessel length was found for spin labeling coronary MRA with SSFP and radial k space sampling.48 Recent work with interleaved and weighted radial imag ing has enabled images with multiple contrasts to be obtained from a single dataset. These 2D and 3D methods enable a radial trajectory to be used in conjunction with preparation pulses for detection of myocardial infarction and viability assessment using late gadolinium enhance ment (LGE) CMR.49

PARALLEL IMAGING Principles One point to note about all MR data acquisition trajectories discussed so far, Cartesian or non Cartesian, is that they are inherently sequential. One point and one line of data are acquired at a time. For sequential acquisitions, increased imaging speed is generally accomplished by reducing the time delay between acquired points, or else by acquiring fewer points (undersampling) and tolerating the level of artifact that may result. Unfortunately, there are limits to how quickly sequential data points may be acquired, because field gradients must be switched or RF pulses applied to move from data point to data point. Current technologic and physiologic limits on gradient switching rate and RF power deposition, then, constrain sequential MRI speed. Within these constraints, undersampling is the only remaining prospect for acceleration, but there are also limits to the level of uncorrected undersampling arti fact that may be tolerated, even for radial trajectories. Parallel imaging achieves its speed by acquiring more than one data point at a time. How it does this may be understood by analogy to another speed enhancing imag ing approach that employs parallelism, namely, multi detector row CT (MDCT). (Both parallel MRI and MDCT have served as potent enablers of cardiovascular imaging in recent times, rendering this analogy a particularly apt one for a consideration of CMR techniques.) Figure 3 7 compares MDCT and parallel MRI, with a schematic view

3 ADVANCED CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES: SPIRAL, RADIAL, AND PARALLEL IMAGING

A

B

C

D Figure 3 6 Montage of cardiovascular magnetic resonance (CMR) applications of radial imaging. A, Short axis electrocardiogram gated (ECG) two dimensional (2D) steady state free precession (SSFP) cine images acquired with radial (left) and traditional Cartesian (right) k space sampling using a 32 element cardiac coil array and a nominal in plane spatial resolution of 1.2
1.2 mm2. Radial acquisition provided an enhanced delineation of the endo and epicardial borders because of its resilience to undersampling, which allowed shorter acquisition windows in each cardiac phase without increasing the breath hold duration. B, Short axis 2D SSFP cine CMR using radial k space trajectories and a frame rate of 20 frames/sec (left) and 10 frames/sec (right). Data were acquired using a 32 element cardiac coil array and a nominal in plane spatial resolution of 2.2
2.2 mm2 (left) or 1.4
1.4 mm2 (right). C, CMR of the left coronary artery system obtained with a free breathing, navigated, three dimensional SSFP technique using radial (left) and Cartesian (right) k space trajectories. An in plane spatial resolution of 1.0
1.0 mm2 was used for the acquisition of the radial and Cartesian datasets. The reduced motion sensitivity of radial imaging results in improved vessel delineation. D, Magnetic resonance angiograms of the renal arteries using radial (left) and Cartesian (right) k space sampling schemes. Signal from the renal parenchyma and veins is completely suppressed by the inversion prepulse used for arterial spin labeling, whereas high contrast visualization of the renal arteries, including the distal subsegmental branches, is enabled. Motion artifacts are reduced in radial k space sampling, resulting in an improved vessel geometry assessment and improved detection of stenoses in the renal arteries. (C, Images courtesy of Rene Botnar, PhD, Guy’s and St. Thomas’ Hospital, London, United Kingdom; D, images courtesy of Marcus Katoh, PhD, RWTH Aachen University, Germany.) Cardiovascular Magnetic Resonance 43

Multiple image slices at once

Parallel MRI

Multiple k-space “slices” at once

RF coils

Rotations y ra Xitt em

Gradient steps

er

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

MDCT

Imaged subject Imaged subject Coil 1

Coil 2

Coil 3 k=0 k=2

Detector rows

k=4

Figure 3 7 Comparison of parallel acquisition approaches in multi detector row computed tomography (MDCT; left) and parallel magnetic resonance imaging (parallel MRI; right). Components contributing sequential image information are generally labeled in reddish orange, whereas parallel components are labeled in pale yellow. The top row shows a schematic side view of an MDCT scanner and a parallel MRI scanner, respectively. In MDCT, multiple image slices are acquired at once using adjacent rows of distinct X ray detectors (pale yellow rectangles below the subject). In parallel MRI, multiple radiofrequency (RF) coils (pale yellow rectangles shown in two representative orientations above or below the subject) effectively enable multiple k space “slices” to be acquired at once. The bottom half shows schematic front views and compares the types of projections of the imaged subject generated using MDCT or parallel MRI. For MDCT, an X ray emitter is rotated on a gantry, and cone like projections of the object are recorded in all detector rows from a variety of projection angles. An image of the subject is then reconstructed from this set of projections. In parallel MRI, each measured signal point is an integration or a projection of the imaged volume against the joint spatial modulations produced by field gradients and RF coils. Use of an array of RF coils provides multiple distinct modulations for each gradient step, thereby increasing the number of generalized projections available for image reconstruction and allowing images to be reconstructed from a reduced number of gradient steps. Sample generalized projection functions with complex spatial variation are shown for three coils and three gradient settings or k space indices. Generalized parallel MRI reconstruction algorithms reconstruct image intensities from the MRI signal data using knowledge of the set of projection and MR functions. The gantry rotations and gradient steps are the sequential components of data acquisition for computed tomography and CMR, respectively, and the detector rows and RF coils represent the corresponding parallel components. (Courtesy of the National Library of Medicine; available at http://www.nlm.nih.gov/copyright.html.)

of each modality shown at the top and a more detailed jux taposition of spatial encoding mechanisms shown at the bottom. Both modalities use arrays of detectors to acceler ate imaging beyond previous sequential limits, and both combine multi detector acquisition (pale yellow coloring in the figure) with more traditional sequential acquisition strategies (indicated by darker orange shading). In MDCT, multiple sequential projections are gathered by gantry rota tion, with additional simultaneous projection information provided by multiple detector rows with coverage of dis tinct image slices. The ability to acquire multiple image 44 Cardiovascular Magnetic Resonance

slices at once enables coverage of a target imaging volume in a fraction of the time that would be required in the pres ence of a single detector row. In parallel MRI, gradients generate sequential information in the form of projections against oscillating functions (the usual Fourier projections that make up the acquired k space matrix). Additional simultaneous information is available from multiple coil array elements, each of whose sensitivity patterns provides a distinct projection function. The extra simultaneous projections from multiple RF coils effectively provide infor mation about multiple k space “slices” at once. This allows

image contents may be determined, once again with a reduced number of sequential gradient steps. From another point of view, parallel MRI returns to the theme of undersampling introduced earlier in the context of radial imaging, and aims to fill in missing data in under sampled acquisitions with information derived from multi ple RF coils. Within this general rubric, a number of particular strategies for parallel MRI have evolved over time. The basic approaches to data acquisition and image reconstruction for several common parallel imaging strate gies are seen in Figure 3 8. First, undersampled data from an imaging sequence and k space trajectory of choice are acquired simultaneously in the multiple elements of a coil array. For Cartesian trajectories, this generally involves the omission of phase encoding gradient steps in a regular pat tern, with only one out of every R lines acquired, where R is the desired acceleration factor. The k space lines corresponding to omitted gradient steps are of course miss ing from the data matrix. In Figure 3 8, a case with R ¼ 2 is illustrated, with odd k space lines (solid red) acquired and even k space lines (dashed red) omitted. As shown on the right side of Figure 3 8, the missing k space data may then

Acquisition

Reconstruction

FFT

SMASH, GRAPPA, etc.

SE

NS

E,

GE

M,

SENSE, ASSET, etc.

etc

.

FFT

GRAPPA, PARS, etc.

c.

, et

EM

E, G

NS

SE

Figure 3 8 Data acquisition and image reconstruction strategies for parallel magnetic resonance imaging (MRI). Parallel acquisition may be used with most existing pulse sequences and k space trajectories. Undersampled data are acquired simultaneously in multiple elements of a radiofrequency (RF) coil array. Each element of a three element array is shown at the top left, along with corresponding coil sensitivities. Undersampled Cartesian, spiral, and radial trajectories are shown below each array element, with solid black lines indicating acquired data and dashed gray lines indicating omitted data. Various approaches to reconstruction of the multi coil undersampled data have been described, and a representative subset is indicated with arrows on the right. Simultaneous acquisition of spatial harmonics (SMASH), generalized autocalibrating partially parallel acquisition (GRAPPA), and similar techniques form fully sampled k space matrices from the multiple undersampled datasets, after which fast Fourier transform (FFT) yields the final image. Cartesian sensitivity encoding (SENSE), array spatial sensitivity encoding technique (ASSET), and so forth perform the FFT first, then operate to unfold aliased component coil images. GRAPPA, parallel imaging with augmented radius in k space (PARS), and other techniques may be used to regrid arbitrary undersampled non Cartesian trajectories onto fully sampled grids before FFT. Alternatively, generalized SENSE, generalized encoding matrix (GEM), or kindred techniques transform arbitrary undersampled datasets directly to a final image. The choice of reconstruction technique in practice will be dictated by the particular imaging situation and by the algorithms available on a particular scanner platform. Cardiovascular Magnetic Resonance 45

3 ADVANCED CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES: SPIRAL, RADIAL, AND PARALLEL IMAGING

the number of time consuming gradient steps to be reduced, and image acquisition can proceed in a fraction of the time that would have been required otherwise. One notable difference between parallel MRI and MDCT is the comparatively broad sensitivity profiles of RF coils as opposed to the collimated views provided by adjacent rows of X ray detectors. The RF coil sensitivities fall off smoothly with distance from each coil’s center, such that sensitivities for adjacent coils typically have a significant degree of over lap. (Note that three sample coil sensitivity patterns shown at the right of Fig. 3 7 have been separated for clarity. If they were overlaid on the same field of view, only their peaks would be clearly distinguishable.) This overlap must be accounted for in parallel image reconstructions. One general means of accounting for overlap is to take a cue, once again, from X ray CT and to formulate the problem as a generalized reconstruction from projections. When the coil related sensi tivity functions and the gradient related Fourier functions are combined, as shown at the bottom right of Figure 3 7, they form rather odd looking projection functions. These known projections may be collected into a matrix, and through suit able matrix inversion or other algebraic techniques, internal

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

be filled in using appropriately weighted combinations of acquired data from multiple coils, with component coil weightings chosen based on the known sensitivity patterns of the coils. This is the basis of the simultaneous acquisi tion of spatial harmonics (SMASH) technique.50 SMASH was the first parallel imaging approach demonstrated in vivo, although various proposals for simultaneous acqui sition had been made earlier,51–57 dating back to the time of the first uses of coil arrays for MRI. Various improve ments and generalizations of k space based parallel image reconstruction strategies have since been described and are somewhat vendor specific. Various improvements and generalizations of k space based parallel image reconstruc tion strategies have since been described. The most com monly used today is the GeneRalized Autocalibrating Partially Parallel Acquisition (GRAPPA) technique.58 Com mercial implementations of parallel imaging have kept pace with these developments; for example, Siemens scanners (Siemens Medical Solutions, Erlangen, Germany) use a commercial adaptation of the GRAPPA technique. The far right side of Figure 3 8 shows a set of component coil images obtained by Fourier transformation of regularly undersampled Cartesian datasets. Regular undersampling in this case results in well defined aliasing in the resulting images (i.e., the left side of the target field of view in each image is folded back around to the right, and vice versa). For any one coil’s image, this aliasing is unavoidable, but when multiple differently aliased images are available, knowledge of each coil’s distinct “view” may be used to undo aliasing artifacts. The Cartesian sensitivity encoding (SENSE) technique59 takes this unfolding approach. Philips scanners (Philips Medical Systems, Best, The Netherlands) use the SENSE algorithm for parallel MRI studies. General Electric systems (General Electric Healthcare, Waukesha, WI) cur rently provide the array spatial sensitivity encoding tech nique (ASSET), which uses principles similar to those of SENSE. Other vendors have their own related offerings. For spiral acquisitions, undersampling generally involves the omission of spiral interleaves; for radial sequences, radial projections are omitted (see Fig. 3 8 bottom left). For these non Cartesian trajectories, undersampling does not result in regular and confined aliasing patterns, and as a result, a Cartesian SENSE unfolding approach is precluded. (This is also true for irregularly undersampled Cartesian trajectories, such as the dense centered trajectories commonly used for GRAPPA and other self calibrating parallel imaging techni ques.) Techniques such as GRAPPA or parallel imaging with augmented radius in k space (PARS)60 may be used to com bine undersampled non Cartesian datasets (or irregularly undersampled Cartesian datasets) into fully sampled Carte sian k space grids that may then be Fourier transformed to yield reconstructed images. Alternatively, generalized approaches with reconstruction from projections may be used to move directly from arbitrary undersampled datasets to final reconstructed images. A suitably generalized SENSE algorithm59,61 may be used for this purpose, as may other techniques, such as sensitivity profiles from an array of coils for encoding and reconstruction in parallel (SPACE RIP),62 the generalized encoding matrix (GEM) technique,63 generalized SMASH,64 and others. These generalized strate gies are typically more computationally intensive than their Cartesian counterparts, and they can result in prolonged image reconstruction times. However, efficient algorithms61 46 Cardiovascular Magnetic Resonance

or distributed computing technology65 may be used to achieve prompt reconstruction. Both before and since the advent of parallel MRI, RF coil arrays have seen extensive use for traditional sequential MRI, but for parallel MRI in particular, coil arrays are strictly required. Since the coil sensitivities share the bur den of spatial encoding in parallel MRI, the details of array design are also particularly important. For example, to be effective for parallel MRI, array elements must have suitably distinct sensitivity profiles along any direction to be tar geted for acceleration, in addition to having a satisfactory SNR over the volume of interest. Much has been written on the subject of array design, and we will not expound further. For an overview, the reader is referred to review articles.66 Various cardiac specialized arrays have been con structed, and some of these are currently available from vendors. One additional requirement that parallel imaging imposes on MR scanner technology is that multiple receiver channels must be available to capture independent data from different coils. Fortunately, the number of commer cially available receiver channels has grown steadily, moti vated in part by the fact that the maximum achievable parallel imaging acceleration is equal to the number of independent array elements, and hence the number of independent receiver channels. Two practical considerations for parallel imaging studies are important to consider. First, once a suitable array is selected, calibration of the component coil sensitivity profiles is required. Second, the acceleration achieved using parallel MRI involves particular SNR trade offs relating both to coil array geometry and to image reconstruction (Fig. 3 9). To perform the sensitivity encoded parallel image recon struction, one needs to know the various “views” provided by the array elements. Various calibration strategies have been described, most involving separate acquisition of in vivo refer ence data in the desired image plane or in an encompassing volume (see Fig. 3 9A). With reference data in hand, varying degrees of image processing may be performed to extract pure sensitivity data59 or to eliminate the contributions of image features shared among all of the coils.63,67 The calibration step does add somewhat to the overall examination time, although rapid low resolution 3D acquisition may often suf fice for an entire series of accelerated acquisitions. However, motion of the patient or the coil array between the time of cal ibration and the time of an accelerated acquisition can result in calibration errors and image artifacts. This is a particular concern for CMR, for which flexible arrays contoured to the mobile chest wall are commonly used. To address this prob lem, self calibrating parallel imaging strategies may be used58,68–70 (see Fig. 3 9B). Self calibrating approaches incor porate a sensitivity reference directly into the accelerated acquisition (in the form of a small set of additional fully sam pled central k space lines). Therefore, they allow accurate par allel imaging even in the presence of vigorous motion. Among the commonly used techniques currently available from ven dors, GRAPPA58 and a modified form of SENSE (mSENSE) are self calibrating,71 and generalized approaches such as the generalized encoding matrix technique may also be used in a self calibrating fashion.70 Even with perfectly calibrated sensitivities, the SNR of a parallel imaging study is always reduced compared with an unaccelerated study obtained using the same coil array.59,72 This is a result of both the data acquisition strategies and

SNRaccelerated ¼

SNRunaccelerated p g R

(1)

Here, R represents the acceleration factor and the square root dependence in Equation 1 reflects the reduced SNR aver aging resulting from a reduced number of acquired data points in an accelerated scan. The “geometry factor,” g, on the other hand, is an additional SNR reduction factor charac terizing the extent to which the linear combinations used in a parallel image reconstruction amplify noise out of proportion to signal. This noise amplification depends sensitively on the choice of image reconstruction algorithm and on coil array design. The value of g varies spatially across the image plane (see Fig. 3 9C), meaning that the noise background is gener ally nonuniform in a parallel imaging study, and this should be taken into account in image interpretation.

Coil sensitivity based parallel imaging strategies may also be supplemented by techniques that operate in the temporal domain. CMR often involves dynamic imaging so that the availability of multiple time frames affords one the opportunity to vary acquisition trajectories as a func tion of time with otherwise identical acquisition parameters and to take advantage of resulting spatiotemporal correla tions. This is the concept behind techniques such as Un aliasing by Fourier encoding the OverLaps using the temporal Dimension (UNFOLD)73 and Broad use Linear Acquisition Speed up Technique (k t BLAST),74 which have been used to achieve accelerations of dynamic imag ing without the need for coil arrays and without the g factor related noise amplification penalty. The k t BLAST concept allows a more general temporal ordering of data acquisition and uses spatiotemporal correlations measured from sequential or interleaved training data to reassemble image components that are distributed in time and space.74,75 Explicit combinations of spatial information

Low-resolution sensitivity calibration volume g (R = 1)

10 5 0

Parallel image reconstruction

A

g (R = 2)

10

Separate accelerated acquisition

5 0

g (R = 3)

10 5 0

g (R = 4)

10 5 0

Parallel image reconstruction Single variable-density

B acquisition

C

Figure 3 9 Coil sensitivity calibration strategies and noise propagation behavior for parallel imaging. A, External calibration. A low resolution calibration data volume is acquired separately from the accelerated scan and is processed to yield sensitivity estimates in the target image plane. These sensitivities are then used for parallel image reconstruction. B, Self calibration. Additional calibration lines (gray) are acquired along with the undersampled accelerated datasets. These calibration lines are generally placed so as to produce a fully sampled region in the center of k space. The fully sampled region for each component coil’s dataset may then be Fourier transformed or otherwise processed to yield the sensitivity information necessary for parallel image reconstruction. C, Noise amplification with increasing acceleration. A four element array is shown at the top with its component coil sensitivities. Surface plots of g factor for acceleration factors R 1 to R 4 using this array are shown on the left of the simulated images. Peaks in the g factor map correspond to areas of increased noise and hence decreased signal to noise ratio in the images. The particular location and shape of noise peaks and valleys are influenced by the choice of coil array geometry, image plane, and acceleration factor. (Reproduced with permission from Sodickson DK. Parallel imaging methods. In: Edelman RR, Hesselink JR, Zlatkin MB, eds. Clinical Magnetic Resonance Imaging. 3rd ed. Philadelphia: Saunders; 2005.) Cardiovascular Magnetic Resonance 47

3 ADVANCED CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES: SPIRAL, RADIAL, AND PARALLEL IMAGING

the image reconstruction algorithms used in parallel MRI. The scaling of SNR may be expressed as follows:

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

from coil arrays with temporal information include k t SENSE,76 UNFOLD SENSE,77 and TSENSE,78 which increase net accelerations for dynamic applications.79–81 Parallel imaging artifacts can result from errors in any of the preparation stages of a parallel imaging study, from

equipment malfunctions, or from “intrinsic” causes, such as motion of the patient or coil arrays. Depending on the acquisition and reconstruction strategy used, these artifacts may be manifested as residual aliasing, increased noise amplification, or diminished temporal fidelity (Fig. 3 10).

A

B

C

D

E

F

Figure 3 10 See legend on next page 48 Cardiovascular Magnetic Resonance

Applications Parallel MRI may be used to improve a wide range of exist ing imaging studies as well as to enable new ones previ ously precluded by constraints on scan time. Today, commercial parallel imaging implementations are available for all modern MR scanners, and parallel imaging acquisi tions are used routinely in a substantial fraction of all CMR examinations. The increased speed and efficiency associated with parallel MRI may be translated into the following:  Shorter examinations  Improved spatial resolution and anatomic coverage  Improved temporal resolution  Enhanced image quality  Relaxed physiologic constraints (e.g., RF power deposi tion, peripheral nerve stimulation, acoustic noise) or physical constraints (e.g., gradient switching rate, dB/dt) Several general areas in which cardiovascular parallel imaging has been applied so far are discussed.

Imaging of Cardiac Anatomy and Structure Imaging of cardiac anatomy and structure using fast spin echo based techniques benefits from parallel imaging, which helps to limit relaxation related blurring by allowing reduced echo train lengths. Parallel imaging also enables anatomic imaging and tissue characterization simulta neously via the acquisition of proton density (short echo time) and T2 weighted (long echo time) images in a single

breath hold, thereby reducing the risk of image misregistra tion compared with the conventional approach.82 Mean while, improved image quality is enabled by the synergy between parallel imaging and the improved baseline SNR available at 3 T as opposed to 1.5 T83 (Fig. 3 11A). Perhaps the greatest benefit of parallel imaging for fast spin echo imaging, particularly at higher field strengths, is the capa bility to reduce the total power deposition by omitting phase encoding steps and corresponding RF refocusing pulses, which can be supplemented by the application of variable flip angles and hyperechoes.84–86 The use of high acceleration factors enabled by many element coil arrays87,88 promises to allow breath hold 3D black blood imaging with whole heart coverage, an approach that would eliminate the risk of 2D slice misregistration.

Assessment of Global and Regional Cardiac Function Assessment of global and regional cardiac function requires high muscle blood contrast, full R R coverage, high tempo ral resolution, and short scan times. Parallel imaging strate gies serve to improve cardiac functional assessments in various ways:  Reduction in the total acquisition time in segmented cine CMR,89 which increases patient comfort and diminishes respiratory artifacts. An acceleration factor of 8 reduces the number of heartbeats required for sin gle slice imaging from 16 to only 2 (see Fig. 3 11B), which permits single breath hold 2D cine CMR with apex to base coverage using multiple slices.  Increase in the number of cardiac phases per heartbeat for enhanced temporal resolution to identify more pre cisely the exact time point of maximal systolic contrac tion and diastolic filling.89,90  Minimization of motion sensitivity through reduction of the acquisition window duration, which supports the

Figure 3 10 Gallery of potential parallel imaging artifacts. A, Artifacts caused by excessive acceleration. As seen in Figure 3 9C, the overall magnitude and the inhomogeneity of the noise background increase with increasing acceleration, and noise amplification can become severe when acceleration factors approach the number of coils. The dotted white circle on the left highlights noise amplification for an acceleration factor of R 4 using a four element array. Noise amplification may be controlled by the use of a more moderate acceleration factor, as shown on the right (R 2 for the same four element array). B, Artifacts caused by inappropriate fields of view. Residual aliasing and relatively high noise near the center of the image (dotted white oval) occur if the imaged subject extends beyond the prescribed field of view (FOV) (left, FOV 24 cm). Image artifacts are not present in the accelerated scans if the target field of view is increased sufficiently (right, FOV 32 cm). (For both images, the same subregion is shown for clarity of comparison, despite the varying FOV and spatial resolution.) This behavior results from the fact that, in Cartesian sensitivity encoding (SENSE) based imaging, the overlap of structures in the target field of view leads to ambiguities in the partitioning of intensities among aliased positions, resulting in image artifacts. These artifacts can be removed by using UNaliasing by Fourier encoding the OverLaps using the temporal Dimension (UNFOLD) or other k t approaches. Some relaxation of the FOV constraint was recently reported to be possible for generalized autocalibrating partially parallel acquisition (GRAPPA) reconstructions.118 C, Artifacts resulting from coverage deficits in an external calibration scan. The accelerated short axis cardiac image on the left used a calibration scan that covers only a limited region marked by the dashed gray rectangle in the sagittal scout image above. This results in blank regions and severe aliasing artifacts highlighted by the dotted white circle. For comparison, the accelerated short axis image on the right used a reference scan that completely covers the prescribed scan plane, resulting in artifact free images. It is generally recommended to include in the calibration scan the entire region over which coils may have appreciable sensitivity. D, Artifacts caused by coil placement errors. Coil arrays used for cardiovascular parallel imaging should be placed so as to bring the target anatomy within the focus of the array’s sensitivity pattern. Examples of inappropriate (left) and appropriate (right) coil positioning are shown, with target position indicated by solid lines and actual position indicated by dashed lines in the sagittal scout images. For the severe offset shown on the left, reconstructed images are noisy and exhibit considerable uncorrected aliasing, as highlighted by the dotted white circle. E, Artifacts resulting from patient and coil array motion (indicated by different chest wall positions in the axial scout images). Any mismatch between the measured coil sensitivities and the actual sensitivities active during an accelerated scan can result in residual aliasing artifacts (left, dotted circle) that disappear if the measured coil sensitivities match the actual sensitivities (right). Self calibrating approaches can limit the incidence and severity of these artifacts. F, Artifacts caused by reduced temporal fidelity in accelerated dynamic imaging using spatiotemporal correlations obtained from low spatial resolution training data. The temporal fidelity of images reconstructed from accelerated data using k t approaches decreases with an increasing acceleration factor. Consequently, image blurring is pronounced and tissue border sharpness is reduced in the faster image on the left as opposed to the slower image on the right. The time course over the full R R interval of a single one dimensional projection along the dotted line in each image also shows evidence of this blurring (dotted white oval), which has the potential to affect the quantitative assessment of dynamic image data. Cardiovascular Magnetic Resonance 49

3 ADVANCED CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES: SPIRAL, RADIAL, AND PARALLEL IMAGING

However, by accelerating the acquisition, the use of parallel imaging can reduce the incidence or severity of other com mon imaging artifacts, such as motional blurring or slice misregistration. This balance should be considered in the planning and interpretation of parallel CMR studies.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Anatomic Imaging Triple IR Fast Spin Echo

Double IR Fast Spin Echo Conventional

SENSE (R = 2)

Conventional

SENSE (R = 2)

A Cardiac function using 2D CINE + SENSE Conventional taoq = 16 R-R

SENSE (R = 4) taoq = 4 R-R

SENSE (R = 2) taoq = 8 R-R

SENSE (R = 8) taoq = 2 R-R

B Myocardial perfusion imaging with k-t BLAST Passage right ventricle

Base line

Passage left ventricle

Passage myocardiuim

C Delayed enhancement Conventional

Coronary artery MRA

SENSE (R = 2) + PSIR

D Figure 3 11 See legend on next page

50 Cardiovascular Magnetic Resonance

Breath hold

E

Free breathing









First-Pass Myocardial Perfusion Imaging Using parallel imaging, one or two heartbeat temporal res olution has been achieved in saturation recovery based techniques used to capture contrast agent kinetics for the assessment of myocardial perfusion.102–105 Moreover, the k t BLAST and k t SENSE approach can be used to double the spatial coverage per unit time while preserving

in plane spatial resolution. Alternatively, k t BLAST and k t SENSE can be put to use to double the in plane matrix size without impairing the temporal resolution (see Fig. 3 11C). The combined SNR and speed improvements associated with high field parallel imaging can be exploited to transition from multi slice 2D acquisitions to whole heart 3D acquisitions or to enable myocardial perfusion imaging with arterial spin labeling rather than exogenous contrast agent injection.106

Detection of Myocardial Infarction and Assessment of Myocardial Viability The established CMR assessment of ischemic heart disease includes LGE CMR using ECG gated, segmented imaging modules preceded by an inversion recovery preparation to provide consistent high contrast between infarcted and healthy myocardium.107 Accelerated LGE imaging is of clin ical importance because unaccelerated approaches have limited spatial coverage of only one to two slices per breath hold, resulting in prolonged examination times of 10 to 15 minutes, with corresponding decay of contrast agent concentration over the course of the examination. Parallel imaging can overcome these difficulties by allowing whole heart coverage in a single breath hold, ensuring uniform suppression of healthy myocardium for all imaged sections. Meanwhile, a phase sensitive reconstruction of inversion recovery technique has been shown to offset the need for perfect evolution time adjustments and to enhance the con trast between healthy and infarcted myocardial tissue.108 This approach doubles the total scan time compared with the conventional one R R interval approach because it requires two R R intervals for the acquisition of a T1 weighted inversion recovery dataset and an extra refer ence image. This trade off can be compensated by using the time savings inherent in parallel imaging to facilitate short breath hold times (see Fig. 3 11D).

Coronary Artery Cardiovascular Magnetic Resonance Parallel imaging strategies provide several means of improv ing coronary artery CMR by minimizing the effect of physi ologic motion. The use of parallel imaging strategies enables 3D free breathing navigator techniques to be accel erated (see Fig. 3 11E) while preserving image quality

Figure 3 11 Montage of parallel cardiovascular magnetic resonance (CMR) applications. A, Short axis cardiac images obtained with double (left) and triple (right) inversion recovery black blood fast spin echo imaging at 3.0 T using a conventional unaccelerated approach and a twofold accelerated parallel imaging approach. B, Short axis cine CMR acquired with an unaccelerated (R 1) conventional two dimensional (2D) steady state free precession (SSFP) approach (left) and 2D SSFP with parallel imaging using one dimensional acceleration factors up to R 8 (right). The increase in the acceleration factor, together with a constant number of phase encoding steps per cardiac cycle, resulted in a significant breath hold time reduction. The total scan time was 16 heartbeats for the unaccelerated approach and 2 heartbeats for R 8. Images were acquired using a 32 channel cardiac coil array. C, Selected short axis first pass perfusion images (spatial resolution of 2.0
2.0
8 mm3) derived from a dataset acquired at three slices per R R interval and reconstructed using Broad use Linear Acquisition Speed up Technique (k t BLAST) with fivefold acceleration. The one R R temporal resolution was used to determine the precontrast baseline (far left), to monitor contrast agent arrival in the right ventricle (center left), and to track the passage of the contrast agent through the left ventricle (center right) and the myocardium (far right). D, Short axis views obtained from late gadolinium enhancement (LGE) CMR at 3.0 T. For the conventional approach (left), one R R interval was used for recovery of the magnetization to achieve a breath hold duration of 12 seconds. The phase sensitive reconstruction of inversion recovery (PSIR) approach (right) required two R R intervals for full magnetization recovery, which was compensated by using twofold accelerated parallel imaging to keep the breath hold time at 12 seconds. E, Maximum intensity projection (3 mm slice thickness) of an image volume showing the right coronary artery obtained from accelerated (R 2) ECG gated, fat saturated, three dimensional SSFP acquisitions using a breath hold (left) and a free breathing, navigated approach (right). The effective scan time was halved in the twofold accelerated scan while preserving the image quality of the unaccelerated acquisition. SENSE, Cartesian sensitivity encoding. Cardiovascular Magnetic Resonance 51

3 ADVANCED CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES: SPIRAL, RADIAL, AND PARALLEL IMAGING



visualisation of small, rapidly moving structures, such as cardiac valves.89 Improvement of anatomic coverage without increasing total acquisition time. For example, rapid segmented 3D techniques have been shown to be capable of scan ning the entire heart in a single breath hold.91,92 Acceleration of real time imaging methods for the assess ment of cardiac function.93,94 In addition to eliminating the need for ECG gating, accelerated real time techniques can also be used to track cardiac motion in a free breathing mode. The effect of motion in such studies may be mini mized through the use of short acquisition windows and the application of self calibration for coil sensitivity mapping.95 Improvement of myocardial tagging methods for the assessment of regional myocardial wall motion.96 The potential to suspend the need for breath holding facili tates the monitoring of transient hemodynamics and car diac mechanics97 and affords the use of true 3D tagging grids for the accurate determination of quantitative 3D motion patterns.98 Acceleration of phase contrast CMR by reducing the scan time for quantitative flow measurements in the great arteries. This approach supports the detection of congenital heart disease99 and the assessment of flow abnormalities.100 Increase in net accelerations using temporal undersam pling strategies. The combination of UNFOLD73 with SENSE has been used to achieve higher acceleration fac tors for a given number of receiver coils or channels than would otherwise be feasible with SENSE alone.77,81 TSENSE with an acceleration factor of R ¼ 6 (3
2) has facilitated single breath hold 3D cine CMR with whole heart coverage at a temporal resolution of 50 msec.101 The k t BLAST and k t SENSE74 approach enables high accelerations, which support single breath hold multi slice and whole heart coverage cine acquisitions without prohibitive noise amplification.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

MRA using HighlY constrained back PRojection (HYPR)

A

From 2D to 3D single breath hold imaging

B Rapid whole heart coverage imaging + reformatting

C

From small volume to moving table whole body imaging

D Figure 3 12 See legend on next page

52 Cardiovascular Magnetic Resonance

FUTURE DIRECTIONS The topic of future directions and the question of what techniques will be considered “advanced” in times to come are tricky topics for a textbook, which can run the risk of

promptly appearing out of date when printed predictions turn out to be off the mark. In a field that evolves as rapidly as MRI has continued to do since its inception, however, it is only prudent to turn our attention briefly to the horizon. One well established precedent drawn from the several decade history of MRI is that imaging speed is likely to con tinue to increase. Figure 3 12 shows four examples of a current trend toward very high accelerations that may well have a significant effect on CMR imaging in the future. In Figure 3 12A, a simulated 50 fold acceleration is shown for an MRA using the highly constrained back projection (HYPR) technique.114 On the left is the reference image obtained with a 400 projection 2D radial acquisition with a standard FBP reconstruction. FBP reconstruction of a highly undersampled trajectory with only eight projections is shown in the center, and the corresponding HYPR recon struction is shown on the right. Clearly, the use of eight projections is insufficient for traditional FBP reconstruc tion, and the image is uninterpretable. However, HYPR incorporates some principles of the k t techniques dis cussed earlier, using information from a time series of undersampled radial acquisitions to constrain each time frame. As the fidelity of the HYPR image suggests, extraor dinary levels of acceleration may be possible, as long as conditions appropriate for HYPR exist, namely, that tempo ral information is available and the field of view is occupied sparsely by narrow, high contrast vessels. Combinations of HYPR principles with 3D VIPR trajectories promise further accelerations for such sparsely populated fields of view. Figure 3 12B and C shows the levels of acceleration that are currently possible and the CMR applications that are thereby enabled using parallel MRI in the absence of these specialized conditions (e.g., for more densely populated fields of view). Both are examples of highly parallel CMR and take advantage of available 32 channel systems and 32 element arrays. Figure 3 12B shows accelerated LGE CMR obtained at 3.0 T. A fourfold acceleration allowed the transition from multiple 2D acquisitions encompassing only one to two slices per breath hold to single breath hold acquisitions with whole heart coverage. Figure 3 12C shows various views from 3D SSFP CMR datasets obtained with eightfold acceleration.87 The high level of acceleration in this case allowed acquisition of a comprehensive axial volume within a single breath hold, enabling visualization

Figure 3 12 Prospects for highly accelerated cardiovascular magnetic resonance (CMR) using undersampled projection reconstruction or parallel MRI. A, Simulated 50 fold acceleration of a magnetic resonance angiogram (MRA) using the highly constrained back projection (HYPR) technique,114 which uses information from a composite image to constrain the reconstruction of each frame in a time series (in this case, containing 16 frames). A reference 400 projection reconstruction image (left) is juxtaposed to an 8 projection projection reconstruction image reconstructed with a traditional filtered back projection algorithm (center) and an 8 projection HYPR image (right). Unlike filtered back projection, HYPR clearly depicts vascular anatomy despite the high level of undersampling. B, Accelerated parallel MRI for late gadolinium enhancement (LGE) at 3.0 T. Acceleration factors of R 4 enabled three dimensional (3D) whole heart coverage in a single breath hold, thereby permitting uniform suppression of the healthy myocardium and preventing slice misregistration. C, Highly accelerated parallel CMR for single breath hold whole heart coronary artery CMR.115 A 32 element array (the top half of which is shown schematically on the left) was used to achieve eightfold (4
2) accelerations for an axial 3D steady state free precession (SSFP) image volume (256
256
60 matrix size, 1.5
1.5
2 mm3 spatial resolution). Several axial images from several sources are shown at center left. Various reformatted views are also shown, including traditional long axis and short axis views (center) as well as depictions of the left and right coronary arteries (center right). The 3D volume rendered view of the right coronary artery (RCA) shown on the right was obtained retrospectively from an automatic segmentation algorithm and attests to the quality of the original individual images as well as the level of contrast achieved with the highly accelerated acquisition. This full complement of information was available after a single 25 second breath hold scan with simple axial planning. D, Schematic depiction of the progression from thin section imaging (left) to rapid volumetric imaging (center) or even to whole body imaging incorporating automatic table motion (right). The general paradigm of highly accelerated imaging with simple axial planning and comprehensive volumetric coverage is highly reminiscent of multi detector row computed tomography (MDCT), but incorporates various advantages associated with MRI. 2D, two dimensional. (A, Images courtesy of Charles Mistretta, University of Wisconsin, Madison.)

Cardiovascular Magnetic Resonance 53

3 ADVANCED CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES: SPIRAL, RADIAL, AND PARALLEL IMAGING

comparable to that of the unaccelerated approach.109 This decreases the total scan time and reduces the susceptibility to patient motion and respiratory drift. Short breath hold 3D coronary artery CMR is challenging because of the com peting constraints of spatial resolution, anatomic coverage, and breath hold time. The efficiency advantage of parallel imaging strategies minimizes the effect of cardiac and respi ratory motion and facilitates shorter data acquisition win dows within the cardiac cycle.110 This is especially beneficial at high heart rates, for which shortening of the mid diastolic cardiac rest period limits the usable acquisi tion window. Parallel imaging also allows reductions in the number of cardiac cycles used for in plane segmenta tion. This leads to very short breath hold times for each coronary artery acquisition and is especially suited for low heart rates. For either free breathing or breath hold approaches, the time savings associated with parallel imaging can be trans lated into an improvement of in plane and through plane spatial resolution, which results in an improved delineation of proximal and especially distal segments of the coronary arteries.109 Conventional coronary artery CMR studies are generally restricted to targeted thin slabs encompassing a particular segment of the coronary artery tree only. Parallel imaging allows the use of a thicker volume, which supports the visualization of long tortuous segments of the coronary arteries and offers the potential to eliminate localization scans.111,112 The T1 prolongation and SNR gain at high magnetic field strengths, together with the speed benefit of parallel imaging, will prove to be useful for flow targeted imaging. The SNR improvements promise to be beneficial not only for MR lumography but also for vessel wall imag ing.113 As accelerated high spatial resolution vessel wall imaging is accomplished at 3.0 T, intrinsic contrast mechanisms and specific contrast agents can be used for plaque detection and plaque characterization.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

of the coronary arteries or any other typical cardiac view by subsequent reformatting. The same scanner and array have also been used for accelerated real time cardiac imaging115 and for rapid volumetric contrast enhanced MRA at 12 to 16 fold accelerations.88 The trend toward still larger num bers of receiver channels and array elements may only be expected to continue. Acceleration levels may not necessarily increase arbitrarily as channel counts increase because SNR losses are expected to mount rapidly as a function of acceler ation caused by fundamental electrodynamic constraints on the shapes of RF coil sensitivities.116,117 However, with suffi cient increases in baseline SNR, through the use of high field MR systems or new hyperpolarized contrast agents, further accelerations may be expected. Of course, one may also com bine parallel imaging at more modest levels with highly undersampled radial acquisitions to further push the limits of imaging speed. The examples in Figure 3 12B and C (as well as Fig. 3 12A if HYPR is used with 3D trajectories) represent efforts at rapid volumetric imaging, another significant trend. As imaging speed has increased over time, larger and larger imaging volumes have become accessible at any given spatial and temporal resolution. This is, of course, an appealing prospect for those interested in

comprehensive surveys of, for example, cardiac or vascular anatomy. It is also a trend that is not restricted to CMR. Rapid advances in MDCT are allowing ever more rapid vol umetric scans, which have been viewed increasingly as a challenge to CMR. As we have seen, however, CMR is capa ble of similarly rapid volumetric acquisitions. Indeed, rapid volumetric imaging enables simplified axial planning and a comprehensive volumetric approach (see Fig. 3 12C), which are reminiscent of MDCT. All that remains to com plete the analogy is to combine highly parallel MRI with existing moving table approaches to enable true whole body coverage, a progression that is shown in Figure 3 12D. Of course, comprehensive volumetric CMR boasts a number of advantages over its kindred tomographic modal ity. By taking advantage of efficient k space trajectories and highly parallel imaging, MRI can provide some of the speed and simplicity of MDCT, while eliminating concerns about radiation dose and maintaining the tissue specificity and biochemical selectivity of the MR phenomenon. For the optimistically inclined practitioner, it should not be diffi cult to imagine the comprehensive cardiovascular examina tion that has long been a holy grail of CMR occupying a scant few minutes. In short, the future of today’s advanced CMR techniques remains promising.

References 1. Ahn BC, Kim JH, Cho ZH. High-speed spiral-scan echo planar NMR imaging. IEEE Trans Med Imaging. 1986;MI-5:2–7. 2. Meyer CH, Hu BS, Nishimura DG, Macovski A. Fast spiral coronary artery imaging. Magn Reson Med. 1992;28:202–213. 3. O’Sullivan JD. A fast sinc function gridding algorithm for Fourier inversion in computer tomography. IEEE Trans Med Imaging. 1985;4:200–207. 4. Jackson JI, Meyer CH, Nishimura DG, Macovski A. Selection of a convolution function for Fourier inversion using gridding. IEEE Trans Med Imaging. 1991;10:473–478. 5. Kerr AB, Pauly JM, Hu BS, et al. Real-time interactive MRI on a conventional scanner. Magn Reson Med. 1997;38:355–367. 6. Noll DC, Pauly JM, Meyer CH, Nishimura DG, Macovski A. Deblurring for non-2D Fourier transform magnetic resonance imaging. Magn Reson Med. 1992;25:319–333. 7. Irarrazabal P, Meyer CH, Nishimura DG, Macovski A. Inhomogeneity correction using an estimated linear field map. Magn Reson Med. 1996;35:278–282. 8. Man LC, Pauly JM, Macovski A. Improved automatic off-resonance correction without a field map in spiral imaging. Magn Reson Med. 1997;37:906–913. 9. Meyer CH, Pauly JM, Macovski A, Nishimura DG. Simultaneous spatial and spectral selective excitation. Magn Reson Med. 1990;15: 287–304. 10. Block W, Pauly J, Nishimura D. RARE spiral T2-weighted imaging. Magn Reson Med. 1997;37:582–590. 11. Bornert P, Aldefeld B, Eggers H. Reversed spiral MR imaging. Magn Reson Med. 2000;44:479–484. 12. Irarrazabal P, Nishimura DG. Fast three dimensional magnetic resonance imaging. Magn Reson Med. 1995;33:656–662. 13. Bornert P, Stuber M, Botnar RM, et al. Direct comparison of 3D spiral vs. Cartesian gradient-echo coronary magnetic resonance angiography. Magn Reson Med. 2001;46:789–794. 14. Foo TK, Ho VB, Hood MN. Vessel tracking: prospective adjustment of section-selective MR angiographic locations for improved coronary artery visualization over the cardiac cycle. Radiology. 2000;214: 283–289. 15. Saranathan M, Ho VB, Hood MN, Foo TK, Hardy CJ. Adaptive vessel tracking: automated computation of vessel trajectories for improved efficiency in 2D coronary MR angiography. J Magn Reson Imaging. 2001;14:368–373.

54 Cardiovascular Magnetic Resonance

16. Sussman MS, Stainsby JA, Robert N, Merchant N, Wright GA. Variable-density adaptive imaging for high-resolution coronary artery MRI. Magn Reson Med. 2002;48:753–764. 17. Santos JM, Cunningham CH, Lustig M, et al. Single breath-hold whole-heart MRA using variable-density spirals at 3T. Magn Reson Med. 2006;55:371–379. 18. Katoh M, Spuentrup E, Buecker A, Manning WJ, Gunther RW, Botnar RM. MR coronary vessel wall imaging: comparison between radial and spiral k-space sampling. J Magn Reson Imaging. 2006;23: 757–762. 19. Kuhl HP, Spuentrup E, Wall A, et al. Assessment of myocardial function with interactive non-breath-hold real-time MR imaging: comparison with echocardiography and breath-hold cine MR imaging. Radiology. 2004;231:198–207. 20. Liao JR, Pauly JM, Brosnan TJ, Pelc NJ. Reduction of motion artifacts in cine MRI using variable-density spiral trajectories. Magn Reson Med. 1997;37:569–575. 21. Spuentrup E, Schroeder J, Mahnken AH, et al. Quantitative assessment of left ventricular function with interactive real-time spiral and radial MR imaging. Radiology. 2003;227:870–876. 22. Nayak KS, Hargreaves BA, Hu BS, Nishimura DG, Pauly JM, Meyer CH. Spiral balanced steady-state free precession cardiac imaging. Magn Reson Med. 2005;53:1468–1473. 23. Nayak KS, Cunningham CH, Santos JM, Pauly JM. Real-time cardiac MRI at 3 tesla. Magn Reson Med. 2004;51:655–660. 24. Nezafat R, Kellman P, Derbyshire JA, McVeigh ER. Real-time blood flow imaging using autocalibrated spiral sensitivity encoding. Magn Reson Med. 2005;54:1557–1561. 25. Lauterbur P. Image formation by induced local interactions: examples employing nuclear magnetic resonance. Nature. 1973;242:190–191. 26. Lauterbur P, Lai C-M. Zeugmatography by reconstruction from projections. IEEE Trans Nucl Sci. 1980;NS-27:1227–1231. 27. Glover GH, Pauly JM. Projection reconstruction techniques for reduction of motion effects in MRI. Magn Reson Med. 1992;28: 275–289. 28. Peters DC, Korosec FR, Grist TM, et al. Undersampled projection reconstruction applied to MR angiography. Magn Reson Med. 2000;43:91–101. 29. Peters DC, Epstein FH, McVeigh ER. Myocardial wall tagging with undersampled projection reconstruction. Magn Reson Med. 2001;45:562–567.

53. Kelton JR, Magin RL, Wright SM. An algorithm for rapid image acquisition using multiple receiver coils. In: Proceedings of the 8th Annual Meeting of the Society for Magnetic Resonance in Medicine. Amsterdam, Netherlands: 1989:1172. 54. Kwiat D, Einav S, Navon G. A decoupled coil detector array for fast image acquisition in magnetic resonance imaging. Med Phys. 1991;18:251–265. 55. Ra JB, Rim CY. Fast imaging using subencoding data sets from multiple detectors. Magn Reson Med. 1993;30:142–145. 56. Carlson JW, Minemura T. Imaging time reduction through multiple receiver coil data acquisition and image reconstruction. Magn Reson Med. 1993;29:681–687. 57. Kwiat D, Einav S. Preliminary experimental evaluation of an inverse source imaging procedure using a decoupled coil detector array in magnetic resonance imaging. Med Eng Phys. 1995;17:257–263. 58. Griswold MA, Jakob PM, Heidemann RM, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med. 2002;47:1202–1210. 59. Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: Sensitivity encoding for fast MRI. Magn Reson Med. 1999;42:952–962. 60. Yeh EN, McKenzie CA, Ohliger MA, Sodickson DK. Parallel magnetic resonance imaging with adaptive radius in k-space (PARS): constrained image reconstruction using k-space locality in radiofrequency coil encoded data. Magn Reson Med. 2005;53:1383–1392. 61. Pruessmann KP, Weiger M, Bornert P, Boesiger P. Advances in sensitivity encoding with arbitrary k-space trajectories. Magn Reson Med. 2001;46:638–651. 62. Kyriakos WE, Panych LP, Kacher DF, et al. Sensitivity profiles from an array of coils for encoding and reconstruction in parallel (SPACE RIP). Magn Reson Med. 2000;44:301–308. 63. Sodickson DK, McKenzie CA. A generalized approach to parallel magnetic resonance imaging. Med Phys. 2001;28:1629–1643. 64. Bydder M, Larkman DJ, Hajnal JV. Generalized SMASH imaging. Magn Reson Med. 2002;47:160–170. 65. Cohen S, Grant AK, Yeh EN, Joshi S, Sodickson DK. A parallel computing solution for rapid reconstruction of highly-accelerated volumetric parallel MRI Data. In: Proceedings of the 12th Scientific Meeting of the International Society for Magnetic Resonance in Medicine. Kyoto: 2004:2631. 66. Ohliger MA, Sodickson DK. An introduction to coil array design for parallel MRI. NMR Biomed. 2006;19:300–315. 67. Sodickson DK. Tailored SMASH image reconstructions for robust In vivo parallel MR imaging. Magn Reson Med. 2000;44:243–251. 68. Jakob PM, Griswold MA, Edelman RR, Sodickson DK. AUTOSMASH: a self-calibrating technique for SMASH imaging. MAGMA. 1998;7: 42–54. 69. Heidemann RM, Griswold MA, Haase A, Jakob PM. VD-AUTOSMASH imaging. Magn Reson Med. 2001;45:1066–1074. 70. McKenzie CA, Yeh EN, Ohliger MA, Price MD, Sodickson DK. Selfcalibrating parallel imaging with automatic coil sensitivity extraction. Magn Reson Med. 2002;47:529–538. 71. Hunold P, Maderwald S, Ladd ME, Jellus V, Barkhausen J. Parallel acquisition techniques in cardiac cine magnetic resonance imaging using TrueFISP sequences: comparison of image quality and artifacts. J Magn Reson Imaging. 2004;20:506–511. 72. Sodickson DK, Griswold MA, Jakob PM, Edelman RR, Manning WJ. Signal-to-noise ratio and signal-to-noise efficiency in SMASH imaging. Magn Reson Med. 1999;41:1009–1022. 73. Madore B, Glover GH, Pelc NJ. Unaliasing by fourier-encoding the overlaps using the temporal dimension (UNFOLD), applied to cardiac imaging and fMRI. Magn Reson Med. 1999;42:813–828. 74. Tsao J, Boesiger P, Pruessmann KP. k-t BLAST and k-t SENSE: dynamic MRI with high frame rate exploiting spatiotemporal correlations. Magn Reson Med. 2003;50:1031–1042. 75. Hansen MS, Kozerke S, Pruessmann KP, Boesiger P, Pedersen EM, Tsao J. On the influence of training data quality in k-t BLAST reconstruction. Magn Reson Med. 2004;52:1175–1183. 76. Tsao J. K-t SENSE for efficient dynamic parallel imaging by jointspace time consideration. In: Second International Workshop on Parallel MRI. Zurich, Switzerland: 2004:51. 77. Madore B. UNFOLD-SENSE: a parallel MRI method with selfcalibration and artifact suppression. Magn Reson Med. 2004;52: 310–320. 78. Kellman P, Epstein FH, McVeigh ER. Adaptive sensitivity encoding incorporating temporal filtering (TSENSE). Magn Reson Med. 2001;45:846–852.

Cardiovascular Magnetic Resonance 55

3 ADVANCED CARDIOVASCULAR MAGNETIC RESONANCE IMAGING TECHNIQUES: SPIRAL, RADIAL, AND PARALLEL IMAGING

30. Shankaranarayanan A, Simonetti OP, Laub G, Lewin JS, Duerk JL. Segmented k-space and real-time cardiac cine MR imaging with radial trajectories. Radiology. 2001;221:827–836. 31. Schaeffter T, Weiss S, Eggers H, Rasche V. Projection reconstruction balanced fast field echo for interactive real-time cardiac imaging. Magn Reson Med. 2001;46:1238–1241. 32. Larson AC, Simonetti OP. Real-time cardiac cine imaging with SPIDER: steady-state projection imaging with dynamic echo-train readout. Magn Reson Med. 2001;46:1059–1066. 33. Barger AV, Block WF, Toropov Y, Grist TM, Mistretta CA. Timeresolved contrast-enhanced imaging with isotropic resolution and broad coverage using an undersampled 3D projection trajectory. Magn Reson Med. 2002;48:297–305. 34. Larson AC, White RD, Laub G, McVeigh ER, Li D, Simonetti OP. Selfgated cardiac cine MRI. Magn Reson Med. 2004;51:93–102. 35. Larson AC, Kellman P, Arai A, et al. Preliminary investigation of respiratory self-gating for free-breathing segmented cine MRI. Magn Reson Med. 2005;53:159–168. 36. Krombach GA, Pfeffer JG, Kinzel S, Katoh M, Gunther RW, Buecker A. MR-guided percutaneous intramyocardial injection with an MRcompatible catheter: feasibility and changes in T1 values after injection of extracellular contrast medium in pigs. Radiology. 2005;235:487–494. 37. Weiss S, Kuehne T, Brinkert F, et al. In vivo safe catheter visualization and slice tracking using an optically detunable resonant marker. Magn Reson Med. 2004;52:860–868. 38. Spuentrup E, Ruebben A, Schaeffter T, Manning WJ, Gunther RW, Buecker A. Magnetic resonance–guided coronary artery stent placement in a swine model. Circulation. 2002;105:874–879. 39. Spuentrup E, Katoh M, Buecker A, et al. Free-breathing 3D steadystate free precession coronary MR angiography with radial k-space sampling: comparison with cartesian k-space sampling and cartesian gradient-echo coronary MR angiography–pilot study. Radiology. 2004;231:581–586. 40. McLeish K, Kozerke S, Crum WR, Hill DL. Free-breathing radial acquisitions of the heart. Magn Reson Med. 2004;52:1127–1135. 41. Stehning C, Bornert P, Nehrke K, Dossel O. Free breathing 3D balanced FFE coronary magnetic resonance angiography with prolonged cardiac acquisition windows and intra-RR motion correction. Magn Reson Med. 2005;53:719–723. 42. Stehning C, Bornert P, Nehrke K, Eggers H, Stuber M. Free-breathing whole-heart coronary MRA with 3D radial SSFP and self-navigated image reconstruction. Magn Reson Med. 2005;54:476–480. 43. Park J, Larson AC, Zhang Q, Simonetti O, Li D. 4D radial coronary artery imaging within a single breath-hold: cine angiography with phase-sensitive fat suppression (CAPS). Magn Reson Med. 2005;54: 833–840. 44. Du J, Carroll TJ, Brodsky E, et al. Contrast-enhanced peripheral magnetic resonance angiography using time-resolved vastly undersampled isotropic projection reconstruction. J Magn Reson Imaging. 2004;20:894–900. 45. Du J, Korosec FR, Thornton FJ, Grist TM, Mistretta CA. High-resolution multistation peripheral MR angiography using undersampled projection reconstruction imaging. Magn Reson Med. 2004;52:204–208. 46. Fain SB, Browning FJ, Polzin JA, et al. Floating table isotropic projection (FLIPR) acquisition: a time-resolved 3D method for extended field-of-view MRI during continuous table motion. Magn Reson Med. 2004;52:1093–1102. 47. Katoh M, Spuentrup E, Stuber M, Hoogeveen R, Gunther RW, Buecker A. Free-breathing renal magnetic resonance angiography with steady-state free-precession and slab-selective spin inversion combined with radial k-space sampling and water-selective excitation. Magn Reson Med. 2005;53:1228–1233. 48. Katoh M, Stuber M, Buecker A, Gunther RW, Spuentrup E. Spinlabeling coronary MR angiography with steady-state free precession and radial k-space sampling: initial results in healthy volunteers. Radiology. 2005;236:1047–1052. 49. Peters DC, Botnar RM, Kissinger KV, Yeon SB, Appelbaum EA, Manning WJ. Inversion recovery radial MRI with interleaved projection sets. Magn Reson Med. 2006;55:1150–1156. 50. Sodickson DK, Manning WJ. Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays. Magn Reson Med. 1997;38:591–603. 51. Carlson JW. An algorithm for NMR imaging reconstruction based on multiple RF receiver coils. J Magn Reson. 1987;74:376–380. 52. Hutchinson M, Raff U. Fast MRI data acquisition using multiple detectors. Magn Reson Med. 1988;6:87–91.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

79. Tsao J, Kozerke S, Boesiger P, Pruessmann KP. Optimizing spatiotemporal sampling for k-t BLAST and k-t SENSE: application to highresolution real-time cardiac steady-state free precession. Magn Reson Med. 2005;53:1372–1382. 80. Kozerke S, Tsao J, Razavi R, Boesiger P. Accelerating cardiac cine 3D imaging using k-t BLAST. Magn Reson Med. 2004;52:19–26. 81. Madore B. Using UNFOLD to remove artifacts in parallel imaging and in partial-Fourier imaging. Magn Reson Med. 2002;48:493–501. 82. Priatna A, Zhu D. Double/triple IR dual contrast FSE of the heart with ASSET. In: Proceedings of the 11th Scientific Meeting of the International Society of Magnetic Resonance in Medicine. Toronto, Ontario, Canada: 2003:1562. 83. Gutberlet M, Noeske R, Schwinge K, Freyhardt P, Felix R, Niendorf T. Comprehensive cardiovascular MR imaging at 3.0 Tesla: feasibility and implications for clinical applications. Invest Radiol. 2006;41. 84. Busse RF. Reduced RF power without blurring: correcting for modulation of refocusing flip angle in FSE sequences. Magn Reson Med. 2004;51:1031–1037. 85. Henning J, Scheffler K. Hyperechoes. Magn Reson Med. 2001;46: 6–12. 86. Hennig J, Weigel M, Thiel T. Optimizing SAR-reduction for high-field TSE with asymmetric hyperechoes combined with partial Fourier parallel imaging. In: Proceedings of the 12th Meeting of the International Society of Magnetic Resonance in Medicine. Kyoto, Japan: 2004:539. 87. Zhu Y, Hardy CJ, Sodickson DK, et al. Highly parallel volumetric imaging with a 32-element RF coil array. Magn Reson Med. 2004;52:869–877. 88. Sodickson DK, Hardy CJ, Zhu Y, et al. Rapid volumetric MRI using parallel imaging with order-of-magnitude accelerations and a 32element RF coil array: feasibility and implications. Acad Radiol. 2005;12:626–635. 89. Noeske R, King KF, Friedrich MG, Niendorf T. Assessment of cardiac function using 2D CINE FIESTA imaging in combination with parallel imaging. In: Proceedings of the 10th Scientific Meeting of the International Society of Magnetic Resonance in Medicine. Honolulu, Hawaii, USA: 2002:1659. 90. Noeske R, Messroghli D, Friedrich MG, Niendorf T. 2D CINE FIESTA imaging for the assessment of cardiac function in combination with parallel imaging: a comparison study. In: Proceedings of the 11th Scientific Meeting of the International Society of Magnetic Resonance in Medicine. Toronto, Ontario, Canada: 2003:1563. 91. Wintersperger BJ, Nikolaou K, Dietrich O, et al. Single breath-hold real-time cine MR imaging: improved temporal resolution using generalized autocalibrating partially parallel acquisition (GRAPPA) algorithm. Eur Radiol. 2003;13:1931–1936. 92. Rettmann DW, Saranathan M, Wu KC, Azevedo CF, Bluemke DA, Foo TK. High temporal resolution breath-held 3D FIESTA CINE: validation of ventricular function in patients with chronic myocardial infarction. In: Proceedings of the 12th Scientific Meeting of the International Society of Magnetic Resonance in Medicine. Kyoto, Japan: 2004:3823. 93. Weiger M, Pruessmann KP, Boesiger P. Cardiac real-time imaging using SENSE: SENSitivity Encoding scheme. Magn Reson Med. 2000;43:177–184. 94. Pruessmann KP, Weiger M, Boesiger P. Sensitivity encoded cardiac MRI. J Cardiovasc Magn Reson. 2001;3:1–9. 95. Zhang Q, Park J, Li D, Simonetti OP. Improving True-FISP Parallel CINE imaging using a new data acquisition scheme for coil sensitivity calibration. In: Proceedings of the 11th Scientific Meeting of the International Society of Magnetic Resonance in Medicine. Toronto, Ontario, Canada: 2003:2329. 96. McVeigh ER. Regional myocardial function. Cardiol Clin. 1998; 16:189–206. 97. Herzka DA, Derbyshire A, Kellman P, McVeigh ER. Myocardial tagging in a single heart-beat with EPI-SSFP and TSENSE. In: Proceedings of the 11th Scientific Meeting of the International Society of Magnetic Resonance in Medicine. Toronto, Ontario, Canada: 2003:374.

56 Cardiovascular Magnetic Resonance

98. Ryf S, Kozerke S, Boesiger P. 3DCSPAMM tagging accelerated with kt-BLAST. In: Proceedings of the 11th Scientific Meeting of the International Society of Magnetic Resonance in Medicine. Toronto, Ontario, Canada: 2003:1565. 99. Beerbaum P, Korperich H, Gieseke J, Barth P, Peuster M, Meyer H. Rapid left-to-right shunt quantification in children by phase-contrast magnetic resonance imaging combined with sensitivity encoding (SENSE). Circulation. 2003;108:1355–1361. 100. Beerbaum P, Korperich H, Gieseke J, Barth P, Peuster M, Meyer H. Blood flow quantification in adults by phase-contrast MRI combined with SENSE–a validation study. J Cardiovasc Magn Reson. 2005;7:361–369. 101. Kellman P, Larson AC, Zhang Q, Simonetti OP, Arai AE, McVeigh ER. Cardiac CINE 3D true-FISP parallel imaging using auto-calibrating 2D TSENSE. In: Proceedings of the 12th Scientific Meeting of the International Society of Magnetic Resonance in Medicine. Kyoto, Japan: 2004:2120. 102. Atkinson DJ, Burstein D, Edelman RR. First-pass cardiac perfusion: evaluation with ultrafast MR imaging. Radiology. 1990;174: 757–762. 103. Wilke N, Jerosch-Herold M, Wang Y, et al. Myocardial perfusion reserve: assessment with multisection, quantitative, first-pass MR imaging. Radiology. 1997;204:373–384. 104. Ding S, Wolff SD, Epstein FH. Improved coverage in dynamic contrast-enhanced cardiac MRI using interleaved gradient-echo EPI. Magn Reson Med. 1998;39:514–519. 105. Slavin GS, Wolff SD, Gupta SN, Foo TK. First-pass myocardial perfusion MR imaging with interleaved notched saturation: feasibility study. Radiology. 2001;219:258–263. 106. An J, Voorhees A, Chen Q. SSFP arterial spin labeling myocardial perfusion imaging at 3 Tesla. In: Proceedings of the 13th Scientific Meeting, International Society for Magnetic Resonance in Medicine. 2005:253. 107. Kim RJ, Fieno DS, Parrish TB, et al. Relationship of MRI delayed contrast enhancement to irreversible injury, infarct age, and contractile function. Circulation. 1999;100:1992–2002. 108. Kellman P, Arai AE, McVeigh ER, Aletras AH. Phase-sensitive inversion recovery for detecting myocardial infarction using gadoliniumdelayed hyperenhancement. Magn Reson Med. 2002;47:372–383. 109. Huber ME, Kozerke S, Pruessmann KP, Smink J, Boesiger P. Sensitivityencoded coronary MRA at 3T. Magn Reson Med. 2004;52:221–227. 110. Niendorf T, Saranathan M, Lingamneni A, et al. Short breath-hold, volumetric coronary MR angiography employing steady-state free precession in conjunction with parallel imaging. Magn Reson Med. 2005;53:885–894. 111. Weber OM, Martin AJ, Higgins CB. Whole-heart steady-state free precession coronary artery magnetic resonance angiography. Magn Reson Med. 2003;50:1223–1228. 112. Niendorf T, Hardy CJ, Giaquinto RO, et al. Towards single breathhold whole heart coverage coronary MRA using highly accelerated parallel imaging with a 32-channel MR system. Magn Reson Med. 2006;56:167–176. 113. Botnar RM, Bucker A, Kim WY, Viohl I, Gunther RW, Spuentrup E. Initial experiences with in vivo intravascular coronary vessel wall imaging. J Magn Reson Imaging. 2003;17:615–619. 114. Mistretta CA, Wieben O, Velikina J, et al. Highly constrained backprojection for time-resolved MRI. Magn Reson Med. 2006;55: 30–40. 115. Hardy CJ, Darrow RD, Saranathan M, et al. Large field-of-view real-time MRI with a 32-channel system. Magn Reson Med. 2004; 52:878–884. 116. Ohliger MA, Grant AK, Sodickson DK. Ultimate intrinsic signal-tonoise ratio for parallel MRI: electromagnetic field considerations. Magn Reson Med. 2003;50:1018–1030. 117. Wiesinger F, Boesiger P, Pruessmann KP. Electrodynamics and ultimate SNR in parallel MR imaging. Magn Reson Med. 2004;52:376–390. 118. Griswold MA, Kannengiesser S, Heidemann RM, Wang J, Jakob PM. Field-of-view limitations in parallel imaging. Magn Reson Med. 2004;52:1118–1126.

Myocardial Perfusion Imaging Theory Michael Jerosch-Herold and Norbert Wilke

The concept of injecting a tracer into the bloodstream and detecting its transit and distribution in the heart muscle for the assessment of myocardial perfusion is well established in nuclear cardiology and X ray densitometry. Both exogenous, injected contrast agents and endogenous contrast mechanisms have been used to assess perfusion with cardiovascular mag netic resonance (CMR). The use of a gadolinium based con trast agent for the assessment of myocardial perfusion with CMR has been extensively validated and successfully applied in patient studies. Recent developments, in particular, the introduction of parallel imaging and high field (3 Tesla) CMR, have made it possible to combine the requirements for spatial and temporal resolution for myocardial perfusion imaging during the first pass, with multi slice coverage. The need for quantitative analysis of perfusion studies is also receiving increasing acceptance.1 This chapter reviews the theoretical foundations of myocardial perfusion imaging with CMR to convince the reader that the techniques have matured to a point where they are applicable in clinical studies, despite the additional time required for quantitative analysis. There is already compelling evidence that CMR is superior to nuclear imaging for the assessment of myocardial perfusion.2

THE PHYSIOLOGIC BASIS FOR MEASURING MYOCARDIAL PERFUSION Under normal conditions, the blood flow resistance of the coronary circulation is determined primarily by the myocar dial microcirculation, meaning vessels that are smaller than 300 mm in diameter. The adequate supply of oxygen and metabolites to the myocytes is tightly coupled to myocardial blood flow. Adequate and approximately constant blood flow is maintained through autoregulation and can compen sate under resting conditions for up to 80% diameter coro nary artery stenosis.3,4 With more severe narrowing in an epicardial vessel, and in the absence of significant collateral flow, the distal perfusion bed is fully vasodilated, even under resting conditions, and no further augmentation of blood flow is feasible. In healthy subjects, myocardial blood flow can increase three to fourfold with maximal vasodilation.5 This means that differences in myocardial blood flow between a region subtended by a stenosed coronary artery and the territory of a normal coronary artery are amplified with maximal vasodilation.

Myocardial perfusion imaging during pharmacologic vasodilation (e.g., with adenosine or dipyridamole) rests on the physiologic observation that the hemodynamic sig nificance of a lesion is most apparent during maximal vaso dilation.3,4 A related measure can be obtained in the catheterization laboratory with an intravascular Doppler flow probe by measuring the coronary flow reserve to assess lesion severity6 or by measuring the fractional flow reserve,7 defined as mean distal coronary artery pressure divided by the aortic pressure during maximal vasodilation. These functional tests of coronary and myocardial blood flow overcome the known limitations of diameter vessel lumen measurements by projection of invasive X ray angi ography for determining the hemodynamic significance of epicardial lesions. Furthermore, myocardial perfusion imaging can be used to assess functional impairments in the microcirculation. For example, both myocardial perfu sion reserve and coronary flow reserve were abnormally low in women with microvascular dysfunction and without hemodynamically significant epicardial lesions.8,9 There is also evidence from epidemiologic studies that CMR perfu sion imaging detects subclinical disease and silent ischemia in subjects without a history or symptoms of atheroscle rotic disease. The myocardial perfusion reserve in response to adenosine was found to be associated with coronary risk factors,10 and the perfusion reserve decreases with increas ing coronary calcium burden.11 In asymptomatic patients, a lower myocardial perfusion reserve is associated with decreased regional left ventricular function12 and also with lower regional myocardial strain.13 With the development of CMR as a new imaging modality for myocardial perfusion imaging, it has become possible to probe with sufficient spatial resolution for more subtle indi cators of regional myocardial ischemia. Blood flow across the myocardial wall is not uniform, but instead favors the suben docardium to accommodate its higher workload and higher rate of oxygen consumption.14,15 Under normal conditions, the ratio of endocardial to epicardial blood flow is approxi mately 1.15:1. With a coronary artery stenosis, blood flow is first diverted away from the subendocardial layer and the endocardial to epicardial blood flow ratio is often less than 1:1, in particular, under stress.16,17 With myocardial ischemia, the subendocardial layer is accordingly most sus ceptible to necrosis. This potential advantage of myocardial perfusion imaging can be sufficiently appreciated only if the imaging modality provides spatial resolution on the order of 2 mm or better. With CMR, it has become feasible to detect flow impairments limited to or most accentuated Cardiovascular Magnetic Resonance 57

4 MYOCARDIAL PERFUSION IMAGING THEORY

CHAPTER 4

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

in the subendocardial layer.18,19 The spatial resolution of conventional imaging modalities, such as positron emission tomography (PET) and single photon emission computed tomography (SPECT), was insufficient to detect blood flow deficits limited to the subendocardial layer. More specifically, the sensitivity of a myocardial perfusion imaging technique is directly related to its spatial resolution and the ability to discern transmural variations of flow.18,20 The use of CMR offers the unique possibility of quanti tatively assessing perfusion, viability, and function with high accuracy to distinguish stunned, hibernating, and infarcted myocardium. Bolli and colleagues showed that even small differences in blood flow during ischemia result in large differences in postischemic function, suggesting that the ability to quantify flow in the low flow range is of importance in predicting the probability of postischemic recovery.21 The extent and incidence of microvascular obstruction observed with CMR was associated with the duration of ischemia before coronary intervention.22

FIRST PASS IMAGING WITH EXOGENOUS TRACERS The success of applying CMR to detect the first pass of an injected contrast agent and to detect perfusion abnormal ities starts with an understanding of the contrast mechan isms that allow detection of the contrast agent. The transit of a contrast agent through the vasculature and tissue leads to changes in the longitudinal (T1) and transverse (T2)* relaxation time constants for the detected 1H signal: the contrast reagent is not detected directly, but rather through its effects on the signal from 1H nuclei, mostly in water and lipids. The chelates of paramagnetic ions used as contrast reagents have a relatively high magnetic susceptibility that causes, on a microscopic scale, magnetic field inhomogene ities and also shortens the T2* relaxation rate of water. The contrast reagent also shortens the T1 magnetization recov ery after a radiofrequency pulse has disturbed the equilib rium state. The CMR methods for perfusion imaging can be subdi vided into T1 and T2* weighted techniques: T1 weighted techniques produce signal enhancement during the transit of the contrast agent, whereas T2* weighted techniques cause signal loss. For contrast agents confined to the vascu lar bed, and in the absence of significant organ motion, T2* weighted perfusion imaging gives rise to relatively larger signal changes than T1 weighted techniques because the magnetic susceptibility effects of the contrast agents extend beyond the capillaries. However, T2* weighted imaging techniques have an inherent sensitivity to motion, leading to signal loss in the presence of motion. For cardiac perfusion imaging, T1 weighted techniques such as gradi ent echo imaging with short echo times and a magnetiza tion preparation for optimal T1 weighting have therefore steadily gained preference. This chapter provides a brief overview of three major sequence techniques for T1 weighted perfusion imaging of the heart. They are, in historical order, spoiled gradient recalled echo imaging (GRE), multi shot T1 weighted echo planar imaging (EPI), and gradient echo imaging with steady state free precession (SSFP). 58 Cardiovascular Magnetic Resonance

Single shot GRE with magnetization preparation is well suited for T1 weighted, quantitative myocardial perfusion imaging. A magnetization preparation for T1 weighting can take the form of an inversion pulse or a saturation pulse; either of them is generally applied in a non slice selective mode so that blood flowing into the image plane during the subsequent image acquisition has been subjected to the same T1 preparation as myocardium. A magnetization preparation with a saturation pulse can be repeated in a sequential multi slice sequence for each slice, and this will result in the same degree of T1 weighting for each slice. After the magnetization preparation, images for one or more slices are rapidly read out within approximately 200 to 300 msec, or even less with parallel imaging techniques. Typical sequence parameters for such rapid readouts are a repetition time (TR) per phase encoding step of 2.0 msec or less, an echo time (TE) of 1 msec or less, a receiver bandwidth on the order of 800 to 1000 Hz/pixel, and an in plane spatial resolution of 2 to 3 mm. Because of the short TE, the signal is relatively insensitive to flow and magnetic susceptibility variations. With a linear ordering for the phase encoding steps from high to low spatial frequencies, the image acqui sition only needs to be delayed by 10 to 20 msec after a sat uration recovery preparation. With a combination of short TR and TE times, the signal initially increases linearly with contrast agent concentration, or as a function of the relaxa tion rate constant. (The relaxivity of the contrast reagent is not appreciably different between blood and tissue.) Even tually, the signal ceases to increase because of the opposite effect of T2* on the signal at higher contrast concentrations, and also because of the limited dynamic range for T1 related signal changes imposed by the pulse sequence technique. Figure 4 1 shows an example of a CMR perfusion study with a spoiled GRE technique. Because EPI eliminates the need for radiofrequency exci tation before each phase encoding step, it is one of the fastest imaging methods for freezing heart motion (50 to 100 msec for single shot acquisitions). In an EPI pulse sequence, an initial radiofrequency excitation creates a coherent precession of transverse magnetization, and a train of gradient echoes is then generated by applying a rapidly oscillating magnetic field gradient along the readout direction. Each of these gradient echoes is preceded by a short, blipped gradient pulse, applied along a direction perpendicular to the oscillating magnetic field gradient. The blipped phase encoding gradients between echoes advance the trajectory in the frequency acquisition space (k space) perpendicular to the readout direction. Instead of just a single line, after each radiofrequency excitation, one acquires a set of parallel lines in frequency space. The T2* related decay of the gradient echo amplitudes in the echo train results in an increasing sensitivity of later echoes to T2* and motion, and causes blurring. The effec tive echo time can be an order of magnitude longer for EPI sequences compared with fast GRE sequences (e.g., 30 to 40 msec vs. 1.2 to 2.0 msec for fast gradient echo imaging). Unfortunately, a longer effective TE gives rise to magnetic susceptibility and flow artifacts in the ventricular cavity. Furthermore, flow and susceptibility related artifacts in the ventricular cavity make it difficult to measure the arte rial transit of the contrast agent. To partially capture the speed advantage of EPI and to minimize T2* and flow related artifacts, one reduces the length of the echo trains

250 200

100 LV RV

50

Signal intensity (a.u.)

150

Tissue 0

0

10

20

30

(IV contrast injection)

40

50

60

70

80

90

Time (sec)

Figure 4 1 The gradient echo cardiovascular magnetic resonance images show the first pass of an extracellular contrast agent (0.03 mmol/kg Omniscan, GE Healthcare Princeton, NJ) in a healthy volunteer, with appearance of the contrast first in the right ventricle (RV), followed by the left ventricle (LV), and finally leading to myocardial contrast enhancement. The contrast agent was injected after acquisition of approximately four pre contrast images. Images were acquired in the short axis view at the level of the papillary muscle. The arrows show the correspondence between images and the time course of changes in signal intensity for a region of interest in the left ventricle (solid circles) and a myocardial segment in the posterior wall (diamonds). Characteristics of the signal curve for a tissue region, such as its increased dispersion, lower amplitude, and reduced rate of signal enhancement compared with the ventricle, result from transit through the coronary microcirculation, where the volume of distribution is limited to either the extracellular or the intravascular space. a.u., arbitrary units; IV, intravenous.

and the images are read out with several radiofrequency excitations/echo trains.23 This results in a 30% to 40% speed advantage compared with conventional GRE sequences and is particularly useful at field strengths of 1.5 T or lower. (At 3.0 T and higher, any increase in TE beyond the minimum allowed by the gradient system nearly inevitably gives rise to susceptibility artifacts, in par ticular, during passage of a contrast bolus.) The signal to noise ratio and the contrast to noise ratio in GRE images are relatively low, when a short TR and wide receiver bandwidths are used. The maximum signal inten sity is reached with relatively low flip angles. (The flip angle corresponding to maximum signal intensity is referred to as Ernst angle.) To overcome this limitation, one can use an ingenious scheme, referred to as SSFP, to preserve the coherence of the transverse magnetization between TRs in the pulse sequence and efficiently convert magnetization between transverse and longitudinal orienta tions. With SSFP, the Ernst angle can approach 90
(i.e., one can reach the maximum theoretical signal amplitude after each radiofrequency excitation while still repeating the excitations with very short TRs). In fact, the TRs should be as short as feasible because the susceptibility of the SSFP technique to any off resonance shifts increases with TR and also with flow.24 Off resonance frequency shifts result in dark bands at locations where the frequency shift (in radians per second) times TR (in seconds) equals an odd

multiple of p/2. Herein lies the Achilles heel of the SSFP technique, and not surprisingly, the passage of a contrast bolus can exacerbate frequency shifts and artifacts. As absolute frequency shifts increase with field strength (assuming the same relative field homogeneity at different field strengths), one can best reap the advantages of SSFP techniques for perfusion imaging at field strengths of up to 1.5 T. Because of the relatively good signal to noise ratio, the SSFP techniques can produce more appealing image quality,25 but the use of the SSFP technique for myocardial perfusion imaging must be approached with caution because the artifacts on SSFP images can be deceptive and mimic hypoperfusion. Ultrafast T1 measurements would provide the most accurate estimates of contrast agent concentration, and with the advent of parallel imaging, this approach may become practical. Chen and coworkers developed such a T1 fast acquisition relaxation mapping (T1 FARM) method to obtain single slice T1 maps of the heart with 1 sec reso lution.26,27 With direct measurement of T1, the problems associated with the saturation of the signal with increasing contrast agent dosage can be avoided. This should allow for more accurate quantification of blood flow with usage of higher contrast agent dosages. Figure 4 2 shows two signal curves for a region of interest in the left ventricle measured during bolus injection of 0.075 mmol/kg gadolinium diethyl triaminepentaacetic acid (Gd DTPA) with the Cardiovascular Magnetic Resonance 59

4 MYOCARDIAL PERFUSION IMAGING THEORY

300

LV blood pool

1000 800 Signal (a.u.)

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

PARALLEL IMAGING AND ACCELERATION TECHNIQUES FOR PERFUSION CARDIOVASCULAR MAGNETIC RESONANCE

COMPARISON OF LV CURVES WITH 0.1 MMOL/KG BOLUS OF GD-DTPA

T1-FARM Fast gradient echo

600 400 200 0 –200 0

10

20

30

40

50

Time (sec)

Figure 4 2 Comparison of signal time curves for a region of interest in the left ventricle (LV) obtained in a canine with quantitative T1 imaging (T1 FARM) and T1 weighted, saturation recovery prepared fast gradient echo (TR/TE 2.4/1.2 msec; flip angle 18
) and a gadolinium diethyl triaminepentaacetic acid (Gd DTPA) dosage of 0.075 mmol/kg at 1.5 T. The curves were normalized so the TurboFLASH and T1 FARM recirculation peaks were equal. a.u., arbitrary units, (Data courtesy of Z. Chen, C.A. McKenzie and F.S. Prato, St. Joseph’s Hospital, London, Ontario.)

T1 FARM and the saturation recovery prepared GRE tech niques, respectively. For better temporal resolution, to achieve multi slice coverage, and to allow simpler image reconstruction, in the future, parallel imaging methods will be used to accelerate T1 FARM image acquisition. All current CMR techniques offer higher spatial resolu tion than is available with nuclear imaging techniques. An illustration is shown in Figure 4 3 from studies in an experimental animal model.

RV

Parallel imaging refers to the use of independent receiver chan nels for the parallel acquisition of data from multiple elements of a receiver coil array (see Chapter 3). Parallel imaging tech niques have led to a leap forward in image acquisition speed. Each element of a receiver coil array has a characteristic spatial sensitivity profile. In the ideal case in which the signal profiles do not overlap, it becomes possible to localize a signal within the dimensions of the coil element, without gradient encod ing. By measuring the local coil profiles, one can reduce the number of phase encoding steps and during image recon struction replace the missing phase encodings with informa tion related to the coil profiles. The algorithms can operate on the sensitivity encoded images calculated for the individual coil elements (SENSE)28,29 or on the spatial frequency data acquired with the coil elements (simultaneous acquisition of spatial harmonics [SMASH]).30 Theoretically, the number of phase encoding steps can be reduced by a factor that equals the number of independent coil elements in the phase encod ing direction. In practice, and in particular for imaging techni ques with relatively poor signal to noise ratio, such as those used for myocardial perfusion imaging, speed up factors on the order of  2 are more realistic.31 Such a speed up results in image acquisition times with GRE techniques on the order of 100 to 150 msec/image, with an in plane spatial resolution on the order of 2 mm. During pharmacologic stress and with heart rates of approximately 100 bpm, one can cover the heart with 4 to 5 slices during stress and up to 10 slices at rest.31

ANT

RV

CMR (1st pass)

FDG-PET

TTC-stained

Figure 4 3 Images from cardiovascular magnetic resonance imaging (CMR), positron emission tomography (PET), and (TTC) staining in a porcine model in which obtuse branches of the left circumflex coronary artery had been ligated 4 weeks before combined MRI and PET studies. From CMR acquired during the first pass of an intravascular iron oxide contrast agent (NC100150 injection, Nycomed), the one shown at left corresponds to the highest peak signal enhancement in tissue and indicates (pointed out by arrow) a subendocardial perfusion defect in the posterior segment, in agreement with 2,3,5 Triphenyltetrazolium chloride (TTC) staining shown at right with sub endocardial absence of staining in sector highlighted by arrow. 13NH2 positron emission tomography (middle image) was carried out 3 hours before the MRI study and shows a fixed defect in the posterior segment (arrow), suggesting a transmural infarct. Fluorodeoxyglucose positron emission tomography images (not shown) also indicated irreversible damage in the posterior segment, in disagreement with the findings from MRI and TTC staining. ANT, anterior; RV, right ventricle. 60 Cardiovascular Magnetic Resonance

WATER EXCHANGE AND ITS EFFECTS ON MYOCARDIAL CONTRAST ENHANCEMENT With 1H magnetic resonance imaging, the presence of a con trast agent is detected indirectly through the change in the 1 H relaxation recovery rates, T1 and T2. The effect of a con trast agent on T1 and T2 within a single uniform closed space can be described by its relaxivity, r1,2, using the follow ing equation: T1;2 ¼ r1;2  ½CR. Here, [CR] denotes the con centration of the contrast reagent (e.g., in units of millimoles), and r1 and r2 are the relaxivity in units of 1/ (mM s). We only focus here on T1 effects because of the pre dominant usage of T1 weighted techniques for myocardial perfusion imaging. In a compartmentalized system such as myocardial tissue, with barriers that are permeable to water, one has to consider not only the local interaction of water with the contrast agent within a compartmental space, but also its exchange between, for example, the plasma space and the interstitial space. For an extracellular agent, the exchange of water between the interstitial and intracellular spaces also comes into play. Water exchange across the cap illary barrier and the cytolemmal membrane, if sufficiently fast, allows 1H spins to sample different environments dur ing spin relaxation. As a result, the observed relaxation recovery rate is no longer determined solely by local contrast

agent concentration, but depends also on the rates at which H spins exchange between spaces and by the relative volumes of those spaces. Thus, even though a contrast agent may be confined to the vascular space, as in the case of an intravascular contrast agent, the 1H spins outside the vascu lar space can enter the plasma space and relax at a faster rate than in the absence of water exchange. (The reverse process, of 1H spins leaving the vascular space and relaxing more slowly in the interstitial space, is equally likely.) Two limiting cases are typically considered: (1) the slow or no exchange limit, where the 1H spins relax at a rate intrinsic to the space they dwell in, and (2) the fast exchange limit, where 1H spins in spaces linked by fast water exchange relax at the same rate, which represents a weighted average of the intrin sic relaxation rates. The limiting cases of no exchange and fast exchange pro vide convenient simplifications for the analysis of myocardial contrast enhancement. If signal curves for myocardial regions of interest are analyzed with the no exchange assumption, then any water exchange will lead to overesti mation of the volume of distribution of the contrast agent.37 Conversely, if the signal curves are analyzed with the fast exchange assumption, then a lower rate of exchange, inter mediate between the fast and no exchange limits, will result in an underestimation of the contrast distribution volume and myocardial blood flow.38 Which limiting case is closer to the true setting depends on the experimental setting. Previous experimental studies suggest that for the capillary barrier, the rate of water exchange is on the order of 7 Hz or less,39 and a study with an intravascular agent gave an esti mate of the intravascular lifetime of a water molecule on the order of 0.3 seconds.40 It would appear that neither the fast exchange limit nor the slow exchange limit approximates well the effects of water exchange in myocardial tissue, at least when the relaxation recoveries proceed undisturbed. Never theless, typical T1 time constants are in the range of 100 to 1000 msec in myocardial tissue, depending on contrast agent concentrations, and these T1 values are much larger than the typical TR used for rapid image readout. The application of radiofrequency pulses during a relaxation recovery can be expected to alter the relaxation recovery. Effectively, the rapid application of radiofrequency pulses during a relaxation recovery can reduce the sensitivity of the observed 1H signal to water exchange.41 With a sufficiently high flip angle (20
), the no exchange assumption is reasonable for the inter pretation of the signal curves measured with an ultrafast (TR ¼ 2.4 msec) saturation recovery prepared GRE sequence.42 In principle, it is also possible to incorporate a description of water exchange into a tracer kinetic model, as shown recently by Li and associates.43 The optimal approach to address the combination of contrast enhancement and water exchange remains an area of active research. 1

ENDOGENOUS CONTRAST FOR THE ASSESSMENT OF MYOCARDIAL PERFUSION Approaches have been developed for an assessment of myo cardial perfusion that are not based on the use of an exoge nous CMR contrast agent, but instead exploit endogenous contrast mechanisms related to blood flow and blood Cardiovascular Magnetic Resonance 61

4 MYOCARDIAL PERFUSION IMAGING THEORY

Another remarkable advance for fast CMR perfusion imag ing is the development of sparse sampling techniques for dynamic imaging applications. An example is the unaliasing by Fourier encoding the overlaps using the temporal dimen sion (UNFOLD) technique, which involves a trade off of tem poral sampling to allow sparser spatial sampling.32,33 Without any other adjustments, a halving of the field of view would speed up the image acquisition by a factor of two, but would result in foldover artifacts. To unfold the overlaps (unaliasing), Madore and associates32 proposed that comple mentary sets of phase encoding steps are interleaved in a dynamic imaging series. The alternation between two sets of phase encoding steps results in a modulation of the aliased image components, whereas un aliased image components remain constant. The aliased image components can be removed by filtering of the temporal frequency spectrum at each pixel location. The UNFOLD approach has been applied by Di Bella and colleagues in myocardial perfusion studies,34 and was found to work well if patients hold their breath because respiratory motion interferes with UNFOLD. A further generalization of sparse sampling in the k space and time domains is provided by the broad use linear acqui sition speed up technique (BLAST).35,36 Briefly, with k t BLAST, one views the acquisition as a sampling process in a higher dimensional k t space. Different sets of phase encodings are acquired at successive time points, and the repetition period for each phase encoding set equals the k t BLAST acceleration factor. Image aliasing is removed by use of filtering or prior information, and as a final result, one obtains a series of images over time, t. Although these sparse sampling techniques have only been applied in initial feasibility studies, it is safe to assume that they will play an increasingly prominent role in myocardial perfusion imaging and will allow unprecedented spatial resolution.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

oxygenation. These approaches fall under the categories of spin labeling,44 blood oxygen level dependent (BOLD) con trast,45,46 and magnetization transfer contrast.47,48 These techniques have to be considered technically more challeng ing than measurements with exogenous contrast agents, or they offer only an indirect measure of blood flow. With spin labeling, the spins are either inverted or saturated in a slab that is generally located upstream of the imaged slice.49–51 The flow dependent change of signal intensity caused by the inflow of saturated or inverted spins into the image slice provides a measure of tissue perfusion. The spin labeling method generally relies on the assump tion that the net arterial blood flow in the myocardium follows the direction from base to apex. Cardiac motion complicates the interpretation of signal changes in a spin labeling experiment because the labeled spins can be trans ported into the imaged slice either through blood flow or by through slice plane motion of the heart. The BOLD technique offers a measure of hemoglobin sat uration that reflects regional oxygen supply and demand (see Chapter 42). Deoxyhemoglobin is paramagnetic, whereas oxyhemoglobin is only diamagnetic. This means that deoxy hemoglobin causes a considerably larger reduction of T2* than oxyhemoglobin and therefore a larger signal intensity attenuation in GRE images. Deoxyhemoglobin can be used as an endogenous intravascular tracer because of the tight coupling between oxygen demand and blood flow. BOLD contrast changes have been observed in the heart after administration of dipyridamole and dobutamine.45,46 The link between BOLD contrast and blood flow depends on the balance between blood flow and oxygen metabolism (i.e., between oxygen supply and demand).52 Spin labeling, BOLD contrast, and magnetization trans fer contrast CMR can be carried out in the steady state, meaning that there is no need to capture with ultrafast imaging the transient signal changes observable in the myo cardium after injection of an exogenous tracer. Ultimately, these techniques may allow an assessment of myocardial perfusion with high spatial resolution ( 1 mm or less).

QUANTITATIVE EVALUATION OF MYOCARDIAL PERFUSION The images acquired in a myocardial perfusion study with a bolus injection of contrast agent can be qualitatively eval uated based on the level of differential signal enhancement in the myocardium. With a T1 weighted imaging tech nique, myocardial segments showing reduced signal inten sity enhancement during the first pass, relative to other myocardial segments, are interpreted as hypoperfused. This type of qualitative judgment of signal enhancement differ ences can suffer from substantial observer bias, and small reductions in perfusion can be missed. Although higher dosages of contrast agent would increase the dynamic range, the images are also more likely to be contaminated by artifacts near the endocardial border. Furthermore, global reductions of blood flow caused by diffuse microvas cular ischemia will be missed by this approach. Therefore, it becomes necessary to evaluate the contrast state during stress or vasodilation, relative to a baseline or resting state, to uncover global impairments of perfusion. 62 Cardiovascular Magnetic Resonance

With an appropriate imaging technique, such as a fast T1 weighted gradient echo sequence and low contrast agent dosages (e.g., 0.03 mmol/kg Gd DTPA for an intravenous bolus injection into an antecubital vein), the signal changes are proportional to the local contrast agent concentration. Under these conditions, the signal curves can be interpreted as or transformed into contrast agent residue curves. Abso lute units for the signal intensity (or related equivalent such as contrast concentration or relaxation rate), do not matter here, because enhancement in the tissue is interpreted rela tive to the enhancement in the blood pool, both measured on the same linear scale. In fact, for the purpose of quantify ing perfusion, the tissue is often viewed as a linear, station ary system, where enhancement in the tissue can be modeled as a linear response to arterial input of contrast. Over the years, investigators using PET have presented compelling arguments for a quantitative approach to myo cardial perfusion imaging, including absolute quantifica tion of myocardial blood flow.53 With CMR, a quantitative analysis of contrast enhancement is based on signal inten sity curves, which can be generated for user defined regions of interest or at the pixel level. CMR perfusion studies require a careful approach for image segmentation and registration. CMR perfusion studies are not motion averaged, the endocardial borders are relatively well defined during transit of contrast through the ventricular cavity, and the large signal enhancement in the ventricular cavity requires accurate image segmentation to avoid spill over effects. Figure 4 4 shows an example of a CMR perfu sion study in a patient that was performed for absolute quantification based on model independent analysis of signal intensity curves.54 In the vein of Zierler’s55–57 original work on the central volume principle, we consider here a tissue region of inter est with a single arterial input and a single venous output. The injection of a tracer is described by the variation of tracer concentration at the (arterial) input cin(t). The amount of tracer that remains in the tissue region at any time t, q(t), represents the difference between the amount supplied to the tissue region up to time t, and the total amount of tracer that has exited the region of interest up to time t, cout(t): ðt qðtÞ ¼ F ½cin ðsÞ  cout ðsÞ  dt

(1)

0

F represents the flow blood (e.g., in units of milliliters per minute per gram of tissue). Zierler showed that if the transport within a tissue region of interest is assumed to be linear and stationary, then the response to an arbitrary arterial input is given by the convolution integral of this input with the tissue impulse response57: ðt qðtÞ ¼ cin ðt  tÞ  RðtÞdt

(2)

0

The impulse response is denoted by R(t). The response in a tissue region to an impulse input, the tissue impulse response, has the useful property that its initial amplitude equals the myocardial blood flow through that region. This can be derived by considering the special case in which the

Myocardial SI (a.u.)

RV

100

50 0.70 mL/g/min (anteriorseptal)

0

A

10

20

30

40

50

60

50

60

Time (sec)

RV

LV SI (a.u.)

600 400 200 0

B

10

C

20

30

40

Time (sec)

Figure 4 4 A 71 year old man presented with a silent non Q wave anterior myocardial infarction. A, An image from a series of 60 dynamic images taken at rest shows the initial contrast enhancement in the myocardium, with reduced contrast uptake in the anterior and anterior septal segments (black arrow) because of an eccentric 90% stenosis with a lesion length of 10 mm in the proximal left anterior descending coronary artery. B, Late gadolinium enhancement was observed in the anterior and anterior septal segments (white arrow) from midlevel to apex, but not at a more basal level, including the level of the slice in shown in A. C, The upper panel shows the change of signal intensity (after subtraction of the baseline offset) in two myocardial segments (solid circles for lateral posterior segment and inverted triangles for anterior septal sector). The curve in the lower panel shows the changes of signal intensity in a central region of the left ventricular blood pool and depicts the arterial input of contrast. Myocardial blood flow was quantified by model independent deconvolution of the myocardial signal intensity curves, using an arterial input measured in the left ventricular blood pool. The solid lines in the upper panel of C show the calculated tissue response, which agree well with the measured data points. Blood flow at rest in an anterior septal segment was 0.7 mL/g/min, compared with 1.2 mL/g/min in the posterior wall. The anterior and anterior septal segments appeared hypokinetic on cine magnetic resonance imaging. After cardiovascular magnetic resonance, the patient was referred for coronary angiography. During percutaneous coronary intervention, a drug eluting stent was deployed in the proximal left anterior descending coronary artery. a.u., arbitrary units; LAT, lateral; LV, left ventricle; RV, right ventricle; SI, signal intensity.

input is a dirac delta impulse, cin(t) ¼ d(t), and requiring that cout(t ¼ 0) ¼ 0 (i.e., the tracer cannot instantaneously reach the output after injection). This property of the impulse response is independent of the vascular and com partmental structure inside the tissue region of interest or the properties of barriers, such as their permeability. For example, leakage of a contrast agent from the capillaries into the interstitial space produces a redistribution of the contrast agent within the region of interest, but does not contribute to flow into and out of the region of interest, as long as transport by diffusion and convection is much slower than by blood flow. For first pass CMR contrast agents such as Gd DTPA, these assumptions hold up well, but for freely diffusible tracers, they break down. The central volume principle indicates that blood flow can be determined by deconvolution of the measured tissue residue curve with the arterial input curve. The deconvolu tion analysis is quite sensitive to noise in the data, and one therefore needs to constrain the deconvolution operation. For example, the impulse response, R(t), should be a monotonically decaying function of time because by defini tion of the impulse input, no additional tracer enters the tissue region after the initial delta function input. Further more, the impulse response should be reasonably smooth. Smoothness of the impulse response follows to the degree that the magnitudes of flow, diffusion, and permeation in the heart impose time scales that exclude abrupt signal intensity changes in myocardial tissue after contrast agent

injection. A Fermi function has been used as an empirical model for the impulse response to fit the first pass portion of the signal curves.42,58 The Fermi model of the impulse response has the following equation: A ; (3) 1 þ exp½ðt  wÞ=t where A, w, and t are the model parameters, with no direct physiologic meaning. From an empirical standpoint, the shape of the Fermi function provides a reasonable approxi mation of the shape of the impulse response of an intravas cular tracer. The portion of the signal intensity curve up to the peak has higher sensitivity to flow changes than later phases of contrast enhancement, and the early enhancement is relatively insensitive to capillary leakage of contrast agent. This initial part of the signal intensity curve can be approxi mated well by the Fermi function model to assess blood flow. The Fermi model for the quantification of myocardial perfusion and perfusion reserve was validated by compari son to blood flow measurements with labeled micro spheres59,60 and also by comparison to invasive coronary flow reserve measurements,42 with similar regional perfu sion in all segments of approximately 1 mL/kg at rest.61 As an alternative to the deconvolution analysis, one can build a mathematical model of tracer transport through tis sue. For an extracellular contrast agent, this involves exchange of tracer between vascular and interstitial spaces. For a compartmental model, it is assumed that the tracer is RðtÞ ¼

Cardiovascular Magnetic Resonance 63

4 MYOCARDIAL PERFUSION IMAGING THEORY

1.19 mL/g/min (lateral-posterior) LAT

these models tends to be rather large. Sensitivity analysis is called for to determine which (preferably few) model para meters need to be adjusted for a best fit of the measured resi due curves. The number of degrees of freedom can be reduced by using an intravascular contrast agent.63,64 The realism of spatially distributed models of blood tissue exchange offer an advantage for the identification and quanti fication of physiologic changes observed indirectly with

BLOOD-TISSUE EXCHANGE UNIT

Blood plasma flow (Fp)

Axial direction (x)

PScap

Plasma volume = Vp

Interstitial space Parenchymal cell

A

Signal intensity (a.u.)

50 LV input

40 30

4 mL/g/min

20 10 1 mL/g/min

0

0

10

20

30

B

40

50

60

70

Time (sec) 200

Signal intensity (a.u.)

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

well mixed within each compartment, meaning that the con trast agent concentration equilibrates instantaneously within each compartment. The rapid injection of a tracer will create a tracer concentration gradient from the arterial to the venous side. Spatially distributed models are described by partial dif ferential equations that account for the variation of tracer con centrations within each tissue region as a function of time and at least one spatial variable.62,63 The number of parameters in

5 mL/g/min 150

Fp ~ impulse amplitude

4 mL/g/min

100

2.7 mL/g/min 2 mL/g/min

50 1 mL/g/min 0 0

C

10

20

30

40

50

60

70

Time (sec)

Figure 4 5 A, Conceptual diagram of the spatially distributed, two region model that was implemented in the JSIM (JSIM, Java based Simulation and Modeling environment) simulation environment (http://www.physiome.org/jsim/) for analysis of tracer residue curves and quantification of myocardial blood flow. The following model parameters are optimized for fitting to experimental data: plasma flow (Fp), plasma volume (Vp), and capillary permeability surface area product (PScap). B, A tissue contrast residue curve measured in a 56 year old female volunteer during vasodilation with adenosine (open circles) was analyzed with the model described in A, and the best fit corresponding to a tissue blood flow of 2.8 mL/g/min is shown as a solid line. Additional model curves for flows of 1, 2, and 4 mL/g/min and for otherwise unchanged model parameters are shown as dashed lines to illustrate the sensitivity of the initial contrast enhancement to blood flow. The rate of the initial contrast enhancement (“up slope”) changes as the flow is varied, whereas the curves cluster together after the first pass, when sensitivity to flow changes is low. Also, for equal increments of blood flow, the peak contrast enhancement changes less as the absolute magnitude of flow increases. C, The measured arterial input was replaced by an impulse input, with approximately the same area under the curve. Now the amplitude of the simulated tissue response to the impulse input increases in proportion to the blood flow (F  impulse amplitude), as predicted by Zierler’s central volume principle. a.u., arbitrary units; LV, left ventricle. 64 Cardiovascular Magnetic Resonance

ARTERIAL INPUT FUNCTION Quantification of myocardial blood flow generally requires the measurement of an arterial input function, which serves as reference for analysis of the myocardial contrast enhancement. The term arterial input function refers to the measurement of contrast enhancement at a location before it enters the perfusion bed under study. Under ideal condi tions, the input function would be measured immediately before blood enters the perfusion bed, but with CMR, it is not feasible to measure reliably the arterial input, even in the epicardial vessels. Therefore, one has to settle for measurement at a more upstream location, either near the aortic root or in the left ventricle. With contrast enhanced CMR, one faces a further con straint in the limited dynamic range of contrast enhancement, which results in signal saturation at higher contrast concen trations, as illustrated in Figure 4 6A. Nevertheless, the dynamic range can be tuned to the concentration range with the sequence parameters, and it is indeed possible to achieve approximate linearity between signal change (above precon trast signal) and tracer concentration by using short inversion times ( 50 msec). However, the same tuning to short TI

values is not optimal for capturing myocardial contrast enhancement because the tracer concentrations are lower than in the blood pool and the short time after inversion (T1) values decrease the sensitivity to small relaxation rate constant (R1) changes. Figure 4 6 shows the effects on con trast enhancement of the delay time after a saturation prepa ration. There are essentially three solutions to the problem of signal saturation: (1) Dual bolus technique: one performs two injections with a low and a relatively higher contrast dos age, each optimized to measure contrast enhancement in the myocardium and the blood pool, respectively, but with other wise identical injection protocols and under stable hemody namic conditions.59 The signal curves for a blood pool region of interest are scaled according to the ratio of contrast dosages used in the dual bolus protocol, and then further quantitative analysis can be performed on curves for tissue and blood pool with an optimized contrast to noise ratio. (2) As a second alternative, one can include in the pulse sequence a rapid low resolution image acquisition with very short inversion time to measure the arterial input with mini mal signal saturation.65,66 Such pulse sequence techniques with arterial input scout impose a small time penalty, but oth erwise show promise for keeping the acquisition protocol simple and not requiring dual bolus injections. (3) Finally, it is often feasible to construct calibration curves of signal versus (R1), or tracer concentration, to correct for signal satu ration,67 as long as the calibration curves are monotonically increasing as a function of R1. An example of calibration based saturation correction appears in Figure 4 6C.

TR/TE = 2.3/1.0 msec Flip angle = 18o

1000

TD = 10

800

2000 1000 0 0

200 400 600 800 1000 1200 1400

B 400

Observed CE (%)

TD = 300 ms 1000

200 0 0

A

3000

TD = 100

600

CE (%)

Contrast enhancement (CE) (%)

1200

Expected CE (%)

MODEL-BASED CALIBRATION CURVE

2

4

6

8

10

500 0

Measured LV CE Corrected LV CE

500

R1 (s)

0

C

10

20

30

40

50

Image #

Figure 4 6 A, Model based calculations based on the Bloch equations were used to calculate contrast enhancement (CE) in the left ventricular (LV) blood pool for a range of relaxation rate constants (R1), assuming a precontrast T1 value of 1100 msec at 1.5 T. The delay time (TD) between saturation preparation and gradient echo image readout varied between 10 and 300 msec and had a marked effect on the nonlinearity of the contrast enhancement curves. Short TD values are most appropriate for measuring contrast enhancement for a wide range of R1 values, as encountered during the first pass of a contrast agent in the blood pool. Longer TD values result in better contrast to noise ratios. The dark blue line corresponds to the ideal case in which contrast enhancement is linearly proportional to R1 and contrast agent concentration. B, Model based calibration curves can be used to convert the observed contrast enhancement to the ideal case in which contrast enhancement is linearly proportional to contrast concentration. Such calibration curves can be generated from the simulations shown in A, using the sequence parameters and precontrast T1 values of an actual study. C, With the model based calibration curve in B, a measured curve of left ventricular contrast enhancement was corrected for saturation effects. Left ventricular contrast enhancement was measured with a bolus of 0.05 mmol/kg gadolinium diethyl triaminepentaacetic acid (Magnevist, Berlex, Wayne, NJ). With this dosage, the peak signal measured in the left ventricle (LV) shows approximately 30% saturation. The contrast concentration change in the blood pool would be underestimated by 30% if the measured signal were taken at face value and not corrected for saturation. TE, echo time; TR, repetition time. Cardiovascular Magnetic Resonance 65

4 MYOCARDIAL PERFUSION IMAGING THEORY

tomographic imaging techniques such as CMR. An example of the use of a tracer kinetic model for analysis of a CMR per fusion study is shown in Figure 4 5, including an illustration of the central volume principle, based on model simulations.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

It may be possible in the future to control the arterial input of an injected contrast agent in a manner similar to the approach used by arterial spin labeling or in myocardial contrast echocardiography with echogenic bubbles that can be destroyed by ultrasound pulses. For CMR, hyperpo larized contrast media68 could be one possible approach because it is relatively easy to selectively quench the hyper polarized signal.

PRACTICAL ASPECTS OF CARDIOVASCULAR MAGNETIC RESONANCE PERFUSION IMAGING This chapter has mostly focused on theoretical aspects of myocardial perfusion imaging, and the presentation was in part motivated by a desire to provide useful insights for optimizing myocardial perfusion studies and for avoid ing some pitfalls. We now summarize some practical rules of thumb for myocardial perfusion imaging protocols that have proven to be of relevance in practice. 1. Contrast agent injection. Because of the dependence of myocardial contrast enhancement on the arterial input of contrast, it becomes important to inject the tracer sufficiently fast so that different levels of tissue blood flow can be sensitively discriminated. The use of a power injector with an injection rate of 3 mL/sec, in particular, for studies during coronary hyperemia, is important. 2. Temporal resolution of measurements. The temporal resolution should allow updating of images from every slice at a rate equal to the heart rate. This rule of thumb automatically compensates for the higher temporal reso lution requirements during stress compared with rest. 3. Spatial resolution. Mild reductions of myocardial blood flow as a result of an epicardial lesion manifest them selves first in the endocardial layer. A spatial resolution of 2.5 mm or less should be used to resolve transmural variations in blood flow. For rest studies, the temporal resolution can be reduced from one R R interval to two R R intervals, to accommodate the requirements for spatial resolution, but for hyperemia, the one R R interval tends to be closer to the critical limit for tempo ral resolution. 4. The CMR sequence for contrast residue detection. T1 weighted imaging can be rendered relatively insensitive to the effects of water exchange by the use of higher flip angles (above the Ernst angle), and this simplifies the conversion of signal time curves into residue curves. There is some evidence that avoidance of high contrast dosages (e.g., > 0.1 mmol/kg for Gd chelates) and high spatial resolution may be effective measures against arti facts and the appearance of a transitional dark rim at the endocardial border. The rationale for these measures is that: (1) the susceptibility difference between blood loaded with contrast and tissue gives rise to signal loss at the endocardial border, which is less likely to be observed with lower contrast dosages; and (2) Gibbs

66 Cardiovascular Magnetic Resonance

ringing, which may be one source of the dark rim arti fact, is reduced by increasing the number of sampling points (i.e., spatial resolution, in this case). 5. Measurement of arterial input. For a semiquantitative or quantitative analysis, contrast dosages need to be kept below approximately 0.05 mmol/kg body weight for Gd chelates. 6. Parallel imaging. For near complete coverage of the heart, the use of parallel imaging techniques has become a blessing and a necessity. Nevertheless, myocardial per fusion studies are relatively signal intensity “starved,” and for this reason, acceleration factors have to be used conservatively to avoid excessive noise and blurring. Acceleration factors on the order of two are recom mended, although higher factors may work well when SSFP techniques are used. 7. Analysis and modeling. The rate of contrast enhance ment in a myocardial tissue region (“up slope”) has empirically been proven to provide a relatively good surrogate measure of blood flow, but it should be appropriately normalized by a measure of contrast enhancement at the arterial input.69 The resulting perfusion index can be determined for rest and stress, and the ratio can provide an approximate measure of the perfusion reserve. Several CMR equipment vendors have now included this up slope method in their cardiac perfusion analysis packages. Other parameters, such as the maximum amplitude of contrast enhance ment, are less sensitive than the up slope parameter to changes in tissue blood flow. Absolute measures of myocardial blood flow can be determined if contrast enhancement is linearly proportional to contrast concentration in blood. Blood flow estimates are obtained by deconvolution of the tissue curves with arterial input, either by using a model independent approach, or by modeling. Currently, these latter approaches for blood flow quantification are used only in a research setting, and commercial software is not yet available.

CONCLUSION In the context of assessing blood flow in the myocardial microcirculation, the CMR first pass imaging technique can address several clinical issues, if the appropriate conditions for spatial and temporal resolution can be met. Valuable quantitative and largely observer independent information related to the pathophysiology of ischemic heart disease becomes available with myocardial perfusion imaging. CMR perfusion reserve measurements can be applied to assess the degree of atherosclerosis, to assess the functional significance of coronary artery lesions, and to evaluate myo cardial viability. In milder forms of ischemic heart disease and ischemic cardiomyopathies, CMR at rest and during stress is a good test to probe for more subtle perfusion lim itations that may be limited to subendocardium or that result in only moderate blood flow reductions. For these applica tions, quantitative CMR methods have been successfully developed and validated to provide a true measure of myo cardial blood flow.

1. Stollfuss JC, Kosa I, Ziegler S, Schwaiger M, et al. Quantitative assessment of myocardial perfusion: is it of clinical relevance? Q J Nucl Med. 1996;40(1):76–84. 2. Schwitter J, Wacker CM, Wilke N, et al. Magnetic resonance Imaging for Myocardial Perfusion Assessment in Coronary artery disease Trial (MR-IMPACT II): a phase III multicenter, multivendor trial comparing perfusion cardiac magnetic resonance versus single photon emission computed tomography for the detection of coronary artery disease. Circulation. 2006;114(18):II–806. 3. Gould KL, Kirkeeide RL, Buchi M. Coronary flow reserve as a physiologic measure of stenosis severity. J Am Coll Cardiol. 1990;15(2): 459–474. 4. Gould KL, Lipscomb K. Effects of coronary stenosis on coronary flow reserve and resistance. Am J Cardiol. 1974;34:48–55. 5. Wilson RF, Wyche K, Christensen BV, et al. Effects of adenosine on human coronary arterial circulation. Circulation. 1990;82(5): 1595–1606. 6. Wilson RF. Assessment of the human coronary circulation using a Doppler catheter. Am J Cardiol. 1991;67(14):44D–56D. 7. Pijls NH, De Bruyne B, Peels K, et al. Measurement of fractional flow reserve to assess the functional severity of coronary-artery stenoses [see comments]. N Engl J Med. 1996;334(26):1703–1708. 8. Panting JR, Gatehouse PD, Yang GZ, et al. Abnormal subendocardial perfusion in cardiac syndrome X detected by cardiovascular magnetic resonance imaging. N Engl J Med. 2002;346(25):1948–1953. 9. Wilke N, Jerosch-Herold M, Wang Y, et al. Myocardial perfusion reserve: assessment with multisection, quantitative, first-pass MR imaging. Radiology. 1997;204(2):373–384. 10. Wang L, Jerosch-Herold M, Jacobs DR, et al. Coronary risk factors and myocardial perfusion in asymptomatic adults: the Multi-Ethnic Study of Atherosclerosis (MESA). J Am Coll Cardiol. 2006;47(3):565–572. 11. Wang L, Jerosch-Herold M, Jacobs DR, et al. Coronary artery calcification and myocardial perfusion in asymptomatic adults the MESA (Multi-Ethnic Study of Atherosclerosis). J Am Coll Cardiol. 2006;48 (5):1018–1026. 12. Rosen BD, Lima JA, Nasir K, et al. Lower myocardial perfusion reserve is associated with decreased regional left ventricular function in asymptomatic participants of the multi-ethnic study of atherosclerosis. Circulation. 2006;114(4):289–297. 13. Rosen BD, Edvardsen T, Lai S, et al. Left ventricular concentric remodeling is associated with decreased global and regional systolic function: the Multi-Ethnic Study of Atherosclerosis. Circulation. 2005; 112(7):984–991. 14. Decking UK, Schrader J. Spatial heterogeneity of myocardial perfusion and metabolism. Basic Res Cardiol. 1998;93(6):439–445. 15. Arai AE, Grauer SE, Anselone CG, et al. Metabolic adaptation to a gradual reduction in myocardial blood flow. Circulation. 1995;92 (2):244–252. 16. Bache RJ, Schwartz JS. Effect of perfusion pressure distal to a coronary stenosis on transmural myocardial blood flow. Circulation. 1982;65 (5):928–935. 17. Canty IJ, Piotrowski B, Klocke F. Relative reductions in subendocardial and full-thickness flow during pharmacologic vasodilation as an index of functional stenosis seventy. J Am Coll Cardiol. 1990;15:182A. 18. Schwitter J, Nanz D, Kneifel S, et al. Assessment of myocardial perfusion in coronary artery disease by magnetic resonance: a comparison with positron emission tomography and coronary angiography. Circulation. 2001;103(18):2230–2235. 19. Muehling OM, Panse P, Jerosch-Herold M, et al. Cardiac magnetic resonance perfusion imaging identifies transplant arteriopathy by a reduced endomyocardial resting perfusion. J Heart Lung Transplant. 2005;24(8):1122–1123. 20. Wagner A, Mahrholdt H, Holly JA, et al. Contrast-enhanced MRI and routine single photon emission computed tomography (SPECT) perfusion imaging for detection of subendocardial myocardial infarcts: an imaging study. Lancet. 2003;361(9355):374–379. 21. Bolli R, Zhu W-X, Thornby JI, et al. Time course and determinants of recovery of function after reversible ischemia in conscious dogs. Am J Physiol. 1988;254:H102–H114. 22. Tarantini G, Cacciavillani L, Corbetti F, et al. Duration of ischemia is a major determinant of transmurality and severe microvascular obstruction after primary angioplasty: a study performed with

23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36.

37.

38.

39. 40.

41. 42. 43. 44. 45.

contrast-enhanced magnetic resonance. J Am Coll Cardiol. 2005;46 (7): 1229–1235. Epstein FH, Wolff SD, Arai AE. Segmented k-space fast cardiac imaging using an echo-train readout. Magn Reson Med. 1999;41(3): 609–613. Li W, Storey P, Chen Q, et al. Dark flow artifacts with steady-state free precession cine MR technique: causes and implications for cardiac MR imaging. Radiology. 2004;230(2):569–575. Hunold P, Maderwald S, Eggebrecht H, et al. Steady-state free precession sequences in myocardial first-pass perfusion MR imaging: comparison with TurboFLASH imaging. Eur Radiol. 2004;14(3):409–416. Chen Z, Prato FS, McKenzie CA. T1 Fast Acquisition Relaxation Mapping (T1-FARM): An Optimised Reconstruction. IEEE Trans Med Imag. 1998;17:155–160. Bellamy DD, Pereira RS, McKenzie CA, et al. Gd-DTPA bolus tracking in the myocardium using T1 fast acquisition relaxation mapping (T1 FARM). Magn Reson Med. 2001;46(3):555–564. Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: sensitivity encoding for fast MRI. Magn Reson Med. 1999;42(5): 952–962. Pruessmann KP, Weiger M, Boesiger P. Sensitivity encoded cardiac MRI. J Cardiovasc Magn Reson. 2001;3(1):1–9. Sodickson DK, Manning WJ. Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays. Magn Reson Med. 1997;38(4):591–603. Niendorf T, Sodickson DK. Parallel imaging in cardiovascular MRI: methods and applications. NMR Biomed. 2006;19(3):325–341. Madore B. Using UNFOLD to remove artifacts in parallel imaging and in partial-Fourier imaging. Magn Reson Med. 2002;48(3):493–501. Madore B. UNFOLD-SENSE: a parallel MRI method with self-calibration and artifact suppression. Magn Reson Med. 2004;52(2):310–320. Di Bella EV, Wu YJ, Alexander AL, et al. Comparison of temporal filtering methods for dynamic contrast MRI myocardial perfusion studies. Magn Reson Med. 2003;49(5):895–902. Hansen MS, Battes C, Tsao J, et al. Accelerated dynamic Fourier velocity encoding by exploiting velocity-spatio-temporal correlations. MAGMA. 2004;17(2):86–94. Tsao J, Kozerke S, Boesiger P, Pruessmann KP. Optimizing spatiotemporal sampling for k-t BLAST and k-t SENSE: application to highresolution real-time cardiac steady-state free precession. Magn Reson Med. 2005;53(6):1372–1382. Donahue KM, Weisskoff RM, Chester DA, et al. Improving MR quantification of regional blood volume with intravascular T1 contrast agents: accuracy, precision, and water exchange. Magn Reson Med. 1996;36(6):858–867. Wiart M, Carme S, Mai W, et al. In vivo quantification of regional myocardial blood flow: validity of the fast-exchange approximation for intravascular T1 contrast agent and long inversion time. Magn Reson Med. 2006;56(2):340–347. Donahue KM, Burstein D, Manning WJ, Gray ML, et al. Studies of GdDTPA relaxivity and proton exchange rates in tissue. Magn Reson Med. 1994;32:66–76. Wacker CM, Wiesmann F, Bock M, et al. Determination of regional blood volume and intra-extracapillary water exchange in human myocardium using Feruglose: first clinical results in patients with coronary artery disease. Magn Reson Med. 2002;47(5):1013–1016. Donahue KM, Weisskoff RM, Bogdonoff AA, et al. Measurement of vascular volume fraction and water permeability with intravascular contrast agents. In: Third Scientific Meeting. Nice: SMR; 1995. Jerosch-Herold M, Wilke N, Stillman AE, Wilson RF. MR quantification of the myocardial perfusion reserve with a Fermi function model for constrained deconvolution. Med Phys. 1998;25(1):73–84. Li X, Huang W, Yankeelov TE, et al. Shutter–speed analysis of contrast reagent bolus–tracking data: preliminary observations in benign and malignant breast disease. Magn Reson Med. 2005;53(3):724–729. Fidler F, Wacker CM, Dueren C, et al. Myocardial perfusion measurements by spin-labeling under different vasodynamic states. J Cardiovasc Magn Reson. 2004;6(2):509–516. Fieno DS, Shea SM, Li Y, et al. Myocardial perfusion imaging based on the blood oxygen level-dependent effect using T2-prepared steady-state free-precession magnetic resonance imaging. Circulation. 2004;110 (10):1284–1290.

Cardiovascular Magnetic Resonance 67

4 MYOCARDIAL PERFUSION IMAGING THEORY

References

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

46. Klocke FJ, Li D. Testing coronary flow reserve without a provocative stress: a “BOLD” approach. J Am Coll Cardiol. 2003;41(5):841–842. 47. Tsekos NV, Wilke N, Xu Y, et al. Magnetization transfer contrast in assessment of myocardial perfusion. Radiology. 1993;189(P):166. 48. Wolff SD, Balaban RS. Magnetization transfer contrast (MTC) and tissue water proton relaxation in vivo. Magn Reson Med. 1989;10:135–144. 49. Detre JA, Zhang W, Roberts DA, et al. Tissue specific perfusion imaging using arterial spin labeling. NMR Biomed. 1994;7(1–2):75–82. 50. Wacker CM, Fidler F, Dueren C, et al. Quantitative assessment of myocardial perfusion with a spin-labeling technique: preliminary results in patients with coronary artery disease. J Magn Reson Imaging. 2003;18(5):555–560. 51. Zhang H, Shea SM, Park V, et al. Accurate myocardial T1 measurements: toward quantification of myocardial blood flow with arterial spin labeling. Magn Reson Med. 2005;53(5):1135–1142. 52. Bauer WR, Nadler W, Bock M, et al. The relationship between the BOLD-induced T(2) and T(2)(*): a theoretical approach for the vasculature of myocardium. Magn Reson Med. 1999;42(6):1004–1010. 53. Schwaiger M. Myocardial perfusion imaging with PET. J Nucl Med. 1994;35(4):693–698. 54. Jerosch-Herold M, Swingen C, Seethamraju RT. Myocardial blood flow quantification with MRI by model-independent deconvolution. Med Phys. 2002;29(5):886–897. 55. Zierler K. Indicator dilution methods for measuring blood flow, volume, and other properties of biological systems: a brief history and memoir. Ann Biomed Eng. 2000;28(8):836–848. 56. Zierler KL. Theoretical basis of indicator-dilution methods for measuring flow and volume. Circ Res. 1962;10:393–407. 57. Zierler KL. Equations for measuring blood flow by external monitoring of radioisotopes. Circ Res. 1965;16:309–321. 58. Axel L. Tissue mean transit time from dynamic computed tomography by a simple deconvolution technique. Invest Radiol. 1983;18(1):94–99. 59. Christian TF, Rettman DW, Aletras AH, et al. Absolute myocardial perfusion in canines measured by using dual-bolus first-pass MR imaging. Radiology. 2004;232(3):677–684.

68 Cardiovascular Magnetic Resonance

60. Muhling OM, Wang Y, Panse P, et al. Transmyocardial laser revascularization preserves regional myocardial perfusion: an MRI first pass perfusion study. Cardiovasc Res. 2003;57(1):63–70. 61. Hsu LY, Rhoads KL, Holly JE, et al. Quantitative myocardial perfusion analysis with a dual-bolus contrast-enhanced first-pass MRI technique in humans. J Magn Reson Imaging. 2006;23:315–322. 62. Bassingthwaighte JB, Raymond GR, Chan JIS. Principles of tracer kinetics. In: Zaret BL, Beller GA, eds. Nuclear Cardiology: State of the Art and Future Directions. St. Louis: Mosby-Year Book; 1993:3–23. 63. Kroll K, Wilke N, Jerosch-Herold M, et al. Accuracy of modeling of regional myocardial flows from residue functions of an intravascular indicator. Am J Physiol (Heart Circ Physiol). 1996;40:H1643–H1655. 64. Jerosch-Herold M, Wilke N, Wang Y, et al. Direct comparison of an intravascular and an extracellular contrast agent for quantification of myocardial perfusion. Int J Card Imaging. 1999;15(6):453–464. 65. Ko¨stler H, Ritter C, Baunach D, et al. Determination of the arterial input function in high dose radial myocardial perfusion imaging. In: Scientific Sessions of the International Society of Magnetic Resonance in Medicine. Miami Beach, FL: ISMRM; 2005. 66. Gatehouse PD, Elkington AG, Ablitt NA, et al. Accurate assessment of the arterial input function during high-dose myocardial perfusion cardiovascular magnetic resonance. J Magn Reson Imaging. 2004;20 (1):39–45. 67. Cernicanu A, Axel L. Theory-based signal calibration with singlepoint T1 measurements for first-pass quantitative perfusion MRI studies. Acad Radiol. 2006;13(6):686–693. 68. Johansson E, Mansson S, Wirestam R, et al. Cerebral perfusion assessment by bolus tracking using hyperpolarized 13C. Magn Reson Med. 2004;51(3):464–472. 69. Al-Saadi N, Nagel E, Gross M, et al. Noninvasive detection of myocardial ischemia from perfusion reserve based on cardiovascular magnetic resonance. Circulation. 2000;101(12):1379–1383.

Cardiovascular Magnetic Resonance Tagging Assessment of Left Ventricular Diastolic Function Matthias Stuber

Left ventricular (LV) diastolic function has been recog nized as an important factor in the pathophysiology of many common cardiovascular diseases. Dilated and hypertrophic cardiomyopathies, coronary artery disease, and systemic hypertension are all associated with abnor mal LV filling dynamics. Diastolic dysfunction has also been increasingly appreciated as a major cause of heart failure, especially in the elderly. Although invasive hemodynamic measures/assessment of diastole are con sidered the gold standard, echocardiographic methods, including tissue Doppler imaging, have gained greater use in the clinical assessment of LV diastolic function because of their noninvasive acquisition, which greatly facilitates serial assessments. With the advent of parallel imaging, cardiovascular magnetic resonance (CMR) tech niques, three dimensional (3D) data collection, spiral and steady state free precession (SSFP) imaging, and higher magnetic field strength, the prolonged tag per sistence permits easier access to diastole. Together with rapid state of the art software analysis tools, quantifi cation of LV diastolic wall motion can now easily be performed.

CARDIAC MOTION During systole, the heart performs a complex motion analogous to “wringing” a towel, and as a result, the base and apex rotate in opposite directions. There is counter clockwise rotation at the apex and clockwise rotation at the base.1 In parallel, the atrioventricular valvular plane (basal LV and right ventricle) descends toward the apex. (The apex is relatively stationary.) The lateral free wall of the right ventricle performs a more pronounced long axis contraction than the lateral wall of the LV.2 During isovolumic relaxation, myofibrils return to their resting state from the contracted state. This process is accompa nied by a rapid untwisting at the apex, whereas the vol ume and cavity shape of the heart remain nearly unchanged. This rapid untwisting typically lasts less than 75 msec and precedes the early passive filling phase of the ventricles. During this filling phase, practically no rotational components can be seen at the apex of the healthy heart.3

Assessment of Cardiac Rotation/ Motion: Non-Cardiovascular Magnetic Resonance Methods With echocardiographic imaging, the myocardium has rel atively poor internal structure because of the absence of structural landmarks, making quantification of parameters, such as rotation, stress, and strain, limited.4,5 Several inva sive methods to examine cardiac motion during diastole have been reported. One approach is the surgical/invasive implantation of tantalum markers into the midwall of the myocardium.6 In combination with X ray angiography, the motion of these markers can then be recorded with high temporal and spatial resolution. Using such an approach, alterations in diastolic untwisting have been observed in patients who have undergone heart transplan tation shortly before rejection.7 Although this method is very powerful, it is invasive, requires ionizing radiation, and is inappropriate for routine clinical use. Alternative angiographic “markers,” such as tracking of the bifurcations of the coronaries,8 suffer from the limited number of land marks and their irregular geometric distribution. Further more, they only provide motion information about the epicardial layers of the myocardium.

Assessment of Cardiac Rotation/ Motion: Cardiovascular Magnetic Resonance Methods Similar to echocardiography, conventional cine CMR images do not provide information about the internal struc ture of the myocardium. However, a CMR myocardial tag ging technique, spatial modulation of magnetization (SPAMM), originally proposed by Zerhouni9 and Axel10 and further developed and refined by others, offers the opportunity to assess strain noninvasively. With these methods, the magnetization of the muscle tissue is spatially modulated, or “tagged,” by the application of a specific time series of radiofrequency (RF) pulses and magnetic field gradients. The tagging is typically applied immediately after the R wave of the electrocardiogram. Images are then Cardiovascular Magnetic Resonance 69

5 CARDIOVASCULAR MAGNETIC RESONANCE TAGGING ASSESSMENT OF LEFT VENTRICULAR DIASTOLIC FUNCTION

CHAPTER 5

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

acquired during successive heart phases in which the tags may be identified as dark lines or grids. Because these tags are spatially fixed with respect to the muscle tissue at the time of the tag application, local myocardial motion can be derived from the translation, rotation, and distortion of the tags on the myocardium. However, because of the relaxation effects, the tags fade rapidly and cannot be reli ably detected after end systole (approximately 300 msec after tag application). This is a serious drawback for the quantification of systolic and diastolic dynamics of the heart wall. Another limitation is that this approach does not compensate for through plane motion. The problems associated with tag fading as well as through plane motion2 can be overcome by the application of complementary spatial modulation of magnetization (CSPAMM) CMR tagging approaches.3,11 This technique is based on a subtraction of two acquisitions. For both acqui sitions, the magnetization of the tissue is modulated or “tagged” locally, and a thin slice and a thick slice are imaged during the subsequent imaging procedure. If the modula tion function preceding the second acquisition is inverted with respect to the first tagging procedure, subtraction of the two acquisitions leads to signal that is derived only from the initially tagged thin slice. This approach both pro longs the persistence of the tags until late diastole and assures that the same tissue plane is imaged in the multiple heart phase images. In this chapter, the technique is described in more detail and initial results associated with diastolic motion components of the human heart are discussed.

METHODS Complementary Spatial Modulation of Magnetization and Slice Following The CSPAMM myocardial tagging technique involves the periodic modulation of the magnetization in a thin slice of the myocardium (dz; Fig. 5 1). A sinusoidal modulation of the magnetization is typically performed immediately after the detection of the R wave of the electrocardiogram.

dz

ds

Figure 5 1 Slice following principle. An initially tagged planar slice of the thickness dz translates and distorts during the cardiac cycle. A volume of the thickness ds is imaged multiple times during the cardiac cycle. This volume must encompass the potential extent of the motion of the tagged, thin slice. 70 Cardiovascular Magnetic Resonance

Subsequently, a thick slice (ds; see Fig. 5 1) encompassing the expected full extent of motion of the selected thin slice is imaged periodically (n heart phase images) during the cardiac cycle. The procedure, consisting of labeling of a thin slice and subsequent imaging of a thicker volume, is performed twice with an inverted modulation of the mag netization for the second experiment. Subtraction of the two acquisitions leads to an image derived from the signal coming from the labeled part of the magnetization in the thin slice. Because of the subtraction procedure, the signal coming from the thick volume outside the tagged slice (ds; see Fig. 5 1) is suppressed. Therefore, the problem of through plane motion is eliminated because only a projec tion of the same tissue elements is visualized in the images. The signal from the tagged thin slice can be decomposed into two parts, with the first part holding the tagging infor mation and a second part (responsible for the fading of the tags) built up as a function of time. For CSPAMM, this sec ond component is also suppressed by the subtraction pro cedure. As a result, only signal derived from the tagged component of the magnetization remains after subtraction, and fading of the tags is avoided. For a constant tag con trast and a maximized signal to noise ratio in systolic and diastolic images, a series of ramped RF excitation angles must be used.11 Typically, double oblique short axis sections of the myo cardium are tagged with a slice thickness of 6 to 8 mm. Subsequently, 16 to 20 sequential heart phase images are acquired with a temporal resolution (Dt) of 35 msec. With this high temporal resolution, rapid motion components, such as diastolic untwisting at the apex, can be identified readily. Because the ratio of wanted to unwanted signal components must be optimized, the thickness of the imaged volume (ds; see Fig. 5 1) must be reduced to a min imum. Therefore, it depends on the level of the tagged slice with respect to its level on the long axis. For basal LV images, where a long axis contraction of more than 20 mm may be expected for the lateral free wall of the right ventricle,2 a slice thickness of 30 mm is typically chosen. For equatorial slices, a slice thickness of 25 mm is appro priate, and at the apex, a slice thickness of 20 mm is used because of reduced through plane motion. For suppression of breathing induced motion artifacts, a repetitive breath hold scheme3,11 or single breath hold techniques12 can be applied. Considering the location of the relevant tagging informa tion in k space, a reduced k space acquisition scheme can be applied.11,13 Hereby, two sets of orthogonally line tagged images are acquired. Subsequent combination of these acquisitions results in grid tagged images (Figs. 5 2 and 5 3). With this method, acquisition time is significantly reduced and image resolution perpendicular to the line tags is not affected. Although traditionally segmented k space gradient echo techniques were used for signal readout, both T1 relaxation and serial RF excitations for imaging are responsible for the fading of the tags. T1 can only be increased by going to a higher magnetic field strength, and the number of RF excitations can be reduced by using alternative imaging sequences with fewer RF excitations. Hereby, echo planar imaging12 (see Fig. 5 2) and more recently also spiral imaging14 proved to be very valuable alternatives (see Fig. 5 3). Especially with spiral imaging, a very high

t=72 msec

t=109 msec

t=146 msec

t=183 msec

t=220 msec

t=257 msec

t=294 msec

t=331 msec

t=368 msec

t=405 msec

t=442 msec

t=479 msec

t=516 msec

t=553 msec

t=590 msec

t=627 msec

t=664 msec

t=701 msec

t=738 msec

Figure 5 2 Twenty apical left ventricular short axis images in a healthy adult subject acquired with complementary spatial modulation of the magnetization cardiovascular magnetic resonance tagging. The systolic images include frames 1 to 9 (331 msec after the R wave of the electrocardiogram), and the diastolic images are shown in frames 10 to 20 (368 to 738 msec). The temporal resolution in these images is 37 msec, and two line tagged acquisitions were combined to create a grid tagged image off line. During systole, a counterclockwise apical rotation is seen, followed by a rapid clockwise untwisting in frames 10 to 15 (368 to 553 msec) during early diastole. This precedes the filling phase (frames 16 to 20; 590 to 738 msec) of the ventricles.

temporal resolution, with up to 70 frames/sec (14 msec temporal resolution), could be obtained, whereas a spatial resolution of 1.5 mm was easily possible14 (see Fig. 5 3). Because spiral imaging is susceptible to off resonance blur ring in the images, spectral spatial RF excitations15 are a stringent requirement.14 Together with echoplanar and spi ral readouts, the use of SSFP was exploited by Herzka and colleagues,16 who found that SSFP leads to an improved tagging contrast compared with more conventional seg mented k space gradient echo imaging. By combining CSPAMM with SSFP imaging, Zwanenburg and associates17 showed that CSPAMM tagged images can easily be acquired in a single breath hold. With the advent of parallel and 3D imaging, 3D assess ment of myocardial motion based on 3D CSPAMM lattice tagging was reported by Ryf and colleagues.18 Simulta neously, the same authors proposed an extension to a pre viously reported analysis procedure19 that enables relatively time efficient analysis and quantification of both systolic and diastolic motion of the heart.20 However, acquisition times were lengthy and only practical in coa ched breathing patterns in well trained subjects.

Evaluation For the extraction of motion data from the tagged time series of images, sophisticated image analysis tools are needed. Although the identification of tags was very labor intensive and took hours in the early years, sophis ticated algorithms that reduce quantification of tagged time series of cardiac images to seconds are now readily available,19 and variants have been successfully imple mented.20 Quantification of tagged images involves the identification of the tags in each heart phase image. With these automated or semiautomated algorithms, the tags may be identified with an accuracy that exceeds the image resolution.21 If the grid intersection points are identified for all heart phase images, local trajectories on the myo cardium (Fig. 5 4) are defined and motion specific para meters (e.g., strain, rotation, rotation velocity, radial or circumferential shortening or shear between epi and endocardial muscle layers of the myocardium) can be derived. Using this approach, several studies of healthy subjects and patients with myocardial infarction,22 Cardiovascular Magnetic Resonance 71

5 CARDIOVASCULAR MAGNETIC RESONANCE TAGGING ASSESSMENT OF LEFT VENTRICULAR DIASTOLIC FUNCTION

t=35 msec

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

17 msec

332 msec

542 msec

17 msec

332 msec

542 msec

24 msec

339 msec

549 msec

Figure 5 3 Spiral complementary spatial modulation of the magnetization myocardial tagging using a spectral spatial excitation for fat suppression in each cine frame. Midventricular short and long axis views of a line tagged and a grid tagged myocardium in a healthy subject. The tagline distance is 4 mm and the temporal resolution is 35 msec. The time after the R wave is indicated on the images. On the grid tagged, short axis view, 50 tagline intersections can be observed.

hypertrophied cardiomyopathy,23 and aortic stenosis with pathologically hypertrophied hearts, as well as athletes with physiologic hypertrophy, were evaluated for apical untwisting during diastole.24 Untwisting velocity, time to peak untwisting velocity, and early diastolic strain can be determined as an index of diastolic function. Time to peak untwisting velocity (untwisting time; Table 5 1) is defined as the time delay between the point in time of minimum inner cavity area and the maximum untwisting velocity (Fig. 5 5). 72 Cardiovascular Magnetic Resonance

RESULTS Images Figure 5 2 shows 20 heart phase images with a temporal res olution of 37 msec acquired at the apex of a healthy subject.12 The grid structure remains visible, with a high contrast up to the last acquisition in late diastole (>700 msec). No fading of the tags is seen in the images. Therefore, the method is well suited for the quantification of diastolic heart wall motion.

At the apex of the healthy heart, a counterclockwise rotation during systole can be seen (see Fig. 5 2, phases 1 to 9; see Fig. 5 5, phase 1). This systolic rotation is followed by a rapid untwisting during isovolumic relaxation (see Fig. 5 2, phases 10 15 and early diastole; see Fig. 5 5, phase 2). This untwisting phase is typically followed by the filling phase of the ventricles, where little rotational component is seen (see Fig. 5 5B, phase 3).3,25 An identical separation of early diastolic apical untwisting and the filling phase of the ventricles is seen in highly competitive athletes with physiologically hypertrophied hearts. Neither the api cal peak rotation angle nor diastolic untwisting time is changed in the hypertrophied hearts of athletes compared with healthy control subjects (see Table 5 1). However, among patients with LV hypertrophy as a result of pressure overload/aortic stenosis, a completely different apical rota tion pattern is seen (see Fig. 5 5). The end systolic peak rotation is significantly increased (P < 0.01) in comparison to the athletes or healthy subjects, and no separation of untwisting and filling can be seen during diastole. Untwist ing and filling occur simultaneously, and the point in time of maximum untwisting velocity is significantly delayed in these patients (P < 0.01; see Fig. 5 5).

Figure 5 4 End diastolic apical cardiovascular magnetic resonance image acquired in a healthy adult subject. The grid tagged image is overlain with the corresponding local trajectories. The arrows start at the beginning of systole and at end diastole.

Table 5-1 Peak Rotation at the Apex, Maximum Rotation Velocity During Diastolic Untwisting and Untwisting Time for 12 Aortic Stenosis Patients, 12 Athletes and 11 Healthy Volunteers* Patients Aortic Stenosis Athletes 22 Volunteers 22

Peak Rotation (Deg)

Untwisting Velocity (Deg/s)

Untwisting Time (% ES)

125 62 72

8029 568 5517

326 178 168

22

14 12 100

10 8 6 4 2 0 –2

Systole

–4

0

A

20

40

60

Diastole

80

Rotation (x100% ES)

Apical rotation angle (degrees)

*The untwisting time is related to the duration of systole. Data are expressed as mean  one standard deviation. Modified and reprinted with permission of the American Heart Association. Stuber et al. Alterations in the local myocardial motion pattern in patients suffering from pressure overload due to aortic stenosis. Circulation. 1999;100:363.

1 80 60

20

0

B Aortic stenosis

3

0

100 120 140 160 180 200

Time (% ES)

2

40

Healthy

10

20

30

40

50

60

70

80

90 100

Apical area reduction (x 100% ES) Rowers

Figure 5 5 A, Cross sectional apical rotation velocity of athletes, healthy adults, and patients with aortic stenosis. The values are mean  1 standard deviation. The time axis is normalized to the end systolic point of the cardiac cycle. The time point of maximum diastolic untwisting velocity (arrows) is delayed in the patients with aortic stenosis compared with the athletes or the healthy control subjects. Apical rotation velocity is identical in athletes and control subjects. B, Left ventricular rotation area loop (apical plane) in healthy control subjects, rowers, and patients with aortic stenosis. The loop is separated in isovolumic contraction (1), ejection (2), isovolumic relaxation (3), and filling of the left ventricle. Both rotation and area change of the inner lumen at the apex are related to their maximum values (100%). ES, end systole. Cardiovascular Magnetic Resonance 73

5 CARDIOVASCULAR MAGNETIC RESONANCE TAGGING ASSESSMENT OF LEFT VENTRICULAR DIASTOLIC FUNCTION

Apical Rotation

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Pressure overload LV hypertrophy results in the addition of new sarcomeres in parallel.26 Furthermore, an increase in the amount of collagen, with a consequently increased elastic stiffness, has been reported.27 This rearranged fiber architec ture, together with the increased stiffness of the muscle tissue, may explain the alterations of the diastolic rotation pattern with a prolonged and delayed apical untwisting in these patients. The prolongation of untwisting results in an overlap with early diastolic filling and presumably impedes normal filling. Thus, a prolongation of early diastolic untwisting may be responsible for the occurrence of diastolic dysfunc tion in these patients. In patients with myocardial infarction, a prolonged untwisting phase with an overlap of diastolic untwisting and filling has also been observed28 and end systolic apical peak rotation is usually severely reduced.

and susceptibility to diaphragmatic drift or misregistration in serial breath holds must be considered. However, with the availability of parallel imaging, echo planar or spiral imag ing, the number of RF excitations for imaging can be mini mized. As an alternative to CSPAMM, more conventional tagging7,31 in combination with SSFP may support an improved tagging contrast to noise ratio, as does 3.0 Tesla (T) imaging because of an increase in T1 (from 850 msec at 1.5 T to 1200 msec at 3.0 T for myocardium and from 1200 msec at 1.5 T to 1650 msec for blood at 3.0 T). For these reasons, access to early diastolic myocardial motion may be feasible, even without the availability of CSPAMM. Nevertheless, the investigation of heart wall motion is still the subject of basic research, and appropriate parameters for the quantification of diastolic dysfunction and threshold values for normal subjects remain to be fully defined.

Strain Measurements In a study by Ennis and colleagues,23 systolic and diastolic strain was computed in both healthy adults and in patients with hypertrophic cardiomyopathy. In this study, CSPAMM was extended with a sophisticated phase reordering algo rithm (CAPTOR) that enables more complete coverage of the entire cardiac cycle. In these patients, a significantly reduced strain was observed around the circumference of the heart in early diastole. In mid diastole, no major differ ence in strain between the patients and the healthy adults was reported. Similarly, the strain rates were significantly reduced in these patients during early diastole. CMR assessment of diastolic function is beginning to be studied in broader populations. A population based study of the multi ethnic subclinical atherosclerosis (MESA) study showed evidence of diastolic dysfunction in asymptomatic individuals with LV hypertrophy.29 Despite similar regional systolic strain and strain rate among those with and with out LV hypertrophy, the regional diastolic strain rate was significantly reduced among those with LV hypertrophy. A provocative study in a small group suggested that diastolic untwisting dysfunction could be identified by SPAMM after a sedentary rest in previously active adult subjects.30

Limitations The additional value of CMR tagging applied for the quantifi cation of diastolic function remains to be more fully investi gated compared with gold standard techniques. Currently, the CMR technique is not widely accessible to clinical cardiol ogists and the CSPAMM sequence is not currently available on all vendors’ systems. Furthermore, because CSPAMM is based on a subtraction technique, scanning time is doubled

CONCLUSION The CSPAMM myocardial tagging technique is a noninvasive method for the quantification of local heart wall motion. Because of the suppressed fading of the tags and the accessi bility to the diastolic phase of the cardiac cycle, CSPAMM is well suited for the characterization of the diastolic portion of the cardiac cycle. Moreover, by the application of a slice fol lowing procedure, the effects of through plane motion can be suppressed and the same tissue elements can be tracked. Because of the relatively high temporal resolution of the data, rapid cardiac motion components, such as apical diastolic untwisting, can be recorded. Data suggest that pathologic LV hypertrophy in patients with aortic stenosis can be differentiated from hypertrophy in athletic hearts.24 Alterations in diastolic untwisting are seen in patients with pressure overload as a result of aortic stenosis, whereas none are observed in athletes’ hearts. Even though the athletes’ hearts were significantly larger than control sub jects’ hearts, apical diastolic untwisting remains unchanged. Similarly, and using CSPAMM, significant alterations in early diastolic strain and strain rate have been reported in patients with hypertrophic cardiomyopathy.23 The current data derived from CSPAMM myocardial tagging clearly suggest that alterations in the diastolic phase of the cardiac cycle can be recorded readily. With the advent of parallel imaging, higher magnetic field strength, and more advanced imaging sequences, persistence of tags can be prolonged, even without CSPAMM, enabling access to early diastolic motion of the left ventricle. Although CMR myocardial tagging requires off line computer analysis, quantification is now a matter of seconds and the technique may offer a new tool for the evaluation of diastolic wall relaxation in healthy and diseased states.

References 1. Maier SE, Fischer SE, McKinnon GC, Hess OM, Krayenbuehl HP, Boesiger P. Evaluation of left ventricular segmental wall motion in hypertrophic cardiomyopathy with myocardial tagging. Circulation. 1992;86:1919–1928. 2. Rogers Jr WJ, Shapiro EP, Weiss JL, et al. Quantification of and correction for left ventricular systolic long-axis shortening by magnetic resonance tissue tagging and slice isolation. Circulation. 1991;84:721–731. 74 Cardiovascular Magnetic Resonance

3. Fischer SE, McKinnon GC, Scheidegger MB, Prins W, Meier D, Boesiger P. True myocardial motion tracking. Magn Reson Med. 1994;31:401–413. 4. D’hooge J, Heimdal A, Jamal F, et al. Regional strain and strain rate measurements by cardiac ultrasound: principles, implementation and limitations. Eur J Echocardiogr. 2000;1:154–170. 5. Gilman G, Khandheria BK, Hagen ME, et al. Strain rate and strain: a step-by-step approach to image and data acquisition. J Am Soc Echocardiogr. 2004;17:1011–1020.

20. Ryf S, Tsao J, Schwitter J, Stuessi A, Boesiger P. Peak-combination HARP: a method to correct for phase errors in HARP. J Magn Reson Imaging. 2004;20:874–880. 21. Atalar E, McVeigh ER. Optimization of tag thickness for measuring position with magnetic resonance imaging. IEEE. 1994;13:152–160. 22. Nagel E, Stuber M, Lakatos M, Scheidegger MB, Boesiger P, Hess OM. Cardiac rotation and relaxation after anterolateral myocardial infarction. Coron Artery Dis. 2000;11:261–267. 23. Ennis DB, Epstein FH, Kellman P, Fananapazir L, McVeigh ER, Arai AE. Assessment of regional systolic and diastolic dysfunction in familial hypertrophic cardiomyopathy using MR tagging. Magn Reson Med. 2003;50:638–642. 24. Stuber M, Scheidegger MB, Fischer SE, et al. Alterations in the local myocardial motion pattern in patients suffering from pressure overload due to aortic stenosis. Circulation. 1999;100:361–368. 25. Rademakers FE, Buchalter MB, Rogers WJ, et al. Dissociation between left ventricular untwisting and filling: accentuation by catecholamines. Circulation. 1992;85:1572–1581. 26. Grossman W, Jones D, McLaurin LP. Wall stress and patterns of hypertrophy in the human left ventricle. J Clin Invest. 1975;56:56–64. 27. Hess OM, Lavelle JF, Sasayama S, Kemper WS, Ross J. Diastolic myocardial wall stiffness of the left ventricle in chronic pressure overload. Eur Heart J. 1982;3:315–324. 28. Matter C, Mandinov L, Kaufmann P, Nagel E, Boesiger P, Hess OM. [Function of the residual myocardium after infarct and prognostic significance]. Z Kardiol. 1997;86:684–690. 29. Edvardsen T, Rosen BD, Pan L, et al. Regional diastolic dysfunction in individuals with left ventricular hypertrophy by tagged magnetic resonance imaging: the Multi-Ethnic Study of Atherosclerosis (MESA). Am Heart J. 2006;151:109–114. 30. Dorfman TA, Rosen BD, Perhonen MA, et al. Diastolic suction is impaired by bed rest: MRI tagging studies of diastolic untwisting. J Appl Physiol. 2008;104:1037–1044. 31. Axel L, Dougherty L. Heart wall motion: improved method of spatial modulation of magnetization for MR imaging. Radiology. 1989;172:349–350.

Cardiovascular Magnetic Resonance 75

5 CARDIOVASCULAR MAGNETIC RESONANCE TAGGING ASSESSMENT OF LEFT VENTRICULAR DIASTOLIC FUNCTION

6. Ingels Jr NB, Daughters GT, Stinson EB, Alderman EL. Measurement of midwall myocardial dynamics in intact man by radiography of surgically implanted markers. Circulation. 1975;52:859–867. 7. Yun KL, Niczyporuk MA, Daughters GT, et al. Alterations in left ventricular diastolic twist mechanics during acute human cardiac allograft rejection. Circulation. 1991;83:962–973. 8. Potel MJ, Rubin JM, MacKay SA, Aisen AM, Al-Sadir J, Sayre RE. Methods for evaluating cardiac wall motion in three dimensions using bifurcation points of the coronary arterial tree. Invest Radiol. 1983;18:47–57. 9. Zerhouni EA, Parish DM, Rogers WJ, Yang A, Shapiro EP. Human heart: tagging with MR imaging: a method for noninvasive assessment of myocardial motion. Radiology. 1988;169:59–63. 10. Axel L, Dougherty L. MR imaging of motion with spatial modulation of magnetization. Radiology. 1989;171:841–845. 11. Fischer SE, McKinnon GC, Maier SE, Boesiger P. Improved myocardial tagging contrast. Magn Reson Med. 1993;30:191–200. 12. Stuber M, Spiegel MA, Fischer SE, et al. Single breath-hold slicefollowing CSPAMM myocardial tagging. MAGMA. 1999;9:85–91. 13. McVeigh ER, Atalar E. Cardiac tagging with breath-hold cine MRI. Magn Reson Med. 1992;28:318–327. 14. Ryf S, Kissinger KV, Spiegel MA, et al. Spiral MR myocardial tagging. Magn Reson Med. 2004;51:237–242. 15. Meyer CH, Pauly JM, Macovski A, Nishimura DG. Simultaneous spatial and spectral selective excitation. Magn Reson Med. 1990;15:287–304. 16. Herzka DA, Guttman MA, McVeigh ER. Myocardial tagging with SSFP. Magn Reson Med. 2003;49:329–340. 17. Zwanenburg JJ, Kuijer JP, Marcus JT, Heethaar RM. Steady-state free precession with myocardial tagging: CSPAMM in a single breathhold. Magn Reson Med. 2003;49:722–730. 18. Ryf S, Spiegel MA, Gerber M, Boesiger P. Myocardial tagging with 3D-CSPAMM. J Magn Reson Imaging. 2002;16:320–325. 19. Osman NF, Kerwin WS, McVeigh ER, Prince JL. Cardiac motion tracking using CINE harmonic phase (HARP) magnetic resonance imaging. Magn Reson Med. 1999;42:1048–1060.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

CHAPTER 6

Cardiovascular Magnetic Resonance Contrast Agents Peter Caravan

Nearly one third of all magnetic resonance imaging (MRI) scans today (2009) use a gadolinium based contrast agent. The contrast agent typically makes diseased tissue appear brighter (or in some cases darker) than the surrounding tissue. Cardiovascular applications, such as magnetic resonance angiography (MRA) and functional imaging of myocardial perfusion and viability, are growing rapidly, and represent the bulk of cardiovascular magnetic reso nance (CMR) scans that use a contrast agent. The first approved contrast agent, gadopentetate (Gd DTPA), appeared in 1988, and several other compounds followed. These first contrast agents were extracellular fluid (ECF) agents. Although frequently used in CMR and an essential component of CMR perfusion and late gadolinium enhancement (LGE) assessment of fibrosis, none of these agents has been approved for cardiac applications. Thus, the use of these agents in CMR is considered “off label.” There are currently several compounds in clinical trials and one recently approved that are intravascular agents designed specifically to enhance contrast enhanced (CE) MRA. At the preclinical stage, there are exciting advance ments in molecular imaging agents, including compounds that detect pH changes, enzymatic activity, specific biomol ecules such as fibrin or integrins, and magnetically labeled cells. This chapter focuses first on the general underlying chemistry and biophysics of contrast agents in clinical CMR. The mechanism of action of different classes of contrast agents is described, with examples drawn from CMR applications. Finally, there is a brief survey of novel contrast agents potentially useful for cardiovascular indi cations that are currently in clinical or preclinical development. The vast majority of MRI agents in clinical use are small molecules based on chelated gadolinium. The bulk of this chapter focuses on gadolinium complexes, including their chemistry, biophysics, and applications. Other exogenous compounds have been used to change signal properties in MRI (e.g., iron particles, hyperpolarized nuclei), and these will be discussed as appropriate to CMR. This chap ter assumes that the reader has a basic understanding of CMR vocabulary, and the reader is referred to Chapter 1 for further clarification.

76 Cardiovascular Magnetic Resonance

INTRODUCTION TO THE BIOPHYSICS OF MAGNETIC RESONANCE IMAGING All contrast agents shorten both T1 and T2. However, it is useful to classify MRI contrast agents into two broad groups based on whether the substance increases the transverse relaxation rate (1/T2) by roughly the same amount that it increases the longitudinal relaxation rate (1/T1) or whether 1/T2 is altered to a much greater extent. The first category is referred to as “T1 agents” because, on a percentage basis, these agents alter 1/T1 of tissue more than 1/T2 because of endogenous transverse relaxation in tissue. With most pulse sequences, this dominant T1 lowering effect gives rise to increases in signal intensity, and thus these agents are referred to as “positive” contrast agents. The T2 agents largely increase the 1/T2 of tissue selectively and cause a reduction in signal intensity, and thus they are known as “negative” contrast agents. Paramagnetic gadolinium based contrast agents are examples of T1 agents, whereas ferromagnetic large iron oxide particles are examples of T2 agents. There are many mechanisms by which contrast agents shorten T1 and T2. Considerable chemistry and biophysics can be applied to understand or predict these mechanisms. However, in many cases, the effect of these mechanisms can be reduced to a single constant, called “relaxivity.” Figure 6 1 shows the effect and definition of relaxivity. Figure 6 1A shows the effect of a typical contrast agent on the relaxation time of two hypothetical tissues, one with T1 ¼ 1200 msec (similar to heart muscle at 1.5 T) and one with T1 ¼ 400 msec. At low concentration (left side of the graph), it appears that the contrast agent has a larger effect (change in T1) on the tissue with the longer T1. At higher concentrations of contrast agent (right side of Fig. 6 1A), both tissues approach approximately the same T1. A simple way to quantify this effect is to consider the rate of relaxa tion, 1/T1 (sometimes denoted “R1”). In most cases in medical imaging, the contrast agent increases the relaxation rate proportional to the amount of contrast agent: 1 1 ¼ þ r1 ½CA T1 T1o

(1)

20 1200 15 1/T1 (s−1)

T1 (msec)

1000 800 600 400

10

5

200 0

0 0

A

2

3

Contrast agent (mM)

where T1 is the observed T1 with contrast agent in the tis sue, T1o is the T1 before addition of the contrast agent, [CA] is the concentration of contrast agent, and r1 is the longitudinal relaxivity, often just “relaxivity.” The conven tional units for r1 are mM-1s-1 (per millimolar per second, sometimes Lmol-1s-1). Thus, the slope of 1/T1 as a func tion of contrast agent concentration (Fig. 6 1B) shows the relaxivity, in this case, 4 mM-1s-1. Figure 6 1B shows that the effect of the contrast agent on the relaxation rate is independent of the initial T1 of the tissue. That is, in terms of relaxation rate, the contrast agent has the same effect, regardless of initial T1. Transverse, or T2, relaxivity is defined in an analogous way: 1 1 ¼ þ r2 ½CA T2 T2o

1

(2)

For all medically used contrast agents, r2 is larger than r1. Relaxivity is dependent on magnetic field strength, is depen dent on temperature, and in some instances can depend on protein binding, pH, or even the presence of enzymes. Contrast agent behavior in vivo is seldom as simple as the pure linear effect relaxation rate shown in Figure 6 1. Even in the simple case of pure linear relaxation, the effect of the contrast agent on the CMR image is generally nonlin ear. In traditional spin echo sequences, nonlinearity can be a result of T1 saturation or T2 signal loss. Once the contrast agent reduces T1 < repetition time (TR)/2, increasing con trast agent concentration will have little effect on increasing the available longitudinal magnetization because the tissue will have nearly fully recovered the magnetization before the next radiofrequency (RF) pulse. Furthermore, because contrast agents affect both T1 and T2 relaxation, at high enough concentration, the contrast agent will reduce T2 to the order of the echo time (TE), and will then decrease magnetic resonance (MR) image intensity. These effects are seen in Figure 6 2, where signal intensity is plotted ver sus contrast agent concentration for T1 and T2 weighted spin echo sequences. Figure 6 2 was generated assuming a contrast agent relaxivity of 4 mM-1s-1, typical of most commercial ECF gadolinium agents, and tissue relaxation times typical of myocardium at 1.5 T (T1 ¼ 1200 msec, T2 ¼ 50 msec). For the T1 weighted sequence (TE/TR ¼ 15/600), Figure 6 2A, signal intensity begins to level out at a contrast agent concentration between 0.5 and 1.0

0

4

B

1

2

3

4

Contrast agent (mM)

mM. From Figure 6 1A, this is the range as which the T1 of the myocardium decreased to approximately 300 msec, or TR/2. At concentrations greater than 1 mM, the T1 effect is saturated, and the only imaging effect of the contrast agent is to make T2 shorter and cause signal loss, even on this T1 weighted sequence. Signal is lost because even a T1 weighted sequence has a finite TE, and T2 effects can enter when T2 is short enough. The signal intensity plateau on the T2 weighted scan (TE/ TR ¼ 80/3000), Figure 6 2B, occurs at much lower contrast agent concentration. Because TR is so long, the only real effect of the contrast agent is to reduce (rather than increase) signal intensity on this T2 weighted scan. However, this neg ative contrast can also be medically useful, and certain con trast agents (notably, the iron oxide particles) create negative contrast exactly by providing enhanced T2 relaxation, and thus darker images on T2 weighted scans. Increasing the relaxivity (r1 or r2) will have the effect of pushing the simulated curves in Figure 6 2A to the left, which means that peak signal and subsequent signal loss will occur at lower contrast agent concentrations. A more linear response of signal to contrast agent can be achieved with a fast three dimensional spoiled gradient recalled echo (GRE) sequence. This is seen in Figure 6 2C, where signal intensity is plotted versus contrast agent concentration using the same tissue relaxation times and relaxivities as in Figure 6 2A and 6 2B for a typical fast three dimensional spoiled GRE sequence, TE/TR/flip ¼ 2.2/9.0/40 . The short TR and very short TE ensure that signal intensity increases across the entire concentration range. At high concentration, the effect of the contrast agent is becoming nonlinear, but signal still is increasing with increasing contrast agent concentration.

Commercial Contrast Agents and Those in Clinical Development The addition of paramagnetic materials to reduce relaxation times goes back to the earliest days of MR. In the 1940s, Bloch and colleagues used ferric nitrate to enhance the relaxation rate of water.1 Exogenous contrast was applied to MRI in 1977 when Lauterbur and associates reported Cardiovascular Magnetic Resonance 77

6 CARDIOVASCULAR MAGNETIC RESONANCE CONTRAST AGENTS

Figure 6 1 Change in (A) longitudinal relaxation time (T1) and (B) longitudinal relaxation rate (1/T1) for typical myocardial tissue (solid line, T1 1200 msec at 1.5 T) and short T1 tissue (dashed line, T1 400 msec).

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

using manganese dichloride to differentiate normal from infarcted myocardium in dogs.2 Carr and colleagues reported the first use of a gadolinium complex, gadopente tate dimeglumine (Gd DTPA; Magnevist, Bayer Schering, Berlin, Germany), in patients with brain tumors in 1984.3 By 1988, Gd DTPA was approved for clinical use.

0.7 0.6 0.5 0.4

Extracellular Agents

0.3 0.2 Spin echo TR/TE/flip = 600/15/90o

0.1 0 0

1

A

2

3

4

Contrast agent (mM)

0.7 Spin echo TR/TE/flip = 3000/80/90o

0.6 0.5 0.4 0.3 0.2 0.1 0 0

1

B

2

3

4

Contrast agent (mM)

0.7 3D GRE TR/TE/flip = 9.0/2.2/40o

0.6 0.5 0.4 0.3 0.2 0.1 0 0

C

1

2

3

4

Contrast agent (mM)

Figure 6 2 Effect of contrast agent on myocardial image intensity on T1 weighted and T2 weighted spin echo scans. (A) T1 weighted spin echo (repetition time [TR] 600 msec) assumes a patient with 100 bpm heart rate and shows a linear increase of signal only for contrast agent concentration < 0.5 mM. (B) T2 weighted spin echo images (TR 3000 msec) show only T2 signal loss effects as a result of contrast agent with no T1 enhancement because of the long TR. (C) Typical short TR fast spoiled gradient recalled echo (GRE) sequence. The very short echo time (TE) and short TR give monotonically increased image intensity across the entire range of contrast agent concentrations typically found in clinical scans. 78 Cardiovascular Magnetic Resonance

The most common contrast agents used clinically are ECF agents. Although several are approved for clinical use, none is specifically approved for cardiac applications. These all behave in a very similar manner, and are typically referred to as “gadolinium” or “gado” agents. Figure 6 3 shows the chemical structures of several approved ECF agents. Chem ically, these compounds exhibit three similar features: they all contain Gd, they all contain an 8 coordinate ligand binding to Gd, and they all contain a single water molecule coordination site to Gd. Nomenclature for contrast agents can be confusing: there is a generic name (e.g., gadopente tate dimeglumine), a trade name (e.g., Magnevist, Bayer HealthCare Pharmaceuticals), and usually a chemical code name or abbreviation (e.g., Gd DTPA). Any of these three names may be used in the scientific literature. The multidentate ligand is required for safety.4 The ligand encapsulates the gadolinium, resulting in a high thermodynamic stability and kinetic inertness with respect to metal loss. This enables the contrast agent to be excreted intact, an important property because these contrast agents tend to be much less toxic than their substituents. For example, the DTPA ligand and gadolinium chloride both have an LD50 of 0.5 mmol/kg in rats (LD50 ¼ dose that causes death in 50% of the animals), whereas the Gd DTPA complex has a safety margin that is higher by nearly a factor of 20, with an LD50 of 8 mmol/kg for the Gd DTPA complex.5 The gadolinium ion and coordinated water molecule are essential to providing contrast. The gadolinium(III) ion has a high magnetic moment and a relatively slow electronic relaxation rate, properties that make it an excellent relaxer of water protons. The proximity of the coordinated water molecule leads to efficient relaxation. The coordinated water molecule is in rapid chemical exchange (106 exchanges/sec) with solvating water molecules.6 This rapid exchange leads to a catalytic effect whereby the Gd complex effectively shortens the relaxation times of the bulk solution. The ECF agents have very similar properties, and these are summarized in Table 6 1. They are all very hydrophilic complexes with similar relaxivities and excellent safety pro files, and they can be formulated at high concentrations. On injection, ECF agents quickly and freely distribute to the extracellular space. Administration of any of these agents (with rare exceptions)7 yields similar diagnostic information. There are some differences among the physical properties. The diamide complexes gadodiamide and gadoversetamide have considerably lower thermodynamic stability (log K17 vs. log K > 21 for other Gd complexes).8,9 The nonionic (neu tral) compounds (gadodiamide, gadoteridol, gadoversetamide, and gadobutrol) were designed to minimize the osmolality of the formulation. This was prompted by the distinct reduction in toxicity and side effects brought on by the development of

N

N

N O O Gd O O O O O O H H

O O O

N

N

NH O O

Gd O O

H

N NH O O O O O H

Gd-DTPA-BMA (Omniscan) Gadodiamide

Gd-DTPA (Magnevist) Gadopentetate O OH

O

OH N OH

N Gd O

OH2

N

N

O

O

O O

O

Gd O

O

H

Gd-DO3A-butrol (Gadovist) Gadobutrol

O

N NH O O O O O

O

H

Gd-DTPA-BMEA (OptiMARK) Gadoversetamide

O

O O

O N

N

O

N OH

N Gd

OH2

N

N

O

O

Gd O

N

N

NH

O O

O

N

OH2

N O

O

O

Gd-DOTA (Dotarem) Gadoterate

Gd-HPDO3A (ProHance) Gadoteridol

Table 6-1 Approved (January 2009) Magnetic Resonance Imaging Contrast Agents—Relaxivity,120 Osmolality, and Viscosity Generic Name Gadopentetate Gadoterate Gadodiamide Gadoteridol Gadobutrol Gadoversetamide

Product Name

Chemical Abbreviation

r1, 0.47 T 40 C

r2, 0.47 T 40 C

Osmolality* (osmol/kg)

Viscosity* (cP)

Magnevist (Bayer HealthCare Pharmaceuticals) Dotarem (Guerbet Group) Omniscan (GE Healthcare) ProHance (Bracco Diagnostics, Inc.) Gadovist (Bayer HealthCare, Inc.) OptiMARK (Mallinckrodt, Inc.)

Gd DTPA

3.4

4.0

1.96121

2.9121

Gd DOTA

3.4

4.1

Gd DTPA BMA

3.5

3.8

Gd HPDO3A

3.1

3.7

Gd DO3A butrol

3.7

5.1

Gd DTPA BMEA

4.2

5.2

1.35122 (4.02)122 0.79121 (1.90)122 0.63122 (1.91)122 0.57123 (1.39)121 1.11124

2.0122 (11.3)122 1.4121 (3.9)122 1.3122 (3.9)122 1.4123 (3.7)121 2.0124

*All concentrations 0.5 M except those in parentheses, which are 1 M.

nonionic X ray contrast media. However, for CMR, the injec tion volumes are much smaller than for X ray, and thus the overall increase in osmolality after injection of a CMR contrast agent is minimal. Unlike with X ray contrast, there is no docu mented safety benefit in using nonionic CMR contrast agents. One benefit of the nonionic compounds is the ability to formu late them at high concentration (1 M)10 without drastically

increasing osmolality or viscosity (see Table 6 1). These high concentration formulations may be useful in fast dynamic studies, such as dynamic MRA or myocardial perfusion. These ECF agents, as with iodinated preparations used for X ray and computed tomography, are excreted by the kidneys. As a result, clearance is impaired in patients with abnormal or depressed renal function (discussed later). Cardiovascular Magnetic Resonance 79

6 CARDIOVASCULAR MAGNETIC RESONANCE CONTRAST AGENTS

Figure 6 3 Chemical structures of commercial (United States or Europe) extracellular fluid contrast agents.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Blood Pool Agents Because CMR contrast agents are administered intravenously, they are all potentially capable of imaging the blood vessels. The ECF agents described earlier are used routinely, if off label, for CE MRA (see Chapters 23, 24, and 34 to 36). One draw back of using ECF agents for MRA is their pharmacokinetics. ECF agents rapidly leak out of the vascular space into all the interstitial spaces of the body. Angiography with ECF agents is thus typically limited to dynamic arterial studies. There has been considerable effort toward designing specific blood pool agents that would be tailored for vascular imaging. The ideal blood pool agent should remain in the vascular compartment and not leak out into the extracellular space. It should be capa ble of being given as a bolus such that a dynamic arterial image can be obtained. At the same time, it should have sufficient relaxivity and blood half life to allow high resolution steady state images to be obtained. Recently, MS 325 (gadofosveset, marketed as VasovistW and more recently as AblavarW in the United States by Lantheus Medical Imaging, Billerica, MA; Fig. 6 4) was approved, and there are several other blood pool agents that have reached various stages of clinical develop ment. Three approaches have been taken to design blood pool agents: protein binding, increased size, and ultrasmall iron oxide particles. These are discussed later. Both MS 325 (gadofosveset)11,12 and B22956 (gadocoletic acid; Fig. 6 5)13,14 are gadolinium based compounds that bind reversibly to serum albumin. Albumin is the most abundant protein in plasma, and its concentration is high enough (600 to 700 mM) to bind reversibly most of the contrast agent after injection. Reversible albumin binding serves four purposes: (1) the albumin slows the leakage of the contrast agent out of the intravascular space; (2) the reversible binding still allows a path for excretion the unbound fraction can be filtered through the kidneys or taken up by hepatocytes; (3) the bound fraction is “hidden” from the liver and kidneys, leading to an extended plasma half life; and (4) the relaxivity of the contrast

O N

N O Gd O O O O O O O H H

O O O

Gd-BOPTA (MultiHance) Gadobenate

O O O P O N

N

O O O

N O O Gd O O O O O O H H

MS-325 (Vasovist; Ablavar) Gadofosveset Figure 6 4 Chemical structures of other commercial contrast agents with weak (Gd BOPTA) or strong (MS 325) serum albumin binding.

agent is increased 4 to 10 fold on binding to albumin (dis cussed later). Gd BOPTA (gadobenate; see Fig. 6 4) has weak affinity15,16 for albumin (10% bound), which leads to a mod est increase in relaxivity relative to the ECF agents.

GdL

LGd O

HO2C

LGd N

N

GdL =

HO

O

N

O

LGd

O

Lys

LGd Lys

Lys Lys

O O

B22956 (Gadocoletic acid)

Lys

O

HN

LGd

Lys = N H

Lys

N

Lys

LGd

Lys

GdL

Lys

Lys

GdL

Lys Lys GdL

Lys

O Lys Lys

Lys Lys

GdL

GdL

N

O

Lys

Lys

GdL

GdL

Lys

GdL Lys Lys

Lys GdL

Figure 6 5 Chemical structures of some investigational gadolinium based blood pool agents.

Lys

Lys

GdL

Gadomer (aka Gadomer-17)

80 Cardiovascular Magnetic Resonance

Lys Lys

Lys

N

N O O Gd O O O O O O H H

GdL Lys Lys

Lys

LGd N

O Lys

N

Lys

LGd O

O

Lys

Lys

Lys OH2

N

GdL

Lys

Lys

LGd

O

Gd O

NH

Lys Lys

Lys

O H

N

Lys GdL

GdL

Agent type r1 buffer (mM 1 s 1) r1 plasma (mM 1 s 1) % bound plasma

MS 325* Gadofosveset

B22956† Gadocoletic Acid

Gd BOPTA Gadobenate

Gadomer 17‡

Strong protein binding 6.612 5012 9112

Strong protein binding 6.5125 27125 95125

Weak protein binding 4.4126 9.7127

Increased size 16.5120 19.0120

*Data at 0.1 mM. {Data at 0.5 mM. {Relaxivity per gadolinium.

The binding and relaxivity features of gadolinium based blood pool agents are listed in Table 6 2. Because binding affinity is moderate to weak for these compounds, the frac tion bound to albumin will depend on the concentrations of albumin and the contrast agent.12,17 Immediately after injection, when the concentration of the contrast agent is high relative to albumin, there will be a greater free frac tion. As the concentration of the contrast agent begins to stabilize (at 0.5 mM) the fraction bound will become con stant. The observed relaxivity will depend on the fraction bound; unlike ECF agents, T1 change in plasma is not lin early related to contrast agent concentration.17,18 Among the albumin binding agents, MS 325 (gadofosveset) has a somewhat lower albumin affinity than B22956 (gadocoletic acid), although the majority of both is bound under steady state conditions. The relaxivity of albumin bound MS 325 (gadofosveset) is higher than that of B22956 (gadocoletic acid). MS 325 (gadofosveset) is mainly renally excreted, whereas B22956 (gadocoletic acid) has significant biliary clearance as well as renal excretion. Early work on blood pool compounds involved gado linium covalently linked to macromolecules, such as poly lysine, dextran, or modified bovine serum albumin.4 The large size of these compounds restricted diffusion out of the vascular space and led to very good vascular imaging properties. However, these compounds cleared very slowly in preclinical studies and there were concerns about a potential immunologic response. This approach was modified by the synthesis of compounds that were large enough to be kept in the vascular compartment, but small enough to be eliminated by glomerular filtration in the kidneys. Gadomer (gadomer 17) is an example of this type of blood pool agent (see Fig. 6 5).19,20 Gadomer is a large, 17 kDa molecule that contains 24 covalently linked gadolinium complexes. The dendritic (branching) approach to synthesis results in a compound that is glob ular. The per gadolinium relaxivity of Gadomer is much higher than that of its monomeric units because of the slow tumbling of the molecule (discussed later). The use of multiple gadolinium chelates to increase the size also increases the molecular relaxivity (24 Gd  18.7 ¼ molecular relaxivity of 450 mM-1s-1), which in turn means that lower doses can be given. Gadomer is a neutral hydrophilic compound that has little affinity for plasma proteins and is excreted renally. All of the gadolinium based vascular agents described earlier are not “true” blood pool compounds. Although far superior to the ECF agents in terms of extravasation and relaxivity, some fraction of the compound leaks out

into the extracellular space. Iron oxide particles, on the other hand, are true blood pool agents. The small particle iron oxide particles (SPIOs) used for liver imaging are large enough to be recognized by the reticuloendothelial system and rapidly removed from the bloodstream. The smallest size fraction of these particles, ultrasmall particle iron oxide (USPIOs), evaded the reticu loendothelial system and could be used to image the blood pool.21,22 Although smaller, these particles are still too large to passively leak out of the vascular space and make very good blood pool agents. Making ultrasmall particles not only changes the biodistribution of the compound, but also changes the relaxation phenomena. SPIOs have a much greater effect on T2 than on T1 (large r2/r1) and are used as T2 or T2* agents. USPIOs have very good T1 relax ation properties (smaller r2/r1) and can be used for bright blood imaging (T1 weighted). The iron oxide particles are not excreted; the iron is eventually resorbed into the body.

RELAXIVITY Because their effect is indirect, CMR contrast agents differ from other diagnostic imaging agents. Water and fat are imaged, and it is the action of the contrast agent on the relaxation properties of the water hydrogen nuclei that generates contrast; in X ray con trast media and nuclear imaging agents, the effect observed is more direct. Because water is present at a very high concentration (55,000 mM) and the contrast agent is typically at a much lower concentration (0.1 to 1 mM), the contrast agent must act catalyti cally to relax the water protons for there to be a measurable effect. Relaxivity, r1 and r2, thus describes this catalytic efficiency. Some compounds are better magnetic catalysts than others (have higher relaxivity). Moreover, relaxivity is dependent on the external mag netic field, B0. This section explains these differences and the molecular basis for them. For discrete ions, such as Gd(III) and manganese (Mn[II]), the factors influencing relaxivity are the same; for iron oxide particles, the relaxation mechanism is differ ent and will be treated separately. Relaxivity can be factored into inner and outer sphere terms. “Inner sphere” r1 ¼

Dð1=T1 Þ OS ¼ rIS 1 þ r1 ½Gd

(3)

refers to the relaxation enhancement due to the exchangeable waters in the inner sphere. “Outer sphere” refers to relaxivity resulting from water in the second and outer coordination spheres. This separation is often used because the inner sphere component is easier to understand from a theoretical framework and the effect of changing specific molecular parameters on relaxivity can be tested. For ECF agents, the inner sphere and Cardiovascular Magnetic Resonance 81

6 CARDIOVASCULAR MAGNETIC RESONANCE CONTRAST AGENTS

Table 6-2 Albumin Binding and Observed Relaxivities (20 MHz) of MS-325, B22956, Gd-BOPTA, Gadomer-17, at 37 C

q=½H2 O (4) rIS 1 ¼ T1m þ tm Inner sphere relaxivity is given by Equation 4, where q repre sents the number of water molecules bound to the metal ion (typ ically q ¼ 1), [H2O] is the water concentration in millimolar, T1m is the relaxation time of the water that is bound to the metal ion, and tm is the lifetime of the water molecule in the inner sphere. Equation 4 shows that the more sites available for water to bind (q), the more efficient the process. Likewise, the T1 of the bound water (T1m) should be very short. This is indicative of how effec tively the metal ion can relax the coordinated water. The lifetime tm of the bound water is the inverse of the water exchange rate (kex ¼ 1/tm). If turnover is slow (tm is long), then it will not mat ter how efficiently the bound water is relaxed if this relaxation cannot be transmitted to the bulk water. Paramagnetic relaxation (1/T1m) occurs via a dipolar mecha nism. At a magnetic field encountered in medical imaging, 1/T1m is shown in Equation 5, where C is a   1 SðS þ 1Þ 3tc (5) ¼C 1 þ o2H t2c T1m r6MH constant,4 S is the spin quantum number, rMH is the metal ion to proton distance, tc is the correlation time, and oH is the hydrogen Larmor frequency. The product S(Sþ1) is proportional to the mag netic moment. All other factors being equal, the higher the mag netic moment, the more efficient the relaxation. This is why Gd3þ (S ¼ 7/2) is preferred to copper (Cu2þ, S ¼ 1/2). The dipo lar effect depends on the distance between the ion and the hydro gen nucleus, rMH, to the inverse sixth power. The inner sphere water is critical; it has the shortest metal to hydrogen distance of water hydrating the metal complex. The correlation time, tc, is the time constant that governs the interaction between the electron spin of the ion and the nuclear spin of the hydrogen. Depending on the ion and the field strength, the correlation time is either the time constant for rota tional diffusion of the molecule, tR, or the electronic T1 of the metal ion, T1e. In the MR experiment, nuclei are excited by apply ing RF energy at the Larmor frequency. To relax these nuclei, there needs to be a resonant source for energy transfer. A paramagnetic molecule tumbling in solution creates a fluctuating magnetic field. Molecules tumble (undergo rotational diffusion) because of ther mal energy, but because they also collide with each other, there is a distribution of rotational diffusion rates, with a mean rate characterized by 1/tR. The closer this tumbling rate is to the Lar mor frequency, the more efficient the relaxation by the contrast agent. Small molecules, such as Gd DTPA, tumble very fast, in the gigahertz range (1 GHz ¼ 1000 MHz), but the Larmor frequency for protons at imaging fields is much slower. For example, at 1.5 T, the Larmor frequency is approximately 65 MHz, so relaxation is not as efficient as it could be. Larger molecules, such as proteins, tumble much more slowly. When contrast agents are made to tumble more slowly, relaxivity is increased. Newer contrast agents take advantage of this phenomenon. Lauffer23 pointed out that if small contrast agents could be made to bind noncovalently to pro tein targets, then their relaxivity would be increased on binding because the contrast agent would take on the rotational character istics of the protein (receptor induced magnetization enhance ment). MS 325 is an example of a contrast agent designed to exploit the receptor induced magnetization enhancement effect.12 MS 325 targets the blood protein serum albumin. In the absence of albumin, the relaxivity of MS 325 is approximately 50% greater than that of Gd DTPA, but in the presence of albumin, the relax ivity is approximately 600% greater than that of Gd DTPA at 1.5 82 Cardiovascular Magnetic Resonance

40

r1 (mM−1 s−1)

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

outer sphere contribution to relaxivity is about the same, whereas for newer contrast agents, the inner sphere component accounts for the majority of overall relaxivity.

30

20

10

0 0.5

1.0 1.5 2.0 B0 (Tesla)

2.5

3.0

Figure 6 6 Magnetic field dependence on relaxivity for MS 325 (gadofosveset; circles) and gadopentetate (Gd DTPA; squares) in either serum albumin solution (filled symbols) or buffered saline (open symbols) at 37 C.

T. This is seen in Figure 6 6, where the magnetic field dependence on relaxivity is plotted for MS 325 and Gd DTPA in either serum albumin solution (filled symbols) or in buffered saline. Increasing the molecular size is another way to increase relax ivity. Gadomer is a dendrimer that contains 24 Gd and has a molecular weight of 17,500. The increased size results in slower motion and higher relaxivity. However, the rotational effect is defined by more than just molecular weight. Linear polymers con taining Gd have lower relaxivities24,25 than may first be expected because of fast rotation around one axis of the molecule. Electrons undergoing relaxation in a paramagnetic ion also generate a fluctuating field that can cause nuclear relaxation. It is the faster of the two processes (rotation or electronic relaxation) that defines how efficiently the hydrogen is relaxed. For Gd and Mn complexes, electronic relaxation depends on the applied field and it gets slower as the magnetic field is increased. At low fields (< 0.5 T), electronic relaxation governs hydrogen relaxation, whereas at higher fields, rotational diffusion is the dominant mechanism. Gd3þ and Mn2þ have symmetrical electronic config urations (half filled f and d shells, respectively), and as a result, electronic relaxation is relatively slow. Other ions, such as dyspro sium (Dy3þ), have a higher magnetic moment, but are rather poor relaxors because electronic relaxation is so fast. Equation 5 also indicates that at higher fields the T1 relaxation mechanism becomes less efficient as oH2tc2 > 1. The field at which this condition is met is of course dependent on the magni tude of the correlation time. This is seen in Figure 6 6, where the relaxivity of MS 325 in serum albumin solution begins to decrease after a peak at approximately 0.7 T. Figure 6 7 further illustrates the effect of correlation time and field strength on relaxivity. In Figure 6 7, r1 and r2 are simulated over a range of fields encoun tered in CMR for correlation times of 0.1 nsec (typical of ECF agents), 1 nsec (intermediate motion), and 10 nsec (typical of albumin bound agents). Figure 6 7 shows that the benefits of very slow rotation are seen at lower field strengths. Note also that r1 does not go to zero because there is also an outer sphere compo nent to relaxivity11 and the correlation times that govern outer sphere relaxivity are quite short. Transverse relaxivity is defined by equations similar to Equations 4 and 5, with the exception that there are other mechanisms that cause r2 to increase at high fields.4,26 The r2 is always greater than r1, and for very slow tumbling systems, the r2/r1 ratio becomes large at high fields. Electronic and nuclear relaxation is described in greater detail in various reviews and books.4,26

100 90 80 70 60 50 40 30 20 10 0

100 90 80 70 60 50 40 30 20 10 0 0

0.5

A

1

1.5

2

2.5

0

0.5

1

B

Bo (Tesla)

Figure 6 7 suggests that slow tumbling T1 agents become less effective at high fields, but relaxation times for tissue are longer at high fields and the signal to noise ratio increases with increased field strength (B0). These factors and the choice of sequence mean that a contrast agent with a lower relaxivity at 3 T than at 1.5 T may still provide greater contrast at 3 T than at 1.5 T. The iron oxide based contrast agents are not discrete mole cules, but crystals of iron oxide (Fe3O4) surrounded by a coating. The individual spins of each iron cooperatively via quantum mechanical interactions build up to give the crystal a very large total spin, and thus relaxivity will be a function of the number of spins. Such a material is called “ferromagnetic,” and its magne tism persists outside the external magnetic field. A weaker form of this is superparamagnetism, in which small particles of iron ox ides with aligned spins occur in a magnetic field. Because the par ticles are small (submicron), the magnetic susceptibility effect is smaller than for large crystals of ferrites. Superparamagnets are no longer magnetic outside of the external field. The iron oxide particles consist of a core of one or more magnetic crystals of Fe3O4 embedded in a coating. Because these are materials, there is a distribution of sizes. USPIOs have a single crystal core and a submicron diameter (e.g., ferumoxtran [AMI 227, SineremW, Guerbet, Roissy, France aka Combidex, Advanced Magnetics, Cambridge, MA] has a crystal diameter of 4.3 to 4.9 nm and a global particle diameter of  50 nm).27 SPIOs have cores containing more than one crystal of Fe3O4 and are larger than USPIOs, but still submi cron in size (e.g., ferumoxide [EndoremW, Guerbet, Roissy, France or FeridexW, Bayer Healthcare, Wayne, NJ] has a crystal diameter of 4.3 to 4.8 nm and a global particle diameter of  200 nm).27 USPIOs and SPIOs are small enough to form a stable suspension and can be admi nistered intravenously. The difference in particle size determines their pharmacokinetic behavior. There are no inner sphere water mole cules in iron particles, and relaxation of water arises from the water molecules diffusing near the particle. However, the mechanism of

3

1.5

2

2.5

3

Bo (Tesla)

outer sphere relaxation differs from that described earlier. One fea ture is that the crystals have a net magnetization, and as the external field is increased, this magnetization is increased (as is true for Gd, but the effect is much smaller). The modulation of this net magneti zation can cause proton relaxation and have field dependence.28 There are some generalities about relaxivity in these particles. For the USPIOs, longitudinal relaxivity (r1) can be quite high and these can function as effective T1 agents. The r2/r1 ratio for USPIOs is significantly larger than for Gd complexes, and r2 increases with increasing magnetic field. When there is aggrega tion of crystals, which is the case in SPIOs, r1 tends to decrease and r2 increases. Thus, for the particles themselves as well as for aggregates of particles, the ratio of r2/r1 typically increases as the size of the particles or aggregates increases, although the T2 relax ivity as a function of particle size can be quite complicated. The effect of aggregation of crystals is that the aggregate itself can be considered a large magnetized sphere whose magnetic moment increases with increasing field strength. This gives rise to suscep tibility effects and the SPIOs can act as T2* relaxation agents. This has important consequences when considering the effects of con trast agent compartmentalization on imaging (discussed later). Table 6 3 gives r1 and r2 values in plasma at 1.5 T and 3.0 T for a range of contrast media. The ECF agents have slightly lower r1 at 3.0 T and low r2/r1 ratios. The weak albumin binders have higher relaxivities, and the slow tumbling Gd compounds, such as Gado mer or albumin bound MS 325, have several fold higher relaxivities than the ECF agents. As Figure 6 7 suggests, the slow tumbling com pounds also have significantly lower r1 at 3.0 T than at 1.5 T and the r2/r1 ratio is increased at 3.0 T. Three iron particle formulations are listed. The SPIOs FeridexW and ResovistW (Bayer Healthcare, Berlin, Germany) have a large r2 and a very large r2/r1 ratio, and this ratio is increased at 3.0 T. The USPIO SHU555C (Bayer Healthcare, Berlin, Germany) has good T1 relaxivity at 1.5 T, but at 3.0 T, the transverse relaxivity dominates.

Table 6-3 Relaxivities120 of Selected Contrast Media (0.25 mM) in Plasma at 1.5 T and 3.0 T at 37 C COMPOUND

1.5 T -1 -1

3T -1

-1

-1

-1

Chemical/Code

Generic name

r1 (mM s )

r2 (mM s )

r1 (mM s )

r2 (mM-1 s-1)

Gd DTPA Gd DTPA BMA Gd HPDO3A Gd BOPTA MS 325* Gadomer

Gadopentetate Gadodiamide Gadoteridol Gadobenate Gadofosveset Gadomer 17

4.1 4.3 4.1 6.3 27.7 16.0

4.6 5.2 5.0 8.7 72.6 19.0

3.7 4.0 3.7 5.2 9.9 13.0

5.2 5.6 5.7 11.0 73 25

*Data from Eldredge HB, Spiller M, Chasse JM, et al. Species dependence on plasma protein binding and relaxivity of the gadolinium-based MRI contrast agent MS-325. Invest Radiol. 2006;41:229–243.

Cardiovascular Magnetic Resonance 83

6 CARDIOVASCULAR MAGNETIC RESONANCE CONTRAST AGENTS

Figure 6 7 Effect of rotational correlation time on longitudinal (r1, 7a) and transverse (r2, 7b) relaxivities as a function of field strength. Long correlation time (tR 10 nsec, solid line), typical of albumin binding, gives high r1 that decreases with increasing field and high r2; intermediate correlation time (tR 1 nsec, short dashed line) shows relaxivity maximum pushed out to a higher field; short correlation time (tR 0.1 nsec, long dashed line), typical of extracellular fluid agents, shows low, roughly field independent r1, r2. Simulations with other parameters typical of gadolinium based agents.11

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

CONTRAST-ENHANCED TISSUE RELAXATION The extent to which a metal complex influences tissue relaxation rates depends on three factors: 1. The chemical environment encountered by the metal com plex. Binding of the agent to macromolecules can cause significant relaxivity enhancement. An example of this is shown in Figure 6 6, comparing the relaxivity of MS 325 in buffer solution and in serum albumin solution. 2. Compartmentalization of the metal complex in tissue. Generally, tissue water is compartmentalized into intra vascular, interstitial (fluid space between cells and capil laries), and intracellular space, constituting roughly 5%, 15%, and 80% of total water, respectively. Cellular orga nelles further subdivide the intracellular component. If water exchange between any of these compartments is slow relative to the relaxation rate in the compartment with the longest T1, multiexponential relaxation may result. This can decrease the effective tissue relaxivity of an agent because not all of the tissue water is encoun tering the paramagnetic center. 3. The magnetic susceptibility of the contrast agent. The contrast agent causes a microscopic field inhomogeneity on a biologic scale of 10 to 1000 nm rather than on the chemical scale of 0.1 to 1 nm. This results in a reduc tion in apparent T2. The result of the first two effects is that the simple relax ivity equation (Equation 1) often is not valid in a biologic setting; likewise, describing the effects on a CMR pulse sequence with a single tissue relaxivity can be misleading. Care must be taken when trying to estimate the concentra tion of a contrast agent from signal intensity changes. The actual relaxivity within a compartment might be affected by the biologic milieu. For some compounds with strong or weak albumin binding (e.g., MS 325 [gadofosve set], B22956 [gadocoletic acid], Gd BOPTA [gadobenate]), local variation in the albumin concentration will affect the amount of contrast agent bound to albumin and thus affect the overall relaxivity. For example, in the plasma space, albumin concentration is typically high (0.6 to 0.7 mM) compared with the extracellular space in the normal heart (0.2 to 0.3 mM). The ECF agents interact very weakly if at all with plasma proteins or membrane structures. Tweedle and colleagues and Wedeking and associates showed that the relaxivity in blood and soft tissue of Gd DTPA (gadopentetate) and Gd DOTA (gadoterate) is the same as the relaxivity in aque ous solution.29,30 In extreme settings, the actual relaxivity could vary.31 Except for pathologic situations, most CMR contrast agents in use today are excluded from intact cells. Thus, the contrast agent will be localized to the extracellular space. As a result, the simple relaxivity equations do not necessarily hold. For a Gd based ECF agent in a test tube, it takes approximately 3 msec for water to diffuse between Gd molecules32; in the time of a typical imaging TR, a given water molecule may interact with thousands or millions of Gd molecules, and all water molecules will interact with approximately the same number of Gd ions. However, if that same ECF agent is compartmentalized solely within the cardiac microvasculature, it takes between 2 and 20 seconds for most of the water in the tissue (85% of it is 84 Cardiovascular Magnetic Resonance

extravascular) to physically diffuse into the microvascula ture; most of the water in the tissue does not have the opportunity to be relaxed by the Gd within TR of an imag ing acquisition, resulting in a lower signal enhancement than that predicted by Equation 1 and assuming a uniform distribution of contrast agent throughout the tissue. To deal with compartmentalization, the concepts of “water exchange” and “exchange time,” t, between com partments are often used.33,34 The water exchange rate and the size of the compartments will determine the effect of the contrast agent on CMR signal. To illustrate this phe nomenon, the two limiting causes of exchange will be described.32,33 In one extreme, water moves so quickly between the biologic compartments that the net effect is as if the contrast agent were uniformly spread throughout the entire tissue. This situation, called “fast exchange,” occurs whenever the exchange rate, 1/t, between the com partments is much faster than the difference in relaxation rates between the compartments.35 This occurs in blood, where red cells have a short water exchange time, on the order of 5 to 10 msec.36 The intact red cell prevents most CMR contrast agents from entering the cell, but as long as the plasma T1 is longer than 20 msec, the two compart ments of the blood (plasma þ red cells) are in fast exchange and blood behaves for CMR purposes as if the contrast agent were spread uniformly through the blood. In this case, in general, the effective relaxation rate will be the weighted average of the relaxation in the two compart ments. That is, if for compartment i the volume fraction is fi, the initial T1 is T1i and the concentration of agent is Ci (which could be zero), the entire tissue together will behave as shown:     1 1 1 ¼ f1 þ r1 C 1 þ f 2 þ r2 C 2 T1 T11 T12

(6)

In “slow exchange,” the water exchange rate is much slower than the difference in relaxation rates between the compartments. In this case, a single relaxation time, and thus a single relaxivity, is meaningless, because the two micro scopic compartments will relax with their own relaxation times. Very few biologic compartments show true slow exchange, whereas the intermediate case, when exchange is neither slow nor fast (“intermediate exchange”) occurs very commonly. With intermediate exchange, relaxation behavior appears biexponential, although both the apparent compart ment size and the effective T1 of the two compartments will vary from their true biologic size and T1. It is possible to model the signal intensity behavior as a function of contrast agent concentration to estimate water exchange times in vivo. Although characterizing human tissue as having only one or two compartments is an oversimplification, these types of models have proved useful for explaining the effects of bio logic water mobility on contrast enhanced scans.37 Biologic compartmentalization also results in susceptibil ity contrast. The contrast agent causes microscopic field inhomogeneities, sometimes called “mesoscopic” inhomo geneities.38 Water diffusion causes the protons to dephase from one another because of the different magnetic fields that they experience. Even in the absence of water diffusion, the field inhomogeneity causes intravoxel dephasing and thus signal loss on GRE images because of the different

NEWER CONTRAST AGENTS The agent Gd BOPTA (gadobenate) has an approximately twofold higher relaxivity than the other ECF agents. Although it was designed for liver imaging, Gd BOPTA dis tributes to extracellular space in the same manner as Gd DTPA. The higher relaxivity can result in greater conspicu ity and higher reader confidence for detecting lesions.39 Similarly, Gd BOPTA has shown efficacy in contrast enhanced dynamic MRA and coronary MR.40,41 Recently, the hepatobiliary contrast agent Gd EOB DTPA (gadoxetic acid, marketed as Primovist in Europe and Eovist in the United States, Bayer Healthcare, Berlin, Germany) was approved for the detection and characteriza tion of liver lesions. Like Gd BOPTA, it has weak binding to plasma proteins and an approximately twofold higher relaxivity compared to ECF agents. In principal it should also be suitable for MRA applications but it is not approved for this purpose. The USPIO ferumoxytol (Feraheme, Advanced Mag netics, Cambridge, MA) was approved in 2009 for the treatment of iron deficiency anemia in adult patients with chronic kidney disease. This compound also produces good T1 and T2 contrast and an MRA indication is being pursued.42,43 MS 325 (gadofosveset, formerly marketed as Vasovist, now marketed as Ablavar by Lantheus Medical Imaging, Billerica, MA) was designed specifically for CE MRA applica tions and is approved for use in Europe, Canada, and Austra lia, and subsequently (December 2008) the first contrast agent approved for CE MRA in the United States. In the United States, gadofosveset is indicated for use as a contrast agent in MRA to evaluate aortoiliac occulusive disease in adults with known or suspected peripheral vascular disease. This agent may well have many uses beyond dynamic MRA. Its extended plasma half life and vascular retention enables steady state CE MRA, offering the potential benefits of imag ing multiple vascular beds with a single injection. It allows acquisition of images with higher spatial resolution than is achievable with dynamic MRA, eliminating the need for bolus timing, and it may be possible to administer the agent several minutes before the patient enters the magnet. It remains to be seen how dependent these benefits will be on the ability to do robust artery and vein separation.

NOVEL CONTRAST AGENTS IN DEVELOPMENT Molecular imaging has been defined as “the in vivo characteriza tion and measurement of biological processes at the cellular and molecular level.”44 Combining the high resolution anatomic and

functional imaging ability of CMR with the specificity of molecular targeting is very appealing, but somewhat limited by the inherent sensitivity limitations of CMR. Examples of molecular CMR con trast agents used for cardiovascular applications are numerous. The vast majority of these agents have only been studied in animal models. A few are described here. Scientists at EPIX Pharmaceuticals (Lexington, MA) designed gadolinium based agents that target fibrin,45,46 an abundant com ponent of thrombus. These molecules contain a peptide sequence specific for fibrin that has been linked to several gadolinium che lates. This class of compounds has been shown to positively enhance thrombus visualization in animal models of atherosclerotic plaque rupture47 and carotid artery thrombus,48 and in embolic models of thrombus in the coronary arteries,49,50 left atrium,51 and lung.52 Figure 6 8 shows an example of acute coronary throm bus visualization with the experimental fibrin targeted agent EP 2104R in a porcine model. This contrast agent has been shown to visualize thrombus in human subjects, one of the first examples of clinical translation of a molecular MR probe.53 The Barnes Hospital Washington University group developed a platform technology based on large particles prepared from a perfluorocarbon emulsion.54,55 The particles can be functiona lized by adding molecules with lipophilic components that are noncovalently incorporated into the particle. Signal generation arises from thousands of lipophilic gadolinium chelates incorporated on the surface. Targeting vectors can similarly be incorporated on the surface to guide the particle’s distribution. This group used antibodies to target the particles to either fibrin56 58 or tissue factor.59 Alternately, small molecules that bind to the avb3 integrin were incorporated and the particle was used to detect overexpression of this integrin that occurs during angio genesis.60 63 Use of the avb3 targeted particle to deliver fumagillin to inhibit angiogenesis has also been described while monitoring the process by molecular MRI.64 Imaging and characterization of atherosclerotic plaque has been an active area of investigation (see Chapters 27 and 28). There are some new contrast agents that have shown efficacy in animal models of atherosclerosis. One interesting class of compounds is the “gadofluorines.”65 67 These amphiphilic small molecules consist of a Gd chelate linked to a perfluorocarbon chain and a hydrophilic group, such as a sugar. Perfluorocarbons do not associate with lipids and impart interesting biodistribu tion properties. Gadofluorine M has been shown to localize in plaques induced in Watanabe heritable hyperlipidemic rabbits, but not in the vessel wall of normal New Zealand white rabbits.67 Figure 6 9 shows an image of a rabbit aorta taken 48 hours after gadofluorine M injection. The lesions are clearly delineated on this inversion recovery image and are confirmed with histology. The Mt. Sinai group has taken many approaches to imaging plaque targeting various plaque components. One approach was to use high density lipoprotein as a carrier for a lipophilic contrast agent.68 The high density lipoprotein is believed to facilitate trans port of the contrast agent to the plaque. They used mixed micelle platforms to nonspecifically target plaques69 and also incorporated antibodies to macrophage scavenging receptor to home the mixed micelles to macrophage in the plaque.70,71 In collaboration with the Guerbet group, they reported a Gd based agent targeting matrix metalloproteases in the plaque.72 A Gd based contrast agent specific to type I collagen was described,73 and this probe was used to show in vivo molecular imaging of fibrosis in a mouse model of healed postinfarction myocardial scarring.74 This probe may find utility in other pathol ogies with elevated collagen levels. The Massachusetts General Hospital group developed a gado linium based agent (MPO Gd) sensitive to myeloperoxidase (MPO) activity. Ischemic injury of the myocardium causes timed recruitment of neutrophils and monocytes/macrophages, which produce substantial amounts of local MPO. MPO forms reactive Cardiovascular Magnetic Resonance 85

6 CARDIOVASCULAR MAGNETIC RESONANCE CONTRAST AGENTS

microscopic magnetic fields within the voxel. The strength of the perturbing magnetic field is directly proportional to contrast agent concentration and its molar magnetic suscep tibility (w). The actual magnitude and even the direction of the magnetic field shifts depend strongly on the size and the shape of the biologic compartment in which the contrast agent resides38; the size of the susceptibility contrast effect depends on how the water diffuses through the tissue.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

A

B

C

D

Figure 6 9 Gadofluorine enhanced M (0.1 mmol/kg, 48 hours postinjection) image of atherosclerotic lesions in an 18 month old Watanabe hyperlipidemic rabbit. Segmented inversion recovery gradient recalled echo cardiovascular magnetic resonance. (Courtesy of Dr. B. Misselwitz, Bayer Schering, Berlin, Germany.) 86 Cardiovascular Magnetic Resonance

Figure 6 8 Acute coronary thrombus visualization with thrombin specific gadolinium based agent, EP 2104R. (A) Double oblique slice from bright blood coronary magnetic resonance angiography sequence (multiplanar reconstruction) demonstrating lumen of left anterior descending (LAD) artery with bright signal (arrows). (C) X ray angiography demonstrating thrombus in LAD (arrow). (B and D) Double oblique slices from an inversion recovery sequence (multiplanar reconstruction, same orientation as in A, before (B) and 2 hours after (D) administration of the agent. The left anterior descending coronary artery thrombus is seen as a “bright” spot on postcontrast images (arrow in B and D). (Courtesy of Dr. A. Wiethof, EPIX Pharmaceuticals, Cambridge, MA.)

chlorinating species capable of inflicting oxidative stress and alter ing protein function by covalent modification. MPO Gd is first radicalized by MPO and then either spontaneously oligomerizes or binds to matrix proteins, all leading to enhanced spin lattice relaxivity and delayed washout kinetics. MPO Gd was used to measure inflammatory responses after myocardial ischemia locally and noninvasively in a murine model.75 There is a rich literature describing the use of iron oxide parti cles as targeted contrast agents. Commercial USPIOs passively accumulate in macrophages. This property has been exploited to provide negative contrast in macrophage rich atherosclerotic pla ques. There appears to be a positive correlation between macro phage uptake and plasma half life.76,77 Iron oxides have also been used to label cells, and then the cells are tracked in vivo using CMR.78,79 Visualization of mesen chymal stem cells engrafted into the myocardium of a pig has been demonstrated using a 1.5 T clinical scanner.80,81 Injection sites containing greater than 105 cells could be detected in vivo.80 In a mouse model of myocardial infarction, a therapeu tic intervention of mouse embryonic stem cells could be followed by visualizing the stem cells and following their effect on LV func tion over a period of 4 weeks.82 The Massachusetts General Hospital group has also shown that the iron oxide coating material can be chemically modified to introduce new targeting vectors.83 85 They termed this contrast agent platform “cross linked iron oxide.” For example, annexin V can be grafted onto a cross linked iron oxide86 to detect apopto tic cardiomyocytes in a mouse model of transient left anterior des cending artery occlusion.87 Another CLIO based probe is targeted

SAFETY The safety of CMR contrast agents, which themselves offer no direct therapeutic benefit, is always a question of appropriate medical concern. The Gd based chelates had initially been considered among the safest injectable compounds in medical use, with a specific reputation for superior safety in patients with renal dysfunction (compared with an iodinated prepara tion). This view has changed with the apparent link between Gd contrast and nephrogenic systemic fibrosis (NSF). A very rare, but potentially devastating condition affect ing patients with chronic kidney disease, NSF has an esti mated prevalence of more than 10% among patients receiving Gd who are on hemodialysis.93 Although present ing primarily with skin thickening, tethering, hyperpig mentation, and disabling joint flexion contractures, patients with NSF can also have multiorgan/systemic fibro sis, leading to organ dysfunction and even death.94,95 The universal feature of NSF is Gd administration and chronic kidney disease, most commonly with an estimated glomer ular filtration rate of less than 30 mL/min/1.73m2.95–98 Gadolinium has been shown to be present in skin biopsy specimens of patients with NSF,99–104 and also in internal organs.105 The prevailing hypothesis is that chronic kidney disease results in prolonged exposure to the contrast agent, providing opportunity for the Gd to dissociate from its che lator, and this dissociated Gd is believed to be responsible for the toxic response.106 The least stable contrast agent, Gd DTPA BMA (gadodiamide), is the agent that has been most frequently associated with NSF,107–109 but it is now established that other Gd compounds also pose a risk.93,110

Although ECF agents are similar in terms of their imaging efficacy, they differ in terms of how well they bind Gd. This may prove to be an important risk factor for NSF because both the metal (Gd) and the chelate have potential toxic effects5 and have shown acute toxicity in animal studies at high enough doses. These various animal studies have been reviewed,111 and a recent high dose study in rats showed NSF like lesions.109 Metal complexes, such as Gd contrast agents, are characterized in terms of their thermody namic stability and kinetic inertness. Stability is a measure of the affinity of the chelator for the metal ion and is expressed as an equilibrium constant for the association of the chelator (ligand) with the metal. Stability constants for approved agents have the order Gd DOTA (gadoterate) > Gd HPDO3A (gadoteridol)  Gd DO3A butrol (gadobutrol)  MS 325 (gadofosveset)  Gd BOPTA (gadobenate)  Gd DTPA (gadopentetate) > Gd DTPA BMA (gadoversetamide)  Gd DTPA BMEA (gadodiamide).4 The compounds with lowest thermodynamic stability are the neutral complexes with acyclic chelators: Gd DTPA BMA (gadodiamide) and Gd DTPA BMEA (gadoversetamide). Stability constants describe equilibrium values, but do not indicate how quickly equilibrium is reached. In the biologic milieu, protons and other metal ions compete to bind the chelator ligand, whereas there are ions such as phosphate and carbonate that have a high affinity for Gd. Under conditions that favor dissociation of the Gd (e.g., low pH), two complexes with comparable stability may have very different rates of dissociation. For example, the macrocyclic complex Gd HPDO3A (gadoteridol) is an example of a compound that is more kinetically inert to Gd loss than Gd DTPA (gadopentetate), which has a similar stability constant. The rate of acid assisted Gd dissociation is 20 times faster for Gd DTPA (gadopentetate) than for Gd HPDO3A (gadoteridol).112 Laurent and colleagues measured the rate of Gd loss in the presence of zinc and phosphate under a standard set of conditions for various approved agents.113,114 The macrocyclic complexes Gd DOTA (gadoterate), Gd HPDO3A (gadoteridol), and Gd DO3A butrol (gadobutrol) were the most inert with respect to Gd loss. The compound that released Gd the most rap idly was Gd DTPA BMA (gadoversetamide) under the con ditions specified by Laurent and colleagues. These differences in kinetics and thermodynamic stabilities may well translate into safety with regard to patients with chronic kidney disease. For instance, there appears to be a much lower incidence of NSF among patients who received Gd HPDO3A (gadoteridol),115 and this may be a result of its combination of stability and inertness. Although care should be taken when considering the use of Gd based contrast in renally compromised patients (estimated glomerular filtration rate < 60 mL/min/1.73m2), it is impor tant to keep this risk in perspective. Gd based contrast agents have been used safely in millions of patients, and to date there have been no reports of Gd associated NSF in patients with normal renal function. Regarding other side effects, Kirchin and Runge116 reviewed the safety record of clinically approved contrast agents as of 2003. The seven approved Gd agents (Gd DTPA [gadopentetate], Gd DOTA [gadoterate], Gd BOPTA [gadobenate], Gd DTPA BMA [gadoversetamide], Gd DTPA BMEA [gadodiamide), Gd DO3A butrol [gadobu trol], and Gd BOPTA [gadobenate]) appeared indistinguish able with respect to their safety profile. The most common Cardiovascular Magnetic Resonance 87

6 CARDIOVASCULAR MAGNETIC RESONANCE CONTRAST AGENTS

to vascular cell adhesion molecule 1 using a peptide specific to this molecule.88 This probe was used to identify inflammatory activation of cells in a mouse model of atherosclerosis. The lack of sensitivity in CMR stems from the very small degree of polarization among the nuclear spins. In a magnetic field there is a net magnetization, but this is small, and approxi mately 0.0006% of the spins are polarized. A technique called “spin exchange” uses a high powered laser (also called “optical pumping”) to increase the polarization by four to five orders of magnitude (hyperpolarization).89 Isotopes with long T1 values can be hyperpolarized and used as contrast agents. The long T1 is necessary to maintain the contrast medium in the hyperpolar ized state long enough to obtain an image before the spins relax back to the equilibrium value. Gases often have long T1 values, and isotopes of the noble gases helium (He 3) and xenon (Xe 129) have been used for imaging. Contrast agents with hyperpolarized C 13 have been reported using a C 13 enriched water soluble compound90 with long relaxation times (in vitro: T1 ¼ approximately 82 seconds, T2 ¼ approxi mately 18 seconds; in vivo: T1 ¼ approximately 38 seconds, T2 ¼ approximately 1.3 seconds). This could be formulated at a C 13 concentration of 200 mM and hyperpolarized to 15%. The authors used this material for CE MRA in rats and swine.91 Hyperpolarized C 13 also offers the potential for investigating metabolic pathways. For instance, Merritt and colleagues studied the metabolism of [1 13C] pyruvate in a perfused rat heart.92 This field is likely to expand in the coming years. A major benefit of hyperpolarized contrast media is the excel lent sensitivity with no background (high signal to noise ratio). Challenges include the distribution and availability of the hyper polarization equipment and imaging hardware compatibility for imaging nonhydrogen nuclei (not available on all clinical scan ners). However, there is now a commercial polarizer available.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

adverse events were headache, nausea, taste perversion, and urticaria (hives). Nearly all adverse events with these agents were transient, mild, and self limiting. Nevertheless, there are reports of serious adverse reactions, including life threat ening anaphylactoid reactions and death. The best estimate puts the rate of these events at between 1 in 200,000 and 1 in 400,000 patient administrations.117 The Gd based ECF contrast agents have been studied, and most are approved for pediatric use in patients older than 2 years, though there are differences in their approval wording. One distinguishing clinical indicator among the Gd based agents is that patients receiving Gd DTPA BMA (gadoversetamide) or Gd DTPA BMEA (gadodiamide) often show spuriously low serum cal cium levels.118 These two contrast agents (but not the other Gd based agents) appear to interfere with the reagent used in the clinical chemistry test for calcium.119 The safety profile for current CMR contrast agents of course will not necessarily be the same for future com pounds. For all contrast agents, the package insert should always be consulted for the latest safety information.

CONCLUSION A large number of contrast agents have been approved for MRI in the last 18 years. In December 2008, the first blood pool agent (gadofosveset) was U.S. Food and Drug Administration approved for a CMR application. A signifi cant number of new agents, including blood pool and tis sue specific agents that are potentially relevant for CMR, are currently in development. The behavior of these agents, which often is summarized by a single effectiveness param eter, r1, is actually quite complex, in terms of both the underlying chemistry and the biophysics in vivo. Although many elements of that complexity remain active areas of research, both for understanding existing agents as well as for creating new agents and new uses for those agents, the relative safety and ease of use of these agents has brought them into routine use in many medical applica tions. As clinical CMR expands, no doubt the use of these contrast agents will expand as well.

References 1. Bloch F, Hansen WW, Packard M. The nuclear induction experiment. Phys Rev. 1948;70:474–485. 2. Lauterbur PC, Mendonca Dias MH, Rudin AM. Augmentation of tissue water proton spin-lattice relaxation rates by in vivo addition of paramagnetic ions. In: Dutton PO, Leigh J, Scarpa A, eds. Frontiers of Biological Energetics. New York: Academic Press; 1978:752–759. 3. Carr DH, Brown J, Bydder GM, et al. Intravenous chelated gadolinium as a contrast agent in NMR imaging of cerebral tumors. Lancet. 1984;1:484–486. 4. Caravan P, Ellison JJ, McMurry TJ, et al. Gadolinium(III) chelates as MRI contrast agents: structure, dynamics, and applications. Chem Rev. 1999;99:2293–2352. 5. Gries H. Extracellular MRI contrast agents based on gadolinium. Top Curr Chem. 2002;221:1–24. 6. Powell DH, Ni Dhubhghaill OM, Pubanz D, et al. High-pressure NMR kinetics. Part 74. Structural and dynamic parameters obtained from 17O NMR, EPR, and NMRD Studies of monomeric and dimeric Gd3þ Complexes of Interest in magnetic resonance imaging: an integrated and theoretically self-consistent approach. J Am Chem Soc. 1996;118:9333–9346. 7. Gillis A, Gray M, Burstein D. Relaxivity and diffusion of gadolinium agents in cartilage. Magn Reson Med. 2002;48:1068–1071. 8. Kumar K, Tweedle MF. Ligand basicity and rigidity control formation of macrocyclic polyamino carboxylate complexes of gadolinium(III). Inorg Chem. 1993;32:4193–4199. 9. White DH, DeLearie LA, Moore DA, et al. The thermodynamics of complexation of lanthanide(III) DTPA-bisamide complexes and their implication for stability and solution structure. Invest Radiol. 1991; 26:S226–S228. 10. Tombach B, Heindel W. Value of 1.0-M gadolinium chelates: review of preclinical and clinical data on gadobutrol. Eur Radiol. 2002;12: 1550–1556. 11. Caravan P, Cloutier NJ, Greenfield MT, et al. The interaction of MS325 with human serum albumin and its effect on proton relaxation rates. J Am Chem Soc. 2002;124:3152–3162. 12. Lauffer RB, Parmelee DJ, Dunham SU, et al. MS-325: albumin: targeted contrast agent for MR angiography. Radiology. 1998;207: 529–538. 13. Deshpande VS, Cavagna F, Maggioni F. Comparison of gradient echo and SSFP for coronary artery MR angiography using a gadolinium based intravascular contrast agent. Invest Radiol. 2006;41:292–298. 14. de Hae¨n C, Anelli PL, Lorusso V, et al. Gadocoletic acid trisodium salt (B22956/1): a new blood pool magnetic resonance contrast agent with application in coronary angiography. Invest Radiol. 2006;41: 279–291. 15. de Hae¨n C, La Ferla R, Maggioni F. Gadobenate dimeglumine 0.5 M solution for injection (MultiHance) as contrast agent for magnetic 88 Cardiovascular Magnetic Resonance

16.

17. 18. 19. 20. 21.

22.

23. 24. 25.

26. 27. 28. 29.

resonance imaging of the liver: mechanistic studies in animals. J Comp Asst Tomog. 1999;23:S169–S179. Vander Elst L, Maton F, Laurent S, et al. A multinuclear MR study of Gd-EOB-DTPA: comprehensive preclinical characterization of an organ specific MRI contrast agent. Magn Reson Med. 1997;38: 604–614. Eldredge HB, Spiller M, Chasse JM, et al. Species dependence on plasma protein binding and relaxivity of the gadolinium-based MRI contrast agent MS-325. Invest Radiol. 2006;41:229–243. Port M, Corot C, Violas X, et al. How to compare the efficiency of albumin-bound and nonalbumin-bound contrast agents in vivo: the concept of dynamic relaxivity. Invest Radiol. 2005;40:565–573. Dong Q, Hurst DR, Weinmann HJ, et al. Magnetic resonance angiography with gadomer-17: an animal study original investigation. Invest Radiol. 1998;33:699–708. Nicolle GM, Toth E, Schmitt-Willich H, et al. The impact of rigidity and water exchange on the relaxivity of a dendritic MRI contrast agent. Chem Eur J. 2002;8:1040–1048. Schnorr J, Wagner S, Abramjuk C, et al. Comparison of the iron oxide-based blood-pool contrast medium VSOP-C184 with gadopentetate dimeglumine for first-pass magnetic resonance angiography of the aorta and renal arteries in pigs. Invest Radiol. 2004;39: 546–553. Tombach B, Reimer P, Bremer C, et al. First-pass and equilibriumMRA of the aortoiliac region with a superparamagnetic iron oxide blood pool MR contrast agent (SH U 555 C): results of a human pilot study. NMR Biomed. 2004;17:500–506. Lauffer RB. Targeted relaxation enhancement agents for MRI. Magn Reson Med. 1991;22:339. Bogdanov AA, Weissleder R, Frank HW, et al. A new macromolecule as a contrast agent for MR angiography: preparation, properties, and animal studies. Radiology. 1993;187:701–706. Toth E, Van Uffelen I, Helm L, et al. Gadolinium-based linear polymer with temperature-independent proton relaxivities: a unique interplay between the water exchange and rotational contributions. Magn Reson Chem. 1998;36:S125–S134. Bertini I, Luchinat C, Parigi G. Solution NMR of Paramagnetic Molecules. Amsterdam: Elsevier; 2001. Jung CW, Jacobs P. Physical and chemical properties of superparamagnetic iron oxide MR contrast agents: ferumoxides, ferumoxtran, ferumoxsil. Magn Reson Imaging. 1995;13:661–674. Muller RN, Roch A, Colet J-M, et al. Particulate magnetic contrast agents. In: Merbach AE, Toth E, eds. Chemistry of Contrast Agents in Medical Magnetic Resonance Imaging. 2001:417–435. Tweedle MF, Wedeking P, Telser J, et al. Dependence of MR signal intensity on Gd tissue concentration over a broad dose range. Magn Reson Med. 1991;22:191–194.

55. Lanza GM, Winter P, Caruthers S, et al. Novel paramagnetic contrast agents for molecular imaging and targeted drug delivery. Curr Pharm Biotechnol. 2004;5:495–507. 56. Winter PM, Caruthers SD, Yu X, et al. Improved molecular imaging contrast agent for detection of human thrombus. Magn Reson Med. 2003;50:411–416. 57. Flacke S, Fischer S, Scott MJ, et al. Novel MRI contrast agent for molecular imaging of fibrin: implications for detecting vulnerable plaques. Circulation. 2001;104:1280–1285. 58. Yu X, Song SK, Chen J, et al. High-resolution MRI characterization of human thrombus using a novel fibrin-targeted paramagnetic nanoparticle contrast agent. Magn Reson Med. 2000;44:867–872. 59. Morawski AM, Winter PM, Crowder KC, et al. Targeted nanoparticles for quantitative imaging of sparse molecular epitopes with MRI. Magn Reson Med. 2004;51:480–486. 60. Schmieder AH, Winter PM, Caruthers SD, et al. Molecular MR imaging of melanoma angiogenesis with alphanubeta3-targeted paramagnetic nanoparticles. Magn Reson Med. 2005;53:621–627. 61. Winter PM, Morawski AM, Caruthers SD, et al. Molecular imaging of angiogenesis in early-stage atherosclerosis with alpha(v)beta3integrin-targeted nanoparticles. Circulation. 2003;108:2270–2274. 62. Winter PM, Caruthers SD, Kassner A, et al. Molecular imaging of angiogenesis in nascent Vx-2 rabbit tumors using a novel alpha(nu) beta3-targeted nanoparticle and 1.5 tesla magnetic resonance imaging. Cancer Res. 2003;63:5838–5843. 63. Anderson SA, Rader RK, Westlin WF, et al. Magnetic resonance contrast enhancement of neovasculature with alpha(v)beta(3)-targeted nanoparticles. Magn Reson Med. 2000;44:433–439. 64. Winter PM, Neubauer AM, Caruthers SD, et al. Endothelial alpha(v) beta3 integrin-targeted fumagillin nanoparticles inhibit angiogenesis in atherosclerosis. Arterioscler Thromb Vasc Biol. 2006;26:2103–2109. 65. Meding J, Urich M, Licha K, et al. Magnetic resonance imaging of atherosclerosis by targeting extracellular matrix deposition with Gadofluorine M. Contrast Media Mol Imaging. 2007;2:120–129. 66. Sirol M, Itskovich VV, Mani V, et al. Lipid-rich atherosclerotic plaques detected by gadofluorine-enhanced in vivo magnetic resonance imaging. Circulation. 2004;109:2890–2896. 67. Barkhausen J, Ebert W, Heyer C, et al. Detection of atherosclerotic plaque with Gadofluorine-enhanced magnetic resonance imaging. Circulation. 2003;108:605–609. 68. Frias JC, Williams KJ, Fisher EA, et al. Recombinant HDL-like nanoparticles: a specific contrast agent for MRI of atherosclerotic plaques. J Am Chem Soc. 2004;126:16316–16317. 69. Briley-Saebo KC, Amirbekian V, Mani V, et al. Gadolinium mixedmicelles: effect of the amphiphile on in vitro and in vivo efficacy in apolipoprotein E knockout mouse models of atherosclerosis. Magn Reson Med. 2006;56:1336–1346. 70. Mulder WJ, Strijkers GJ, Briley-Saboe KC, et al. Molecular imaging of macrophages in atherosclerotic plaques using bimodal PEG-micelles. Magn Reson Med. 2007;58:1164–1170. 71. Amirbekian V, Lipinski MJ, Briley-Saebo KC, et al. Detecting and assessing macrophages in vivo to evaluate atherosclerosis noninvasively using molecular MRI. Proc Natl Acad Sci USA. 2007;104:961–966. 72. Lancelot E, Amirbekian V, Brigger I, et al. Evaluation of matrix metalloproteinases in atherosclerosis using a novel noninvasive imaging approach. Arterioscler Thromb Vasc Biol. 2008;28:425–432. 73. Caravan P, Das B, Dumas S, et al. Collagen-targeted MRI contrast agent for molecular imaging of fibrosis. Angew Chem Int Ed Engl. 2007;46:8171–8173. 74. Helm PA, Caravan P, French BA, et al. Postinfarction myocardial scarring in mice: molecular MR imaging with use of a collagen-targeting contrast agent. Radiology. 2008;247:788–796. 75. Nahrendorf M, Sosnovik D, Chen JW, et al. Activatable magnetic resonance imaging agent reports myeloperoxidase activity in healing infarcts and noninvasively detects the antiinflammatory effects of atorvastatin on ischemia-reperfusion injury. Circulation. 2008;117: 1153–1160. 76. Corot C, Petry KG, Trivedi R, et al. Macrophage imaging in central nervous system and in carotid atherosclerotic plaque using ultrasmall superparamagnetic iron oxide in magnetic resonance imaging. Invest Radiol. 2004;39:619–625. 77. Yancy AD, Olzinski AR, Hu TC, et al. Differential uptake of ferumoxtran-10 and ferumoxytol, ultrasmall superparamagnetic iron oxide contrast agents in rabbit: critical determinants of atherosclerotic plaque labeling. J Magn Reson Imaging. 2005;21:432–442. 78. Modo M, Hoehn M, Bulte JW. Cellular MR imaging. Mol Imaging. 2005;4:143–164.

Cardiovascular Magnetic Resonance 89

6 CARDIOVASCULAR MAGNETIC RESONANCE CONTRAST AGENTS

30. Wedeking P, Sotak CH, Telser J, et al. Quantitative dependence of MR signal intensity on tissue concentration of Gd(HP-DO3A) in the nephrectomized rat. Magn Reson Imaging. 1992;10:97–108. 31. Stanisz GJ, Henkelman RM. Gd-DTPA relaxivity depends on macromolecular content. Magn Reson Med. 2000;44:665–667. 32. Donahue KM, Weisskoff RM, Burstein D. Water diffusion and exchange as they influence contrast enhancement. J Magn Reson Imaging. 1997;7:102–110. 33. Li X, Rooney WD, Springer Jr CS. A unified magnetic resonance imaging pharmacokinetic theory: intravascular and extracellular contrast reagents. Magn Reson Med. 2005;54:1351–1359. 34. Hazlewood CF, Chang DC, Nichols BL. Nuclear magnetic resonance transverse relaxation times of water protons in skeletal muscle. Biophys J. 1974;14:583–606. 35. McLaughlin AC, Leigh JS. Relaxation times in systems with chemical exchange. J Magn Reson. 1973;9:296–304. 36. Wright GA, Hu BS, Macovski A. Estimating oxygen saturation of blood in vivo with MR imaging at 1.5 T. J Magn Reson Imaging. 1991;1: 275–283. 37. Donahue KM, Weisskoff RM, Chesler DA, et al. Improving MR quantification of regional blood volume with intravascular T1 contrast agents: accuracy, precision, and water exchange. Magn Reson Med. 1996;36:858–867. 38. Sukstanskii AL, Yablonskiy DA. Gaussian approximation in the theory of MR signal formation in the presence of structure-specific magnetic field inhomogeneities: effects of impermeable susceptibility inclusions. J Magn Reson. 2004;167:56–67. 39. Knopp MV, Runge VM, Essig M, et al. Primary and secondary brain tumors at MR imaging: bicentric intraindividual crossover comparison of gadobenate dimeglumine and gadopentetate dimeglumine. Radiology. 2004;230:55–64. 40. Goyen M, Debatin JF. Gadobenate dimeglumine (MultiHance) for magnetic resonance angiography: review of the literature. Eur Radiol. 2003;13:N19–N27. 41. Wikstrom J, Wasser MN, Pattynama PM, et al. Gadobenate dimeglumine-enhanced magnetic resonance angiography of the pelvic arteries. Invest Radiol. 2003;38:504–515. 42. Neuwelt EA, Varallyay CG, Manninger S, et al. The potential of ferumoxytol nanoparticle magnetic resonance imaging, perfusion, and angiography in central nervous system malignancy: a pilot study. Neurosurgery. 2007;60:601–611. 43. Li W, Salanitri J, Tutton S, et al. Lower extremity deep venous thrombosis: evaluation with ferumoxytol-enhanced MR imaging and dual-contrast mechanism—preliminary experience. Radiology. 2007;242:873–881. 44. Weissleder R, Mahmood U. Molecular imaging. Radiology. 2001;219: 316–333. 45. Caravan P, Kolodziej AF, Greenwood JM, et al. EP-1242: A Fibrin Targeted Contrast Agent for Thrombus Imaging. In: 10th ISMRM Scientific Sessions. HI, USA: Honolulu; 2002. 46. Overoye-Chan K, Koerner S, Looby RJ, et al. EP-2104R: a fibrinspecific gadolinium-based MRI contrast agent for detection of thrombus. J Am Chem Soc. 2008;130:6025–6039. 47. Botnar RM, Perez AS, Witte S, et al. In vivo molecular imaging of acute and subacute thrombosis using a fibrin-binding magnetic resonance imaging contrast agent. Circulation. 2004;109:2023–2029. 48. Sirol M, Aguinaldo JG, Graham PB, et al. Fibrin-targeted contrast agent for improvement of in vivo acute thrombus detection with magnetic resonance imaging. Atherosclerosis. 2005;182:79–85. 49. Spuentrup E, Buecker A, Katoh M, et al. Molecular magnetic resonance imaging of coronary thrombosis and pulmonary emboli with a novel fibrin-targeted contrast agent. Circulation. 2005;111:1377–1382. 50. Botnar RM, Buecker A, Wiethoff AJ, et al. In vivo magnetic resonance imaging of coronary thrombosis using a fibrin-binding molecular magnetic resonance contrast agent. Circulation. 2004;110: 1463–1466. 51. Spuentrup E, Katoh M, Wiethoff AJ, et al. Molecular magnetic resonance imaging of pulmonary emboli with a fibrin-specific contrast agent. Am J Respir Crit Care Med. 2005;172:494–500. 52. Spuentrup E, Katoh M, Buecker A, et al. Molecular MR imaging of human thrombi in a swine model of pulmonary embolism using a fibrin-specific contrast agent. Invest Radiol. 2007;42:586–595. 53. Spuentrup E, Botnar RM, Wiethoff AJ, et al. MR imaging of thrombi using EP-2104R, a fibrin-specific contrast agent: initial results in patients. Eur Radiol. 2008;18:1995–2005. 54. Cyrus T, Winter PM, Caruthers SD, et al. Magnetic resonance nanoparticles for cardiovascular molecular imaging and therapy. Expert Rev Cardiovasc Ther. 2005;3:705–715.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

79. Bulte JW, Kraitchman DL. Monitoring cell therapy using iron oxide MR contrast agents. Curr Pharm Biotechnol. 2004;5:567–584. 80. Hill JM, Dick AJ, Raman VK, et al. Serial cardiac magnetic resonance imaging of injected mesenchymal stem cells. Circulation. 2003;108: 1009–1014. 81. Kraitchman DL, Heldman AW, Atalar E, et al. In vivo magnetic resonance imaging of mesenchymal stem cells in myocardial infarction. Circulation. 2003;107:2290–2293. 82. Arai T, Kofidis T, Bulte JW, et al. Dual in vivo magnetic resonance evaluation of magnetically labeled mouse embryonic stem cells and cardiac function at 1.5 t. Magn Reson Med. 2006;55:203–209. 83. Wunderbaldinger P, Josephson L, Weissleder R. Crosslinked iron oxides (CLIO): a new platform for the development of targeted MR contrast agents. Acad Radiol. 2002;9:S304–S306. 84. Moore A, Sun PZ, Cory D, et al. MRI of insulitis in autoimmune diabetes. Magn Reson Med. 2002;47:751–758. 85. Hogemann D, Josephson L, Weissleder R. Improvement of MRI probes to allow efficient detection of gene expression. Bioconjug Chem. 2000;11:941–946. 86. Schellenberger EA, Bogdanov Jr A, Hogemann D. Annexin V-CLIO: a nanoparticle for detecting apoptosis by MRI. Mol Imaging. 2002;1:102–107. 87. Sosnovik DE, Schellenberger EA, Nahrendorf M, et al. Magnetic resonance imaging of cardiomyocyte apoptosis with a novel magnetooptical nanoparticle. Magn Reson Med. 2005;54:718–724. 88. Nahrendorf M, Jaffer FA, Kelly KA, et al. Noninvasive vascular cell adhesion molecule-1 imaging identifies inflammatory activation of cells in atherosclerosis. Circulation. 2006;114:1504–1511. 89. Moller HE, Chen XJ, Saam B, et al. MRI of the lungs using hyperpolarized noble gases. Magn Reson Med. 2002;47:1029–1051. 90. Svensson J, Mansson S, Johansson E, et al. Hyperpolarized 13C MR angiography using trueFISP. Magn Reson Med. 2003;50:256–262. 91. Olsson LE, Chai CM, Axelsson O, et al. MR coronary angiography in pigs with intraarterial injections of a hyperpolarized 13C substance. Magn Reson Med. 2006;55:731–737. 92. Merritt ME, Harrison C, Storey C, et al. Hyperpolarized 13C allows a direct measure of flux through a single enzyme-catalyzed step by NMR. Proc Natl Acad Sci USA. 2007;104:19773–19777. 93. Todd DJ, Kagan A, Chibnik LB, et al. Cutaneous changes of nephrogenic systemic fibrosis: predictor of early mortality and association with gadolinium exposure. Arthritis Rheum. 2007;56:3433–3441. 94. Cowper SE. Nephrogenic systemic fibrosis: a review and exploration of the role of gadolinium. Adv Dermatol. 2007;23:131–154. 95. Cowper SE, Kuo PH, Bucala R. Nephrogenic systemic fibrosis and gadolinium exposure. Association and lessons for idiopathic fibrosing disorders. Arthritis Rheum. 2007;56:3173–3175. 96. Grobner T. Gadolinium–a specific trigger for the development of nephrogenic fibrosing dermopathy and nephrogenic systemic fibrosis? Nephrol Dial Transplant. 2006;21:1104–1108. 97. Thomsen HS. Nephrogenic systemic fibrosis: a serious late adverse reaction to gadodiamide. Eur Radiol. 2006;16:2619–2621. 98. Grobner T, Prischl FC. Gadolinium and nephrogenic systemic fibrosis. Kidney Int. 2007;72:260–264. 99. Boyd AS, Zic JA, Abraham JL. Gadolinium deposition in nephrogenic fibrosing dermopathy. J Am Acad Dermatol. 2007;56:27–30. 100. High WA, Ayers RA, Chandler J, et al. Gadolinium is detectable within the tissue of patients with nephrogenic systemic fibrosis. J Am Acad Dermatol. 2007;56:21–26. 101. High WA, Ayers RA, Cowper SE. Gadolinium is quantifiable within the tissue of patients with nephrogenic systemic fibrosis. J Am Acad Dermatol. 2007;56:710–712. 102. Saussereau E, Lacroix C, Cattaneo A, et al. Hair and fingernail gadolinium ICP-MS contents in an overdose case associated with nephrogenic systemic fibrosis. Forensic Sci Int. 2008;176:54–57. 103. Thakral C, Alhariri J, Abraham JL. Long-term retention of gadolinium in tissues from nephrogenic systemic fibrosis patient after multiple gadolinium-enhanced MRI scans: case report and implications. Contrast Media Mol Imaging. 2007;2:199–205. 104. Abraham JL, Thakral C, Skov L, et al. Dermal inorganic gadolinium concentrations: evidence for in vivo transmetallation and long-term persistence in nephrogenic systemic fibrosis. Br J Dermatol. 2008; 158:273–280.

90 Cardiovascular Magnetic Resonance

105. Kay J, Bazari H, Avery LL, et al. Case records of the Massachusetts General Hospital. Case 6-2008: a 46-year-old woman with renal failure and stiffness of the joints and skin. N Engl J Med. 2008;358: 827–838. 106. Kurtkoti J, Snow T, Hiremagalur B. Gadolinium and nephrogenic systemic fibrosis: association or causation. Nephrology (Carlton). 2008;. 107. Richmond H, Zwerner J, Kim Y, et al. Nephrogenic systemic fibrosis: relationship to gadolinium and response to photopheresis. Arch Dermatol. 2007;143:1025–1030. 108. Rydahl C, Thomsen HS, Marckmann P. High prevalence of nephrogenic systemic fibrosis in chronic renal failure patients exposed to gadodiamide, a gadolinium-containing magnetic resonance contrast agent. Invest Radiol. 2008;43:141–144. 109. Sieber MA, Pietsch H, Walter J, et al. A preclinical study to investigate the development of nephrogenic systemic fibrosis: a possible role for gadolinium-based contrast media. Invest Radiol. 2008;43: 65–75. 110. Leiner T, Herborn CU, Goyen M. Nephrogenic systemic fibrosis is not exclusively associated with gadodiamide. Eur Radiol. 2007;17: 1921–1923. 111. Caravan P, Lauffer RB. Contrast agents: Basic principles. In: Edelman RR, Hesselink JR, Zlatkin MB, et al, eds. Clinical Magnetic Resonance Imaging. 3rd ed. Philadelphia: Saunders; 2005:357–375. 112. Kumar K, Chang CA, Tweedle MF. Equilibrium and kinetic studies of lanthanide complexes of macrocyclic polyamino carboxylates. Inorg Chem. 1993;32:587–593. 113. Laurent S, Elst LV, Muller RN. Comparative study of the physicochemical properties of six clinical low molecular weight gadolinium contrast agents. Contrast Media Mol Imaging. 2006;1:128–137. 114. Laurent S, Elst LV, Copoix F, et al. Stability of MRI paramagnetic contrast media: a proton relaxometric protocol for transmetallation assessment. Invest Radiol. 2001;36:115–122. 115. Reilly RF. Risk for nephrogenic systemic fibrosis with gadoteridol (prohance) in patients who are on long-term hemodialysis. Clin J Am Soc Nephrol. 2008;3:747–751. 116. Kirchin MA, Runge VM. Contrast agents for magnetic resonance imaging: safety update. Top Magn Reson Imaging. 2003;14:426–435. 117. Carr JJ. Magnetic resonance contrast agents for neuroimaging. Safety issues. Neuroimaging Clin N Am. 1994;4:43–54. 118. Prince MR, Erel HE, Lent RW, et al. Gadodiamide administration causes spurious hypocalcemia. Radiology. 2003;227:639–646. 119. Emerson J, Kost G. Spurious hypocalcemia after Omniscan- or OptiMARK-enhanced magnetic resonance imaging: an algorithm for minimizing a false-positive laboratory value. Arch Pathol Lab Med. 2004;128:1151–1156. 120. Rohrer M, Bauer H, Mintorovitch J, et al. Comparison of magnetic properties of MRI contrast media solutions at different magnetic field strengths. Invest Radiol. 2005;40:715–724. 121. Tweedle MF, Hagan JJ, Kumar K, et al. Reaction of gadolinium chelates with endogenously available ions. Magn Reson Imaging. 1991;9:409–415. 122. Tweedle MF. Physicochemical properties of gadoteridol and other magnetic resonance contrast agents. Invest Radiol. 1992;27:2–6. 123. Vogler H, Platzek J, Schuhmann-Giampieri G, et al. Pre-clinical evaluation of gadobutrol: a new, neutral, extracellular contrast agent for magnetic resonance imaging. Eur J Radiol. 1995;21:1–10. 124. OptiMARK Package Insert. St. Louis: Mallinckrodt, Inc.. 125. Cavagna FM, Lorusso V, Anelli PL, et al. Preclinical profile and clinical potential of gadocoletic acid trisodium salt (B22956/1), a new intravascular contrast medium for MRI. Acad Radiol. 2002;9: S491–S494. 126. Uggeri F, Aime S, Anelli PL, et al. Novel contrast agents for magnetic resonance imaging: synthesis and characterization of the ligand BOPTA and Its Ln(III) Complexes (Ln Gd, La, Lu). X-ray Structure of Disodium (TPS-9-145337286-C-S)-[4-Carboxy-5,8,11-tris(carboxymethyl)-1-phenyl-2-oxa- 5,8,11-triazatridecan-13-oato(5-)]gadolinate(2-) in a mixture with its enantiomer. Inorg Chem. 1995;34:633–642. 127. de Hae¨n C, Gozzini L. Soluble-type hepatobiliary contrast agents for MR imaging. J Magn Reson Imaging. 1993;1993:3.

Blood Flow Velocity Assessment David Firmin

The idea of mapping measurements of blood flow onto a magnetic resonance (MR) image was first discussed in an article by Singer in 1978.1 The methods that followed could generally be categorized into time of flight or phase shift types, and were based on the techniques that had previously been described for nonimaging MR flow stud ies.2 A number of review articles have covered the subject and described the variety of methods3–5 that have been used and validated both in vitro and in vivo. The interest in flow in MR imaging has not been solely directed toward the goal of quantitative flow measurement. A large amount of effort has also been devoted to understanding the appearance of a flowing fluid on an image because this can often be indicative of the type of flow present and therefore can give important information on the diagnosis of a particular disorder. Also, the development of MR angiography techniques has required a full understanding of these effects. In 1984, soon after the development of the first clinical MR scanners, there was an increase in interest in the search for an MR method of imaging flow. Review articles were published and a number of techniques described.6–10 This chapter provides a description of and a brief historical over view of the methods that have been used to measure blood flow in the heart and great vessels.

TIME-OF-FLIGHT METHODS There are two categories of time of flight techniques. The first, often known as washin/washout, or flow enhancement, methods, normally rely on the saturation or partial satura tion of material in a selected slice or volume being replaced by fully magnetized “high signal” spins as a result of flow (Fig. 7 1A). The second involves some form of tagging and then imaging to follow the motion of the tagged mate rial (see Fig. 7 1B). Singer and Crooks11 adopted the first approach in an attempt to measure flow in the internal jugular veins, although quantification was questionable because of other factors affecting the flow signal. The first report to describe a tagged time of flight approach was by Feinberg and colleagues.10 Their method involved a variation on a dual echo spin echo sequence; the first 180
selected slice was displaced by 3 mm from the initial excitation slice, and the second was displaced by 9 mm. The first 180
selection overlapped sufficiently with the 90
selection to produce a good anatomic image. The second 180
pulse selection did not overlap with the 90
or the first 180
pulse selection and therefore produced no anatomic image,

but gave high signal from the blood that had experienced all of the preceding radiofrequency (RF) pulses (i.e., that had passed between the different selected planes). The technique was used to identify flow in the carotid and vertebral arteries of a volunteer’s neck, although flow velocities had to be within a specific range and could not be defined accurately. Methods have also been described in which the time of flight flow movement can be visualized directly on an image.12,13 The methods involved the application of slice selection and frequency encoding in the same axis. In this way, material that had moved in this axis between selection and readout would be displaced relative to the stationary material. These techniques were therefore making use of signal misregistration, an effect that is often seen as a problem in other methods of flow imaging. Another time of flight approach is to saturate a band of tissue, for example, in a transverse plane, and then to follow the progress of this dark band in the coronal or sagittal plane.14 The major limitation of these saturation methods is that they are limited by the T1 of the various tissues being saturated. The contrast of the saturated blood will decrease with time, eventually making it difficult to measure accurately the distances traveled. Also, motion during the sampling gradients results in signal and thus image distortion.15 In addition, for arterial flow measurements in which cardiac gating is required, only two dimensional (2D) images can be acquired in a reasonable time so that only limited details of the flow profile can be studied.

PHASE FLOW IMAGING METHODS Considerable knowledge had been gained on the measure ment of flow from the phase of the CMR signal from the nonimaging studies after an original suggestion by Hahn in 1960 of a method of measuring the slow flow of currents in the sea.16 In 1984, the first attempts of measuring blood flow were described by van Dijk8 and Bryant and associ ates,9 using methods based on the theory suggested by Moran 2 years earlier.17 The imaging methods that followed fell broadly into two categories: 1. Phase contrast velocity mapping methods that mapped the phase of the signal directly to measure the flow. 2. Fourier flow imaging methods that phase encoded flow velocity to produce an image after Fourier transfor mation with velocity resolved on one image axis. Cardiovascular Magnetic Resonance 91

7 BLOOD FLOW VELOCITY ASSESSMENT

CHAPTER 7

Signal intensity

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

location will take up a frequency shift that depends on their position in the direction of the field gradient. When the gradient is turned off, the phase of the flowing and stationary materials can be considered equal. In the period between the positive and negative gradients, the flowing material moves away from its stationary neighbor. During the period of the negative gradient, the stationary material takes up an equal but opposite frequency shift and returns to the phase it had before the first gradient. However, the flowing fluid takes up a different frequency shift that is dependent on the distance it has moved; its final phase, therefore, also depends on this distance and hence its velocity. The relationship between the phase of the signal and flow velocity is:

A

Flow rate

f ¼ gvDAg

B

Time 1 (tagging)

Time 2 (detection)

Figure 7 1 Time of flight approaches to magnetic resonance flow measurement. In the saturation wash in method (A), the signal is enhanced by an amount related to the in flow of fully magnetized blood. In the true time of flight method (B), there is a time of spatial labeling or tagging, followed after a defined period by detection of the movement of the labeled blood.

Both of these methods rely on the same principles that cause flowing material to attain a phase shift that is related to its motion. Figure 7 2 shows these principles for a fluid flowing down a tube surrounded by stationary material. A bipolar gradient pulse is applied, consisting of a positive magnetic field gradient, followed a certain time later by an equal but opposite negative magnetic field gradient in the direction of the flow. During the period of the positive gradient, flowing and stationary materials in a particular

Gradient waveform

Time

Time

Magnetic field

Position

Position

Flow

Flow Signal phase Stationary Flowing material material

Stationary material

Time 1

Flowing material

Time 2

Figure 7 2 Principles of phase velocity encoding. At time 1, a positive magnetic field gradient is applied that results in an equal frequency and associated phase shift for neighboring stationary and flowing spins. At time 2, an equal but opposite magnetic field is applied. By this time, the spins in the flowing blood have separated from their original neighbors to be in a different strength of magnetic field during the gradient application. The result is that although the phase of the stationary spins will be returned to zero, the flowing spins will accumulate a phase shift proportional to the distance moved and hence the velocity. 92 Cardiovascular Magnetic Resonance

(1)

where Ag is the area of one gradient pulse (amplitude  duration), D is the time between the centers of the two gra dient pulses, v is the velocity, and g is the gyromagnetic ratio. Ag, D and g are all constants for a particular imaging sequence so that a quantitative measure of velocity can be determined if the phase shift can be measured. The two principal approaches of using the phase shift to produce a quantitative flow image, phase contrast velocity mapping and Fourier flow imaging, are discussed in this chapter.

Phase Contrast Velocity Mapping The early phase contrast velocity mapping methods8,9 used a spin echo sequence, which was not ideal because of problems in repeating the sequence rapidly, and signal loss as a result of shear and other more complex flows. These problems were reduced and the methods were made clinically more useful, partly by the use of a gradient echo sequence18,19 and, more importantly, by the introduction of velocity compensated gradient waveforms.20,21 Normally, two images are acquired with different gradi ent waveforms in the direction of desired flow measure ment. The difference in the waveforms is calculated to produce a well defined velocity related phase difference between the two images. A phase reconstruction is pro duced for each of the images, which are then subtracted pixel by pixel to produce the final velocity map. This pro cess of subtraction removes any phase variations that are not related to flow. The velocity phase sensitivity of the final image is normally set such that the expected veloc ity related phase shifts are within the range of  p radians. If a larger range of velocities is present, then aliasing or wrap around will occur, resulting in measurement of ambig uous velocities. This problem can be avoided by reducing the velocity sensitivity or potentially can be corrected by a process known as phase unwrapping (described later). Figure 7 3A shows the phase velocity images of a series of time frames from a slice just above the heart. Flow can be seen in the ascending and descending aortas, the pulmonary artery, and the superior vena cava. Flow versus time curves throughout the cardiac cycle are shown in Figure 7 3B. The stroke volume can be measured by

250 msec

40

MPA

AA MPA DA SVC

35 30

DA

300 msec

350 msec

400 msec

Flow (L/min)

SVC

25 20 15 10 5 0 –5

A

B

0

200

400

600

800

Time (msec)

Figure 7 3 A, Flow velocity images of a transverse slice at the level of the right pulmonary artery showing head/foot flow velocities at six times in the cardiac cycle. B, Plot of measured volume flow versus time for the ascending aorta (AA), descending aorta (DA), main pulmonary artery (MPA), and superior vena cava (SVC).

integrating under the aortic or pulmonary flow curve. The technique has been validated both in vitro and in vivo,22 and is now routinely used to provide useful measurements in clinical and physiologic flow studies.23

Fourier Flow Imaging Fourier flow imaging normally involves the addition of a bipolar velocity phase encoding gradient that is stepped through a range of defined amplitudes (Fig. 7 4). Because this increases the scan time by a multiple of the number of steps, it is often applied as a replacement of one of the spatial phase encoding gradients. The image is normally reconstructed and displayed with velocity information in one dimension. Stationary material is positioned in the center of the image, with faster velocities toward the edge. The method was first described by Redpath and col leagues24 in 1984; in this case, eight velocity phase encoding steps were added to a 2D imaging sequence, to image velocities in a circle of fluid filled tubing that rotated in the image plane. Different segments of the circle, each corresponding to different velocity ranges, were seen on the eight resultant images. A year later, Feinberg and coworkers25 applied the method, both in vitro and

SPATIAL PHASE ENCODING IS REPLACED BY VELOCITY PHASE ENCODING

Spatial phase encoding gradient waveform

Velocity phase encoding gradient waveform

Figure 7 4 Gradient waveforms required for spatial and velocity phase encoding.

in vivo, and increased the velocity range and resolution by increasing the number of flow phase encoding steps. In this case, however, to maintain a tolerable scan time, only one spatial phase encoding direction was used so that there was only spatial resolution in one direction. The accuracy of the method was shown with a phantom, whereas the in vivo study, which showed the flow in the descending aorta, highlighted the problem of very high zero velocity signal from the large amount of stationary tissue imaged. In 1988, Hennig and colleagues.26 described a development of this method in which the signal from stationary tissue was saturated and the sequence was repeated much more rapidly. The issue of time precluding the use of spatial phase encoding remains a problem, although 2D radiofrequency (RF) pulses have been used successfully to locate signals within a column.27,28 More recently, Luk Pat and associates.29 showed a method of real time Fourier velocity imaging that used an excited col umn to localize the signals. In vivo aortic flow waveforms were presented with a temporal resolution of only 33 msec.

IMPROVING THE ACCURACY OF PHASE CONTRAST VELOCITY MEASUREMENTS The vast majority of MR flow imaging applications have used the method of phase contrast velocity mapping. The accuracy of this method is highly dependent on such factors as flow pulsatility, velocity, and size and tortuosity of the vessel. One simple approach to improving the over all accuracy of the method is to adjust the velocity sensi tivity of the sequence so that the velocity related phase shift is close to 2p for the maximum expected velocity. Buonocore30 extended this approach by varying the velocity sensitivity during the cardiac cycle, based on the knowledge that the arterial flow velocity is high in systole but low in diastole. The accuracy can be improved further by allowing a velocity related phase shift greater Cardiovascular Magnetic Resonance 93

7 BLOOD FLOW VELOCITY ASSESSMENT

45 150 msec

AA

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Figure 7 5 Method of phase unwrapping. A, Systolic image in which high velocities result in aliasing in both the positive and negative directions. B and C, Adjustment of the velocity window to remove aliasing in the positive and negative directions, respectively. D, The same image data after processing by the anti aliasing algorithm.

A

B

C

D

than 2p that will result in aliasing that can be corrected by use of a phase unwrapping algorithm (Fig. 7 5).31 Another approach to improving accuracy has been sug gested by Bittoun and associates.32 The method is a combina tion of phase contrast velocity mapping and Fourier velocity imaging with a small number of velocity phase encoding steps. The final phase contrast velocity map is calculated from the best fit through the Fourier velocity encoded result. One potential problem with this method that can also affect conventional phase contrast velocity imaging occurs if there is any beat to beat variation in flow velocity. As a result of the high velocity phase sensitivity used with this method, significant phase variations can occur, with resulting ghosting artifacts and loss of flow information. Another method that was originally described to give a measure of velocity and flow quantification was phase contrast angiography. This technique again involves acquiring two image datasets with sequences that have opposite phase velocity sensitivities, although in this case, the raw data are subtracted before reconstruction. This technique has the advantage of subtracting out signal from stationary tissue, which removes errors caused by partial volumes where voxels contained a mixture of flowing and stationary tissue. The method is, however, generally less accurate because the signal and hence the velocity measurement can be affected by factors such as in flow enhancement and intravoxel dephasing (signal loss). In the mid 1990s, Polzin and colleagues.33 suggested a method of combining this method with phase contrast velocity mapping, which they showed to be more accurate in a number of phantom studies. The methods are yet to be fully validated in vivo, however, and are likely to be affected by problems of signal loss and motion, particu larly when imaging is performed on small mobile vessels, such as the coronary arteries.34 94 Cardiovascular Magnetic Resonance

One of the most significant factors that can affect the accuracy of the flow measurement methods is flow related signal loss. This is normally the result of loss of phase coherence within a voxel, and eventually it results in an inability to detect the encoded phase of the flow signal above the random phase of the background noise. Even if a velocity compensated imaging sequence is used, the acceleration and even the higher orders of motion present in complex flows can result in loss of phase coherence. Figure 7 6 shows an example of a long axis image of a patient with a mitral valve stenosis in which the valve is also regurgitant. In this case, a region of blood signal is lost from the ventricle during diastole as a result of the stenosis generating complex flows and from the atrium during systole because of a regurgitant jet of flow through the valve. Partial signal loss, however, does not greatly affect the accuracy of the phase contrast velocity mapping measurement unless it is accompanied by partial volume errors. When signal loss is the result of a spread of phase within a voxel, the mean phase will be detected, although this will be affected by differential saturation effects.35 The phase contrast velocity mapping techniques are most susceptible to signal loss of one form or another, although this can normally be minimized by appropriate gradient profile design. A good way to reduce signal loss is to use a symmetrical gradient waveform that nullifies phase shifts caused by all of the odd order derivatives of position and then to shorten the sequence as much as possible to reduce the effects of the even order derivatives.36 Signal loss of the type described is much less of a problem with the Fourier flow imaging method. In this case, the Fourier transform is used to separate out constituent velocities. Errors can be caused by phase differences for reasons other than the velocity encoding pulses resulting in the phase map velocity values being offset from zero, even for stationary

tissues.37 This background phase offset generally varies grad ually with position across the image, and it also varies with image plane orientation and other sequence parameters that affect the gradient waveforms, such as Venc. Distortion of the requested magnetic field gradients is unavoidable because of the fundamental laws of electromagnetism, and these are known in MR imaging as Maxwell, or concomitant, gradients. These background phase shifts become more significant when high amplitude gradients are used and also when imaging is performed at lower main magnetic field strengths. However, with the latest gradient systems, Maxwell gradient effects are certainly a factor for phase velocity mapping at 1.5 Tesla. However, these velocity map offsets can be cor rected precisely and automatically in software, with no user intervention required.38 A second common reason for back ground phase shifts is the presence of small uncorrected side effects of the gradient pulses in the magnet, known as eddy currents. These phase shifts became more of a problem with the advent of higher performance gradients, although more recently, the problem appears to have been reduced. Software is sometimes provided that allows the user to place markers identifying stationary tissues so that this background phase error can be calculated and removed from the entire image. Sometimes, however, if the phase shifts are nonlinear or if there is little signal from stationary tissue, then the only way to correct for them is to acquire an additional set of images of a large static phantom using the same sequence parameters as those used in vivo, and then to subtract out the phase errors on a pixel by pixel basis.39 Pixels without any signal in velocity maps have a ran dom phase or show the phase of a weak ghost that may not be visible on the magnitude image with normal bright ness settings. Particularly for poststenotic jet images, it is important to check that the magnitude image pixels of the jet are not affected by signal loss. Avoiding the inclusion of noise pixels can be problematic in regions of interest around the great vessels, and ideally, the image analysis software enables the user to set the velocity to zero for pix els whose magnitude is below a user defined threshold. Provided that the sequence parameters are carefully chosen such that the potential errors and artifacts discussed earlier can be minimized or avoided, phase velocity mapping has been shown to be accurate and reproducible. Validation has been reported in phantoms by comparison with true measured flow and with Doppler36,40,41 in animal models by comparison with in vivo flow meter measurements.42

7 BLOOD FLOW VELOCITY ASSESSMENT

Figure 7 6 Systolic and diastolic long axis frames from cine datasets acquired in a patient with a stenotic and regurgitant mitral valve. Signal loss can be seen in the left atrium (LA) during systole and the left ventricle (LV) during diastole.

Validation has also been reported in humans by comparison with methods such as Doppler ultrasound or catheteriza tion.43–45 Perhaps the most convincing forms of validation were those performed initially by comparing the aortic flow with the left ventricular stroke volume and later flow in both the aorta and pulmonary artery with left and right ventricu lar stroke volumes in normal subjects.22,46,47 For the latter, the four measurements should be the same except for small differences caused by coronary and bronchial flow, and it can be calculated that flow measurements in large vessels are accurate to within 6%. The effect of breath holding on flow measurement is another factor that could affect more recent studies. Sakuma and associates showed a significant change in both pulmonary and aortic cardiac output during a large lung volume breath hold.48 Conversely, flows measured during a small volume breath hold were found to be similar to those measured during normal breathing. There are other potential sources of error that have been reported. These include misalignment of the vessel with the direction of velocity encoding and misregistration of flow signal caused by flow between excitation and readout. Because of these and the other sources of errors discussed earlier, care is required when setting up the scan para meters, to minimize their effect.

Rapid Phase Flow Imaging Methods With the very rapid scanning hardware available today, it is possible to repeat a phase contrast velocity sequence so fast that low resolution images can be acquired in 100 msec or high resolution images can be acquired in a breath hold. The major problem is for pulsatile flow where the accuracy of the measurements and the temporal resolution can be limited if the acquisition period per cardiac cycle is too long. Also, if high spatial resolution is required, the cardiac motion of structures, such as coronary arteries, can cause blurring, with subsequent errors in flow measurement.49 For this reason, it is likely that more efficient k space cover age methods, such as interleaved spirals, will be important. Ultra fast flow imaging techniques have also been devel oped, either by combining a phase mapping type approach with imaging methods, such as single shot echo planar and Cardiovascular Magnetic Resonance 95

Visualizing Flow and Flow Parameters The method used to visualize MR flow data has depended on the method used for acquisition. For Fourier velocity measurement, each voxel may contain a range of measured velocities, and the Fourier velocity image normally takes the form of a plot of velocity versus time or velocity versus position in one direction. Figure 7 8 shows an example in

Magnitude Velocity

600

Aortic flow

400

Diaphragm position

20 Expiration

10 0

200

(mm)

(mm)/sec

–10

SVC flow

Inspiration

–20 –30

0 1

4

7

10

13

16

19

22

Cardiac cycles

Figure 7 7 Real time magnitude and flow images from the excited region containing the aorta and the superior vena cava. The plot shows the variation in flow during a period of respiratory maneuver. 96 Cardiovascular Magnetic Resonance

ECG Gate Delay 100 msec 105 msec 115 msec 125 msec 135 msec 145 msec 155 msec 165 msec 175 msec

250 mm FOV

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

spiral imaging,50,51 or by imaging only one spatial dimen sion.52 A compromise generally has to be made in temporal or spatial resolution and probably also in the signal to noise ratio. However, taking into account these constraints, the methods have generally been shown to be accurate. One complication with the echo planar sequence is flow signal loss because of its inherent phase sensitivity, even when additional flow compensation is applied. However, this has been used to advantage for more qualitative flow imaging showing flow disturbances, for example.53 The one dimensional rapid acquisition mode, real time acquisition and velocity evaluation (RACE),52 can be used to measure flow perpendicular to the slice. The technique can be repeated rapidly throughout the cardiac cycle to give near real time flow information. One problem with this type of approach is that data are acquired from a projection through the patient; this means that any signal overlapping with the flow signal will combine and introduce errors to flow measurement. Several strategies have been suggested for localizing the signal to avoid this: they include spatial presaturation, projection dephasing (applying a gradient to suppress stationary tissue), and collecting a cylinder of data and multiple oblique measurements. Yang and colleagues used a 2D RF excitation scheme to excite a narrow rectangular X section column and used only 16 echoes to spatially resolve the other dimension in high resolution.54 This approach allowed real time flow measurements to be acquired. The authors used the method to show the effect of controlled breathing on flow in the ascending aorta and superior vena cava (Fig. 7 7).

0

168 cm/sec

Figure 7 8 Series of nine Fourier velocity images at 10 msec intervals showing the velocity pulse wave propagating down the descending aorta. ECG, electrocardiogram.

which velocity images were acquired from a column of excited tissue, including the descending aorta.55 The front edge of the aortic pulse wave can be seen on successive frames as it travels down the vessel. Phase contrast velocity images contain only one velocity measure per image voxel. Historically, these have been shown with a grayscale such that flow in one direction tends toward white, flow in the other direction tends toward black, and stationary material is mid gray, as shown previously in Figure 7 3A. When flow is measured in more than one direction, then more sophisticated methods of display can be used. Figure 7 9 shows a vector map of flow in the root of the aorta of a patient with an atherosclerotic aneurysm. This systolic image, shown alongside a pressure map (described later), shows high velocity flow impinging on the wall of the aneurysm. An alternative type of repre sentation would be to use the cine velocity images to calcu late the path of a seed over time.56 The ability to study flow in such detail and at any site in the body is unique to MR imaging. For this reason, a large amount of interest is being generated from those who wish to understand the physiology of blood flow and its interaction with blood vessels and the cardiac chambers. Despite the relatively poor spatial resolution, a number of groups have studied methods of extracting a measure of the wall shear stress from the MR images. Both Oshinski and colleagues and Oyre and associates developed fitting methods to derive the velocity profile at subpixel distances from the vessel wall.57,58 Both groups presented expected values of stress, although it is difficult to suggest a method of validating the accuracy of these measurements. Frayne and Rutt suggested an alternative approach that potentially gave more informa tion about the flow within a voxel that straddled the vessel wall.59 Their method used Fourier velocity encoding to distinguish the distribution of flow velocities, so that only the spatial location had to be considered. There has also been considerable interest in the possibility of deriving pressure measurements from MR images. Urchuk and coworkers considered vessel compliance and the flow

66 (msec)

A

162 (msec)

B

Pressure

7 BLOOD FLOW VELOCITY ASSESSMENT

Figure 7 9 Flow vector map showing the systolic flow pattern in the aortic root of a patient with an atherosclerotic aneurysm. The associated image shows the corresponding pressure distribution.

Flow Pattern

258 (msec)

C

354 (msec)

D

450 (msec)

E

Figure 7 10 Selected frames from a cine series of calculated flow pressure maps showing the variation in pressure at different times in the cardiac cycle (A to E). Each grayscale band from white to black represents a pressure gradient of 1 mm Hg. A positive pressure gradient during systole (A) reverses during the deceleration phase of diastole (D).

pulse wave to calculate the pressure waveform and showed a good correlation with catheter pressure measurements made in a pig model.60 In contrast, Yang and colleagues derived flow pressure maps from the cine phase contrast velocity maps using the Navier Stokes equations.61 Figure 7 10 shows an example of the changing flow pressure around the aortic arch during the first half of the cardiac cycle. Figure 7 11 shows an interesting example of a flow pressure map, showing the descending aorta in a patient who underwent polyester fiber (Dacron, Dupont, Wilmington, DE) graft repair of aortic coarctation. In contrast to the rest of the aorta, no pressure gradient can be seen in the repaired region, pos sibly because of the reduced compliance. To fully understand, visualize, and measure flow para meters in blood vessels, a method of acquisition is required that measures velocity in three dimensions and three direc tions over time, with high spatial and temporal resolution. Even with the fastest gradient systems available today, this pre sents a significant problem in terms of acquisition time and data handling. A method has been suggested, however, using a combination of echo planar and k space view sharing that suggests that the acquisition times can be reduced to accept able levels.62 These types of methods are now being used to visualize and study flow patterns in much more detail.63,64

Figure 7 11 Systolic pressure map of the aortic arch in a patient with a polyester fiber (Dacron, Dupont, Wilmington, DE) repair. No pressure gradient is seen in the region of the repair. Cardiovascular Magnetic Resonance 97

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

References 1. Singer JR. NMR diffusion and flow measurement and an introduction to spin phase graphing. J Phys E: Sci Instrum. 1978;11:281–291. 2. Jones DW, Child TF. NMR in flowing systems. Adv Magn Reson. 1976;8:123–148. 3. Bradley WG. Flow phenomenon in MR imaging. AJR. 1988;150: 983–994. 4. Alfidi RJ, Masaryk TJ, Haacke EM, et al. MR angiography of peripheral, carotid, and coronary arteries. AJR. 1987;149:1097. 5. Firmin DN, Dumoulin C, Mohiaddin RH. Quantiative MR flow measurement. In: Haacke EM, Potchen EJ, Gottschalk A, Siebert JE, eds. Magnetic Resonance Angiography: Concepts and Applications. St Louis: Mosby; 1993, pp. 187–219. 6. Crooks LE, Kaufman L. NMR imaging of blood flow. Brit Med Bul. 1984;40:167–169. 7. Axel L. Blood flow effects in magnetic resonance imaging. AJR. 1984;143:1157–1166. 8. van Dijk P. Direct cardiac NMR imaging of heart wall and blood flow velocity. J Comput Assist Tomogr. 1984;8:429–436. 9. Bryant DJ, Payne JA, Firmin DN, Longmore DB. Measurement of flow with NMR imaging using a gradient pulse and phase difference technique. J Comput Assist Tomogr. 1984;8:588–593. 10. Feinberg DA, Crooks LE, Hoenninger J, et al. Pulsatile blood velocity in human arteries displayed by magnetic resonance imaging. Radiology. 1984;153:177–180. 11. Singer JR, Crooks LE. Nuclear magnetic resonance blood flow measurements in the human brain. Science. 1983;221:654–656. 12. Shimizu K, Matsuda T, Sakurai T, et al. Visualisation of moving fluid: quantitative analysis of blood flow velocity using MR imaging. Radiology. 1986;159:195–199. 13. Axel L, Shimakawa A, MacFall J. A time-of-flight method of measuring flow velocity by magnetic resonance imaging. Magn Reson Imaging. 1986;4:199–205. 14. Edelman RR, Mattle HP, Kleefield J, Silver MS. Quantification of blood flow with dynamic MR imaging and presaturation bolus tracking. Radiology. 1989;171:551–556. 15. Izen SH, Haacke EM. Measuring non-constant flow in magnetic resonance imaging. IEEE Trans Med Imaging. 1990;9:450–460. 16. Hahn EL. Detection of sea-water motion by nuclear precession. Geophys Res. 1960;65:776–777. 17. Moran PR. A flow zeugmatographic interlace for NMR imaging in humans. Magn Reson Imaging. 1982;1:197–203. 18. Young IR, Bydder GM, Payne JA. Flow measurement by the development of phase differences during slice formation in MR imaging. Magn Reson Med. 1986;3:175–179. 19. Ridgway JP, Smith MA. A technique for velocity imaging using magnetic resonance imaging. Brit J Radiol. 1986;59:603–607. 20. Nayler GL, Firmin DN, Longmore DB. Blood flow imaging by cine magnetic resonance. J Comput Assist Tomogr. 1986;10:715–722. 21. Haacke EM, Lenz GW. Improving MR image quality in the presence of motion by using rephasing gradients. AJR. 1987;148:1251–1258. 22. Firmin DN, Nayler GL, Klipstein RH, et al. In vivo validation of MR velocity imaging. J Comput Assist Tomogr. 1987;11:751–756. 23. Mohiaddin RH, Pennell DJ. MR Blood flow measurement: clinical application in the heart and circulation. Cardiol Clin. 1998;16: 161–187. 24. Redpath TW, Norris DG, Jones RA, Hutchinson MS. A new method of NMR flow imaging. Phys Med Biol. 1984;29:891–895. 25. Feinberg DA, Crooks LE, Sheldon P, et al. Magnetic resonance imaging and velocity vector components of fluid flow. Magn Reson Med. 1985;2:555–566. 26. Hennig J, Mueri M, Brunner P, Friedburg H. Quantitative flow measurement with the fast Fourier flow technique. Radiology. 1988; 166:237–240. 27. Gatehouse PD, Link K, Bebbington MWP, et al. Pulse-wave and stenosis studies by cylinder excitation with Fourier velocity encoding [Abstract]. In: Proceedings of the Third Annual Meeting of the International Society of Magnetic Resonance. 1995;318. 28. Hardy CJ, Bolster Jr BD, McVeigh ER, et al. Pencil excitation with interleaved Fourier velocity encoding: NMR measurement of aortic distensibility. Magn Reson Med. 1996;35:814–819. 29. Luk Pat GT, Pauly JM, Hu BS, Nishimura DG. One-shot spatially resolved velocity imaging. Magn Reson Med. 1998;40:603–613. 30. Buonocore MH. Blood flow measurement using variable velocity encoding in the RR interval. Magn Reson Med. 1993;29:790–795.

98 Cardiovascular Magnetic Resonance

31. Yang GZ, Burger P, Kilner PJ, et al. Dynamic range extension of cine velocity measurements using motion registered spatio-temporal phase unwrapping. J Magn Reson Imag. 1996;6:495–502. 32. Bittoun J, Bourroul E, Jolivet O, et al. High-precision MR velocity mapping by 3D-Fourier phase encoding with a small number of encoding steps. Magn Reson Med. 1993;29:674–680. 33. Polzin JA, Alley MT, Korosec FR, et al. A complex-difference phasecontrast technique for measurement of volume flow rates. J Magn Reson Imaging. 1995;5:129–137. 34. Frayne R, Polzin JA, Mazaheri Y, Grist TM, Mistretta CA. Effect of and correction for in-plane myocardial motion on estimates of coronaryvolume flow rates. J Magn Reson Imaging. 1997;7:815–828. 35. Polzin JA, Korosec FR, Wedding KL, et al. Effects of through-plane myocardial motion on phase-difference and complex-difference measurements of absolute coronary artery flow. J Magn Reson Imaging. 1996;6:113–123. 36. Firmin DN, Nayler GL, Kilner PJ, Longmore DB. The application of phase shifts in NMR for flow measurement. Magn Reson Med. 1990;14:230–241. 37. Kilner PJ, Gatehouse PD, Firmin DN. Flow measurement by magnetic resonance: a unique asset worth optimising. J Cardiovasc Magn Reson. 2007;9:723–728. 38. Bernstein MA, Zhou XJ, Polzin JA, et al. Concomitant gradient terms in phase contrast MR: analysis and correction. Magn Reson Med. 1998;39:300–308. 39. Chernobelsky A, Shubayev O, Comeau CR, Wolff SD. Baseline correction of phase contrast images improves quantification of blood flow in the great vessels. J Cardiovasc Magn Reson. 2007;9:681–685. 40. Kilner PJ, Firmin DN, Mohiaddin RH, Underwood SR, Rees RSO, Longmore DB. Valve and great vessel stenosis: assessment with MR jet velocity mapping. Radiology. 1991;178:229–235. 41. Meier D, Maier S, Boesiger P. Quantitative flow measurements on phantoms and on blood vessels with MR. Magn Reson Med. 1988;8: 25–34. 42. Pettigrew RI, Dannels W, Galloway JR, et al. Quantitative phase-flow imaging in dogs by using standard sequences: comparison with in vivo-flow meter measurements. Am J Roentgenol. 1987;148: 411–414. 43. Kilner PJ, Manzara CC, Mohiaddin RH, et al. Magnetic resonance jet velocity mapping in mitral and aortic valve stenosis. Circulation. 1993;87:1239–1248. 44. Maier SE, Meier D, Boesiger P, Moser UT, Vieli A. Human abdominal aorta: comparative measurements of blood flow with MR imaging and multigated Doppler US. Radiology. 1989;171:487–492. 45. Van Rossum A, Sprenger KH, Peels FC, et al. In vivo validation of quantitative flow imaging in arteries and veins using magnetic resonance phase shift techniques. Eur Heart J. 1991;12:117–126. 46. Bogren HG, Klipstein RH, Firmin DN, et al. Quantitation of antegrade and retrograde blood flow in the human aorta by magnetic resonance. Am Heart J. 1989;117:1214–1222. 47. Bogren HG, Klipstein RH, Mohiaddin RH, et al. Pulmonary artery distensibility and blood flow patterns: a magnetic resonance study of normal subjects and of patients with pulmonary arterial hypertension. Am Heart J. 1989;118:990–999. 48. Sakuma H, Kawada N, Kubo H, et al. Effect of breath holding on blood flow measurement using fast velocity encoded cine MRI. Magn Reson Med. 2001;45:346–348. 49. Hofman MB, van Rossum AC, Sprenger M, Westerhof N. Assessment of flow in the right human coronary artery by magnetic resonance phase contrast velocity measurement: effects of cardiac and respiratory motion. Magn Reson Med. 1996;35:521–531. 50. Firmin DN, Klipstein RH, Hounsfield GL, et al. Echo-planar high-resolution flow velocity mapping. Mag Reson Med. 1989; 12:316–327. 51. Gatehouse PD, Firmin DN, Collins S, Longmore DB. Real time blood flow imaging by spiral scann phase velocity mapping. Magn Reson Med. 1994;31:504. 52. Mueller E, Laub G, Grauman R, Loeffler W. RACE—Real time Acquisition and Evaluation of pulsatile blood flow on a whole body MRI unit [Abstract]. In: Proceedings of the Seventh Annual Meeting of the International Society of Magnetic Resonance in Medicine. 1988;729. 53. Kose K. One shot velocity mapping using multiple spin-echo EPI and its application to turbulent flow. J Magn Reson. 1991;92:631–635.

59. Frayne R, Rutt BK. Measurement of fluid-shear rate by Fourierencoded velocity imaging. Magn Reson Med. 1995;34:378–387. 60. Urchuk SN, Fremes SE, Plewes DB. In vivo validation of MR pulse pressure measurement in an aortic flow model: preliminary results. Magn Reson Med. 1997;38:215–223. 61. Yang GZ, Kilner PJ, Wood NB, Underwood SR, Firmin DN. Computation of flow pressure fields from magnetic resonance velocity mapping. Magn Reson Med. 1996;36:520–526. 62. Firmin DN, Gatehouse PD, Yang GZ, Jhooti P, Keegan J. A 7-Dimensional echo-planar flow imaging technique using a novel k-space sampling scheme with velocity compensation [Abstract]. In: Proceedings of the Fifth Annual Meeting of the International Society of Magnetic Resonance in Medicine. 1997;118. 63. Canstein C, Cachot P, Faust A, et al. 3D MR flow analysis in realistic rapid-prototyping model systems of the thoracic aorta: comparison with in vivo data and computational fluid dynamics in identical vessel geometries. Magn Reson Med. 2008;59:535–546. 64. Bolger AF, Heiberg E, Karlsson M, et al. Transit of blood flow through the human left ventricle mapped by cardiovascular magnetic resonance. J Cardiovasc Magn Reson. 2007;9:741–747.

Cardiovascular Magnetic Resonance 99

7 BLOOD FLOW VELOCITY ASSESSMENT

54. Yang GZ, Gatehouse PD, Mohiaddin RH, et al. Zonal echo-planar flow imaging with respiratory monitoring [Abstract]. In: Proceedings of the Fifth Annual Meeting of the International Society of Magnetic Resonance in Medicine. 1997;1885. 55. Gatehouse PD, Link K, Bebbington MWP, et al. Pulse-wave and stenosis studies by cylinder excitation with Fourier velocity encoding [Abstract]. In: Proceedings of the Third Annual Meeting of the International Society of Magnetic Resonance and the Twelth Annual Meeting of the European Society for Magnetic Resonance in Medicine and Biology. 1995;318. 56. Napel S, Lee DH, Frayne R, Rutt BK. Visualizing three-dimensional flow with simulated streamlines and three-dimensional phase-contrast MR imaging. J Magn Reson Imaging. 1992;2:143–153. 57. Oshinski JN, Ku DN, Mukundan Jr S, et al. Determination of wall shear stress in the aorta with the use of MR phase velocity mapping. J Magn Reson Imaging. 1995;5:640–647. 58. Oyre S, Ringgaard S, Kozerke S, et al. Accurate noninvasive quantitation of blood flow, cross-sectional lumen vessel area and wall shear stress by three-dimensional paraboloid modeling of magnetic resonance imaging velocity data. J Am Coll Cardiol. 1998;32:128–134.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

CHAPTER 8

Special Considerations for Cardiovascular Magnetic Resonance: Safety, Electrocardiographic Setup, Monitoring, and Contraindications Jeroen J. Bax and Ernst E. van der Wall

During the last decade, cardiovascular magnetic resonance (CMR) has developed into an important diagnostic clinical tool in cardiology. Not only the anatomy of the heart but also its function, metabolism, and perfusion, as well as the coronary arteries, can be evaluated with CMR. CMR offers some special advantages over other diagnostic imag ing methods. First, CMR does not use ionizing radiation. Second, the radiofrequency (RF) radiation penetrates bony structures without attenuation through relaxation para meters. Third, CMR gives additional diagnostic information about tissue characteristics. Finally, CMR provides three dimensional images or images of arbitrarily oriented slices. However, when performing CMR, particular precautions must be taken. Because CMR operates with high static and gradient magnetic fields, special safety regulations must be taken into account and certain contraindications must be considered. This chapter reviews the safety, electrocar diographic (ECG) setup, patient monitoring, and contrain dications to CMR; in particular, the issue of pacemakers and implantable cardiac defibrillators (ICDs) is addressed.

SAFETY OF CARDIOVASCULAR MAGNETIC RESONANCE General Issues CMR generally takes longer than other diagnostic modalities (although the time is significantly shortened with real time imaging1), and the confined space in which the patient is placed is rather narrow, which some patients find uncom fortable. During CMR, communication with the patient may be difficult because of interfering noise from the gradi ent coils. On the other hand, CMR is entirely noninvasive. Overall, the safety issues during CMR that may pose poten tial safety concerns2 can be summarized as follows:

100 Cardiovascular Magnetic Resonance

1. Biologic effects of the static magnetic field 2. Ferromagnetic attractive effects of the static magnetic field on certain devices 3. Potential effects on the relatively slowly time varying magnetic field gradients 4. Effects of the rapidly varying RF magnetic fields, includ ing RF power deposition concerns 5. Auditory considerations from noise from the gradients 6. Safety considerations concerning superconducting mag net systems 7. Psychological effects 8. Possible effects of the intravenous use of MR contrast agents 9. Patient safety during stress conditions The safety concerns inherent to these issues are discussed.

Biologic Effects Concerning the biologic effects of a static magnetic field, many structures in animals and humans are affected by magnetic fields. Many potential biologic effects and differ ent magnitudes of magnetic fields have been examined, including the effect of the field on cardiac contractility and function. Gulch and colleagues concluded that static magnetic fields used in CMR do not constitute any hazard in terms of cardiac contractility.3 These magnetic fields do not increase ventricular vulnerability, as assessed by the repetitive response threshold and the ventricular fibrilla tion threshold.4 In one of the investigations, however, car diac cycle length was shown to be altered.5 Numerous biologic effects on other systems have been investigated extensively, and it may be concluded that no deleterious biologic effects from static magnetic fields used in CMR have yet been established. However, as in all aspects of safety monitoring for patients, further research needs to be conducted.

The physical effect of the static magnetic field consists of a potential health hazard from the attractive effect on fer romagnetic objects. Ferromagnetic objects can be defined as those in which a strong intrinsic magnetic field can be induced when they are exposed to an external magnetic field. The existence of different kinds of scanners with different shielding makes the discussion about this topic even more crucial. When dealing with a static magnetic field, two types of physical concerns exist. First, there are concerns about forces exerted on ferro magnetic objects within, on, or distant from the patient. These forces result in rotational (torque) or translational (attractive) motion of the object. Within the human body, a ferromagnetic metallic structure may be sufficiently attracted, or have a sufficient amount of torque exerted, to create a hazardous situation. These factors should be carefully considered before subjecting a patient with a ferromagnetic implant or material to CMR, particularly if the device is located in a potentially dangerous area of the body, where movement or dislodgment of the device could injure the patient. Another potentially injurious effect is known as the projectile, or missile, effect. This refers to the fact that ferromagnetic objects have the potential to gain sufficient speed during attraction to the magnet that the accumulated kinetic energy could be injurious or even lethal if the object were to strike a patient. Numerous studies have been performed to assess the ferromagnetic qualities of various metallic implants and materials.6–10 The results indicate that patients with certain metallic implants or prostheses that are nonferromagnetic or are minimally deflected by static magnetic fields can safely undergo CMR. The literature on this topic has been exten sively reviewed and compiled.8 However, there are com mon misconceptions about what types of objects are ferromagnetic. The most important misconception is that stainless steel is ferromagnetic, when it is not. Patients with stainless steel implants can therefore be imaged safely, except for a small number of well described exceptions,

A

B

as discussed later. The implant will interfere locally with the images; for example, signal loss occurs around metallic prosthetic valves (Fig. 8 1) and sternal wires (Fig. 8 2) after bypass surgery, but this does not make the imaging hazard ous. Non stainless steel, which may be ferromagnetic, is not used for human implants, but is commonly used for oxygen cylinders, for example. Finally, batteries are typi cally attracted to the magnet, and this is one of the prob lems of imaging pacemakers. The second type of physical concern deals with magnet ically sensitive equipment, the functioning of which may be adversely affected by the magnetic field. The most common of these is the cardiac pacemaker. Most pacemakers include a reed relay switch whereby the sensing mechanism can be bypassed and excitation in the asynchronous mode can occur. This switch is activated when a magnet of sufficient strength is held over the pacemaker.11 In addition, the function of cardiac pacemakers may be influenced by field strengths as low as 17 gauss.11 In practice, reed switch clo sure can be expected in all pacemakers placed in the bore of the scanner. Pacemaker and ICD function is considered again later in this chapter.

Effect of Rapidly Switched Magnetic Fields CMR exposes the patient to rapid variations of magnetic fields by the transient application of magnetic gradients during imaging. The effect of rapidly switched magnetic fields may be the induction of currents within the body or any other electrical conductor, according to Faraday’s law. The current is dependent on the time rate of change of the magnetic field (dB/dt), the cross sectional area of the conducting tissue loop, and the conductivity of the tis sue. Biologic effects of induced currents can be caused either by power deposition by the induced currents (ther mal effects) or by direct effects of the current (nonthermal effects). Thermal effects as a result of switching gradients are not believed to be clinically significant.12–14 Possible

C

Figure 8 1 Cardiovascular magnetic resonance image of a patient with a Starr Edwards prosthetic valve in the mitral position. A, Horizontal long axis. B, Vertical long axis. C, Basal short axis. The dark artifact is obvious on the images, but does not interfere with assessment of ventricular function. These images were acquired at 0.5 T with a gradient echo cine sequence and an echo time of 14 msec. Shorter echo time gradient echo sequences and spin echo sequences typically show less artifact. The Starr Edwards valve causes the largest artifact because of the large amount of metal present in its construction. Other valves cause considerably less disturbance, especially the tissue valves. (Images courtesy of Dr. Dudley Pennell, Royal Brompton Hospital, London.) Cardiovascular Magnetic Resonance 101

8 SPECIAL CONSIDERATIONS FOR CARDIOVASCULAR MAGNETIC RESONANCE: SAFETY, ELECTROCARDIOGRAPHIC SETUP, MONITORING, AND CONTRAINDICATIONS

Ferromagnetism

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

A

B

C

Figure 8 2 Cardiovascular magnetic resonance image of a patient with sternal wires after bypass grafting. The artifact is clearly seen (arrows) on the horizontal long axis (A) and short axis (B) gradient echo cine images, and to a much lesser extent, on the transaxial spin echo image (C). (Images courtesy of Dr. Dudley Pennell, Royal Brompton Hospital, London.)

nonthermal effects include stimulation of nerve or muscle cells. The threshold currents for nerve stimulation and ven tricular fibrillation are known to be much higher than the estimated current densities induced under clinical CMR conditions. The echo planar imaging method, however, involves more rapidly changing magnetic field gradients, and peripheral muscle stimulation in humans has been reported.15 Such considerations have become more impor tant as new technology has allowed the introduction of commercially available ultrafast gradient switching systems, and guidelines for maximum magnetic field variation are under development.

Radiofrequency Time Varying Field The transmitted RF time varying field induces electrical cur rents within the tissue of the patient. The majority of this power is transformed into heat within the patient’s tissue as a result of ohmic heating. The time varying magnetic gra dients have the potential to cause either thermal or nonther mal biologic effects. The distinction between these two is a matter of frequency, waveform shape, and magnitude. The discussion about nonthermal effects from RF magnetic fields is controversial because of questions about the relationship between chronic exposure to electromagnetic fields over many years and the causation of cancer or developmental abnormalities. The most recent evidence suggests that prox imity to power lines is not injurious.16 Of course, acute exposure of a patient to short term RF fields for a diagnostic CMR examination is different from chronic exposure. The induced currents from RF magnetic fields are unable to cause nerve excitations. One of the difficulties faced by med icine is proving that a procedure is not injurious because of anecdotal case reports of adverse events and publication bias toward non neutral reports.17 This issue is also faced by such well established technology as ultrasound, where safety concerns have been raised over acoustic exposure.18 In contrast to the insignificant thermal effects caused by switched gradients, however, thermal effects as a result of 102 Cardiovascular Magnetic Resonance

RF pulses are of significant concern. The main biologic effects induced by RF fields are therefore related to the thermogenic qualities of the RF field. A general point of discussion is the appropriate safety regulations for levels of magnetic field strength in CMR imaging. Application of the fundamental law of electrostimulation is well estab lished, both on theoretical and experimental grounds. Application of this law, in combination with Maxwell’s law, yields an equation called the fundamental law of magne tostimulation, which has the hyperbolic form of a strength duration curve and allows an estimation of the lowest possible value of the magnetic flux density capable of stimulating nerves and muscles. Calculations have shown that the threshold for heart excitation is more than 200 times higher than for nerve and muscle stimulation, depending on pulse duration.19 However, in clinical prac tice, some precautions are necessary. First and most impor tantly, the specific absorption rate of the imaging sequence being operated is monitored by the scanner software and must be kept below limits set by such bodies as the U.S. Food and Drug Administration (FDA). Second, circum stances that could enhance the possibility of heating injury should be avoided. This includes ensuring the prevention of loops that could act as aerials within the scanner and enhance the heating effect locally. Therefore, patients should not be allowed to cross their legs (loop via the pel vis) or clasp their hands (loop via the shoulder and upper chest). The simple use of pillows prevents such problems. Other possible loops include the ECG leads, which should always be run out of the scanner parallel to the main field, and not looped across the chest. Finally, pacemaker and ICD leads make excellent aerials. In most cases, MR is con traindicated in patients with pacemakers and ICDs, although recent developments are promising (see the dis cussion of pacemakers and ICDs). The pacemaker lead can heat significantly during CMR and become a potential hazard (discussed later). Another consideration in patients after cardiac surgery is the effect of retained epicardial pacemaker leads. These leads can be left in place after sur gery, and they may therefore act as an antenna during CMR. Recent studies have suggested that such short retained epicardial wires do not pose a significant problem.20,21

Auditory Considerations During CMR, the gradient coils and adjacent conductors produce a repetitive sound because they act essentially as loudspeakers, with current being driven through them while they are in a magnetic field. Auditory considerations should therefore be taken into account when imaging a patient. The amplitude of this noise depends on factors such as the physical configuration of the magnet, the pulse sequence type, timing specifications of the pulse sequence, and the amount of current passing through these coils.24 In general, the amplitude of the generated noise from the clinical CMR scanners remains between 65 and 95 dB. However, there have been reported instances of temporary hearing impairment as a result of CMR. Magnet safe head phones or wax earplugs are readily available and have been shown to prevent hearing loss,25 and these are in common use. Systems combining sound attenuation with the facility to play music of the patient’s choice are also available. Research on the reduction of noise in MR scanners is ongo ing, and the use of anti noise is one area of interest.26

Superconducting System Issues Most superconducting CMR scanner systems use liquid helium. The helium maintains the magnet coils in their superconducting state. Helium achieves the gaseous state at approximately 269 C (4 K). If for any reason the tem perature within the cryostat rises, or in a system quench, the helium will enter the gaseous state. This means a marked increase in volume and thereby pressure within the cryostat. A pressure sensitive valve is designed to give way to the gaseous helium, which is always vented outside the CMR scanner room. However, it is possible that some helium gas is released into the imaging room should the system not work perfectly. Asphyxia and frostbite are potential hazards if a patient is exposed to helium vapor for a prolonged time, although there are no reports of such an occurrence in the medical community. For older scanners that still use a buffer of liquid nitrogen within the system (boils at 77 K), an oxygen monitor is recommended in the scanner room. Cryogenic dewars should be stored away from the scanner and in well ventilated areas.

Psychological Effects Claustrophobia or other psychological problems may be encountered in up to 10% of patients undergoing CMR,27 although on average, the incidence is closer to 2% to 4%, and this can be reduced further to a small number of intractably anxious patients by the use of explanation, reassurance, and where necessary, light seda tion with, for example, 2 to 5 mg intravenous (IV) diaze pam.28 In addition, the development of shorter magnets

as well as open designs is proving to be helpful. Such problems are related to a variety of factors, including the restrictive dimensions of the scanner, the duration of the examination, the noise, and the ambient conditions within the magnet bore.29 Fortunately, adverse psychological effects with CMR are usually transient. In a study reported by Weinreb and colleagues,30 based on the experience of 450 patients undergoing CMR and computed tomography examinations, it was clearly shown that patients often pre fer the CMR study, although CMR imaging took longer. Furthermore, the patient is placed into a confined space and there are difficulties in communicating with the patient during CMR scanning because of the noise from the gradient coils and the necessity of eliminating all extraneous RF sources from the examination room. To a certain extent, this can be avoided when the patient assumes a prone position in the scanner,31 facilitating communication with the outside surroundings. Simple maneuvers, such as using mirrors, also help in allowing the patient a clear view of the scanning room. Allowing the anxious patient to visit the scanner before the appoint ment gives the patient an opportunity to become familiar with the facility and staff.

SAFETY CONSIDERATIONS ASSOCIATED WITH GADOLINIUM-BASED CONTRAST AGENTS The safety profile of the contrast agents containing gadolin ium currently on the market is extremely good. Gadopente tate dimeglumine (Gd DTPA) is the best established, and its safety profile is well documented,32,33 but similar safety results have been shown with the other commercially avail able agents. The median lethal dose of Gd DTPA is roughly 10 mmol/kg, which is 100 times the diagnostic dose and shows the wide safety margin that the contrast agent enjoys. Patient tolerance of this drug is also high, and the prevalence of adverse reactions is approximately 2%. Among the reactions related to the IV administration of this drug are headache, nausea, vomiting, local burning or cool sensation, and hives. There have been reported incidents of anaphylactoid reactions associated with IV injection,34 although the frequency of this appears to be approximately 1 per 100,000 doses. The safety margins with these agents appear to be considerably better than with X ray contrast agents, although issues of nephrogenic septemic fibrosis are a concern in patients with severely impaired renal func tion. (See Chapter 6.) For CMR, these agents are used to increase contrast between blood and soft tissue for cine imaging for functional studies or angiography, to enhance cardiac tumors and cysts, to assess myocardial perfusion, and to examine for myocardial infiltration. In summary, FDA registered gadolinium complexes, such as Gd DTPA, can be safely used in patients with cardiac disorders. Multiple new CMR contrast agents are being developed and investigated. These are mainly gadolinium complexes, sometimes with novel binding molecules for special actions, but in addition, iron based compounds are being developed. Some of these agents are retained in the Cardiovascular Magnetic Resonance 103

8 SPECIAL CONSIDERATIONS FOR CARDIOVASCULAR MAGNETIC RESONANCE: SAFETY, ELECTROCARDIOGRAPHIC SETUP, MONITORING, AND CONTRAINDICATIONS

Finally, the use of ECG electrodes, which are essential for cardiac gating, must be considered. Metallic ECG electrodes may cause burns during CMR,22,23 but this risk can be reduced with the use of carbon fiber electrodes, and these have now become standard.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

vascular system and do not leak into the extravascular space. This suggests that they may have clinical utility for angiography, possibly in the coronaries, and for functional imaging.

PATIENT SAFETY DURING STRESS CONDITIONS A concern with CMR stress studies has been the ability to handle emergency situations. Patient monitoring during stress conditions is a critical issue because myocardial ischemia can be provoked in patients with coronary artery disease. Commercial equipment is available for noninvasive monitoring of blood pressure, heart rate, oxygen saturation, and other vital parameters in CMR scanners. The most cru cial difference compared with conventional exercise testing outside a magnetic field is the lack of a diagnostic ECG, in particular, at high levels of stress, precluding the proper assessment of stress induced ST segment changes. This holds for both conventional exercise using a specially adapted bicycle ergometer and pharmacologically induced stress. Under these circumstances, only heart rate can be monitored reliably. When performing pharmacologic stress CMR (e.g., with dipyridamole, adenosine, or dobutamine), an experienced physician should be present during the examination, and appropriate treatments for complications should be in direct proximity. Dipyridamole (half life, 30 minutes) and adenosine (half life, 10 seconds) are both vasodilators. Both agents have similar side effects, such as bronchospasm, hypotension, arrhythmias, and bradycar dia. In particular during adenosine infusion, atrioventricu lar heart block may develop in a small percentage of patients (0.7% to 2.8%), although this is usually asymp tomatic and self limited. When patients are symptomatic, the short physical half life of adenosine means that heart rhythm and symptoms can be restored very quickly by halting the infusion. As a suitable antagonist to both dipyr idamole and adenosine, aminophylline may be given slowly at an initial dose of 50 mg IV up to a maximum dose of 250 mg if necessary. In the case of persisting advanced heart block, 0.5 mg atropine IV should be administered up to a total dose of 3 mg. Dipyridamole and adenosine should not be administered to patients with asthma. Dobu tamine (half life, 2 minutes) is a beta agonist leading to an increase in cardiac inotropy (contractility) and chronotropy (heart rate). Common side effects are cardiac pounding and palpitations, and less commonly, arrhythmias, such as sup raventricular tachycardia and (nonsustained) ventricular tachycardia, are seen. Dobutamine can be safely adminis tered to patients with asthma.35 The actions of dobutamine can be counteracted by IV administration of a short acting beta blocking agent, such as esmolol. In the case of cardiac arrest or ventricular fibrillation, the recommendations should be followed according to published guidelines, such as those proposed by the European Resuscitation Council.36 In every CMR facility, an alarm system and a written flow chart should be visually available with the necessary instructions in case of emergency. It is necessary to be able to remove the patient from the examination room quickly (preferably within 20 seconds) to an area where emergency treatment can be performed safely, away from 104 Cardiovascular Magnetic Resonance

the hazards of the magnetic field. A nonferromagnetic stretcher stored in the scanner room or a detachable scan ner table is ideal. A cardiac arrest trolley must be main tained in close proximity to the scanner room, and all staff should undergo regular training in cardiopulmonary resuscitation techniques. Regular checks should be made of both the resuscitation equipment and the alarm system.

PATIENT MONITORING AND ELECTROCARDIOGRAPHIC SETUP Patient monitoring during CMR poses problems that will not be familiar to users of other technologies, such as echo cardiography. Ferrous metal, which is present in most mon itoring equipment, can distort the magnetic field, and such an item has the potential to become a projectile. In addi tion, monitoring wires that are attached to the patient, leav ing the scanner, and passing to another room may act as an antenna for stray RF signals. Electrical equipment in the scanner room also can act as a source of RF noise. All of these disturbances may result in image degradation. There fore, specific solutions to these problems have been designed. Commercially available CMR compatible monitoring equi pment, including that used to measure ECG, blood pres sure, and chest wall movements, as well as for general anesthesia, has been tested in several studies.37,38 Satisfactory monitoring can be obtained and images obtained during its use can be evaluated adequately.39 For some monitoring, simple solutions work, such as that reported by Roth and associates,39 who measured arterial blood pressure outside the CMR scanner by lengthening the rubber tubing con nected to a blood pressure cuff. The newest monitoring equipment eliminates the need for wires and tubes to leave the scanning room by using a microwave transmitter communicating with a slave display unit in the operating room. The CMR procedure depends on a high quality ECG sig nal for routine imaging, and each manufacturer has devel oped its own solution to the problems posed. Fiberoptic transmission of ECG signals for gating is now common place, and this significantly reduces RF pulse artifacts in the ECG. Felblinger and coworkers showed that this type of system could yield signals almost free from inter ference,40 during both conventional41 and high gradient activity sequences, such as during echo planar imaging.42 From this signal, the authors also developed a method for respiration monitoring during MR sequences. Third party ECG solutions are also now being incorporated into the lat est generation of scanners, and these come with specific ECG recommendations for lead placement. Carbon fiber electrodes are required to eliminate the risk of burning that has been reported with standard metallic ECG electrodes.22 Typical lead placement is the result of compromise. A better signal results from widely spaced electrodes, but this results in more artifact from the gradients. In general, therefore, the leads are kept relatively close together, and on the left side, which reduces the magnetohydrodynamic effect (the effect of systolic aortic flow causing surface potentials on the ECG that distort the ST segment; Fig. 8 3). A typical

Inside magnet

Figure 8 3 The magnetohydrodynamic effect. The top trace was recorded in a patient with atrial fibrillation outside of the magnet, and the bottom trace, inside. Note the distortion of the ST segment (black arrows) caused by added potentials arising from systolic flow in the aorta. (Courtesy of Dr. Dudley Pennell, Royal Brompton Hospital, London.)

lead placement that is commonly adopted is shown in Figure 8 4. Some centers have found ECG gating using elec trodes on the back to be successful, but this is not widely used. The ECG leads should not be allowed to form loops, which could present a burning hazard, and they should be braided together and brought out of the magnet while aligned parallel to the bore to reduce electrical interference. Keeping the electrical cables short is helpful, and fiberoptic conversion modules are therefore often very close to the patient’s chest. Switching between the ECG traces sometimes allows flexibility to reduce gating errors from tall T waves or electrical interfer ence. One thing is certain, however, and that is that time spent ensuring that the ECG is stable and working correctly at the start of the scan is time very well spent. An alternative technique to routine surface ECG record ing has been described by Fischer and colleagues using vectorcardiography, and this may prove to be a useful advance and has been widely implemented.43 The system examines the three dimensional orientation of the ECG V2

V3 V4

V1 V5 V3R

V4R V6

Figure 8 4 Typical electrode placement for cardiovascular magnetic resonance. The conventional chest leads (solid circles) are shown for comparison, and the positions of the four carbon electrocardiographic electrodes over the left chest are indicated (white circles). (Courtesy of Dr. Dudley Pennell, Royal Brompton Hospital, London.)

CONTRAINDICATIONS TO CARDIOVASCULAR MAGNETIC RESONANCE In general, there are potential hazards and artifacts of ferro magnetic and nonferromagnetic materials in CMR, for exam ple, with neurosurgical clips and ocular implants.46–48 The (relative) contraindications of these materials for CMR are dependent on factors such as the degree of ferromagne tism, the geometry of the material, the gradient (for force), and the field strength (for torque) of the imaging magnet and many other factors.

General Contraindications to Cardiovascular Magnetic Resonance There are a number of circumstances in which CMR is better avoided. This is because of reports of death or harm that have occurred. Prominent among these are patients with Cardiovascular Magnetic Resonance 105

8 SPECIAL CONSIDERATIONS FOR CARDIOVASCULAR MAGNETIC RESONANCE: SAFETY, ELECTROCARDIOGRAPHIC SETUP, MONITORING, AND CONTRAINDICATIONS

Outside magnet fast atrial fibrillation

signal and uses the calculated vector of the QRS complex as a filter mechanism to ignore electrical signals, which are of similar timing and similar magnitude in the cardiac cycle, but a different vector. The system as reported identi fied the QRS complex correctly in 100% of cases, with 0.2% false positive findings. In a subsequent study in normal individuals and patients with supraventricular extrasystoles, the same authors showed that vector cardio graphy based triggering provided nearly 100% triggering performance during CMR examinations.44 This system represents a significant improvement for CMR in stabilizing this important gating signal. Should interpretation prove impossible for technical rea sons, a standard vascular Doppler can be used to monitor heart rate during CMR. The Doppler and telemetric ECG do not contain enough ferromagnetic material to cause visible image degradation. Jorgensen and associates evalu ated whether patients could be monitored during CMR with 1.5 Tesla (T) machines in a manner that complies with monitoring standards.45 The high magnetic field can interfere with normal func tioning of equipment, not only monitoring equipment, but also smaller items, such as infusion pumps used for stress testing. In general, the influence of the CMR scanner on nearby equipment depends on several factors, such as the strength of the CMR magnetic field, the proximity of the equipment to the scanner, the amount of ferromagnetic material in the equipment, and the design of the electrical circuitry. Finally, simple devices, such as closed circuit television and a two way intercommunication system, also aid in monitoring by allowing a constant view of the patient and easy communication if the patient is in discomfort. How ever, the latter may be impaired during imaging because of the noise of the gradients. Because sequences are now commonly being reduced in duration to a breath hold, however, this limitation is becoming much less important.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

cerebral aneurysm clips, which have become dislodged by MR, causing fatal cerebral hemorrhage. Modern clips are not ferromagnetic and are safe, but the problem is establish ing the type of clip with certainty before performing CMR. In general, therefore, CMR should be avoided in these patients unless written information is available and appro priate advice from a neurosurgical center is obtained. There have been cases of bleeding in the eye in patients with pre vious injury with metallic shards (usually metal workers), and again a history of this should be sought. A skull X ray can be helpful in cases of doubt. Electronic implants are the other major problem. These may dysfunction or may be damaged by CMR, which is therefore best avoided. This applies to cochlear implants, neurostimulation systems, pacemakers/ICDs, and a number of other modern implants. In general, the main rule to observe is to determine the risk benefit ratio for the proposed procedure. If the information can only be obtained by CMR, and if it is very important, then the risk benefit ratio may be positive for some patients (as has been shown in some patients with pacemakers). In many clinical circumstances, however, the information required can be obtained by other means. Reference texts on the safety of specific devices are available. For CMR, a number of other specific device issues require mention. Swan Ganz catheters and temporary pac ing wires in general preclude CMR, but sternal wires and vascular clips on grafts do not present safety problems, although localized artifacts occur on the images. Three spe cific areas are discussed in more detail, namely, stents, valve prostheses, and pacemakers/ICDs.

Stents The first intracoronary stents were implanted in human coronary arteries in 1986 by Sigwart and colleagues.49 Recently, the indications for intracoronary stents have expanded and the use of stents has grown dramatically. One of the reasons for this development is the reduced restenosis rate compared with conventional balloon angio plasty.50–53 Because of the increased indications for intra coronary stents, the population of patients with these stents in situ who need to undergo CMR for various diag nostic reasons is rapidly expanding. Coronary stents are metallic structures that remain in situ for life. As a result, there have been concerns about stent dislodgment during CMR as well as concern about possible heating effects. Several factors determine the risk of placing these materials in a magnetic field. These factors include the ferromagnetism of the material, the strength of the static and gradient magnetic fields, the metal mass, and the geometry of the material.10,54 The most commonly used stents are made of stainless steel or tantalum, which is not ferromagnetic.55 In vitro experiments performed by Scott and Pettigrew56 evaluated and quantified the influence of magnetic fields used in clinical MR scanners on widely used coronary stents. The authors used a 1.5 T magnet and the results did not show any significant deflection of the stents used. In vivo studies have been carried out in dogs with tantalum stents in the aorta using a 1.5 T scanner.55 This study examined the CMR compatibility of tantalum stents and evaluated the feasibility of using vascu lar MR to evaluate the patency of the stented vessel in vivo. In this study, the animals underwent CMR repeatedly in the 106 Cardiovascular Magnetic Resonance

first 8 weeks post implantation. No evidence was found of any stent migration after CMR immediately post implantation in the animals. Angiographic evaluation showed no inter val development of luminal narrowing or thrombus in the region of the vascular stent. CMR artifacts produced by the stent were increased with longer echo times. Further experiments by Friedrich and colleagues have shown that the common stents do not heat up during CMR.57 Large studies have subsequently confirmed the safety of stent imaging with CMR. Gerber and coworkers evaluated 111 patients with MR within 8 weeks of stent implantation (median, 8 days; range, 0 to 54 days); during follow up, four noncardiac deaths occurred as well as three revascu larization procedures.58 These events were not related to the CMR procedure. This finding agrees with widespread clinical experience in which stent imaging has been performed in the day after implantation, or soon after ward.59,60 Moreover, the feasibility of assessing changes in coronary flow reserve after percutaneous intervention with stenting was shown recently.61 Another issue regarding CMR procedures in patients with coronary stents is the image distortion caused by the ferromagnetic properties of the stent. In general, the greater the ferromagnetism of a metallic implant, the greater its magnetic susceptibility artifact. However, recent studies have shown that CMR angiography could reliably be used for noninvasive imaging and evaluation of blood flow after stent placement.62,63

Valvular Prosthesis Heart valve prostheses are all safe for CMR.64 These are the most recent recommendations, and they supersede those that suggested that pre model 6000 series Starr Edwards valves might cause problems during MR. This conclusion is backed up by numerous data. Several studies in 1.5 to 2.35 T static magnetic fields have shown that for a number of prosthetic valves there is no hazardous deflection during exposure of the magnetic field.65,66 Heating of small metallic implants was tested in a study reported by Davis and associ ates.46 The authors found no significant increase in temper ature in steel and copper clips that were exposed to changing magnetic fields 6.4 times as strong as those expected to be used in the CMR scanner. As a result, there is no contraindication to CMR in patients with the currently used prosthetic valves.67–69 However, prosthetic material may lead to artifacts on CMR images. To evaluate the influ ence of prosthetic valves on the interpretation of CMR images and the capability of functional valve analysis, in a group of 89 patients and 100 heart valve prostheses, Bachmann and colleagues showed convincingly that all patients could be imaged with CMR without any risk and that prosthesis induced artifacts did not interfere with image interpretation.68 In particular, physiologic valvular regurgitation was easy to differentiate from pathologic or transvalvular regurgitation. DiCesare and coworkers studied 14 patients who were surgically treated with nine biologic and seven mechanical aortic and mitral valves.69 Three classes of artifacts were distinguished and graded as mini mal, moderate, or significant. The biologic valves produced minimal artifacts and the mechanical valves showed only moderate artifacts. In all 16 prosthetic valves, CMR allowed adequate semi quantitative analysis of flow over the valve.

The number of patients with cardiac pacemakers and ICDs has increased exponentially in recent years, with 2.4 mil lion patients in the United States having a permanent pace maker in 2002 and more than 370,000 having an implant in 2003. Many centers consider MR absolutely contraindi cated in these patients, and none of the pacemakers or ICDs has been approved by the FDA for CMR examina tions. A total of 10 deaths have been attributed to MR examinations in patients with pacemakers.69 However, as is often the case, such a dogmatic approach is not entirely correct. First, the reported fatalities were poorly character ized and ECGs are not available; moreover, CMR related deaths during physician supervised examinations have not been reported. Second, many patients with pacemakers have safely undergone CMR.70 Therefore, by all normal rules of semantics, the presence of a pacemaker is not an absolute contraindication. However, the presence of a pacemaker is a strong relative contraindication to scanning, and such procedures still require a great deal more research. They should be undertaken only after careful evaluation of the risk benefit ratio to the patient, and should be performed only in expert cardiovascular centers. The issues surrounding CMR of pacemakers are com plex. In general, three hazardous MR interactions with pacemakers and ICDs should be considered. First, the static magnetic fields exert mechanical forces on the ferro magnetic components of the devices, including the pace maker and shock leads; the static magnetic fields also can induce asynchronous pacing. Second, a pulsed RF field may result in oversensing or may induce currents in the leads, resulting in thermal damage at the tissue electrode interface. Third, the gradient magnetic fields may induce voltages on leads, resulting in over and undersensing. Combined fields may also result in device damage and failure. Various generations of pacemakers and ICDs are cur rently implanted in patients, and studies of the effects of MR on cardiac pacemakers, in both animal models and patients, have been reported with varying results.71 In an in vitro study by Lauck and associates, it was concluded that no disturbances arise when the systems are tested in the asynchronous mode at 0.5 T CMR under standard examination conditions with ECG triggered imaging.72 In a study by Achenbach and colleagues, the effect of MR on pacemakers and electrodes was investigated with phan toms.73 Twenty five electrodes were exposed in a 1.5 T scanner, with continuous registration of temperature at the tip of the electrode. Eleven pacemakers were exposed to MR and the pacemaker output was monitored. Tempera ture increases of up to 63.1 C were observed. Furthermore, no pacemaker malfunctions were observed in the asynchro nous mode. Inhibition or rapid pacing was observed dur ing spin echo CMR if the pacemakers were set to VVI or DDD mode. During scanning with gradient echo CMR, pacemaker function was not impaired. Conversely, Erlebacher and associates reported signifi cant adverse effects of CMR on DDD pacemakers.74 All units paced normally in the static magnetic field, but dur ing MR, all units malfunctioned. All malfunctions were the result of RF interference, whereas gradient and static

magnetic fields had no effects. Thus, despite magnetic field strengths adequate to close pacemaker reed switches, RF interference during MR may cause total inhibition of atrial and ventricular output in DDD pacemakers, and may also lead to dangerous atrial pacing at high rates. Gambel and coworkers studied the effect of MR in five patients with permanent cardiac pacemakers, one of whom was pacemaker dependent.75 A variety of pacing config urations was studied, but none of the patients experienced any torque or heat sensation. Four non pacemaker depen dent patients remained in sinus rhythm throughout the MR procedure. During and after CMR, all pacemakers continued to function normally, except for one transient pause of 2 seconds toward the end of the procedure. The authors concluded that, when appropriate strategies are used, CMR may be performed with an acceptable risk benefit ratio for the patient.75 Pennell reported four patients with pacemakers and urgent clinical problems who underwent CMR. No significant problems occurred in three patients (Fig. 8 5). However, CMR was not attempted in one patient because the pacemaker switched into full output mode near the magnet.76 This study was unusual because all of the patients with pacemakers underwent heart scans,77 whereas most studies report the outcomes of noncardiac scans. After these preliminary data in small groups of patients, a large prospective study was performed in 54 patients undergoing a total of 62 examinations using 1.5 T CMR scanners.78 During the MR examinations, only two patients experienced mild, clinically insignificant symptoms. After the MR examina tions, no loss of capture, changes in lead impedance, or battery voltages were noted. A total of 107 leads were evaluated, including 48 atrial and 59 ventricular leads. A significant change in pacing threshold was noted in 10 (9.4%) leads, and only 2 (1.9%) needed a change in pro grammed output. Threshold changes were not related to cardiac chamber or anatomic location. An important issue is that mid and long term follow up of patients was not obtained and effects may occur at a later stage; this is of particular concern in patients with an increase in pacing thresholds. Also, the effects of CMR related heating were not evaluated in this study. In vivo heating of pacemaker leads was evaluated recently in nine pigs undergoing CMR.79 Significant temperature increases were noted, with significant changes in impedance and minor changes in the stimulation threshold. However, histologic chan ges were not observed. Thus, despite the observed increases in temperature, significant tissue damage was not reported. Finally, pacemaker leads may serve as an antenna, which could result in pacing the heart during scanning at the fre quency of the applied imaging pulses. This could poten tially lead to hypotension and dysrhythmias. This effect was shown in experiments and in several patients while positioned in a CMR system,80–82 but this effect must be separated from excitation caused by the pulse generator. Although recent data have reported minimal effects of CMR on pacemakers, there is too little experience and there are too many types of pacemakers to allow general state ments to be made about their suitability for CMR. Prefera bly, pacemaker dependent patients should not be scanned, but if needed, it has been suggested to program the pace makers in the asynchronous mode (VOO or DOO).83 Cardiovascular Magnetic Resonance 107

8 SPECIAL CONSIDERATIONS FOR CARDIOVASCULAR MAGNETIC RESONANCE: SAFETY, ELECTROCARDIOGRAPHIC SETUP, MONITORING, AND CONTRAINDICATIONS

Pacemakers and Implantable Cardioverter Defibrillators

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Diastole

Patients with pacemakers should not undergo scanning unless special circumstances arise, and then only in centers with special expertise and cardiologic backup. Pacemaker dependent patients should not be scanned. Information on patients with ICDs undergoing MR examinations is scarce. In a study of the safety of devices in an animal model, Roguin and colleagues84 included 17 different ICDs and reported that 9 (53%) ICDs had interrogation or battery problems after the scan. Incidental reports showed that patients with ICDs could safely undergo CMR exami nation.85 Naehle and colleagues reported battery voltage declines after MR scanning86 in 18 patients with ICDs. With the rapid growth in ICD implantations, there is a clear need for more studies, in patients particularly, on the safety of ICDs in CMR.

CONCLUSION In general, CMR is safe and no long term ill effects have been reported. Very rapidly changing gradients may induce nerve excitations that may result in muscle twitch ing. However, clinical scanners operate below the thresh old for such effects. The threshold of excitation of the myocardium is approximately 200 times higher than that for other muscles, and so the heart will not be stimulated

108 Cardiovascular Magnetic Resonance

Systole

Figure 8 5 Images from a young patient with a pacemaker. The patient had numerous ventricular fibrillation arrests. Cardiovascular magnetic resonance imaging suggested sarcoidosis (diagnostic images not shown). The pacemaker artifact can be seen in the transaxial gradient echo cine images in the top row and in the right ventricular outflow tract cine images in the bottom row (straight arrows). Curved arrows show the artifact from the pacing lead in the apex of the right ventricle. Imaging was performed at 0.5 T with echo time of 14 msec. (Images courtesy of Dr. Dudley Pennell.)76

by the rapidly changing gradients. Most metallic implants, such as intracoronary stents, prosthetic valves, and sternal sutures, present no hazard because most materials used are nonferromagnetic. In general, patients with pace makers and ICDs should not undergo CMR because of the unquantifiable risks. According to the recently pub lished Council Panel Report on Clinical Indications for CMR, the following statements were made about the safety of CMR: CMR is safe and no long term ill effects have been demon strated. Claustrophobia occurs in about 2% of patients, but mild anxiolysis is often effective. One of the most important safety issues for CMR is the prevention of introduction into the scanner area of ferromagnetic objects which can become projectiles. Metallic implants such as hip pros theses, prosthetic heart valves, coronary stents and sternal sutures present no hazard since the materials used are not ferromagnetic (although an artefact local to the implant may be present). Care is required in patients with many cerebrovascular clips however, and specialist advice is needed for such patients. Patients with pacemakers, ICDs, retained permanent pacemaker leads and other electronic implants are not scanned, although some reports of success do exist, and there is progress towards manufacture of CMR compatible devices.87

1. Yang PC, Kerr AB, Liu AC, et al. New real time interactive magnetic resonance imaging system complements echocardiography. J Am Coll Cardiol. 1998;32:2049–2056. 2. Kanal E, Shellock FG, Talagala L. Safety considerations in MR imaging. Radiology. 1990;176:593–606. 3. Gulch RW, Lutz O. Influence of strong static magnetic fields on heart muscle contraction. Phys Med Biol. 1986;31:763–769. 4. Doherty JU, Whitman GJR, Robinson MD, et al. Changes in cardiac excitability and vulnerability in NMR fields. Invest Radiol. 1985;20:129–135. 5. Jehenson P, Duboc D, Lavergne T, et al. Change in human cardiac rhythm induced by a 2-T static magnetic field. Radiology. 1988;166:227–230. 6. New PFJ, Rosen BR, Brady TJ, et al. Potential hazards and artifacts of ferromagnetic and nonferromagnetic surgical and dental materials and devices in nuclear magnetic resonance imaging. Radiology. 1983; 147:139–148. 7. Shellock FG, Crues JV. High-field strength MR imaging and metallic biomedical implants: an ex vivo evaluation of deflection forces. AJR. 1988;151:389–392. 8. Shellock FG. MR imaging of metallic implants and materials: a compilation of the literature. AJR. 1988;151:811–814. 9. Randall PA, Kohman LJ, Scalzetti EM, et al. Magnetic resonance imaging of prosthetic cardiac valves in vitro and in vivo. Am J Cardiol. 1988;62:973–976. 10. Teitelbaum GP, Bradley WG, Klein BD. MR imaging artefacts, ferromagnetism, and magnetic torque of intravascular filters, coils and stents. Radiology. 1988;166:657–664. 11. Pavlicek W, Geisinger M, Castle L, et al. The effects of nuclear magnetic resonance on patients with cardiac pacemakers. Radiology. 1983;147:149–153. 12. Bottomley PA, Edelstein WA. Power deposition in whole body NMR imaging. Med Phys. 1981;8:510–512. 13. Safety aspects of magnetic resonance imaging. In: Schaefer DJ, Wehrli FW, Shaw D, Kneeland BJ, eds. Biomedical Magnetic Resonance Imaging: Principles, Methodology, and Applications. New York: VCH; 1988:553. 14. Extremely low frequency (ELF) magnetic fields. In: Persson BRR, Stahlberg F, eds. Health and Safety of Clinical NMR Examinations. Boca Raton: CRC; 1989:49. 15. Cohen M, Weisskoff R, Rzedzian RR, Cantor HL. Sensory stimulation by time-varying magnetic fields. Magn Reson Med. 1990;14:409–414. 16. UK childhood cancer study investigators: exposure to powerfrequency magnetic fields and the risk of childhood cancer. Lancet. 1999;354:1925–1931. 17. Easterbrook PJ, Berlin JA, Gopalan R, Matthews DR. Publication bias in clinical research. Lancet. 1991;337:867–872. 18. Newnham JP, Evans SF, Michael CA, Stanley FJ, Landau LI. Effects of frequent ultrasound during pregnancy: a randomised controlled trial. Lancet. 1993;342:887–891. 19. Irnich W, Schmitt F. Magnetostimulation in MRI. Magn Reson Med. 1995;33:619–623. 20. Hartnell GG, Spence L, Hughes LA, Cohen MC, Saouaf R, Buff B. Safety of MR imaging in patients who have retained metallic materials after cardiac surgery. Am J Roentgenol. 1997;168:1157–1159. 21. Murphy KJ, Cohan RH, Ellis JH. MR Imaging in patients with epicardial pacemaker wires. AJR. 1999;172:727–728. 22. Boutin RD, Briggs JE, Williamson MR. Injuries associated with MR imaging: survey of safety records and methods used to screen patients for metallic foreign bodies before imaging. AJR. 1994;162:189–194. 23. Jones S, Jaffe W, Alvi R. Burns associated with electrocardiographic monitoring during magnetic resonance imaging. Burns. 1996; 22:420–421. 24. Hurwitz R, Lane SR, Bell RA, Brant-Zawadzki MN. Acoustic analysis of gradient-coil noise in MR imaging. Radiology. 1989;173:545–548. 25. Brummett RE, Talbot JM, Charuhas P. Potential hearing loss resulting from MR imaging. Radiology. 1988;169:539–540. 26. McJury M, Stewart RW, Crawford D, Toma E. The use of active noise control (ANC) to reduce acoustic noise generated during MRI scanning: some initial results. Magn Reson Imaging. 1997;15:319–322. 27. Flaherty JA, Hoskinson K. Emotional distress during magnetic resonance imaging. N Eng J Med. 1989;320:467–468. 28. Francis JM, Pennell DJ. The treatment of claustrophobia during cardiovascular magnetic resonance: use and effectiveness of mild sedation. J Cardiovasc Magn Reson. 2000;2:139–141.

29. Quirk ME, Letendre AJ, Ciottone RA, Lingley JF. Anxiety in patients undergoing MR imaging. Radiology. 1989;170:463–466. 30. Weinreb JC, Maravilla KR, Peshock R, Payne J. Magnetic resonance imaging: improving patient tolerance. AJR. 1984;143:1285–1287. 31. Hricak H, Amparo EG. Body MRI: alleviation of claustrophobia by prone positioning. Radiology. 1984;152:819. 32. Goldstein HA, Kashanian FK, Blumetti RF, et al. Safety assessment of gadopentetate dimeglumine in U.S. clinical trials. Radiology. 1990; 174:17–23. 33. Sullivan ME, Goldstein HA, Sansone KJ, et al. Hemodynamic effects of Gd-DTPA administered via rapid bolus or slow infusion: a study in dogs. AJNR. 1990;11:537–540. 34. Weiss KL. Severe anaphylactoid reaction after IV Gd-DTPA. Magn Reson Imaging. 1990;8:817–818. 35. Pennell DJ, Underwood SR, Ell PJ. Safety of dobutamine stress for thallium myocardial perfusion tomography in patients with asthma. Am J Cardiol. 1993;71:1346–1350. 36. Guidelines for advanced life support. A statement by the advanced life support working party of the European Resuscitation Council. Resuscitation. 1992;24:111–121. 37. Sellden H, de Chateau P, Ekman G, et al. Circulatory monitoring of children during anaesthesia in low-field magnetic resonance imaging. Acta Anaesthesiol Scand. 1990;34:41–43. 38. Lindberg LG, Ugnell H, Oberg PA. Monitoring of respiratory and heart rates using a fibre-optic sensor. Med Biol Eng Comput. 1992;30:533–537. 39. Roth JL, Nugent M, Gray JE, et al. Patient monitoring during magnetic resonance imaging. Anesthesiology. 1985;62:80–83. 40. Felblinger J, Boesch C. Amplitude demodulation of the electrocardiogram signal (ECG) for respiration monitoring and compensation during MR examinations. Magn Reson Med. 1997;38:129–136. 41. Felblinger J, Lehmann C, Boesch C. Electrocardiogram acquisition during MR examinations for patient monitoring and sequence triggering. Magn Reson Med. 1994;32:523–529. 42. Felblinger J, Debatin JF, Boesch C, et al. Synchronization device for electrocardiography-gated echo-planar imaging. Radiology. 1995;197:311–313. 43. Fischer SE, Wickline SA, Lorenz CH. Novel real-time R-wave detection algorithm based on the vectorcardiogram for accurate gate magnetic resonance acquisitions. Magn Reson Med. 1999;42:361–370. 44. Chia JM, Fischer SE, Wickline SA, et al. Performance of QRS detection for cardiac magnetic resonance imaging with a novel vectorcardiographic triggering method. J Magn Reson Imaging. 2000;12:678–688. 45. Jorgensen NH, Messick JM, Gray J, et al. ASA monitoring standards and magnetic resonance imaging. Anesth Analg. 1994;79:1141–1147. 46. Davis PL, Crooks L, Arakawa M, et al. Potential hazards in NMR imaging: heating effects of changing magnetic fields and RF fields on small metallic implants. AJR. 1981;137:857–860. 47. Laakman RW, Kaufman B, Han JS, et al. MR imaging in patients with metallic implants. Radiology. 1985;157:711–714. 48. Mechlin M, Thickman D, Kressel HY, et al. Magnetic resonance imaging of postoperative patients with metallic implants. AJR. 1984;143:1281–1284. 49. Sigwart U, Puel J, Mirkovitch V, et al. Intravascular stents to prevent occlusion and restenosis after transluminal angioplasty. N Engl J Med. 1987;316:701–706. 50. Fischman DL, Leon MB, Baim DS, et al. A randomized comparison of coronary stent placement and balloon angioplasty in the treatment of coronary artery disease. N Engl J Med. 1994;331:496–501. 51. Serruys PW, de Jaegere P, Kiemeneij F, et al. A comparison of balloonexpandable stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med. 1994;331:489–495. 52. Kimura T, Yokoi H, Nakagawa Y, et al. Three-year follow up after implantation of metallic coronary artery stents. N Engl J Med. 1996;334:561–566. 53. Carrozza JP, Kuntz RE, Levine MJ, et al. Angiographic and clinical outcome of intra-coronary stenting: immediate and longterm results from a large single-center experience. J Am Coll Cardiol. 1992;20:328–337. 54. Shellock FG, Morisoli S, Kanal E. MR procedures and biomedical implants, materials and devices: an update. Radiology. 1993;189:587–599. 55. Matsumoto AH, Teitelbaum GP, Barth KH, et al. Tantalum vascular stents: in vivo evaluation with MR imaging. Radiology. 1989;170:753–755. 56. Scott NA, Pettigrew RI. Absence or movement of coronary stents after placement in a magnetic resonance imaging field. Am J Cardiol. 1994;73:900–901.

Cardiovascular Magnetic Resonance 109

8 SPECIAL CONSIDERATIONS FOR CARDIOVASCULAR MAGNETIC RESONANCE: SAFETY, ELECTROCARDIOGRAPHIC SETUP, MONITORING, AND CONTRAINDICATIONS

References

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

57. Strohm O, Kivelitz D, Gross W, et al. Safety of implantable coronary stents during H-1 magnetic resonance imaging at 1.0 and 1.5 T. J Cardiovasc Magn Reson. 2000;1:239–245. 58. Gerber TC, Fasseas P, Lennon RJ, et al. Clinical safety of magnetic resonance imaging early after coronary artery stent placement. J Am Coll Cardiol. 2003;42:1295–1298. 59. Nagel E, Hug J, Bunger S, et al. Coronary flow measurements for evaluation of patients after stent implantation. MAGMA. 1998; 6:184–185. 60. Kramer CM, Rogers WJ, Reichek N, et al. Magnetic resonance contrast enhancement versus dobutamine tagging response for assessment of myocardial viability after infarction [abstract]. J Am Coll Cardiol. 1999;33(suppl a):485A. 61. Al Saadi N, Nagel E, Gross M, et al. Improvement of myocardial perfusion reserve early after coronary intervention: assessment with cardiac magnetic resonance imaging. J Am Coll Cardiol. 2000;36: 1557–1564. 62. Duerinckx AJ, Atkinson D, Hurwitz R, et al. Coronary MR angiography after coronary stent placement. AJR. 1995;165:662–664. 63. Kotsakis A, Tan KH, Jackson G. Is MRI a safe procedure in patients with coronary stents in situ? Int J Clin Practice. 1997;51:349. 64. Shellock F. Pocket Guide to MR Procedures and Metallic Objects: Update 1998. Philadelphia, USA: Lippincott-Raven; 1998. 65. Hassler M, Le Bas JF, Wolf JE, et al. Effects of magnetic fields used in MRI on 15 prosthetic heart valves. J Radiol. 1986;67:661–666. 66. Soulen RL, Budinger TF, Higgins CB. Magnetic resonance imaging of prosthetic heart valves. Radiology. 1985;154:705–707. 67. Globits S, Higgins CB. Assessment of valvular heart disease by magnetic resonance imaging. Am Heart J. 1995;129:369–381. 68. Bachmann R, Deutsch HJ, Jungehulsing M, et al. Magnetic resonance tomography in patients with a heart valve prosthesis. Rofo. 1991; 155:499–505. 69. DiCesare E, Enrici RM, Paparoni S, et al. Low field magnetic resonance imaging in the evaluation of mechanical and biological heart valve function. Eur J Radiol. 1995;20:224–228. 70. Avery JK. Loss prevention case of the month: not my responsibility! J Tenn Med Assoc. 1988;81:523. 71. Gimbel JR, Kanal E. Can patients with implantable pacemakers safely undergo magnetic resonance imaging? J Am Coll Cardiol. 2004;43:1325–1327. 72. Lauck G, Smekal AV, Wolke S, et al. Effects of nuclear magnetic resonance imaging on cardiac pacemakers. PACE. 1995;18:1549–1555. 73. Achenbach S, Moshage W, Diem B, et al. Effects of magnetic resonance imaging on cardiac pacemakers and electrodes. Am Heart J. 1997;134:467–473.

110 Cardiovascular Magnetic Resonance

74. Erlebacher JA, Cahill PT, Pannizzo F, Knowles RJ. Effect of magnetic resonance imaging on DDD pacemakers. Am J Cardiol. 1986;57: 437–440. 75. Gimbel JR, Johnson D, Levine PA, Wilkoff BL. Safe performance of magnetic resonance imaging on five patients with permanent cardiac pacemakers. PACE. 1996;19:913–919. 76. Pennell . Cardiac magnetic resonance with a pacemaker in-situ: can it be done [abstract]. J Cardiovasc Magn Reson. 1999;1:72. 77. Heatlie G, Pennell DJ. Cardiovascular magnetic resonance at 0.5 T in five patients with permanent pacemakers. J Cardiovasc Magn Reson. 2007;9:15–19. 78. Martin ET, Coman JA, Shellock FG, et al. Magnetic resonance imaging and cardiac pacemaker safety at 1.5-Tesla. J Am Coll Cardiol. 2004;43:1315–1324. 79. Luechinger R, Zeijlemaker VA, Pedersen EM, et al. In vivo heating of pacemaker leads during magnetic resonance imaging. Eur Heart J. 2005;26:376–383. 80. Hayes DL, Holmes DR, Gray JE. Effect of 1.5 tesla nuclear magnetic resonance imaging scanner on implanted permanent pacemakers. J Am Coll Cardiol. 1987;10:782–786. 81. Holmes DR, Hayes DL, Gray JE, Merideth J. The effects of magnetic resonance imaging on implantable pulse generators. PACE. 1986;9:360–370. 82. Fetter J, Aram G, Holmes DR, et al. The effects of nuclear magnetic resonance imagers on external and implantable pulse generators. PACE. 1984;7:720–727. 83. Gimbel JR, Bailey SM, Tchou PJ, et al. Strategies for the safe magnetic resonance imaging of pacemaker-dependent patients. Pacing Clin Electrophysiol. 2005;28:1041–1046. 84. Roguin A, Zviman MM, Meininger GR, et al. Modern pacemaker and implantable cardioverter/defibrillator systems can be magnetic resonance imaging safe: in vitro and in vivo assessment of safety and function at 1.5 T. Circulation. 2004;110:475–482. 85. Naehle CP, Sommer T, Meyer C, et al. Strategy for safe performance of magnetic resonance imaging on a patient with implantable cardioverter defibrillator. Pacing Clin Electrophysiol. 2006;29:113–116. 86. Naehle CP, Strach K, Thomas D, et al. Magnetic resonance imaging at 1.5T in patients with implantable cardioverter-defibrillators. J Am Coll Cardiol. 2009;54:549–555. 87. Pennell DJ, Sechtem UP, Higgins CB, et al. Clinical indications for cardiovascular magnetic resonance (CMR): consensus panel report. Eur Heart J. 2004;25:1940–1965.

Special Considerations: Cardiovascular Magnetic Resonance in Infants and Children Mark A. Fogel

Cardiovascular magnetic resonance (CMR) has been in use for more than two decades and has become firmly established in the clinical evaluation of anatomy, physiology, and function in patients with congenital heart disease (CHD).1–29 Currently, CMR is used as an adjunct to other imaging modalities, such as echocardiography and invasive angiography. When requested, CMR studies are directed to clarify specific ques tions about the morphology and function of known anatomy rather than de novo anatomic issues.30 However, in areas such as the rings25–29 and in the assessment of left ventricular (LV) volume and mass and right ventricular (RV) volume,31,32 CMR is the established “gold standard.” It offers numerous advantages over other imaging modalities including lack of ionizing radiation, excellent soft tissue contrast, a capacity for true three dimensional (3D) imaging of anatomy and physiology as well as function, accurate flow quantification, noninvasive labeling of the myocardium or blood (myocardial tagging), assessment of myocardial viability and perfusion, coronary imaging, and freely selectable viewing planes with out limitations to “acoustic windows” or overlapping struc tures. These advantages and continued advances in CMR hardware, software, and imaging techniques have brought CMR into mainstream pediatric cardiology. Fast imaging with steady state free precession (SSFP) and real time cine,33 dark blood spin echo,34 and 3D contrast enhanced magnetic reso nance angiography (CE MRA)35 provide additional benefits to CMR. The application of CMR to CHD uses nearly all of the techniques discussed in the “adult” chapters of this text, but it nevertheless stands on its own as a separate discipline in the world of CMR for the following reasons: 1. Technical challenges: Imaging of children is more demanding than imaging of adults. Children require increased spatial resolution because of their smaller size as well as increased temporal resolution because of their higher heart rates compared with adults. Children youn ger than 7 to 9 years old often need to be sedated, making the children incapable of sustained breath holding. “Workarounds” to breath holding have been developed for use in imaging the pediatric patient. 2. Anatomy of CHD: The anatomy of native CHD is different from what is seen in adult cardiology, and the correct diag nosis demands a rigorous and systematic approach as well as a global appreciation of 3D thoracic anatomy. Many types of CHD rely on surgical or catheter based therapies.

The anatomic information obtained from the CMR exami nation must accommodate these interventions. Thus, the physician must be knowledgeable about the pre and post operative anatomy of CHD. For example, the connections between the major cardiac segments (atria, ventricles, and great vessels), along with venous anatomy, can be altered in many types of CHD (e.g., transposition of the great arteries). Communications are present where they should not be (e.g., ventricular septal defect [VSD]), persistent blood vessels from fetal life may remain (e.g., patent ductus arteriosus), and cardiac structures may be markedly hypo plastic (e.g., hypoplastic left heart syndrome). In the evalu ation of CHD after surgery, conduits and baffles constructed to separate the circulations have little parallel in the adult world. 3. Physiology: The unique physiology of CHD is most often a result of the altered anatomy. Evaluation of shunts plays a critical role in the assessment of many types of CHD, yet has a minor role in adults. Assess ment of ventricular function in ventricles with uncom mon shapes is routine in the practice of CMR in CHD. Similarly, the assessment of postoperative physiology where the surgeon has created a systemic to pulmonary artery shunt or an anastomosis between the cava and the pulmonary artery must be addressed. As with other imaging modalities, CMR has limitations and challenges in CHD. Sedation and, in some rare instances, general anesthesia to allow young patients to participate in a 45 to 60 minute scan is always a consider ation. Even in older preteens or teens, cooperation may be problematic (e.g., breath holding). Although many intra vascular coils, wires, stents, and clips are safe for CMR imaging (see www.mrisafety.com), they may cause image artifacts near the structure of interest. The lack of portabil ity of CMR is disadvantageous, especially for the critically ill infant for whom transportation to the CMR suite is cum bersome. Finally, arrhythmias may not allow proper data acquisition, whereas abnormal T waves may not allow for proper triggering. An alternative for some situations is the use of real time CMR, in which sequential images are con structed without regard to electrocardiogram (ECG) trigger ing and sequences with “arrhythmia rejection.” Pacemakers remain a problem, and patients with CHD who have these devices often do not undergo CMR, although there are increasing data to suggest that CMR is safe.36–39 Cardiovascular Magnetic Resonance 111

9 SPECIAL CONSIDERATIONS: CARDIOVASCULAR MAGNETIC RESONANCE IN INFANTS AND CHILDREN

CHAPTER 9

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

This chapter discusses the current state of the art of CMR in CHD. For simplicity, the imaging techniques dis cussed are generic. The major manufacturers of CMR equipment have similar sequences with proprietary nomen clature (see Appendix I.).

GENERAL PROTOCOL FOR CARDIOVASCULAR MAGNETIC RESONANCE IN CONGENITAL HEART DISEASE The CMR procedure in patients with CHD is not standar dized because unsuspected findings are common in this population. Nevertheless, certain basic principles can be formulated into a generalized protocol, as outlined in Figure 9 1. Although the specific protocol for each patient must be individualized, depending on the suspected disease process, this basic outline has been found to be efficient and comprehensive.

Anatomic Imaging After scout localizer examinations are performed, anatomic images are acquired both to survey the thoracic anatomy

GENERALIZED CMR PROTOCOL FOR CHD • Localizers • Static SSFP-axial • HASTE* • Dark blood • Coronary imaging • Cine • Velocity mapping • Gadolinium • Tagging • Viability

• Anatomy • Physiology • Perfusion • Viability

*MPR performed on stack of axial SSFP during HASTE to determine slice orientations and positions of future sequences

Figure 9 1 Generalized protocol for cardiovascular magnetic resonance imaging in patients with congenital heart disease. This is a generalized protocol and must be individualized to the patient and the specific disease. HASTE, half Fourier acquisition single shot turbo spin echo; MPR, multiplanar reconstruction; SSFP, steady state free precession.

and to make an “anatomic diagnosis.” These images also serve as localizers for subsequent physiologic and func tional imaging. A contiguous set of 40 to 50 axial images from the diaphragm to the thoracic inlet is obtained (Fig. 9 2). The volume may extend outside the thorax for special cases, such as to the neck if arch anatomy is being evaluated (because the innominate artery may branch in the neck) or if total anomalous pulmonary venous connec tions are suspected (e.g., connection below the diaphragm).

Ao

RA

RA-PA

RPA

Anatomy Morphology

RV

LPA

C

B

A

RA

LV

D

E

F

Figure 9 2 Stack of contiguous axial steady state free precession images. A selected set of axial CMR images obtained in a patient with an atriopulmonary Fontan connection using static steady state free precession. Images progress from superior (A) to inferior (F). Note the very dilated right atrium (RA) and the RA to pulmonary artery (PA) connection. Ao, aorta; LPA, left pulmonary artery; LV, left ventricle; RPA, right pulmonary artery; RV, right ventricle. 112 Cardiovascular Magnetic Resonance

Cine Cardiovascular Magnetic Resonance Cine SSFP or spoiled gradient recalled echo (GRE) acqui sitions are tailored to the lesion under study. Cine CMR (Fig. 9 5) and velocity mapping are the major sequences used to delineate physiology and function in CHD. These bright blood techniques readily show motion of the heart and blood and can detect turbulence in vessels (e.g., in the pulmonary arteries if they are stenotic) or at the valve level (e.g., aortic stenosis or regurgitation). For example, to determine LV volume overload in a patient with a VSD, a stack of LV short axis images is acquired to mea sure LV volume, mass, ejection fraction, stroke volume, and cardiac index. If, instead of the VSD, the LV volume overload was caused by aortic regurgitation, cine CMR of the LV outflow tract in multiple views can be used to visualize the regurgitant jet. Cine CMR can also be used to delineate the anatomy. If coarctation of the aorta is visualized on dark blood imaging or MPR of the axial dataset, turbulence at the coarctation site should be noted on cine imaging (unless there are

multiple collateral vessels or if the coarctation is not severe). Caution must be used, however, when this signal void is used to assess the degree of stenosis (or valvular regurgitation). The size of the signal void is related to many factors, including echo time (TE). Longer TE increases the signal void size, whereas shorter TE decreases it. The size of the signal void is also a function of the direction of the stenotic jet relative to the orientation of the image voxel. Finally, SSFP imaging may underestimate the signal void seen on GRE sequences. With advances in hardware and software, cine CMR is also used to visualize valvular anatomy, often en face (Fig. 9 6). To visualize a bicuspid aortic valve, the most common type of CHD, an en face aortic valve view is obtained by setting the imaging plane to be perpendicular to the LV outflow tract in two orthogonal views at the level of the aortic valve. This typically results in the equivalent of an echocardiographic parasternal short axis orientation. Either SSFP or GRE can be used; GRE is preferred, because the high signal of the blood flowing into the imaging plane outlines the valve leaflets and gives exceptionally fine valve detail (see Fig. 9 6).

Velocity Mapping After cine imaging, blood flow data are obtained. “Retro spectively” acquired velocity mapping (compared with pro spectively triggered imaging) is preferred because the former allows data to be gathered throughout the entire cardiac cycle (prospective gating does not allow for data acquisition in the 50 to 100 mg immediately before the next R wave). Through plane velocity mapping (flow into and out of the imaging plane) is the most useful type of velocity mapping, although in plane velocity mapping may also be valuable (discussed later). For example, in a patient with aortic regurgitation, a velocity map across the ascending aorta measures the regurgitant fraction (the area under the curve of the reverse flow divided by the area under the curve of the forward flow). Care should be taken to place the imaging plane as close to the sinotubular junc tion as possible. For VSDs, through plane velocity mapping can measure the pulmonary to systemic flow ratio (Qp/Qs) by placing velocity maps across the proximal aorta and proximal pulmonary artery. Measurement of Qp/Qs by velocity mapping has been validated against oximetry.19 One of the strengths of CMR velocity mapping is that it is possible to perform an independent verification to ensure data quality. In the absence of an intracardiac shunt, forward stroke volume from through plane velocity mapping across the ascending aorta should equal forward stroke volume across the pulmonary artery. If a shunt is present, the sum of the flows from each branch pulmo nary artery should equal the flow in the main pulmonary artery. LV stroke volume using cine CMR (in the absence of mitral regurgitation) should equal the aortic forward flow. Velocity mapping can also be used to assess the anat omy of valves (see Fig. 9 6). Through plane velocity mapping can outline the leaflets of valves when imaged en face and can be successful when routine cine imaging is not. Cardiovascular Magnetic Resonance 113

9 SPECIAL CONSIDERATIONS: CARDIOVASCULAR MAGNETIC RESONANCE IN INFANTS AND CHILDREN

Imaging with SSFP depends on contrast between the blood pool and the surrounding tissue, which is enhanced by using a high flip angle. Therefore, the scanner is allowed to select the highest flip angle possible. One disadvantage of using “static” SSFP is that diastolic turbulence may be present on the image and may appear as an absence of a structure (signal loss), as in the case of the pulmonary arteries in a patient with a single ventricle after a Blalock Taussig shunt. These may be very difficult to visualize because the diastolic turbulence creates a major signal loss. This drawback can be compensated for by performing dark blood half Fourier acquisition single shot turbo spin echo imaging or another type of dark blood imaging. It is performed while multiplanar reconstruction (MPR) is being performed (discussed later) on the “static” SSFP images. Alternatively, a set of axial cines can be per formed to visualize these structures in the hope that a phase of the cardiac cycle will show the “missing” structure. With the MPR technique, a set of contiguous images (e.g., axial) are stacked or combined to create a large volume dataset from which software can then reconstruct any arbitrary plane. The exact slice orientation and position can therefore be used for subsequent imaging during the study. In addition, the anatomy can be inspected from multiple views from just the axial dataset. An interesting feature of this approach is that one is not restricted to planar surfaces. A “curved” cut can be created to show the important anatomic features (Fig. 9 3). A set of high resolution, double inversion recovery dark blood images is obtained for the regions of interest (e.g., candy cane view of the aorta for suspected coarcta tion) after delineation via MPR (Figs. 9 3 and 9 4). This type of imaging is used judiciously because it requires relatively long scanning times. If there are time con straints and if there are no turbulence artifacts in the area of interest, this step can be skipped and cine imaging can be performed instead to delineate the region of interest.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

A

B

C

D

E

Figure 9 3 Curved multiplanar reconstruction. This patient has a cervical, circumflex right aortic arch, where the transverse aortic arch passes over the right main stem bronchus and then crosses over to the left side of the thorax posteriorly to descend on the left side of the spine. A to D, Double inversion dark blood axial CMR images that progress from superior (A) to inferior (D). The transverse arch crosses over to the left of the thorax posteriorly. E, When the axial double inversion dark blood images are stacked one atop the other, a curved plane (white line) can be drawn tracing the course of this very tortuous aortic arch that shows the entire aortic arch anatomy (F) in the reformatted image. G, Similar curved plane reconstruction from the bright blood steady state free precession dataset. AAo, ascending aorta; DAo, descending aorta; LV, left ventricle.

F

AAo DAo

LV

G

114 Cardiovascular Magnetic Resonance

PDA

AAo

MPA

DAo

DAo

A

C

B

Figure 9 4 Dark blood imaging. Three examples of cardiac pathology using double inversion dark blood CMR imaging. A, Patent ductus arteriosus (PDA) in a 2 week old infant obtained while free breathing. B, Aortic coarctation in the same patient. C, Tetralogy of Fallot in a 2 day old infant with pulmonary atresia imaged while free breathing to look for collateral vessels. A collateral vessel is seen from underneath the transverse aortic arch (TAo) connecting to the left pulmonary artery (LPA). AAo, ascending aorta; DAo, descending aorta; MPA, main pulmonary artery.

RV LV RV

RV

LV

A

C

B RVOC AAo

LV

RV LV

D

E

Figure 9 5 Cine cardiovascular magnetic resonance. Steady state free precession CMR images from cines of four chamber (A), short axis (B), and long axis (C) images of a patient with Uhl anomaly showing an extremely dilated and thinned right ventricle (RV). D, Systolic frame from the patient in Figure 9 4 with tetralogy of Fallot and pulmonary atresia showing the ventricular septal defect (arrow) and the overriding aorta (Ao). E, Short axis cine of a patient with superoinferior ventricles and a right ventricular outflow chamber (RVOC). AAo, ascending aorta; LV, left ventricle.

Cardiovascular Magnetic Resonance 115

9 SPECIAL CONSIDERATIONS: CARDIOVASCULAR MAGNETIC RESONANCE IN INFANTS AND CHILDREN

TAo

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Figure 9 6 Cine cardiovascular magnetic resonance and velocity mapping across a systemic semilunar valve. Phase encoded velocity maps (A and C) and gradient echo cines of a quadricuspid truncal valve (A and B) in a patient with truncus arteriosus (A and B) and a bicuspid aortic valve (C and D). Arrows point to the various leaflets. Both datasets were acquired with free breathing.

A

B

C

D

Gadolinium-Based Anatomic Cardiovascular Magnetic Resonance The two types of 3D gadolinium techniques for anatomy (both static [Fig. 9 7] and time resolved35 [Fig. 9 8] ver sions), not including late gadolinium enhancement (LGE; discussed later), are useful. If both are to be performed in the same patient, they should be separated in time (
15 to 20 minutes) to allow the gadolinium of the first injection to be washed out before the second injection. These techni ques not only add a special type of 3D dataset that can be rotated and manipulated in many different ways, but also can be used to image smaller vessels much more effectively than other techniques. This may also be used to create a shaded surface display or a volume rendered 3D image (see Fig. 9 7). In addition, the gadolinium can “label” the collateral vessels. Special techniques that do not fall into this generalized protocol may be inserted at different points, depending on the lesion:  Myocardial and blood tagging (Fig. 9 9): When there is a question of regional wall motion abnormalities (e.g., cardiomyopathy, as in Duchenne muscular 116 Cardiovascular Magnetic Resonance

dystrophy,40 or a single ventricle11,22,23) or mass (e.g., tumor or hypertrophic cardiomyopathy), CMR tagging can be used both qualitatively and quantitatively. With myocardial tagging, the ventricle is divided up nonin vasively into “cubes of magnetization” and local defor mation of the myocardium can be visualized (see Fig. 9 9).  If there are concerns about a possible atrial septal defect (ASD) or VSD, a saturation band can be proscribed tag ging blood on a GRE image.41 This “tagged”/dark blood can be used to visualize a shunt (see Fig. 9 9).  Perfusion (see Fig. 9 8): Assessment of myocardial perfusion is also a consideration for the patient with CHD.42 These may include patients with congenital coronary anomalies (e.g., anomalous left coronary artery from the pulmonary artery) and patients who have had surgical manipulation of the coronary arteries (e.g., transposition of the great arteries after an arterial switch procedure or a Ross procedure) as well as patients with Kawasaki disease or other inflammatory disorders. A first pass gadolinium based method typically involves imaging with adenosine (140 mg/kg/min for 4 to 6 minutes), followed 20 minutes later with resting perfusion without vasodilator and 10 minutes later with LGE imaging (discussed later). Cine CMR dobutamine stress43 is a non gadolinium based option for those with impaired renal function and detects wall motion abnormalities.

RPA MPA MPA

LA

C

LV

B RPV

C LV

B

A

E

C AAo

AAo

AAo C C

RV RV

LV

C

DAo

D

F

Figure 9 7 Static three dimensional contrast enhanced magnetic resonance angiography. A D, A child with transposition of the great arteries (S, L, L; “corrected transposition”) with pulmonic stenosis who underwent apical left ventricular (LV) apical to pulmonary artery conduit (C). “Right sided” (A and B) and “left sided” (C and D) phases, including the right ventricle (RV) and ascending aorta (AAo). E, Maximum intensity projection image of the gadolinium injection of a patient after scimitar vein repair with a baffle (B) placed between the right pulmonary vein (RPV) and the left atrium (LA). F, Shaded surface display of the gadolinium injection of a patient with aortic coarctation (arrow) and a conduit (C) between the AAo and descending aorta (DAo). MPA, main pulmonary artery; RPA, right pulmonary artery.

 Late gadolinium enhancement imaging (see Fig. 9 8): Similar to myocardial perfusion, LGE imaging (viability) has a role in CHD.42,44,45 Scar tissue preferentially accu mulates gadolinium. LGE imaging takes advantage of the contrast between the gadolinium laden scar tissue and the gadolinium poor normal myocardium to “label” the scar tissue. Pulse sequences, first described in the mid 1980s,46 have taken advantage of this property, which is unique in noninvasive, nonionizing imaging. With the development of segmented inversion recovery GRE, differences in signal intensity between normal and infarcted myocardium of up to 500% have been achieved.47  The same patients who are candidates for perfusion imaging are also candidates for LGE imaging. In addi tion, hearts that have undergone surgical reconstruction are candidates for LGE imaging to assess for myocardial scar tissue in both the myocardium and the areas that were reconstructed, such as the infundibulum and the pulmonary annulus in a patient after transannular patch repair of tetralogy of Fallot. LGE imaging can also be

used to delineate anatomy because surgically placed patches and valves can become bright with this technique.  Coronary artery imaging45 (Fig. 9 10): Using navigator sequences,48 and for some patients, breath hold imag ing, coronary artery CMR may be used to image patients with CHD who have coronary manipulation (e.g., trans position of the great arteries after arterial switch45), native disease (e.g., anomalous left coronary artery from the pulmonary artery), or acquired disease (e.g., coro nary aneurysm as a result of Kawasaki disease). Both tar geted 3D and whole heart coronary CMR has been successfully applied in infants and children.43,49  Two special situations that are worth mentioning occur not infrequently in pediatrics and require special proto cols that use many of the techniques noted earlier as well as some that were not mentioned:  Tumor/mass characterization50: Many cardiac tumors and masses can be differentiated from each other not only by where they occur in the heart, what symp toms they cause, and at what age they occur, but also Cardiovascular Magnetic Resonance 117

9 SPECIAL CONSIDERATIONS: CARDIOVASCULAR MAGNETIC RESONANCE IN INFANTS AND CHILDREN

LPA

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

F

F

A

B

D

E

C

F

Figure 9 8 Late gadolinium enhancement, perfusion, and time resolved gadolinium imaging. A, Late gadolinium enhancement (LGE) image of a patient who underwent complete atrioventricular canal repair with high signal intensity in the region of the patch (arrow). B, LGE image of a 1 year old child with a right ventricular (RV) fibroma (F) showing how the tumor enhances with this sequence. C, Single perfusion frame of the patient with the RV fibroma showing a ring of high intensity signal in the tumor and much lower signal intensity in the middle of the tumor. D to F, Three phases of a time resolved gadolinium injection (D, earlier phase; F, later phase) in a 5 year old child with single ventricle physiology after Fontan completion. The patient has a left superior vena cava to coronary sinus that was connected to the inferior margin of the Fontan baffle (D and E). F, Later arterial and early venous phases, including the right superior vena cava connection to the right pulmonary artery (arrows).

by their characteristics on CMR. For example, a fibroma accumulates gadolinium and shows signal intensity on LGE imaging and T1 weighted imaging after gadolinium administration, whereas a lipoma shows signal intensity on noncontrast T1 weighted images and becomes signal poor with the application of a fat saturation pulse. Tumor characterization proce dures with CMR are a protocol in themselves and typ ically include T1 and T2 weighted images, images with fat saturation, GRE imaging (e.g., thrombus), perfusion (e.g., hemangiomas), and LGE imaging, T1 weighted images after gadolinium administration, and myocardial tagging. Functional imaging can be used to assess for effects of the tumor, such as obstruction to flow.  Arrhythmogenic right ventricular cardiomyopathy: Arrhythmogenic right ventricular cardiomyopathy is characteristically associated with replacement of the RV myocardium with fatty or fibrofatty tissue.51 In its most flagrant form, the following are seen: (1) left bundle branch block tachycardia; (2) RV dilation; (3) dyskinetic RV wall motion, especially in the right ventricular outflow tract, diaphragmatic surface, and septum; and (4) RV conduction delay on ECG, inverted T waves, and e waves. Imaging fulfills some of the criteria set forth in the 1994 Task Force 118 Cardiovascular Magnetic Resonance

report.52 CMR has been used successfully in adults. However, in the pediatric population, there is uncer tainty as to its utility.53–55 The CMR protocol includes T1 weighted imaging, cine for RV volume and func tion, myocardial tagging if needed to assess regional wall motion, and LGE.

TECHNICAL CONSIDERATIONS IN PEDIATRIC CARDIOVASCULAR MAGNETIC RESONANCE Because of the demands of high spatial and temporal reso lution in infants and children, along with the inability of these patients to hold their breath while sedated, many adult CMR sequences must be adapted for the pediatric population.

Spatial and Temporal Resolution Small voxels needed to image infants and small children can be problematic because of the poor signal to noise ratio

RV

LV

RV LV

C

LAD

B

A

LCx

E

D

RCA

LAD

RCA Ao

PA

LCA LCx

LCA

A

B

C

Figure 9 10 Navigator gated, electrocardiogram gated CMR image of a right coronary artery anomaly obtained with free breathing. A and B, A 3 year old patient with a single right coronary artery (RCA) giving rise to a retroaortic left main coronary artery (LCA) with a retroaortic (benign) course and followed by bifurcation into the left circumflex (LCx) and left anterior descending (LAD) coronary arteries. C, A 1 year old child with anomalous origin of the RCA from the left coronary cusp. Ao, aorta; PA, pulmonary artery.

(SNR). With the use of parallel imaging techniques, the SNR is further impaired, leading to poor quality, “grainy” images. To increase the SNR, a number of strategies have been used, either in isolation or in combination, such as using multiple averages, phase oversampling, decreasing the bandwidth, and avoiding or limiting the use of parallel

imaging. These modifications come at the cost of imaging time for the benefit of image quality. At times, for the smallest voxels, even these strategies are not sufficient and the CMR imager must be satisfied with larger voxels to improve the SNR by decreasing the matrix size. Typically, when imaging the adult, a base matrix Cardiovascular Magnetic Resonance 119

9 SPECIAL CONSIDERATIONS: CARDIOVASCULAR MAGNETIC RESONANCE IN INFANTS AND CHILDREN

Figure 9 9 Myocardial and blood tagging. Diastolic (A) and systolic (B) CMR images of a “one dimensional” tagging sequence in the four chamber view of the patient described in Figure 9 5 with Uhl anomaly. Regional wall motion can be visualized in this manner. C and D, Myocardial “grid” tagging (spatial modulation of magnetization) in the short axis of a patient with hypoplastic left heart syndrome, again showing regional wall motion. E, A set of three images of a patient with an incomplete atrioventricular canal and an ostium primum atrial septal defect. Images with a saturation band are shown on the right side of the heart (upper right, arrow), and a saturation band is seen on the left side of the heart (arrowhead). Black blood can be seen shunting from the left atrium to the right atrium in the lower left image (arrowhead), and bright blood can be seen shunting the same way in the upper right image (lower left, arrow). LV, left ventricle; RV, right ventricle.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

(in the frequency encoding direction) of 256 or 192 is used and this can be decreased to 128 to improve the SNR and still yield sufficient spatial resolution with a field of view of 200 mm or less (pixel size of 1.5 to 2.0 mm or less). Slice thicknesses generally are not less than 2.5 to 3.0 mm because of SNR considerations. Because heart rates can be very high in infants (not uncommonly, 140 to 170 bpm), as a general rule, for an R R interval of less than 500 msec, 15 to 20 phases of the cardiac cycle should be acquired. For an R R interval of 500 to 800 msec, 20 to 25 phases should be acquired. For an R R interval of greater than 800 msec, 25 to 30 phases should be acquired. To be able to obtain this kind of temporal resolution at fast heart rates, both the repeti tion time (TR) and the number of segments must be low. If the TR and the number of segments are not low enough, in retrospectively gated imaging, the walls of the heart appear blurry and with a double or triple shadow, whereas in prospectively triggered imaging, the heart motion appears stilted and jerky. Adequate temporal resolution can be obtained at fast heart rates with three segments at a minimum and TR of 10 to 30 msec. In retrospective gating sequences, the CMR scanner is acquiring data continuously and recording the ECG simul taneously. After all of the lines of k space are acquired, the computer then “bins” each line of k space to the closest phase of the cardiac cycle it is calculating, interpolating the data as it is performing the “bin.” It is, therefore, impor tant to recognize that the number of calculated phases should not exceed two times the measured phases (essen tially the R R interval [msec]/TR [msec], where TR is the line TR  lines of k space obtained) because there should be two measured points between each interpolated point to obtain robust data. The formula used is: 2  R R interval ðmsecÞ=number of calculated phases ¼ TR ðmsecÞ where TR is defined as the line TR  lines of k space obtained. This is especially important in velocity mapping. At times, it is advantageous to acquire single shot, real time CMR in which all of the lines of k space are acquired in one heartbeat because of respiratory motion or arrhyth mia. If the heart rate is too fast, it may be advantageous to obtain the image over two heartbeats (i.e., imaged at the end of two heartbeats [2  R R interval] instead of one.

Inability of Pediatric Patients to Hold Their Breath To allow CMR to be performed in patients who cannot hold their breath, some institutions have instituted general anes thesia during which the anesthesiologist can suspend respi ration. This approach is both aggressive and invasive and adds considerable expense. At the Children’s Hospital of Philadelphia, deep sedation has been successfully used for many years without untoward effects and allows the infant to breathe freely during the CMR examination. We reserve general anesthesia for patients with cardiorespiratory com promise, those who have failed deep sedation, and those situations in which deep sedation would need to be mark edly prolonged. 120 Cardiovascular Magnetic Resonance

Imaging the patient who is allowed to breathe freely often is used with the following: 1. Multiple excitations to “average out” the respiratory motion. Typically, three excitations (which increase the SNR) are needed to obtain good image quality, although more excitations may be needed for the vigorous breather. 2. Navigator based respiratory gating techniques that mon itor diaphragmatic motion with data acquisition during expiration. In addition to these strategies, placing a saturation band over the chest wall can minimize respiratory artifacts. These common strategies allow high quality CMR images to be obtained, albeit at the cost of longer acquisition times.

Gadolinium-Based Techniques The information discussed earlier applies to all types of imaging, but the gadolinium based techniques have addi tional considerations. 1. When performing “static” 3D gadolinium imaging, many CMR systems use a “bolus tracking” technique in which the major structure of interest is imaged in real time to detect the arrival of gadolinium with the static 3D sequence cued up. When the imager visualizes the gado linium bolus arriving in this structure, the real time sequence is stopped and the static 3D sequence is initiated. This ensures that the maximum concentration of gadolin ium is in the structure of interest during imaging. In chil dren, because of the quick circulatory time, intravenously administered gadolinium will reach the systemic circula tion (e.g., aorta) much more rapidly than in adults. 2. The perfusion technique obtains images at a single slice position at different time points in the cardiac cycle. With the high heart rates found in infants and children, it may be possible to acquire only one or two perfusion slices within a single R R interval. If this occurs, the imager can perform the sequence over two R R intervals without degradation of diagnostic quality and can obtain more slices (three to four) during the scan.

EXAMPLES OF CARDIOVASCULAR MAGNETIC RESONANCE IN CONGENITAL HEART DISEASE Transposition of the Great Arteries With the aorta arising from the RV and the pulmonary artery (PA) arising from the LV, the state of the art surgical procedure for simple transposition of the great arteries is the arterial switch procedure, during which both great ves sels are surgically transposed to arise normally and the cor onary arteries are implanted above the native pulmonary (neoaortic) valve. The pulmonary arteries are typically “draped” over the ascending aorta in the Lecompte maneu ver (Figs. 9 11 to 9 13), which may result in left PA stenosis. It is advantageous for the surgeon to perform a

LPA

RPA

AAo

DAo

B A LPA

RPA

RVOT

RVOT

C

Figure 9 12 Cine steady state free precession CMR imaging of a patient with transposition of the great arteries. A, The patient seen in Figure 9 11A with no left pulmonary artery (LPA) stenosis. B, The patient in Figure 9 11C showing a narrowed right ventricular outflow tract (RVOT). This is a diastolic frame, so no turbulence is seen. RPA, right pulmonary artery.

D

LPA

RVOT

RPA RPA

A

Lecompte maneuver rather than making the right PA course tortuous by bringing it directly posterior and then under neath the aortic arch. For a typical assessment, after the set of axial images is obtained, both dark blood imaging (see Fig. 9 11) and cine of the PA (see Fig. 9 12) are performed in both long axes to

B

assess for stenoses, which generally can occur at the take off of the left PA or in the right PA between the superior vena cava (SVC) and the aorta. Cines of the RV and LV out flow tracts are used to evaluate for obstruction. In addition, cines of the four chamber view and a set of contiguous short axis continuous cines of both ventricles are used to Cardiovascular Magnetic Resonance 121

9 SPECIAL CONSIDERATIONS: CARDIOVASCULAR MAGNETIC RESONANCE IN INFANTS AND CHILDREN

Figure 9 11 Dark blood CMR imaging of a patient with transposition of the great arteries. The Lecompte maneuver and anatomy of the pulmonary arteries. A, “Draping” of the right (RPA) and left pulmonary arteries (LPA) over the ascending aorta (AAo) in a patient without LPA stenosis. B, Image in a patient with similar anatomy but a proximal LPA stenosis (arrow). Long axis view of the RPA (C) and LPA (D) in a patient with a narrowed right ventricular outflow tract (RVOT). DAo, descending aorta.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Figure 9 13 Three dimensional volume rendered contrast enhanced magnetic resonance angiography of a patient with transposition of the great arteries. A, The pulmonary anatomy relative to the aorta is seen in various views rotated around the superoinferior axis of the patient. Note the narrowed right ventricular outflow tract (A, C, and D). Ao, aorta; LPA, left pulmonary artery; RPA, right pulmonary artery; RVOT, right ventricular outflow tract.

RPA LPA

RVOT Ao

RPA

B

A

RPA Ao

RVOT

RVOT

LPA

C

D

quantitate ventricular performance and to search for residual VSD. Data suggest that including RV trabeculations and pap illary muscles as ventricular volume (rather than wall/mass) results in shorter analysis time and better interobserver reproducibility for ejection fraction.53 Velocity mapping across the neoaortic and neopulmonary valves is used to quantify cardiac output. This technique is used across the right and left PAs to determine flow distribution to both lungs and to assess gradients across the stenotic PA.20 Gad olinium is used for 3D reconstruction of the PA and aorta (see Fig. 9 13), and LGE44 is performed to evaluate for myocardial scarring. In the case shown in Figures 9 11C and D, 9 12A and B, and 9 13, the patient had an RV outflow tract obstruc tion and was being evaluated by CMR to determine whether the coronary arteries would be compromised by stent insertion. 3D reconstruction, along with coronary imaging, showed nearly 14 mm between the nearest coro nary artery and the RV outflow tract. Thus, the coronary arteries were far enough away from the region of stent deployment that there would be no concerns. In general, coronary imaging can be performed successfully to show the manipulated coronary arteries in the surgical treatment of this disease.49 122 Cardiovascular Magnetic Resonance

Single Ventricles When only one usable ventricle is present to pump blood effectively and the other is hypoplastic, or when both ven tricles are linked in such a way that separation of the circu lations into two pumping chambers is impossible, the heart falls into the category of functional single ventricle. The aorta may be hypoplastic, and there may be other asso ciated anomalies, such as anomalous venous connections. Single ventricles undergo staged surgical reconstruction, culminating in the Fontan procedure.56 The CMR imager must be familiar not only with the various forms of single ventricles, associated anomalies, and the physiologic and functional sequelae, but also with the various surgical reconstructive techniques to allow for the best possible medical and surgical management of these very complex patients. A comprehensive discussion of this evaluation is beyond the scope of this chapter, and the reader is referred to more specialized sources.6,8,11,12,18,21,22,24,53 As the patient with a single ventricle moves through staged reconstruction, specific targets of examination change, but the overall goal of assessing anatomy, physiol ogy, and function remains. At all stages of surgical

the semilunar (and indirectly) atrioventricular valve, and for ventricular outflow tract obstruction (see Fig. 9 16). In the native state of the single ventricle, because much less is known about the anatomy than at other stages, it is important to perform an even more detailed anatomic assess ment. Anomalous venous structures, such as a decom pressing vein from the LA, the presence of a left SVC, delineation of a visceral situs for heterotaxy, the presence of an inferior vena cava, and hepatic venous drainage are all important details. In addition, because some patients may have needed resuscitation, assessment of ventricular func tion and valve insufficiency is also important. After the stage I procedure for hypoplastic left heart syndrome, for example, aortic arch imaging is used to evaluate the initial repair (see Fig. 9 14). Besides assess ment of the distal aortic arch for obstruction and the aor tic to pulmonary anastomosis, visualization of the aortic to pulmonary shunt (typically, a right Blalock Taussig shunt) or the right ventricle to PA shunt (i.e., Sano proce dure) is important. This is generally done with double inversion dark blood imaging (or CE MRA) because the turbulence across this structure by cine generally causes

nAo

nPA

nAo

nPA

C

B

A

D LPA Inn v

LPA

SVC SVC SVC B RPA

E

F

G

Figure 9 14 Aortic arch, pulmonary artery, and venous pathway imaging in a patient with hypoplastic left heart syndrome. A to C, Various views of the reconstructed aorta with a three dimensional shaded surface display with the native aorta (nAo) connected to the native main pulmonary artery (nPA). A narrowing of the aortic arch is seen (arrow). D, Two dimensional steady state free precession (SSFP) image of the same aortic to pulmonary anastomosis. E, Static SSFP image at the level of the right (RPA) and left pulmonary arteries (LPA) showing their size and relationship to the superior vena cava (SVC). F, Three dimensional shaded surface display of the same connection (SVC to RPA) with the innominate vein (Inn v) in view. G, Dark blood image of the Fontan baffle showing no clot and a widely patent systemic venous pathway. The LPA in long axis can be seen. Cardiovascular Magnetic Resonance 123

9 SPECIAL CONSIDERATIONS: CARDIOVASCULAR MAGNETIC RESONANCE IN INFANTS AND CHILDREN

reconstruction, including the single ventricle in the native state, the following is the minimum assessment:  Aortic arch imaging, aimed mostly at patients with an aor tic to pulmonary anastomosis, to assess for aortic arch obstruction (Figs. 9 14 and 9 15).  Imaging of the PA6,18,24 to assess for stenosis, hypo plasia, and discontinuity (Figs. 9 14 to 9 16).  Anatomic assessment of the Fontan baffle7,8 (see Figs. 9 14 to 9 16); ASD (as in a patient with hypoplas tic left heart syndrome; see Fig. 9 15); ventricular out flow tract obstruction (especially in patients with a bulboventricular foramen; see Fig. 9 16); aortic pulmo nary collateral vessels; anomalous venous structures; and pulmonary or systemic venous obstruction (espe cially in the surgically created bidirectional Glenn shunt or the Fontan baffle itself). (see Figs. 9 14 to 9 16)  Biventricular function,11,21,22,23 including regional wall motion abnormalities, ejection fraction, end diastolic vol ume and mass, stroke volume, cardiac index, and atrio ventricular valve regurgitant fraction (see Fig. 9 15).  Velocity mapping18 to assess for cardiac index, Qp/Qs, relative flow to both lungs, and regurgitant fraction of

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Figure 9 15 Cine, late gadolinium enhancement, and perfusion in patients with a single ventricle. Still frame steady state free precession cines in off axis, two chamber (A), short axis (B), and (C) long axis views of a 2 year old child with hypoplastic left heart syndrome after hemi Fontan procedure, imaged while free breathing. The superior vena cava to the right pulmonary artery connection can be seen in C (arrow) along with the hemi Fontan “dam,” blocking flow from the superior vena cava into the right atrium. D, Late gadolinium enhancement in single ventricles, where patch material such as that used to reconstruct the aorta shows high signal intensity (arrow).

RV

A

C

B

D

signal loss. Qp/Qs is obtained by velocity mapping in the ascending aorta (over each semilunar valve) and by veloc ity mapping across the shunt as well as velocity maps in each PA using high velocity encoding (e.g., 400 cm/sec) and a low TE. The status of the ASD should be assessed, and because this is a volume loaded stage, ventricular function is also a key imaging goal (see Fig. 9 15). After the bidirectional Glenn or hemi Fontan stage (see Fig. 9 15), imaging the SVC to PA anastomosis is the major difference with patients undergoing the stage I procedure. This can be done with cine, dark blood, or CE MRA because the flows are generally low velocity. Qp/Qs uses flow in the SVC with flow in both branches of the PA as an internal check along with flow in the aorta. If a hemi Fontan procedure was performed, imaging should assess whether any leak was present from the SVC to PA anasto mosis into the atrium. Collated flow is assessed with veloc ity maps in the pulmonary veins. After the Fontan procedure, the most important structure to image is the entire systemic venous pathway (Fontan baf fle) for obstruction or clot (see Fig. 9 16). Visualization of 124 Cardiovascular Magnetic Resonance

fenestration flow (using cine) and assessment of this struc ture for thrombus are important steps. Because it is known that patients who have undergone the Fontan procedure have poor ventricular function, cine imaging is an essential part of the CMR study. In addition, CE MRA can help deter mine the presence of collateral arteries and assess the aortic arch. LGE imaging57,58 can be used to visualize myocardial as well as patches in both the reconstructed aorta and the atrioventricular valves (see Fig. 9 15). Perfusion may also be performed to evaluate for defects.

Coarctation of the Aorta One of the more common congenital heart lesions and a very common reason for referral for CMR, coarctation of the aorta is defined as an obstruction in the isthmus or des cending aorta. It may have a number of associated abnorm alities, including bicuspid aortic valve (see Fig. 9 6), complex CHD (e.g., Shone complex), double orifice mitral valve, or a posteriorly malaligned VSD. In older children,

PA

LV

A

E

RVOC

B

C

VENC=3 m/s

D

VENC=5 m/s

F

Figure 9 16 Ventricular outflow obstruction, Fontan baffle, and pulmonary arteries in single ventricles. This patient with heterotaxy has a functional single left ventricle (LV) and a right ventricular outflow chamber (RVOC) with the aorta (Ao) arising from it, whereas the native pulmonary artery (PA) arises from the LV (anatomic transposition of the great vessels). The only connection between the LV and the RVOC, which connects to the Ao is through two small ventricular septal defects (VSD; arrows on A). A, Coronal steady state free precession (SSFP) cine. B, Off axis axial cine through the more superior VSD. Note the turbulence/dephasing through the VSD (arrow). C and D, Still frame images from in plane velocity mapping at a velocity encoding (VENC) of 3 m/sec and 5 m/sec, respectively, of the VSD. The velocity exceeds 3 m/sec as indicated by the white signal intensity in the otherwise black jet (arrow) in C, whereas the velocity does not exceed the VENC at 5 m/sec in D. The instantaneous maximum velocity was 4.5 m/sec. E and F, Cine SSFP and X ray angiography of the Fontan baffle and pulmonary arteries (arrows), respectively. RV, right ventricle.

multiple collateral arteries can be seen. Surgery for the coarctation can be as simple as placement of a patch over the narrowed portion or as complex as an ascending to descending aortic conduit. The CMR procedure can provide exquisite detail of the aortic anatomy and physiology needed for diagnosis and treatment of coarctation (Fig. 9 17). An axial stack of SSFP images will show a decreased descending aortic diameter at the level of the coarctation, whereas the candy cane view with dark blood imaging will show narrowing. It is impor tant to obtain parallel images on both sides of the candy cane view to ensure that the narrowing is not misinterpreted by deviation of the aorta out of the imaging plane. Cine CMR is used to assess for turbulence, whereas the four chamber and ventricular short axis stack is used to obtain the LV mass to identify possible hypertrophy. Cine CMR is also used to show aortic valve morphology and visualize a possible bicus pid aortic valve (see Fig. 9 6). Aortic stenosis or regurgita tion secondary to a bicuspid valve (see Fig. 9 6) can be assessed by cine as well as phase encoded velocity mapping. Through plane phase encoded velocity mapping across the

aortic valve is used to obtain the cardiac index, and regurgi tant fraction and peak velocity is obtained for aortic regurgi tation and stenosis, respectively. In addition, through plane velocity mapping just below the coarctation and at the dia phragm can quantify collateral blood flow.57–60 In plane velocity mapping can be used to obtain peak instantaneous velocity as well. Finally, 3D CE MRA is used to assess for collateral arteries and to create a 3D image of the aorta and the coarctation (see Fig. 9 17).

THE FUTURE With the many ongoing advances in CMR, the future holds great promise for those with CHD. Developments include both diagnostic and therapeutic or interventional CMR (e.g., balloon valvuloplasty, stent placement), molecular imaging, and T2* techniques to evaluate the heart for myo cardial iron and oxygen content as well as real time flow assessment. Functional fetal CMR (Fig. 9 18) with real time imaging shows the feasibility of this approach in patients Cardiovascular Magnetic Resonance 125

9 SPECIAL CONSIDERATIONS: CARDIOVASCULAR MAGNETIC RESONANCE IN INFANTS AND CHILDREN

Ao

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Figure 9 17 Cardiovascular magnetic resonance of aortic coarctation. A, Still frame image from a cine gradient echo in diastole showing narrowing of the aorta at the apex of the aortic arch. B, Accompanying in plane phase encoded velocity map showing the jet at the coarctation site (arrow). C, Dark blood image in a 3 year old child with a typical coarctation of the proximal descending aorta (DAo) (arrow). D and E, Shaded surface display from a three dimensional contrast enhanced magnetic resonance angiography showing extensive collateral vessels. The coarctation is readily seen (arrow). AAo, ascending aorta.

DAo

AAo

A

B

AAo

DAo

C

E

D

A

B

C

Figure 9 18 Functional real time fetal cardiovascular magnetic resonance. At 36 weeks’ gestation, the patient has a diagnosis of tetralogy of Fallot and pulmonary atresia. A, CMR image from a four chamber cine steady state free precession (SSFP) of the heart (arrow). The mother’s anterior abdomen is at the top of the image. B, CMR image of a left ventricular outflow tract cine SSFP (arrow) with the mother’s anterior abdomen to the left of the image. The head is at the bottom. C, Cine SSFP image showing a short axis view of the heart (arrow) with the mother’s anterior abdomen to the left of the image. The head is at the bottom. 126 Cardiovascular Magnetic Resonance

CONCLUSION The use of the CMR technique for infants, children, adoles cents, and adults with CHD is a unique subspecialty field

in the world of CMR. The imager must understand the complex anatomy, physiology, and function of native CHD as well as the treatments (surgical, catheter based, medical) to perform the CMR examination accurately and efficiently. In addition to addressing the myriad of technical issues associated with imaging in pediatrics, ongoing hardware and software advances will likely lead to increasing use of CMR in the CHD population.

References 1. Didier D, Higgins CB, Fisher M, Osakai L, Silverman NH, Cheitlin MD. Congenital heart disease: gated magnetic resonance imaging in 72 patients. Radiology. 1986;158:227–235. 2. Fletcher BD, Jacobstein MD, Nelson AD, Riemenschneider TA, Alfidi RJ. Gated magnetic resonance imaging of congenital cardiac malformations. Radiology. 1984;150:137–140. 3. Higgins CB, Byrd BF, Farmer DW, Osakai L, Silverman N, Cheitlin MD. Magnetic resonance imaging in patients with congenital heart disease. Circulation. 1984;70:851–860. 4. Bank ER. Magnetic resonance of congenital cardiovascular disease. An update. Radiol Clinic North Am. 1993;31:553–572. 5. Adams R, Fellows KE, Fogel MA, Weinberg PM. Anatomic delineation of congenital heart disease with 3D magnetic resonance imaging. In: Proceedings SPIE-Medical Imaging, 1994: Physiology and Function from Multidimensional Images. 1994;2168:184–194. 6. Fogel MA, Ramaciotti C, Hubbard AM, Weinberg PW. Magnetic resonance and echocardiographic imaging of pulmonary artery size throughout stages of Fontan reconstruction. Circulation. 1994;90:2927–2936. 7. Bornemeier RA, Weinberg PM, Fogel MA. Angiographic, echocardiographic and three-dimensional magnetic resonance imaging of extracardiac conduits in congenital heart disease. Am J Cardiol. 1996;78 (6):713–717. 8. Fogel MA, Hubbard A, Weinberg PM. A simplified approach for assessment of intracardiac baffles and extracardiac conduits in congenital heart surgery with two- and three-dimensional magnetic resonance imaging. Am Heart J. 2001;142(6):1028–1036. 9. Fogel MA, Rychik J, Chin A, Hubbard A, Weinberg PM. Evaluation and follow-up of patients with left ventricular apical to aortic conduits using two and three-dimensional magnetic resonance imaging and Doppler echocardiography: a new look at an old operation. Am Heart J. 2001;141:630–636. 10. Weinberg PM, Hubbard AM, Fogel MA. Aortic arch and pulmonary artery anomalies in children. Semin Roentgenol. 1998;33(3):262–280. 11. Fogel MA, Weinberg PM, Gupta KB, et al. Mechanics of the single left ventricle: a study in ventricular-ventricular interaction II. Circulation. 1998;98:330–338. 12. Fogel MA, Weinberg PM, Rychik J, et al. Caval contribution to flow in the branch pulmonary arteries of Fontan patients using a novel application of magnetic resonance presaturation pulse. Circulation. 1999;99:1215–1221. 13. Donofrio MT, Clark BJ, Ramaciotti C, Jacobs ML, Fellows KE, Weinberg PM, et al. Regional wall motion and strain of transplanted hearts in pediatric patients using magnetic resonance tagging. Am J Physiol Regul Integr Comp Physiol. 1999;277:R1481–R1487. 14. Fogel MA, Weinberg PM, Hubbard A, Haselgrove J. Diastolic biomechanics in normal infants utilizing MRI tissue tagging. Circulation. 2000;102:218–224. 15. Eyskens B, Reybrouck T, Bagaert J, et al. Homograft insertion for pulmonary regurgitation after repair of tetralogy of Fallot improves cardiorespiratory exercise performance. Am J Cardiol. 2000;85:221–225. 16. Niezen RA, Helgbing WA, van der Wall EE, van der Geest RJ, Rebergen SA, de Roos A. Biventricular systolic function and mass studied with MRI imaging in children with pulmonary regurgitation after repair for tetralogy of Fallot. Radiology. 1996;201:135–140. 17. Rebergen SA, Chin JGJ, Ottenkamp J, van der Wall EE, de Roos A. Pulmonary regurgitaion in the late postoperative follow-up of tetralogy of Fallot: volumetric quantification by MR velocity mapping. Circulation. 1993;88:2257–2266.

18. Rebergen SA, Ottenkamp J, van der Wall EE, Chin JDJ, de Roos A. Postoperative pulmonary flow dynamics after Fontan surgery: assessment with nuclear magnetic resonance velocity mapping. J Am Coll Cardiol. 1993;21:123–131. 19. Beerbaum P, Korperich H, Barth P, Esdorn H, Gieseke J, Meyer H. Non-invasive quantification of left-to-right shunt in pediatric patients: phase-contrast cine magnetic resonance imaging compared with invasive oxymetry. Circulation. 2001;10:2476–2482. 20. Gutberlet M, Boeckel T, Hosten N, Vogel M, Kuhne T, Oellinger H, et al. Arterial switch procedure for d-transposition of the great arteries: quantitative midterm evaluation of hemodynamic changes with cine MR imaging and phase-shift velocity mapping — initial experience. Radiology. 2000;214:467–475. 21. Fogel MA, Weinberg PM, Chin AJ, Fellows KE, Hoffman EA. Late ventricular geometry and performance changes of functional single ventricle throughout staged Fontan reconstruction assessed by magnetic resonance imaging. J Am Coll Cardiol. 1996;28(1):212–221. 22. Fogel MA, Weinberg PM, Fellows KE, Hoffman EA. Study in ventricular–ventricular interaction: single right ventricles compared with systemic right ventricles in a dual chambered circulation. Circulation. 1995;92(2):219–230. 23. Fogel MA, Gupta KB, Weinberg PW, Hoffman EA. Regional wall motion and strain analysis across stages of Fontan reconstruction by magnetic resonance tagging. Am J Physiol Heart Circ Physiol. 1995;269(38):H1132–H1152. 24. Fogel MA, Weinberg PM, Hoydu A, et al. The nature of flow in the systemic venous pathway in Fontan patients utilizing magnetic resonance blood tagging. J Thorac Cardiovasc Surg. 1997;114:1032–1041. 25. Ho V, Prince M. Thoracic MR aortography: imaging techniques and strategies. Radiographics. 1998;18(2):287–309. 26. Beekman R, Hazekamp M, Sobotka M, et al. A new diagnostic approach to vascular rings and pulmonary slings: the role of MRI. Magn Reson Imaging. 1998;16(2):137–145. 27. van Son J, Julsrud P, Hagler D, et al. Imaging strategies for vascular rings. Ann Thorac Surg. 1994;57(3):604–610. 28. Didier D, Ratib O, Beghetti M, et al. Morphologic and functional evaluation of congenital heart disease by magnetic resonance imaging. J Magn Reson Imaging. 1999;10:639–655. 29. Fleenor JT, Weinberg PM, Kramer SS, Fogel M. Vascular rings and their effect on tracheal geometry. Pediatr Cardiol. 2003;24:430–435. 30. Fratz S, Hess J, Schuhbaeck A, et al. Routine clinical cardiovascular magnetic resonance in paediatric and adult congenital heart disease: patients, protocols, questions asked and contributions made. J Cardiovasc Magn Reson. 2008;10:46. 31. Mor-Avi V, Sugeng L, Weinert L, et al. Fast measurement of LV mass with real-time 3-dimensional echocardiography: comparison with MRI. Circulation. 2004;110:1814–1818. 32. Bu L, Munns S, Zhang H, et al. Rapid full volume data acquisition by real-time 3-dimensional echocardiography for assessment of LV indexes in children: a validation study compared with MRI. J Am Soc Echocardiogr. 2005;18:299–305. 33. Lee VS, Resnick D, Bundy JM, Simonetti OP, Lee P, Weinreb JC. MR evaluation in one breath hold with real-time true fast imaging with steady-state precession. Radiology. 2002;222:835–842. 34. Edelman RR, Chien D, Kim D. Fast selective black blood MR imaging. Radiology. 1991;181:655–660. 35. Finn JP, Baskaran V, Carr JC, et al. Low-dose contrast-enhanced threedimensional MR angiography with subsecond temporal resolution– initial results. Radiology. 2002;224:896–904.

Cardiovascular Magnetic Resonance 127

9 SPECIAL CONSIDERATIONS: CARDIOVASCULAR MAGNETIC RESONANCE IN INFANTS AND CHILDREN

with in utero diagnosis of hypoplastic left heart syndrome to quantify ventricular volumes and cardiac output.61

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

36. Martin ET, Coman JA, Shellock FG, Pulling CC, Fair R, Jenkins K. Magnetic resonance imaging and cardiac pacemaker safety at 1.5-Tesla. J Am Coll Cardiol. 2004;43:1315–1324. 37. Pulver AF, Puchalski MD, Bradley DJ, et al. Safety and imaging quality of MRI in pediatric and adult congenital heart disease patients with pacemakers. Pacing Clin Electrophysiol. 2009;32:450–456. 38. Nazarian S, Roguin A, Zviman MM, et al. Clinical utility and safety of a protocol for noncardiac and cardiac magnetic resonance imaging of patients with permanent pacemakers and implantable-cardioverter defibrillators at 1.5 Tesla. Circulation. 2006;114:1277–1284. 39. Sommer T, Naehle CP, Yang A, et al. Strategy for safe performance of extrathoracic magnetic resonance imaging at 1.5 T in the presence of cardiac pacemakers in non-pacemaker-dependent patients: a prospective study with 115 examinations. Circulation. 2006;114:1285–1292. 40. Ashford MW, Liu W, Lin SJ, et al. Occult cardiac contractile dysfunction in dystrophin-deficient children revealed by cardiac magnetic resonance strain imaging. Circulation. 2005;112:2462–2467. 41. Harris MA, Weinberg PM, Fogel MA. Cardiac magnetic resonance atrial level shunt detection utilizing presaturation tagging. Presented at the 4th World Congress of Pediatric Cardiology and Cardiac Surgery. Argentina: Buenos Aires; 2005. 42. Prakash A, Powell AJ, Krishnamurthy R, Geva T. Magnetic resonance imaging evaluation of myocardial perfusion and viability in congenital and acquired pediatric heart disease. Am J Cardiol. 2004;93:657–661. 43. Paetsch C, Jahnke A, Wahl R, et al. Comparison of dobutamine stress magnetic resonance, adenosine stress magnetic resonance, and adenosine stress magnetic resonance perfusion. Circulation. 2004;110:835–842. 44. Harris MA, Ghods S, Weinberg PM, Fogel MA. Magnetic resonance delayed enhancement for detection of fibrous tissue in postoperative pediatric patients with various forms of congenital heart disease. J Cardiovasc Magn Reson. 2005;7:157. 45. Taylor AM, Dymarkowski S, Hamaekers P, et al. MR coronary angiography and late-enhancement myocardial MR in children who underwent arterial switch surgery for transposition of great arteries. Radiology. 2005;234:542–547. 46. McNamara MT, Tscholakoff D, Revel D, et al. Differentiation of reversible and irreversible myocardial injury by MR imaging with and without gadolinium-DTPA. Radiology. 1986;158:765–769. 47. Simonetti OP, Kim RJ, Fieno DS, et al. An improved MRI technique for the visualization of myocardial injury. Radiology. 2001;218:215–223. 48. Kim WY, Danias PG, Stuber M, et al. Coronary magnetic resonance angiography for the detection of coronary stenoses. N Engl J Med. 2001;345:1863–1869.

128 Cardiovascular Magnetic Resonance

49. Tangcharoen T, Hegde S, Bell A, et al. Feasibility of whole-heart, steady-state free precession magnetic resonance coronary angiography (MRCA) in infants and children with congenital heart disease [abstract 219]. J Cardiovasc Magn Resonan. Abstracts from the 11th annual SCMR scientific sessions. 2008;80–81. 50. Kiaffas MG, Powell AJ, Geva T. Magnetic resonance imaging evaluation of cardiac tumor characteristics in infants and children. Am J Cardiol. 2002;89:1229–1233. 51. Basso C, Thiene G, Corrado D, Angelini A, Nava A, Valente M. Arrhythmogenic right ventricular cardiomyopathy: dysplasia, dystrophy or myocarditis. Circulation. 1996;94:983–991. 52. McKenna WJ, Thiene G, Nava A, et al. Diagnosis of arrhythmogenic right ventricular dysplasia/cardiomyopathy. Task Force of the Working Group Myocardial and Pericardial Disease of the European Society of Cardiology and of the Scientific Council on Cardiomyopathies of the International Society and Federation of Cardiology. Br Heart J. 1994;71:215–218. 53. Winter MM, Bernink FJP, Groenink M, et al. Evaluating the systemic right ventricle by CMR: the importance of consistent and reproducible delineation of the cavity. J Cardiovasc Magn Reson. 2008;10:40. 54. Fogel MA, Weinberg PM, Rhodes L. Usefulness of magnetic resonance imaging for the diagnosis of right ventricular dysplasia in children. 2006;97(8):1232–1237. 55. Midiri M, Finazzo M, Brancato M, et al. Arrhythmogenic right ventricular dysplasia: MR features. Eur Radiol. 1997;7:307–312. 56. Fontan F, Baudet E. Surgical repair of tricuspid atresia. Thorax. 1971;26:240–248. 57. Fogel MA, ed. Ventricular Function and Blood Flow in Congenital Heart Disease. Malden MA: Blackwell Futura; 2005. 58. Fogel MA. Cardiac magnetic resonance of single ventricles. J Cardiovasc Magn Reson. 2006;8(4):661–670. 59. Harris M, Johnson T, Weinberg P, Fogel M. Delayed enhancement cardiovascular magnetic resonance identifies fibrous tissue in children after congenital heart surgery. J Thorac Cardiovasc Surg. 2007; 133:676–681. 60. Steffens JC, Bourne MW, Sakuma H, O’Sullivan M, Higgins CB. Quantification of collateral blood flow in coarctation of the aorta by velocity encoded cine magnetic resonance imaging. Circulation. 1994; 90:937–943. 61. Fogel MA, Wilson DR, Flake A, et al. A new method of functional assessment of the fetal heart using a novel application of “real time” cardiac magnetic resonance imaging. Fetal Diagn Ther. 2005;20:475–480.

Use of Navigator Echoes in Cardiovascular Magnetic Resonance and Factors Affecting Their Implementation David Firmin and Jennifer Keegan

Respiration has been shown to be an important factor influ encing the quality of cardiovascular magnetic resonance (CMR) images. In addition to cardiac motion, which can be addressed reasonably well by electrocardiographic (ECG) triggering, respiratory motion moves the position and distorts the shape of the heart by several millimeters between inspiration and expiration. In 1991, Atkinson and Edelman1 showed the detrimental effects of breathing on the quality of cardiac studies by showing improved detail (fast low angle shot) in breath hold segmented fast gradient echo images compared with conventional non breath hold images. Although breath holding produces images that are free of respiratory motion artifact, it is not without problems. The breath hold position may vary from one breath hold scan to the next, giving rise to mis registration effects, and it may also vary during the breath hold period itself,2 resulting in image blurring and arti facts. In addition, the scan parameters are limited by the need to perform imaging within the duration of a comfort able breath hold period, and for a number of patients, this period may be very short. An alternative to breath holding is to monitor respira tory motion throughout the data acquisition period and to correct the data for that motion, either in real time or through postprocessing, with the efficacy of both techni ques being strongly dependent on the accuracy of the method of motion assessment. During normal tidal respira tion, the superior inferior (SI) motion of the diaphragm is approximately four to five times the anterior posterior motion of the chest wall,3 and consequently, diaphragm motion is the most sensitive measure of respiratory motion. In 1989, Ehman and Felmlee4 were the first to introduce navigator echoes for measuring the displacement of a moving structure and to demonstrate their use in determin ing diaphragm motion during respiration. The navigator echo is the signal from a column of material oriented per pendicular to the direction of the motion to be monitored. On Fourier transformation, this signal results in a well defined edge of the moving structure. The navigator echoes may be interleaved with the imaging sequence and conse quently enable the motion to be determined throughout the data acquisition period.

In CMR, there have been a number of developments in the use of navigator measurement to reduce the problems of respiratory motion. This chapter discusses these devel opments, considers the various choices that have been studied in the implementation of navigators, and discusses their importance. There are many variables in the applica tion and use of navigator echoes, and although there have been some attempts to study these, it is unlikely that we are close to optimizing their application.

USE OF NAVIGATOR INFORMATION There are two distinct ways of using navigator echoes to reduce the problems of respiratory motion in CMR, which are multiple breath holding with feedback and free breath ing methods. The first of these uses the navigator informa tion to provide visual feedback on the diaphragm position to subjects to allow them to hold their breath at the same point repeatedly.5 The second uses the navigator echo mea surement as an input to some form of respiratory gating algorithm while the patient breathes normally. Figure 10 1 shows actual respiratory trace data in a subject when performing multiple breath holds and when free breathing. In both cases, a navigator acceptance window, typically 5 mm wide, is defined, and all data acquired when the nav igator is outside of this window are ignored. The resulting image therefore consists of data acquired over a narrow range of respiratory positions. The respiratory or scan effi ciency is defined as the percentage of ECG triggers that fall within the navigator acceptance window and is a measure of the data rejection rate, which in turn determines the overall scan duration. As the navigator acceptance window is reduced, the rejection rate increases and the scan efficiency decreases. Figure 10 2 shows the residual dia phragm displacements that occurred during data acquisi tions performed during conventional breath holding, breath holding with navigator feedback, and navigator free breathing in normal subjects.6 Both navigator techniques result in images acquired over a reduced range of Cardiovascular Magnetic Resonance 129

10 USE OF NAVIGATOR ECHOES IN CARDIOVASCULAR MAGNETIC RESONANCE AND FACTORS AFFECTING THEIR IMPLEMENTATION

CHAPTER 10

Standard deviation

0

±0.7 mm

±0.7mm

30

20

20 40

60

A

0

Time (s)

60

0

Diaphragm displacement (mm)

Diaphragm displacement (mm)

±2.4 mm

Diaphragm displacement (mm)

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

5mm NE acceptance window

10

0

−10 BH

20

LED

−20

+ 40

B Figure 10 1 Navigator echo respiratory trace data during multiple breath holding with navigator feedback (A) and free breathing (B). In each case, the shaded region shows the position of a 5 mm navigator acceptance window outside of which data is rejected. (Adapted from Taylor AM, Keegan J, Jhooti P, Gatehouse PD, Firmin DN, Pennell DJ. Differences between normal subjects and patients with coronary artery disease for three different MR coronary angiography respiratory suppression techniques. J Magn Reson Imaging. 1999;9:786 793, with permission.)

diaphragm displacement compared with those acquired using repeated conventional breath holding. In addition, they allow a longer overall scan time. This allows for aver aging of data to improve the signal to noise ratio, increas ing the k space coverage for improved spatial resolution and increasing the temporal resolution by reducing the number of individual image views acquired per cardiac cycle.

Multiple Breath Hold Methods Wang and colleagues7 were the first to show the use of a respiratory feedback monitor to reduce misregistration arti facts in consecutive breath hold segmented fast gradient echo coronary artery images and to show improved image quality from averaging scans acquired over multiple breath holds. When used in informed healthy volunteers, this tech nique has been shown to produce good results with reason able scan efficiency.8 However, a period of training is required, and the process can be problematic, particularly with patients who have difficulty holding their breath because of a combination of illness and anxiety.6 Although 130 Cardiovascular Magnetic Resonance

FR

−30 Figure 10 2 Mean diaphragm displacement in 17 normal subjects with conventional breath holding (open circles), breath holding with navigator feedback (closed circles), and free breathing (plus marks). The navigator controlled studies used a 5 mm navigator acceptance window. (Adapted from Taylor AM, Keegan J, Jhooti P, Gatehouse PD, Firmin DN, Pennell DJ. Differences between normal subjects and patients with coronary artery disease for three different MR coronary angiography respiratory suppression techniques. J Magn Reson Imaging. 1999; 9:786 793, with permission.)

it might be expected that breath holding with respiratory feedback would enable the completion of a cardiac study much more quickly than when using the free breathing methods described later, because of the time required for training and the required rest periods between breath holds, the overall examination times are longer than anticipated. In fact, in a group of patients with coronary artery disease, there was no significant difference between the overall examination times with the two techniques,6 although the same study showed that, in a group of normal healthy sub jects, multiple breath holding resulted in a time reduction of 20%.

Free Breathing Methods Free breathing methods require very little cooperation from the patient. The main disadvantage is the potential for respiratory drift, which can cause considerably reduced scan efficiency.9 Recently, therefore, most effort has gone toward improving scan efficiency with this approach. Much of the early work used retrospective respiratory gating.10 With this method, data acquisition was over sampled, typically by a factor of five, and then sorted retro spectively so that the final image was constructed from data acquired over the narrowest possible range of respiratory

much greater scan efficiency than other methods, while retaining scan quality (Table 10 1 and Figure 10 3). An alternative method, initially developed by Sinkus and Bornert to address general respiratory motion16 and more recently applied to imaging of the coronary arteries, used a tailored acceptance window through k space as opposed to phase encode ordering to obtain a very similar result.17 Both of these phase ordering or windowing techniques use a predefined navigator acceptance window, and scan efficiency is reduced when the respiratory pattern changes during study acquisition. This has been more recently addressed by Jhooti and colleagues, who developed a tech nique that combines the benefits of phase ordering with an automatic window selection that enables the highly effi cient acquisition of high quality coronary artery images without the need for a predefined acceptance window.18 Three dimensional motion adapted gating19 is a similar technique that yields images comparable to standard pro spective navigator gating, with significantly improved scan efficiency.20

NAVIGATOR ECHO IMPLEMENTATION Method of Column Selection Two methods have been used for the generation of a navi gator echo. With the spin echo technique, a spin echo signal is gen erated from the column of material formed by the intersec tion of two planes, one excited by a 90
radiofrequency (RF) pulse and the other by a 180
RF pulse. The column cross section may be either rectangular or rhombic, depending on the orientation of the two planes. This approach is very robust and produces an extremely well defined column. However, it cannot be repeated rapidly, and care must be taken to ensure that the column selection planes do not impinge on the region of interest. The alternative approach is to use a selective two dimensional (2D) RF pulse to excite a column of approxi mately circular cross section.21 Although this technique is much more sensitive to factors such as shimming errors, which can potentially cause blurring and distortion of the column, with a reduced flip angle, it can be repeated more rapidly and the navigator artifact is less extensive.

Table 10-1 Image Quality Scores and Scan Efficiencies{ for Three-Dimensional Magnetic Resonance Angiography* Image Quality Mean Score Scan Efficiency

Phase Ordered

ARA

DVA{

RRG

4.4 72 ( 11.6)

4.7 48 ( 11.5)

6.6 72 ( 11.6)

6.8 20

*Mean image quality scores (1 ¼ excellent, 10 ¼ very poor) and scan efficiencies{ ( SD) for data acquired using phase ordering, an accept/reject algorithm (ARA), the diminishing variance algorithm (DVA), and retrospective respiratory gating (RRG) in 15 subjects. { Scan time for the DVA technique is set to that of the phase ordered technique. (Adapted from Jhooti P, Keegan J, Gatehouse PD, et al. 3D coronary artery imaging with phase reordering for improved scan efficiency. Magn Reson Med. 1999; 41: 555.)

Cardiovascular Magnetic Resonance 131

10 USE OF NAVIGATOR ECHOES IN CARDIOVASCULAR MAGNETIC RESONANCE AND FACTORS AFFECTING THEIR IMPLEMENTATION

positions. In 1995, Hofman and associates11 showed that, using this approach, the image quality of three dimensional (3D) coronary acquisitions was improved over those acquired with multiple averages. However, scan efficiency was poor (20% for an oversampling ratio of five), and although the final image was constructed from the narrow est respiratory window possible, the range was highly dependent on the subject’s breathing pattern during the long acquisition period and was often still unacceptably high. After the introduction of prospective control techni ques, navigators have most commonly been used with a simple accept reject algorithm where data are acquired or not, depending on whether the navigator measurement is within a predefined acceptance window. Oshinski and coworkers12 were the first investigators to show high quality coronary artery images with such an approach. The problem with this method, however, is that for reasonably high scan quality, a narrow acceptance win dow of 5 mm or less is required, and this generally results in relatively poor scan efficiency. In addition, as noted ear lier, many subjects and patients undergo a drift in dia phragm position over time,9 such that the predefined acceptance window becomes less and less suitable as the study progresses. The diminishing variance algorithm overcomes this problem because it does not use a prede fined acceptance window.13 With this method, one com plete scan is acquired and the navigator positions are saved for each line of data. At the end of the initial scan, the most frequent diaphragm position during that scan is determined, and a process of reacquiring lines of data that were acquired with diaphragm positions furthest off set from this position begins. As time progresses, the range of diaphragm positions for the data making up the final set is considerably reduced. In addition to the lack of requirement of an acceptance window, this method has the advantage that an image can be reconstructed at any time after the initial dataset is complete. Another alternative to the simple accept reject algorithm that can improve both image quality and scan efficiency is to use a k space ordering that depends on diaphragm posi tion. Two similar approaches have been suggested, based on the finding that the center of k space appears to be more sensitive to motion than the edges.14 Jhooti and colleagues developed a phase encode ordered method that used a dual acceptance window of 5 mm for the center of k space and 10 mm for the outer regions.15 This approach allowed

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

A

C DVA (scan eff. = 73%)

Phase reordered (scan eff. = 73%)

B

Figure 10 3 A single slice from a three dimensional dataset showing a long section of the right coronary artery. The phase ordered images (A) are of comparable quality to those acquired with the accept reject algorithm (B) and better than those acquired with both the diminishing variance algorithm (C) and retrospective respiratory gating (D). Scan efficiency is also significantly higher for phase ordering than for both the accept reject algorithm and retrospective respiratory gating techniques. (Adapted from Jhooti P, Keegan J, Gatehouse PD, Collins S, Rowe A, Taylor AM, Firmin DN. 3D coronary artery imaging with phase reordering for improved scan efficiency. Magn Reson Med 1999. 41:555 562, with permission.)

D ARA (scan eff. = 49%)

RRG (scan eff. = 20%)

Both methods are used routinely for research studies on coronary imaging, without any reported problems.

Correction Factors In CMR, navigator echoes are most frequently used to mea sure the position of the diaphragm. However, the motion of the heart is not straightforward, and only the inferior bor der that sits on the diaphragm will move to the same extent, whereas superiorly, the relative motion will be reduced. This was first studied by Wang and associates,3 who measured the displacement of the right coronary artery root, the origin of the left anterior descending artery, and the superior and inferior margins of the heart relative to the diaphragm in 10 healthy subjects. For the right cor onary artery origin, the mean ( SD) relative displacement (or correction factor) was 0.57 ( 0.26). McConnell and coworkers22 first used this correction factor to track the position of the imaging slice during breath holding and showed improved image registration relative to untracked acquisitions. In free breathing studies, the correction factor was first applied by Danias and colleagues,23 who showed that tracked image quality was maintained as the navigator acceptance window increased from 3 mm to 7 mm, whereas in untracked images, it decreased significantly. This technique, called real time prospective slice following, 132 Cardiovascular Magnetic Resonance

or slice tracking, is now used routinely for both 2D and 3D methods of acquisition. Of note, however, is the rela tively high standard deviation of the correction factor noted earlier, which reflects considerable intersubject variation in the degree of cardiac motion with respiration. This was also observed by Danias and coworkers, who used real time 2D echo planar imaging to study the SI motion of the heart as a function of navigator position.24 The accuracy of slice fol lowing techniques will obviously depend on the accuracy of the correction factor implemented. In 1997, Taylor and colleagues25 showed how a subject specific factor could be measured rapidly with end inspiratory and end expira tory breath hold scans before the coronary imaging proto col. Figure 10 4 shows the relationship between the motion of the right hemi diaphragm and the coronary ostia measured in one subject, with the slope of the graph giving the correction factor. Figure 10 5 shows two examples of subjects with very different correction factors, showing how a wider acceptance window can be used, thus improv ing scan efficiency. The need for a subject specific correc tion factor has further been confirmed in 3D coronary angiography, where its use was found to yield optimal image quality.26 In 2002, Keegan and associates further developed this area of work by studying the variability of correction factors in the SI, anterior posterior, and right left directions for both breath holding and free breathing.27 The study concluded that subject variability in correction

Downward coronary displacement (mm)

y = –0.08 – (0.45 x) r = 0.99

4

8

12

16 0

10

20

30

Downward diaphragm displacement (mm) Figure 10 4 Plot of superior inferior right coronary artery displacement against superior inferior diaphragm displacement for a single subject. The gradient of the linear regression line is the subject specific correction factor. (Modified from Taylor AM, Keegan J, Jhooti P, Firmin DN, Pennell DJ. Calculation of a subject specific adaptive motion correction factor for improved real time navigator echo gated magnetic resonance coronary angiography. J Cardiovasc Magn Reson. 1999; 1:131 138, with permission.)

factors, together with within subject differences between breath holding and free breathing, is such that slice follow ing should be performed with subject specific factors deter mined from free breathing acquisitions. An additional or alternative approach to the real time slice following described earlier is to use a postprocessing adap tive motion correction technique4 to correct an image retro spectively for movement occurring during data acquisition. This technique, which can be used to correct a 2D acquisi tion for in plane displacement or a 3D acquisition for in plane and through plane displacement, may not appear to be an attractive option initially, but it has the advantages of allowing the correction factor to be optimized for each indi vidual patient and provides an alternative approach to those centers with scanners that do not have a real time decision making capability. This approach has been implemented with both segmented gradient echo28 and interleaved spi ral29 coronary artery acquisitions, with promising results.

Column Positioning The degree of diaphragm motion detected by the navigator echo is dependent on the positioning of the navigator col umn. The dome of the right hemi diaphragm is higher than that of the left, and the two move coherently with respira tion, but to differing degrees.30 Motion of the diaphragm is also greater posteriorly than anteriorly (anterior and dome excursions are 56% and 79%, respectively, of posterior excursions), and at the level of the dome, it is greater later ally than medially.31 The correction factor implemented in real time slice following or postprocessing adaptive motion correction, as described earlier, is strongly dependent on the positioning of the navigator column and further

supports the use of a subject specific factor, as described in the previous section. McConnell and colleagues32 studied the effects of vary ing the navigator location on the image quality of coronary artery studies. Navigators were positioned through the dome of the right hemi diaphragm, through the posterior portion of the left hemi diaphragm, through the anterior and posterior left ventricular walls, and through the ante rior left ventricular wall, as shown in Figure 10 6. The advantage of the latter navigator position is that it would eliminate the need for a correction factor, as described in the previous section, relating the navigator echo measured displacement to the coronary artery motion. The results are summarized in Table 10 2 and show no significant differences in the image quality scores obtained with varying navigator location. There was a tendency for the anterior left ventricular wall navigator scans to be lon ger in duration, but the difference did not reach statistical significance. One of the problems with monitoring the heart itself is the complex anatomy, making it more diffi cult to find a suitable position for the navigator column. More sophisticated methods of positioning the column may further improve this method of monitoring cardiac motion.

MULTIPLE COLUMN ORIENTATIONS There is a linear relationship between the SI and anterior posterior motions of the heart, with the SI motion being approximately five times that of the anterior posterior motion.3 For this reason, the real time slice following methods first used by McConnell and colleagues22 and by Danias and associates23 included a correction for anterior posterior motion of the heart, assuming it to be equal to 20% of the SI motion. Unfortunately, there is not always such a strong relationship between the direc tions of motion of the heart with respiration. Sachs and colleagues showed this by using three navigators to measure the SI, anterior posterior, and right left motions of the heart.33 Figure 10 7 shows an example from this study illustrating the scatter of SI, right left, and anterior posterior measurements, made over a period of approxi mately 10 minutes. The group went on to compare the use of one, two, and three navigators for imaging the right coronary artery and showed an improvement when multiple directions of motion were considered. This improvement in image quality, however, must be offset against the main disadvantage, which is that scan effi ciency is reduced, potentially introducing more problems associated with long term drift in the breathing pattern. A more recent study by Jahnke and coworkers used a new cross correlation based approach and showed the poten tial advantage of combining three orthogonal navigators.34

Navigator Timing Navigator timing is one of the more important parameters; however, flexibility to alter this is often limited by the com puting architecture of the scanner being used (discussed Cardiovascular Magnetic Resonance 133

10 USE OF NAVIGATOR ECHOES IN CARDIOVASCULAR MAGNETIC RESONANCE AND FACTORS AFFECTING THEIR IMPLEMENTATION

0

6 mm 16 mm

0.70

0.00

1.00

CF

0.25

0.00

1.00

6 mm

CF

16 mm

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

A

B Figure 10 5 Right coronary artery origin images acquired with navigator acceptance windows of 6 mm and 16 mm in subjects with subject specific correction factors (CFs) of 0.7 (A) and 0.25 (B). For both subjects, images were also acquired with CFs of 0 and 1. In the absence of slice following (CF 0), image quality is reduced as the navigator acceptance window increases from 6 mm to 16 mm. When slice following with a subject specific CF is used, however, image quality is maintained. (Modified from Taylor AM, Keegan J, Jhooti P, Firmin DN, Pennell DJ. Calculation of a subject specific adaptive motion correction factor for improved real time navigator echo gated magnetic resonance coronary angiography. J Cardiovasc Magn Reson. 1999; 1:131 138, with permission.)

134 Cardiovascular Magnetic Resonance

C

B

D

Figure 10 6 Navigator column locations positioned on tranverse, coronal and sagittal pilot images: (A) through the dome of the right hemi diaphragm, (B) through the posterior left hemi diaphragm, (C) through both anterior and posterior left ventricular walls and (D) through the anterior left ventricular wall.

Table 10-2 Image Quality Scores, Registration Errors, and Total Scan Times* Parameter Image Quality Score (0 4) Registration error (mm) Craniocaudal Anteroposterior Total Scan Time{ (sec)

Right Diaphragm Navigator

Left Diaphragm Navigator

Left Ventricle Navigator

Anterior LV Wall Navigator

2.3  0.1

2.3  0.1

2.4  0.1

2.2  0.2

0.5  0.1 0.3  0.1 294  28

0.4  0.1 0.3  0.1 314  30

0.6  0.1 0.3  0.1 342  62

0.4  0.1 0.4  0.1 427  111

*Image quality scores (0 ¼ very poor, 4 ¼ excellent) registration errors, and total scan times for different navigator column positions during free breathing MR coronary angiography. There were no significant differences between the navigator column locations. Data are presented as mean  standard error of the mean (SEM); LV ¼ left ventricle { Total scan time is the time from start to finish for 6 scans. Adapted from McConnell MV, Khasgiwala VC, Savord BJ, et al. Comparison of respiratory suppression methods and navigator locations for MR coronary angiography. Am J Roentgenol. 1997;168:1369.

Figure 10 7 Anterior posterior (A/P; A) and right left (R/L; B) navigator echo measurements as a function of superior inferior navigator echo measurements in a healthy subject. (Data provided by Todd Sachs, Stanford University.)

80 60

R/L

A/P

40

A

0

S/I

later). Figure 10 8 shows the three main alternatives: (1) pre , (2) pre and post , and (3) navigators repeated regu larly throughout the cardiac cycle. A simple prenavigator provides the highest scan efficiency when a navigator acceptance window is used, but may not be reliable if there is a sudden change in breathing between the navigator measurement and image data acquisition. Pre and

150

B

0

S/I

150

postnavigators overcome this problem, but of course, they also reduce scan efficiency. In our experience, the use of prenavigators only produces acceptable results for free breathing studies, whereas multiple breath hold acquisi tions certainly require both pre and postnavigators. An important factor that depends on the computer hardware and architecture is the time required after the navigator Cardiovascular Magnetic Resonance 135

10 USE OF NAVIGATOR ECHOES IN CARDIOVASCULAR MAGNETIC RESONANCE AND FACTORS AFFECTING THEIR IMPLEMENTATION

A

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

D. acq. PrePre- and postD. acq.

D. acq.

D. acq.

D. acq.

Rep. Figure 10 8 Timing of the navigators for pre , pre and post , and repeated (Rep.) navigator echo controlled data acquisition (D. acq.).

acquisition before the start of the imaging sequence. Partic ularly if prenavigators only are being used, the longer this interval, the greater the potential for errors caused by respi ratory motion. Also, for ECG R wave triggered scans, this may also have implications for the minimum gating delay that can be obtained and for short gating delay or cine scans, post only navigators can be used as an alternative. This approach was recently implemented in left ventricular function studies, where it was found that image quality in a group of patients with heart failure was significantly improved over conventional breath hold scans.35 Repeated navigators allow for improved cine or multi slice imaging and also provide some potential for estimat ing internavigator respiratory motion. The potential prob lem with this is that the navigator signal to noise ratio could be reduced and this may affect the accuracy of navi gator edge detection. In addition, as the time for navigator output increases, the time for imaging decreases and the number of phases or slices that can be acquired is reduced.

Precision of Navigator Measurement Commonly, a spatial resolution of 1 mm is used along the navigator echo column, for example, having a field of view of 512 mm and sampling 512 points on the navigator read out. However, the precision of the measurement is depen dent to a large extent on the signal to noise ratio of the navigator measurement. The most important factor affect ing the signal to noise ratio is the coil arrangement. If, for example, a single coil is used for imaging and navigator detection, it must be large enough to cover both the imag ing area of interest and the region of navigator detection. On the other hand, if phased array coils are used, it is pos sible to position one coil specifically for navigator detec tion, possibly over the region of the right diaphragm. Another important factor in the precision of the measure ment is the quality of the edge on the navigator trace. To obtain a well defined edge of the diaphragm, for example, it is important to have a reasonably small column cross sec tion and to position it through the dome of the diaphragm, so that the column is perpendicular to the diaphragm edge, rather than more posteriorly, where motion is greatest. 136 Cardiovascular Magnetic Resonance

Finally, the diaphragm edge may be detected by edge detec tion, correlation, or least squares fit algorithms. For rapid tracking (repetition time < 100 msec) or narrower col umns, the signal to noise ratio in the diaphragm trace could be too poor for simple edge detection methods to succeed. Of the remaining two techniques, the least squares fit method has been shown to be more resistant to the effects of noise and to the diaphragm profile deformation that occurs during respiration than the correlation method and therefore would be the technique of choice.36 How ever, most navigator techniques acquire only one or two navigators per cardiac cycle, and in such cases, the signal to noise ratios are usually relatively high and edge detec tion algorithms are generally adequate.

MORE RECENT APPROACHES Other Forms of Navigators As has been mentioned, there are problems with the con ventional navigators that have been described earlier because they generally do not give a direct measure of the respiratory related motion of the heart and they can not be implemented simply and efficiently to give a mea sure of this motion in 3D.37 A number of ingenious alternative approaches have been described. In 2003, Nguyen and colleagues38 described a method that selec tively excited the epicardial fat, followed by a rapid read out scheme that instantaneously gave three 1D images of its position in the x, y, and z directions. Tested on six sub jects, the method showed a slight improvement over con ventional navigators; however, the authors noted a number of problems that would need to be resolved before it could be in routine use. In the same year, another method of rapidly localizing heart signals for measure ment of its position was suggested by Pai and Wen, who used a phase contrast angiographic approach to selectively image the flowing blood in the heart chambers.39 These blood signals were used to define the heart position in the SI and anterior posterior directions. Despite the potential advantages of truly tracking the heart position, this method does not appear to have been developed fur ther or used. Subsequently, Stehning and associates used radial imaging for “self navigation.” They developed an inter leaved 3D radial acquisition modified in such a way that the first readout was always in the SI direction.40 This readout could then be reconstructed every cardiac cycle to give an SI projection for motion extraction. In this work, the authors showed improved definition compared with conventional navigators when imaging a moving phantom and similar image quality on initial in vivo cor onary scans. Hardy and colleagues41 used cross correla tion of low resolution real time 2D spiral coronary artery images to accept or reject images for averaging. This adaptive averaging technique was extended to high resolution segmented acquisitions by cross correlating subi mages reconstructed from individual data segments. Breath ing autocorrection with spiral interleaves (BACSPIN) is a similar technique42 that involves the acquisition of a multi slice spiral dataset during breath holding, followed by repeated acquisition of the same slices during free breathing.

Motion Models Because of the complexity of cardiac motion and the diffi culty in extracting measures to correct for it, there has been an interest in developing modeling methods to assist with this process. Manke and coworkers compared one dimensional (1D) translation (SI direction), 3D translation, and 3D affine transformation motion models in a group of healthy subjects.44 By using an elastic image registration algorithm on 3D coronary images acquired at different breath hold positions, the superiority of the 3D transla tional model over the 1D translational model was shown. The authors also used a fast model based image registration to extract motion information from a time series of low resolution 3D images. This was used in conjunction with conventional navigators to calibrate a respiratory motion model that allowed the prediction of affine transformation parameters, including 3D translation, rotation, scale, and shear motion from the navigator measurements, and was shown with coronary artery imaging.45 McLeish and col leagues46 acquired images at a number of breath hold posi tions and studied the accuracy of both rigid and nonrigid registration methods in registering the other breath hold images with those acquired at end expiration. They used principal component analysis to produce patient specific statistical motion models and suggest how this could be used to assist motion correction in CMR. In 2005, Nehrke and Bornert described their study in which they used a patient specific model to control the acquisition.47 Initially, multiple low resolution 3D images were preac quired during free breathing. An affine model of the respi ratory related cardiac motion was then extracted from these images, and this was steered by real time navigators to con trol the high resolution acquisition. The method was demonstrated in both phantoms and volunteers when a 20 mm navigator acceptance window was used. The results in the volunteers showed slightly inferior image quality to images acquired simply with a 5 mm navigator acceptance window. However, scan efficiency was considerably higher

and image quality was also considerably better than for those acquired simply with a 20 mm acceptance window. A more recent study showed that the residual coronary artery motion observed using affine navigators with a 10 mm acceptance window is similar to that observed with conventional navigator gating with a 5 mm window and that observed using a single breath hold.48

Computer Architecture The computer architecture of modern CMR scanners can be very complex, generally incorporating three main com puters. The host computer runs the user interface and allows connection to the image database, the reconstruction computer is a dedicated rapid processor for reconstruction of the CMR image data, and the scan computer allows con trol and adjustment of parameters associated with the scan ning sequence. The architecture of these computers can significantly affect the potential and usefulness of navigator echoes. On many systems, for example, the navigator signal must be reconstructed and processed on the reconstruction computer, but the measurement made must be passed through the host to control the parameters on the scan computer. This arrangement inevitably adds a variable and unknown delay that is dependent on other tasks being performed by the host operating system. To overcome this, either a direct and rapid link is required, allowing transfer of data from the reconstruction computer to the scan com puter, or the scan computer itself must be capable of acquiring and reconstructing the navigator data, so that no data transfer is required. Newer scanners are generally being designed with rapid acquisition, processing, and con trol in mind.

CONCLUSION Navigator echo has been shown to be an important method for monitoring respiration that has been used for defining the position of the heart, enabling improved coronary and other cardiac imaging. The limited number of studies and the many parameters and variables involved in their use suggest that an optimal method of application may not yet have been developed. With future system development, there will be minimal cost, in imaging time or other factors, involved in obtaining this positional information. There fore, it would seem worthwhile to collect and use it where appropriate. The methods are not robust, probably because of their relative lack of sophistication. One of the major advantages of this technique is that navigators allow images to be acquired during free respira tion, eliminating the need for patient cooperation. They also allow longer acquisition times, enabling higher spatial and temporal resolution and increasing the potential for more sophisticated techniques, such as detailed flow imag ing.49,50 A balance must be maintained, however, and imaging time should not be increased so much that increased respiratory drift cancels any potential benefit.

Cardiovascular Magnetic Resonance 137

10 USE OF NAVIGATOR ECHOES IN CARDIOVASCULAR MAGNETIC RESONANCE AND FACTORS AFFECTING THEIR IMPLEMENTATION

Each heavily undersampled free breathing spiral interleaf is compared with the same interleaf acquired during breath holding, and those that match closely are incorporated into a multi slice, multi average dataset. This technique has been shown in a group of six healthy volunteers where increased signal to noise ratio has been achieved with minimal motion blurring. An alternate approach to respiratory motion cor rection that has been applied to spiral imaging is to acquire a low resolution 3D dataset on the fat resonance immedi ately before a high resolution interleaf on the water reso nance. Cross correlation of a selected region of interest of the low resolution fat images from beat to beat is then used to determine the x, y, and z translations of that region and can be used to correct the next high resolution interleaf retrospectively. This approach has been shown in dark blood coronary vessel wall imaging, where high quality images have been obtained without the need for a gating window.43

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

References 1. Atkinson DJ, Edelman RR. Cineangiography of the heart in a single breath hold with a segmented turboFLASH sequence. Radiology. 1991;178:357–360. 2. Holland AE, Goldfarb JW, Edelman RR. Diaphragmatic and cardiac motion during suspended breathing: preliminary experience and implications for breath-holding. Radiology. 1998;209:483–489. 3. Wang Y, Riederer SJ, Ehman RL. Respiratory motion of the heart: kinematics and the implications for the spatial resolution in coronary imaging. Magn Reson Med. 1995;33:713–719. 4. Ehman RL, Felmlee JP. Adaptive technique for high-definition MR imaging of moving structures. Radiology. 1989;173:255–263. 5. Liu YL, Riederer SJ, Rossman PJ, Grimm RC, Debbins JP, Ehman RL. A monitoring, feedback, and triggering system for reproducible breath-hold MR imaging. Magn Reson Med. 1993;30:507–511. 6. Taylor AM, Keegan J, Jhooti P, Gatehouse PD, Firmin DN, Pennell DJ. Differences between normal subjects and patients with coronary artery disease for three different MR coronary angiography respiratory suppression techniques. J Magn Reson Imaging. 1999;9:786–793. 7. Wang Y, Grimm RC, Rossman PJ, et al. angiography in multiple breath-holds using a respiratory feedback monitor. Magn Reson Med. 1995;34:11–16. 8. Keegan J, Gatehouse PD, Taylor AM, Yang GZ, Jhooti P, Firmin DN. Coronary artery imaging on a mobile 0.5Tesla scanner: implementation of real-time navigator-echo controlled segmented k-space FLASH and interleaved spiral sequences. Magn Reson Med. 1999; 41:392–399. 9. Taylor AM, Jhooti P, Wiesmann FW, Keegan J, Firmin DN, Pennell DJ. MR navigator-echo monitoring of temporal changes in diaphragm position: implications for MR coronary angiography. J Magn Reson Imaging. 1997;7:629–636. 10. Lenz GW, Haacke EM, White RD. Retrospective cardiac gating: a review of technical aspects and future directions. Magn Reson Imaging. 1989;7:445–455. 11. Hofman MB, Paschal CB, Li D, Haacke EM, van Rossum AC, Sprenger M. MRI of coronary arteries: 2D breath-hold vs. 3D respiratory-gated acquisition. J Comput Assist Tomogr. 1995;19:56–62. 12. Oshinski JN, Hofland L, Mukundan S, Dixon WT, Parks WJ, Pettigrew RI. Two-dimensional coronary MR angiography without breath-holding. Radiology. 1996;201:737–743. 13. Sachs TS, Meyer CH, Irarrazabal P, Hu BS, Nishimura DG, Macovski A. The diminishing variance algorithm for real-time reduction of motion artifacts in MRI. Magn Reson Med. 1995;34:412–422. 14. Maki JH, Prince MR, Londy FJ, Chenevert TL. The effects of time varying intravascular signal intensity and k-space acquisition order on three-dimensional MR angiography image quality. J Magn Reson Imaging. 1996;6:642–651. 15. Jhooti P, Keegan J, Gatehouse PD, et al. 3D coronary artery imaging with phase reordering for improved scan efficiency. Magn Reson Med. 1999;41:555–562. 16. Sinkus R, Bornert P. Motion pattern adapted real-time respiratory gating. Magn Reson Med. 1999;41:148–155. 17. Sinkus R, Bo¨rnert P. Extension of real-time MR gating to cope with changes in motion pattern: making MR gating autarkic. In: Proceedings of the Sixth Scientific Meeting of ISMRM, Sydney. 1998:2127. 18. Jhooti P, Gatehouse PD, Keegan J, Bunce HH, Taylor AM, Firmin DN. Phase ordering with automatic window selection (PAWS): a novel motion-resistant technique for 3D coronary imaging. Magn Reson Med. 2000;43:470–480. 19. Hackenbroch M, Nehrke K, Gieseke J, et al. 3D motion adapted gating (3D MAG): a new navigator gated 3D coronary MR-angiography. Eur Radiol. 2005;1598–16606. 20. Langreck H, Schnackenburg B, Nehrke K, et al. MR coronary artery imaging with 3D motion adapted gating (MAG) in comparison to a standard prospective navigator technique. J Cardiovasc Magn Reson. 2005;7:793–797. 21. Pauly J, Nishimura D, Macovski A. A k-space analysis of small tipangle excitation. J Magn Reson. 1989;81:43–56. 22. McConnell MV, Khasigawala VC, Savord BJ, et al. Prospective adaptive navigator correction for breath-hold MR coronary angiography. Magn Reson Med. 1997;37:148–152. 23. Danias PG, McConnell MV, Khasigawal VC, Chuang ML, Edelman RR, Manning WJ. Prospective navigator correction of image position for coronary MR angiography. Radiology. 1997;203:733–736.

138 Cardiovascular Magnetic Resonance

24. Danias PG, Stuber M, Botnar RM, Kissinger KV, Edelman RR, Manning WJ. Relationship between motion of coronary arteries and diaphragm during free breathing: lessons from real-time MR imaging. Am J Roentgenol. 1999;172:1061–1065. 25. Taylor AM, Keegan J, Jhooti P, Firmin DN, Pennell DJ. Calculation of a subject-specific adaptive motion-correction factor for improved realtime navigator echo-gated magnetic resonance coronary angiography. J Cardiovasc Magn Reson. 1999;1:131–138. 26. Nagel E, Bornstedt A, Schnackenburg B, Hug J, Oswald H, Fleck E. Optimisation of real-time adaptive navigator correction for 3D magnetic resonance coronary angiography. Magn Reson Med. 1999;42: 408–411. 27. Keegan J, Gatehouse P, Yang GZ, Firmin D. Coronary artery motion with the respiratory cycle during breath-holding and free-breathing: implications for slice-followed coronary artery imaging. Magn Reson Med. 2002;47:476–481. 28. Wang Y, Ehman RL. Retrospective adaptive motion correction for navigator-gated 3D coronary MR angiography. J Magn Reson Imaging. 2000;11:208–214. 29. Keegan J, Gatehouse PD, Yang GZ, Firmin DN. Adaptive motion correction of interleaved spiral images and velocity maps: implications for coronary imaging. Proceedings of the 15th Annual Meeting of the ESMRMB. MAGMA. 1998;6(suppl 1):67. 30. Korin HW, Ehman RL, Riederer SJ, Felmlee JP, Grimm RC. Respiratory kinematics of the upper abdominal organs: a quantitative study. Magn Reson Med. 1992;23:172–178. 31. Gierada DS, Curtin JJ, Erickson SJ, Prost RW, Strandt JA, Goodman LR. Diaphragmatic motion: fast gradient recalled echo MR imaging in healthy subjects. Radiology. 1995;194:879–884. 32. McConnell MV, Khasgiwala VC, Savord BJ, et al. Comparison of respiratory suppression methods and navigator locations for MR coronary angiography. AJR Am J Roentgenol. 1997;168:1369–1375. 33. Sachs TS, Meyer CH, Pauly JM, Hu BS, Nishimura DG, Macovski A. The real-time interactive 3D-DVA for robust coronary MRA. IEEE Trans Med Imag. 2000;19:73–79. 34. Jahnke C, Nehrke K, Paetsch I, et al. Improved bulk myocardial motion suppression for navigator-gated coronary magnetic resonance imaging. J Magn Reson Imaging. 2007;26:780–786. 35. Bellenger NG, Gatehouse PD, Rajappan K, Keegan J, Firmin DN, Pennell DJ. Left ventricular quantification in heart failure by cardiovascular MR using prospective respiratory navigator gating: comparison with breath-hold acquisition. J Magn Reson Imaging. 2000;11:411–417. 36. Wang Y, Grimm RC, Felmlee JP, Riederer SJ, Ehman RL. Algorithms for extracting motion information from navigator echoes. Magn Reson Med. 1996;36:117–123. 37. Nehrke K, Bornert P, Manke D, Bock JC. Free-breathing cardiac MR imaging: study of implications of respiratory motion: initial results. Radiology. 2001;220:810–815. 38. Nguyen TD, Nuval A, Mulukutla S, Wang Y. Direct monitoring of coronary artery motion with cardiac fat navigator echoes. Magn Reson Med. 2003;50:235–241. 39. Pai VM, Wen H. Rapid-motion-perception based cardiac navigators: using the high flow blood volume as a marker for the position of the heart. J Cardiovasc Magn Reson. 2003;5:531–543. 40. Stehning C, Bornert P, Nehrke K, Eggers H, Stuber M. Free-breathing whole-heart coronary MRA with 3D radial SSFP and self-navigated image reconstruction. Magn Reson Med. 2005;54:476–480. 41. Hardy CJ, Saranathan M, Zhu Y, Darrow RD. Coronary angiography by real-time MRI with adaptive averaging. Magn Reson Med. 2000;44:940–946. 42. Hardy CJ, Zhao L, Zong X, Saranathan M, Yucel EK. Coronary MR angiography: respiratory motion correction with BACSPIN. J Magn Reson Imaging. 2003;17:170–176. 43. Keegan J, Gatehouse PD, Yang GZ, Firmin DN. Non-model-based correction of respiratory motion using beat-to-beat 3D spiral fat-selective imaging. J Magn Reson Imaging. 2007;26:624–629. 44. Manke D, Rosch P, Nehrke K, Bornert P, Dossel O. Model evaluation and calibration for prospective respiratory motion correction in coronary MR angiography based on 3D image registration. IEEE Trans Med Imaging. 2002;21:1132–1141. 45. Manke D, Nehrke K, Bornert P. Novel prospective respiratory motion approach for free breathing coronary angiography using patientadapted affine motion model. Magn Reson Med. 2003;50:122–131.

49. Nagel E, Bornstedt A, Hug J, Schnackenburg B, Wellnhofer E, Fleck E. Noninvasive determination of coronary blood flow velocity with magnetic resonance imaging: comparison of breath-hold and navigator techniques with intravascular ultrasound. Magn Reson Med. 1999;41: 544–549. 50. Keegan J, Gatehouse P, Mohiaddin RH, Yang GZ, Firmin D. Comparison of spiral and FLASH phase velocity mapping, with and without breath-holding, for the assessment of right and left coronary artery blood flow velocity. J Magn Reson Imaging. 2004;20:953–960.

Cardiovascular Magnetic Resonance 139

10 USE OF NAVIGATOR ECHOES IN CARDIOVASCULAR MAGNETIC RESONANCE AND FACTORS AFFECTING THEIR IMPLEMENTATION

46. McLeish K, Hill DL, Atkinson D, Blackall JM, Razavi R. A study of the motion and deformation of the heart due to respiration. IEEE Trans Med Imaging. 2002;21:1142–1150. 47. Nehrke K, Bornert P. Prospective correction of affine motion for arbitrary MR sequences on a clinical scanner. Magn Reson Med. 2005;54: 1130–1138. 48. Fischer RW, Botnar RM, Nehrke K, Boesiger P, Manning WJ, Peters DC. Analysis of residual coronary artery motion for breath hold and navigator approaches using real-time coronary MRI. Magn Reson Med. 2006;55:612–618.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

CHAPTER 11

Normal Cardiac Anatomy, Orientation, and Function Michael L. Chuang, Warren J. Manning, and Ronald M. Peshock

Cardiovascular magnetic resonance (CMR) can be used to obtain images of the heart in any plane. Thus, to define normal anatomy and function, it is important to define standard imaging planes in order to develop knowledge of normal anatomy, anatomic variants, and potential arti facts. Standard CMR planes have evolved from other imag ing modalities, including body computed tomography (CT) imaging, echocardiography, and X ray contrast angiogra phy, and consistent nomenclature across imaging mod alities is important1 for accurate and unambiguous communication. The problem is often one of determining the appropriate plane as rapidly as possible to make the diagnosis. As with most other cardiac imaging techniques, it is important to know as much as possible regarding the clinical question prior to determining the protocol. All examinations therefore should be planned to answer a spe cific clinical question. The basic imaging planes can be grouped into planes oriented with respect to the heart, such as horizontal and vertical long axis and short axis, and planes oriented with respect to the major axes of the body, such as the transaxial, sagittal, and coronal planes. Cardiac oriented planes are essential for evaluation of cardiac chamber size and func tion and are familiar from other cardiac imaging techni ques. With CMR, these planes can be positioned very accurately. As shown in Figure 11 1A, a breath hold scout image in the coronal or sagittal plane is the usual starting point. An axial scout (Fig. 11 1B) is used to define the ver tical long axis (also known as the two chamber view, Fig. 11 1C). The horizontal long axis (Fig. 11 1D), which depicts both atria and both ventricles but is slightly differ ent from the true four chamber view, is then planned and is followed by the short axis (Fig. 11 1E and F), which can be used to generate the left ventricular (LV) outflow tract view (Fig. 11 1G), which is similar to the parasternal long axis view of echocardiography. The main structures of normal cardiac anatomy in the coronal, axial, and sagittal planes are shown for spin echo sequences in Figures 11 2A to J. There are many atlases of cross sectional anatomy by CMR that can be helpful2 and web sites (e.g., www.scmr.org) with interactive learning of the cross sectional anatomy can be very useful teaching aids. It is recommended that the reader refer to these for further details. From the standpoint of tissue char acterization, the spin echo images typically permit the dif ferentiation of fat (white) from muscle (intermediate gray). Black regions in spin echo CMR studies may repre sent several tissues or materials, including air, bone, fibrous 140 Cardiovascular Magnetic Resonance

tissue, metal, or rapidly moving blood. It is important to note that if fluid moves relatively slowly (for example, in an aneurysm), its signal intensity will increase, which can mimic more solid tissue such as thrombus or muscle. The placement of imaging planes, slice thickness, and in plane resolution are determined by the size of the struc ture of interest. Presaturation bands can be added to remove specific artifacts. Other preparatory prepulses can be applied to emphasize or deemphasize the signal contri bution of specific tissues. For example, in the evaluation of arrhythmogenic right ventricular cardiomyopathy, high spatial resolution spin echo images of the anterior right ventricular (RV) wall that are free from respiratory artifact are needed. This can be achieved by using a surface coil to improve signal to noise ratio compared with the body coil and thus facilitate higher spatial resolution. Breath hold, double inversion recovery spin echo techniques can also be very effective in removing respiratory and flow artifacts. Imaging planes oriented with respect to the principal axes of the body are particularly useful in the evaluation of the aorta, pericardium, RV free wall, and paracardiac masses. Coronal images can also be quite useful because they present tomographic information in an orientation similar to that of the chest X ray, which is familiar to most clinicians (see Fig. 11 2A and B). In general, axial planes are also useful because they are familiar to the clinician from knowledge of CT (see Fig. 11 2C H). Specific vascu lar structures of interest that can be evaluated well with axial imaging include the thoracic aorta and its branches, the pulmonary artery and veins, and the superior vena cava (see Fig. 11 2C and D). Axial images through the heart can be particularly useful in the evaluation of the pericardium and RV free wall (see Fig. 11 2E H). They are of limited value in the assessment of myocardial wall thickness and chamber size because of the variable orientation of the heart relative to the principal axes of the body. Sagittal images are in general the least familiar to clinicians and are often more difficult to interpret (see Fig. 11 2I and J). However, sagittal images are useful in depicting the RV out flow tract and are therefore helpful in the evaluation of patients with congenital heart disease and RV cardiomyop athy. Oblique sagittal planes are useful in the evaluation of the thoracic aorta, and these planes can be easily defined from the transaxial images, especially if three point plane definition is available by using the arch and lower ascending and descending aorta as the reference points (see Fig. 11 2K and L). Black blood images (see Fig. 11 2M P) oriented

C

11 NORMAL CARDIAC ANATOMY, ORIENTATION, AND FUNCTION

A

B

D

E

F

G H Figure 11 1 For legend see next page.

Cardiovascular Magnetic Resonance 141

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Figure 11 1 A, Scout image 1, coronal: typical breath hold CMR image used to begin study (alternatively, a sagittal image could be used). The white line indicates the location of an axial image used to locate the mitral valve plane and interventricular septum. B, Scout image 2, axial: Typical breath hold image obtained to set up a vertical long axis (VLA) image. The white line indicates the position of the VLA imaging plane and is drawn to pass through the middle of the mitral valve and the ventricular apex. C, Vertical long axis image from breath hold steady state free precession (SSFP) cine CMR oriented as described above. The white line indicates the position of the horizontal long axis (HLA) imaging plane and is selected to pass through the ventricular apex and between the attachment points of the mitral valve. D, Horizontal long axis breath hold SSFP cine CMR image oriented based on the prescription in panel C. The white lines indicate the positions of a stack of images oriented in the left ventricular (LV) short axis orientation, which will be obtained next. E, A representative end diastolic breath hold SSFP cine LV short axis image from the middle of the stack shown in panel D. The white line perpendicular to the imaginary line between the insertion points of the right ventricular (RV) free wall is used to select the imaging plane to obtain a four chamber view of the heart. F, Basal image in the LV short axis orientation showing an oblique view of the aortic valve. The white line shows the orientation of the LV outflow tract (LVOT) view. G, An LVOT view at end diastole. This imaging plane is comparable to the parasternal long axis view of transthoracic echocardiography. H, Four chamber breath hold SSFP cine CMR image at end diastole. This view is similar to the HLA view, but note that typically, less of the aortic outflow tract is seen.

along the functional axes introduced in Figure 11 1 can be useful in the definition and tissue characterization of intra cardiac and paracardiac masses. In addition, depiction of these planes with double inversion recovery black blood imaging is useful for characterization of valvular disease and the coronary artery wall.3–5 Myocardial function is typically assessed by using cine steady state free precession (SSFP) imaging, which has largely supplanted the older cine segmented gradient echo methods. SSFP provides improved contrast between blood pool and myocardium,6 particularly in the presence of impaired ventricular function, as it is dependent mainly on T1/T2 ratio rather than inflow of unsaturated protons. Both SSFP and gradient echo cine methods depict blood as bright (white), while muscle is an intermediate gray, and air, bone, fibrous tissue, and metal are dark. The main findings using bright blood cine CMR of the heart are shown in Figures 11 1C H and 11 2Q T. These cines are typically used to assess myocardial and valve function. The LV outflow tract view, for example, is used to visualize the mitral and aortic valves (see Fig. 11 1G). One advan tage of CMR is the ability to precisely position the long axis planes to avoid the foreshortening that can occur in con trast X ray ventriculography or echocardiography (Fig. 11 1G and H). Short axis views are planned from the long axis views to span the entire LV. The short axis views in Figure 11 2R T are useful in the evaluation of ventricular size and regional function. Using the same ori entation to obtain views of the atria can be useful for asses sing atrial masses as well as chamber size and function. To most accurately assess LV function and size, it is important to obtain correctly oriented images that encom pass the entire LV throughout the cardiac cycle. To obtain true LV short axis views, we go through the following steps. From an axial scout image, the vertical long axis or two chamber view is obtained. The horizontal long axis (HLA) view is planned from the two chamber view, ensuring that the imaging plane passes through the apex of the LV and through the center of the mitral valve annulus. A stack of short axis images is planned from the HLA view (Fig. 11 1D), with the short axis planes perpendicular to an imagi nary line passing through the LV apex distally and midway between the visualized portions of the mitral valve annulus basally. It is important to plan the short axis stack so that it extends just distal to the apex and slightly above the base of the LV to ensure coverage of the entire LV. Failure to do so results in an incomplete dataset of suboptimal use for quantitative volumetric LV measures. The stack of images obtained as described above will be oriented in the LV short 142 Cardiovascular Magnetic Resonance

axis orientation. Though the long axis of the RV and LV are not parallel, from a practical perspective, the LV short axis dataset is used to calculate and assess RV volumes and sys tolic function. Coronary artery CMR requires yet another set of imaging planes to plane the coronary arteries in tomographic slices in the atrioventricular groove and axial planes when the tar geted slab approach is used. Whole heart methods, which encompass the entire coronary tree in a 3D axial volume, yield a volumetric dataset similar to that obtained by ECG gated cardiac CT. This subject is discussed in more detail in Chapter 21.

ANATOMIC VARIANTS Given the ability to obtain images in many planes, it is important to be aware of normal structures and anatomic variants that can complicate interpretation of studies. Sev eral potential confusing features have been described:
Prominence of the lateral border of the right atrial wall (Fig. 11 3A). This structure is a prominence of the tra beculae carneae and crista terminalis and does not rep resent an atrial mass.7
Lipomatous hypertrophy of the atrial septum (Fig. 11 3B). Fat deposition in the atrial septum is occasionally seen, particularly in the elderly. This process spares the region of the fossa ovalis, leading to the characteristic “dumbbell” appearance.8,9 It is benign but is associated with atrial arrhythmias. Severe and extensive lipomatous hypertrophy may extend outside the heart.10 Imaging with and without fat saturation readily characterizes this abnormality.
Superior pericardial recess (Fig. 11 3C). The pericar dium normally extends up the ascending aorta, and this space may contain fluid. This recess can be mistaken for aortic dissection or potentially an anomalous coronary vessel.

COMMON ARTIFACTS A number of CMR artifacts can complicate image interpre tation. These artifacts relate primarily to several features of CMR. The acquisition time is often relatively long in com parison to physiologic processes, which leads to cardiac and respiratory motion artifacts. These motion artifacts must be recognized and minimized at the acquisition stage if possible. Also, because the strength of the local magnetic

Pulmonary artery Left ventricle

RVOT Right atrium

Left ventricular apex

A

B Ascending aorta

Superior vena cava

Superior vena cava

Main pulmonary artery

Transverse aortic arch Trachea

C

Descending aorta

D Pericardium

Right atrial appendage

Pericardium Epicardial fat

Right ventricular outflow tract

Tricuspid valve

RV RA

LV LA

Left atrium

Mitral valve Pulmonary vein

F

E

Right ventricular wall

Inferior vena cava

Right atrium Coronary sinus

G Figure 11 2

Coronary sinus

H

For legend see page 145.

Cardiovascular Magnetic Resonance 143

11 NORMAL CARDIAC ANATOMY, ORIENTATION, AND FUNCTION

Ascending aorta

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Ascending aorta

Right pulmonary artery

Pericardium

Left atrium

RVOT Pericardium Epicardial fat

Right atrium

Right ventricular wall

J

I

Transverse aortic arch

Ascending aorta

K

L

RCA origin Aortic valve leaflets PDA

M

N Left ventricular apex

Anterior AV groove

O Figure 11 2 For legend see page 145.

144 Cardiovascular Magnetic Resonance

Posterior AV groove

P

Descending aorta

R

S

T

11 NORMAL CARDIAC ANATOMY, ORIENTATION, AND FUNCTION

Q

Figure 11 2 A, Coronal breath hold double inversion recovery, spin echo CMR images. Fat is white, myocardium is intermediate gray in intensity, and blood is dark. The slice is positioned anteriorly and cuts through the RV, the RV outflow tract (RVOT), the interventricular septum, and the LV apex. B, Coronal image positioned more posteriorly than the image in panel A. This shows the right atrium, LV, ascending aorta, and pulmonary artery. The aortic valve leaflets are also seen. C, Transverse conventional gated spin echo image at the left of the transverse aortic arch. The trachea and superior vena cava are also demonstrated. D, Transverse conventional gated spin echo image at the level of the main pulmonary artery. The views in panels C and D are useful in the evaluation of possible aortic dissection. E, Transverse conventional spin echo image at the level of the aortic valve. F, Transverse conventional spin echo image at the level of the interatrial septum. The pericardium and epicardial fat are clearly demonstrated. This view can be useful in evaluating atrial masses and pericardial disease. G, Transverse conventional spin echo image at the level of the coronary sinus. The RV wall, epicardial fat, and pericardium are also demonstrated. This view can be helpful in evaluating patients for constrictive pericarditis and RV dysplasia. H, Transverse conventional spin echo image at the level of the entrance of the inferior vena cava into the right atrium. I, Sagittal conventional spin echo image obtained through the ascending aorta. The pericardium is clearly demonstrated. This view can be helpful in the evaluation of the ascending aorta and pericardium. J, Sagittal conventional spin echo image obtained through the RVOT. This view can be helpful in evaluating the pericardium, RVOT, and RV free wall. K, Transverse breath hold double inversion recovery, spin echo images obtained at the level of the transverse portion of the aortic arch (left) and the main pulmonary artery (right). The white line indicates the position of a parasagittal oblique plane used to obtain a “candy cane” view of the aorta (next panel). L, Parasagittal view of the aorta. The ascending aorta, transverse aorta, and descending aorta are seen in a single slice. The vessels to the head and neck are also well seen. This view can be helpful in the evaluation of aortic disease. M, Long axis view using breath hold double inversion recovery technique. This image is comparable with the parasternal long axis view in transthoracic echocardiography. Both the RV and LV are well demonstrated. The origin of the right coronary artery (RCA) is seen in the fat of the anterior atrioventricular groove. The aortic valve leaflets are also well seen. This view can be useful in the evaluation of hypertrophic cardiomyopathy with septal asymmetry. N, Short axis view using breath hold double inversion recovery technique. The LV and RV walls are well demonstrated. In this image the posterior descending artery (PDA) is also seen in cross section in the posterior interventricular groove. O, Four chamber view using breath hold double inversion recovery technique. P, Vertical long axis or two chamber view using breath hold double inversion recovery technique. Q, End diastolic image in the HLA orientation from an steady state free precession (SSFP) breath hold cine CMR sequence. The white lines indicate the locations of short axis imaging planes in subsequent panels R T, all of which were obtained by using breath hold cine SSFP imaging). R, Basal RV and LV. The left anterior descending coronary artery is seen in the anterior interventricular groove. S, End diastolic short axis image obtained at the mid level of the left ventricle. T, End systolic image at the same imaging level as in panel S. LA, left atrium, LV, left ventricle; RA, right atrium, RV, right ventricle.

field determines the position of an object in a CMR image, alterations in the local magnetic field can shift the image position of the structure. Therefore, metal on or in the body can alter the local magnetic field, leading to distortion and local signal loss. Finally, hydrogen nuclei in fat experi ence a slightly different magnetic field in comparison with hydrogen nuclei in water molecules because of the local

chemical environment. This chemical shift is used in CMR spectroscopy to differentiate one compound from another. However, in CMR, this results in what is known as a chemical shift artifact at the interface of water and fatty tissues. This artifact results from sharing of fat and water components within a pixel, leading to signal cancellation. Specific examples are given for each type of artifact. Cardiovascular Magnetic Resonance 145

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

A

Figure 11 3 A, Transverse gradient echo CMR image at the level of the aortic valve obtained by using respirator gating with a navigator echo. A right atrial ridge is noted in the lateral wall of the right atrium (arrow). This finding is normal and should not be mistaken for a right atrial mass. B, Single frame from horizontal long axis steady state free precession cine CMR. Lipomatous hypertrophy of the atrial septum is demonstrated (arrow). There is fatty infiltration of the septum that does not involve the region of the fossa ovalis, resulting in the typical “dumbbell” appearance. C, Oblique double inversion recovery breath hold CMR image obtained at the level of the right pulmonary artery. The extension of the pericardial space both anterior and posterior to the ascending aorta is demonstrated (arrows). The pericardial recess should not be mistaken for evidence of aortic dissection.

B

C

Cardiac Motion Artifacts Except for single shot echo planar imaging (EPI) or other real time imaging approaches, CMR requires gating to the electro cardiogram (ECG) or peripheral pulse. Inaccurate cardiac gating can result in ghosting and other artifacts (Fig. 11 4A and B). Focused efforts to confirm accurate QRS detection are well worth the additional time and effort with regards to image quality. Vectorcardiographic techniques have been implemen ted to take advantage of the difference in the normal vector and the vector of the artifact from the magnetohydrodynamic effect to improve ECG gating.11 Surprisingly good quality cine images can be obtained in patients with atrial fibrillation, which may be related to the relatively consistent length of systole rela tive to changes in heart rate.12 In contrast, bigeminy (trigeminy, etc.) often results in poor quality cine images, in that every other beat is activated differently, resulting in combining data from two different activation patterns. Some CMR systems pro vide arrhythmia rejection in an attempt to reduce these effects; however, use of these tools generally results in increased scan time because of rejection of cardiac cycles.

Respiratory Motion Artifacts Respiration is associated with marked displacement of the heart. Motion in the craniocaudal direction is on the order 146 Cardiovascular Magnetic Resonance

of 1 to 1.5 cm in normal individuals.13 This motion can result in image degradation with ghosting and blurring, particularly in patients with inconsistent respiratory pat terns. Strategies to reduce respiratory artifact include the use of sustained breath hold, presaturation of the high intensity signal from fat in the chest wall, and the use of free breathing with respiratory gating. Respiratory gating may be accomplished by using a thoracic bellows or by tracking the diaphragm position by using a navigator echo.14 These methods accept cardiac cycles only during some portion (typically end expiration) of the respiratory cycle. It can substantially improve image quality and can be useful in coronary imaging without breath hold15 and in patients with heart failure.16 All respiratory gating methods increase total scan time.

Metal Artifact Pieces of metal outside or inside the body alter the local magnetic field and can result in artifacts (Fig. 11 4C I). Patients must be screened carefully for the presence of metal, but despite vigilance, objects that are common in the hospital might still go with the patient into the scanner. Figure 11 4C shows an artifact related to a safety pin on the patient’s gown. Note that signal loss and distortion are pres ent in both the fast spin echo and gradient echo images. The severity of the artifact is larger in the gradient echo

B

C

D

E

F

G

H

11 NORMAL CARDIAC ANATOMY, ORIENTATION, AND FUNCTION

Figure 11 4

A

For legend see next page.

Cardiovascular Magnetic Resonance 147

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

I

J

K

Figure 11 4 A, Artifacts due to respiration and poor gating. In this gated spin echo CMR image there is mottling of the ventricular wall and loss of edge sharpness. B, The same image as in panel A, but with the window and level adjusted to accentuate the artifact. There are ghosts of the chest wall related to respiratory motion and additional artifact over the heart as a result of poor ECG gating. C, Metal artifact. The upper images were obtained with a safety pin present on the subject’s gown. The resultant signal void is very evident. The bottom row shows corresponding images after removal of the safety pin. Distortion from metal artifact is markedly more prominent in the gradient echo images (right column) than in the spin echo images (left column). D, Plain film X ray showing sternal wires (dashed arrow) and metallic coronary artery bypass graft (CABG) markers (solid arrow) in a patient with prior CABG surgery. E, Artifact from sternal wires (dashed arrow) and CABG markers (solid arrow) on T1 weighted spin echo CMR imaging. F, Signal voids (arrows) in two views of a bioprosthetic aortic valve replacement by breath hold cine steady state free precession imaging. The artifact results from the nonorganic struts. G, Metal in bileaflet mitral valve prosthesis produces signal voids (arrows). H, There is minimal artifact from the tricuspid (dashed arrow) and mitral (solid arrow) annuloplasty rings. I, Metal artifact from a coronary artery stent in the left anterior descending coronary artery (arrow) seen on a scout image. J, Chemical shift artifact. The image on the left is done with a relatively short signal acquisition time (wide bandwidth). The image on the right is done with a longer signal acquisition time (narrow bandwidth). This display accentuates the effect of the difference in chemical shift of water and fat, creating the artifactual space between the aortic wall at fat (arrow). K, Chemical shift artifact in echo planar imaging (EPI). In EPI, the chemical shift artifact occurs in the frequency encoding direction (right to left in these images). The image on the left is obtained by using a multishot EPI sequence with a relatively short EPI acquisition with each shot. The chemical shift artifact is indicated by the white line in the posterior chest wall. The image on the right is obtained by using fewer shots with a longer EPI acquisition. The chemical shift is larger, as indicated by the longer white line posteriorly. The image is degraded by superimposition of anterior subcutaneous fat onto the heart. This problem can be addressed by adding fat saturation to the sequence.

images, severely compromising interpretation of the RV and septum. Figure 11 4D and E shows the artifacts related to sternal wires and coronary artery bypass graft markers. Fig ure 11 4F shows the artifact related to a bioprosthetic aor tic valve while Figure 11 4G is a mechanical bileaflet mitral valve prosthesis. Figure 11 4H shows the minimal artifacts associated with mitral and tricuspid annuloplasty rings, and Figure 11 4I depicts artifact from a stent in the left anterior descending coronary artery. 148 Cardiovascular Magnetic Resonance

Chemical Shift Artifact A chemical shift artifact occurs because the hydrogen nuclei in fat experience a slightly different magnetic field than hydrogen nuclei in water because of the different local chemical environment (Fig. 11 4J,K).17 This process results in displacement of the fat signal in the frequency encoding direction relative to water and is accentuated with narrow bandwidth sequences, which can present a diagnostic

NORMAL CARDIAC SYSTOLIC AND DIASTOLIC FUNCTION The management of cardiovascular disease is critically dependent on the assessment of cardiac function. Thus, every cardiac imaging technique has been used to assess systolic and diastolic function. There is an extensive body of evidence to indicate that CMR provides highly accurate and reproducible assessments of global and regional car diac function, and CMR is often considered as the gold standard for the noninvasive evaluation of cardiac func tion,19 a standard by which other noninvasive methods are validated. An important consideration in determining ventricular function is the temporal resolution or frame rate of the cine CMR sequence. A frame rate of at least 25 frames/sec (i.e., temporal resolution of 40 msec/frame) is required to accu rately identify end systole. Historically, X ray left ventricu lography has been obtained at a frame rate of 30 frames/ sec (temporal resolution of 33 msec). The frame rate in echocardiography is dependent on the speed of ultrasound in the body and the distance of the heart from the trans ducer but typically is at least 30 frames/sec (temporal reso lution of 33 msec or better). With modern CMR scanners with high performance gradients, frame rates over 30 frames/sec for breath hold sequences are routine. Further, the apparent frame rate can be increased by using view sharing techniques that reconstruct intermediate images by combining recent, but previously acquired, k space data with selectively updated current data.20 Finally, partially parallel imaging methods (e.g., SENSE, SMASH, GRAPPA) are now routinely used in clinical imaging to decrease acquisition time and increase frame rate.21 These are described in greater detail in Chapter 3. Real time CMR approaches have inferior spatial and temporal (50 to 70 msec) resolution but may be preferred for patients with fre quent arrhythmias and in examining respiratory changes in interventricular septal motion (e.g., respiratory variation with tamponade or constriction).22 While the CMR cine loop appears to display a single car diac cycle, the image data are generally acquired over mul tiple heartbeats. As a result, image quality can be markedly degraded in the presence of arrhythmias (see above) or unreliable ECG triggering or gating. Cardiac function is typically assessed during breath holding, to minimize bulk cardiac translation; so if the goal is to assess the effect of respiration on chamber size or function (as in evaluating

for possible constriction, for example), then non breath hold real time methods22 may be more appropriate, as such techniques acquire images during a fraction of a single heartbeat and not as a composite over multiple cardiac cycles as with usual segmented k space sequences. How ever, the trade off with real time CMR is a decrease in both spatial and temporal resolution.

Left Ventricle Assessment of LV function includes global and regional function. Assessment of global LV function is based on measuring changes in chamber volumes. These changes can be estimated from unidimensional (linear) measure ments as in echocardiography, but with CMR, more accu rate measures of chamber volume can be made by using three dimensional methods. The two dimensional methods (e.g., the area length method or biplane angiographic for mulas) have no advantages over echocardiography and are not widely used.23–25 More accurate measures, particularly in deformed ventricles, which do not fit common geometric formula based models, can be obtained by using the sum mation of disks method. (In the cardiac imaging literature, this method is often referred to as the Simpson’s rule method. The mathematical Simpson’s rule is a fourth order polyno mial approximation for numerical integration;26 failure to distinguish between the mathematical and medical defini tions of Simpson’s rule leads to confusion between clini cians and engineers or medical physicists. The term modified Simpson’s rule is sometimes used in the echocardiog raphy literature to refer to a simplified summation of disks method.) With the summation of disks method, short axis images are obtained that span the entire ventricle; the cross sectional area in each slice is measured, multiplied by the slice thickness (and interslice gap if applicable), and summed over the entire ventricle.27 This approach is highly accurate and reproducible and is widely used both clinically and in research.26 Regional LV systolic function can be assessed both qual itatively (“eyeball” method similar to echocardiography) and quantitatively. As shown in Figure 11 5, the standard long axis, four chamber, two chamber, and short axis views can be mapped onto the 17 segment American Heart Asso ciation1 model for qualitative assessment of wall thickening. Wall thickening can also be determined quanti tatively by using centerline or other methods, and commer cial software is available for such analyses. Myocardial tagging methods28–30 can be used to quantify myocardial contractility using strain without the need to identify endo cardial or epicardial borders explicitly. However, software for tracking deformation of the tag lines over time none theless requires human interaction, prolonging analysis times. The harmonic phase (HARP) technique was pro posed to obviate the human postprocessing issue by com puting strain based on local spatial frequency of the tag lines.31 With appropriate bandpass filtering and subsequent transformation of the k space signal, strain can be extracted automatically. An alternative method, dis placement encoding with stimulated echoes (DENSE),32 combines a stimulated echo with bipolar gradient to encode displacement, from which strain can be calculated. These methods are detailed in Chapter 13. Cardiovascular Magnetic Resonance 149

11 NORMAL CARDIAC ANATOMY, ORIENTATION, AND FUNCTION

problem in spin echo imaging of patients with suspected aortic dissection.18 It can be addressed by using a wider bandwidth sequence or repeating the sequence with fre quency encoding in the alternate direction, which will result in changing the orientation of the artifact and thus help exclude the presence of an aortic dissection. In echo planar images, chemical shift effects lead to artifacts dis placed in the phase encoding direction. As shown in Fig ure 11 4K, this effect can be minimized using multishot EPI techniques. In single shot EPI with long acquisition times, the chemical shift effects can be quite large. For this reason, single shot echo planar images often employ fat sat uration to suppress this artifact.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Regional LV function

Ap-AS

Mid- Bs-AS AS

BsSept

Ap Ap-Inf/ Lat

Bs-AS

BsMidInf/Lat Inf/Lat

Bs-Ant

BsLat

Bs-Inf

Bs-Inf/Lat

MidAS

Mid-Ant

Figure 11 5 A, Cine CMR images in each of the standard planes typically used for scoring wall thickening and segmental function. The segments visualized correspond to those depicted in the standard American Heart Association (AHA) 17 segment model. B, AHA 17 segment model of the LV (2). Ant, anterior; Ap, apical; As, anteroseptal; Inf; inferior; Lat, lateral; Sept, septal; Bs, basal.

Apex Ap-Sept

Ap-Lat

Mid-Sept MidLat Bs-Lat

Bs-Sept

MidSept

MidLat

Mid-Inf

Inf/Lat

Ap-Sept

Ap-Ant

Ap-Inf

Ap-Lat

Apex Ap-Inf Mid-Inf Bs-Inf

A

Ap-Ant MidAnt Bs-Ant

B

Right Ventricle Historically, measurement of RV volumes has been largely qualitative. This situation is due to the lack of standard geometric models for the RV. A distinct advantage of CMR is that the same summation of disks approach may be read ily applied to the RV. Studies in ventricular casts have shown an excellent correlation between CMR and displace ment measurements.33

Stroke Volume Stroke volume is the amount of blood ejected from the heart with each cardiac cycle. It can be readily calculated by sub tracting the end systolic volume from the end diastolic vol ume. Multiplying the stroke volume by the heart rate yields the ventricular cardiac output, typically reported in liters per minute. Studies comparing CMR cardiac output with invasive thermodilution methods have shown good correlation.34–37 The cardiac output that is determined in this way from the stroke volume is the same as the cardiac output that is deter mined in the catheterization laboratory. In the setting of aortic or mitral regurgitation, however, part of this volume does not result in the net delivery of blood to the periphery. In this setting, the cardiac output based on the summation of disks LV stroke volume is greater than the forward flow (which can be measured by using flow techniques (see Chapter 37), but the regurgitant volume can be determined by subtracting the forward flow from the apparent stroke volume.38

Left Ventricular Mass CMR is similarly considered the gold standard for the assess ment of in vivo LV mass and is often used for validation of other methods.39 The mass of the LV wall can be estimated 150 Cardiovascular Magnetic Resonance

by measuring the volume of the myocardium and multiply ing it by the specific gravity of myocardium, 1.05 g/mL. CMR, in conjunction with a volumetric method such as summation of disks, provides accurate estimates of myocar dial mass over a broad range of heart sizes and deformed ventricles both in animals and in man.40–42 In cadaver heart studies, linear correlation analysis demonstrated a correla tion coefficient of 0.99 with a standard error of 6.8 g. Intraobserver, interobserver, and interstudy variability are excellent. Notably, LV mass that is determined by using lin ear measurements and cubed power geometric formulas, such as the echocardiographic Penn formula,43 generally overestimate volumetric mass even when the linear mea surements are made from CMR images.44 The volumetric summation of disks approach has been used to determine RV mass with good results as well.45–46

EFFECT OF IMAGING SEQUENCE AND MAGNETIC FIELD STRENGTH ON VENTRICULAR VOLUMES AND MASS AND IMPLICATIONS FOR REFERENCE STANDARDS Compared with cine gradient echo cine sequences, cine SSFP imaging provides superior contrast between blood pool and myocardium. Perhaps owing to this difference in delineation of endocardial contours, ventricular volumes and stroke volume by SSFP imaging are slightly but sys tematically greater than corresponding values measured using cine gradient echo cine methods.47 Conversely, LV

imaging sequences is used. Table 11 1 shows cine gradient echo LV reference (“normal”) values for healthy adults (free of any history of hypertension and cardiac disease) from the primarily Caucasian Framingham population and Dal las Heart Study populations. Table 11 2 shows cine SSFP based LV reference values from Framingham among adults strictly free of hypertension or clinical cardiovascular disease. Table 11 3 presents RV reference values, obtained by using cine gradient echo imaging, from the MESA cohort for differ ent ethnic groups. Higher field strength CMR systems (i.e., 3 Tesla[T]) offer signal to noise advantages over “conventional” 1.5 Tesla systems and are well established for neurologic imag ing. As the installed base of 3 T systems increases, there has been increasing interest in 3 Tesla CMR (see Chapter 13). With respect to cardiac size and function, the magnetic field strength of the scanner (i.e., 1.5 T versus 3 Tesla) does not appear to have any significant effect on measured ventricular volumes and mass, but the systematic difference

Table 11-1 Breath Hold Cine First Gradient Echo CMR Reference Values for the Left Ventricle Based on Community-Dwelling Adult Subjects, Aged 57  9 Years and Strictly Free of Cardiovascular Disease and Any History of Hypertension, Drawn from the Framingham Heart Study Offspring Cohort and Dallas Heart Study Raw LV EDV (mL) LV ESV (mL) LVM (g) LV EF LV EDV/HT (mL/m) LV ESV/HT (mL/m) LVM/HT (g/m) LV EDV/BSA (mL/m2) LV ESV/BSA (mL/m2) LVM/BSA (mLm2)

Men (n

63)

Men 95th Percentile

115 36 155 0.69 0.70 (DHS) 66 21 89 58 54 (DHS) 18 78

169 65 201 0.59* 0.55{ 94 36 114 80 31 95

Women (n

79)

84 25 103 0.70 0.75 (DHS) 52 16 64 50 49 (DHS) 15 61

Women 95th Percentile 117 41 134 0.60* 0.61{ 70 25 82 66 24 75

*Fifth percentile (lower limit) for ejection fraction. { Below the 2.5th percentile of the DHS. BSA, body surface area; DHS, Dallas Heart Study data; EDV, end-diastolic volume; EF, ejection fraction; ESV, end-systolic volume; HT, height; LV, left ventricle; LVM, left ventricular mass. Source: Adapted from CJ Salton et al. J Am Coll Cardiol. 2002;39:1055; Chung et al. Circulation. 2006;113:1597.

Table 11-2 Breath Hold Cine Steady-State Free Precession CMR Reference Values for the Left Ventricle Based on Community-Dwelling Adult Subjects, Aged 61  8 Years and Strictly Free of Cardiovascular Disease and Any History of Hypertension, Drawn from the Framingham Heart Study Offspring Cohort Raw LV EDV (mL) LV ESV (mL) LV SV (mL) LVM (g) LV EF LV EDV/HT (mL/m) LV ESV/HT (mL/m) LVM/HT (g/m) LV EDV/BSA (mL/m2) LV ESV/BSA (mL/m2) LVM/BSA (g/m2)

Men (n

239)

144  26 54  14 93  17 123  22 0.65  0.05 81  14 29  8 70  12 71  12 25  7 61  10

Men 95th Percentile 196 78 126 167 0.55* 109 43 93 95 38 79

Women (n

367)

106  19 35  10 71  11 81  14 0.67  0.05 65  10 22  6 50  8 61  9 20.5 47  7

Women 95th Percentile 143 55 93 109 0.57* 86 33 66 78 30 60

*Fifth percentile (lower limit) for ejection fraction. BSA, body surface area; EDV, end-diastolic volume; EF, ejection fraction; ESV, end-systolic volume; HT, height; LV, left ventricle; LVM, left ventricular mass; SV, stroke volume. Source: Adapted from Salton CJ et al. Circulation. 2006;114: II-669.

Cardiovascular Magnetic Resonance 151

11 NORMAL CARDIAC ANATOMY, ORIENTATION, AND FUNCTION

mass with SSFP CMR is smaller than CMR gradient echo mass. This has implications for what constitutes normal range ventricular volumes and mass, as the reference values must be imaging sequence specific. Gradient echo reference values for LV volumes and mass have been published based on data from the community dwelling Framingham Heart Study.48 The Multiethnic Study of Atherosclerosis (MESA) and Dallas Heart group have also published RV49 and LV50,51 reference values based on gradient echo imaging with a slightly higher LV ejection fraction demonstrated in women.51 Reports on SSFP based reference values for RV volumes and mass also demonstrate increased RV ejection fraction in women.52 Other publications have proposed reference values partitioned by age as well as sex, using SSFP imaging.53–55 These are useful, though the data are somewhat limited by small sample sizes or scanning of younger, athletically fit populations. Despite greater cavity volumes by SSFP as compared with gradient echo methods, LV ejection fraction is similar regardless of which of these

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Table 11-3 Breath Hold Cine Gradient Echo CMR Reference Values for the Right Ventricle Based on Community-Dwelling Adult Subjects, Aged 61  10 Years, Drawn from the Multi-Ethnic Study of Atherosclerosis (MESA) cohort Raw RV RV RV RV RV RV RV RV RV RV

EDV (mL) ESV (mL) SV (mL) EF EDV/HT (mL/m) ESV/HT (mL/m) SV/HT (mL/m) EDV/BSA (mL/m2) ESV/BSA (mL/m2) SV/BSA (mL/m2)

Men (n

219)

142  31 54  17 88  22 0.62  0.1 73  14 28  8 45  10 82  16 31  9 51  11

Men 95th Percentile 201 85 125 0.50* 98 43 63 101 48 70

Women (n

268)

Women 95th Percentile

110  24 35  13 75  18 0.69  0.1 65  11 21  7 44  8 69  14 22  8 47  10

155 57 106 0.58* 83 33 58 95 34 63

*Fifth percentile (lower limit) for ejection fraction. BSA, body surface area; EDV, end-diastolic volume; EF, ejection fraction; ESV, end-systolic volume; HT, height; LVM, left ventricular mass; RV, right ventricle; SV, stroke volume. Source: Adapted from Tandri H et al. Am J Cardiol. 2006;98:1660.

between gradient echo and SSFP imaging sequences is pre served regardless of field strength.56,57 It should be noted that SSFP imaging at 3 T is more subject to artifacts than imaging at 1.5 T; this results in part from greater sensitivity to field inhomogeneities, longer T1, and limitations on radiofrequency power deposition.58 Cardiac SSFP imaging at 3 T at present requires more careful sequence optimiza tion and understanding of how to identify and correct arti facts than CMR at 1.5 T.

Aortic Flow An important feature of CMR is its sensitivity to motion, and this feature can be harnessed to allow the measure ments of velocity and vessel flow without the constraints of Doppler methods. Although there is potential for arti facts, CMR allows accurate and reproducible measurement of vessel flow in vivo.37,59–61 The approach for the mea surement of aortic flow is straightforward. The coronal scout image is used, and the imaging plane is placed per pendicular to the direction of flow several centimeters above the aortic valve, typically at approximately the level of the bifurcation of the main pulmonary artery (Fig. 11 6A). The pulse sequence uses bipolar gradients oriented in the expected direction of flow, so tissues, such as blood, moving through the imaging plane accrue a nonzero net phase, while stationary tissues gain and lose phase equally for zero net phase change. The phase change corresponds to velocity. The velocity encoding value (VENC) should be chosen to be above the anticipated maximum velocity (approximately 1.5 to 2 m/sec in normal individuals and higher in patients with aortic valve disease). Setting the VENC too low can result in aliasing, in which phase change exceeds þ180 , so there is an abrupt discontinuity in apparent velocity, also known colloquially as wrap around. For example, a phase change of, for example, þ270 will be misinterpreted as a phase of 90 , owing to wrap around. The reconstructed images are typically presented as a set of magnitude images that are used to determine the cross sectional area of the aorta in each frame (Fig. 11 6B). There is also a set of phase encoded (velocity map) images in 152 Cardiovascular Magnetic Resonance

which the gray scale indicates the velocity of motion in each voxel (Fig. 11 6C). Velocity is measured at each voxel across the vessel, integrated over the cross sectional area of the vessel, and then integrated over the cardiac cycle (Fig. 11 6D). The integrated flow across the slice at each point in time can be graphed. Note that some retrograde flow is normal in the ascending aorta during early diastole, owing to closure of the aortic valve, diastolic ascent of the base of the heart, and diastolic coronary flow. The differ ence between LV stroke volume and aortic systolic forward flow can also be used to quantify mitral regurgitation.

Pulmonary Artery Flow Pulmonary artery flow can be measured by using techni ques similar to those for aortic flow. This is particularly valuable in evaluation of patients with left to right intracar diac shunting to determine the ratio of pulmonary to sys temic flow (Qp/Qs). Data demonstrated a very good correlation with invasive techniques.62,63 Depending on the pulmonary artery orientation, the location of the flow images can be planned from the axial and/or sagittal scout with a perpendicular plane positioned several centimeters distal to the pulmonary valve. The velocity profile is then integrated over the cross sectional area of the artery over time to determine the volume flow per cardiac cycle in a manner similar to that used in the ascending aorta. The dif ference between RV stroke volume and pulmonary systolic forward flow can also be used to quantify tricuspid regurgitation.

NORMAL VALVULAR FUNCTION Assessment of valve function involves evaluation of mor phology, motion, competence, and effects on ventricular function. Imaging cardiac valve morphology poses signifi cant problems for CMR.64,65 First, the normal valve is a thin, fibrous structure often less than 1 mm thick, leading to potential for partial volume effects. Second, it is

11 NORMAL CARDIAC ANATOMY, ORIENTATION, AND FUNCTION

B

A

C Scan time 00:02:51 PHILIPS MR 1.5 T SESSION INFORMATION: Q-Flow: AorticFlow (not validated).

Flux results (slice 1) ml/s 400 350

RESULTS SUMMARY: Heart rate : RR-interval :

300 250

60 bpm 1000 ms

(from heart rate)

ANALYSIS RESULTS: slice 1 Vessel 1

200 150 100 50 0 0 100 200 300 400 500 600 700 800 9001000

time (ms)

RR-interval: 1000 ms (from heart rate)

Stroke volume (ml) Forward flow vol. (ml) Backward flow vol. (ml) Regurgitant fract. (%) Abs. stroke volume (ml) Mean flux (ml/s) Stroke distance (cm) Mean velocity (cm/s)

73.5 73.8 0.3 0.4 74.1 73.5 9.5 9.5

Vessel 1, slice 1

Q-Flow: AorticFlow (not validated) Flux: Peak vel: 12.82 cm/s 7.19 ml/s Mean vel: 0.94 cm/s Max. vel: 12.82 cm/s Area: Min vel: –6.96 cm/s 7.65 cm2 Pixels: 557 pixels Vel stddev: 3.38 cm/s

300 cm/s 200 100 0 –100 –200

FFE/M SI 1 Ph 1/000 ms

D Figure 11 6

PCA/P SI 1 Ph 1/000 ms PCv FH 300 cm/s

–300 cm/s

For legend see next page. Cardiovascular Magnetic Resonance 153

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

Figure 11 6 A, Coronal scout image for measuring aortic flow. The white line indicates the anatomic position of a flow sequence. The plane is positioned at the level of the pulmonary artery well above the aortic valve and perpendicular to the walls of the aorta. B, Magnitude reconstruction from the flow sequence positioned in panel A. The imaging plane is transverse and positioned at the level of the bifurcation of the main pulmonary artery. The ascending aorta is seen anteriorly (arrow). C, Velocity map reconstruction from the same flow sequence as shown in panel B. The gray scale in this image indicates the velocity of motion toward the head as bright (solid arrow, ascending aorta) and away as dark (dashed arrow, descending aorta). D, Typical dataset for semiautomated analysis. The area of the ascending aorta is determined in each frame and the velocity over the area is integrated to calculate flow volume per frame. The top left subpanel shows a graph of flow volume over the cardiac cycle in the ascending aorta. The forward stroke volume, shown in the results listed in upper right subpanel, is calculated by integrating the flow over the cardiac cycle.

constantly in motion. In most cine CMR, the image is acquired over a number of cardiac cycles, requiring that the valve return to the same position with each cycle. With breath hold imaging, one can obtain very high resolution images, indicating that normal valve motion appears to be quite reproducible over a limited number of cycles when the effects of respiratory motion are removed. However, vegetations and other valve pathology are characterized by chaotic valve motion, which is less reproducible, leading to loss of signal and motion artifacts. Third, valve pathol ogy frequently involves fibrosis and calcification, both of which are characterized by loss of signal on CMR, making it difficult to detect, particularly in spin echo imaging, Last, regions of turbulence are associated with loss of signal on CMR, which may lead to overestimation of the extent of abnormality on gradient echo imaging. In spite of these concerns, CMR can be used to obtain high resolution images of valves using bright blood SSFP techniques and/

or black blood breath hold, double inversion recovery techniques.66 An example of a normal valve image obtained by using the latter technique is shown in Figure 11 7A. The LVOT view in Figure 11 7B shows aortic and mitral valves by SSFP imaging, and Figure 11 7C is an en face view of a normal trileaflet aortic valve; this frame from the SSFP cine loop shows the closed valve at end diastole. Figure 11 7D is another normal aortic valve open during early systole. Figure 11 7E is an en face view of a sclerotic, mildly ste nosed aortic valve during early systole. Figure 11 7F shows a dark signal void due to turbulent flow of aortic regurgitation. Imaging the valvular abnormality is only one part of the evaluation of the patient with valvular disease. It is essential to quantify the functional severity of the lesion and to determine its impact on ventricular size and function. Full details of assessing valvular abnormalities are given in Chapter 37, but some general comments are useful. As

A

B

C

D

Figure 11 7 A, Double inversion recovery spin echo CMR long axis image. The aortic valve leaflets are demonstrated (arrow). B, An LVOT view image acquired using breath hold steady state free precession (SSFP) CMR imaging. C, A normal trileaflet aortic valve depicted en face at end diastole (valve closed) by SSFP CMR imaging. D, An open, normal trileaflet aortic valve seen during early systole. continued 154 Cardiovascular Magnetic Resonance

F

Figure 11 7 cont’d E, An en face view of a sclerotic, mildly stenosed, trileaflet aortic valve also during early systole. Note the small opening and deformed leaflets as compared with panel D. F, Aortic regurgitation is visualized qualitatively in the dephasing jet (arrow) in this SSFP image.

was noted earlier, CMR is highly accurate in determining ventricular dimensions and volumes. In addition, phase contrast quantitative flow techniques provide effective means for determining velocity and blood flow. Thus, in addition to demonstrating that valvular disease is present, CMR can be used to quantify the degree of dysfunction and to determine its effect on ventricular size and function. Valve pressure gradients estimated by using CMR correlate well with ultrasound measurements.67,68 In addition, mea surements of aortic valve area and cardiac output by CMR agree with measurements of valve area at catheterization. Regurgitant jets are generally well visualized, owing to tur bulence (dephasing) creating dark regions of signal loss in gradient echo cines, but it is hazardous to estimate even qualitative severity of regurgitation from the size of the region of signal loss on cine CMR, as the apparent size of the jet is highly dependent on the details of the particular imaging sequence used. Specifically, although SSFP imaging generally provides excellent depiction of myocardial and valve anatomy, the size of dephasing jets is smaller (see Fig. 11 7F) than that with gradient echo sequences. Regur gitant jets can be underestimated or overlooked entirely based on visual assessment alone. Quantitative flow techni ques are more useful in determining the regurgitant volume, for example, by measuring retrograde flow in diastole in the aorta.69–71 Interestingly, measures of chamber volume and cardiac output by CMR in patients with atrial fibrillation agree well with invasive measures.12 Mitral regurgitation

has also been studied using quantitative measures.72,73 Inter rogation of the aortic and pulmonic valves is relatively straightforward using velocity encoding methods as described above, but the application of this method to the mitral and tricuspid valves is not necessarily straightforward, owing to the through plane translation (10 to 20 mm in a normal heart) of the base of the ventricles and thus the mitral and tricuspid annuli. Assessment of mitral regurgita tion is more reliably achieved by calculating the difference between stroke volume, by summation of disks method, and net aortic forward flow, by phase encoded velocity mapping, as long as there is no intracardiac shunt. The presence of a prosthetic valve is not a contraindica tion for CMR74 except in the case of probable valve dehis cence,75 although mechanical valves (see Fig. 11 4G) or the struts of bioprosthetic valves will results in artifacts (see Fig. 11 4F). Similarly, rings used for valve repair may pro duce local artifacts (see Fig. 11 4H).

CONCLUSION CMR can be used to clearly delineate cardiac anatomy and to assess function. As with any imaging technique, it is important to have a strategy for imaging and be familiar with the normal anatomy and potential artifacts. When this knowledge is in hand, CMR can be a very effective tool in the evaluation of patients with cardiovascular disease.

References 1. Cerqueira MD, Weissman NJ, Dilsizian V, et al. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: a statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation. 2002;105:539–542. 2. El-Khoury GY, Bergman RA, Montgomery WJ. Sectional Anatomy by MRI. New York: Churchill Livingstone; 1995. 3. Arai AE, Epstein FH, Bove KE, Wolff SD. Visualization of aortic valve leaflets using black blood MRI. J Magn Reson Imaging. 1999;10:771–777.

4. Fayad ZA, Fuster V, Fallon JT, et al. Noninvasive in vivo human coronary artery lumen and wall imaging using black-blood magnetic resonance imaging. Circulation. 2000;102:506–510. 5. Botnar RM, Stuber M, Kissinger KV, et al. Non-invasive coronary vessel wall and plaque imaging with magnetic resonance imaging. Circulation. 2000;102:2582–2587. 6. Thiele H, Nagel E, Paetsch I, et al. Functional cardiac MR imaging with steady-state free precession (SSFP) significantly improves endocardial border delineation without contrast agents. J Magn Reson Imag. 2001;14:362–367. Cardiovascular Magnetic Resonance 155

11 NORMAL CARDIAC ANATOMY, ORIENTATION, AND FUNCTION

E

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

7. Jarvinen VM, Kupari MM, Hekali PF, Poutanen VP. Right atrial MR imaging studies of cadaveric atrial casts and comparison with right and left atrial volumes and function in healthy subjects. Radiology. 1994;191:137–142. 8. Mortele KJ, Mergo PJ, Williams WF. Lipomatous hypertrophy of the atrial septum: diagnosis with fat suppressed MR imaging. J Magn Reson Imaging. 1998;8:1172–1174. 9. Shirani J, Roberts WC. Clinical, electrocardiographic and morphologic features of massive fatty deposits (“lipomatous hypertrophy”) in the atrial septum. J Am Coll Cardiol. 1993;22:226–238. 10. Noma M, Kikuchi Y. Images in cardiovascular medicine: lipomatous hypertrophy of the atrial septum. Circulation. 1999;100:684. 11. Fischer SE, Wickline SA, Lorenz CH. Novel real-time R-wave detection algorithm based on the vectorcardiogram for accurate gated magnetic resonance acquisitions. Magn Reson Med. 1999;42:361–370. 12. Hundley WG, Meshack BM, Willett DL, et al. Comparison of quantitation of left ventricular volumes, ejection fraction and cardiac output in patients with atrial fibrillation by cine magnetic resonance imaging: versus invasive measurements. Am J Cardiol. 1996;78:1119–1123. 13. Danias PG, Stuber M, Botnar RM, et al. Relationship between motion of coronary arteries and diaphragm during free breathing: lessons from real-time MR imaging. AJR Am J Roentgenol. 1999;172:1061–1065. 14. Taylor AM, Jhooti P, Wiesmann F, et al. Magnetic resonance navigatorecho monitoring of temporal changes in diaphragm position: implications for magnetic resonance coronary angiography. J Magn Reson Imaging. 1997;7:629–636. 15. Botnar RM, Stuber M, Danias PG, et al. Improved coronary artery definition with T2-weighted, free-breathing, three-dimensional coronary MRA. Circulation. 1999;99:3139–3148. 16. Bellenger NG, Gatehouse PD, Rajappan K, et al. Left ventricular quantification in heart failure by cardiovascular MR using prospective respiratory navigator gating: comparison with breath-hold acquisition. J Magn Reson Imaging. 2000;11:411–417. 17. Babcock EE, Brateman L, Weinreb JC, et al. Edge artifacts in MR images: chemical shift effect. J Comput Assist Tomogr. 1985;9:252–257. 18. Lotan CS, Cranney GB, Doyle M, Pohost GM. Fatshift artifact simulating aortic dissection on MR images. AJR Am J Roentgenol. 1989;152: 385–386. 19. Mor-Avi V, Jenkins C, Kuhl HP, et al. Real-time 3-dimensional echocardiographic quantification of left ventricular volumes: multicenter study for validation with magnetic resonance imaging and investigation of sources of error. J Am Coll Cardiol Img. 2008;1:413–423. 20. Foo TK, Bernstein MA, Aisen AM, et al. Improved ejection fraction and flow velocity estimates with use of view sharing and uniform repetition time excitation with fast cardiac techniques. Radiology. 1995;195:471–478. 21. Heidemann RM, Ozsarlak O, Parizel PM, et al. A brief review of parallel magnetic resonance imaging. Eur Radiol. 2003;13:2323–2327. 22. Yang PC, Kerr AB, Liu AC, et al. New real-time interactive cardiac magnetic resonance imaging system complements echocardiography. J Am Coll Cardiol. 1998;32:2049–2056. 23. Dodge HT, Sandler H, Ballew DW, et al. The use of biplane angiocardiography for the measurement of left ventricular volume in man. Am Heart J. 1960;60:762–776. 24. Sandler H, Dodge HT. The use of single plane angiocardiograms for the calculation of left ventricular volume in man. Am Heart J. 1968;75:325–334. 25. Chuang ML, Hibberd MG, Beaudin RA, et al. Importance of imaging method over imaging modality in noninvasive determination of left ventricular volumes and ejection fraction: assessment by two and three dimensional echocardiography and magnetic resonance imaging. J Am Coll Cardiol. 2000;35:477–484. 26. Kaplan W. Advanced Calculus. New York: Addison-Wesley; 2002. 27. Strohm O, Schulz-Menger J, Pilz B, et al. Measurement of left ventricular dimensions and function in patients with dilated cardiomyopathy. J Magn Reson Imaging. 2001;13:367–371. 28. Zerhouni EA, Parish DM, Rogers WJ, Yang A, Shapiro EP. Human heart: tagging with MR imaging—a method for noninvasive assessment of myocardial motion. Radiology. 1988;169:59–63. 29. Axel L, Dougherty L. MR imaging of motion with spatial modulation of magnetization. Radiology. 1989;171:841–845. 30. Fischer SE, McKinnon GC, Maier SE, Boesiger P. Improved myocardial tagging contrast. Magn Reson Med. 1993;30:191–200. 31. Osman NF, Kerwin WS, McVeigh ER, Prince JL. Cardiac motion tracking using CINE harmonic phase (HARP) magnetic resonance imaging. Magn Reson Med. 1999;42:1048–1060.

156 Cardiovascular Magnetic Resonance

32. Aletras AH, Ding S, Balaban RS, Wen H. DENSE: Displacement encoding with stimulated echoes in cardiac functional MRI. J Magn Reson. 1999;137:247–252. 33. Boxt LM, Katz J, Kolb T, et al. Direct quantitation of right and left ventricular volumes with nuclear magnetic resonance imaging in patients with primary pulmonary hypertension. J Am Coll Cardiol. 1992;19:1508–1515. 34. Utz JA, Herfkens RJ, Heinsimer JA, et al. Cine MR determination of left ventricular ejection fraction. AJR Am J Roentgenol. 1987;148:839–843. 35. Culham JAG, Vince DJ. Cardiac output by MR imaging: an experimental study comparing right ventricle and left ventricle with thermodilution. J Can Assoc Radiol. 1988;39:247–249. 36. Hunter GJ, Hamberg LM, Weisskoff RM, et al. Measurement of stroke volume and cardiac output within a single breath hold with echoplanar MR imaging. J Magn Reson Imaging. 1994;4:51–58. 37. Hundley WG, Li HF, Hillis LD, et al. Quantitation of cardiac output with velocity-encoded, phase-difference magnetic resonance imaging. Am J Cardiol. 1995;75:1250–1255. 38. Hundley WG, Li HF, Willard JE, et al. Magnetic resonance imaging assessment of the severity of mitral regurgitation: a comparison with invasive techniques. Circulation. 1995;92:1151–1158. 39. Takeuchi M, Nishikage T, Mor-Avi V, et al. Measurement of left ventricular mass by real-time three-dimensional echocardiography: validation against magnetic resonance and comparison with twodimensional and M-mode measurements. J Am Soc Echo. 2008;21: 1001–1005. 40. Franco F, Dubois SK, Peshock RM, Shohet RV. Magnetic resonance imaging accurately estimates cardiac mass in a transgenic mouse model of cardiac hypertrophy. Am J Physiol. 1998;43:H679–H683. 41. Maddahi J, Crues J, Berman DS, et al. Noninvasive quantification of left ventricular myocardial mass by gated proton nuclear magnetic resonance imaging. J Am Coll Cardiol. 1987;10:682–692. 42. Katz J, Milliken MC, Stray-Gundersen J, et al. Estimation of human myocardial mass with MR imaging. Radiology. 1988;169:495–498. 43. Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation. 1977;55:613–618. 44. Bottini PB, Carr AA, Prisant LM, et al. Magnetic resonance imaging compared to echocardiography to assess left ventricular mass in the hypertensive patient. Am J Hyperten. 1995;8:221–228. 45. Katz J, Whang J, Boxt LM, Barst RJ. Estimation of right ventricular mass in normal subjects and in patients with primary pulmonary hypertension by nuclear magnetic resonance imaging. J Am Coll Cardiol. 1993;21:1475–1481. 46. McDonald KM, Parrish T, Wennberg P, et al. Rapid, accurate and simultaneous noninvasive assessment of right and left ventricular mass with nuclear magnetic resonance imaging using the snapshot gradient method. J Am Coll Cardiol. 1992;19:1601–1607. 47. Moon JCC, Lorenz CH, Francis JM, Smith GC, Pennell DJ. Breathhold FLASH and FISP cardiovascular MR imaging: left ventricular volume differences and reproducibility. Radiology. 2002;223:789–797. 48. Salton CJ, Chuang ML, O’Donnell CJ, et al. Gender differences and normal left ventricular anatomy in an adult population free of hypertension: a cardiovascular magnetic resonance study of the Framingham Heart Study offspring cohort. J Am Coll Cardiol. 2002;39:1055–1060. 49. Tandri H, Daya SK, Nasir K, et al. Normal reference values for the adult right ventricle by magnetic resonance imaging. Am J Cardiol. 2006;98:1660–1664. 50. Natori S, Lai S, Finn JP, et al. Cardiovascular function in multi-ethnic study of atherosclerosis: normal values by age, sex and ethnicity. AJR Am J Roengtenol. 2006;186:S357–S365. 51. Chung AK, Das SR, Leonard D, et al. Women have higher left ventricular ejection fractions than men independent of differences in left ventricular volume: the Dallas Heart Study. Circulation. 2006;113: 1597–1604. 52. Salton CJ, Gona P, Chuang ML, et al. Gender-specific reference values for left ventricular anatomy and systolic function using modern cardiovascular magnetic resonance methods in a longitudinally followed normotensive population: the Framingham Heart Study. Circulation. 2006;114(suppl S):669. 53. George KP, Birch KM, Pennell DJ, Myerson SG. Magnetic-resonanceimaging-derived indices for the normalization of left ventricular morphology by body size. Magn Reson Imaging. 2009;27:207–213. 54. Maceira AM, Prasad SK, Khan M, et al. Reference right ventricular systolic and diastolic function normalized to age, gender and body surface area from steady-state free precession cardiovascular magnetic resonance. Eur Heart J. 2006;27:2879–2888.

65. Globits S, Higgins CB. Assessment of valvular heart disease by magnetic resonance imaging. Am Heart J. 1995;129:369. 66. Arai AE, Epstein FH, Bove KE, Wolff SD. Visualization of aortic valve leaflets using black blood MRI. J Magn Reson Imaging. 1999;10: 771–777. 67. Kilner PJ, Firmin DN, Rees RSO, et al. Valve and great vessel stenosis: assessment with magnetic resonance jet velocity mapping. Radiology. 1991;178:229–235. 68. Eichenberger AC, Jenni R, von Shulthess GK. Aortic valve pressure gradients in patients with aortic valve stenosis: quantification with velocityencoded cine MR imaging. AJR Am J Roentgenol. 1993;160:971–977. 69. Dulce MC, Mostbeck GH, O’Sullivan M, et al. Severity of aortic regurgitation: interstudy reproducibility of measurements with velocityencoded cine MR imaging. Radiology. 1992;185:235–240. 70. Honda N, Machida K, Hashimoto M, et al. Aortic regurgitation: quantitation with MR imaging velocity mapping. Radiology. 1993; 185:189–194. 71. Walker PG, Oyre S, Pedersen EM, et al. A new control volume method for calculating valvular regurgitation. Circulation. 1995;92:579–586. 72. Fujita N, Chazouilleres AE, Hartiala JJ, et al. Quantification of mitral regurgitation by velocity-encoded cine nuclear magnetic resonance imaging. J Am Coll Cardiol. 1994;23:951–958. 73. Hundley WG, Li HF, Willard JE, et al. Magnetic resonance imaging assessment of the severity of mitral regurgitation: a comparison with invasive techniques. Circulation. 1995;92:1151–1158. 74. Deutsch HJ, Bachmann R, Sechtern U, et al. Regurgitant flow in cardiac valve prostheses: diagnostic value of gradient echo nuclear magnetic resonance imaging in reference to transesophageal twodimensional color doppler echocardiography. J Am Coll Cardiol. 1992;19:1500–1507. 75. Shellock FG, Kanal E. Magnetic Resonance: Bioeffects, Safety, and Patient Management. 2nd ed. Philadelphia: Lippincott-Raven; 1996.

Cardiovascular Magnetic Resonance 157

11 NORMAL CARDIAC ANATOMY, ORIENTATION, AND FUNCTION

55. Maceira AM, Prasad SK, Khan M, Pennell DJ. Normalized left ventricular systolic and diastolic function by steady state free precession cardiovascular magnetic resonance. J Cardiovasc Magn Reson. 2006;8:415–426. 56. Hudsmith LE, Petersen SE, Tyler DJ, et al. Determination of cardiac volumes and mass with FLASH and SSFP cine sequences at 1.5 vs. 3 Tesla: a validation study. J Magn Reson Imag. 2006;24:312–318. 57. Grothues F, Boenigk H, Graessner J, Kanowski M, Klein HU. Balanced steady-state free precession vs. segmented fast low-angle shot for the evaluation of ventricular volumes, mass and function and 3 Tesla. J Magn Reson Imag. 2007;26:392–400. 58. Scha¨r M, Kozerke S, Fischer SE, Boesiger P. Cardiac SSFP imaging at 3 Tesla. Magn Reson Med. 2004;51:799–806. 59. Stahlberg F, Sondergaard L, Thomsen C, Henriksen O. Quantification of complex flow using MR phase imaging: a study of parameters influencing the phase/velocity relation. Magn Reson Imaging. 1992;10: 13–23. 60. Tang C, Blatter DD, Parker DL. Accuracy of phase-contrast flow measurements in the presence of partial volume effects. J Magn Reson Imaging. 1993;3:377–385. 61. Buonocore MH, Bogren H. Factors influencing the accuracy and precision of velocity-encoded phase imaging. Magn Reson Med. 1992;26:141–154. 62. Brenner LD, Caputo GR, Mostbeck G, et al. Quantification of left to right atrial shunts with velocity-encoded cine nuclear magnetic resonance imaging. J Am Coll Cardiol. 1992;20:1246–1250. 63. Hundley WG, Li HE, Lange RA, et al. Assessment of left-to-right intracardiac shunting by velocity-encoded, phase-difference magnetic resonance imaging: a comparison with oximetric and indicator dilution techniques. Circulation. 1995;91:2955–2960. 64. Duerinckx AJ, Higgins CB. Valvular heart disease. Radiol Clin North Am. 1994;32:613–630.

BASIC PRINCIPLES OF CARDIOVASCULAR MAGNETIC RESONANCE

CHAPTER 12

Comprehensive Cardiovascular Magnetic Resonance in Coronary Artery Disease Sven Plein

Cardiovascular magnetic resonance (CMR) imaging is a highly versatile imaging modality, capable of assessing sev eral manifestations of coronary artery disease (CAD).1,2 Furthermore, CMR data are acquired in freely definable and highly reproducible imaging planes, so the different components of a CMR study can be readily registered and correlated. These unique characteristics of CMR have led to the expectation that the technology will provide a comprehensive assessment of CAD in a single imaging study. However, until recently, the realization of such compre hensive CMR studies has been hindered by long image acqui sition times and a lack of studies demonstrating the clinical effectiveness of CMR. In the past years, these limitations have been largely overcome. Several CMR methods have now been shown to provide clinically relevant and often complemen tary information in CAD patients.3–12 At the same time, advances in technology and acquisition techniques have shortened scan durations for most CMR methods13–15 More data can therefore be acquired within tolerable examination times, and protocols integrating several CMR methods have become feasible. It is expected that such comprehensive pro tocols will further improve the clinical utility and diagnostic accuracy of CMR in the management of patients with CAD. In this chapter, the most relevant CMR methods for the assessment of CAD are briefly discussed, with particular emphasis on their acquisition times and suitability for a single session CMR study. This is followed by a discussion of current comprehensive protocols for typical clinical scenarios.

CARDIOVASCULAR MAGNETIC RESONANCE IN CORONARY ARTERY DISEASE The following measurements are of relevance to the man agement of most patients with CAD: cardiac morphology, global and regional myocardial function, myocardial ische mia, viability status of dysfunctional myocardium, and the presence of flow limiting coronary stenoses. CMR can pro vide data in all of these aspects of CAD, and acquisition times have been reduced for most of the relevant CMR methods (Table 12 1). 158 Cardiovascular Magnetic Resonance

Morphology and Function CMR can accurately assess cardiac morphology, global and regional cardiac function in normal as well as deformed ven tricles.16,17 Cine imaging forms an essential component of any CMR study in CAD. With the introduction of steady state free precession (SSFP) pulse sequences, data acquisi tion times have been reduced to around 5 minutes for acqui sition of a stack of 10 to 12 two dimensional (2D) short axis sections.13,14 Combined with spatial or spatiotemporal undersampling methods, acquisition times can be reduced further, albeit at the expense of signal to noise ratio. These acceleration methods can also be employed to achieve either 2D real time or three dimensional (3D) whole heart acquisi tion.15,18 Combined with myocardial tagging, cine CMR imaging can provide a further detailed assessment of regional cardiac function in similar acquisition times.19

Ischemia Myocardial ischemia as the principal manifestation of CAD can be detected by two different CMR techniques: contrast enhanced myocardial perfusion and dobutamine stress MR (DSMR). The place of myocardial perfusion imaging in a compre hensive CMR study, and its duration, are determined mainly by the need for patient preparation and monitoring, the length of the administration of the stress agent (3 to 4 min utes for adenosine as the preferred pharmacologic stress agent),20,21 and some time to allow for hemodynamics and symptoms to recover after the study. Approximately 10 min utes should therefore be considered for an adenosine stress perfusion acquisition. With the use of a variety of acquisition protocols, the accuracy of CMR in the published literature is at least comparable to that of nuclear scintigraphy.3,8,10,22–25 Dobutamine stress CMR (DCMR) detects inducible wall motion abnormalities with accuracy at least equal to that of stress echocardiography (See Chapter 15.)9,26,27 The duration of a DCMR study is determined primarily by the need to increase the dose of dobutamine incrementally in steps of 5 to 10 mg/kg/min up to 30 or 40 mg/kg/min. Data acquisition itself, usually using 2D SSFP pulse sequences and acquiring several orthogonal views, is rapidly accomplished at each stress level. Recently, 3D cine imaging has been proposed to

Indication

CMR method

Pulse sequence

Wall motion Ischemia

Cine function or tagging DSMR Perfusion LGE DSMR Angiography

SSFP SSFP T1 GRE IR GRE SSFP T2 prep, GRE or SSFP, navigator or breath hold Velocity encoded GRE

Viability Coronary imaging

Flow

Acquisition time/slice

Total acquisition time

5 15 sec 5 15 sec 40 mm Hg) or symptomatic. To provide a marker of goodness of pharmacologic stress testing, the peak rate pressure product (heart rate  systolic blood pressure) can be calculated, as this pro vides a reasonable estimate of myocardial workload (a heart rate pressure product greater than 20,000 mm Hg/min usu ally indicating an adequate hemodynamic response). DOBUTAMINE: STRESS-INDUCIBLE WALL MOTION ABNORMALITIES Over the years, several investigators have consistently reported on the high diagnostic value of dobutamine stress inducible wall motion abnormalities for the detection of significant epicardial coronary disease (>50% luminal narrowing) with an overall diagnostic accuracy of around 86% (range: 83% to 90%). In general, in patients who are amenable to DCMR wall motion imaging, the method has 232 Cardiovascular Magnetic Resonance

the potential to become the preferred imaging technique for the detection of inducible wall motion abnormalities and can be regarded as the method of first choice in patients with limited echocardiographic image quality.23,24 DOBUTAMINE: STRESS-INDUCIBLE PERFUSION ABNORMALITIES Dobutamine in high doses (20 to 40 mg/kg/min) increases the three main determinants of myocardial oxygen demand (i.e., heart rate, systolic blood pressure, and myocardial contractility), thereby eliciting a secondary increase in myo cardial blood flow and provoking myocardial perfusion abnormalities. The flow increase (twofold to threefold base line values) is reportedly less than that elicited by adeno sine. However, dobutamine stress myocardial perfusion imaging using radioactive isotopes including 201Tl, 99m TC sestamibi, and 99mTC tetrofosmin has been success fully performed. The overall sensitivity, specificity, and accuracy of dobutamine stress nuclear studies were 85%, 72%, and 83%, respectively.25 Data on DCMR perfusion imaging is relatively limited; in a feasibility study, Al Saadi and colleagues26 showed that a quantitatively derived myocardial perfusion reserve index differentiated ischemic and nonischemic myocardial seg ments reaching a sensitivity, specificity, and diagnostic accu racy of 81%, 73%, and 77%, respectively. In this study, however, stress imaging has been done at low dose dobuta mine (20 mg/kg/min level) only to avoid the marked increase of heart rate that occurs at higher doses. The main obstacle for performing myocardial perfusion imaging at high dose dobutamine levels has been the inability of previous perfusion CMR techniques to acquire multiple slices at high heart rates (i.e., >120 beats per min ute) while preserving a high spatial resolution. With the more recent development of parallel imaging techniques (e.g., sensitivity encoding, SENSE),27–29 dynamic, multi slice CMR perfusion imaging has become much faster and therefore can be used to assess myocardial perfusion abnormalities occurring under high dose dobutamine stress (Fig.17 1C). Though no large scale study on the diagnostic value of dobutamine stress CMR perfusion imag ing has been published yet, we do perform CMR perfusion imaging as an integral part of dobutamine stress testing for wall motion assessment (as exemplified in Fig. 17 2C). In our experience, the diagnostic accuracy of high dose dobu tamine myocardial perfusion CMR imaging is comparable to the accuracy of high dose dobutamine stress nuclear studies. Hence, DCMR might be used as an alternative pharmacologic stress test in patients with contraindications to vasodilator stress.

Safety Aspects Adenosine The vasodilatory effect of adenosine may result in a mild to moderate reduction in systolic, diastolic, and mean arterial blood pressure (15 min)

Adenosine (140 μg/kg/min)

≈31 min

≈45 min

Stop

Cine

B 0 min

Cine

Rest

Start Survey

Perfusion ≈30 min

12 min

Stop

Start

Stop

Start Survey

Dobutamine (up to 40 μg/kg/min, +/− atropine)

Rest

Perfusion 6 min

10 min

Break (10 min) 12 min

Perfusion

≈22 min

Break (10 min)

Scar imaging

≈23 min

≈40 min

Survey

C 0 min

Cine

Perfusion

Stop

Start

Dobutamine (up to 40 μg/kg/min, +/− atropine )

Cine (repetitive) 6 min

9 min

12 min

Perfusion 15 min

18 min

Break (10 min) Scar imaging 19 min

≈30 min

≈35 min

Figure 17 1 Time course of stress testing (dobutamine and adenosine administration) and corresponding CMR (cine, dynamic perfusion, and late gadolinium enhancement imaging). A, Combined adenosine stress perfusion and high dose dobutamine/atropine stress CMR. B, Assessment of rest and adenosine stress perfusion CMR. Optionally, stress testing may be followed by late gadolinium enhancement CMR for myocardial scar detection; if done so, rest perfusion might be spared, and the observer needs to rely on judging the fixed portion of a perfusion deficit from scar extent only. C, Assessment of high dose dobutamine/atropine stress perfusion CMR at rest and during maximal stress in combination with repetitive cine imaging during graded dobutamine infusion. Though dobutamine stress testing is already capable of accurately diagnosing viable myocardium, stress testing may be followed by scar imaging.

Since adenosine exerts a direct depressant effect on the sinoatrial and atrioventricular (AV) nodes, the transient occurrence of first , second , and third degree AV block and sinus bradycardia has been reported in 2.9%, 2.6%, and 0.8% of patients, respectively.30 Also, adenosine can cause significant hypotension. Patients with an intact baro receptor reflex are able to maintain blood pressure in response to adenosine by increasing cardiac output and heart rate. Adenosine can also cause a paradoxical increase in systolic and diastolic blood pressure, which develops mostly in individuals with significant left ventricular hyper trophy, but these effects are usually transient and resolve spontaneously. Because adenosine is a respiratory stimulant primarily through activation of carotid body chemorecep tors, intravenous administration showed increases in minute ventilation and a reduction in arterial pCO2, resulting in respiratory alkalosis.12 Approximately 20% to

30% of patients complain of dyspnea and an urge to breathe deeply during adenosine infusion. Because of the above reported adverse effects, a number of studies have been carried out investigating the safety of intravenous adenosine infusions in different diagnostic modalities of cardiac imaging.30,31 So far, there is evidence that was accumulated in over 10,500 patients who were studied with thallium radionu clide imaging, echocardiography, SPECT, and CMR that shows that pharmacologic stress with adenosine presents a safe method of acquiring stress imaging data. Safety of an adenosine infusion at 140 mg/kg/min was evaluated during radionuclide imaging of 9256 consecutive patients.30 The infusion protocol was completed in 80% of patients, required dose reduction in 13%, and was termi nated early in 7%. Interpretable imaging studies were obtained in 98.7% of patients, and 0.8% of patients Cardiovascular Magnetic Resonance 233

17 COMPARISON OF PERFUSION AND WALL MOTION CARDIOVASCULAR MAGNETIC RESONANCE IMAGING

Adenosine (140 μg/kg/min)

Rest

508) without alteration in the adenosine infusion. There were no sustained episodes of AV block.

Dobutamine The safety of high dose DCMR has been investigated. In a rou tine clinical setting, high dose DCMR was attempted on 1075

DOBUTAMINE WALL MOTION Low dose

Max

End diastole

Rest

End systole

ISCHEMIC HEART DISEASE

received aminophylline. Minor and well tolerated side effects were reported in 81.1% of patients. There were no deaths, one myocardial infarction, seven episodes of severe bronchospasm, and one episode of pulmonary edema. Transient AV node block occurred in 706 patients (first degree in 256, second degree in 378, and third degree in 72) and resolved spontaneously in most patients (n ¼

ADENOSINE PERFUSION Rest

SCAR Stress

A Figure 17 2 Stress CMR examples corresponding to the proposed imaging protocols in Figure 17 1A C. A, Dobutamine stress CMR showed an extensive inducible wall motion abnormality at maximum stress level (anterior and septal segments, white arrows). Adenosine stress CMR demonstrated an extensive inducible perfusion deficit; note that the inducible perfusion deficit covered the inferolateral segment as well (white arrowheads), but no visually assessable wall motion abnormality is demonstrated (segmental mismatch). Since myocardial scar was not present, the absence of resting perfusion abnormalities could be confirmed. 234 Cardiovascular Magnetic Resonance

Stress

B DOBUTAMINE WALL MOTION Low dose

Max

End systole

End diastole

Rest

ADENOSINE PERFUSION Rest

SCAR Stress

C Figure 17 2 cont’d B, Perfusion CMR at rest was found to be normal; under adenosine stress, though, a subendocardial perfusion deficit of varying transmurality is demonstrated (inferior and inferolateral segment, white arrows). Myocardial scarring was absent. C, Dobutamine stress CMR detected an inducible wall motion abnormality of the inferolateral segment (white arrows). Dobutamine stress perfusion CMR demonstrated an inducible perfusion deficit in identical location. Myocardial scarring was absent. Cardiovascular Magnetic Resonance 235

17 COMPARISON OF PERFUSION AND WALL MOTION CARDIOVASCULAR MAGNETIC RESONANCE IMAGING

SCAR

ADENOSINE PERFUSION Rest

ISCHEMIC HEART DISEASE

occasions in 1035 consecutive patients using the graded stan dard dobutamine protocol of 10, 20, 30, and 40 mg/kg/min dobutamine for 3 minutes each plus atropine if needed to reach target heart rate. A high number of diagnostic examina tions was found (89.5%). Reasons for nondiagnostic tests included failure of ECG triggering (n ¼ 4), suboptimal image quality (n ¼ 6), submaximal stress (n ¼ 21), and limiting side effects (n ¼ 74), including chest pain, dyspnea, and nausea. Atrial fibrillation was provoked in five patients, and nonsus tained atrial fibrillation was provoked in 16. Only one patient suffered sustained ventricular tachycardia and underwent suc cessful emergent defibrillation, while no cases of death or myocardial infarction occurred over the observed 5 year period. The patients who were being examined were typical of those evaluated for coronary disease, and over half had ischemia induced at the time of dobutamine stress, thereby closely reflecting the clinical reality. This study demonstrated that the technique can be undertaken with a safety profile vir tually identical to that previously reported for dobutamine stress echocardiography.32–35

Drug Interactions

Adenosine infusion should be exercised with caution in patients with chronic lung diseases not associated with bronch oconstriction (emphysema, bronchitis, etc.) and should be avoided in patients with bronchoconstriction or bronchospasm (e.g., allergic asthma). If a patient develops severe respiratory difficulties, the adenosine infusion is ceased immediately, and theophylline can be applied intravenously to rapidly dissolve bronchoconstriction (Tables 17 2 and 17 3).

Dobutamine In general, the dobutamine infusion is to be stopped if tar get heart rate defined as age predicted submaximal heart rate ([220  age]  0.85) is reached. Additional termina tion criteria are given in Table 17 3. Contraindications include acute coronary syndromes, severe aortic stenosis, hypertrophic cardiomyopathy, uncontrolled hypertension, and uncontrolled heart failure. If atropine is needed to reach the target heart rate, its contraindications must be taken into consideration (i.e., narrow angle glaucoma, myasthenia gravis, obstructive uropathy, or obstructive gastrointestinal diseases; see also Tables 17 2 and 17 3).

Dobutamine

Practicability

It is very important that the patient be instructed that beta blockers be withheld for at least 24 hours prior to the examination, since beta blockers reduce the inotropic and chronotropic effects of dobutamine and thus reduce the sensitivity of the stress test. Preferably, other antianginal medication should be withdrawn as well.

A detailed description of the cognitive and training skills neces sary for competent performance of pharmacologic stress testing has been released by the AHA/ACC36; in general, these skills do apply to CMR stress testing as well. However, CMR has some unique complexities resulting from the static magnetic field and the examination of a patient in an enclosed magnet.

Adenosine The patient should refrain from caffeine containing beverages or food (tea, coffee, chocolate, etc.), smoking, and any antian ginal medication 24 hours prior to the examination. Caffeine and other methylxanthine derivatives act like theophylline, which is the clinically used antidote of adenosine.

Contraindications and Termination Criteria Adenosine Adenosine should be used with caution in patients with preexisting AV block or bundle branch block and should be avoided in patients with high grade AV block or sinus node dysfunction. Adenosine should be used with caution if a patient is receiving any medication that already depresses the sinus node and/or AV conduction (e.g., beta blockers, cal cium channel blockers, cardiac glycosides). Adenosine needs to be discontinued in patients who develop persistent or symptomatic high grade block or a significant drop in systolic blood pressure (>20 mm Hg). The drug should be discontinued in case of persistent or symptomatic hypotension. Also, adenosine should be used with caution in patients with autonomic dysfunction, stenotic valvular heart disease, pericar ditis and pericardial effusion, high grade carotid artery stenoses and cerebrovascular insufficiency or uncorrected hypovolemia. 236 Cardiovascular Magnetic Resonance

Monitoring Monitoring needs are similar for the different stressors. Heart rate and rhythm need to be registered throughout

Table 17-2 Contraindications for Dobutamine/ Atropine and Adenosine Stress DOBUTAMINE

ADENOSINE

General Contraindications

General Contraindications

Unstable angina Severe arterial hypertension Significant aortic stenosis (gradient > 50 mm Hg or area < 1 cm2) Significant obstructive hypertrophic cardiomyopathy Complex cardiac arrhythmias

Unstable angina Arterial hypotension Myocardial infarction < 3 days Asthma

Myocardial inflammation (perimyocarditis) Caution Comedication with diuretics (hypokalemia) Atropine Contraindications Narrow angle glaucoma, myasthenia gravis, obstructive uropathy

Severe obstructive pulmonary disease Preexisting AV block Caution Stenotic valvular disease Cerebrovascular insufficiency Comedication with b blockers/Ca antagonists/ digitalis Sinoatrial disease

Dobutamine

Adenosine

Target heart rate ([220 age]  0.85) reached Hypertension (blood pressure > 240/120 mm Hg) Systolic blood pressure drop > 40 mm Hg Intolerable symptoms (chest pain, dyspnea, etc.) Significant supraventricular/ ventricular arrhythmias

Persistent or symptomatic AV block or other arrhythmia Persistent or symptomatic hypotension Significant blood pressure drop > 20 mm Hg Severe respiratory difficulties

the stress examination. Basically, changes of the ST segment are nondiagnostic as a result of the ECG wave distortion in the static magnetic field. However, since wall motion abnormalities precede ST segment changes and the former can readily be detected with CMR imaging, monitoring with rapid cine sequences is effective without a diagnostic ECG (e.g., cine CMR real time scans can be run repetitively to detect new or worsening wall motion abnormalities at the very first occurrence).37 Blood pressure monitoring can easily be done with a conventional monitoring system outside the scanner room with an extension line placed through a waveguide in the radiofrequency cage or special CMR compatible equipment may be used.

Patient Evacuation and Emergency Equipment CMR stress testing can be done safely and successfully if certain caveats are closely followed. In general, monitoring during a stress CMR examination requires the same precau tions and emergency equipment as any other stress test.36 A physician who is appropriately trained in advanced cardiac life support must be present throughout the stress examination and during the recovery phase (personal supervision). Precautions for rapid patient evacuation must be taken; that is, a trolley should be permanently placed under the table, and a button for manual table release must be present. The staff must regularly practice the maneuver for rapid patient evacuation consisting of the immediate cessation of the dobutamine infusion, disconnection of the cardiac receiver coil and the ECG cable, and evacuation of the patient using the table trolley unit, all of which are usually achievable in less than 30 sec by two staff members. Out side the scan room, cardiac resuscitation can be performed according to emergency guidelines.

Image Display and Analysis During the pharmacologic stress procedure, the examiner continuously evaluates the CMR cine loops in an automatic view window as displayed on the console of the scanner. The reconstruction speed of the cine images has become very fast and allows on line visual assessment of wall motion, with the cine loops being displayed almost instan taneously. Alternatively, with a workstation next to the scanner console, the cine loops can be automatically trans ferred and displayed in a synchronized quadscreen. For

assessment of perfusion abnormalities with CMR, the inter pretation of both stress and rest images is required. Thus, on line monitoring of the very first occurrence of perfusion abnormalities is limited.

Duration of Stress CMR Examinations A comparison of the examination times required for the different imaging strategies is shown in Figure 17 1. Whereas the adenosine stress CMR perfusion protocol per se is quite rapid (maximal hyperemic stress is achieved after 4 minutes, and imaging takes approximately 2 min utes), most centers add either a perfusion study at rest (required for the calculation of myocardial perfusion reserve) or a delayed enhancement study for the detection of viable and scarred myocardium. In contrast, DCMR takes longer (12 to 15 minutes of graded dobutamine infusion with repetitive cine imaging). However, a dobutamine stress test provides all information (i.e., presence of infarction, viability, ischemia) without the need for additional, contrast enhanced studies. Conse quently, for both pharmacologic stress CMR imaging approaches, the duration of time when the patient is within the scanner is similar.

Pitfalls and Advanced Issues Coverage CMR perfusion imaging usually covers 16 out of 17 myocar dial segments according to the standardized myocardial seg mentation of the heart using three short axis views (apical, equatorial, and basal).38 Although segment 17 (i.e., the apical cap) is not covered, studies incorporating the apex with imag ing of an additional long axis view during CMR perfusion imaging failed to show ischemia in this region.39 Further more, the analysis of only the inner three out of five to eight short axis slices resulted in a significantly improved diagnos tic accuracy for the detection of coronary artery disease.18 These findings illustrate that diagnostic quality may differ with the order of slice acquisition and/or slice location. Thus, the question of optimal coverage and segmentation cannot be answered definitely at the present time.

Functional Assessment of Viable Myocardium With CMR, myocardial viability can be assessed from a mor phologic and a functional perspective. The morphologic approach aims at detecting regions of scarred myocardium after intravenous injection of an extracellular, gadolinium contrast agent. This late gadolinium enhancement (LGE) technique distinguishes scarred myocardium (appearing bright) from nonscarred myocardium (appearing dark). The transmural extent of LGE can then be used to predict the likelihood of functional recovery after restoration of blood flow to the respective myocardial territory. As an alternative, the functional, contractile response to low dose dobutamine stimulation can be determined. The low dose dobutamine challenge was found to be superior in predicting recovery of function, particularly in segments with a scar transmurality of 1% to 74%.40 As a possible explanation for this finding, it was suggested that even Cardiovascular Magnetic Resonance 237

17 COMPARISON OF PERFUSION AND WALL MOTION CARDIOVASCULAR MAGNETIC RESONANCE IMAGING

Table 17-3 Termination Criteria for Dobutamine and Adenosine Stress

ISCHEMIC HEART DISEASE

though scar imaging depicts the area of myocardial fibrosis, it does not assess the functional state of the surrounding (potentially viable) myocardial tissue. As a consequence, its capability for the prediction of functional recovery of nontransmurally scarred myocardium was found to be lim ited. As can be seen in Figure 17 1, LGE can easily be integrated when combined stress CMR perfusion and/or stress CMR wall motion imaging is performed.

myocardial blood flow reserve.43 Therefore, judging the induction of a maximal hyperemic response from these parameters alone might be misleading.

Route and Duration of Administration

Combined Adenosine Perfusion and Dobutamine Wall Motion CMR

In general, for noninvasive stress testing, both dobutamine and adenosine are given intravenously as continuous infu sions via a peripheral line. DOBUTAMINE For dobutamine, the likelihood of adverse effects increases with infusion duration rather than dosage. Hence, the exam iner is advised to abide by the 3 minute intervals for each stress level. In addition, in a minority of patients (15%), no relevant increase in heart rate might be observed up to the 30 mg/kg/min level (premaximal level); in this case, an early, low dose application of atropine (e.g., 0.25 mg) should be considered, since with alleviating a dominant vagotonic state, dobutamine rapidly exerts its full effects. Side effects of dobutamine should preferably be reverted by using a short acting beta blocker (e.g., esmolol), since then the nearly identical half life of stressor and antagonist (both 2 to 3 minutes) ensures that side effects of the beta blocker do not occur long after the dobutamine effect is gone. Nitroglycerin might be used to treat chest pain symp tomatically, but it does not specifically antagonize the effects of dobutamine and should be reserved for patients in whom beta blockers are contraindicated. ADENOSINE The standard intravenous dosage of adenosine is 140 mg/ kg/min applied during a total infusion duration of 6 minutes at maximum. In a detailed invasive dose response study, this dose regimen was compared with intracoronary papaverine, which is considered to be the strongest coro nary vasodilator.41 Of note, with the standard adenosine regimen, 16% of patients did not reach a maximal hyper emic response as compared with papaverine. In addition, the reagibility of the coronary system to adenosine may vary considerably between patients not only with regard to the maximally achievable hyperemic response, but also with regard to the onset of maximal hyperemia (after 112  48 sec).42 At our department, care is taken to start perfusion CMR not earlier than after 4 min utes of adenosine infusion. In addition, the patient is asked to abstain from substances that antagonize the adenosine effects (i.e., methylxanthine derivatives such as caffeine, theophylline, and theobromine) for at least 24 hours prior to the examination. Recently, the common practice of using heart rate and blood pressure to assess adenosine response during noninvasive stress testing has been questioned. In a study examining the relationship between myocardial blood flow as assessed by PET and changes in heart rate and blood pressure during adenosine infusion, the investigators found that both parameters were extremely poor predictors of 238 Cardiovascular Magnetic Resonance

COSTS It should be recognized that in most countries, the cost of direct vasodilators (in particular adenosine) is higher than that of dobutamine.

In 100 patients, Wahl and colleagues used a combined single session protocol for adenosine perfusion and high dose DCMR.44 The combined protocol was reported to be safe and feasible. Reasons for nondiagnostic tests were poor image quality on perfusion CMR (n ¼ 3 patients) and inability to reach the target heart rate during dobutamine/atropine stress CMR (n ¼ 1 patient). Invasive X ray angiography was the standard of reference for determination of diagnostic accu racy, with significant epicardial stenoses defined as 50% diameter reduction. Figure 17 3 shows the diagnostic performance of DCMR alone, perfusion CMR alone, and their respective combina tion (dobutamine and adenosine stress test pathologic, dobutamine or adenosine stress test pathologic). Perfusion CMR and dobutamine stress CMR reached similarly high diagnostic accuracies (86% and 85%, respectively). Com pared with the combined analysis, only a marginal improvement in diagnostic accuracy (89%) could be achieved. Yet the number of diagnostic examinations

100 80

99 97 96100

97 87 88

90

84 77

77

86 85

89 81

71

60 40 20 0 Sensitivity, % Specificity, %

Accuracy, %

Diagnostic examinations, %

Dobutamine stress CMR Adenosine stress perfusion CMR Dobutamine stress and adenosine perfusion pathologic Dobutamine stress or adenosine perfusion pathologic

Figure 17 3 Preliminary data on the diagnostic performance of a combined stress examination (high dose dobutamine stress wall motion and adenosine stress perfusion CMR) as assessed during a single session examination in 100 patients (see also Fig. 17 1A). Dobutamine stress wall motion and adenosine stress perfusion CMR showed a similarly high diagnostic accuracy (86% and 85%, respectively) for the detection of significant epicardial coronary stenosis of 50% or more. If significant coronary artery disease was assumed in case of at least one pathologic stress test (dobutamine stress wall motion or adenosine stress perfusion CMR), diagnostic accuracy increased only marginally (89%).

CONCLUSION CMR offers the unique possibility of assessing perfusion and wall motion during pharmacologic stress in a single session examination. For this purpose, one can rely on either vasodilator (adenosine) or adrenergic stimulating agent (dobutamine) stress testing. Vasodilator agents cause heterogeneous myocardial perfusion, which under specific conditions (transmurality of perfusion deficit > 75%) is sufficient to cause dysfunc tional myocardial motion. However, vasodilator stress

should be used mainly in conjunction with perfusion CMR for which high diagnostic accuracies have been consistently reported. The adrenergic stimulating agent dobutamine causes an increase in myocardial oxygen demand by increasing con tractility, blood pressure, and heart rate, which are the main determinants of myocardial workload. As such, the detection of inducible wall motion abnormalities can best be accom plished by using adrenergic stimulants and a consistently high diagnostic performance for the detection of inducible, dysfunctional wall motion related to the presence of epicardial coronary stenoses; more than 50% has been demonstrated by dobutamine stress wall motion CMR. Preliminary data suggest that a combined single session evaluation of vasodilator and adrenergic stimulating stress testing compared to stress testing with either of the agents alone does not lead to a clear improvement in the detection of significant coronary artery disease. A single session CMR examination of wall motion and perfusion during high dose dobutamine stress is feasible and represents an alternative for patients with contraindica tions to vasodilator stress. Although initial results are encouraging, the value of high dose dobutamine stress perfusion CMR still requires investigation in large, unselected patient populations.

References 1. Fearon WF, Yeung AC. Evaluating intermediate coronary lesions in the cardiac catheterization laboratory. Rev Cardiovasc Med. 2003;4(1):1–7. 2. Wilson RF. Assessing the severity of coronary-artery stenoses. N Engl J Med. 1996;334(26):1735–1737. 3. de Bruyne B, Bartunek J, Sys SU, Pijls NH, Heyndrickx GR, Wijns W. Simultaneous coronary pressure and flow velocity measurements in humans: feasibility, reproducibility, and hemodynamic dependence of coronary flow velocity reserve, hyperemic flow versus pressure slope index, and fractional flow reserve. Circulation. 1996;94(8): 1842–1849. 4. De Bruyne B, Pijls NH, Heyndrickx GR, Hodeige D, Kirkeeide R, Gould KL. Pressure-derived fractional flow reserve to assess serial epicardial stenoses: theoretical basis and animal validation. Circulation. 2000;101(15):1840–1847. 5. Pijls NH, De Bruyne B, Peels K, et al. Measurement of fractional flow reserve to assess the functional severity of coronary-artery stenoses. N Engl J Med. 1996;334(26):1703–1708. 6. Pijls NH, De Bruyne B, Bech GJ, et al. Coronary pressure measurement to assess the hemodynamic significance of serial stenoses within one coronary artery: validation in humans. Circulation. 2000;102(19): 2371–2377. 7. Nesto RW, Kowalchuk GJ. The ischemic cascade: temporal sequence of hemodynamic, electrocardiographic and symptomatic expressions of ischemia. Am J Cardiol. 1987;59(7):23C–30C. 8. Kaul S, Kiess M, Liu P, et al. Comparison of exercise electrocardiography and quantitative thallium imaging for one-vessel coronary artery disease. Am J Cardiol. 1985;56(4):257–261. 9. Carlson RE, Kavanaugh KM, Buda AJ. The effect of different mechanisms of myocardial ischemia on left ventricular function. Am Heart J. 1988;116(2 Pt 1):536–545. 10. Elhendy A, Geleijnse ML, Roelandt JR, et al. Dobutamine-induced hypoperfusion without transient wall motion abnormalities: less severe ischemia or less severe stress? J Am Coll Cardiol. 1996;27(2): 323–329. 11. Forster T, McNeill AJ, Salustri A, et al. Simultaneous dobutamine stress echocardiography and technetium-99m isonitrile single-photon emission computed tomography in patients with suspected coronary artery disease. J Am Coll Cardiol. 1993;21(7):1591–1596.

12. Biaggioni I, Olafsson B, Robertson RM, Hollister AS, Robertson D. Cardiovascular and respiratory effects of adenosine in conscious man: evidence for chemoreceptor activation. Circ Res. 1987;61(6):779–786. 13. Belardinelli L, Linden J, Berne RM. The cardiac effects of adenosine. Prog Cardiovasc Dis. 1989;32(1):73–97. 14. Nguyen T, Heo J, Ogilby JD, Iskandrian AS. Single photon emission computed tomography with thallium-201 during adenosine-induced coronary hyperemia: correlation with coronary arteriography, exercise thallium imaging and two-dimensional echocardiography. J Am Coll Cardiol. 1990;16(6):1375–1383. 15. Marwick T, Willemart B, D’Hondt AM, et al. Selection of the optimal nonexercise stress for the evaluation of ischemic regional myocardial dysfunction and malperfusion: comparison of dobutamine and adenosine using echocardiography and 99mtc-MIBI single photon emission computed tomography. Circulation. 1993;87(2):345–354. 16. Paetsch I, Jahnke C, Wahl A, et al. Comparison of dobutamine stress magnetic resonance, adenosine stress magnetic resonance, and adenosine stress magnetic resonance perfusion. Circulation. 2004;110 (7):835–842. 17. Al-Saadi N, Nagel E, Gross M, et al. Noninvasive detection of myocardial ischemia from perfusion reserve based on cardiovascular magnetic resonance. Circulation. 2000;101(12):1379–1383. 18. Nagel E, Klein C, Paetsch I, et al. Magnetic resonance perfusion measurements for the noninvasive detection of coronary artery disease. Circulation. 2003;108(4):432–437. 19. Schwitter J, Nanz D, Kneifel S, et al. Assessment of myocardial perfusion in coronary artery disease by magnetic resonance: a comparison with positron emission tomography and coronary angiography. Circulation. 2001;103(18):2230–2235. 20. Wolff SD, Schwitter J, Coulden R, et al. Myocardial first-pass perfusion magnetic resonance imaging: a multicenter dose-ranging study. Circulation. 2004;110(6):732–737. 21. Fung AY, Gallagher KP, Buda AJ. The physiologic basis of dobutamine as compared with dipyridamole stress interventions in the assessment of critical coronary stenosis. Circulation. 1987;76(4):943–951. 22. Bartunek J, Wijns W, Heyndrickx GR, de Bruyne B. Effects of dobutamine on coronary stenosis physiology and morphology: comparison with intracoronary adenosine. Circulation. 1999;100(3):243–249.

Cardiovascular Magnetic Resonance 239

17 COMPARISON OF PERFUSION AND WALL MOTION CARDIOVASCULAR MAGNETIC RESONANCE IMAGING

(i.e., at least one of the two stress tests was performed successfully with adequate image quality) for the combined approach increased to 100%. Thus, although this study represents preliminary data only, it may serve as an indicator that adenosine stress per fusion and dobutamine stress wall motion CMR might not necessarily detect the same “ischemic” abnormality, which may at least in part be attributed to the different pathophy siologies being detected (maldistribution of myocardial blood flow during vasodilator induced hyperemia versus myocardial oxygen supply demand mismatch during adrenergic stimulation).

ISCHEMIC HEART DISEASE

23. Hundley WG, Hamilton CA, Thomas MS, et al. Utility of fast cine magnetic resonance imaging and display for the detection of myocardial ischemia in patients not well suited for second harmonic stress echocardiography. Circulation. 1999;100(16):1697–1702. 24. Nagel E, Lehmkuhl HB, Klein C, et al. [Influence of image quality on the diagnostic accuracy of dobutamine stress magnetic resonance imaging in comparison with dobutamine stress echocardiography for the noninvasive detection of myocardial ischemia]. Z Kardiol. 1999;88(9):622–630. 25. Elhendy A, Bax JJ, Poldermans D. Dobutamine stress myocardial perfusion imaging in coronary artery disease. J Nucl Med. 2002;43 (12):1634–1646. 26. Al-Saadi N, Gross M, Paetsch I, et al. Dobutamine induced myocardial perfusion reserve index with cardiovascular MR in patients with coronary artery disease. J Cardiovasc Magn Reson. 2002;4(4):471–480. 27. Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: sensitivity encoding for fast MRI. Magn Reson Med. 1999;42(5):952–962. 28. Plein S, Radjenovic A, Ridgway JP, et al. Coronary artery disease: myocardial perfusion MR imaging with sensitivity encoding versus conventional angiography. Radiology. 2005;235(2):423–430. 29. Weiger M, Pruessmann KP, Boesiger P. Cardiac real-time imaging using SENSE. Sensitivity Encoding scheme. Magn Reson Med. 2000;43 (2):177–184. 30. Cerqueira MD, Verani MS, Schwaiger M, Heo J, Iskandrian AS. Safety profile of adenosine stress perfusion imaging: results from the Adenoscan Multicenter Trial Registry. J Am Coll Cardiol. 1994;23(2): 384–389. 31. Cortigiani L, Picano E, Coletta C, et al. Safety, feasibility, and prognostic implications of pharmacologic stress echocardiography in 1482 patients evaluated in an ambulatory setting. Am Heart J. 2001;141(4): 621–629. 32. Mathias Jr W, Arruda A, Santos FC, et al. Safety of dobutamineatropine stress echocardiography: a prospective experience of 4,033 consecutive studies. J Am Soc Echocardiogr. 1999;12(10):785–791. 33. Secknus MA, Marwick TH. Evolution of dobutamine echocardiography protocols and indications: safety and side effects in 3,011 studies over 5 years. J Am Coll Cardiol. 1997;29(6):1234–1240. 34. Geleijnse ML, Fioretti PM, Roelandt JR. Methodology, feasibility, safety and diagnostic accuracy of dobutamine stress echocardiography. J Am Coll Cardiol. 1997;30(3):595–606. 35. Picano E, Mathias Jr W, Pingitore A, Bigi R, Previtali M. Safety and tolerability of dobutamine-atropine stress echocardiography:

240 Cardiovascular Magnetic Resonance

36.

37.

38.

39.

40. 41.

42.

43.

44.

a prospective, multicentre study. Echo Dobutamine International Cooperative Study Group. Lancet. 1994;344(8931):1190–1192. Rodgers GP, Ayanian JZ, Balady G, et al. American College of Cardiology/American Heart Association Clinical Competence Statement on Stress Testing. A Report of the American College of Cardiology/ American Heart Association/American College of Physicians-American Society of Internal Medicine Task Force on Clinical Competence. Circulation. 2000;102(14):1726–1738. Nagel E, Lorenz C, Baer F, et al. Stress cardiovascular magnetic resonance: consensus panel report: detecting left ventricular myocardial ischemia during intravenous dobutamine with cardiovascular magnetic resonance imaging (MRI). J Cardiovasc Magn Reson. 2001;3 (3):267–281. Cerqueira MD, Weissman NJ, Dilsizian V, et al. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: a statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation. 2002;105(4):539–542. Elkington AG, Gatehouse PD, Prasad SK, Moon JC, Firmin DN, Pennell DJ. Combined long- and short-axis myocardial perfusion cardiovascular magnetic resonance. J Cardiovasc Magn Reson. 2004; 6(4):811–816. Wellnhofer E, Olariu A, Klein C, et al. Magnetic resonance low-dose dobutamine test is superior to SCAR quantification for the prediction of functional recovery. Circulation. 2004;109(18):2172–2174. Wilson RF, Marcus ML, White CW. Prediction of the physiologic significance of coronary arterial lesions by quantitative lesion geometry in patients with limited coronary artery disease. Circulation. 1987;75 (4):723–732. De Bruyne B, Pijls NH, Barbato E, et al. Intracoronary and intravenous adenosine 5’-triphosphate, adenosine, papaverine, and contrast medium to assess fractional flow reserve in humans. Circulation. 2003;107(14):1877–1883. Mishra RK, Dorbala S, Logsetty G, et al. Quantitative relation between hemodynamic changes during intravenous adenosine infusion and the magnitude of coronary hyperemia: implications for myocardial perfusion imaging. J Am Coll Cardiol. 2005;45(4):553–558. Wahl A, Paetsch I, Roethemeyer S, Gebker R, Klein C, Nagel E. A combined single session analysis of adenosine perfusion and of high-dose dobutamine stress cardiovascular magnetic resonance improves diagnosis of ischemia. Circulation. 2003;108(17):403. [Supplement IV].

Acute Myocardial Infarction: Cardiovascular Magnetic Resonance Detection and Characterization Andrew E. Arai

Noninvasive imaging in patients presenting with acute myocardial infarction (AMI) can serve several useful roles. Cardiovascular magnetic resonance (CMR) in particular is useful in assessing a wide range of clinically relevant issues. Measurement of left ventricular (LV) ejection fraction (LVEF) provides prognostic value and is a critical element in determining which patients should be treated with implanted cardiodefibrillators. CMR is well accepted as a reference standard modality of determining myocardial via bility and the methods work well in patients with recent AMI. In most patients, the culprit vessel can be predicted on the basis of the spatial distribution of abnormalities on the late gadolinium enhancement (LGE) scan or other find ings. CMR stress testing is capable of detecting residual myocardial ischemia. Various important complications of myocardial infarction can be imaged, such as LV dysfunc tion, LV thrombus, ventricular septal defect, aneurysm, pseudoaneurysm, and valvular complications. Thus, infor mation that has prognostic value or is useful in risk stratifi cation can be acquired in a single CMR examination in patients who survive acute myocardial infarction. This chapter will focus on the role of CMR in the detec tion and characterization of AMI. This goal overlaps to some degree with the goals of other chapters, such as those on stress testing and viability assessment, so the material presented will focus on validations, particularly on the per spective of detecting or characterizing AMI.

GLOBAL AND REGIONAL LEFT VENTRICULAR FUNCTION: CINE CARDIOVASCULAR MAGNETIC RESONANCE At 1.5 T, global and regional LV systolic function is rela tively easily imaged with steady state free precession (SSFP) cine CMR.1,2 SSFP cine CMR has been accepted as a refer ence standard for assessing LV mass, volumes, and LVEF.3–6 Cine CMR is a powerful method for detecting regional wall motion abnormalities, as is evidenced by the diagnostic accu racy of dobutamine stress tests. Recent comparisons indicate that cine CMR performs significantly better than non contrast enhanced echocardiography and invasive X ray left

ventriculography in assessing regional LV systolic function.7 Only contrast enhanced echocardiography performed slightly better than cine CMR. Cine CMR can assess regional wall motion with high image quality, allowing diagnosis of very small wall motion abnormalities with confidence that might be missed by other methods. For example, Figure 18 1 illustrates a case in which a presumptive diagnosis of clinically unrecog nized myocardial infarction could be made on the basis of a focal anteroseptal wall motion abnormality that was missed on a good quality non contrast enhanced transtho racic echocardiogram. Dobutamine stress testing provides the strongest evidence supporting the accuracy of cine CMR in diagnosing regional wall motion abnormalities. In publications summarizing experience using dobutamine stress testing to detect or man age coronary artery disease (CAD) in a combined experience in 877 patients, cine CMR performed with a sensitivity rang ing from 82% to 87% and a specificity between 80% and 90%.8–15 This is an important set of validations, because dobutamine stress testing poses some of the most extreme challenges for assessing regional wall motion. The diagnostic information must be acquired quickly usually within about 2 minutes at a time when the heart rate is 85% of pre dicted maximum and patients may be experiencing angina. Cine CMR also performed well in detecting acute coro nary syndrome in patients presenting to an emergency department with at least 30 minutes of chest pain. Both qualitatively and quantitatively, regional wall motion abnormalities by cine CMR had an accuracy of 82% for acute coronary syndrome, 89% for non ST elevation myo cardial infarction, and 98% for ischemic heart disease.16 One important reason cine CMR performed so well relates to the acuity of the CMR scan in this study. Specifically, 87 of subjects were scanned before the 4 hour troponin was available. By studying patients within 6 hours of presenta tion to the emergency department, stunned myocardium offers an additional mechanism that is capable of detecting infarction or unstable angina. The statistics summarizing use of cine CMR to detect acute coronary syndrome illustrate an important concept related to the diagnostic utility of using regional wall motion abnormalities. Most importantly, a single study of regional wall motion does not differentiate acute from chronic conditions. In patients with no history of Cardiovascular Magnetic Resonance 241

18 ACUTE MYOCARDIAL INFARCTION: CARDIOVASCULAR MAGNETIC RESONANCE DETECTION AND CHARACTERIZATION

CHAPTER 18

ISCHEMIC HEART DISEASE

End diastole

infarction, a definite regional wall motion abnormality is reasonably specific for CAD even though there are other conditions that can cause regional wall motion abnormal ities such as myocarditis. On the other hand, in a patient with a prior infarction, the presence or absence of a regional wall motion abnor mality frequently does not alter the clinical approach to a patient. In the latter situation, evidences of inducible ische mia or residual viability become additional pieces of infor mation that are needed to understand how to deal with the regional wall motion abnormality. Fortunately, CMR is well suited to imaging these other important aspects of ischemic heart disease. In summary, cine CMR is a reference standard method of assessing global and regional LV systolic function. Cine CMR is both sensitive and specific for acute coronary syn drome in patients without prior infarction but cannot unequivocally differentiate AMI from chronic infarction. Focal myocarditis may mimic wall motion abnormalities associated with myocardial infarction.

LATE GADOLINIUM ENHANCEMENT TO DETECT ACUTE MYOCARDIAL INFARCTION LGE CMR has become accepted as a reference standard method for imaging myocardial scar. Before discussing applications of LGE, it is important to understand the mechanisms leading to hyperenhancement of AMI and chronic myocardial infarction, as these are somewhat differ ent mechanisms. Finally, it is important to differentiate per fusion abnormalities from viability abnormalities because CMR is capable of differentiating these different clinically relevant scenarios. Current generation gadolinium based contrast agents are generally a combination of gadolinium with a chelating agent that facilitates excretion and minimizes likelihood of toxic effects from free gadolinium in the body17 (see Chapter 6). These contrast agents are designated as extra cellular contrast agents. After intravenous injection, the 242 Cardiovascular Magnetic Resonance

End systole

Figure 18 1 Small dyskinetic wall motion abnormality associated with myocardial infarction that was missed by transthoracic echocardiography. The cine CMR is shown at end diastole and end systole. The arrow points to the wall motion abnormality.

extracellular gadolinium based contrast agent rapidly dif fuses into the interstitial space but is excluded from the intracellular space in healthy tissues. Thus, viable myocar dium enhances to a degree based on the amount of gado linium that is in the intravascular volume and the interstitial space. In contrast, AMI and chronic myocardial infarctions enhance to a much greater degree and thus can be distinguished as regions that are generally brighter than normal myocardium on T1 weighted images obtained typically 10 to 20 minutes after contrast administration.18 A gadolinium dose of 0.1 and 0.2 mmol/kg and a time win dow of 5 to 30 minutes after contrast administration pro vide similar results, provided that the TI is adjusted.18a The vascular, interstitial, and intracellular distribution or exclusion is summarized schematically in Figure 18 2. In AMI, nonviable or infarcted cardiomyocytes lose cell membrane integrity, leading to a larger percentage of the myocardium that enhances with gadolinium (see Fig. 18 2) as gadolinium enters what had previously been intracellular space. Thus, there is enhancement in the area of AMI as gadolinium is found in the intravascular space, in the interstitial space, and inside nonviable cells. With chronic myocardial infarction, LGE enhancement is due to the characteristics of collagen scar (see Chapter 1). The collagen scar that forms after myocardial infarction is relatively cellular in comparison with normal myocardium. There is an extensive increase in interstitial collagen deposi tion but a relatively small intracellular space within fibro blasts. Thus, both AMI and chronic myocardial infarctions enhance to a greater extent than normal myocardium does. Inversion recovery methods have become the de facto standard in obtaining LGE images of myocardial infarc tion.19 While gadolinium deposition and T1 shortening in the area of infarction had been recognized for many years, the inversion recovery approach provides much higher contrast than previous T1 weighted images did. Based on the physics of inversion recovery, there is an optimal inver sion time that results in uniform nulling of normal myocar dium (see Chapter 1). Nulling normal myocardium refers to a situation whereby the signal intensity of normal myo cardium becomes uniformly very dark. Thus, any abnorm alities of excess gadolinium accumulation appear as a bright patch on a dark background, a situation that leads to high sensitivity in detecting myocardial infarction.

LGE: Nulled to define normal T2: Dark

Viable myocardium within area at risk

LGE: Essentially normal T2: Bright

Acutely infarcted myocardium

LGE: Bright T2: Bright

A

B

C

D

Infarcted myocardium with microvascular obstruction

LGE: Dark or paradoxically gray on IR T2: Bright (sometimes dark)

Figure 18 2 In the setting of acute myocardial infarction, different types of myocardium have distinctive appearances on late gadolinium enhancement (LGE) images and T2 weighted images. A, Normal myocardium outside the ischemic area at risk is characterized by healthy cardiomyocytes (striped cylinders) with intact cell membranes and a characteristic intracellular to extracellular ratio and blood volume. The extracellular gadolinium contrast agents (white crosses) arrive via the bloodstream and rapidly enter the interstitial space but are excluded from the intracellular space. Thus, normal myocardium has a shorter T1 after contrast administration, but the amount of enhancement is modest, and the inversion recovery time can be adjusted to make normal myocardium appear uniformly dark (nulled). B, Viable myocardium within the area at risk has intact cell membranes and therefore excludes gadolinium contrast agents from the intracellular space. However, some aspect of the ischemic period reversibly damaged this part of the heart enough to raise the water content and thus lengthen the T2 of the tissue. Therefore, the area at risk is brighter than normal myocardium on T2 weighted images. Because LGE images are essentially normal, there must be relatively balanced increases in the intracellular and extracellular volumes. Thus, the myocardium appears essentially normal on LGE CMR. C, Acutely infarcted myocardium is characterized by loss of cell membrane integrity. Hence, gadolinium contrast agents rapidly enter not only the extracellular space but also what used to be the intracellular space as long as there has been adequate reperfusion. As indicated by the large number of white crosses, acutely infarcted myocardium enhances significantly in comparison to normal myocardium and appears bright on LGE CMR. Acutely infarcted myocardium also appears bright on T2 weighted images, owing to the tissue swelling. D, Infarcted myocardium with microvascular obstruction has a very different appearance. Despite opening of the epicardial vessels, sometimes the microvessels remain occluded (black no entry of white crosses symbols). With severe enough occlusions, the gadolinium contrast agents cannot enter the myocardium, and the T1 remains long. Sometimes it remains long enough that the signal intensity is slightly positive and is gray on conventional magnitude inversion recovery images. In our experience thus far, non reperfused infarcts are bright on T2 weighted images, but there are times where dark patches are present within the otherwise bright edematous zone due to microvascular obstruction or intramyocardial hemorrhage.

VALIDATION OF LATE GADOLINIUM ENHANCEMENT CARDIOVASCULAR MAGNETIC RESONANCE FOR ACUTE MYOCARDIAL INFARCTION LGE CMR is currently the highest resolution reliable method for detecting myocardial infarction and provides an exquisitely sensitive diagnostic test. Although it was known for many years that myocardial infarctions enhance with gadolinium, there were important advances in meth odology,19 basic validations, and clinical testing (Table 18 1).20–30 These provide important clues as to why LGE CMR is such a powerful clinical tool. As was previously discussed, the spatial localization of gad olinium corresponds closely to areas of both AMI and chronic myocardial infarction and the LGE mass appears highest in patients with Q wave AMI, intermediate in patients with ST

elevation but non Q wave AMI, and lowest in patients with non ST elevation AMI.31 In animal studies in which histopa thology can be used as a reference standard for determining what is normal or infarcted myocardium, the bright zones on LGE CMR correlate closely in size and location to both AMI and chronic myocardial infarction.18,32 In patients, AMI size correlates well with the size of myocardial infarction deter mined by single photon emission tomography (SPECT) imag ing.25 Although acute infarcts have not been studied by both positron emission tomography (PET) scans and CMR, there are good correlations between these two modalities in measur ing chronic infarct size.33,34 However, investigators have noted that small subendocardial infarcts are routinely missed on SPECT and PET scans, owing to issues related to spatial reso lution.33–35 One must recognize that it is possible for a patient to have an AMI despite a “normal” SPECT scan. Very small myocardial infarctions ( 1 mm N = 1713 DWT > 5.5 mm (viable)

WTh < 1 mm N = 234

47% WTh < 1 mm N = 248

DWT > 5.5 mm DWT < 5.5 mm (viable) (scar)

Figure 20 10 Normalized uptake of 18 F fluorodeoxyglucose (FDG) uptake stratified by end diastolic wall thickness as measured from gradient echo magnetic resonance images. Regions with preserved wall thickening (WTh) of 1 mm or more and preserved end diastolic wall thickness (DWT) of > 5.5 mm had a relative FDG uptake similar to that of regions without wall thickening (WTh < 1 mm) but preserved end diastolic wall thickness. In contrast, akinetic regions (WTh < 1 mm) with reduced end diastolic wall thickness had significantly reduced FDG uptake (mean: 47%), indicating scar formation. Source: Data from Baer FM, Voth E, Schneider CA, Theissen P, Schicha H, Sechtem U. Comparison of low dose dobutamine gradient echo magnetic resonance imaging and positron emission tomography with [18F] fluorodeoxyglucose in patients with chronic coronary artery disease: a functional and morphological approach to the detection of residual myocardial viability. Circulation. 1995;91:1006 1015.

Contractile Reserve During Low-Dose Dobutamine Infusion Although severely reduced end diastolic wall thickness is helpful in identifying scarred myocardium, the positive predictive value of a preserved end diastolic wall thickness for predicting recovery of function following revasculariza tion is disappointingly low. However, CMR offers the possi bility of measuring wall thickening not only at rest but also during low dose dobutamine infusion. Until recently, a protocol with acquisition of cine CMR images in multiple short axes and two long axes sections at rest and at 5 and 10 mg/kg/min dobutamine required an imaging time of more than 60 minutes. The advent of fast CMR sequences now permits completion of the same protocol within approximately 30 to 45 minutes, and image quality is often better with breath hold cine CMR images than with con ventional techniques. The sensitivity of dobutamine CMR for detection of viable myocardium as defined by a normal ized FDG uptake on PET images is 81% with a specificity of 95%.69 When recovery of wall thickening following revascularization was considered to be the gold standard, the sensitivity of dobutamine CMR in predicting recovery of function after revascularization was 89% with a specific ity of 94%. The latter analysis was patient related, which is clinically more meaningful than a segment by segment analysis.70 New techniques such as tissue tagged MR imag ing with three dimensional strain analysis may further enhance the accuracy of dobutamine CMR.74 Using a healthier patient group and a different methodol ogy, Trent and colleagues74a found less satisfactory values for sensitivity and specificity of wall motion (50% and 72%, Cardiovascular Magnetic Resonance 275

20 MYOCARDIAL VIABILITY

residual viable myocardium in the infarct area. However, cau tion must be used in observing a small area of pronounced wall thinning in order not to assume that the entire region per fused by an occluded coronary artery is completely scarred. Frequently, myocardial cells in the border zone survive, and ischemia of this border zone alone may cause substantial symptoms in a patient. Therefore, in a patient with single ves sel disease, previous myocardial infarction and anginal symp toms, restoration of blood flow by reestablishing patency of the occluded artery may be justified, despite evidence of com plete necrosis in the center of the infarct zone.69

ISCHEMIC HEART DISEASE

respectively) or wall thickening (50% and 68%, respectively). This was possibly related to the fact that they used higher dobutamine doses. They also included segments with wors ening wall dynamics, which were considered viable. In con trast, Baer and colleagues70 looked only at akinetic segments, which by definition could not become worse. Baer and coworkers also presented data on the relative value of dobutamine CMR and dobutamine transesophageal echocardiography (TEE).75 Normalized FDG uptake on PET images was used as the standard against which both techni ques were compared. The sensitivity and the specificity of dobutamine TEE and dobutamine CMR for FDG PET defined myocardial viability were 77% versus 81% and 94% versus 100%, respectively. Thus, the two imaging techniques provide similar accuracy. In choosing the appro priate technique, patient acceptance becomes an important consideration. Although claustrophobia may be a problem with CMR, only a small fraction of patients are affected. In contrast, many patients do not like the experience of a transesophageal echocardiographic examination. On the other hand, there is a clear cost advantage for transesopha geal echocardiography because the echo probe costs only a fraction of a CMR scanner, and additional investment is not necessary.

Late Gadolinium Enhancement in Chronic Infarction Similar principles apply to the assessment of myocardial infarction in the chronic phase as to assessment shortly after the event. Late enhancement of the infarct zone is also seen in chronic infarcts,76,77 because of the continuing increase in partition coefficient, and delayed contrast agent kinetics. The tissue in chronic scar consists mainly of fibro blasts and collagen surrounded by a large intercellular space. Because gadolinium contrast agents will diffuse into this space, enhancement may simply be related to the larger volume of distribution in comparison to normal myocar dium with its tightly packed myocytes, which the contrast agent cannot enter. LGE CMR is able to identify patients with healed myo cardial infarction with high accuracy. Moreover, this tech nique may permit noninvasive differentiation between ischemic and nonischemic cardiomyopathy. In a study of 71 subjects, 40 patients with healed myocardial infarction were prospectively enrolled after enzymatically proven necrosis and were imaged 3  1 months (N ¼ 32) and/or 14  7 months (N ¼ 19) later.74 They were compared with 20 patients with nonischemic cardiomyopathy and 11 normal volunteers. Twenty nine of 32 patients (91%) with 3 month old infarcts (13 non Q wave) and all 19 with 14 month old infarcts (8 non Q wave) exhibited LGE. In patients in whom the infarct related artery was determined at angiography, 24 of 25 patients with 3 month old infarcts (96%) and all 14 with 14 month old infarcts had LGE in the correct territory. None of the 20 patients with nonis chemic cardiomyopathy or the 11 volunteers had LGE. Regardless of the presence or absence of Q waves, the majority of patients with LGE had only nontransmural involvement. Normal LV contraction was visualized in 7 patients with 3 month old infarcts (22%) and 3 with 14 month old infarcts (16%), but in these cases, LGE was limited to the subendocardium. McCrohon and 276 Cardiovascular Magnetic Resonance

colleagues78 demonstrated in a cohort of 90 patients with heart failure that all patients with coronary artery disease had LGE primarily with a subendocardial or transmural pattern, whereas patients with dilated cardiomyopathy (no stenosis by coronary angiography) had either no enhancement (59%), patchy or midwall enhancement not corresponding to a coronary perfusion bed (28%), or enhancement patterns indistinguishable from those of patients with ischemic cardiomyopathy (13%). The patchy and midwall enhancement patterns are often found in patients with biopsy proven myocarditis.79 This indicates that CMR may give profound new insights into the causes of LV failure. Prediction of the myocardial response to revasculariza tion remains the holy grail of viability testing. Late enhancement CMR performs well in this discipline. In the first study looking at the ability of contrast CMR to predict improvement in LV function after revascularization, Kim and colleagues80 showed that the likelihood of functional recovery decreased progressively as the transmural extent of LGE observed before revascularization increased (Fig. 20 11). Approximately 80% of segments with no LGE improved function after revascularization (Fig. 20 12), whereas if more than 75% of the transmural tissue was enhanced, only a small percentage improved with revascularization (Fig. 20 13).

Thickness of the Viable Epicardial Rim and Recovery of Function The clear depiction of myocardial scar by gadolinium enhanced CMR also permits clear identification of epicar dially located viable tissue. Knuesel and colleagues80a determined the segmental amount of viable tissue by LGE CMR and compared this to 18F FDG uptake and resting perfusion using 13N ammonia. FDG uptake of 50% or more corresponded to a viable rim thickness of 4.5 mm by LGE CMR. Segments that had both a thick viable rim and a FDG uptake of 50% or more showed functional recovery in 85% of segments (Fig. 20 14), whereas thin metabolically nonviable segments (FDG uptake 2 mm) indicative of an increased atherosclerotic plaque burden. The inner and outer walls are indicated by the arrows. Source: Kim WY, Stuber M, Bornert P, Kissinger KV, Manning WJ, Botnar RM. Three dimensional black blood cardiac magnetic resonance coronary vessel wall imaging detects positive arterial remodeling in patients with nonsignificant coronary artery disease. Circulation. 2002;106: 296 299.

ISCHEMIC HEART DISEASE

Prospective studies to determine the predictive value of contrast enhanced coronary plaque imaging are needed. Pre liminary data in patients with acute myocardial infarction suggest that LGE CMR can monitor local inflammation.38a Angiographically normal segments showed no change in CNR within a week of infarction as compared with 3 months post infarction. In comparison, there was a significant decrease in CNR in stenotic segments not revascularized. The decline paralleled the decline in C reactive protein.

MOLECULAR CARDIOVASCULAR MAGNETIC RESONANCE OF ATHEROSCLEROSIS Different CMR probes have been developed to study vari ous biologic processes (e.g., thrombosis, angiogenesis, inflammation, neoplasia) and diseases (e.g., cancer, cardio vascular disease, stroke, diabetes) by targeting a spectrum of molecular markers such as fibrin, selectins, and integ rins. Molecular probes generally target either changes in receptor expression or alterations in metabolic processes.39 A molecular imaging agent is typically composed of a ligand such as an antibody, a short peptide or a sugar mol ecule, and a signal element (e.g., Gd3þ, iron oxides) that are attached to each other by a linker or spacer (Fig. 26 6). Larger molecular imaging probes often include a carrier or nanoparticle (e.g., liposomes, perfluorocarbon emul sions, cross linked iron oxide) that can be loaded with sev eral to a few thousand signal elements and ligands, thereby (1) Gd molecule attached to ligand (ligand = carrier)

increasing signal amplification and the affinity to the target of interest (see Fig. 26 6). The pharmacokinetics of those agents can be modulated by the size of the carrier, the number of ligands, or additional pharmacokinetic modula tion units. Many such agents are in the preclinical stage. So far, only a few plaque specific contrast agents (e.g., gadofluorin, iron oxides) have been investigated.

MOLECULAR CARDIOVASCULAR MAGNETIC RESONANCE OF THROMBOSIS Noninvasive visualization of evolving arterial thrombus may facilitate detection of unstable plaques and thrombosis burden in vulnerable patients. Recently, advances have been made with in vivo imaging of arterial thrombus by fibrin binding molecular CMR contrast agents.40,41 Direct imaging of arterial thrombus by targeted or “molecular” contrast agents (which are engineered to bind to specific target molecules) is advantageous to classical noncontrast multispectral CMR, since the demand for high spatial reso lution and motion compensation strategies is less stringent. Furthermore, even though several studies have shown high sensitivity of CMR to the detection of carotid and aortic thrombi,42–44 differentiation between complex atheroscle rotic plaques and mural thrombosis remains difficult because of the complex composition (e.g., platelets, fibrin, and red blood cells) of thrombus and resultant complex CMR signal characteristics on T1 , T2 , and proton (2) Nanoparticle labeled with multiple Gd molecules and ligands (nanoparticle = carrier)

Signal element (e.g. gadolinium) Nanoparticle Linker Ligand Target Endothelium

Ligand: antibody, small peptides

Signal element: Gd3+ chelates or iron oxide

Carrier Liposomes, Perfluorocarbon emulsions Cross-linked iron oxide (CLIO)

Target: adhesion molecules, integrins, receptors, fibrin Figure 26 6 Schematic of molecular contrast agents targeted against endothelial activation. The basic components of a typical molecular contrast agent consist of a ligand that binds to a specific target and a signal element, which, in case of CMR, is made of a Gd3þ chelate or an iron oxide. These two basic components can be directly linked to each other, as in (1), or may be attached to or incorporated within a larger nanoparticle (carrier) as demonstrated in (2). 356 Cardiovascular Magnetic Resonance

Figure 26 7 A, Reformatted coronary artery CMR from a coronal 3D dataset shows subrenal aorta 20 hours after EP 1873 administration (EPIX Pharmaceuticals, Inc., Cambridge, MA) in a rabbit model of atherosclerosis and plaque rupture. Three well delineated mural thrombi (arrows) can be observed, with good contrast between thrombus (numbered), arterial blood (dashed arrow), and vessel wall (solid arrow). The in plane view of the aorta allows simultaneous display of all thrombi, showing head, tail, length, and relative location. B D, Corresponding cross sectional views show good agreement with histopathology (E G). Source: Botnar RM, Perez AS, Witte S, et al. In vivo molecular imaging of acute and subacute thrombosis using a fibrin binding magnetic resonance imaging contrast agent. Circulation. 2004;109:2023 2029.

In another study by Sirol and colleagues,50,51 similar agents were used in an animal model of acute and chronic thrombo sis. Signal intensity decreased with thrombus age, allowing differentiation between the acute and chronic stages. Combining the advent of fibrin binding molecular CMR con trast agents and advances in coronary artery CMR techniques offers the potential for direct imaging of coronary thrombo sis. The feasibility of this approach was demonstrated by using the gadolinium based fibrin binding contrast agent EP 2104R ((EPIX Pharmaceuticals, Inc., Cambridge, MA), in a swine model of native coronary thrombus (Fig. 26 8) and in stent thrombosis using CMR lucent stents.40 Potential applications for direct thrombus imaging include detection and evaluation of acute coronary syndromes and ischemic strokes. Platelets are also key to thrombus formation and play a role in the development of atherosclerosis. Though it has not yet been studied in the coronary arteries, von zur Muh len and colleagues51a reported on a CMR contrast agent consisting of microparticles of iron oxide and single chain antibody targeting ligand induced binding sites on acti vated glycoprotein IIb/IIIa to image the carotid artery

H

1)

F Lumen Microthrombus

B

1)

E 2)

Wall

2)

C

F 3)

3)

A

5 mm

5 mm

D

G

Cardiovascular Magnetic Resonance 357

26 ATHEROSCLEROTIC PLAQUE IMAGING: CORONARIES

density weighted images of arterial thrombi. The concept of target specific imaging or molecular imaging was first introduced over a decade ago, and has since been further developed by Weissleder45,46 and others47 for MRI and for optical imaging in recent years. The advantage of CMR molecular plaque imaging is the potential for relatively high spatial resolution. The limitation is the inherently low sen sitivity of CMR contrast enhancement technology, requiring a relatively high target molecule concentration (>50 100 mM Gd at target site; r1 ffi 21 mM 1  s 1 per Gd) for sufficient signal amplification. Initial attempts were made with targeting fibrin,47–49 which is abundant in arte rial clots and therefore plays an important role in acute cor onary syndromes and stroke. In vivo CMR of acute and subacute thrombosis following plaque rupture in an animal model of aortic atherosclerosis has been implemented by using a small molecule fibrin binding peptide derivative, EP 1873 (EPIX Pharmaceuticals, Inc., Cambridge, MA).41 This molecular agent allowed for imaging of large lumen encroaching thrombi as well as submillimeter mural thrombi with signal enhancement of the entire thrombus and excellent differentiation from the vessel wall (Fig. 26 7).

ISCHEMIC HEART DISEASE

In-stent thrombus Stent LAD LCX

LAD

LM

A

B

C

In-stent thrombus Stent LAD In-stent thrombus

LCX

LM

D

LAD

E

F

Figure 26 8 In vivo CMR of in stent thrombosis in a swine model of coronary thrombosis. A, D, Coronary artery CMR before (A) and after (D) injection of a fibrin binding contrast agent, EP 2104R (EPIX Pharmaceuticals, Inc., Cambridge, MA). On both scans, no apparent thrombus is visible (arrows, circle). (B, E) Black blood inversion recovery segmented k space gradient echo CMR before (B) and after (E) contrast administration of EP 2104R (same view as A and D). After contrast injection (E), a bright area is readily visible (arrows, circle), consistent with the location of the in stent thrombus. No apparent thrombus was visible on prethrombus (B) images (arrow, circle). (C, F) X ray angiogram confirming CMR finding of in stent thrombus in mid LAD (circle). LAD, left anterior descending coronary artery; LCX, left circumflex coronary artery; LM, left main coronary artery. Source: Botnar RM, Buecker A, Wiethoff AJ, et al. In vivo molecular imaging of acute and subacute thrombosis using a fibrin binding magnetic resonance imaging contrast agent. Circulation. 2004;110:1463 1466.

thrombi and atherosclerosis plaques in a mouse model of atherothrombosis studied at high field (9.4 T). In vivo visu alization of platelet activation was demonstrated, along with signal loss after thrombolysis.

INFLAMMATION Vascular inflammation and associated endothelial activation are believed to play an integral role in the initiation and pro gression of atherosclerosis. Endothelial activation is charac terized by the upregulation of leukocyte adhesion molecules (intercellular adhesion molecule 1 (ICAM 1) and vascular adhesion molecule 1 (VCAM 1), E selectin, and P selectin) on the endothelial cell surface.52 The adhesion molecules facilitate tethering, adhesion, and transendothelial migration of leukocytes, including monocytes. Differentia tion of monocytes into macrophages and subsequent diges tion of lipoproteins by macrophages occur in a later stage and eventually lead to the accumulation of lipid filled macrophages, which are believed to be a precursor of rupture prone vulnerable plaque. While the specific 358 Cardiovascular Magnetic Resonance

mechanisms by which the adhesion molecules contribute to this process have not yet been determined, studies of ath erogenic mice with various adhesion molecule deficiencies have indicated some role for each of the inducible adhesion molecules, particularly VCAM 1 and P selectin.52–54 Work by Sibson and colleagues55 and Barber and col leagues56 demonstrated that early endothelial activation occurs in focal ischemia in mice brains55 and in brain inflammation in rats (after IL 1b and TNF a induced E and P selectin upregu lation) using a novel MR contrast agent. This novel gadolinium labeled contrast agent, Gd DTPA B(sLex)A,57 consists of the Sialyl Lewisx (sLex) carbohydrate, which interacts with both E and P selectin. The relaxivity was measured as 3.5 mM 1  sec 1 at 1.5 T and thus is similar to gadolinium DTPA.

ANGIOGENESIS Identification of angiogenesis may be useful in studying tumor growths and atherosclerotic plaque development. Integrins, such as aVb3, are overexpressed in activated neo vascular endothelial cells, which are believed to play an

CLINICAL STUDIES Reproducibility studies have shown excellent results in measuring total plaque volumes of the thoracic and abdom inal aorta as surrogate markers of atherosclerosis.61 Also, coronary plaque burden can be quantified with good repro ducibility.62 Since CMR permits highly reproducible mea sures of aortic anatomy and atherosclerosis, serial studies to investigate the effect of, for example, statin therapy on regression of aortic plaque burden have been successfully performed.63

OUTLOOK With the broader availability of higher field (>1.5 T) whole body CMR systems, in vivo coronary vessel wall imaging is likely to benefit from the expected near linear increase in SNR.64–66 Improved SNR may potentially improve spatial

resolution and/or reduce imaging time. In a preliminary study, we demonstrated the feasibility (see Fig. 26 5) of high field in vivo coronary vessel wall imaging and found an SNR increase of approximately 50% compared to previ ous reports at 1.5 T.15,22 Other investigators67,68 have reported similar results. With continued advances in hard ware technology and pulse sequence design, further advances in coronary plaque imaging should be expected in the years to come.

CONCLUSION Atherothrombosis, defined as atherosclerotic plaque dis ruption with superimposed thrombus formation, is the major cause of acute coronary syndromes and cardiovascu lar death. CMR is emerging as the most comprehensive noninvasive imaging technique for imaging of atherothrom bosis in large and medium sized arteries, including the aorta and the carotid, coronary, and peripheral arteries. CMR imaging of coronary wall atherothrombosis is particu larly challenging, owing to the small caliber of the vessels combined with respiratory and cardiac motion. Free breathing 3D CMR coronary vessel wall imaging has enabled in vivo quantification of coronary plaque burden and remodeling as a marker of subclinical coronary artery disease. Molecular imaging utilizing target specific contrast agents such as fibrin binding agents to detect arterial thrombus shows great promise as the new frontier in non invasive imaging. Advances in molecular imaging and CMR techniques offer the potential for direct imaging of coronary thrombosis and in stent thrombosis using fibrin binding molecular CMR contrast agents. While the current role of noninvasive CMR imaging of atherothrombosis remains investigational, integration of vascular biology with CMR should enhance our understanding of the natural history of acute coronary syndromes and thereby facilitate strate gies to prevent acute coronary syndromes and cardiovascu lar death in vulnerable patients.

References 1. Yusuf S, Reddy S, Ounpuu S, Anand S. Global burden of cardiovascular diseases: part I: general considerations, the epidemiologic transition, risk factors, and impact of urbanization. Circulation. 2001;104: 2746–2753. 2. Buffon A, Biasucci LM, Liuzzo G, D’Onofrio G, Crea F, Maseri A. Widespread coronary inflammation in unstable angina. N Engl J Med. 2002;347:5–12. 3. Mauriello A, Sangiorgi G, Fratoni S, et al. Diffuse and active inflammation occurs in both vulnerable and stable plaques of the entire coronary tree: a histopathologic study of patients dying of acute myocardial infarction. J Am Coll Cardiol. 2005;45:1585–1593. 4. Goldstein JA, Demetriou D, Grines CL, Pica M, Shoukfeh M, O’Neill WW. Multiple complex coronary plaques in patients with acute myocardial infarction. N Engl J Med. 2000;343:915–922. 5. Rioufol G, Finet G, Ginon I, et al. Multiple atherosclerotic plaque rupture in acute coronary syndrome: a three-vessel intravascular ultrasound study. Circulation. 2002;106:804–808. 6. Naghavi M, Libby P, Falk E, et al. From vulnerable plaque to vulnerable patient: a call for new definitions and risk assessment strategies: Part I. Circulation. 2003;108:1664–1672. 7. Toussaint JF, LaMuraglia GM, Southern JF, Fuster V, Kantor HL. Magnetic resonance images lipid, fibrous, calcified, hemorrhagic, and thrombotic components of human atherosclerosis in vivo. Circulation. 1996;94:932–938.

8. Fayad ZA, Fallon JT, Shinnar M, et al. Noninvasive in vivo high-resolution magnetic resonance imaging of atherosclerotic lesions in genetically engineered mice. Circulation. 1998;98:1541–1547. 9. Yuan C, Mitsumori LM, Ferguson MS, et al. In vivo accuracy of multispectral magnetic resonance imaging for identifying lipid-rich necrotic cores and intraplaque hemorrhage in advanced human carotid plaques. Circulation. 2001;104:2051–2056. 9a. Sosnovik DE, Nahrendoft M, Weissleder R. Molecular magnetic resonance imaging in cardiovascular disease. Circulation. 2007;115: 2076–2086. 10. Kim WY, Danias PG, Stuber M, et al. Coronary magnetic resonance angiography for the detection of coronary stenoses. N Engl J Med. 2001;345:1863–1869. 11. Kim WY, Stuber M, Kissinger KV, Andersen NT, Manning WJ, Botnar RM. Impact of bulk cardiac motion on right coronary MR angiography and vessel wall imaging. J Magn Reson Imaging. 2001;14: 383–390. 12. Wang Y, Vidan E, Bergman GW. Cardiac motion of coronary arteries: variability in the rest period and implications for coronary MR angiography. Radiology. 1999;213:751–758. 13. Stehning C, Bornert P, Nehrke K, Dossel O. Free breathing 3D balanced FFE coronary magnetic resonance angiography with prolonged cardiac acquisition windows and intra-RR motion correction. Magn Reson Med. 2005;53:719–723. Cardiovascular Magnetic Resonance 359

26 ATHEROSCLEROTIC PLAQUE IMAGING: CORONARIES

integral role in tumor growth and the initiation and devel opment of atherosclerosis. Anderson and coworkers have developed perfluoro nanoparticles that can carry as many as 90,000 paramagnetic Gd chelates per particle and can be targeted against various biomarkers by attaching appro priate ligands. They have developed methods for identify ing angiogenic vessels by targeting the aVb3 integrin.58 A probe consisting of Gd perfluorocarbon nanoparticle linked to anti aVb3 monoclonal antibody produced MRI signal enhancement of the capillary bed in a corneal micro pocket model of angiogensis.58 A similar approach was taken by Guccione and colleagues, who attached anti aVb3 antibody to a lipid based polymerized vesicle to image angiogenic tumor.39 A nonantibody based approach was also taken by Winter and coworkers, who attached a peptidomimetic vitronectin antagonist to the nanoparticles to again target aVb3 integrin. Using this approach, they were able to image angiogenic vessels in nascent Vx 2 rabbit tumors59 and in early stage atherosclerosis.60

ISCHEMIC HEART DISEASE

14. Taylor AM, Jhooti P, Wiesmann F, Keegan J, Firmin DN, Pennell DJ. MR navigator-echo monitoring of temporal changes in diaphragm position: implications for MR coronary angiography. J Magn Reson Imaging. 1997;7:629–636. 15. Fayad ZA, Fuster V, Fallon JT, et al. Noninvasive in vivo human coronary artery lumen and wall imaging using black-blood magnetic resonance imaging. Circulation. 2000;102:506–510. 16. Ehman RL, Felmlee JP. Adaptive technique for high-definition MR imaging of moving structures. Radiology. 1989;173:255–263. 17. Li D, Kaushikkar S, Haacke EM, et al. Coronary arteries: threedimensional MR imaging with retrospective respiratory gating. Radiology. 1996;201:857–863. 18. Stuber M, Botnar RM, Danias PG, Kissinger KV, Manning WJ. Submillimeter three-dimensional coronary MR angiography with real-time navigator correction: comparison of navigator locations. Radiology. 1999;212:579–587. 19. Botnar RM, Kim WY, Bornert P, Stuber M, Spuentrup E, Manning WJ. 3D coronary vessel wall imaging utilizing a local inversion technique with spiral image acquisition. Magn Reson Med. 2001;46:848–854. 20. Manke D, Nehrke K, Bornert P. Novel prospective respiratory motion correction approach for free-breathing coronary MR angiography using a patient-adapted affine motion model. Magn Reson Med. 2003;50:122–131. 21. Jahnke C, Paetsch I, Nehrke K, et al. Rapid and complete coronary arterial tree visualization with magnetic resonance imaging: feasibility and diagnostic performance. Eur Heart J. 2005;26:2313–2319. 22. Botnar RM, Stuber M, Kissinger KV, Kim WY, Spuentrup E, Manning WJ. Noninvasive coronary vessel wall and plaque imaging with magnetic resonance imaging. Circulation. 2000;102:2582–2587. 23. Edelman RR, Chien D, Kim D. Fast selective black blood MR imaging. Radiology. 1991;181:655–660. 24. Botnar RM, Stuber M, Kim WY, Danias PG, Manning WJ. Magnetic resonance coronary lumen and vessel wall imaging. Rays. 2001;26: 291–303. 25. Stuber M, Botnar RM, Danias PG, et al. Double-oblique free-breathing high resolution three-dimensional coronary magnetic resonance angiography. J Am Coll Cardiol. 1999;34:524–531. 26. Katoh M, Spuentrup E, Buecker A, Manning WJ, Gu¨nther RW, Botnar RM. MR coronary vessel wall imaging: comparison between radial and spiral k-space sampling. J Magn Reson Imaging. 2006;23: 757–762. 27. Kim WY, Stuber M, Bornert P, Kissinger KV, Manning WJ, Botnar RM. Three-dimensional black-blood cardiac magnetic resonance coronary vessel wall imaging detects positive arterial remodeling in patients with nonsignificant coronary artery disease. Circulation. 2002;106: 296–299. 28. Kim WY, Astrup AS, Stuber M, et al. Subclinical coronary and aortic atherosclerosis detected by magnetic resonance imaging in type 1 diabetes with and without diabetic nephropathy. Circulation. 2007;115: 228–235. 29. Yuan C, Kerwin WS, Ferguson MS, et al. Contrast-enhanced high resolution MRI for atherosclerotic carotid artery tissue characterization. J Magn Reson Imaging. 2002;15:62–67. 30. Wasserman BA, Smith WI, Trout 3rd HH, Cannon 3rd RO, Balaban RS, Arai AE. Carotid artery atherosclerosis: in vivo morphologic characterization with gadolinium-enhanced double-oblique MR imaging initial results. Radiology. 2002;223:566–573. 31. Cai J, Hatsukami TS, Ferguson MS, et al. In vivo quantitative measurement of intact fibrous cap and lipid-rich necrotic core size in atherosclerotic carotid plaque: comparison of high-resolution, contrastenhanced magnetic resonance imaging and histology. Circulation. 2005;112:3437–3444. 32. Kramer CM, Cerilli LA, Hagspiel K, DiMaria JM, Epstein FH, Kern JA. Magnetic resonance imaging identifies the fibrous cap in atherosclerotic abdominal aortic aneurysm. Circulation. 2004;109:1016–1021. 33. Bley TA, Wieben O, Uhl M, Thiel J, Schmidt D, Langer M. Highresolution MRI in giant cell arteritis: imaging of the wall of the superficial temporal artery. AJR Am J Roentgenol. 2005;184:283–287. 34. Desai MY, Stone JH, Foo TK, Hellmann DB, Lima JA, Bluemke DA. Delayed contrast-enhanced MRI of the aortic wall in Takayasu’s arteritis: initial experience. AJR Am J Roentgenol. 2005;184:1427–1431. 35. Kerwin W, Hooker A, Spilker M, et al. Quantitative magnetic resonance imaging analysis of neovasculature volume in carotid atherosclerotic plaque. Circulation. 2003;107:851–856.

360 Cardiovascular Magnetic Resonance

35a. Itskovich V, Mani V, Mizsei G, et al. Parallel and nonparallel simultaneous multislice black-blood double inversion recovery techniques for vessel wall imaging. J Magn Reson Imaging. 2004;19:459–467. 36. Maintz D, Ozgun M, Hoffmeier A, et al. Selective coronary artery plaque visualization and differentiation by contrast-enhanced inversion prepared MRI. Eur Heart J. 2006;27:1732–1736. 37. Yeon SB, Sabir A, Clouse M, et al. Delayed-enhancement cardiovascular magnetic resonance coronary artery wall imaging: comparison with multislice computed tomography and quantitative coronary angiography. J Am Coll Cardiol. 2007;31;50:441–447 . 38. Schar M, Kim WY, Stuber M, Boesiger P, Manning WJ, Botnar RM. The impact of spatial resolution and respiratory motion on MR imaging of atherosclerotic plaque. J Magn Reson Imaging. 2003;17:538–544. 38a. Ibrahim T, Makowski MR, Jankauskas A, et al. Serial contrastenhanced MRI demonstrates regression of hyperenhancement within the coronary artery wall in patients after acute myocardial infarction. JACC: Cardiovasc Imaging. 2009;2:580–588. 39. Guccione S, Li KC, Bednarski MD. Vascular-targeted nanoparticles for molecular imaging and therapy. Methods Enzymol. 2004;386:219–236. 40. Botnar RM, Buecker A, Wiethoff AJ, et al. In vivo magnetic resonance imaging of coronary thrombosis using a fibrin-binding molecular magnetic resonance contrast agent. Circulation. 2004;110:1463–1466. 41. Botnar RM, Perez AS, Witte S, et al. In vivo molecular imaging of acute and subacute thrombosis using a fibrin-binding magnetic resonance imaging contrast agent. Circulation. 2004;109:2023–2029. 42. Johnstone MT, Botnar RM, Perez AS, et al. In vivo magnetic resonance imaging of experimental thrombosis in a rabbit model. Arterioscler Thromb Vasc Biol. 2001;21:1556–1560. 43. Corti R, Osende JI, Fayad ZA, et al. In vivo noninvasive detection and age definition of arterial thrombus by MRI. J Am Coll Cardiol. 2002;39:1366–1373. 44. Moody AR, Murphy RE, Morgan PS, et al. Characterization of complicated carotid plaque with magnetic resonance direct thrombus imaging in patients with cerebral ischemia. Circulation. 2003;107:3047–3052. 45. Weissleder R. Molecular imaging: exploring the next frontier. Radiology. 1999;212:609–614. 46. Weissleder R, Elizondo G, Wittenberg J, Rabito CA, Bengele HH, Josephson L. Ultrasmall superparamagnetic iron oxide: characterization of a new class of contrast agents for MR imaging. Radiology. 1990;175:489–493. 47. Flacke S, Fischer S, Scott MJ, et al. Novel MRI contrast agent for molecular imaging of fibrin: implications for detecting vulnerable plaques. Circulation. 2001;104:1280–1285. 48. Johansson LO, Bjornerud A, Ahlstrom HK, Ladd DL, Fujii DK. A targeted contrast agent for magnetic resonance imaging of thrombus: implications of spatial resolution. J Magn Reson Imaging. 2001;13:615–618. 49. Yu X, Song SK, Chen J, et al. High-resolution MRI characterization of human thrombus using a novel fibrin-targeted paramagnetic nanoparticle contrast agent. Magn Reson Med. 2000;44:867–872. 50. Sirol M, Aguinaldo JG, Graham PB, et al. Fibrin-targeted contrast agent for improvement of in vivo acute thrombus detection with magnetic resonance imaging. Atherosclerosis. 2005;182:79–85. 51. Sirol M, Fuster V, Badimon JJ, et al. Chronic thrombus detection with in vivo magnetic resonance imaging and a fibrin-targeted contrast agent. Circulation. 2005;112:1594–1600. 51a. von zur Muhlen C, von Elverfeldt D, Moeller JA, et al. Magnetic resonance imaging contrast agent targeted toward activated platelets allows in vivo detection of thrombosis and monitoring of thrombolysis. Circulation. 2008;118:358–367. 52. Cybulsky MI, Charo IF. Leukocytes, adhesion molecules, and chemokines. In: Fuster V, ed. Atherothrombosis and Coronary Artery Disease. Philadelphia: Lippincott Williams & Wilkins; 2005:489–503. 53. Dansky HM, Barlow CB, Lominska C, et al. Adhesion of monocytes to arterial endothelium and initiation of atherosclerosis are critically dependent on vascular cell adhesion molecule-1 gene dosage. Arterioscler Thromb Vasc Biol. 2001;21:1662–1667. 54. Dong ZM, Brown AA, Wagner DD. Prominent role of P-selectin in the development of advanced atherosclerosis in ApoE-deficient mice. Circulation. 2000;101:2290–2295. 55. Sibson NR, Blamire AM, Bernades-Silva M, et al. MRI detection of early endothelial activation in brain inflammation. Magn Reson Med. 2004;51:248–252. 56. Barber PA, Foniok T, Kirk D, et al. MR molecular imaging of early endothelial activation in focal ischemia. Ann Neurol. 2004;56: 116–120.

63. Corti R, Fayad ZA, Fuster V, et al. Effects of lipid-lowering by simvastatin on human atherosclerotic lesions: a longitudinal study by highresolution, noninvasive magnetic resonance imaging. Circulation. 2001;104:249–252. 64. Singerman RW, Denison TJ, Wen H, Balaban RS. Simulation of B1 field distribution and intrinsic signal-to-noise in cardiac MRI as a function of static magnetic field. J Magn Reson. 1997;125:72–83. 65. Dougherty L, Connick TJ, Mizsei G. Cardiac imaging at 4 Tesla. Magn Reson Med. 2001;45:176–178. 66. Stuber M, Botnar RM, Fischer SE, et al. Preliminary report on in vivo coronary MRA at 3 Tesla in humans. Magn Reson Med. 2002;48: 425–429. 67. Koktzoglou I, Simonetti O, Li D. Coronary artery wall imaging: initial experience at 3 Tesla. J Magn Reson Imaging. 2005;21:128–132. 68. Priest AN, Bansmann PM, Kaul MG, Stork A, Adam G. Magnetic resonance imaging of the coronary vessel wall at 3 T using an obliquely oriented reinversion slab with adiabatic pulses. Magn Reson Med. 2005;54:1115–1122.

Cardiovascular Magnetic Resonance 361

26 ATHEROSCLEROTIC PLAQUE IMAGING: CORONARIES

57. Laurent S, Vander Elst L, Fu Y, Muller RN. Synthesis and physicochemical characterization of Gd-DTPA-B(sLex)A, a new MRI contrast agent targeted to inflammation. Bioconjug Chem. 2004;15:99–103. 58. Anderson SA, Rader RK, Westlin WF, et al. Magnetic resonance contrast enhancement of neovasculature with alpha(v)beta(3)-targeted nanoparticles. Magn Reson Med. 2000;44:433–439. 59. Winter PM, Caruthers SD, Kassner A, et al. Molecular imaging of angiogenesis in nascent Vx-2 rabbit tumors using a novel alpha(nu) beta3-targeted nanoparticle and 1.5 tesla magnetic resonance imaging. Cancer Res. 2003;63:5838–5843. 60. Winter PM, Morawski AM, Caruthers SD, et al. Molecular imaging of angiogenesis in early-stage atherosclerosis with alpha(v)beta3integrin-targeted nanoparticles. Circulation. 2003;108:2270–2274. 61. Chan SK, Jaffer FA, Botnar RM, et al. Scan reproducibility of magnetic resonance imaging assessment of aortic atherosclerosis burden. J Cardiovasc Magn Reson. 2001;3:331–338. 62. Desai MY, Lai S, Barmet C, Weiss RG, Stuber M. Reproducibility of 3D free-breathing magnetic resonance coronary vessel wall imaging. Eur Heart J. 2005;26:2320–2324.

ISCHEMIC HEART DISEASE

CHAPTER 27

Assessment of the Biophysical Mechanical Properties of the Arterial Wall Raad H. Mohiaddin

Arteries are elastic tubes whose diameter varies with the pulsating pressure. In addition, they propagate the pulse created by ejection of blood by the heart, at a velocity that is determined largely by the elastic properties of the arterial wall. The vascular wall can be deformed by pressure and shear stress forces exerted by the blood as well as the teth ering imposed by the surrounding tissues. Biophysical mechanical properties of the arterial wall play an important role in the pathogenesis of cardiovascular diseases. Sclero sis (or stiffness), for example, is an important aspect of ath erosclerotic vascular disease that can be demonstrated in experimental disease both in animals1 and in humans,2 and regression of the disease leads to reduced stiffness.3,4 Rupture of atherosclerotic plaque (a common mechanism of myocardial infarction) and aortic dissection can be viewed as mechanical failures in the diseased vessels. In addition, the effectiveness of interventional procedures such as angioplasty is often achieved by causing mechani cal injury to the vessel wall, and the injury itself may lead to restenosis. A number of common conditions are asso ciated with changes in arterial mechanical properties, although their importance is not always recognized. Systemic hypertension is almost always associated with altered mechanical properties of the peripheral vasculature. Though it is not clear which of the two is the inceptive event, one begets the other, fostering a vicious cycle. Compliance (the reciprocal of the resistance offered to deformation) of the proximal aorta is reduced as a result. This causes waves reflected from the periphery to return prematurely and coincide with the incident or forward traveling wave produced by the ventricle. This augments aortic systolic pressure, increases pulse pressure, and reduces diastolic pressure (as the reflected wave no longer contributes to diastolic pressure). Decreased subendocar dial/subepicardial flow ratio has been demonstrated with reduced aortic compliance.5 Thus, decreased aortic disten sibility may increase the risk of subendocardial ischemia in the presence of coronary artery stenosis, left ventricular hypertrophy, or both. Aortic compliance represents an important determinant of left ventricular performance. Measurement of ventricular vascular coupling takes into account the pulsatile load imposed on the left ventricle as well as the systemic vascular resistance.6 Metabolic disorders such as Ehlers Danlos and Marfan syndromes,7 diabetes mellitus,8 familial 362 Cardiovascular Magnetic Resonance

hypercholesterolemia,9 and growth hormone deficiency10 are also known to alter arterial compliance. Similarly, the distensibility of the pulmonary arteries is reduced in pulmonary hypertension. Postmortem studies of externalized pulmonary arterial strips have shown that the extensibility of the pulmonary trunk is decreased in pulmonary arterial hypertension.11 The wall of an artery becomes less extensible, the more it is stretched from its natural length. An increased stretching of the circumfer ence of the vessel will diminish the distensibility. When the pulmonary artery resistance increases in vivo, the ves sels become more distended and less distensible. Indirect measurements of pulmonary artery compliance have sug gested that pulmonary arterial distensibility decreases with rising pulmonary artery pressure.12,13 The ability of cardio vascular magnetic resonance (CMR) to image flow within any medium sized vessel in any plane provides a unique opportunity to study the pulmonary arteries. Distensibility and flow have already been assessed in patients with pul monary arterial hypertension.14,15 In this chapter, the clinical importance of arterial bio physical function and its assessment by CMR will be exam ined in detail as a complement to Chapter 26.

ARTERIAL STRUCTURE A normal artery consists of three morphologically distinct layers. The intima consists of a single continuous layer of endothelial cells bounded peripherally by a fenestrated sheet of elastic fibers. The media consists entirely of diago nally oriented smooth muscle cells, surrounded by variable amounts of collagen, elastin, and proteoglycans. The adventitia consists predominantly of fibroblasts intermixed with smooth muscle cells loosely arranged between bun dles of collagen and proteoglycans. Each structural compo nent has its own characteristic properties. Smooth muscle is the physiologically active element, and by contracting or developing force, it can alter the diameter of the vessel or the tension in the wall. The other components are essen tially passive in their mechanical behavior. Elastin, which can be stretched to up to 300% of its length at rest without rupturing,16 behaves mechanically more closely to a linear elastic material such as rubber than other connective tissue components do. When elastin fibers are stretched and

DEFINITION OF VASCULAR WALL STIFFNESS Vascular mechanics have been described by using different elastic moduli and assumptions and for different purposes. Several approaches have been described that use clinically available methods for in vivo characterization of the stiff ness of the vessel wall. The ability to measure vascular stiff ness has been greatly improved by the recent advances in imaging, such as high frequency ultrasound and CMR. The relationship between vascular wall deformation (strain) and the pressure exerted on the inner surface of the vascular wall (stress) is commonly used for the measure ment of arterial wall biophysical properties (elastic modu lus). A plethora of terminology for the description of different elastic moduli, which can be confusing, has been described.18 The pressure strain elastic modulus of the arterial wall (Ep) described by Peterson and colleagues18a

is commonly used. This elastic modulus that applies to an open ended vessel in the absence of reflection is defined as Ep ¼ 2DP/(DV/V). This is the inverse of the fractional distensibility DV/V of the arterial lumen per unit pulse pressure DP. Arterial compliance, C, which is defined as the change in volume DV per unit change in pressure DP, has been also used in the literature. It has been argued that this definition is appropriate to measurement of ventricular compliance and not to the compliance of an open ended arterial seg ment. For the latter, the inverse of Peterson’s modulus was suggested 1/Ep. The average arterial compliance of a particular vessel pathway can also be determined by measuring the speed of propagation of the pulse in the vessel pathway. The velocity of such waves depends principally on the distensi bility of the vessel wall. In real terms, this pulse is measur able by the disturbances in pressure, flow, or vessel diameter that it causes. The propagation of flow waves has not been studied as extensively as that of pressure waves, partly because, unlike flow, accurate methods of pulsatile pressure measurements have been available for a long time and partly because the distinction between flow wave velocity and blood velocity has not always been clearly recognized. Blood velocity means the speed of an average drop of blood, while flow wave velocity means the speed with which motion is transmitted. The wave velocity is usually much faster than that of the blood itself. While CMR is unable to assess pressure changes, altera tions in the flow within (or diameter of) a vessel can be measured accurately.

MEASUREMENT OF ARTERIAL WALL STIFFNESS Arterial stiffness, which describes the resistance of arterial wall to deformation, is difficult to measure because of the complex mechanical behavior of arterial wall. As a result, a bewildering number of choices abound. Though smooth muscle tone is not considered, in vitro human arterial com pliance has been measured from pressure volume curves in postmortem arteries.19–22 In vivo estimation of arterial wall compliance is more difficult, however, and has been per formed by using indirect and invasive techniques, includ ing pulse wave velocity measurements in animals and in humans23,24 the pressure radius relationship using the Peterson transformer coil in animals,25 X ray contrast angi ography in humans,26,27 and ultrasonography.28–30 The contributions of CMR to the assessment of arterial wall mechanics are discussed in the following paragraphs. Other noninvasive measures are forced to rely on the assessment of accessible and often superficial vessels. Under the assumption that central and peripheral arteries behave in a similar fashion, the properties of these arteries are then used as surrogates for those of central arteries. There is, however, considerable heterogeneity between peripheral and central sites. CMR circumvents this problem by allow ing the study of central arteries. Its ability to identify ana tomic landmarks suggests that reproducibility between studies should be improved, allowing more effective fol low up. Cardiovascular Magnetic Resonance 363

27 ASSESSMENT OF THE BIOPHYSICAL MECHANICAL PROPERTIES OF THE ARTERIAL WALL

released, they return promptly to their original state. Elas tin fibers are important for maintaining normal pulsatile behavior, but they fracture at very low stresses and are probably much less important in determining the overall strength of the vessel wall. Collagen fibers, on the other hand, are much stiffer, but they are much stronger. The proportion of these components varies from artery to artery. In the thoracic aorta, the elastin forms 60% of total fibrous element, and collagen forms 40%. The collagen proportion increases with increasing distance from the heart, reaching 30% elastin and 70% collagen in the extrathoracic ves sels.17 The collagen to elastin ratio increases with age, which is one reason why vascular stiffness increases with age. The human thoracic aorta is supplied by vasa vasorum and grows by increasing the number of lamellar units. The abdominal aorta, in comparison, is avascular, as it lacks vasa vasorum and grows by increasing the thickness of each lamellar unit. The avascular thickness and the ele vated tension per lamellar unit of the abdominal aorta pre dispose it to atherosclerosis. The distensibility of a blood vessel depends on the pro portions and interconnections of these materials and on the contractile state of the vascular smooth muscle. Elasticity is a material’s ability to return to its original shape and dimen sions after deformation, the deformation being proportional to the force applied. This proportionality was first described by Hooke in 1676 and is known as Hooke’s law. The point at which Hooke’s law ceases to apply is known as the elastic limit, and when a solid has been deformed beyond this point, it cannot regain its original form and acquires a permanent distortion. With larger loads still, the yield point is reached when the deformation continues to increase without further load and usually rap idly leads to breakage. In purely elastic bodies, stress (the force per unit area that produces deformation) produces its characteristic strain (the deformation of a stressed object) instantaneously, and strain vanishes immediately on removal of the stress. Some materials, however, require a finite time to reach the state of deformation appropriate to the stress and a similar time to regain their unstressed shape. Blood vessels typically exhibit such behavior, which is called viscoelasticity.

ISCHEMIC HEART DISEASE

CARDIOVASCULAR MAGNETIC RESONANCE OF REGIONAL AORTIC COMPLIANCE CMR provides a direct, noninvasive way of studying local aortic compliance.31,32 High resolution cine imaging or elec trocardiogram gated spin echo imaging in a plane perpen dicular to the ascending and/or descending aorta allows measurement of aortic cross sectional area during systole and diastole. Measurement of regional aortic compliance by CMR is calculated from the change in volume of an aortic segment and from aortic pulse pressure estimated by a sphyg momanometer at the level of the brachial artery. The lumen of the aorta is outlined manually on the computer screen to mea sure the change in aortic area (DA) between diastole and sys tole. Regional aortic compliance (C) (microl/mm Hg, m3/ Nm 2) is calculated from the change in volume (DV ¼ DA  slice thickness) of the aortic segment divided by the aortic pulse pressure (DP) measured by a sphygmomanometer (Fig. 27 1). Automatic measurement of aortic cross sectional area is also possible.33 Other indices of aortic stiffness that can be derived from these measurements include distensibility, Peterson’s elastic modulus, and stiffness index b ([systolic blood pressure/diastolic blood pressure]/area strain).

The accuracy of the indirect measurement of the pressure change that is needed to compute compliance is limited, as it ignores the changes in the pressure wave as it propagates through the arterial tree (a process known as amplification). Further, it is important to obtain this pressure data on patients who are ideally lying in the cardiovascular magnetic resonance imaging (MRI) scanner using CMR compatible apparatus. Despite the limitations of the pressure measurement, there is a good correlation between measurement of local aortic com pliance and measurement of global compliance from the speed of the propagation of the flow wave within the vessel.34

CARDIOVASCULAR MAGNETIC RESONANCE OF FLOW WAVE VELOCITY Flow wave velocity is defined as the speed with which a flow wave propagates along a vessel and is regarded as the purest measure of arterial stiffness. It is the quotient of distance traveled divided by the time taken for the flow wave to move between the two points and represents an average for that length of vessel (Figs. 27 2 and 27 3). The approach is dependent on assessment of path length

A

AA

DA

B

C

Figure 27 1 A, Oblique sagittal image of the ascending aorta, arch and descending thoracic aorta showing the sites where flow wave velocity and regional compliance are measured. The oblique transverse plane shown in the top image is represented in diastole (B) and systole (C). This shows the change in area of a 22 year old healthy subject. AA, ascending aorta; DA descending aorta. 364 Cardiovascular Magnetic Resonance

27 ASSESSMENT OF THE BIOPHYSICAL MECHANICAL PROPERTIES OF THE ARTERIAL WALL

Distance (mm)

2

1

A

1 1

2

MPA

2

2

RPA

LPA

B 0.8 0.7

Mean velocity (m/s)

0.6 0.5 0.4 1

0.3

2

0.2 0.1 0 0

C

D

10

20

30

40

50

60

70

80

90

100

110 120 130 140 150 160

170 180 190 200 210

Time (ms)

PWV = transit time (AT2-AT1)/distance

Figure 27 2 Assessment of pulse wave velocity. A, Phase velocity acquisition across a vessel is undertaken at two points (1, 2), and the distance between the two is measured. B, Examples of slice prescription are given for the proximal pulmonary arteries and aorta; note that coverage of the ascending aorta and descending aorta is enabled by a single slice. C, Transit time is defined by the difference in arrival time for the flow wave at both points and is divided by the distance to give pulse wave velocity (D). LPA, left pulmonary artery; MPA, main pulmonary artery; RPA, right pulmonary artery.

Cardiovascular Magnetic Resonance 365

ISCHEMIC HEART DISEASE

6

5

Flow (L/sec)

4

3

2

1

0 17

traveled and accurate measurement of pulse arrival time. The latter requires recognition of equivalent features or points on leading edges of the proximal and distal flow waveforms (see Fig. 27 2), a process that is made compli cated by alterations that occur in flow wave morphology and magnitude as it progresses down the vessel. Unlike noninvasive measurements that rely on linear, transcuta neous measurements, CMR makes no assumptions about the shape of the artery and can accurately measure the path length traveled. Mohiaddin and colleagues showed the feasibility of using CMR phase shift velocity mapping to measure aortic flow wave velocity in humans.34 By taking advantage of the anatomy of the aorta, cine two dimensional phase shift velocity maps were acquired with high temporal resolution in a single plane perpendicular to the ascending and des cending aorta, and the time taken for the flow wave to travel between the two points was measured (Fig. 27 4). Instantaneous flow (in liters per second) in the mid ascending aorta and mid descending aorta was calculated by multiplying the aortic cross sectional area and the mean velocity within that area. Pulse wave velocity (PWV) was calculated in meters per second from the transit time (T) of the foot of the flow wave (see Fig. 27 4) and from the distance (D) between the two points obtained from an oblique sagittal image. The distance is determined manu ally on the computer screen by drawing a line in the center of the aorta joining the two points. In Figure 27 4, the foot of the flow wave was defined by extrapolation of the rapid upstroke of the flow wave to the baseline as opposed to the midpoint of the upslope method used in Figure 27 2. Others have used different MR flow imaging techniques to assess arterial compliance. Tarnawski and colleagues35

366 Cardiovascular Magnetic Resonance

76

106 136 165 195 225 254 284

Time delay from R-wave (msec)

A

Ascending aorta

Descending aorta

4.5 4.0 3.5 3.0 Flow (L/sec)

Figure 27 3 Cine velocity mapping in a plane equivalent to that shown in Figure 27 1. The first frame was acquired 50 msec after the R wave of the electrocardiogram and represents the onset of left ventricular systole. The velocity maps indicate zero velocity as medium gray, caudal velocities in the descending aorta as light gray to white, and cranial velocities as darker shades of gray to black, gray level intensity being proportional to velocity.

47

2.5 2.0 1.5 1.0 0.5 0.0 −0.5 17

B

47

76

106 136 165 195 225 254 284

Time delay from R-wave (msec) Ascending aorta

Descending aorta

Figure 27 4 The foot of the flow wave is defined by extrapolation of the rapid upstroke of the flow wave to the baseline, and this is performed for flow in both the ascending aorta and descending aorta. The transit time needed for the flow wave to propagate from a point in the mid ascending aorta to a point in the mid descending aorta can then be measured. The distance around the arch from the plane in the ascending aorta to the plane in the descending aorta can be measured from the oblique sagittal image shown in Figure 27 1, and the flow wave velocity is calculated by division of this distance by the propagation time. A, Transit time from a normal subject with good compliance. B, Transit time from an elderly patient with poor compliance. Note that the transit time is significantly shorter in the patient with poor compliance. Source: Mohiaddin RH, Firmin DN, Longmore DB. Age related changes of human aortic flow wave velocity measured non invasively by magnetic resonance imaging. J Applied Physiol. 1993;74:492 497.

Figure 27 5 Fourier velocity measurements in a healthy subject. A, In this method, the magnetic resonance signal is obtained from a column aligned with the descending aorta. B, The position along the column is shown horizontally in each of the 12 cine frames. The velocity is shown by the vertical position of the signal. I, inferior; S, superior.

REFLECTED WAVES When the incident pulse wave from the ventricle reaches the periphery, it may be reflected and return toward the heart as a backward running wave. Reflected waves express properties of the peripheral circulation; if the peripheral resistance is elevated, reflected waves will be greater in magnitude and will return sooner. Since its shape and mag nitude will be defined by the complex interaction between forward incident wave and backward reflected waves, the measured flow wave will also be altered in pathologic con ditions. As a result, the timing and location of measure ments become important considerations in the CMR assessment of arterial properties. If work is concerned only with the structure of proximal arteries, measurement should be made as far from the periphery and as early in systole as possible to avoid the influence of reflections (see diagram). These requirements become more demanding when the available length of ves sel is limited (e.g., in the pulmonary arteries) or baseline data particularly noisy. Conversely, study of the influence

A

Velocity

I

Position

S 21

25

18

22

26

19

23

27

20

24

28

B FOV 24 cm, 4.0 ms/frame, peak velocity 120 cm/s

Cardiovascular Magnetic Resonance 367

27 ASSESSMENT OF THE BIOPHYSICAL MECHANICAL PROPERTIES OF THE ARTERIAL WALL

methods, the interleaved repeats needed to achieve high temporal resolution make them highly sensitive to physio logic variability. In vitro experiments showed CMR measurements of pulse wave velocity in a tube phantom to be very reproduc ible and in good agreement with pulse wave velocity measurements made with a pressure catheter.39 These methods have not been widely applied in clinical research, nor has their in vivo reproducibility been proven yet.

used the comb excited Fourier velocity encoded method, previously reported by Dumoulin and colleagues,36 to mea sure local arterial wave speed in the femoral artery in healthy men. In this method, simultaneous Fourier velocity encoded data from multiple stations were acquired. The technique employs a comb excitation radio frequency pulse that excites an arbitrary number of slices. This causes the signals from the spin in a particular slice to appear at a position in the phase encoding direction, which is the sum of the spin velocity and an offset arising from the phase increment given to that excitation slice. Acquisition of spin velocity information occurs simulta neously for all slices, permitting the calculation of wave velocities arising from the pulsatile flow. Hardy and colleagues37 studied aortic flow wave velocity using a two dimensional CMR selective excitation pulse to repeatedly excite a cylinder of magnetization in the aorta, with magnetization read out along the cylinder axis each time. A toggled bipolar flow encoding pulse was applied prior to readout to produce a one dimensional phase con trast flow image. Cardiac gating and data interleaving were employed to improve the effective time resolution to 2 msec. Wave velocities were determined from the slope of the leading edge of flow measured on the resulting M mode velocity image. Aortic pulse wave velocity was also measured by the same group using a combination of cylin der of magnetization with Fourier velocity encoding and readout gradients applied along the cylinder axis (aorta) (Fig. 27 5),38 with the advantage of eliminating partial vol ume effects that hindered their previous approach, but the Fourier method has the drawback that it is no longer in real time and errors occur, owing to accumulation of flow data over several (typically 16 to 32) cardiac cycles. For both

Mohiaddin and colleagues were the first to use CMR for measurement of aortic compliance.40 They demonstrated that aortic compliance in asymptomatic subjects falls with age and that patients with coronary artery disease have abnormally low compliance (Fig. 27 6). The results sug gested a possible role for compliance in the assessment of cardiovascular fitness and the detection of coronary artery disease. Because there is overlap between normal compli ance and compliance in patients with coronary artery dis ease above the age of 50 years, the test cannot have perfect sensitivity and specificity. Below the age of 50, however, there is much less overlap, and the test is more specific. Abnormally low aortic compliance has also been demonstrated in patients with aortic coarctation41 and in patients with Marfan syndrome.42 The same group also showed the feasibility of using CMR velocity mapping for measurement of aortic flow wave velocity. Aortic flow wave velocity increased linearly with age, and there was a signifi cant difference between the youngest decade and the oldest

decade studied. Flow wave velocity was negatively corre lated with regional ascending aortic compliance measured in the same subjects (Fig. 29 7). In a study employing a single slice phase contrast acqui sition of the ascending, arch, proximal aorta, and distal thoracic aorta, Rogers and colleagues43 demonstrated an age related increase in PWV among their cohort. In addi tion, among the older patients, stiffness increased, the more proximally the aorta was studied. The researchers ascribed these changes to the disproportionate effect of elastin deg radation with age on the more proximal parts of the aorta, where the elastic to collagen ratio is at its greatest. Regression of atheroma with reduction of cholesterol levels is recognized to occur, but less is known about rever sal of sclerosis. Noninvasive indices of sclerosis have largely

8

Pulse wave velocity (m/sec)

THE CLINICAL USE OF CARDIOVASCULAR MAGNETIC RESONANCE FOR ASSESSING ARTERIAL WALL STIFFNESS

7

6

5

4

3 10

20

30

A

100 50

40 50 Age (years)

60

70

8

10 5

20

40

60

80

Age Normals

Athletes

r = −0.77 Slope = −0.07 SEE = 0.88 7

5

CAD

Figure 27 6 Ascending aortic compliance displayed by using a logarithmic scale and plotted against age in three groups: normals, athletes, and patients with coronary artery disease (CAD). The 95% confidence limits are shown for the normals. The athletes’ compliance is abnormally high, and that in coronary disease patients is abnormally low. Source: Mohiaddin RH, Underwood SR, Bogren HG, Firmin DN, Klipstein RH, Rees RSO, Longmore DB. Regional aortic compliance studied by magnetic resonance imaging: the effects of age, training, and coronary artery disease. Br Heart J. 1989;62:90 96. 368 Cardiovascular Magnetic Resonance

Pulse wave velocity (m/sec)

Compliance (mL/mm Hg)

ISCHEMIC HEART DISEASE

of reflected waves on measured flow waves might provide interesting insights into the nature of the more distal vessels.

3 0

B

20

40 60 Compliance (mL/mm Hg)

80

Figure 27 7 Flow wave velocity is directly proportional to age (A) and inversely proportional to regional aortic compliance (B). Source: Mohiaddin RH, Firmin DN, Longmore DB. Age related changes of human aortic flow wave velocity measured non invasively by magnetic resonance imaging. J Applied Physiol. 1993;74:492 497.

assessment of aortic area was made at the level of the pul monary bifurcation by CMR and plotted against a surrogate for central aortic pressures (using a finger arterial blood pressure derived brachial blood pressure). Using pres sure area lines, a subgroup with a transition point (a depar ture from a linear pressure area relationship) was identified in 6 out of 32 patients. It was suggested that this point represented the recruitment of collagen to load bearing elements. These patients demonstrated a trend toward reduced distensibility, though this relationship might have been stronger had the study group been larger. It was hypothesized that patients in whom the transition point is seen at higher blood pressures might experience greater improvements in distensibility with beta blockade than those whose transition points occurred at lower blood pressures. Savolainen and colleagues51 used CMR and indirect bra chial artery blood pressure measurements to assess aortic elastic modulus in patients with essential hypertension prior to therapy and after 3 weeks and 6 months of therapy with cilazapril (an angiotensin converting enzyme inhibi tor) or atenolol (a beta 1 adrenergic blocker). The authors concluded that 6 months of treatment with either cilazapril or atenolol reduces the stiffness of the ascending aorta in essential hypertension. No statistically significant differ ences between the effects of the two drugs were observed. Honda and associates used CMR to measure aortic distensi bility in patients with systemic hypertension and demon strated that the antihypertensive drugs nicardipine and alacepril have a beneficial effect on aortic distensibility.52 Resnick and colleagues53 assessed aortic distensibility, left ventricular mass index, abdominal fat (subcutaneous and visceral), and free magnesium levels in the brain and skel etal muscle by CMR. In patients with essential hyperten sion, the following were concluded: Systolic hypertension and increased left ventricular mass index may result from arterial stiffness; arterial stiffness may be one mechanism by which abdominal visceral fat contributes to cardiovascu lar risk; and decreased magnesium contributes to arterial stiffness in hypertension. Chelsky and coworkers54 used CMR to measure aortic compliance in nine premenopausal women before and after menotropin therapy. They demonstrated that a short term rise in estrogen induced by menotropin treatment was associated with an increase in aortic compliance. Aortic size was not significantly increased within this time frame. Bogren and colleagues55 used CMR to study pulmonary artery distensibility in healthy subjects (Fig. 29 8) and in patients with pulmonary arterial hypertension (Fig. 29 9). The distensibility was found to be significantly lower in pulmonary arterial hypertension than in normal subjects, but there was no age related difference. The results also demonstrated a small retrograde flow (2%) in the pulmo nary trunk of normal subjects close to the pulmonic valve. Antegrade plug flow occurred in most normal subjects but varied among individuals. There were also other variations in the flow pattern among normal individuals. All patients with pulmonary arterial hypertension had a markedly irreg ular antegrade and retrograde flow and a large retrograde flow (average 26%).

Cardiovascular Magnetic Resonance 369

27 ASSESSMENT OF THE BIOPHYSICAL MECHANICAL PROPERTIES OF THE ARTERIAL WALL

been based on carotid ultrasound measurements. Forbat and colleagues44 measured aortic compliance, coronary cal cification, and carotid intimal medial thickness during reduction of cholesterol level in hypercholesterolemic patients with and without coronary artery disease. All received fluvastatin for 1 year. Aortic compliance was assessed by using CMR, and the coronary calcification score was determined by electron beam computed tomog raphy. Carotid intimal medial thickness was measured by carotid ultrasound. The authors showed an improvement in aortic compliance over 1 year, which indicates that the lipid changes induced by fluvastatin (an increase in high density lipoprotein level, decrease in very low density lipo protein level, and improvement in low: high density lipoprotein ratio) beneficially influenced vascular patho physiology. In the patients who were studied with carotid ultrasound means, carotid intimal medial thickness decreased from 1.09 to 0.87 mm (p ¼ 0.004), corroborat ing these results. Kupari and colleagues45 measured aortic elastic modu lus by CMR in asymptomatic subjects and correlated these measurements with physical activity, ethanol consumption, systolic blood pressure, fasting blood lipids, and serum insulin. They showed that the average value of the ascend ing and descending aortic elastic modulus was associated positively and statistically significantly with blood pressure, physical inactivity, serum insulin, and high density lipopro tein. The elastic modulus was associated negatively with the ratio of low density lipoprotein cholesterol to high density lipoprotein cholesterol. No association between aortic elastic modulus and either smoking or ethanol con sumption was demonstrated in this study. The same group demonstrated a higher aortic elastic modulus in patients with Marfan syndrome than in healthy subjects, indicating a relative decrease in the distensibility of the thoracic aorta.46 Kupari and colleagues also demonstrated that aortic flow wave velocity was more reproducible (interob server and intraobserver) than measurement of the pulsatile aortic area change or the elastic modulus. However, inter study reproducibility has not been tested.46 Adams and colleagues demonstrated abnormal aortic distensibility and stiffness index in patients with Marfan syndrome using CMR.47 These findings were confirmed by Groenink and associates using a CMR derived measure of distensibility and aortic PWV. Arrival time was deter mined by the point at which flow had reached half its maximum value.48 Beta adrenergic blocking agents may reduce the rate of aortic root dilation and the development of aortic complications in patients with Marfan syndrome. This may be due to beta blocker induced changes in aortic stiffness. To investigate this, Groenink and colleagues used CMR to measure aortic distensibility and aortic pulse wave velocity to assess aortic stiffness in Marfan syndrome and healthy volunteers before and after beta blocker therapy.49 They showed that in both groups, mean blood pressure decreased significantly but only the Marfan syndrome patients had a significant increase in aortic distensibility at multiple levels and a significant decrease in pulse wave velocity after beta blocker therapy. The same group50 sought to explain why some of these patients responded to beta blockers while others did not. Cross sectional

ISCHEMIC HEART DISEASE

A

B

C

D

Figure 27 8 Right ventricular outflow tract image (A) and oblique sagittal and transverse image (B) of the main pulmonary artery, showing the sites where main pulmonary artery distensibility was measured. In the bottom row, the change in the main pulmonary artery cross sectional area between diastole (C) and systole (D) demonstrated a large change in a 25 year old healthy subject.

Changes in the cross sectional area of the pulmonary artery in a magnitude image from a phase velocity sequence have been measured to derive pulmonary artery PWV,56–58 Values for normal subjects and for patients with pulmonary hypertension have been derived.57 In the latter group, max imum and minimum values for mean pulmonary artery pressure (mPAP) were predicted that “framed” the actual mPAP reliability.58 Finally, Stefanides and colleagues have showed that aor tic stiffness has prognostic power in determining the likeli hood of future cardiac events (Fig. 29 10).59 This interesting study merits further examination, as it uses a simple marker of disseminated arterial disease that is sim ple to measure in large populations.

ASSESSMENT OF ENDOTHELIAL FUNCTION Brachial artery reactivity testing (BAR) has been proposed as a biomarker of endothelial function.60 In the normal

370 Cardiovascular Magnetic Resonance

subject, release of a previously occluded brachial blood pressure cuff results in postischemic hyperemia (Fig. 29 11) 61 and subsequent increased shear stress of the local arterial wall. This, in turn, causes the release of endothe lium derived nitric oxide, a potent vasodilator. The extent to which nitric oxide is released (itself dependent on the amount produced and stored locally) can be indirectly assessed by measuring the degree of vessel dilation that results from this provocation. Additionally, endothelium independent function can be assessed by using dilation mediated by glyceryl trinitrate (GTN) (normally adminis tered sublingually). The noninvasive, uncomplicated and well tolerated nature of this examination makes it an attractive endpoint in epidemiologic studies. This has been further strength ened as alterations in flow mediated response (by ultra sound) have been shown to precede clinical manifestations of disease. These findings suggest a useful role for this biomarker as a screening tool in the future; allowing risk stratification and intervention at an earlier stage than might previously have been thought possible.

27 ASSESSMENT OF THE BIOPHYSICAL MECHANICAL PROPERTIES OF THE ARTERIAL WALL

A

B

C

D

Figure 27 9 Right ventricular outflow tract image (A) and oblique sagittal and transverse image (B) of the main pulmonary artery, showing the sites where main pulmonary artery distensibility was measured. In the bottom row, the change in the main pulmonary artery cross sectional area between diastole (C) and systole (D) demonstrated little change in a patient with pulmonary hypertension.

While the peripheral location of the brachial artery allows ultrasound ready access to such a measure, CMR can also be employed in a similar fashion. In this way, BAR can be used as an adjunct, combining with more rou tine measures of ventricular and vascular function to pro vide a powerful means by which to evaluate patients in research (Fig. 29 12). CMR offers other advantages over ultrasonography for this measure as it is more reproducible and less operator dependent.62 This has logistical and eco nomic consequences for researchers undertaking such work, as it allows sample size to be reduced without compromising the ability to identify statistically significant changes. Also CMR measures the true cross sectional area, whereas ultrasound usually measures diameter only. In addition, assessment of reactive hyperemic response using real time CMR flow imaging has been shown to be asso ciated with cardiovascular increased risk.63 CMR has been used to demonstrate pertubations of endothelial function in a variety of different scenarios. Sorenson and coworkers64 were able to demonstrate impairment of flow mediated dilation, whose degree was

inversely related to circulating levels of estradiol, in patients who were given depot medroxyprogesterone ace tate when compared to controls. Wiesmann and col leagues64 showed that brachial artery flow mediated dilation was significantly reduced in smokers in compari son with nonsmokers. As with Sorenson and colleagues’ work, impairment of dilation was endothelium dependent only as degree of GTN mediated dilation was similar in the two groups. Conversely, reduced endothelium indepen dent (but not endothelium dependent) function has been demonstrated in young elite rowers.65

ARTERIAL WALL SHEAR STRESS The use of detailed anatomic model and boundary condi tions, such as the inlet and outlet flow, captured in CMR are important for the generation of a patient specific computational fluid dynamic (CFD) simulation. The CFD Cardiovascular Magnetic Resonance 371

Ratio free of all cardiac endpoints

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

10

20

30 Months

40

50

60

Figure 27 10 Relationship of aortic stiffness to occurrence of any cardiac end point, with the study population divided into terciles (p 0.001). Source: Stefanides C, Dernellis J, Tsiamis E, et al. Aortic stiffness as a risk factor for recurrent acute coronary events in patients with ischaemic heart disease. Eur Heart J. 2000;21:390.

method allows the calculation of features and properties such as wall shear stress (WSS) and mass transfer rate, which are difficult to measure directly with imaging but are important to the understanding of basic hemodynam ics. CFD is the technique of using numerical methods to solve equations that govern the fluid flow. The basic premise of CFD is to split the domain that is to be ana lyzed into small volumes or elements. For each of these, a set of partial differential equations is solved, which approximates a solution for the flow in order to achieve the basics of conservation of mass, momentum, and energy for each volume or element. The three basic equa tions are solved simultaneously, with any additional

Velocity (mm/sec)

ISCHEMIC HEART DISEASE

1.0

800 700 600 500 400 300 200 100 0 −100 −200 −300

equations implemented in a particular model to obtain the flow velocities and pressure and therefore any derived quantities, such as shear stress. CFD has been extensively used to model physiologic flow and arterial WSS, particu larly in the carotid bifurcation, in attempts to better understand the mechanism of atherosclerosis in this region66 and to give insight into the pathogenesis of arterial aneurysms.67 WSS is defined as the mechanical frictional force exerted on the vessel wall by flowing blood and is the product of the blood viscosity with the velocity gradient at the vessel wall. Low or oscillating arterial WSS is correlated with atherosclerosis.68 In the aorta, once an aneurysm is formed, fluctuation in blood flow within it can induce vibrations of the aneu rysm wall that can contribute to progression and eventual rupture. Volume flow measured by CMR alone is not yet capable of sufficient spatial and temporal resolution for accurate measurement of WSS. Simulation with CFD pro vides more information than current diagnostic tools. In particular, it allows the determination of the shear stress level, which in turn helps in identifying regions that are susceptible to aneurysm formation, growth, and rupture. The only criterion that has been used so far for the selec tion of surgical patients has been the maximum diameter and the rate of change of aortic diameter.69 This is based on Laplace’s law for hollow circular pipes, which states that the maximum stress within the arterial wall is propor tional to the radius. Works in literature suggest that the peak wall stress is a more reliable parameter for the assessment of the rupture risk of aortic aneurysms. Fill inger and colleagues reported that the peak wall stress in aneurysms has a higher sensitivity and specificity for pre dicting the rupture than maximum diameter.70 CMR and CFD are promising tools for stress and velocity patterns simulation using patient specific flow condition. Fig ures 27 13 and 27 14 show how the stress pattern and values are influenced by the shape of the aneurysm and the presence of intraluminal thrombus.

Cuff release

Cuff inflated 1-minute gap

Baseline Cardiac cycles (R-R ≈ 900 msec)

Figure 27 11 Mean velocity in real time shown graphically for 72 cardiac cycles. The first 5 cardiac cycles are before cuff inflation. Imaging was suspended for the cuff inflation and a 5 minute delay. The next 10 cycles were acquired with the cuff inflated and show a shorter forward peak (i.e., reduced forward flow) in the waveform, although the peak velocity does not reduce. The occlusion cuff was placed more than 10 cm distal to the imaging plane. Note the increased reverse flow during the occlusion. After release of the cuff, the forward peak is longer, and the reverse flow changes to forward flow. Both effects recover to baseline after approximately 40 sec. Source: Mohiaddin RH, Gatehouse PD, Moon JCC, Youssuffidin M, Yang GZ, Firmin DN and Pennell DJ. Assessment of reactive hyperemia using real time zonal echo planar flow imaging. J Cardiovasc Magn Reson. 2002;4:283 287.

372 Cardiovascular Magnetic Resonance

B

C

D

27 ASSESSMENT OF THE BIOPHYSICAL MECHANICAL PROPERTIES OF THE ARTERIAL WALL

A

Figure 27 12 Assessment of brachial artery reactivity. The top panel demonstrates the arrangement of the surface coil and cuff on a subject’s right arm whose motion is restricted by sandbags. The transxial plane in which the brachial artery is at its most superficial (A) is used to prescribe a plane perpendicular to the brachial artery (B). Arterial reactivity is then assessed through peripheral reactive hyperemia (C) and sublingual glyceryl trinitrate (D). Source: Sorenson MB, Collins P, Ong PJL, et al. Long term use of contraceptive depot medroxyprogesterone acetate in young women impairs arterial endothelial function assessed by cardiovascular magnetic resonance. Circulation. 2002;106:1646 1651.

Cardiovascular Magnetic Resonance 373

ISCHEMIC HEART DISEASE

A

B Smoothed Effective Stress RST CALC Time 1.000

64.50 57.00 49.50 42.00 34.50 27.00 19.50 12.00 4.50

C

Stress pattern at peak systole (units:kPa)

Figure 27 13 A, Surface rendered contrast enhanced magnetic resonance angiography in the left anterior oblique and right anterior oblique views in a patient with aortic arch aneurysm at the site of previous coarctation repair (left subclavian artery flap). B, The aortic arch morphology was segmented from the CMR images, and the inflow outflow flow conditions were extracted from CMR flow mapping. C, Stress pattern at peak systole (units: kilopascals). Regions of high shear stress (arrows) are found in the aortic arch and at the entry of the aneurysmal bulge in the first model, with low values only in the distal region of the aneurysm.

374 Cardiovascular Magnetic Resonance

27 ASSESSMENT OF THE BIOPHYSICAL MECHANICAL PROPERTIES OF THE ARTERIAL WALL

A

B

Smoothed Effective Stress RST CALC Time 1.000

64.50 57.00 49.50 42.00 34.50 27.00 19.50 12.00 4.50

C Figure 27 14 A, Surface rendered contrast enhanced magnetic resonance angiography in the coronal and sagittal views in a patient with atherosclerotic aneurysm in the distal descending thoracic aorta. B, The aneurysm contains a large crescent shape mural thrombus. C, Stress pattern at peak systole (units: kilopascals). Low values of shear stress throughout the aneurysm are noted where the thrombus appears to absorb part of the pressure load resulting in low stress in the wall.

CONCLUSION Atherosclerosis consists of two components, the most often discussed being atherosis, relating to plaque genesis and composition. Sclerosis is often the forgotten partner in clini cal practice, but measurement of vessel wall stiffening has now been shown to be useful both diagnostically and prog nostically. Coupled with CMR’s powerful role in excluding secondary causes of hypertension and accurately assessing ventricular function, CMR is an ideal tool for its investiga tion, and further studies of the clinical role of parameters

of sclerosis can be expected. The capability of CMR to assess brachial artery reactivity and to measure WSS, in conjunc tion with CFD, will enhance further its role in the studies of the biophysical properties of the arterial wall.

ACKNOWLEDGMENTS The author acknowledges the helpful contributions of Peter Gatehouse, PhD, and William Bradlow, MD, in the prepara tion of this manuscript.

Cardiovascular Magnetic Resonance 375

ISCHEMIC HEART DISEASE

References 1. Band W, Goedhard WJ, Knoop AA. Comparison of effects of high cholesterol intake on viscoelastic properties of the thoracic aorta in rats and rabbits. Atherosclerosis. 1973;18:163–172. 2. Banga I, Balo J. Elasticity of the vascular wall. 1 The elastic distensibility of the human carotid as a function of age and arteriosclerosis. Acta Physiol Acad Sci Hung. 1961;20–21:237–247. 3. Farrar DJ, Green HD, Wanger WD, Bond MG. Reduction in pulse wave velocity and improvement of aortic distensibility accompanying regression of atherosclerosis in Rhesus monkey. Circ Res. 1980;47:425–432. 4. Farrar DJ, Bond GM, Riley WA, Sawyer JK. Anatomic correlates of aortic pulse wave velocity and carotid artery elasticity during atherosclerosis progression and regression in monkeys. Circulation. 1991;83:1754–1763. 5. Ohtsuka S, Kakihana M, Watanabe H, et al. Chronically decreased aortic distensibility causes deterioration of coronary perfusion during increased left ventricular contraction. J Am Coll Cardiol. 1994;24:1406–1414. 6. Isnard RN, Pannier BM, Laurent S, et al. Pulsatile diameter and elastic modulus of the aortic arch in essential hypertension: a non-invasive study. J Am Coll Cardiol. 1989;13:399–405. 7. Handler CE, Child A, Light NM. Mitral valve prolapse, aortic compliance, and skin collagen in joint hypermotility syndrome. Br Heart J. 1985;54:501–508. 8. Lehman ED, Gosling RG, Sonksen PH. Arterial compliance in diabetes. Diabetes Care. 1986;9:27–31. 9. Lehman ED, Watts GF, Gosling RG. Aortic distensibility and hypercholesterolaemia. Lancet. 1992;340:1171–1172. 10. Lehman ED, Hopkins KD, Weissberger AJ, Gosling RG, Sonksen PH. Aortic distensibility in growth hormone deficient adults. Lancet. 1993;341:309. 11. Harris P, Heath D. The relation between structure and function in the blood vessels of the lung in pulmonary hypertension. In: Harris P, Heath D, eds. The Human Pulmonary Circulation. 3rd ed. London: Churchill Livingstone; 1986:284–297. 12. Reuben SR. Compliance of the human pulmonary arterial system in disease. Circ Res. 1971;29:40–50. 13. Harris P, Heath D, Apostolopoulos A. Extensibility of the pulmonary trunk in heart disease. Br Heart J. 1965;27:660–666. 14. Paz R, Mohiaddin RH, Longmore DB. Magnetic resonance assessment of pulmonary trunk: anatomy, flow, pulsatility and distensibility. Eur Heart J. 1993;14:1524–1530. 15. Mohiaddin RH, Paz R, Theodoropolus S, Firmin DN, Longmore DB, Yacoub MH. Magnetic resonance characterization of pulmonary arterial blood flow following single lung transplantation. J Thorac Cardiovasc Surg. 1991;101:1016–1023. 16. Mukherjee DP, Kagan HM, Jordan RE, Franzblau C. Effect of hydrophobic elastin ligands on the stress-strain properties of elastin fibers. Connect Tissue Res. 1976;4:177. 17. Harkness ML, Harkness RD, McDonald DA. The collagen and elastin content of the arterial wall in the dog. Proc Roy Soc [Biol]. 1957;146:541–551. 18. Lee RT, Kamm RD. Vascular mechanics for the cardiologist. J Am Coll Cardiol. 1994;23:1289–1295. 18a. Peterson LN, Jensen RE, Parnell R. Mechanical properties of arteries in vivo. Circ Res. 1968;8:622–639. 19. Bergel DH. The dynamic elastic properties of the arterial wall. J Physiol. 1961;156:458–469. 20. Hardung V. Vergleichende messungen der dynamischen elastizitat und viskositat von blutegfassen, kautschuk und synthetischen elastomeren. Helv Physiol Acta. 1953;11:194–211. 21. Learoyd BM, Taylor MG. Alterations with age in the visco-elastic properties of human arterial walls. Circ Res. 1966;18:278–292. 22. Remington JW. Hysteresis loop, behaviour of the aorta and other extensible tissues. Am. J Physiol. 1955;180:83–95. 23. Bramwell JC, Hill AV, McSwiney BA. The velocity of the pulse wave in man in relation to age as measured by hot-wire sphygmograph. Heart. 1923;10:233–255. 24. Hallock P. Arterial elasticity in man in relation to age as evaluated by the pulse wave velocity method. Arch Int Med. 1934;54:770–798. 25. Remington JW. Pressure-diameter relations of the in vivo aorta. Am J Physiol. 1962;203:440–448.

376 Cardiovascular Magnetic Resonance

26. Luchsinger PC, Sachs M, Patel D. Pressure-radius relationship in large blood vessels of man. Circ Res. 1962;11:885–887. 27. Stefanadis C, Stratos C, Boudoulas H, et al. Distensibility of the ascending aorta: comparison of invasive and non-invasive techniques in healthy men and in men with coronary artery disease. Eur Heart J. 1990;11:990–996. 28. Gosling RG, King DH. Arterial assessment by Doppler-shift ultrasound. Proc Roy Soc Med. 1974;67:447–449. 29. Kok WEM, Peters RJG, Prins MH, et al. Contribution of age and intimal lesion morphology to coronary artery wall mechanics in coronary artery disease. Clin Sci. 1995;89:239–246. 30. Hanrath P, Heinzt B, Dahl V, et al. Evaluation of segmental elastic properties of the aorta in normotensive and medically treated patients by intravascular ultrasound. In: Boudoulas P, Toutouzas PK, Wooley C, eds. Functional Abnormality of the Aorta. New York: Futura Publishing Company; 1996:221–231. 31. Mohiaddin RH, Underwood SR, Bogren HG, et al. Regional aortic compliance studied by magnetic resonance imaging: the effects of age, training, and coronary artery disease. Br Heart J. 1989;62:90–96. 32. Chien D, Saloner D, Laub G, Anderson CM. High resolution cine MRI of vessel distension. J Comput Assist Tomogr. 1994;18:576–580. 33. Rueckert D, Burger P, Yang GZ, Forbat SM, Mohiaddin RH. Automatic tracking of the aorta in cardiovascular MR images using deformable models. IEEE Trans Med Imaging. 1997;16:581–590. 34. Mohiaddin RH, Firmin DN, Longmore DB. Age-related changes of human aortic flow wave velocity measured non-invasively by magnetic resonance imaging. J Appl Physiol. 1993;74:492–497. 35. Tarnawski M, Cybulski G, Doorly D, Dumoulin C, Darrow R, Caro C. Noninvasive determination of local wavespeed and distensibility of the femoral artery by comb-excited Fourier velocity-encoded magnetic resonance imaging: measurements on athletic and nonathletic human subjects. Heart Vessels. 1994;9:194–201. 36. Dumoulin CL, Doorly DJ, Caro CG. Quantitative measurement of velocity at multiple positions using comb excitation and Fourier velocity encoding. Magn Reson Med. 1993;29:44–52. 37. Hardy CJ, Bolster BD, McVeigh ER, Adams WJ, Zerhouni EA. A onedimensional velocity technique for NMR measurement of aortic distensibility. Magn Reson Med. 1994;31:513–520. 38. Hardy CJ, Bolster Jr BD, McVeigh ER, Iben IE, Zerhouni EA. Pencil excitation with interleaved Fourier velocity encoding: NMR measurement of aortic distensibility. Magn Reson Med. 1996;35:814–819. 39. Bolster Jr BD, Atalar E, Hardy CJ, McVeigh ER. Accuracy of arterial pulse-wave velocity measurement using MR. J Magn Reson Imaging. 1998;8:878–888. 40. Mohiaddin RH, Underwood SR, Bogren HG, et al. Regional aortic compliance studied by magnetic resonance imaging: the effects of age, training, and coronary artery disease. Br Heart J. 1989;62:90–96. 41. Rees RSO, Somerville J, Ward C, et al. Magnetic resonance imaging in late post-operative assessment of coarctation of the aorta. Radiology. 1989;173:499–502. 42. Manzara CC, Mohiaddin RH, Pennell DJ, et al. Magnetic resonance assessment of thoracic aorta in Marfan’s syndrome. American Heart Association, Dallas. Circulation. 1990;82(suppl III):497(Abstract). 43. Rogers WJ, Hu YL, Coast D, et al. Age-associated changes in regional aortic pulse wave velocity. JACC. 2001;38(4):1123–1129. 44. Forbat SM, Naoumova RP, Sidhu PS, et al. The effect of cholesterol reduction with fluvastatin on aortic compliance, coronary calcification and carotid intimal-medial thickness: a pilot study. J Cardiovasc Risk. 1998;5:1–10. 45. Kupari K, Hekali P, Keto P, et al. Relation of aortic stiffness to factors modifying the risk of atherosclerosis in healthy persons. Arterioscler Thromb. 1994;14:386–394. 46. Kupari K, Keto P, Hekali P, Poutanen V, Savolainen A, StandertskjoldNordenstam CG. Cine magnetic resonance imaging in the assessment of aortic distensibility. In: Boudoulas P, Toutouzas PK, Wooley C, eds. Functional Abnormality of the Aorta. New York: Futura Publishing Company; 1996:247–268. 47. Adams JN, Brooks M, Redpath TW, et al. Aortic distensibility and stiffness index measured by magnetic resonance imaging in patients with Marfan’s syndrome. Br Heart J. 1995;73:265–269.

59. Stefanides C, Dernellis J, Tsiamis E, et al. Aortic stiffness as a risk factor for recurrent acute coronary events in patients with ischaemic heart disease. Eur Heart J. 2000;21:390–396. 60. Deanfield J, Donald A, Ferri C, et al. Working Group on Endothelin and Endothelial Factors of the European Society of Hypertension. Endothelial function and dysfunction. Part II: Association with cardiovascular risk factors and diseases. A statement by the Working Group on Endothelins and Endothelial Factors of the European Society of Hypertension. J Hypertens. 2005;23(2):233–246. 61. Mohiaddin RH, Gatehouse PD, Moon JCC, et al. Assessment of reactive hyperemia using real time zonal echo planar flow imaging. J Cardiovasc Magn Reson. 2002;4:283–287. 62. Sorenson MB, Collins P, Ong PJL, et al. Long term use of contraceptive depot medroxyprogesterone acetate in young women impairs arterial endothelial function assessed by cardiovascular magnetic resonance. Circulation. 2002;106:1646–1651. 63. Schwitter J, Oelhafen M, Wyss BM, et al. 2D spatially selective real time magnetic resonance imaging for the assessment of microvasculature function and its relation to the cardiovascular risk profile. J Cardiovasc Magn Reson. 2006;8:759–769. 64. Wiesmann F, Petersen SE, Leeson PM, et al. Global impairment of brachial, carotid, and aortic vascular function in young smokers: direct quantification by high-resolution magnetic resonance imaging. J Am Coll Cardiol. 2004;16;44(10):2056–2064. 65. Petersen SE, Wiesmann F, Hudsmith LE, et al. Functional and structural vascular remodeling in elite rowers assessed by cardiovascular magnetic resonance. J Am Coll Cardiol. 2006;48(4):790–797. 66. Milner JS, Moore JA, Rutt BK, Steinman DA. Hemodynamics of human carotid artery bifurcations: computational studies with models reconstructed from magnetic resonance imaging of normal subjects. J Vasc Surg. 1998;28(1):143–156. 67. Borghi A, Wood NB, Mohiaddin RH, Xu XY. 3D geometric reconstruction of thoracic aortic aneurysms. Biomed Eng Online. 2006;5:59. 68. Malek AM, Alper SL, Izumo S. Hemodynamic shear stress and its role in atherosclerosis. JAMA. 1999;282:2035–2042. 69. Elefteriades JA. Natural history of thoracic aortic aneurysms: indications for surgery and surgical versus nonsurgical risks. Ann Thorac Surg. 2002;74:S1877–S1880. 70. Fillinger MF, Marra PS, Raghavan ML, Kennedy EF. Prediction of rupture risk in abdominal aortic aneurysm during observation: wall stress versus diameter. J Vasc Surg. 2003;37(4):724–732.

Cardiovascular Magnetic Resonance 377

27 ASSESSMENT OF THE BIOPHYSICAL MECHANICAL PROPERTIES OF THE ARTERIAL WALL

48. Groenink M, de Roos A, Mulder BJ, et al. Biophysical properties of the normal-sized aorta in patients with Marfan syndrome: evaluation with MR Flow mapping. Radiology. 2001;219:535–540. 49. Groenink M, de Roos A, Mulder BJ, Spaan JA, van der Wall EE. Changes in aortic distensibility and pulse wave velocity assessed with magnetic resonance imaging following beta-blocker therapy in the Marfan syndrome. Am J Cardiol. 1998;82:203–208. 50. Nollen GJ, Westerhof BE, Groenink M, Osnabrugge A, Van der Wall EE, Mulder BJM. Aortic pressure-area relation in Marfan patients with and without B blocking agents: a new invasive approach. Heart. 2004;90:314–318. 51. Savolainen A, Keto P, Poutanen VP, et al. Effects of angiotensin-converting enzyme inhibition versus beta-adrenergic blockade on aortic stiffness in essential hypertension. J Cardiovasc Pharmacol. 1996;27:99–104. 52. Honda T, Hamada M, Shigematsu Y, Matsumoto Y, Matsuoka H, Hiwada K. Effect of antihypertensive therapy on aortic distensibility in patients with essential hypertension: comparison with trichlormethiazide, nicardipine and alacepril. Cardiovasc Drugs Ther. 1999;13:339–346. 53. Resnick LM, Militianu D, Cunnings AJ, Pipe JG, Evelhoch JL, Soulen RL. Direct magnetic resonance determination of aortic distensibility in essential hypertension: relation to age, abdominal visceral fat, and in situ intracellular free magnesium. Hypertension. 1997;30 (3 Pt 2):654–659. 54. Chelsky R, Wilson RA, Morton MJ, et al. Alteration of ascending thoracic aorta compliance after treatment with menotropin. Am J Obstet Gynecol. 1997;176:1255–1259. 55. Bogren HG, Klipstein RH, Mohiaddin RH, et al. Pulmonary artery distensibility and blood flow patterns: a magnetic resonance study of normal subjects and of patients with pulmonary arterial hypertension. Am Heart J. 1989;118:990–999. 56. Laffon E, Bernard V, Montaudon M, Marthan R, Barat JL, Laurent F. Tuning of pulmonary arterial circulation evidenced by MR phase mapping in healthy volunteers. J Appl Physiol. 2001;90(2):469–474. 57. Laffon E, Laurent F, Bernard V, De Boucaud L, Ducassou D, Marthan R. Noninvasive assessment of pulmonary arterial hypertension by MR phase-mapping method. J Appl Physiol. 2001;90 (6):2197–2202. 58. Laffon E, Vallet C, Bernard V, et al. A computed method for noninvasive MRI assessment of pulmonary arterial hypertension. J Appl Physiol. 2004;96(2):463–468.

Cardiovascular Magnetic Resonance Assessment of Right Ventricular Anatomy and Function Alicia M. Maceira and Dudley J. Pennell

Accurate noninvasive assessment of right ventricular (RV) mass and systolic function is important in several patholo gies, such as grown up congenital heart disease, pulmonary hypertension, and arrhythmogenic RV cardiomyopathy. This chapter aims to summarize the features of the normal right ventricle, briefly describe cardiovascular magnetic res onance (CMR) techniques for assessing RV dimensions and function, and give reference values for the assessment of the right ventricle.

NORMAL RIGHT VENTRICULAR ANATOMY The right ventricle is a thin, highly trabeculated structure that is triangular in form and, on gross inspection, appears to be wrapped around the left ventricle. The anterosuperior wall of the right ventricle is rounded and convex, its infe rior surface is flattened and forms a small part of the dia phragmatic surface of the heart, and its posterior wall is formed by the ventricular septum, which bulges into the right ventricle, owing to the much greater left ventricular (LV) systolic pressure,1 so a transverse section of the cavity presents a semilunar outline. The right ventricle has a con tinuum of muscle bands that rotate by approximately 160 from the epicardium to the endocardium.2 The principal axis of these fibers is oblique to the long axis of the right ventricle. In the normal adult, the total RV free wall mass is 26  5 g/m2. The right ventricle has several distinctive features. In its upper left portion, there is a conical pouch called the conus arteriosus or infundibulum, from which the pulmonary artery arises. A tendinous band connects the posterior sur face of the conus arteriosus to the aorta. Also, the RV wall is thinner than the LV wall, the proportion between them being as 1 to 3;3 it is thickest at the base and gradually becomes thinner toward the apex. The whole inner surface except the conus arteriosus is covered by more or less prominent muscular columns called trabeculae carneae and from some of them (papillary muscles), the chordae tendinae connect the myocardium to the tricuspid valve, which is more apically placed than the septal leaflet of

the mitral valve. Finally, a muscular band frequently extends from the base of the anterior papillary muscle to the ventricular septum. This band is considered to prevent overdistension of the ventricle and is called the moderator band.4 The depictions of the moderator band, the infundibu lum, and the different levels of insertion of the tricuspid and mitral septal leaflets are important diagnostic features for identification of the right ventricle, which can be diffi cult in some congenital cardiomyopathies.

Importance of Assessing Right Ventricular Dimensions and Function The measurement of RV dimensions, morphology, and function is important in several situations, such as congen ital heart disease, LV heart failure, pulmonary hypertension, pulmonary embolism, valvular heart disease, lung disease, and arrhythmogenic RV cardiomyopathy. RV failure may result from conditions that lead to impaired RV contractility, such as RV infarction, right sided cardiomyopathies, or severe sepsis; RV pressure overload, including pulmonic stenosis, pulmonary primary hyperten sion, and pulmonary hypertension with left heart disease, lung disease, or thromboembolic disease; and RV volume overload, for instance, tricuspid regurgitation. Many disor ders, such as corrected and uncorrected adult congenital heart disease and intracardiac shunts, may result in right ventricle failure through a combination of pathophysiologic mechanisms. Also, decompensated right ventricle (both acute and acute on chronic) is increasing as the prevalence of predisposing conditions grows.5 The prognostic value of RV function has been shown in several conditions such as left heart failure due to coronary or valvular heart disease,6,7 pulmonary hypertension,8 or myocardial infarction.9 Thus, the early detection of RV dys function can have an impact on therapeutic decision making and on prognosis. Finally, improved understanding of the RV response to pressure and volume overload might lead to more optimal surgical and medical treatments.

Cardiovascular Magnetic Resonance 381

28 CARDIOVASCULAR MAGNETIC RESONANCE ASSESSMENT OF RIGHT VENTRICULAR ANATOMY AND FUNCTION

CHAPTER 28

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

Techniques for Assessing Right Ventricular Dimensions and Function Angiography used to be the gold standard for assessment of RV volumes and regional and global function. But this technique is invasive, involves ionizing radiation and the use of potentially nephrotoxic contrast, and is not as accu rate as CMR.10 Echocardiography and radionuclide ventri culography have been used for the assessment of RV dimensions and function. More recently, “nongeometric” techniques such as three dimensional (3D) echocardiogra phy, CMR, and multidetector row computed tomography (CT) permit accurate assessment of RV volumes, function, and mass.

Echocardiography Echocardiography is the most frequently used technique for assessing the right ventricle. and it can be used bed side in very ill patients. It provides information about RV morphology, dimensions, septum convexity, function, tricuspid regurgitation, and estimates of pulmonary arte rial pressure and RV pressure.11 But the assessment of the right ventricle with echocardiography has several lim itations. First, the location of the right ventricle behind the sternum restricts the window that can be accessed by the ultrasound beam. Second, the complex shape and thin walls of the right ventricle make it necessary to image the right ventricle from several projections. Third, the thick trabeculations in the chamber may be confused with a thrombus, a tumor, or hypertrophic cardiomyopathy. Finally, there is a lack of accurate mathematical models to quantify RV mass and volumes with M mode or two dimensional (2D) echocardiography, as quantitative mea surements are based on geometric assumptions that do not apply to the right ventricle. Other indicators of RV function are Doppler derived indices such as the myocar dial performance index,12 tissue Doppler measurements of myocardial velocities and time intervals, and strain and strain rate measurements of contractility.13 Transeso phageal echocardiography is another echocardiographic method of RV assessment, but it is semi invasive, is not well suited for evaluation of anteriorly positioned RVs, and requires special skills. Three dimensional echocardi ography has emerged as a more accurate and reproducible approach to ventricular quantitation, mainly by avoiding the use of geometric assumptions of the ventricular shape. Three methods have been proposed for the acquisition of temporal and positional data: real time volumetric scan ning, the use of positional locators or free hand scanning, and rotational systems.14 Real time 3D echocardiography is an on line acquisition of a 3D dataset of the heart with out the need for electrocardiographic and respiratory gat ing, which has a great potential for immediate assessment of ventricular function. However, these methods need a stable cardiac rhythm and constant cardiac function dur ing image acquisition, and there are other practical pro blems, such as full cardiac visualization, good quality endocardial border recognition for manual endocardial 382 Cardiovascular Magnetic Resonance

tracing, and time consumption. Faster data acquisition, acceleration of data processing and reconstruction, use of automatic border detection algorithms, improvement of spatial resolution, and development of stable intrave nous contrast agents that enhance the endocardial delin eation are being developed that should provide automatic, even on line, volume measurement. Three dimensional echocardiography has been compared with CMR for the evaluation of RV function, and improved results in comparison with 2D echocardiography have been obtained.15,16 Nonetheless, 3D echocardiography has been used mainly for the left ventricle, and little has been reported on 3D echocardiography of the RV.

Radionuclide Angiography This technique provides a reliable quantitative measure ment of ventricular function not based on geometric assumptions with good agreement with CMR.17 This tech nique works well for the left ventricle but not so well for the right ventricle, owing to problems such as the limited count numbers in this chamber and the overlap of other cardiac chambers.18–20 It also has disadvantages, such as poor resolution compared to other imaging modalities, the use of ionizing radiation, and the need for an adequate bolus injection for first pass studies and a regular rhythm. Therefore, it has been of limited use for the study of the right ventricle so far.

Multislice Computed Tomography Multislice CT (MSCT) is emerging as an alternative tech nique, especially for patients with implantable devices (a contraindication for CMR). However, MSCT uses ionizing radiation and potentially nephrotoxic contrast and requires a low and stable heart rate for image acquisition.21

CMR CMR has some important advantages over other imaging techniques, which have led to the growing enthusiasm for its use. CMR offers accurate and reproducible tomographic, static, or cine images of high spatial and temporal resolution in any desired plane without exposure to contrast agents or ionizing radiation. It allows the acquisition of true RV short axis images encompassing the entire RV with high spatial and temporal resolution, thereby providing highly accurate and reproducible quantitative RV mass and functional data regardless of its position in the thorax.22–25 Nowadays, this technique is considered the gold standard for quantitative assessment of RV volume, mass, and function.

Imaging Strategies for Cardiovascular Magnetic Resonance of the Right Ventricle Before the study begins, it is essential to obtain an accurate electrocardiographic gating with minimal ectopy and to instruct the patient in breath holding. Sometimes oxygen may be applied to improve breath hold length. Ventricular

ectopy can be a problem, mainly in patients with congenital heart disease or with suspicion of arrhythmogenic RV car diomyopathy (ARVC). If this condition is present, pretreat ment with an antiarrhythmic agent should be considered. Spin echo (black blood) sequences (including turbo spin echo, half Fourier acquisition single shot turbo spin echo, or spin echo echo planar imaging) are used for ana tomic assessment (Fig. 28 1) as well as to rule out possible fatty replacement/infiltration of the RV free wall as can be seen in RV dysplasia.26 Functional evaluation of the RV is performed by using gradient echo (white blood) cine imaging. In the past, to achieve full 3D coverage of the ventricle with conventional nonsegmented free breathing gradient echo cine sequences, a total scanning time of 30 minutes or more was required. On modern scanners with segmented fast imaging, a single cine can be acquired in just one breath hold of about 8 to 10 sec, allowing the whole stack of images to be acquired in 5 to 10 minutes,27 thereby reducing breathing and movement artifact (Fig. 28 2). Moreover, real time steady state free precession (SSFP) imaging can acquire the whole stack in just one breath hold with acceptable accuracy and image quality.28 In patients who are unable to hold their breath consistently, solutions using the same sequence with more signal averages or combined with navigator echo are successful with free breathing,29 with a slight increase in the scanning time. The RV is well shown in the transaxial plane from the tricuspid valve to the apex. This plane has been used for analysis of RV function and has been shown to yield good

agreement with pulmonary flow and LV stroke volume30,31 as well as better interobserver and intraobserver reproduc ibility than the short axis plane32 (Fig. 28 3). However, it can be problematic because it is subject to partial volume effects in the anterior and inferior walls. The question has been raised as to whether the RV volumes should always be measured in the axial orientation. Yet the interstudy reproducibility of RV measurements in the short axis orien tation is good,33 and in practice, this orientation allows both the LV and RV dimensions to be measured simultaneously. Simpson’s rule is used for measuring volumes and func tion. A standardized method of combined ventricular func tional analysis is described in Chapters 11 and 14. Briefly, a stack of contiguous tomographic slices is acquired that encompass the entire ventricles. It is critical to ensure that the most basal part of the free wall of the right ventricle in diastole is included in the most basal cine slice, as it can be easily truncated. Manual or semiautomatic planimetry of the endocardial borders of each ventricle at both end dias tole and end systole and of the epicardial borders at end diastole is done. Care must be taken to exclude the right and left atria as they come into the basal imaging planes during systole. The ventricular volume is equal to the sum of the endocardial areas multiplied by the distance between the centers of each slice (Fig. 28 4). The stroke volume is equal to the difference between the end diastolic volume (EDV) and the end systolic volume (ESV). The ejection fraction (EF) is calculated as the stroke volume divided by the EDV. The mass is calculated as the volume Cardiovascular Magnetic Resonance 383

28 CARDIOVASCULAR MAGNETIC RESONANCE ASSESSMENT OF RIGHT VENTRICULAR ANATOMY AND FUNCTION

Figure 28 1 Short axis and transaxial spin echo images of the heart in a healthy subject. The right ventricular free wall is perfectly depicted between the black blood pool and the black pericardium. Note that there is epicardial fat deposition around the right coronary artery (anterior atrioventricular groove), left anterior descending artery (anterior interventricular groove), and left circumflex (posterior atrioventricular groove). It is important to know the normal patterns of fat distribution to prevent false positive reading of scans in assessing patients for possible fat infiltration in arrhythmogenic right ventricular cardiomyopathy.

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

RV

LV

Figure 28 2 SSFP cine images in eight short axis planes from cines acquired from base to apex. The series of short axis slices begins in the atrioventricular groove in diastole and goes through the ventricles down to the apex, thus covering the whole left ventricle and right ventricle. In the most basal slice (upper left), the pulmonary artery will appear in systole because of atrioventricular ring descent. The high contrast between blood and myocardium allows the drawing of the endocardial and epicardial borders for volume and mass determinations. For the images to be accurate and reproducible, it is important to use always the same slice thickness and the same gap between slices. Also, retrospective gating should be used in order to acquire the whole cardiac cycle. LV, left ventricle; RV, right ventricle.

PA PV

RV

A Figure 28 3 A, To plan the transversal slices, a cine sequence through the right ventricular (RV) outflow tract aligned in an oblique sagittal plane should be acquired. This view is useful for examining the function of the right ventricular free wall from the apex to the pulmonary valve (PV) and visualizing pulmonary regurgitation. (Continued) 384 Cardiovascular Magnetic Resonance

PA

RV LV

RA LA

B Figure 28 3 Cont’d B, Multiple transaxial steady state free precession cine sequences in contiguous planes should be acquired from the pulmonary valve level down to the inferior right ventricular wall, using the right ventricular outflow tract view to pilot them. These planes are invaluable for examining regional right ventricular wall motion. Again, the slice thickness and gap between slices should be consistent among studies. Two features of the normal right ventricle are seen in this transversal stack: the muscular outflow tract and the more apical insertion of the tricuspid valve in the septum compared with the septal leaflet of the mitral valve. Ao, aorta; LA, left atrium; LV, left ventricle; PA, pulmonary artery; RA, right atrium.

of tissue occupied by the free wall multiplied by an assumed density of 1.05 g/cc. The volumes obtained by this method are independent of geometric assumptions and dimensionally accurate.34 Mass measurements agree well with autopsy studies.35,36 RV parameters obtained with this technique are reproducible, with an interobserver variabil ity of 6.3% for EDV, 8.6% for ESV, 7% for stroke volume, 4.4% for EF, and 7.8% for RV mass and an intraobserver variability of 3.6% for EDV, 6.5% for ESV, 5.9% for stroke volume, 4% for EF, and 5.7% for RV mass.37 With semiau tomated analysis, all images throughout the cardiac cycle can be planimetered, and diastolic function parameters can also be calculated. If necessary, other CMR techniques should be used in the assessment of the right ventricle. Flow velocity maps allow for accurate assessment of pulmonary valve regur gitation (regurgitant fraction) and systemic to pulmonary flow ratio. Magnetic resonance angiography must be done in case assessment of great vessel anatomy is advisable, such as in a number of congenital cardiac conditions.38 CMR with late gadolinium enhancement can detect myo cardial fibrosis in both ischemic and nonischemic cardiomyopathies.39,40

NORMAL RIGHT VENTRICULAR VOLUMES AND SYSTOLIC FUNCTION There have been a number of human series describing the normal characteristics of RV size and function using

autopsy,41,42 echocardiography,43–45 X ray angiography,19 radionuclide angiography,46 CMR,38,47–51 and ultrafast CT.52,53 We have reported on RV reference parameters for mass, volumes, and systolic and diastolic function using cine SSFP CMR techniques and analysis from 120 healthy adult subjects.37 These gender specific data are summarized in Tables 28 1 to 28 5. We observed that many clinical parameters of RV volumes and systolic and diastolic function are significantly dependent on gender, age, and body surface area (BSA). On multivariable analy sis, BSA was found to significantly influence RV mass, EDV, ESV, stroke volume, and early tricuspid peak filling rate (PFRE). Gender had a significant independent influ ence on absolute and normalized RV mass, EDV, and stroke volume (Figs. 28 5 and 28 6). It was also an inde pendent predictor of absolute and normalized active tri cuspid peak filling rate (PFRA, PFRA/BSA). There was a significant decrease with age of normalized RV mass and of absolute and normalized EDV and ESV in both males and females. There was a significant increase with age in absolute right ventricular ejection fraction (RVEF) in both males and females and a significant increase in normalized EF in males. For diastolic function, absolute and normal ized PFRE decreased significantly with age in males and females, while absolute and normalized PFRA increased in males. Accordingly, PFRE/ PFRA decreased significantly. In a multivariable analysis, age was an independent predictor of absolute and normalized ventricular mass and volumes (EDV, ESV, stroke volume, EDV/BSA, ESV/BSA, stroke vol ume/BSA) and of EF. It was also an independent predictor of diastolic variables (PFRE, PFRA, PFRE/ PFRA, PFRE/EDV, PFRA/EDV, PFRE/BSA, PFRA/BSA). Thus, the interpretation Cardiovascular Magnetic Resonance 385

28 CARDIOVASCULAR MAGNETIC RESONANCE ASSESSMENT OF RIGHT VENTRICULAR ANATOMY AND FUNCTION

Ao

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

RV

LV

A

B Figure 28 4 Summation of discs is used to quantify ventricular volumes and mass. The ventricular volume is equal to the sum of the endocardial areas multiplied by the distance between the centers of each slice, both for the end diastolic volume (A) and the end systolic volume (B). The ventricular mass is calculated as the sum of the epicardial minus the endocardial areas multiplied by the distance between the centers of each slice, as shown in A. LV, left ventricle; RV, right ventricle.

Table 28-1 SSFP Right Ventricular Volumes, Systolic Function and Mass (Absolute and Normalized to Body Surface Area) by Age Decile (Mean, 95% Confidence Interval): Males 20 29 Years

30 39 Years

40 49 Years

50 59 Years

60 69 years

70 79 Years

160 (111 210) 55 (25 85) 106 (72 140) 66 (53 79) 65 (37 94)

155 (105 205) 50 (20 80) 105 (71 139) 68 (55 81) 63 (35 92)

150 (100 200) 46 (16 76) 104 (70 138) 70 (57 83) 62 (33 90)

EDV (mL) SD 25.4 ESV (mL) SD 15.2 SV (mL) SD 17.4 EF (%) SD 6.5 Mass (g) SD 14.4

Absolute Values 177 (127 227) 171 (121 221) 68 (38 98) 64 (34 94) 108 (74 143) 108 (74 142) 61 (48 74) 63 (50 76) 70 (42 99) 69 (40 97)

166 (116 216) 59 (29 89) 107 (73 141) 65 (52 77) 67 (39 95)

EDV/BSA (mL/m2) SD 11.7 ESV/BSA (mL/m2) SD 7.4 SV/BSA (mL/m2) SD 8.2 EF/BSA (%/m2) SD 4 Mass/BSA (g/m2) SD 6.8

Normalized to Body Surface Area 91 (68 114) 88 (65 111) 35 (21 50) 33 (18 47) 56 (40 72) 55 (39 71) 32 (24 40) 32 (25 40) 36 (23 50) 35 (22 49)

(BSA) 85 (62 30 (16 55 (39 33 (25 34 (21

108) 45) 71) 41) 48)

82 28 54 34 33

(59 (13 (38 (26 (20

105) 42) 70) 42) 46)

79 25 53 35 32

(56 (11 (37 (27 (19

101) 40) 69) 42) 45)

75 23 52 35 31

(52 98) (8 37) (36 69) (27 43) (18 44)

BSA, body surface area; EDV, end-diastolic volume; EF, ejection fraction; ESV, end-systolic volume; SD, standard deviation; SV, stroke volume.

386 Cardiovascular Magnetic Resonance

20 29 Years

30 39 Years

EDV (mL) SD 21.6 ESV (mL) SD 13.3 SV (mL) SD 13.1 EF (%) SD 6 Mass (g) SD 10.6

Absolute Values 142 (100 184) 136 (94 178) 55 (29 82) 51 (25 77) 87 (61 112) 85 (59 111) 61 (49 73) 63 (51 75) 54 (33 74) 51 (31 72)

EDV/BSA (mL/m2) SD 9.4 ESV/BSA (mL/m2) SD 6.6 SV/BSA (mL/m2) SD 6.1 EF/BSA (%/m2) SD 5.2 Mass/BSA (g/m2) SD 5.2

Normalized to Body Surface Area 84 (65 102) 80 (61 98) 32 (20 45) 30 (17 43) 51 (39 63) 50 (38 62) 37 (27 47) 38 (27 48) 32 (22 42) 30 (20 40)

40 49 Years

50 59 Years

60 69 years

70 79 Years

130 (87 46 (20 84 (58 65 (53 49 (28

124 (81 42 (15 82 (56 67 (55 47 (26

117 (75 37 (11 80 (55 69 (57 45 (24

111 (69 153) 32 (6 58) 79 (53 105) 71 (59 83) 43 (22 63)

76 27 49 38 29

(57 (14 (37 (28 (19

172) 72) 109) 77) 70) 94) 40) 61) 49) 39)

72 24 48 39 27

(53 (11 (36 (29 (17

166) 68) 108) 79) 68) 90) 37) 60) 49) 37)

68 21 46 40 26

160) 63) 106) 81) 66)

(49 86) (8 34) (34 58) (30 50) (16 36)

64 19 45 41 24

(45 82) (6 32) (33 57) (31 51) (14 35)

BSA, body surface area; EDV, end-diastolic volume; EF, ejection fraction; ESV, end-systolic volume; SD, standard deviation; SV, stroke volume.

Table 28-3 SSFP Right Ventricular Diastolic Function and Atrioventricular Plane Descent (Absolute and Normalized Values) by Age Decile (Mean, 95% Confidence Interval): Males 20 29 Years

30 39 Years

40 49 Years

50 59 Years

60 69 years

70 79 Years

PFRE (mL/s) SD137 PFRA (mL/s) SD 175 PFRE/PFRA SD* 0.49 Septal AVPD (mm) SD 4.1 Lateral AVPD (mm) SD 4.4

Absolute Values 545 (277 814) 491 (223 760) 366 (23 709) 413 (70 756) 1.6 (0.6 2.5) 1.2 (0.3 2.2) 16 (8 24) 15 (7 24) 23 (14 32) 23 (14 31)

438 (169 706) 461 (118 804) 1.0 (0.0 1.9) 15 (7 23) 22 (14 31)

384 (116 652) 508 (165 852) 0.7 ( 0.2 1.7) 14 (6 22) 22 (13 30)

330 (62 599) 556 (213 899) 0.6 ( 0.4 1.5) 14 (6 22) 21 (13 30)

276 (8 545) 604 (260 947) 0.5 ( 0.5 1.4) 13 (5 21) 21 (12 29)

PFRE/BSA (mL/s/m2) SD 71 PFRE/EDV/s SD 0.75 PFRA/BSA (mL/s/m2) SD 94 PFRA/EDV/s SD 1.07 Septal AVPD/long length (%) SD 4.5 Lateral AVPD/long length (%) SD 4.1

Normalized Values 280 (142 419) 252 (114 390) 3.1 (1.6 4.6) 2.8 (1.4 4.3) 190 (6 374) 213 (29 397) 2.1 (0.0 4.2) 2.5 (0.4 4.6) 18 (9 27) 18 (9 27) 23 (15 31) 23 (15 31)

224 (85 362) 2.6 (1.1 4.1) 236 (52 420) 2.9 (0.8 4.9) 17 (9 26) 23 (15 31)

195 (57 334) 2.3 (0.9 3.8) 259 (75 443) 3.2 (1.1 5.3) 17 (8 26) 23 (15 31)

167 (29 306) 2.1 (0.6 3.6) 283 (98 467) 3.6 (1.5 5.7) 17 (8 26) 23 (15 31)

139 (1 277) 1.9 (0.4 3.3) 306 (122 490) 4.0 (1.9 6.1) 16 (8 25) 23 (15 31)

A, active; AVPD, atrioventricular plane descent; BSA, body surface area; E, early; PFR, peak filling rate; SD, standard deviation; SD*, standard deviation of log transformed data.

Table 28-4 SSFP Right Ventricular Diastolic Function and Atrioventricular Plane Descent (Absolute and Normalized Values) by Age Decile (Mean, 95% Confidence Interval): Females 20 29 Years PFRE (mL/s) SD 117 PFRA (mL/s) SD 153 PFRE/PFRA SD* 0.46 Septal AVPD (mm) SD 3.0 Lateral AVPD (mm) SD 3.5

30 39 Years

40 49 Years

50 59 Years

60 69 Years

70 79 Years

Absolute Values 471 (241 701) 419 (189 649) 368 (137 598) 316 (86 546) 264 (34 494) 213 ( 17 443) 355 (54 656) 360 (59 660) 365 (64 665) 370 (69 670) 374 (74 675) 379 (79 680) 1.6 (0.7 2.5) 1.3 (0.4 2.2) 1.0 (0.1 1.9) 0.8 ( 0.1 1.7) 0.7 ( 0.2 1.6) 0.5 ( 0.4 1.4) 16 (10 22) 15 (9 20) 13 (7 19) 12 (6 18) 11 (5 17) 10 (4 16) 22 (15 29) 21 (14 28) 21 (14 28) 20 (13 27) 20 (13 27) 19 (12 26)

Normalized Values 278 (145 411) 247 (114 380) 216 (83 349) PFRE/BSA (mL/s/m2) SD 68 3.4 (1.8 5.1) 3.1 (1.5 4.8) 2.8 (1.2 4.5) PFRE/EDV/s SD 0.85 211 (36 386) 212 (37 388) 214 (39 389) PFRA/BSA (mL/s/m2) SD 89 2.4 (0.4 4.4) 2.6 (0.6 4.6) 2.8 (0.8 4.8) PFRA/EDV/s SD 1.03 Septal AVPD/long length (%) SD 3.9 19 (11 27) 18 (11 26) 17 (10 25) Lateral AVPD/long length (%) SD 4.0 24 (16 32) 24 (16 32) 24 (16 32)

185 (52 318) 2.5 (0.9 4.2) 215 (40 390) 3.0 (1.0 5.0) 17 (9 24) 24 (16 32)

153 (20 286) 2.2 (0.6 3.9) 217 (42 392) 3.2 (1.2 5.2) 16 (8 23) 24 (16 32)

122 ( 11 255) 1.9 (0.3 3.6) 218 (43 393) 3.4 (1.4 5.4) 15 (7 22) 24 (16 31)

A, active; AVPD, atrioventricular plane descent; BSA, body surface area; E, early; PFR, peak filling rate; SD, standard deviation; SD*, standard deviation of log transformed data.

Cardiovascular Magnetic Resonance 387

28 CARDIOVASCULAR MAGNETIC RESONANCE ASSESSMENT OF RIGHT VENTRICULAR ANATOMY AND FUNCTION

Table 28-2 SSFP Right Ventricular Volumes, Systolic Function and Mass (Absolute and Normalized to Body Surface Area) by Age Decile (Mean, 95% Confidence Interval): Females

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

Table 28-5 SSFP Right Ventricular Summary Data for All Ages (Mean  Standard Deviation, 95% Confidence Interval) EDV (mL) EDV/BSA (mL/m2) ESV (mL) ESV/BSA (mL/m2) SV (mL) SV/BSA (mL/m2) EF (%) EF/BSA (%/m2) Mass (g) Mass/BSA (g/m2) PFRE (mL/s) PFRE/BSA (mL/m2) PFRE/EDV/s PFRA (mL/s) PFRA/BSA (mL/m2) PFRA/EDV/s PFRE/PFRA Septal AVPD (mm) Septal AVPD/long length (%) Lateral AVPD (mm) Lateral AVPD/long length (%)

All

Males

Females

144  23 (98 190) 78  11 (57 99) 50  14 (22 78) 27  7 (13 41) 94  15 (64 124) 51  7 (37 65) 66  6 (54 78) 36  5 (27 45) 48  13 (23 73) 31  6 (19 43) 371  125 (126 615) 202  69 (67 337) 2.6  0.8 (1.0 4.1) 429  168 (99 759) 233  93 (50 415) 3.0  1.0 (1.0 5.1) 0.9  0.47 ( 0.1 1.8) 14  3.6 (6 21) 17  4.2 (9 25) 21  3.9 (13 29) 23  4.0 (15 31)

163  25 (113 213) 83  12 (60 106) 57  15 (27 86) 29  7 (14 43) 106  17 (72 140) 54  8 (38 70) 66  6 (53 78) 34  4 (26 41) 66  14 (38 94) 34  7 (20 47) 405  137 (137 674) 207  70 (68 345) 2.4  0.75 (1.0 3.9) 489  175 (146 833) 250  94 (66 434) 3.1  1.0 (1.0 5.2) 0.8  0.49 ( 0.1 1.8) 15  4.1 (6 23) 17  4.5 (8 26) 22  4.4 (13 30) 23  4.1 (15 31)

126  21 (84 168) 73  9 (55 92) 43  13 (17 69) 25  7 (12 38) 83  13 (57 108) 48  6 (36 60) 66  6 (54 78) 39  5 (29 49) 48  11 (27 69) 28  5 (18 38) 337  117 (107 567) 197  68 (64 330) 2.7  0.85 (1.0 4.3) 368  153 (67 668) 215  89 (40 390) 2.9  1.0 (0.9 5.0) 0.9  0.46 (0.0 1.8) 13  3.0 (7 19) 17  3.9 (9 25) 21  3.5 (14 27) 24  4.0 (16 32)

A, active; AVPD, atrioventricular plane descent; BSA, body surface area; E, early; EDV, end-diastolic volume; EF, ejection fraction; ESV, end-systolic volume; PFR, peak filling rate; SV, stroke volume.

of RV parameters in borderline clinical cases, especially in arrhythmogenic RV cardiomyopathy, cardiovascular shunt ing, and adult congenital heart disease, should be referred to age , gender , and BSA normalized values to determine normality or severity of abnormality. Another CMR approach that may be particularly valuable for quantifying regional RV free wall systolic function is myocardial tagging,54–59 a technique that enables the clini cian to assess the complex mechanism of myocardial con traction and to quantify myocardial strain. Klein and colleagues analyzed the RV free wall motion and contraction in humans with CMR tagging.54 In this study percent seg mental shortening (PSS) was obtained to measure the amount of contraction and a vector analysis was used to show the trajectory of the RV free wall in systole. PSS increased through time to an average of 12% across all seg ments (inferior, mid, and superior wall) at the base, 14% at the mid ventricle, and 16% at the apex, with a wave of motion toward the septum and outflow tract. Naito and col leagues determined PSS only at the midventricle55 and found a PSS of 6.7% in the superior wall segment and 20% for the midwall segment. Fayad and colleagues reported similar PSS values: 24.7% in the midwall segment of the midventricular slice and 28.7% in the midwall seg ment of the apical slice.56 A 3D reconstruction of RV con traction with CMR tagging58 shows a primary contraction of the RV tangential to its own surface plane and a circum ferential contraction as the RV moves apically, with a twis ting motion similar to that described by Klein and colleagues.54 Still, while CMR tagging seems a promising method for the assessment of regional function, especially with stress techniques in ischemic heart disease, it remains to be seen whether it provides clinically relevant information beyond that provided by standard cine CMR. Further studies are needed to better define the clinical role of CMR tagging. 388 Cardiovascular Magnetic Resonance

CARDIOVASCULAR MAGNETIC RESONANCE ASSESSMENT OF RIGHT VENTRICULAR ANATOMY AND FUNCTION IN DISEASE Right Ventricular Assessment in Heart Failure In left heart failure, atrial pressure rises, forcing open a num ber of pulmonary capillaries. When all reserve capillaries are open, the increase in pulmonary pressure leads to an increased load on the RV. Therefore, the function of the RV during exercise in heart failure is important.60 The prognostic value of RV function in advanced heart failure of various causes has been reported;61,62 therefore, the estimation of RV function is now warranted in the standard evaluation of patients with heart failure, since it is helpful in the clinical assessment and prognostic stratification of such patients. Di Salvo and colleagues studied 67 patients with heart failure who had been referred for cardiac transplantation with ischemic (46%) or dilated (54%) cardiomyopathy.63 An RVEF of 35% or more at rest and with exercise pre dicted overall survival. Maximal oxygen consumption was also predictive of survival, with a modest correlation between RVEF and maximal oxygen consumption. Also, in patients with moderate heart failure, de Groote and col leagues62 showed that three variables, NYHA classification, percent of maximal predicted VO2, and RVEF, were inde pendent predictors of both survival and event free cardiac survival. Left ventricular EF and peak VO2 normalized to

RV end-systolic volume/BSA−−Females (mL/m2)

RV end-diastolic volume/ BSA−−Females (mL/m2)

+95% CI 100 Mean 80 −95% CI 60

40 20

30

40

50

60

70

60 50

+95% CI

40 Mean 30 −95% CI

20 10 0 20

80

30

40

60

70

80

60

70

80

60

70

80

60

80

RV mass/BSA−−Females (mL/m2)

RV ejection fraction−−Females (%)

90

+95% CI

70 Mean 60 −95% CI

50 40

50 +95% CI 40 Mean 30 −95% CI

20 10 0

20

30

40

50

60

70

80

20

30

40

Age (years)

50 Age (years)

500

600 RV active peak filling rate/BSA−−Females (mL/m2)

RV early peak filling rate/BSA−−Females (mL/m2)

50 Age (years)

Age (years)

+95% CI 400 Mean

300 200

−95% CI

100 0

500 +95% CI

400 300

Mean 200 100

−95% CI

0 20

30

40

50

60

70

80

Age (years)

20

30

40

50 Age (years)

RV early/active peak filling rate ratio−−Females

3 +95% CI 2 Mean

1

−95% CI

0 20

30

40

50

60

70

80

Age (years) Figure 28 5 Normal values for right ventricular (RV) end diastolic volume, end systolic volume, mass, and parameters of diastolic function for females normalized to body surface area (BSA). Cardiovascular Magnetic Resonance 389

28 CARDIOVASCULAR MAGNETIC RESONANCE ASSESSMENT OF RIGHT VENTRICULAR ANATOMY AND FUNCTION

FEMALES 120

60 RV end-systolic volume/BSA−−Males (mL/m2)

RV end-diastolic volume/ BSA−−Males (mL/m2)

120 +95% CI 100

Mean

80 −95% CI 60

40

+95% CI

50 40

Mean

30 −95% CI

20 10 0

20

30

40

50

60

70

80

20

30

40

90

60

80

50

+95% CI

70 Mean 60 −95% CI

50 40

60

70

80

60

70

80

60

70

80

+95% CI

40 Mean

30 20

−95% CI

10 0

20

30

40

50

60

70

20

80

30

40

50 Age (years)

Age (years) 500

600 +95% CI

RV active peak filling rate/BSA−−Males (mL/m2)

RV early peak filling rate/BSA−−Males (mL/m2)

50 Age (years)

RV mass/BSA−−Males (g/m2)

RV ejection fraction−−Males (%)

Age (years)

400 Mean

300 200

−95% CI

100 0

500 +95% CI

400 300

Mean 200 100

−95% CI

0 20

30

40

50

60

70

80

Age (years)

20

30

40

50 Age (years)

3 RV early/active peak filling rate ratio−−Males

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

MALES

+95% CI 2 Mean

1 −95% CI 0 20

30

40

50

60

70

80

Age (years) Figure 28 6 Normal values for right ventricular (RV) end diastolic volume, end systolic volume, mass, and parameters of diastolic function for males normalized to body surface area (BSA). 390 Cardiovascular Magnetic Resonance

Right Ventricular Assessment in Ischemic Heart Disease Isolated RV infarction is relatively rare, but concurrent RV infarction in the setting of an inferior infarction due to proximal right coronary occlusion64,65 occurs in up to half of LV infarctions.66 RV infarction can be detected and eval uated in extent by using late enhancement gadolinium CMR. RV necrosis causes a loss of contractile mass, and if the inferior interventricular septum is involved, there is also a loss of septal augmentation of RV function. The exis tence of RV dysfunction in patients with inferior myocar dial infarction is associated with high rates of morbidity and mortality.65–67 Cardiogenic shock is also more frequent if the RV is involved in inferior infarctions.66 Reperfusion of acute RV infarcts by primary angioplasty has been shown to greatly improve RV function.68 These studies also high light the importance of CMR in characterizing the RV in the setting of acute infarction, not only for RV mass and function quantification but also for detection and quantifi cation of necrosis with late gadolinium techniques.

Right Ventricular Assessment in Arrhythmogenic Right Ventricular Cardiomyopathy In ARVC, normal myocardium is replaced by fibrofatty tis sue, and electrical instability develops. This disorder usu ally involves the right ventricle, but the left ventricle and septum may also be affected. After hypertrophic heart dis ease, ARVC is the number one cause of sudden cardiac death in young people, especially athletes. Evident forms of the disease are straightforward to diagnose on the basis of a series of diagnostic criteria proposed by the Interna tional Task Force for Cardiomyopathy69; however, the diag nosis of early and mild forms of the disease often is difficult. CMR is regarded as the best imaging technique for detecting RV structural and functional abnormalities.70 The main advantage of CMR is the possibility of planning any desired view so that regions such as the outflow tract, which are hard to visualize with other techniques, can be assessed with great precision. The most common findings in ARVC are RV wall motion abnormalities and dilation.71 CMR can also detect fatty infiltration; however, this alone does not allow a definitive diagnosis of ARVC, as fatty infiltration occurs in a high pro portion of healthy people, particularly elderly subjects.

Several authors have reported variable diagnostic sensitivity of RV fatty infiltration in patients with ARVC. The presence of fatty infiltration is often associated with RV structural abnormalities or RV wall motion abnormalities. It has been suggested that the findings of studies of fatty infiltration vary greatly from observer to observer72; therefore, diagno sis should be done carefully by experts to avoid erroneous diagnosis of ARVD, particularly as fatty infiltration is a major criterion. Fat suppressed sequences may be useful for confirming the diagnosis in doubtful cases. Diagnosis of ARVC should always be made according to the criteria proposed by the International Task Force and never according to findings from a single test such as CMR.70

Congenital Heart Disease A comprehensive summary of the role of CMR in congen ital heart disease is detailed in Chapters 29 and 30. Assessment of RV size, location, and connections as well as of function and pulmonary flow are important.73 In congenital heart disease, the right ventricle may support the pulmonary (subpulmonary right ventricle) or the sys temic circulation (systemic right ventricle). In many of these patients, RV dysfunction develops and leads to con siderable morbidity and mortality. Therefore, RV function in certain conditions needs close surveillance and timely and appropriate intervention to optimize outcomes. Many of these patients have come into the adult age, and this has created a patient population in which the right ventri cle is often the center of attention. Despite major progress being made, assessing the RV in either the subpulmonary or the systemic circulation remains challenging, often requiring a multi imaging approach. CMR is of use not only in the assessment of the anatomy and physiology of CHD but also, in some cases, in the risk stratification.39,74

Pulmonary Hypertension and Lung Transplantation The fact that CMR is a radiation free, highly accurate and reproducible technique for quantitative assessment of RV mass and volume makes it the most appropriate imaging modality for serial studies on the same patient. Cine gradi ent echo CMR has been used to study pulmonary hyperten sion,75,76 its response to therapy77 and the progression of RV failure in this condition.75 CMR has been used to con firm the diagnosis of cor pulmonale through increased RV mass measurements above 60 g.78 Pulmonary flow patterns are known to be abnormal in pulmonary hypertension,79 and this may affect RV afterload, while diastolic function has also been found to be abnormal in pulmonary fibrosis by using tricuspid flow patterns.80 CMR has been used to determine the time course of changes in ventricular mass and function after lung transplantation.81–83 Frist and col leagues observed that RVEF normalized in the early post lung transplantation period.81 Other early changes included a decrease in RV end diastolic volume to below normal levels, with persistence at this level even in the late

Cardiovascular Magnetic Resonance 391

28 CARDIOVASCULAR MAGNETIC RESONANCE ASSESSMENT OF RIGHT VENTRICULAR ANATOMY AND FUNCTION

body weight had no predictive value. The event free sur vival rates from cardiovascular mortality and urgent trans plantation at 1 year were 80%, 90%, and 95% in patients with an RVEF less than 25%, with an RVEF of 25% or more and less than 35% and with an RVEF of 35% or more, respectively. At 2 years, survival rates were 59%, 77%, and 93%, respectively, in the same subgroups. To date, we are unaware of any published data on the prognostic value of CMR derived indices of RV function, but these data suggest that CMR quantitative RV volumetric assess ment should be measured on studies that are performed on patients with heart failure.

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

studies. RV mass also regressed early but remained increased in comparison to healthy control subjects. The authors concluded that RV anatomic normalization occurred later than functional normalization, and RV mass remained increased. Fayad and colleagues also studied RV tagging in patients with chronic pulmonary hypertension.84 Regional short axis shortening was reduced in patients in comparison to healthy controls, and the greatest reductions in shortening were found in the outflow tract and basal septal region.

CONCLUSION The role of the RV in acquired and congenital heart disease is being increasingly recognized. CMR is a highly accurate, reproducible, and versatile technique that is considered the ideal imaging modality for the comprehensive evaluation of RV dimensions and global and regional function. However, further clinical studies are needed to establish standards for the best use of CMR for predicting patient outcome and the role of serial evaluation in the case of known RV dysfunction.

References 1. Guyton AC. The pulmonary circulation. In: Textbook of Medical Physiology. 7th ed. Philadelphia: WB Saunders; 1986:287–294. 2. Armour JA, Randall WC. Structural basis for cardiac function. Am J Physiol. 1970;218:1517–1523. 3. Lorenz CH, Walker ES, Morgan VL, Klein SS, Graham TP. Normal human right and left ventricular mass, systolic function and gender differences by cine magnetic resonance imaging. J Cardiovasc Magn Reson. 1999;1:7–22. 4. Lewis WH. Gray’s anatomy of the human body. 20th ed. Philadelphia. 2000. 5. Piazza G, Goldhaber SZ. The acutely decompensated right ventricle. Chest. 2005;128:1836–1852. 6. Ghio S, Gavazzi A, Campana C, et al. Independent and additive prognostic value of right ventricular systolic function and pulmonary artery pressure in patients with chronic heart failure. J Am Coll Cardiol. 2001;37:183–188. 7. Meluzin J, Spinarova L, Hude P, et al. Prognostic importance of various echocardiographic right ventricular functional parameters in patients with symptomatic heart failure. J Am Soc Echocardiogr. 2005;18:435–444. 8. Gavazzi A, Ghio S, Scelsi L, et al. Response of the right ventricle to acute pulmonary vasodilation predicts the outcome in patients with advanced heart failure and pulmonary hypertension. Am Heart J. 2003;145:310–316. 9. Zornoff LA, Skali H, Pfeffer MA, et al. SAVE Investigators. Right ventricular dysfunction and risk of heart failure and mortality after myocardial infarction. J Am Coll Cardiol. 2002;39:1450–1455. 10. Rumberger JA, Behrenbeck T, Bell MR, et al. Determination of ventricular ejection fraction: a comparison of available imaging methods: the cardiovascular imaging working group. Mayo Clin Proc. 1997;72:860–870. 11. Baker BJ, Scovil JA, Kane JJ, Murphy ML. Echocardiographic detection of right ventricular hypertrophy. Am Heart J. 1983;505:611–614. 12. Kaul S, Tei C, Hopkins JM, Shah PM. Assessment of right ventricular function using two-dimensional echocardiography. Am Heart J. 1984;107:526–531. 13. Borges AC, Knebel F, Eddicks S, et al. Right ventricular function assessed by two-dimensional strain and tissue Doppler echocardiography in patients with pulmonary arterial hypertension and effect of vasodilator therapy. Am J Cardiol. 2006;98:530–534. 14. Gopal AS, Keller AM, Shen Z, et al. Three-dimensional echocardiography: in vitro and in vivo validation of left ventricular mass and comparison with conventional echocardiographic methods. J Am Coll Cardiol. 1994;24:504–513. 15. Fujimoto S, Mizuno R, Nagakawa Y, Dohi K, Nakano H. Estimation of the right ventricular volume and ejection fraction by transthoracic three-dimensional echocardiography: a validation study using magnetic resonance imaging. Int J Cardiac Imaging. 1998;14:385–390. 16. Vogel M, Gutberlet M, Dittrich S, Hosten N, Lange PE. Comparison of transthoracic three dimensional echocardiography with magnetic resonance imaging in the assessment of right ventricular volume and mass. Heart. 1997;78:127–130. 17. Hornung TS, Anagnostopoulos C, Bhardwaj P, et al. Comparison of equilibrium radionuclide ventriculography with cardiovascular magnetic resonance for assessing the systemic right ventricle after Mustard or Senning procedures for complete transposition of the great arteries. Am J Cardiol. 2003;92:640–643.

392 Cardiovascular Magnetic Resonance

18. Ohsuzu F, Handa S, Kondo M, et al. Thallium-201 myocardial imaging to evaluate right ventricular overloading. Circulation. 1980;61: 620–625. 19. Pietras RJ, Kondos GT, Kaplan D, Lam W. Comparative angiographic right and left ventricular volumes. Am Heart J. 1985;109:321–326. 20. Dehmer GJ, Firth DG, Hillis LD, Nicod P, Willerson JT, Lewis SE. Non geometric determination of right ventricular volumes from equilibrium blood pool scans. Am J Cardiol. 1982;49:79–84. 21. Schmermund A, Rensing BJ, Sheedy PF, Rumberger JA. Reproducibility of right and left ventricular volume measurements by electronbeam CT in patients with congestive heart failure. Int J Card Imaging. 1998;14:201–209. 22. Pignatelli RH, McMahon CJ, Chung T, Vick 3rd GW. Role of echocardiography versus MRI for the diagnosis of congenital heart disease. Curr Opin Cardiol. 2003;18:357–365. 23. Grothues F, Moon JC, Bellenger NG, Smith GS, Klein HU, Pennell DJ. Interstudy reproducibility of right ventricular volumes, function, and mass with cardiovascular magnetic resonance. Am Heart J. 2004;147:218–223. 24. Katz J, Whang J, Boxt LM, Barst RJ. Estimation of right ventricular mass in normal subjects and in patients with pulmonary hypertension by nuclear magnetic resonance imaging. J Am Coll Cardiol. 1993;21:1475–1478. 25. Lorenz CH, Walker ES, Graham TP, Powers TA. Right ventricular performance and mass in adults late after atrial repair of transposition of the great arteries using cine magnetic resonance imaging. Circulation. 1995;92(suppl II):233–239. 26. Sen-Chowdhry S, Lowe MD, Sporton SC, McKenna WJ. Arrhythmogenic right ventricular cardiomyopathy: clinical presentation, diagnosis, and management. Am J Med. 2004;117:685–695. 27. Maceira AM, Prasad SK, Khan M, Pennell DJ. Reference right ventricular systolic and diastolic function normalized to age, gender, and body surface area from steady-state free precession cardiovascular magnetic resonance. Eur Heart J. 2006;27:2879–2888. 28. Lee VS, Resnick D, Bundy JM, Simonetti OP, Lee P, Weinreb JC. Cardiac function: MR evaluation in one breath hold with real-time true fast imaging with steady-state precession. Radiology. 2002;222: 835–842. 29. Bellenger NG, Gatehouse PD, Rajappan K, Keegan J, Firmin DN, Pennell DJ. Left ventricular quantification in heart failure by CMR using prospective respiratory navigator gating: comparison with breath-hold acquisition. J Magn Reson Imaging. 2000;11:411–417. 30. Helbing WA, Rebergen SA, Maliepaard C, et al. Quantification of right ventricular function with magnetic resonance imaging in children with normal hearts and with congenital heart disease. Am Heart J. 1995;130:828–837. 31. Jauhiainen T, Jarvinen VM, Hekali PE, Poutanen VP, Penttila A, Kupari M. MR Gradient echo volumetric analysis of human cardiac casts: focus on the right ventricle. J Comput Assist Tomogr. 1998;22:899–903. 32. Alfakih K, Plein S, Bloomer T, Jones T, Ridgway J, Sivananthan M. Comparison of right ventricular volume measurements between axial and short axis orientation using steady-state free precession magnetic resonance imaging. J Magn Reson Imaging. 2003;18:25–32. 33. Grothues F, Moon JC, Bellenger NG, Smith GS, Klein HU, Pennell DJ. Interstudy reproducibility of right ventricular volumes, function and mass with cardiovascular magnetic resonance. Am Heart J. 2004;147:218–223.

56. Fayad ZA, Kraitchman DL, Ferrari VA, Axel L. Right ventricular regional function in normal subjects using magnetic resonance tissue tagging. In: Book of Abstracts: Society of Magnetic Resonance. Berkeley, CA:1994:1504. 57. Young AA, Fayad ZA, Axel L. Right ventricular midwall surface motion and deformation using magnetic resonance tagging. Am J Physiol. 1996;271:H2677–H2688. 58. Young AA, Cowan BR, Occleshaw CJ, Oxenham HC, Gentles TL. Temporal evolution of left ventricular strain late after repair of coarctation of the aorta using 3D MR tissue tagging. J Cardiovasc Magn Reson. 2002;4:233–243. 59. Menteer J, Weinberg PM, Fogel MA. Quantifying regional right ventricular function in tetralogy of Fallot. J Cardiovasc Magn Reson. 2005;7:753–761. 60. Brieke A, DeNofrio D. Right ventricular dysfunction in chronic dilated cardiomyopathy and heart failure. Coronary Artery Disease. 2005;16: 5–11. 61. Meluzin J, Spinarova L, Hude P, et al. Prognostic importance of various echocardiographic right ventricular functional parameters in patients with symptomatic heart failure. J Am Soc Echocardiogr. 2005;18:435–444. 62. de Groote P, Millaire A, Foucher-Hossein C, et al. Right ventricular ejection fraction is an independent predictor of survival in patients with moderate heart failure. J Am Coll Cardiol. 1998;32:948–954. 63. Di Salvo TG, Mathier M, Semigran MJ, Dec GW. Preserved right ventricular ejection fraction predicts exercise capacity and survival in advanced heart failure. J Am Coll Cardiol. 1995;25:1143–1153. 64. Zehender M, Kasper W, Kauder E, et al. Right ventricular infarction as an independent predictor of prognosis after acute inferior myocardial infarction. N Engl J Med. 1993;328:981–988. 65. Bueno H, Lopez-Palop R, Bermejo J, Lopez-Sendon JL, Delcan JL. Inhospital outcome of elderly patients with acute inferior myocardial infarction and right ventricular involvement. Circulation. 1997;96:436–441. 66. Kinch JW, Ryan TJ. Right ventricular infarction. N Engl J Med. 1994;330:1211–1217. 67. Moazami N, Hill L. Right ventricular dysfunction in patients with acute inferior MI: role of RV mechanical support. Thorac Cardiovasc Surg. 2003;51:290–292. 68. Bowers TR, O’Neill WW, Grines C, Pica MC, Safian RD, Goldstein JA. Effect of reperfusion on biventricular function and survival after right ventricular infarction. N Engl J Med. 1998;338:933–940. 69. Diagnosis of arrhythmogenic right ventricular dysplasia/cardiomyopathy. Task Force of the Working Group Myocardial and Pericardial Disease of the European Society of Cardiology and of the Scientific Council on Cardiomyopathies of the International Society and Federation of Cardiology. Br Heart J. 1994;71(3):215–218. 70. Pennell DJ, Sechtem UP, Higgins CB, et al. European Society of Cardiology; Society for Cardiovascular Magnetic Resonance. Clinical indications for cardiovascular magnetic resonance (CMR): Consensus Panel report. J Cardiovasc Magn Reson. 2004;6:727–765. 71. Tandri H, Calkins H, Nasir K, et al. Magnetic resonance imaging findings in patients meeting task force criteria for arrhythmogenic right ventricular dysplasia. J Cardiovasc Electrophysiol. 2003;14:476–482. 72. Bluemke DA, Krupinski EA, Ovitt T, et al. MR imaging of arrhythmogenic right ventricular cardiomyopathy: morphologic findings and interobserver reliability. Cardiology. 2003;99:153–162. 73. Rebergen SA, Ottenkamp J, Doornbos J, van der Wall EE, Chin JG, de Roos A. Postoperative pulmonary flow dynamics after Fontan surgery: assessment with nuclear magnetic resonance velocity mapping. J Am Coll Cardiol. 1993;21:123–131. 74. Babu-Narayan SV, Kilner PJ, Li W, et al. Ventricular fibrosis suggested by cardiovascular magnetic resonance in adults with repaired tetralogy of fallot and its relationship to adverse markers of clinical outcome. Circulation. 2006;113:405–413. 75. Boxt LM, Katz J, Kolb T, Czegledy FP, Barst RJ. Direct quantitation of right and left ventricular volumes with nuclear magnetic resonance imaging in patients with primary pulmonary hypertension. J Am Coll Cardiol. 1992;19:1508–1515. 76. Saito H, Dambara T, Aiba M, Suzuki T, Kira S. Evaluation of cor pulmonale on a modified short-axis section of the heart by magnetic resonance imaging. Am Rev Respir Dis. 1992;146:1576–1581. 77. Wilkins MR, Paul GA, Strange JW, et al. Sildenafil versus Endothelin Receptor Antagonist for Pulmonary Hypertension (SERAPH) study. Am J Respir Crit Care Med. 2005;171:1292–1297. 78. Pattynama PMT, Willems LNA, Smit AH, van der Wall EE, de Roos A. Early diagnosis of cor pulmonale with MR imaging of the right ventricle. Radiology. 1992;182:375–379.

Cardiovascular Magnetic Resonance 393

28 CARDIOVASCULAR MAGNETIC RESONANCE ASSESSMENT OF RIGHT VENTRICULAR ANATOMY AND FUNCTION

34. Longmore DB, Klipstein RH, Underwood SR, et al. Dimensional accuracy of magnetic resonance in studies of the heart. Lancet. 1985;1:1360–1362. 35. Cutrone JA, Georgiou D, Khan S, et al. Comparison of electron beam computed tomography scanning and magnetic resonance quantification of right ventricular mass: validation with autopsy weights. Acad Radiol. 1996;3:395–400. 36. Sinitsyn VE, Mareeva GI, Galakhov IE, Veselova SP. The determination of ventricular myocardial mass by magnetic resonance tomography. Med Radiol. 1990;35:9–11. 37. Maceira AM, Prasad SK, Khan M, Pennell DJ. Reference right ventricular systolic and diastolic function normalized to age, gender and body surface area from steady state free precession cardiovascular magnetic resonance. Eur Heart J. (Accepted). 38. Prasad SK, Soukias N, Hornung T, et al. Role of magnetic resonance angiography in the diagnosis of major aortopulmonary collateral arteries and partial anomalous pulmonary venous drainage. Circulation. 2004;109:207–214. 39. Babu-Narayan SV, Goktekin O, Moon JC, et al. Late gadolinium enhancement cardiovascular magnetic resonance of the systemic right ventricle in adults with previous atrial redirection surgery for transposition of the great arteries. Circulation. 2005;111:2091–2098. 40. Sen-Chowdhry S, Prasad SK, McKenna WJ. Arrhythmogenic right ventricular cardiomyopathy with fibrofatty atrophy, myocardial oedema, and aneurysmal dilation. Heart. 2005;91:784. 41. Fulton RM, Hutchinson EC, Morgan-Jones A. Ventricular weight in cardiac hypertrophy. Br Heart J. 1952;4:413–420. 42. Hangartner JRW, Marley NJ, Whitehead A, Thomas AC, Davies MJ. The assessment of cardiac hypertrophy at autopsy. Histopathology. 1985;9:1295–1306. 43. Daniels SR, Meyer RA, Liang Y, Bove KE. Echocardiographically determined left ventricular mass index in normal children, adolescents and young adults. J Am Coll Cardiol. 1988;12:703–708. 44. Am K, Gopal AS, King DL. Left and right atrial volume by freehand three-dimensional echocardiography: in vivo validation using magnetic resonance imaging. Eur J Echocardiogr. 2000;1:55–65. 45. Schvartzman PR, Fuchs FD, Mello AG, Coli M, Schvartzman M, Moreira LB. Normal values of echocardiographic measurements: a population-based study. Arq Bras Cardiol. 2000;75:107–114. 46. Kjaer A, Lebech AM, Hesse B, Petersen CL. Right-sided cardiac function in healthy volunteers measured by first-pass radionuclide ventriculography and gated blood-pool SPECT: comparison with cine MRI. Clin Physiol Funct Imaging. 2005;25:344–349. 47. Lorenz CH, Walker ES, Morgan VL, Klein SS, Graham Jr TP. Normal human right and left ventricular mass, systolic function, and gender differences by cine magnetic resonance imaging. J Cardiovasc Magn Reson. 1999;1:7–21. 48. Rominger MB, Bachmann GF, Pabst W, Rau WS. Right ventricular volumes and ejection fraction with fast cine MR imaging in breath-hold technique: applicability, normal values from 52 volunteers and evaluation of 352 adult cardiac patients. J Magn Reson Imaging. 1999;10:908–918. 49. Alfakih K, Plein S, Thiele H, Jones T, Ridgway JP, Sivananthan MU. Normal human left and right ventricular dimensions for MRI as assessed by turbo gradient echo and steady-state free precession imaging sequences. J Magn Reson Imaging. 2003;17:323–329. 50. Beygui F, Furber A, Delepine S, et al. Routine breath-hold gradient echo MRI-derived right ventricular mass, volumes and function: accuracy, reproducibility and coherence study. Int J Cardiovasc Imaging. 2004;20:509–516. 51. Hudsmith LE, Petersen SE, Francis JM, Robson MD, Neubauer S. Normal human left and right ventricular and left atrial dimensions using steady state free precession magnetic resonance imaging. J Cardiovasc Magn Reson. 2005;7:775–782. 52. Hajduczok ZD, Weiss RM, Stanford W, Marcus ML. Determination of right ventricular mass in humans and dogs with ultrafast cardiac computed tomography. Circulation. 1990;82:202–212. 53. Wachspress JD, Clark NR, Untereker WJ, Kraushaar BT, Kurnik PB. Systolic and diastolic performance in normal human subjects as measured by ultrafast computed tomography. Cathet Cardiovasc Diagn. 1988;15:277–283. 54. Klein SS, Graham TP, Lorenz CH. Noninvasive delineation of normal right ventricular contractile motion with MRI myocardial tagging. Ann Biomed Eng. 1998;26:756–763. 55. Naito H, Arisawa J, Yamagami H, Kozuka T, Tamura S. Assessment of right ventricular regional contraction and comparison with left ventricle in normal humans: a cine magnetic resonance study with presaturation myocardial tagging. Br Heart J. 1995;74:186–191.

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

79. Bogren HG, Klipstein RH, Mohiaddin RH, et al. Pulmonary artery distensibility and blood flow patterns: a magnetic resonance study of normal subjects and of patients with pulmonary arterial hypertension. Am Heart J. 1989;118:990–999. 80. Kroft LJ, Simons P, van Laar JM, de Roos A. Patients with pulmonary fibrosis: cardiac function assessed with MR imaging. Radiology. 2000;216:464–471. 81. Frist WH, Lorenz CH, Walker ES, et al. MRI complements standard assessment of right ventricular function after lung transplantation. Ann Thorac Surg. 1995;60:268–271.

394 Cardiovascular Magnetic Resonance

82. Moulton JM, Creswell LL, Ungacta FF, Downing SW, Szabo BA, Pasque MK. Magnetic resonance imaging provides evidence for remodeling of the right ventricle after single-lung transplantation for pulmonary hypertension. Circulation. 1996;94(suppl II):312–319. 83. Lorenz CH, Loyd JE, Klein SS, et al. Characterization of different time courses of left and right ventricular recovery after lung transplantation. J Am Coll Cardiol. 1997;29(Suppl A):23A. 84. Fayad ZA, Ferrari VA, Kraitchman DL, et al. Right ventricular regional function using MR tagging: normals versus chronic pulmonary hypertension. Magn Reson Med. 1998;39:116–123.

Cardiovascular Magnetic Resonance of Simple Congenital Cardiovascular Defects Arno A. W. Roest, Lucia J. M. Kroft, and Albert de Roos

The evaluation of congenital heart and large vessel disease is one of the well established clinical applications of cardio vascular magnetic resonance (CMR). Accurate determina tion of cardiac anatomy and function is crucial for patient management at initial diagnosis, during intervention, and at follow up after repair of cardiovascular malformations. The incidence of moderate to severe forms of congenital heart disease (CHD) is estimated to be 5 to 12 per 1000 live births; worldwide, 1.5 million children are born each year with a congenital cardiac malformation.1 The preva lence increases to 19 per 1000 live births if bicuspid aortic valves are included and to 75 per 1000 live births if tiny ventricular septal defects (VSDs) and other forms of trivial lesions are taken into account.2 The estimated distribution of various types of congenital cardiac malformations is listed in Table 29 1. The initial diagnosis of CHD is most often made with transthoracic echocardiography. Before correction, CMR is especially useful to evaluate complex vascular malformations, assess possible airway compres sion, provide tissue characterization of cardiac tumors,3 and assess the spatial relationships in complex CHD.4 Fur thermore, left to right shunting of blood frequently occurs in congenital heart defects, such as VSD, patent ductus arteriosus (PDA), atrial septal defect (ASD), aortopulmon ary window, and partial or total anomalous pulmonary venous return. Frequently used methods to evaluate shunt volume, such as invasive oximetry, first pass radionuclide angiography, and echocardiography, have limitations,5 whereas CMR is safe, does not expose the patient to radia tion or iodinated contrast, and is accurate and reproducible for quantifying left to right shunting.6 Most types of CHD require surgical or catheter based inter vention.7 Since the introduction of various types of interven tions for CHD, the survival rate of these patients has increased dramatically. In developed countries, more than 85% of infants with CHD now reach adulthood.8 In the planning of interventional procedures, such as closure of an ASD, CMR is a noninvasive and accurate method for anatomic delinea tion9,10 and can be used in the stratification of patients for either interventional or surgical correction.9–12 The long term outcome of patients with corrected or pal liated CHD is determined by residua (preoperative abnor mality intentionally unaffected by intervention), sequelae (unintended but foreseen result of intervention), and com plications after surgical intervention.7 Timely detection of

these morphologic and functional abnormalities requires accurate and preferably noninvasive imaging methods. CMR is ideally suited to assess morphologic and functional abnormalities after correction or palliation of CHD because this technique is not hampered by anatomic limitations and does not use ionizing radiation. CMR can provide detailed anatomic information and quantitative data on vas cular stenosis and valvular function and can accurately assess the dimensions and function of the left ventricle (LV) and right ventricle (RV). Especially in corrected CHD, echocardiography may be hampered by the presence of scar tissue, rib and chest deformations, and interposed lung tis sue. Cardiac catheterization using X ray is not suited for routine follow up after correction because of its invasiveness and limitations or repeated radiation exposure. Especially in patients with CHD, multiple cardiac catheterizations may lead to high radiation exposure, a risk factor for the develop ment of cancer.13 The further development and application of interventional CMR guided cardiac catheterization will further decrease the use of radiation in patients with a congenital cardiac defect.14,15

CARDIOVASCULAR MAGNETIC RESONANCE IN PEDIATRIC PATIENTS The duration of a “typical” CMR study is 45 to 60 minutes during which the patient is located within the bore of the scanner. Claustrophobia is less common in children than in adults,16 and most children older than 10 years can hold still for 60 minutes and can cooperate with breath hold maneuvers. For children 5 to 10 years of age, lying still is accomplished by proper instructions and by allowing a parent to remain with them in the scanner room. Breath holding, however, is difficult at this age, and free breathing techniques, in combination with navigator gated respira tory techniques, can be used. In patients younger than 5 years, some form of anesthe sia is needed to perform a CMR study. The application of sedation has been proposed as a safe form of “conscious” anesthesia,17–19 although some advocate general anesthesia to protect the airway and control respiration.20 Specialized personnel who are familiar with providing anesthesia to Cardiovascular Magnetic Resonance 395

29 CARDIOVASCULAR MAGNETIC RESONANCE OF SIMPLE CONGENITAL CARDIOVASCULAR DEFECTS

CHAPTER 29

Ventricular septal defect* Secundum atrial septal defect* Patent ductus arteriosus* Pulmonary valve stenosis* Coarctation of the aorta* Tetralogy of Fallot Aortic valve stenosis* Transposition of the great arteries Atrioventricular septal defect Hypoplastic left heart Hypoplastic right heart

31% 7.5% 7.1% 7.0% 5.6% 5.5% 4.9% 4.5% 4.4% 3.1% 2.4%

(CE MRA) is particularly of use in the evaluation of the thoracic vasculature at initial diagnosis as well as after cor rection of a congenital vascular malformation, such as coarctation of the aorta.25 CMR guided cardiac catheteriza tion with intervention is a new, exciting area and is espe cially applicable in the catheter based closure of ASDs,26– 28 in the nonsurgical replacement of semilunar valves,29 and in guiding intervention for native or recurrent coarcta tion of the aorta.30–32 This chapter reviews the value of CMR for the evaluation of the most common simple congenital cardiovascular defects.

*Cardiovascular magnetic resonance evaluation discussed in this chapter.

Ventricular Septal Defect patients with CHD and CMR compatible respiratory equip ment and monitoring systems are needed to perform CMR studies with general anesthesia. An advantage of using gen eral anesthesia is that the CMR examination is not limited to free breathing techniques because ventilation is con trolled by the anesthesiologist and suspended respiration studies can be performed.3 Few reports exist on normal values for cardiac dimen sions and function obtained with CMR in healthy children. Because the evaluation of CHD and large vessel disease is one of the well established clinical applications of CMR, more studies must be performed to establish reliable infor mation on normal cardiac dimensions and function in children.21

CARDIOVASCULAR MAGNETIC RESONANCE TECHNIQUES IN CONGENITAL HEART DEFECTS Spin echo CMR is a black blood technique and is used to assess the cardiac and vascular anatomy under investiga tion, whereas gradient recalled echo (GRE) and steady state free precession (SSFP) CMR are bright blood techniques often used for assessment of LV and RV function or to study flow phenomena across stenoses and to depict vascular dis ease. Flow mapping is useful to quantify flow in large ves sels as well as across valves and is useful to quantify valvular regurgitation. It can also be used to quantify flow velocity for calculation of pressure gradients in case of a stenosed vascular segment or valve. Flow measurements can be readily performed in vascular areas that may not routinely be accessible by Doppler echocardiography. CMR can directly measure flow in the aorta and pulmonary circulation, thereby allowing quantification of shunt lesions that manifest themselves by a discrepancy between aortic and pulmonary flow. For example, atrial left to right shunts can be assessed by quantifying the stroke volume in the aorta and pulmonary artery, thereby allowing direct shunt quantification with high precision and accuracy.22–24 The use of contrast enhanced magnetic resonance angiography 396 Cardiovascular Magnetic Resonance

The most common CHD malformation and the major cause of left to right shunts, a VSD can occur as an isolated anomaly or in combination with other cardiac malforma tions, such as coarctation of the aorta, tetralogy of Fallot, double outlet RV, or truncus arteriosus. VSDs are classified according to the part of the RV surface of the interventric ular septum in which they are located.33 The four parts of the ventricular septum are the inlet septum; the muscu lar, or trabecular, septum; the outlet septum; and the mem branous septum (Fig. 29 1). A defect can occur in each part of the ventricular septum. Most VSDs are small and result in a small left to right shunt without any signs and symptoms other than a sys tolic murmur. VSDs often become smaller or close sponta neously. At least 70% of defects that are present at birth will spontaneously close, usually within the first years of life.34 With a larger VSD, high pulmonary vascular resistance in the first weeks of life will prevent significant left to right shunting of blood. As pulmonary vascular resistance declines, the left to right shunt will increase and tachyp nea, dyspnea, and feeding difficulties will develop. Such infants often present 2 to 6 weeks after birth. Indications for interventional or surgical closure and the age at which

PA 4 Outlet

*3 Trabecular TV

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

Table 29-1 Distribution of Types of Congenital Heart Disease in Liveborn Children

1 2 Inlet

Figure 29 1 Most common locations of ventricular septal defect shown over the four parts of the interventricular septum, viewed from the right ventricular septal surface.* Membranous part of the septum. 1, inlet defect; 2, trabecular or muscular defect; 3, perimembranous defect; 4, outlet defect, also called supracristal or subaortic defect; PA, pulmonary artery; TV, tricuspid valve.

Figure 29 2 Transverse spin echo cardiovascular magnetic resonance image at the level of the aortic root in a patient with corrected tetralogy of Fallot. Note the position of the aortic orifice, overriding the ventricular septum and the pericardial patch (arrow), closing the outlet ventricular septal defect. (From Roest AA, Helbing WA, van der Wall EE, de Roos A. Postoperative evaluation of congenital heart disease by magnetic resonance imaging. J Magn Reson Imaging. 1999;10:656 666, with permission.)

Figure 29 3 Sagittal gradient recalled echo cardiovascular magnetic resonance image obtained during systole in a patient after correction of coarctation of the aorta. The area of signal loss (arrowhead) in the right ventricle from the interventricular septum is caused by turbulent flow across a perimembranous ventricular septal defect. (From Roest AA, Helbing WA, van der Wall EE, de Roos A. Postoperative evaluation of congenital heart disease by magnetic resonance imaging. J Magn Reson Imaging. 1999; 10:656 666, with permission.)

Estimation of the Qp:Qs ratio with echocardiography is not reliable because it is highly operator dependent.36 In addition, it is difficult to assess the cross sectional area of the vessels throughout the cardiac cycle, the limited sample volume does not necessarily represent mean velocity across the vessels, and the main pulmonary arteries are sometimes difficult to assess.22 Catheterization in combination with oximetry can be used; however, this remains an invasive procedure that exposes the infant or young child to poten tially harmful ionizing radiation. Radionuclide angiocardi ography is restricted to simple shunt lesions with normal ventricular function37 and also exposes the young child to radiation. Several reports have shown that CMR is ideal for the measurement of flow volume through the aorta and pulmo nary trunk using velocity encoded CMR from which the shunt ratio can be extracted (Fig. 29 4).22–24 Earlier reports assessed shunt volume in adult patients. In pediatric patients, higher peak velocities and blood pulsation rates may influence velocity and flow quantification.38 Data show that CMR accurately provides quantitative shunt volume data in the pediatric population.37,39 Advances in interventional cardiology have allowed for transcatheter closure of VSDs.40,41 CMR guided catheteri zation15 and intervention are now feasible,26–28,32 and the combination of X ray fused with CMR (XMR) has proven beneficial in the closure of perimembranous VSD.42 In swine models, the use of XMR guided antegrade catheter crossing and closure of the VSD is both easier and faster, and is associated with reduced radiation compared with conventional techniques.42 Cardiovascular Magnetic Resonance 397

29 CARDIOVASCULAR MAGNETIC RESONANCE OF SIMPLE CONGENITAL CARDIOVASCULAR DEFECTS

the correction is performed are based on the clinical status of the patient. Although the magnitude of the left to right shunt is important, there is controversy about the threshold Qp:Qs ratio (i.e., flow volume through the pulmonary cir culation divided by flow volume through the systemic cir culation) at which correction is indicated. Ratios greater than 2:1 are generally accepted as an indication for inter vention. However, a chronic moderate left to right shunt of greater than 1.5:1 to 1.8:1 also justifies closure because the LV is subjected to volume overload for a longer period and may remain dysfunctional after the defect is closed.34 Therefore, precise estimation of shunt flow is essential in the management of patients with a VSD. In addition, for surgical management of complex cardiac malformations, it is of utmost importance to know the spatial relationship between the VSD and the orifices of the great arteries. Depending on the location of the VSD, surgical manage ment can vary from a simple patch, inserted from the right atrium (Fig. 29 2), to a complicated biventricular correction.4,35 For anatomic delineation of VSDs, particularly those associated with complex malformations, CMR is superior to two dimensional echocardiography, especially in the visualization of the spatial relationship with surrounding structures, the atrioventricular valves and great arteries. To determine the location of the VSD, a black blood fast spin echo CMR imaging sequence is used and images are made in the three orthogonal planes. The transverse plane is the most useful for the identification of all types of VSDs.33 When precise information is needed about the shape and dimensions of the VSD and the spatial relation ship between the VSD and great arteries, these images should be completed with en face images of the VSD, made by planning a series of slices parallel to the VSD.35 A bright blood GRE CMR imaging sequence is used to visualize a jet caused by turbulent blood flow through the VSD (Fig. 29 3). As mentioned previously, precise shunt quantification is essential in the management of patients with a VSD.

250

Flow volume (mL/sec)

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

Figure 29 4 By subtracting the aortic flow curve from the pulmonary flow curve, one can quantify the amount of shunting in each heartbeat. The Qp:Qs ratio in this case was 1.7. (From Roest AA, Helbing WA, van der Wall EE, de Roos A. Postoperative evaluation of congenital heart disease by magnetic resonance imaging. J Magn Reson Imaging. 1999; 10:656 666, with permission.)

300

Pulmonary trunk 105 mL

200

Ascending aorta 63 mL

150 100 50 0

−100

0 −50

100

200

300

400

500

600

700

Time after R wave (msec)

−50

Atrial Septal Defect An ASD is a common cause of left to right intracardiac shunting. Patients with an ASD are generally asymptomatic through infancy and childhood. When symptoms occur in infancy, patients present with frequent respiratory infec tions, symptoms of volume overload of the pulmonary cir culation, and sometimes congestive heart failure. Later in childhood, a murmur is often heard because of high pul monary artery flow, and patients are referred for cardiac evaluation. After the age of 20 years, symptoms are more common and include dyspnea on exertion, fatigue, palpita tions, sustained atrial arrhythmias,43 and cryptogenic stroke.44 Several types of ASDs are recognized (Fig. 29 5). The most common type is the ostium secundum or fossa ovalis defect, which is located in the central part of the atrial septum (Fig. 29 6C). The sinus venosus defect is situated high or low on the septum near the entrance of, respectively, the superior vena cava or inferior vena cava into the right atrium (RA). This type may be associated with anomalous drainage of the right upper lobe pulmo nary vein into the RA (see Fig. 29 6A and B).45 The ostium primum defect is situated low in the atrial septum, close to the atrioventricular valves. This defect belongs to the spec trum of atrioventricular septal defects referred to as endo cardial cushion defects and is accompanied by abnormal position and structure of the atrioventricular valves. Transthoracic echocardiography is the first line clinical tool for detecting an ASD and is likely more sensitive for detecting a very small ASD. Transesophageal echocardiogra phy is a moderately invasive but even more accurate diag nostic tool. CMR can be useful, especially in older patients, who may have suboptimal echocardiographic windows.43 The presence of an ASD can be established definitively by using cine GRE CMR to identify the ASD jet across the sep tum, with visualization of a low signal jet on the right side of the interatrial septum.46 The definition of such flow related signal void can be enhanced by a spatial saturation slab at the inflow region of the ASD.47 The diameter of the defect 398 Cardiovascular Magnetic Resonance

SVC

3

2 1

3

TV

4

IVC

Figure 29 5 Most common locations of atrial septal defect, viewed from the right atrium. 1, ostium primum defect (also known as atrioventricular septal defect); 2, ostium secundum defect, or fossa ovalis defect; 3, sinus venosus defect; 4, coronary sinus defect; IVC, inferior vena cava; SVC, superior vena cava; TV tricuspid valve.

can be derived from the maximum width of the transseptal flow acquired with multiple parallel and intersecting CMR acquisitions.48 Especially phase contrast CMR is used to determine the size of the ASD, whereas spin echo CMR tech niques tend to overestimate the size of the defect.9 The size and morphologic features of the ASD and its anatomic

RV

RVOT

AA

AA

RA

RA

*

RPA

A

LA

LA

B

LV

C

Figure 29 6 Axial fast spin echo images in a 26 year old woman with partial abnormal pulmonary venous return showing the right upper lobe pulmonary vein (RULPV) draining into the superior vena cava (SVC; A). A large sinus venosus type atrial septal defect (ASD) is present high in the atrial septum (*; B). At the lower level, a fossa ovale, or ostium secundum, ASD is shown (arrow; C). A thin membrane may be present at this fossa ovale level despite visual interruption of the atrium septum, as shown in these images. Sizing of the ASD is best performed with velocity encoded or cine gradient recalled echo cardiovascular magnetic resonance. AA, ascending aorta; L, left; LV, left ventricle; P, posterior; RA, right atrium; RPA, right pulmonary artery; RV, right ventricle; RVOT, right ventricular outflow tract.

relationships are more important because percutaneous clo sure of ostium secundum ASDs is now routinely performed. The location, size of the defect, and age of the patient are the major determining factors for the choice between sternot omy/surgical correction and percutaneous interventional clo sure.49 Transthoracic echocardiography is not always accurate in the precise evaluation of the ASD.10 Transeso phageal echocardiography is more accurate, but is moder ately invasive.50 CMR is an accurate and noninvasive method for eva luation of the size and morphologic features of ASDs. Fur thermore, CMR provides precise information about the surrounding structures, such as the atrioventricular valves and the entrance of the systemic and pulmonary veins.9–12 In children in whom transesophageal echocardiography was inconclusive for deciding on percutaneous or surgical closure of the ASD, CMR provided additional information, thereby assisting in the planning of interventional ASD clo sure.10 In adults with a patent foramen ovale and crypto genic ischemic events, CMR currently appears to be inferior to transesophageal echocardiography in the detec tion of agitated saline demonstrated right to left shunt shunting and identification of an atrial septal aneurysm.50 The advantage of CMR in evaluating patients with left to right shunt at the atrial level is the ability to assess the func tional significance by determining flow through the shunt. In ASDs, the Qp:Qs ratio and shunt flow can be extracted from the discrepancy between RV and LV stroke volumes and by measuring aortic and pulmonary flow with flow mapping (Fig. 29 7).22,24,37,51,52 This provides for independent confir mation of the size of the intracardiac shunt (in the absence of significant mitral regurgitation and tricuspid regurgitation). Direct measurement of flow across the ASD is also possi ble.37,48,53 Furthermore, the effects of shunting through an ASD on RV size can be readily quantified by CMR. The use of CMR guided transcatheter closure of ASDs has been reported in animal models using real time CMR.26–28 This is a promising application of CMR to fur ther decrease the use of ionizing radiation in patients with CHD. The acute effect of transcatheter closure of an ASD

on atrial and ventricular parameters can be assessed by performing CMR shortly before and after the interven tion.54 Within 24 hours after closure of an ASD, the mean size of the RA and RV decreases, whereas the LA and LV volume remains unchanged.54

[mL/sec] 0

Flow Volume 40 60

20

300

80

100

59.19 mL/beat (4555.0 mL/min)

250 200

Qs

150 100 50 0 -50

A

0

100

[mL/sec] 0

200

400

500

Flow Volume 40 60

20

700

300

600

700

80

800 100

169.74 mL/beat (13010.7 mL/min)

600 500

Qp

400 300 200 100 0

B

0

100

200

300

400

500

600

700

800

Figure 29 7 Same patient as shown in Figure 29 6, with partial abnormal pulmonary venous return and atrial septal defect, both contributing to a left to right shunt. Graphic reconstruction representing flow over one heart phase through the aortic valve (A) and through the pulmonary valve (B). The shunt size was calculated by dividing the flow through the pulmonary valve, representing the pulmonary circulation Qp, by flow over the aortic valve, representing the systemic flow Qs. Qp was 13.0 L/min and Qs was 4.6 L/min. The shunt size was therefore 2.8 (13.0/4.6). Cardiovascular Magnetic Resonance 399

29 CARDIOVASCULAR MAGNETIC RESONANCE OF SIMPLE CONGENITAL CARDIOVASCULAR DEFECTS

SVC RULPV

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

After closure of the ASD, CMR can be used for the evalu ation of cardiac function. If a septal occluder has been used for ASD closure, CMR can clearly show the location of the device, its effect on systemic and pulmonary venous return, and the presence of valve insufficiency.55 Furthermore, CMR can be used for follow up of ventricular volumes and func tion after surgical or interventional ASD closure.56–58,58a

Patent Ductus Arteriosus The ductus arteriosus is the connecting vessel between the pulmonary trunk and the descending aorta. In utero, most of the RV stroke volume bypasses the still unexpanded lungs and enters the systemic circulation through the duc tus arteriosus. In the vast majority of infants, the ductus arteriosus closes within the first week of life. There is delayed closure of the ductus arteriosus in preterm infants. Spontaneous closure occurs in most preterm infants in the first month of life. In term infants, spontaneous closure of a PDA is rare because of abnormalities in the structure of the ductus.59 Interventional closure of a PDA is needed when the shunt flow is hemodynamically significant and is also recommended in the case of a small ductus without hemo dynamic significance because of the increased risk of infec tive endocarditis.59 A PDA results in left to right shunt and therefore volume overload of the pulmonary circulation and LV, which may lead to pulmonary hypertension. The amount of shunting depends on the ductus size and the difference between pulmonary vascular resistance and systemic vascular resistance.59 Although a PDA is a frequently encountered CHD mal formation, there are few reports on evaluation with CMR.60–64 Transthoracic echocardiography is the first line clinical tool for detecting a PDA, especially in infants and young children. In adult patients, transthoracic echocardi ography may be less sensitive than CMR.60 Using CE MRA, a PDA is readily seen.61,62 Measurement of shunt vol ume can be performed by assessing LV and RV stroke volumes with cine GRE or SSFP CMR or by evaluation of flow volume over the pulmonary and aortic valves with flow mapping.61 The Qp:Qs ratio is less than 1.0 in patients with a PDA. Furthermore, CMR can be used to assess the complications of chronic left to right shunting caused by a PDA, such as dilation of the pulmonary arteries, LV dysfunction as a result of chronic volume overload, and dilation and hypertrophy of the RV second ary to pulmonary artery hypertension.60,63 Similar results appear to be available at 3 Tesla.64a

poststenotic dilation is frequently observed. A coarctation can occur in isolation, but can be associated with VSDs, aortic valve stenosis, mitral valve anomalies, double outlet RV, transposition of the great arteries, hypoplastic left heart syndrome, and tricuspid atresia.65 In severe coarctation, signs of cardiac failure and decreased or absent femoral pulsations develop within the first weeks of life as the ductus arteriosus closes. In milder coarctation, after closure of the ductus arteriosus, cardio vascular adaptation will occur, including LV hypertrophy and cavity dilation, increased sympathetic activation causing hypertension, and development of a collateral circulation to bypass the coarctation.65 Currently, CMR is considered the standard noninvasive technique for the evaluation of native and repaired aortic coarctation.66,67 Associated abnormalities, such as arch hypoplasia, bicuspid aortic valve, and VSD, can also be assessed with CMR.68 CMR readily identifies the site and extent of coarctation, involvement of arch vessels, postste notic dilation, and dilated collateral vessels (Figs. 29 8 to 29 10).69,70 CE MRA has emerged as a valuable technique to assess the thoracic aorta and is now routinely performed in older children and adults, resulting in high quality aor tograms with the aid of an intravenous infusion of gadolin ium to shorten the arterial blood T1 relaxation time (see Figs. 29 9 and 29 10).25 Velocity encoded CMR measure ment of peak jet velocity across the coarctation provides comparable estimates of the pressure gradient to those obtained from continuous wave Doppler echocardiography (by application of the modified Bernoulli formula).71,72 Significant narrowing will impair blood flow into the descending aorta; therefore, collateral vessels are required

Coarctation of the Aorta Coarctation of the aorta most commonly occurs as a dis crete stenosis of the proximal descending aorta, just oppo site the (former) insertion of the ductus arteriosus (juxtaductal location). Most commonly, coarctation of the aorta is congenital, although coarctation of the aorta can occur after Takayasu arteritis. The gross morphologic fea tures of the coarctation may vary from a discrete narrowing to long segment stenosis. Distal to the coarctation, 400 Cardiovascular Magnetic Resonance

Figure 29 8 Oblique sagittal fast spin echo image of a 10 year old boy showing severe aortic coarctation (arrow) at the classic position in the aorta distal from the origin of the left subclavian artery.

B

C

Figure 29 9 Same patient as shown in Figure 29 8. Contrast enhanced magnetic resonance angiography after injection of gadolinium chelate. There are several postprocessing options for displaying the site of coarctation. A, Maximum intensity projection. B and C, Shaded surface display. Note the large collateral arteries entering the descending aorta distal from the coarctation (arrows).

A

B

C

Figure 29 10 Same patient as shown in Figure 29 8. A, Volume rendering image. B and C, Shaded surface display. Changing the window width and window level settings allows more collateral arteries to be observed. Large collateral arteries include the internal mammarian arteries to the abdomen (double arrows in A and C) and large thoracic wall collateral artery (single arrow in A, B, and C) entering an intercostal artery (B). Note the large intercostal arteries (B). Also note other collateral flow pathways to the lower body, such as internal mammarian artery flow via the epigastric arteries (arrowheads in C).

to reestablish aortic flow distal to the coarctation. The inter costal, lateral thoracic, internal mammary, anterior spinal, and epigastric arteries can all serve as collateral pathways. Flow mapping has proven to be a valuable adjunct to assess the severity of coarctation by measuring flow in the proxi mal and distal descending thoracic aorta (Figs. 29 11 and

29 12).73,74 This method for assessing the severity of coarc tation is based on the increase of flow in the distal aorta above the diaphragm with regard to flow near the coarcta tion site. Measurement within the coarctation tends to over estimate flow, whereas measurement of flow immediately above and below the coarctation yields similar findings Cardiovascular Magnetic Resonance 401

29 CARDIOVASCULAR MAGNETIC RESONANCE OF SIMPLE CONGENITAL CARDIOVASCULAR DEFECTS

A

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

[mL/sec]

0.6 L/min

[mL/sec]

0.6 L/min

[mL/sec]

2.6 L/min

A

B

C

Figure 29 11 Same patient as shown in Figure 29 8. A, Maximum intensity projection image. B, Double oblique transverse gradient recalled echo cardiovascular magnetic resonance at the aortic level indicated with phase and modulus images just proximal to the coarctation (upper panel), just distal to the coarctation (middle panel), and in the abdominal aorta (lower panel). C, Graphic reconstructions representing flow during the cardiac cycle at the corresponding levels, where flow volumes are calculated per minute. The flow volume at the level immediately before (and after) the coarctation was 0.6 L/min. Flow volume at the level of the abdominal aorta was 2.6 L/min. This indicates that almost 80% of the blood flow through the descending aorta was derived from collateral artery flow (2.6 0.6)/2.6. The flow volume measured just above the aortic valve was 5.0 L/min.

(see Fig. 29 11).74 Retrograde flow in collateral channels will increase distal aortic flow, depending on the severity of coarctation, thereby providing a direct estimate of the hemodynamic severity of the coarctation. Combining ana tomic and flow techniques, CMR provides a sensitive and specific test for predicting a catheterization gradient of greater than 20 mm Hg, which is considered an important criterion for a significant coarctation.75 Repair of native coarctation of the aorta can be per formed either surgically or with catheter based balloon angioplasty and stenting. Comparison of the long term out come of native coarctation in children has shown that bal loon angioplasty is associated with a higher incidence of aneurysm formation and iliofemoral artery injury compared with surgery.76 After initial coarctation repair, restenosis and aneurysm formation frequently occur. CMR guidance can be used to monitor for these complications and to guide subsequent repair.30,32,77 Because CMR is superior to Doppler echocardiography in evaluating patients with a corrected coarctation,78 it is ideally suited for long term follow up. CMR is now consid ered the standard method for the evaluation of children and adults with corrected coarctation of the aorta because there is no exposure to ionizing radiation.79–81 Spin echo CMR can be used to assess the anatomy of the aorta, but the addition of GRE and SSFP cine CMR as well as CE MRA provides more detailed information than spin echo CMR.78,80,82 Furthermore, velocity encoded CMR, preferably in plane as well as through plane, provides an 402 Cardiovascular Magnetic Resonance

accurate estimate of the gradient over a restenosis.78,83 After repair, information on collateral flow, as assessed with CMR, proved more accurate than arm/leg blood pressure gradient in the assessment of the hemodynamic severity of restenosis and may be helpful for planning treatment options and monitoring patient outcome.84

Figure 29 12 Transverse spin echo cardiovascular magnetic resonance image at the level of the atrioventricular valves in a patient with tricuspid valve atresia. The single headed arrow indicates the right coronary artery surrounded by fat, located at the site of the atretic tricuspid valve.

Valvular Heart Disease Valvular abnormalities are frequently seen in patients with CHD, either as a congenital malformation or as a result of treatment. CMR can be used for evaluation of valvular anat omy and function. Valve abnormalities frequently seen are atresia of one of the valves (Fig. 29 12), Ebstein anomaly, bicuspid semilunar valve (Figs. 29 13 and 29 14), single atrioventricular valve, and congenital mitral valve malfor mations. Functional valve abnormalities occur either con genitally or because of intervention. Both valvular stenosis and regurgitation can be quantified by velocity encoded CMR. Estimation of the severity of stenosis is possible using velocity encoded CMR.86 As discussed previously, measure ment of peak velocity, preferably in the orthogonal plane,

A

allows estimation of the pressure gradient over the valve by applying the modified Bernoulli equation.66,87 In the case of insufficiency of a valve, the volume of forward and backward flow can be measured, allowing quantifica tion of regurgitation volume (Fig. 29 15).88 A major diffi culty associated with using CMR in the evaluation of valvular function is movement of the valve during the car diac cycle, influencing flow and velocity measure ments.89,90 This problem can be overcome by applying dedicated scanning protocols that allow more accurate measurements.91–93 The most frequently encountered valvular abnormality is bicuspid aortic valve, occurring in 1% of the adult popula tion.94 Complications of bicuspid aortic valve are aortic valve stenosis and regurgitation, progressive aortic dilation, aortic aneurysm formation, and aortic dissection. Dilation of the aortic root is associated with intrinsic aortic wall pathology in patients with bicuspid aortic valve.94 CMR can be used to evaluate valvular competency and the effect of the bicus pid valve on LV volume and systolic function.94 Further more, aortic distensibility and pulse wave velocity can be

B

PT AA

RV LV

C Figure 29 13 Double oblique sagittal gradient recalled echo cardiovascular magnetic resonance in a 23 year old woman at the aortic valve level showing the slit like opening of the bicuspid aortic valve. A, Modulus image. B, Corresponding phase image. C, Coronal fast spin echo image showing poststenotic dilation of the ascending aorta (AA). Also, the left ventricle (LV) was dilated with a calculated end diastolic volume of 285 mL. F, feet; L, left; PT, pulmonary trunk; RA, right atrium; RV, right ventricle. Cardiovascular Magnetic Resonance 403

29 CARDIOVASCULAR MAGNETIC RESONANCE OF SIMPLE CONGENITAL CARDIOVASCULAR DEFECTS

In conclusion, the combination of clinical assessment and CMR in every patient after surgical or interventional repair of coarctation of the aorta is the most cost effective way to diagnose complications.85

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

assessed by CMR. Reduced aortic elasticity and aortic dila tion have been shown to correlate with aortic valve regurgi tation and LV hypertrophy in patients with nonstenotic bicuspid aortic valves.94 In patients with aortic regurgitation, quantification of regurgitant volume is superior to indices of LV volume and systolic function for identification of patients requiring surgery.95 The pulmonary valve is frequently affected in CHD, and the application of dedicated CMR protocols with long and short axis views allows detailed anatomic and functional evaluation of this valve and the influence of dysfunction on the RV (Fig. 29 14).96 Additional informa tion on all sources of blood supply to the lungs is essential in the evaluation of pulmonary stenosis and pulmonary atresia. Multiplanar GRE and SSFP cine CMR and CE MRA are fast and accurate techniques for this purpose (see Fig. 29 14C).97 As a result of stenosis or regurgitation, the cardiac chamber situated upstream of the abnormal valve will react to the increased workload. Pressure overload

A

resulting from stenosis causes myocardial hypertrophy and eventually ventricular dilation (see Fig. 29 14D), whereas chronic regurgitation causes chamber enlarge ment because of volume overload. With GRE and SSFP cine CMR, the severity and progression of these secondary abnormalities can be analyzed by measuring LV and RV volumes, wall thickness, and mass.88 After valve replacement, CMR can be used to evaluate the amount of residual regurgitation or stenosis and the effect of replacement on cardiac function.98 Nonsurgical valve replacement has been introduced for correction of ste nosed or insufficient pulmonary valves.99,100 CMR proved to be essential in the selection of patients for percutaneous pulmonary valve replacement.101 In a swine model, it was shown that CMR can be used to guide stenting of the pulmonary and aortic valve and the pulmonary arteries, with immediate postinterventional evaluation of flow within the stent.29,102,103 In the future, CMR may be used for routine guidance and follow up of these interventional procedures.

B

C

RV RA

D

a

RV

LV

RA

LV

b

Figure 29 14 Gradient recalled echo cardiovascular magnetic resonance images in a 22 year old man with an isolated pulmonary valve stenosis caused by a bicuspid pulmonary valve (arrow, axial image A). A turbulent flow jet caused by this valve is visible in the main pulmonary artery (arrow, axial image B). Note the asymmetrically developed branch pulmonary arteries: the large pulmonary artery on the left and the hypoplastic pulmonary artery on the right. Contrast enhanced magnetic resonance angiography shows asymmetrical pulmonary perfusion with preferential flow to the left lung and an apical perfusion defect in the right lung (coronal image C). D, Gradient recalled echo imaging, axial orientation. End diastolic (a) and end systolic (b) images at the midventricular level. Note the severely hypertrophied right ventricle (RV) with the interventricular septum bulging toward the left ventricle (LV). RA, right atrium.

404 Cardiovascular Magnetic Resonance

Flow Volume

0

20

40

60

80

100

500

AA

56.4 mL/beat (3.9L/min)

PA

400 300 200

100

B

100

9

0 -100

53

AA PA

-200 -300 0

100

200

300

400

500

600

700

A

800

900

C

Figure 29 15 Follow up cardiovascular magnetic resonance study in a 19 year old woman with corrected tetralogy of Fallot and pulmonary regurgitation. A, Graphic reconstruction representing flow over one heart phase through the pulmonary valve. Forward flow during systole was 100 mL, and regurgitant flow during diastole was 53 mL. End diastolic forward flow was 9 mL because of a restrictive right ventricle. Total regurgitant fraction was 53/109 49%. B and C, Double oblique transverse gradient recalled echo magnetic resonance phase and corresponding modulus images on pulmonary valve level showing the forward flow as a bright area and the regurgitant flow as a dark area measured at the level just above the pulmonary valve in the main pulmonary artery (PA). AA, ascending aorta.

CONCLUSION In conclusion, CMR has emerged as an indispensable tool in the management of patients with CHD. CMR in pediat ric patients requires dedicated personnel and equipment, but when available, CMR is a valuable tool in the diag nosis and follow up of patients with CHD. Several CMR techniques are available for the accurate delineation of anatomy of the heart and the great vessels, allowing determination of the location of intracardiac shunts and vascular abnormalities and their relationship to surrounding

structures. Additional to providing morphologic infor mation, CMR has the unique capability of quantification of ventricular function and blood flow velocity and volume, allowing quantification of intra and extracardiac shunting, valvular function, and ventricular performance. Because of its non invasive, non ionizing nature, CMR is ideally suited for monitoring of cardiovascular function during follow up or to evaluate the effect of intervention in patients with CHD. Advances in interventional cardiology and XMR will further expand the application of CMR in patients with CHD.

References 1. Hoffman JI. Incidence of congenital heart disease: I. Postnatal incidence. Pediatr Cardiol. 1995;16(3):103–113. 2. Hoffman JI, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol. 2002;39(12):1890–1900. 3. Tsai-Goodman B, Geva T, Odegard KC, Sena LM, Powell AJ. Clinical role, accuracy, and technical aspects of cardiovascular magnetic resonance imaging in infants. Am J Cardiol. 2004;94(1):69–74. 4. Beekman RP, Roest AA, Helbing WA, et al. Spin echo MRI in the evaluation of hearts with a double outlet right ventricle: usefulness and limitations. Magn Reson Imaging. 2000;18(3):245–253. 5. Boehrer JD, Lange RA, Willard JE, Grayburn PA, Hillis LD. Advantages and limitations of methods to detect, localize, and quantitate intracardiac right-to-left and bidirectional shunting. Am Heart J. 1993;125(1):215–220. 6. Powell AJ, Tsai-Goodman B, Prakash A, Greil GF, Geva T. Comparison between phase-velocity cine magnetic resonance imaging and invasive oximetry for quantification of atrial shunts. Am J Cardiol. 2003;91(12):1523–1525. 7. Stark J. Do we really correct congenital heart defects? J Thorac Cardiovasc Surg. 1989;97(1):1–9. 8. Moodie DS. Adult congenital heart disease. Curr Opin Cardiol. 1994;9(1):137–142. 9. Holmvang G, Palacios IF, Vlahakes GJ, et al. Imaging and sizing of atrial septal defects by magnetic resonance. Circulation. 1995;92 (12):3473–3480.

10. Beerbaum P, Korperich H, Esdorn H, et al. Atrial septal defects in pediatric patients: noninvasive sizing with cardiovascular MR imaging. Radiology. 2003;228(2):361–369. 11. Durongpisitkul K, Tang NL, Soongswang J, Laohaprasitiporn D, Nanal A. Predictors of successful transcatheter closure of atrial septal defect by cardiac magnetic resonance imaging. Pediatr Cardiol. 2004;25(2):124–130. 12. Piaw CS, Kiam OT, Rapaee A, et al. Use of non-invasive phase contrast magnetic resonance imaging for estimation of atrial septal defect size and morphology: a comparison with transesophageal echo. Cardiovasc Intervent Radiol. 2006;29(2):230–234. 13. Einstein AJ, Henzlova MJ, Rajagopalan S. Estimating risk of cancer associated with radiation exposure from 64-slice computed tomography coronary angiography. JAMA. 2007;298:317–323. 14. Schalla S, Saeed M, Higgins CB, Martin A, Weber O, Moore P. Magnetic resonance–guided cardiac catheterization in a swine model of atrial septal defect. Circulation. 2003;108(15): 1865–1870. 15. Razavi R, Hill DL, Keevil SF, et al. Cardiac catheterisation guided by MRI in children and adults with congenital heart disease. Lancet. 2003;362(9399):1877–1882. 16. Samyn MM. A review of the complementary information available with cardiac magnetic resonance imaging and multi-slice computed tomography (CT) during the study of congenital heart disease. Int J Cardiovasc Imaging. 2004;20(6):569–578.

Cardiovascular Magnetic Resonance 405

29 CARDIOVASCULAR MAGNETIC RESONANCE OF SIMPLE CONGENITAL CARDIOVASCULAR DEFECTS

[mL/sec]

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

17. Beekman RP, Beek FJ, Hazekamp MG, Meijboom EJ. The value of MRI in diagnosing vascular abnormalities causing stridor. Eur J Pediatr. 1997;156(7):516–520. 18. Boothroyd A. Magnetic resonance: its current and future role in paediatric cardiac radiology. Eur J Radiol. 1998;26(2):154–162. 19. Didier D, Ratib O, Beghetti M, Oberhaensli I, Friedli B. Morphologic and functional evaluation of congenital heart disease by magnetic resonance imaging. J Magn Reson Imaging. 1999;10(5):639–655. 20. Odegard KC, DiNardo JA, Tsai-Goodman B, Powell AJ, Geva T, Laussen PC. Anaesthesia considerations for cardiac MRI in infants and small children. Paediatr Anaesth. 2004;14(6):471–476. 21. Lorenz CH, Walker ES, Morgan VL, Klein SS, Graham Jr TP. Normal human right and left ventricular mass, systolic function, and gender differences by cine magnetic resonance imaging. J Cardiovasc Magn Reson. 1999;1(1):7–21. 22. Mohiaddin RH, Underwood R, Romeira L, et al. Comparison between cine magnetic resonance velocity mapping and first-pass radionuclide angiocardiography for quantitating intracardiac shunts. Am J Cardiol. 1995;75(7):529–532. 23. Hundley WG, Li HF, Lange RA, et al. Assessment of left-to-right intracardiac shunting by velocity-encoded, phase-difference magnetic resonance imaging: a comparison with oximetric and indicator dilution techniques. Circulation. 1995;91(12):2955–2960. 24. Arheden H, Holmqvist C, Thilen U, et al. Left-to-right cardiac shunts: comparison of measurements obtained with MR velocity mapping and with radionuclide angiography. Radiology. 1999;211 (2):453–458. 25. Prince MR, Narasimham DL, Jacoby WT, et al. Three-dimensional gadolinium-enhanced MR angiography of the thoracic aorta. AJR Am J Roentgenol. 1996;166(6):1387–1397. 26. Buecker A, Spuentrup E, Grabitz R, et al. Magnetic resonance-guided placement of atrial septal closure device in animal model of patent foramen ovale. Circulation. 2002;106(4):511–515. 27. Rickers C, Jerosch-Herold M, Hu X, et al. Magnetic resonance imageguided transcatheter closure of atrial septal defects. Circulation. 2003;107(1):132–138. 28. Schalla S, Saeed M, Higgins CB, Weber O, Martin A, Moore P. Balloon sizing and transcatheter closure of acute atrial septal defects guided by magnetic resonance fluoroscopy: assessment and validation in a large animal model. J Magn Reson Imaging. 2005;21(3): 204–211. 29. Kuehne T, Saeed M, Higgins CB, et al. Endovascular stents in pulmonary valve and artery in swine: feasibility study of MR imagingguided deployment and postinterventional assessment. Radiology. 2003;226(2):475–481. 30. Ince H, Petzsch M, Rehders T, et al. Percutaneous endovascular repair of aneurysm after previous coarctation surgery. Circulation. 2003;108(24):2967–2970. 31. Raval AN, Telep JD, Guttman MA, et al. Real-time magnetic resonance imaging-guided stenting of aortic coarctation with commercially available catheter devices in Swine. Circulation. 2005;112(5): 699–706. 32. Krueger JJ, Ewert P, Yilmaz S, et al. Magnetic resonance imagingguided balloon angioplasty of coarctation of the aorta: a pilot study. Circulation. 2006;113(8):1093–1100. 33. Didier D, Higgins CB. Identification and localization of ventricular septal defect by gated magnetic resonance imaging. Am J Cardiol. 1986;57(15):1363–1368. 34. Wilkinson JL. Ventricular septal defect. In: Moller JH, Hofman JIE, eds. Pediatric Cardiovascular Medicine. 1st ed. 2000:289–310. 35. Yoo SJ, Lim TH, Park IS, et al. MR anatomy of ventricular septal defect in double-outlet right ventricle with situs solitus and atrioventricular concordance. Radiology. 1991;181(2):501–505. 36. Boehrer JD, Lange RA, Willard JE, Grayburn PA, Hillis LD. Advantages and limitations of methods to detect, localize, and quantitate intracardiac left-to-right shunting. Am Heart J. 1992;124(2):448–455. 37. Beerbaum P, Korperich H, Barth P, Esdorn H, Gieseke J, Meyer H. Noninvasive quantification of left-to-right shunt in pediatric patients: phase-contrast cine magnetic resonance imaging compared with invasive oximetry. Circulation. 2001;103(20):2476–2482. 38. Rebergen SA, van der Wall EE, Doornbos J, de Roos A. Magnetic resonance measurement of velocity and flow: technique, validation, and cardiovascular applications. Am Heart J. 1993;126(6): 1439–1456. 39. Korperich H, Gieseke J, Barth P, et al. Flow volume and shunt quantification in pediatric congenital heart disease by real-time magnetic

406 Cardiovascular Magnetic Resonance

40.

41. 42.

43. 44. 45. 46.

47. 48. 49. 50.

51. 52.

53.

54.

55. 56.

57.

58.

58a. 59. 60.

resonance velocity mapping: a validation study. Circulation. 2004;109(16):1987–1993. Holzer R, Balzer D, Cao QL, Lock K, Hijazi ZM. Device closure of muscular ventricular septal defects using the Amplatzer muscular ventricular septal defect occluder: immediate and mid-term results of a U.S. registry. J Am Coll Cardiol. 2004;43(7):1257–1263. Knauth AL, Lock JE, Perry SB, et al. Transcatheter device closure of congenital and postoperative residual ventricular septal defects. Circulation. 2004;110(5):501–507. Ratnayaka K, Raman VK, Faranesch AR, et al. Antegrade percutaneous closure of membraneous ventricular septal defect using X-ray fused with magnetic resonance imaging. JACC Cardiovasc Imaging 2008;2:224–230. Latson LA. Atrial septal defect. In: Moller JH, Hofman JIE, eds. Pediatric Cardiovascular Medicine. 1st ed. 2000:311–322. Wechsler LR. PFO and stroke: what are the data? Cardiol Rev. 2008;16 (1):53–57. Valente AM, Sena L, Powell AJ, Del Nido PJ, Geva T. Cardiac magnetic resonance imaging evaluation of sinus venosus defects: comparison to surgical findings. Pediatr Cardiol. 2007;28(1):51–56. Theissen P, Sechtem U, Mennicken U, Hilger HH, Schicha H. [Noninvasive diagnosis of atrial septal defects and anomalous pulmonary venous return using magnetic resonance tomography]. Nuklearmedizin. 1989;28(5):172–180. Hartnell GG, Sassower M, Finn JP. Selective presaturation magnetic resonance angiography: new method for detecting intracardiac shunts. Am Heart J. 1993;126(4):1032–1034. Holmvang G. A magnetic resonance imaging method for evaluating atrial septal defects. J Cardiovasc Magn Reson. 1999;1(1):59–64. Ferreira SM, Ho SY, Anderson RH. Morphological study of defects of the atrial septum within the oval fossa: implications for transcatheter closure of left-to-right shunt. Br Heart J. 1992;67(4):316–320. Nusser T, Hoher M, Merkle N, et al. Cardiac magnetic resonance imaging and transesophageal echocardiography in patients with transcatheter closure of patent foramen ovale. J Am Coll Cardiol. 2006;48:322–329. Brenner LD, Caputo GR, Mostbeck G, et al. Quantification of left to right atrial shunts with velocity-encoded cine nuclear magnetic resonance imaging. J Am Coll Cardiol. 1992;20(5):1246–1250. Sieverding L, Jung WI, Klose U, Apitz J. Noninvasive blood flow measurement and quantification of shunt volume by cine magnetic resonance in congenital heart disease: preliminary results. Pediatr Radiol. 1992;22(1):48–54. Thomson LEJ, Crowley AL, Heitner JF, et al. Direct en face imaging of secundum atrial septal defects by velocity-encoded cardiovascular magnetic resonance in patients evaluated for possible transcatheter closure. Circ Cardiovasc Imaging. 2008;1:31–40. Burgstahler C, Wohrle J, Kochs M, et al. Magnetic resonance imaging to assess acute changes in atrial and ventricular parameters after transcatheter closure of atrial septal defects. J Magn Reson Imaging. 2007;25(6):1136–1140. Lapierre C, Raboisson MJ, Miro J, Dahdah N, Guerin R. Evaluation of a large atrial septal occluder with cardiac MR imaging. Radiographics. 2003;23:S51–S58. Bolz D, Lacina T, Buser P, Buser M, Guenthard J. Long-term outcome after surgical closure of atrial septal defect in childhood with extensive assessment including MRI measurement of the ventricles. Pediatr Cardiol. 2005;26(5):614–621. Schoen SP, Kittner T, Bohl S, et al. Transcatheter closure of atrial septal defects improves right ventricular volume, mass, function, pulmonary pressure, and functional class: a magnetic resonance imaging study. Heart. 2006;92(6):821–826. Weber M, Dill T, Deetjen A, et al. Left ventricular adaptation after atrial septal defect closure assessed by increased concentrations of N-terminal pro-brain natriuretic peptide and cardiac magnetic resonance imaging in adult patients. Heart. 2006;92 (5):671–675. Teo KSL, Dundon BK, Molaee P, et al. Percutaneous closure of atrial septal defects leads to normalization of atrial and ventricular volumes. J Cardiovasc Magn Resonan. 2008;10:55. Gersony WM, Apfel HD. Patent ductus arteriosus and other aortopulmonary anomalies. In: Moller JH, Hofman JIE, eds. Pediatric Cardiovascular Medicine. 1st ed. 2000:323–334. Schmidt M, Theissen P, Deutsch HJ, Erdmann E, Schicha H. Magnetic resonance imaging of ductus arteriosus Botalli apertus in adulthood. Int J Cardiol. 1999;68(2):225–229.

82. Riquelme C, Laissy JP, Menegazzo D, et al. MR imaging of coarctation of the aorta and its postoperative complications in adults: assessment with spin-echo and cine-MR imaging. Magn Reson Imaging. 1999;17 (1):37–46. 83. Henk CB, Grampp S, Koller J, et al. Elimination of errors caused by first-order aliasing in velocity encoded cine-MR measurements of postoperative jets after aortic coarctation: in vitro and in vivo validation. Eur Radiol. 2002;12(6):1523–1531. 84. Araoz PA, Reddy GP, Tarnoff H, Roge CL, Higgins CB. MR findings of collateral circulation are more accurate measures of hemodynamic significance than arm-leg blood pressure gradient after repair of coarctation of the aorta. J Magn Reson Imaging. 2003;17(2):177–183. 85. Therrien J, Thorne SA, Wright A, Kilner PJ, Somerville J. Repaired coarctation: a “cost-effective” approach to identify complications in adults. J Am Coll Cardiol. 2000;35(4):997–1002. 86. Kilner PJ, Manzara CC, Mohiaddin RH, et al. Magnetic resonance jet velocity mapping in mitral and aortic valve stenosis. Circulation. 1993;87(4):1239–1248. 87. Caruthers SD, Lin SJ, Brown P, et al. Practical value of cardiac magnetic resonance imaging for clinical quantification of aortic valve stenosis: comparison with echocardiography. Circulation. 2003;108(18):2236–2243. 88. Niezen RA, Helbing WA, van der Wall EE, van der Geest RJ, Rebergen SA, de Roos A. Biventricular systolic function and mass studied with MR imaging in children with pulmonary regurgitation after repair for tetralogy of Fallot. Radiology. 1996;201(1):135–140. 89. Reid SA, Walker PG, Fisher J, et al. The quantification of pulmonary valve haemodynamics using MRI. Int J Cardiovasc Imaging. 2002;18 (3):217–225. 90. Chatzimavroudis GP, Walker PG, Oshinski JN, Franch RH, Pettigrew RI, Yoganathan RI. Slice location dependence of aortic regurgitation measurements with MR phase velocity mapping. Magn Reson Med. 1997;37(4):545–551. 91. Kozerke S, Schwitter J, Pedersen EM, Boesiger P. Aortic and mitral regurgitation: quantification using moving slice velocity mapping. J Magn Reson Imaging. 2001;14(2):106–112. 92. Friedrich MG, Schulz-Menger J, Poetsch T, Pilz B, Uhlich F, Dietz R. Quantification of valvular aortic stenosis by magnetic resonance imaging. Am Heart J. 2002;144(2):329–334. 93. Westenberg JJ, Doornbos J, Versteegh MI, et al. Accurate quantitation of regurgitant volume with MRI in patients selected for mitral valve repair. Eur J Cardiothorac Surg. 2005;27(3):462–466. 94. Grotenhuis HB, Ottenkamp J, Westenberg JJ, Bax JJ, Kroft LJ, de Roos A. Reduced aortic elasticity and dilatation are associated with aortic regurgitation and left ventricular hypertrophy in nonstenotic bicuspid aortic valve patients. J Am Coll Cardiol. 2007;49(15):1660–1665. 95. Myerson SG, Karamitsos TD, Francis JM, Banning AP, Neubauer S. Quantifying aortic regurgitation with CMR can predict patients requiring aortic valve surgery. J Cardiovasc Magn Reson. 2008. 10(suppl). 96. Kivelitz DE, Dohmen PM, Lembcke A, et al. Visualization of the pulmonary valve using cine MR imaging. Acta Radiol. 2003;44(2):172–176. 97. Greil GF, Powell AJ, Gildein HP, Geva T. Gadolinium-enhanced threedimensional magnetic resonance angiography of pulmonary and systemic venous anomalies. J Am Coll Cardiol. 2002;39(2):335–341. 98. Vliegen HW, van SA, de Roos A, et al. Magnetic resonance imaging to assess the hemodynamic effects of pulmonary valve replacement in adults late after repair of tetralogy of Fallot. Circulation. 2002;106 (13):1703–1707. 99. Khambadkone S, Bonhoeffer P. Nonsurgical pulmonary valve replacement: why, when, and how? Catheter Cardiovasc Interv. 2004;62(3):401–408. 100. Khambadkone S, Coats L, Taylor A, et al. Percutaneous pulmonary valve implantation in humans: results in 59 consecutive patients. Circulation. 2005;112(8):1189–1197. 101. Schievano S, Coats L, Migliavacca F, et al. Variations in right ventricular outflow tract morphology following repair of congenital heart disease: implications for percutaneous pulmonary valve implantation. J Cardiovasc Magn Reson. 2007;9(4):687–695. 102. Kuehne T, Yilmaz S, Meinus C, et al. Magnetic resonance imaging-guided transcatheter implantation of a prosthetic valve in aortic valve position: feasibility study in swine. J Am Coll Cardiol. 2004;44(11):2247–2249. 103. Kuehne T, Saeed M, Reddy G, et al. Sequential magnetic resonance monitoring of pulmonary flow with endovascular stents placed across the pulmonary valve in growing swine. Circulation. 2001;104(19): 2363–2368.

Cardiovascular Magnetic Resonance 407

29 CARDIOVASCULAR MAGNETIC RESONANCE OF SIMPLE CONGENITAL CARDIOVASCULAR DEFECTS

61. Wang ZJ, Reddy GP, Gotway MB, Yeh BM, Higgins CB. Cardiovascular shunts: MR imaging evaluation. Radiographics. 2003;23:S181–S194. 62. Goitein O, Fuhrman CR, Lacomis JM. Incidental finding on MDCT of patent ductus arteriosus: use of CT and MRI to assess clinical importance. AJR Am J Roentgenol. 2005;184(6):1924–1931. 63. Frydrychowicz A, Bley TA, Dittrich S, Hennig J, Langer M, Markl M. Visualization of vascular hemodynamics in a case of a large patent ductus arteriosus using flow sensitive 3D CMR at 3T. J Cardiovasc Magn Reson. 2007;9(3):585–587. 64. Rigatelli G, Zamboni A, Cardaioli P. Three-dimensional rotational digital angiography in a complicated case of patent ductus arteriosus transcatheter closure. Catheter Cardiovasc Interv. 2007;70(6):900–903. 64a. Frydrychowicz A, Bley TA, Dittrich S, et al. Visualization of vascular hemodynamics in a case of a large patent ductus arteriosus using flow sensitive 3D CMR at 3T. J Cardiovasc Magn Reson. 2007;9:585–587. 65. Rocchini AP. Coarctation of the aorta and interrupted aortica arch. In: Moller JH, Hofman JIE, eds. Pediatric Cardiovascular Medicine. 1st ed. 2000:567–593. 66. Varaprasathan GA, Araoz PA, Higgins CB, Reddy GP. Quantification of flow dynamics in congenital heart disease: applications of velocityencoded cine MR imaging. Radiographics. 2002;22(4):895–905. 67. Konen E, Merchant N, Provost Y, McLaughlin PR, Crossin J, Paul NS. Coarctation of the aorta before and after correction: the role of cardiovascular MRI. AJR Am J Roentgenol. 2004;182(5):1333–1339. 68. Haramati LB, Glickstein JS, Issenberg HJ, Haramati N, Crooke GA. MR imaging and CT of vascular anomalies and connections in patients with congenital heart disease: significance in surgical planning. Radiographics. 2002;22(2):337–347. 69. von Schulthess GK, Higashino SM, Higgins SS, Didier D, Fisher MR, Higgins CB. Coarctation of the aorta: MR imaging. Radiology. 1986;158(2):469–474. 70. Simpson IA, Chung KJ, Glass RF, Sahn DJ, Sherman FS, Hesselink J. Cine magnetic resonance imaging for evaluation of anatomy and flow relations in infants and children with coarctation of the aorta. Circulation. 1988;78(1):142–148. 71. Mohiaddin RH, Kilner PJ, Rees S, Longmore DB. Magnetic resonance volume flow and jet velocity mapping in aortic coarctation. J Am Coll Cardiol. 1993;22(5):1515–1521. 72. Oshinski JN, Parks WJ, Markou CP, et al. Improved measurement of pressure gradients in aortic coarctation by magnetic resonance imaging. J Am Coll Cardiol. 1996;28(7):1818–1826. 73. Steffens JC, Bourne MW, Sakuma H, O’Sullivan M, Higgins CB. Quantification of collateral blood flow in coarctation of the aorta by velocity encoded cine magnetic resonance imaging. Circulation. 1994;90(2):937–943. 74. Holmqvist C, Stahlberg F, Hanseus K, et al. Collateral flow in coarctation of the aorta with magnetic resonance velocity mapping: correlation to morphological imaging of collateral vessels. J Magn Reson Imaging. 2002;15(1):39–46. 75. Nielsen JC, Powell AJ, Gauvreau K, Marcus EN, Prakash A, Geva T. Magnetic resonance imaging predictors of coarctation severity. Circulation. 2005;111(5):622–628. 76. Cowley CG, Orsmond GS, Feola P, McQuillan L, Shaddy RE. Longterm, randomized comparison of balloon angioplasty and surgery for native coarctation of the aorta in childhood. Circulation. 2005;111(25):3453–3456. 77. Saeed M, Henk CB, Weber O, et al. Delivery and assessment of endovascular stents to repair aortic coarctation using MR and X-ray imaging. J Magn Reson Imaging. 2006;24(2):371–378. 78. Didier D, Saint-Martin C, Lapierre C, et al. Coarctation of the aorta: pre and postoperative evaluation with MRI and MR angiography; correlation with echocardiography and surgery. Int J Cardiovasc Imaging. 2006;22(3–4):457–475. 79. Schmidt M, Theissen P, Klempt G, et al. Long-term follow-up of 82 patients with chronic disease of the thoracic aorta using spin-echo and cine gradient magnetic resonance imaging. Magn Reson Imaging. 2000;18(7):795–806. 80. Bogaert J, Kuzo R, Dymarkowski S, et al. Follow-up of patients with previous treatment for coarctation of the thoracic aorta: comparison between contrast-enhanced MR angiography and fast spin-echo MR imaging. Eur Radiol. 2000;10(12):1847–1854. 81. Hager A, Kaemmerer H, Leppert A, et al. Follow-up of adults with coarctation of the aorta: comparison of helical CT and MRI, and impact on assessing diameter changes. Chest. 2004;126(4): 1169–1176.

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

CHAPTER 30

Cardiovascular Magnetic Resonance in Complex Congenital Heart Disease Jens Bremerich, Rolf Wyttenbach, Peter T. Buser, and Charles B. Higgins

Cardiovascular magnetic resonance (CMR) is an attractive tool for noninvasive imaging of congenital heart disease (CHD) because it provides comprehensive cardiac morphologic and functional data without ionizing radiation, a particularly valuable attribute when considering imaging in children and young adults.1 The number of patients with CHD is continuously increasing as a result of extended survival produced by palliative and corrective surgical procedures. The role of CMR in the assessment of CHD is primarily defined by the use and limitations of two dimensional transthoracic echocardiography (TTE), which has become the accepted standard for evaluation of most forms of CHD. Advantages of CMR compared with TTE include the following: (1) The entire thorax can be imaged with sequential high resolution tomograms; (2) the information represents a continuous three dimensional (3D) dataset that can be obtained in any plane; (3) the technique pro vides superior depiction of the central pulmonary arteries,2 including the systemic and pulmonary veins3 as well as the aorta.4 The major requirement for evaluation of CHD is the pre cise depiction of cardiovascular anatomy. Electrocardio graphically (ECG) gated spin echo CMR can be used with high diagnostic accuracy for assessment of the morphologic features in simple and complex CHD.5 Cine gradient echo CMR permits accurate measurement of functional parameters such as left ventricular (LV) and right ventricular (RV) stroke volume, ejection fraction, regional wall motion, and wall thickening. Real time imaging methods are becoming increasingly popular.6 Moreover, cine CMR can detect the jet flow associated with valvular lesions and intracardiac shunts. Operator independent isotropic 3D cine sequences are now available for assessment of the morphologic features and function in any plane reconstructed after the scan.7 With velocity encoded CMR, measurement of flow velocity and flow vol ume is possible, allowing quantification of pulmonary artery flow, shunt lesions, valvular regurgitation, and ven tricular filling. These CMR techniques are ideally suited for sequential follow up studies of surgically treated patients.8 Contrast enhanced 3D magnetic resonance angiography (MRA) provides precise depiction of vessels such as pulmo nary artery stenoses in tetralogy of Fallot9 and extracardiac thoracic vessels.10 Reconstruction of targeted maximum 408 Cardiovascular Magnetic Resonance

intensity projections (MIP) and multiplanar images (MPR) in any plane allows for precise assessment of the mor phologic features of stenosis and measurement of the cross sectional area. In aortic coarctation, the extent of col lateral circulation through the intercostal and mammarian arteries provides an estimate of the hemodynamic relevance of the stenosis. Vascular function and abnormal connec tions are evaluated with contrast enhanced dynamic time resolved MRA.11 Tissue characterization is obtained through different sequence weighting and preparation pulses. Late gadolinium enhancement is visible in scar tissue or inflammation 20 minutes after injection of contrast agent. Inversion recovery sequences are applied to minimize signal from normal myo cardium and to optimize contrast. After surgical repair of tetralogy of Fallot, dilation and scar formation in the RV outflow tract adversely affect RV hemodynamics.12,13 This chapter discusses the CMR appearance of cardiovas cular morphology with the use of a segmental approach that represents the most rational way to evaluate complex CHD. The reader is also referred to consensus documents regard ing the use of CMR in CHD.14 The segmental approach is based on morphologic identification of the great arteries, atria, and ventricles, and the visceroatrial relationship as well as the type of connection among these structures. There is also a brief review of the morphologic and functional evalu ation of complex CHD with CMR pre and postoperatively.

ATRIAL MORPHOLOGY AND DETERMINATION OF SITUS Axial spin echo CMR images extending from the base of the heart to the dome of the liver show the segmental cardio vascular anatomy. A segmental approach is based on locali zation of the three cardiac segments (atria, ventricles, and great arteries), the type of atrioventricular (AV) and ventri culoarterial (VA) connections, and the detection of asso ciated anomalies (e.g., shunts, valve atresia).15,16 This segmental approach to CHD provides a precise description of the cardiac morphologic features, allowing accurate diag nosis of CHD with CMR.17 Atrial situs solitus is the normal situation in which the morphologic right (systemic venous) atrium is positioned

VENTRICULAR MORPHOLOGY AND ISOMERISM Transverse images at the midventricular level permit defini tion of the ventricular loop. The morphologic left ventricle (LV) and morphologic right ventricle (RV) have different anatomic characteristics that allow their differentiation. The morphologic RV can be recognized by its anterior and right sided location (D ventricular loop). The RV has prominent septomarginal trabecula passing from the apical portion of the ventricular septum to the anterior wall of the morphologic RV. The septal leaflet of the tricuspid valve is attached more anteriorly toward the cardiac apex than is the septal leaflet of the mitral valve. The depiction of differ ent levels of insertion of the septal leaflets is an important diagnostic feature for distinction of the ventricles.19 Proba bly the most reliable sign for defining the RV is the pres ence of an infundibulum, or conus, that separates the tricuspid and pulmonary valves. This characteristic feature of the RV is best appreciated on axial images. The morpho logic LV is located posteriorly and to the left. The septal leaflet of the mitral valve is located more distant from the cardiac apex compared with the tricuspid valve. The ana tomic LV has a smoothly contoured apical portion of the ventricular septum and is characterized by a lack of com plete muscular infundibulum. Therefore, the morphologic LV is characterized by a direct (fibrous) continuity between the mitral and aortic valves. The relationship of the great vessels can be determined by axial sections through the base of the heart. The ascend ing aorta can be identified by its continuity with the aortic arch and the brachiocephalic arteries. The main pulmonary artery is characterized by its bifurcation into the pulmonary arteries. Normally, the aorta lies posterior and to the right of the pulmonary trunk, and both vessels are similar in size.

The term isomerism means that both atria have features of the RA or the LA. In general, both atria develop with the same side as the thoracic and abdominal viscera (vis ceral atrial rule). Therefore, bilateral left pulmonary artery anatomy with the artery passing over the left bronchus indicates left sided isomerism (bilateral left sidedness). These findings are best shown on coronal sections. Con versely, bilateral right pulmonary anatomy indicates right sided isomerism (bilateral right sidedness). The former is usually associated with polysplenia and the latter with asplenia. A previous report showed that gated spin echo CMR is highly accurate for determination of relationships among the great arteries, visceroatrial situs, and type of ventricular loop in patients with CHD.5 In another study, surgical planning was altered in some patients with het erotaxy syndrome, based on additional data supplied by CMR and not available by TTE or invasive cardiac catheterization.20

ABNORMALITIES OF THE ATRIOVENTRICULAR CONNECTION Once the atrial and ventricular morphology is determined, the next step is to determine whether the AV connection is concordant or discordant. Concordance is defined by a con nection between the morphologic RA and the morphologic RV or between the morphologic LA and the morphologic LV, regardless of the positions of the atria and ventricles. Thus, discordance means that the morphologic RA drains to the morphologic LV and the morphologic LA drains to the morphologic RV, again regardless of the chamber posi tion within the chest. An example of AV (and VA) discor dance is corrected transposition (L transposition), which is described later. AV discordance in conjunction with VA concordance is a rare malformation called isolated ventricu lar inversion. Other abnormalities of AV connections in which the terms concordant and discordant are not appropri ate include the following: (1) double inlet AV connection in which both atria are connected to a single ventricular chamber; (2) straddling AV valves in which one of the valves overlies the septum and drains blood into both ven tricles; and (3) AV valve atresia in which one of the valves does not form and the AV ring may be replaced by fat (Fig. 30 1).21

ABNORMALITIES OF VENTRICULOARTERIAL CONNECTIONS Transposition Transposition of the great arteries (TGA) is one of the most common cyanotic forms of CHD. CMR has been shown to define the pathoanatomy of transposition and other abnormalities of VA connections accurately in several stud ies.5,17,19,22 In complete transposition, or D transposition Cardiovascular Magnetic Resonance 409

30 CARDIOVASCULAR MAGNETIC RESONANCE IN COMPLEX CONGENITAL HEART DISEASE

on the right of the spine and the morphologic left (pulmo nary venous) atrium is positioned to the left of the spine. In atrial situs inversus, the mirror image of the normal situation occurs, with the anatomic right atrium (RA) on the left side and the anatomic left atrium (LA) on the right side. The morphologic RA has distinctive anatomic char acteristics, such as the broad based, triangular appendage, whereas the morphologic LA contains an appendage with a narrow ostium and a more tubular configuration.18 The RA can also be identified by its connection to the inferior vena cava. In virtually all individuals, the side of the infe rior vena cava defines the side of the RA. Furthermore, the situs of the atria can be defined by the visceral situs because they are nearly always concordant.16 Thus, the morphologic RA is determined by the side of the short main bronchus and the liver, whereas the long main bronchus, spleen, stomach, and aorta define the morpho logic LA. In situs solitus, the left pulmonary artery courses crani ally over the left main bronchus and the right pulmonary artery runs anterior and slightly inferior to the right bronchus.

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

A

Figure 30 1 Electrocardiogram (ECG) gated spin echo transverse CMR at the level of the cardiac base (A) demonstrates side by side relationship of the aorta (A) and the main pulmonary artery (P) in a patient with DORV. Note the complete muscular ring surrounding both great arteries. Image B shows a stenosis of the left pulmonary artery (straight arrow) and an occluded right pulmonary artery (curved arrow).

P

A

B

show the anterior aorta arising from the morphologic RV and the posterior pulmonary artery arising from the mor phologic LV.17 In patients with levo TGA, also known as congenitally corrected TGA, the developing embryonic cardiac tube bends initially toward the left, rather than to the right, called an L loop. In L transposition, both AV and VA con nections are discordant. This results in an aorta that is posi tioned anterior and leftward toward the pulmonary artery (Fig. 30 3). In addition, the aorta arises from the left sided morphologic right ventricle and the pulmonary artery from the right sided morphologic LV because the position of the ventricles is inverted (L loop) in L transposition. Therefore,

of the great arteries, VA discordance exists in the presence of AV concordance. The ventricles receive blood from the correct atrium, although the pulmonary artery is connected to the LV and the aorta is connected to the RV. CMR of the normal anatomy of the great artery at the base of the heart shows the aorta posterior and to the right of the pulmonary artery. Contrary to the normal anatomy, the transaxial images at the base of the heart in TGA show the anterior position of the aorta relative to the main pulmonary artery (Fig. 30 2). The aorta is located to the right of the pul monary artery in D transposition and to the left of the pulmonary artery in L transposition. In patients with trans position of the great vessels, sagittal images can clearly

Figure 30 2 Complex congenital heart disease shown on T1 weighted fast spin echo (A) and gradient recalled echo (B to D) images. D Transposition with aorta (Ao) arising from the right ventricle (RV), which is connected to the left ventricle (LV) through a large ventricle septum defect (arrow in B). There is atresia of the tricuspid valve (D). Blood from the right atrium (RA) is directed through a large atrial septal defect to the left atrium (LA). Moreover, this patient has a left superior vena cava (asterisk in C and D) draining into the coronary sinus. Ao, aorta.

Ao

RV LV

A

B

Ao

RV RA

*

C

410 Cardiovascular Magnetic Resonance

LV LA

*

D

Ao Ao LA RV RV

B

A

RVOT

Ao RA

LA

C

D

the severity of valvular and subvalvular stenosis as well as pulmonary artery stenosis or atresia can be determined with CMR.23 Reversal in muscle thickness and shape in the RV compared with the LV, which is characteristic of transposition, can also be shown by CMR. In this respect, CMR is now regarded as the most accurate method to quantify ventricular mass. Consequently, cine CMR can be used to determine LV mass in children or adults in whom an arterial switch procedure is under consideration (Table 30 1).

systemic venous blood is pumped to the lungs by the LV through the pulmonary arteries, and oxygenated blood is pumped to the systemic circulation by the RV through the aorta. These anatomic features can be assessed readily by transverse spin echo CMR. In the coronal plane, the left sided ascending aorta typically forms the upper heart border with L transposition. Furthermore, CMR has the capability to show anomalies associated with TGA. Axial cine CMR images can assess the presence and severity of atrial and ventricular septal defects (VSDs). Additionally,

Table 30-1 Surgical Procedures Used to Treat Congenital Heart Disease Name

Description

Indication

Jatene

Arterial switch (Ao , PA) Reanastomose coronaries Atrial switch Conduit RV ) PA Anastomosis/conduit RA ) PA Anastomosis ascending Ao ) right PA Anastomosis descending Ao ) left PA Shunt subclavian artery ) PA Anastomosis SVC ) PA

TGA

Mustard, Senning Rastelli Fontan Waterston Cooley Potts Blalock Taussig Glenn

TGA Pulmonary atresia, TGA with PA stenosis, infundibular stenosis RVOT Tricuspid stenosis/atresia, hypoplastic RV, solitary ventricle TOF TOF TOF Tricuspid atresia, solitary ventricle, hypoplastic RV, septum intact

Ao, aorta; PA, pulmonary artery; RA, right atrium; RV, right ventricle; RVOT, right ventricular outflow tract; SVC, superior vena cava; TGA, transposition of the great arteries; TOF, tetralogy of Fallot.

Cardiovascular Magnetic Resonance 411

30 CARDIOVASCULAR MAGNETIC RESONANCE IN COMPLEX CONGENITAL HEART DISEASE

Figure 30 3 Congenitally corrected transposition (L transposition) on steady state free precession (SSFP) images. Blood from the left atrium (LA) is directed through the right ventricle (RV) to the aorta (Ao) as shown in A and B. Blood from the right atrium (RA) is directed through the left ventricle (arrow in D) to the pulmonary trunc. Note the aorta (C) arises from the right ventricular outflow tract (RVOT), which is characterized by a complete muscular ring.

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

semilunar valves (uncommitted VSD). Double outlet RV may be associated with valvular or subvalvular pulmonic stenosis. Especially if this arises in combination with a subaortic VSD, differentiation from tetralogy of Fallot may be difficult clinically and angiographically. Axial CMR images provide direct visualization of this type of double outlet right ventricle by showing a complete circle of muscle separating the aortic from the mitral valve rather than indirectly gathering its presence from the distance between the aortic and mitral valve, as shown on left ventriculography.30 CMR can be used to determine the size and location of the VSD relative to the great arteries and to define subpulmonary or subaortic stenosis, the spatial relationships of the great vessels, and the status of the pulmonary arteries and the aortic arch (Fig. 30 4). These features are important for clinical and surgical management. Additionally, double outlet right ventricle may be associated with atresia of the right AV valve. Because this condition will affect surgical repair, it is crucial to identify both AV valves in the axial plane.31

Double-Outlet Right Ventricle Double outlet right ventricle is defined as an abnormal VA connection in which more than half of both the aorta and the pulmonary artery arise from the morphologic RV.24,25 On axial CMR images, double outlet RV is typically characterized by side by side positioning of the great arteries at the semilunar valve level, with the aorta to the right of the pulmonary artery, although this relationship may be variable.26,27 An additional important feature of this condition is that neither semilunar valve is in direct fibrous continuity with the mitral valve. There is a com plete rim of muscle separating both semilunar valves from the anterior mitral valve leaflet. On transverse CMR images, this side by side positioning of two muscular circles in the outflow region of the RV is diagnostic of double outlet RV (see Fig 30 1). Coronal images define the side by side rela tionship of the aorta and the pulmonary artery at the level of the semilunar valves and their origin from the RV. The LV is shown to be separated from the semilunar valves. The only outlet for the LV becomes the requisite VSD. To plan a surgical repair adequately, it is important to determine the relative relationship of the VSD to the great vessels.28,29 This can usually be assessed on axial images. The VSD may be localized near the aortic valve (subaortic VSD), the pulmonic valve (subpulmonic VSD; Bing Taus sig anomaly), or both semilunar valves (doubly com mitted VSD). It may also be found remote from both

Truncus Arteriosus Persistent truncus arteriosus results from failure of division of the embryonic truncus into a separate aorta and pulmo nary artery. This abnormality is infrequent, representing 0.4% of all cases of CHD.32 On axial, sagittal, and coronal

TA

RV LV

*

A

B

Ao W

Ao

C

412 Cardiovascular Magnetic Resonance

D

Figure 30 4 Pulmonary atresia with truncus arteriosus. Axial T1 weighted fast spin echo CMR (A) shows large ventricular septal defect between right (RV) and left (LV) ventricles. Axial gradient recalled echo image cranial to the ventricles (B) shows one large truncus arteriosus (TA) arising from both ventricles. No central pulmonary arteries can be identified. Palliation was done in childhood with a Waterston anastomosis (W in C) connecting the right pulmonary arteries to the ascending aorta (Ao). The Potts anastomosis connects the left pulmonary arteries to the descending aorta (arrow), but shows a stenosis. Contrast enhanced time resolved dynamic 2D MR angiography (D) shows perfusion of the right lung (arrowheads) but hypoperfusion of the left lung, confirming the hemodynamic relevance of the stenosis.

TETRALOGY OF FALLOT Consecutive transverse tomograms (spin echo; 5 to 7 mm slice thickness) through the entire heart and pulmonary hili show the following: (1) RV hypertrophy; (2) unequal divi sion of the outflow tracts, with an enlarged and anteriorly displaced aorta; (3) membranous VSD; and (4) multilevel narrowing of the infundibulum, pulmonary annulus, main, and central pulmonary arteries. The infundibulum and pul monary annulus are best shown on sagittal tomograms. The degree of pulmonary stenosis varies, and in extreme cases, the pulmonary trunk may not be identifiable. Severe pul monary stenosis and atresia are usually associated with numerous large collateral channels arising from the aorta, principally, the descending aorta, and proceeding to the pulmonary hili.35 These vessels may be seen on flow sensitive transverse gradient echo CMR at the level of the carina or on coronal images. GRE techniques provide a bright signal from flowing blood. Stenoses of the central pulmonary arteries are frequent in tetralogy of Fallot and not uncommonly remain after ini tial surgical correction. These stenoses are best shown on a set of very thin tomograms with 3 mm slice thickness acquired in a plane parallel to the long axis of the right and left pulmonary arteries. The image plane should be parallel to the long axis of the right or left pulmonary artery. Most adult patients with tetralogy of Fallot have already undergone one or more corrective surgical procedures. CMR is an ideal technique for monitoring these patients after surgery. Cine CMR is used to monitor RV volume, mass, and ejection fraction.36 Velocity encoded cine CMR has been used to monitor pulmonary regurgitation, which occurs in most patients after total correction of the anomaly. One study has shown that the severity of pulmonary regurgitation, as quantified by velocity

encoded CMR, correlates with RV volume, mass, and ejec tion fraction.37 Failure of the RV can occur in tetralogy of Fallot. RV mass or functional parameters, such as ejection fraction and stroke volume, can be readily calculated from a stack of consecutive cine gradient echo CMR images acquired in the short axis of the heart and covering the entire RV. RV mass calculated from such CMR images has been shown to correlate with the width of the QRS complex. A wide QRS complex is a harbinger of RV arrhythmias.

EBSTEIN ANOMALY OF THE TRICUSPID VALVE Ebstein anomaly is an uncommon congenital developmen tal abnormality of the tricuspid valve that has a wide spec trum of pathologic anatomy. Although the diagnosis of Ebstein anomaly is usually straightforward and is based on transthoracis echocardiography findings, CMR may be helpful in defining the pathologic anatomy and tailoring the surgical method for each patient (Figs. 30 5 and 30 6). Axial CMR images are most informative. The tricus pid valve is almost always dysplastic in addition to being abnormally inserted. In fact, the dominant feature is dys plasia of the valve rather than displacement.38 The anterior leaflet usually attaches normally to the AV junction and is enlarged,39 and these features can be shown on CMR images in most cases.40 Distally, the anterior leaflet may be attached to an abnormal anterolateral papillary muscle. Therefore, it may be mobile, may have a continuous mus cular connection with restricted mobility, or may be broadly plastered to the anterior wall of the RV and there fore would not be distinguishable on CMR.41,42 Septal and posterior leaflets are displaced downward in Ebstein anomaly,42 and they are best assessed on axial and coronal images, respectively.43 Both leaflets, however, may be defi cient or absent in Ebstein anomaly and therefore would not be detectable on CMR.42,44 Planning of surgical repair must include assessment of the morphologic and functional features of the atrialized and hemodynamically effective portion of the RV. This assessment is best achieved with coronal T1 weighted spin echo and cine CMR. Reconstruction of the tricuspid valve with a prosthetic ring and vertical plication of the RA and the AV annulus may be complemented by an additional plication of the atrialized ventricle, depending on the size and function of the atrialized ventricle.44 Moreover, cine CMR enables assessment of tricus pid regurgitation, tricuspid stenosis, and shunts through the atrial septal defect (see Fig. 30 5).

COMPLEX VENTRICULAR ABNORMALITIES (SINGLE VENTRICLE) Compared with invasive angiography, CMR has many advantages for the definition of segmental anatomy and other defects in patients with a suspected diagnosis of sin gle ventricle.45 The specific goals of CMR imaging in com plex ventricular abnormalities are: (1) determination of Cardiovascular Magnetic Resonance 413

30 CARDIOVASCULAR MAGNETIC RESONANCE IN COMPLEX CONGENITAL HEART DISEASE

images, a single large vessel can be seen arising from the expected position of the semilunar valves just above the VSD. The truncus arteriosus gives rise to the aorta and the pulmonary and coronary arteries. Based on the configuration of the pulmonary arteries, three different types of truncus arteriosus are recognized. The most common, type I, is char acterized by a short main pulmonary artery arising from the truncus. Bidirectional shunting occurs at the truncus, result ing in early cyanosis and eventual congestive heart failure. Axial T1 weighted CMR images can define the anatomy of the truncus and the pulmonary arteries.4 Sagittal and coro nal planes are useful for demonstrating the origin of the pul monary arteries from the truncus. For precise assessment of pulmonary artery caliber, a reduction of slice thickness to 3 mm is needed. Furthermore, CMR can show ventricular size and wall thickness as well as associated abnormalities, such as VSD, right aortic arch, or interrupted aortic arch. Axial and sagittal CMR tomograms are most effective for postoperative evaluation of the caliber of the conduit between the right ventricle and the pulmonary artery (Ras telli procedure). Narrowing of the conduit, stenosis at the origin of the right or left pulmonary artery, and pseudoa neurysm are complications that can be detected readily by CMR.33,34 Truncal insufficiency, which is common, may be detected by the use of cine CMR.

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

*

A

B

RV RA LV RV

C

D

Ao

PT

Figure 30 5 Large atrial septal defect. Conventional X rays show enlarged pulmonary (arrowhead) in PA (A) and enlarged right ventricle and right pulmonary artery in lateral view (B). Ebstein anomaly shown on axial T1 weighted fast spin echo (A, B) and gradient echo (C, D) images. Axial gradient recalled echo CMR (C) shows large atrial septal defect (arrow in C) and enlarged right atrium (RA) and ventricle (RV), also shown on short axis view (D). Velocity encoded magnitude (E) and phase (F) image show enlarged pulmonary trunc (PT). Through plane flow measurement in the ascending aorta (circle) and pulmonary trunc yield a left to right shunt of 1:5.

visceral situs; (2) assessment of the type of ventricular loop, the morphologic features of the predominant ventricle (right, left, or primitive), and the position of the rudimen tary ventricle; (3) definition of the AV and VA connections; (4) determination of the size of interventricular communi cation; and (5) definition of the connections of the systemic and pulmonary veins and arteries. Transverse gated spin echo tomograms with 7 mm slice thickness are, in general, the most useful for evaluation of complex cardiac anomalies.46,47 AV connections can be identified as double inlet, absent left AV, or absent right AV connections. Stenoses or regurgitation can be detected on 414 Cardiovascular Magnetic Resonance

cine bright blood gradient echo CMR as flow void caused by spin dephasing in turbulent blood flow. After identifica tion of the AV connection, the ventricular morphology must be determined. Distinction between dominant LV or domi nant RV is usually possible by transverse and coronal CMR. When there is no detectable muscle separating either AV valve from the adjacent semilunar valve, the chamber is considered an LV. The position of a rudimentary RV is usu ally anterior and superior to the dominant ventricle, whereas the rudimentary LV is usually posterior and inferior to the dominant ventricle. A dominant LV is most common in adult patients. A primitive type of single ventricle has morphologic

* ** RA

A

B

*

C

features characteristic of neither the RV nor the LV. The com munication between the dominant and the rudimentary ven tricle can be assessed on both T1 weighted spin echo and cine gradient echo CMR.

POSTOPERATIVE EVALUATION Many patients with complex CHD have survived to adult hood after various palliative and corrective procedures (see Table 30 1). These patients require frequent monitoring at regular intervals with imaging studies. Transthoracic echocar diography is usually the initial technique used for this pur pose, but it is not as ideal in adults as in children and is less quantitative. Moreover, many surgical procedures involve supracardiac as well as cardiac structures. Because the supra cardiac structures are sometimes not well depicted by echo cardiography, CMR is increasingly recognized as more effective for postoperative monitoring of older children and adults after complex surgical procedures.48 Several reports confirm the effectiveness of CMR for postoperative evaluation of complex CHD, suggesting that this noninvasive technique may obviate the serial use of invasive studies for postoperative follow up in many cases.34,49,50 CMR not only is capable of visualizing cardiac

D

and extracardiac morphology, but also can quantify blood flow in the pulmonary arteries and conduits. Compared with echocardiography, CMR has the advantage of superior demonstration of conduits and anastomoses at the level of the great arteries. Furthermore, it is unaffected by postsur gical changes or graft material that can make echocardiog raphy difficult.51 In addition, CMR has been found to be effective for monitoring pulmonary arterial status postoper atively and to be superior to echocardiography for evalua tion of the pulmonary arteries.52,53 Many adult patients with TGA were treated with Mustard and Senning procedures (see Fig. 30 2). This atrial switch procedure is accomplished with a complex baffle con structed in the atria to direct blood flow from the pulmonary veins across the tricuspid valve into the RV and then to the aorta. Blood from the superior vena cava and the inferior vena cava flows through the baffle and then across the mitral valve into the pulmonary artery. On postoperative follow up after a Mustard procedure, CMR in the coronal plane is particularly useful for evaluation of the superior systemic venous channel, which is the most common site of systemic venous obstruction. The sagittal and transaxial planes can show obstruction of pulmonary venous return or narrowing of the connection between the dorsal and ventral parts of the pulmonary venous atrium. In both procedures, the RV continues to work against systemic load. This may eventually result in RV systolic Cardiovascular Magnetic Resonance 415

30 CARDIOVASCULAR MAGNETIC RESONANCE IN COMPLEX CONGENITAL HEART DISEASE

Figure 30 6 Ebstein anomaly shown on axial T1 weighted fast spin echo (A and B) and gradient echo (C and D) images. Anterior displacement of the septal leaflet (arrow in A) of the tricuspid valve apically of the normal tricuspid plane (dashed line) results in the following changes: The functional right atrium comprises the morphologic right atrium (RA) plus the atrialised right ventricle (asterisks). The functional right ventricle is small (asterisk), which explains why such patients frequently suffer from right heart failure. Apical displacement of the septal leaflet of the tricuspid valve also leads to tricuspid insufficiency shown by a flow void on gradient echo images (arrow in D).

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

dysfunction and tricuspid insufficiency because the RV and the tricuspid valve are not structured for systemic pressure load.54 Consequently, cine CMR can be used to monitor RV volume, mass, and ejection fraction.55 This method is also used to detect and estimate the severity of tricuspid regurgitation. Function of the systemic ventricle may also be monitored by through plane velocity encoded cine measurements perpendicular to the ascending aorta.56 Currently, the favored procedure for TGA is the Jatene, or arterial switch, procedure, in which the aorta and the pulmonary artery are transected above the sinus portion and switched to redirect blood flow. The coronary arteries are then transplanted onto the neoaorta. This is usually done in the neonatal period. However, many older children or adults previously treated with the Senning or Mustard procedure are now candidates for the arterial switch proce dure to avert or relieve RV pressure overload or failure. CMR has a major role in the long term follow up of patients after surgical repair of TGA.50,57 CMR can assess various complications after an atrial switch proce dure for d TGA, including baffle leaks, systemic or pul monary venous obstructions, LV outflow tract stenosis, tricuspid regurgitation, and coronary anatomy.58 Transverse and sagittal spin echo CMR is used for assess ment of the great vessel anatomy after a Jatene procedure. Because this procedure is frequently performed in the neo natal period, thin sections of 3 mm thickness are preferable. Postoperatively, the position of the aorta, posterior to the main pulmonary artery, occasionally results in proximal ste nosis of the right or left pulmonary artery. This stenosis can be clearly seen on axial CMR images. In addition, CMR can visualize other complications after a Jatene procedure, such as narrowing of the RV outflow region, dilation of the aortic root, and supravalvular aortic stenosis.34,57 Compared with transthoracic echocardiography, CMR has been shown to be virtually equal in depicting stenoses of the RV outflow region. However, compared with echocardiography, CMR was superior in the detection of proximal pulmonary artery stenosis in patients who underwent the Jatene procedure (41% vs. 94%).59 Furthermore, CMR is unaffected by post surgical changes, graft material, or an inadequate acoustic window, which may render echocardiography difficult. The Fontan procedure consists of an anastomosis between the RA or right atrial appendage and the main pul monary artery.60 Numerous variations of the Fontan proce dure have been devised. Many patients initially have a bidirectional Glenn shunt (SVC to right pulmonary artery anastomosis) followed months or years later by placement of a conduit from the inferior vena cava to the right or left pulmonary artery. Therefore, systemic venous blood is for warded directly to the pulmonary circulation, bypassing the functional single ventricle, which is functioning as a systemic pumping chamber. The concomitant ASD is also closed. The major indications for Fontan procedure are tri cuspid atresia or severe stenosis, single ventricle, and hypo plastic left heart. CMR can be used to show the size of the atriopulmonary connection, assess the distribution of pul monary perfusion, and recognize the presence of obstruc tion.61,62 Axial and coronal images are usually effective for this purpose. Besides obstruction of the conduit, com plications of the Fontan procedure include residual ASD, systemic venous hypertension, and thrombosis.63 The for mer can be diagnosed with cine CMR, and the latter may 416 Cardiovascular Magnetic Resonance

result in right atrial enlargement, venous stasis, pleural and peritoneal effusion, and edema. Severe right atrial enlargement may even compress the right pulmonary veins at the entrance to the LA.34 Determination of the size of the pulmonary arteries is important in patients undergoing Fontan reconstruction because artery size is considered a major indicator of prognosis. CMR has been shown to be use ful in determining pulmonary artery size in these patients and has been found to be superior to echocardiograpy.54 Surgical repair of tetralogy of Fallot is achieved by clos ing the VSD and enlarging the pulmonary outflow tract with patches. Axial CMR scans can show abnormalities in the RV outflow tract, including residual stenosis or aneurys mal patch dilation. In the case of concomitant pulmonary artery atresia, surgical repair is more complex, necessitating systemic to pulmonary shunts to allow blood flow to the lungs and to promote the growth of pulmonary vessels. The subclavian to pulmonary artery shunt (Blalock Taus sig), which was used in the past, has been largely replaced by the modified Blalock shunt, representing a graft con necting the aorta or brachiocephalic artery with the pulmo nary artery. An earlier report showed the usefulness of electrocardiographically gated CMR to assess the size, course, and patency of Blalock Taussig, Glenn, and aorto pulmonary shunts.50 Coronal and axial CMR imaging planes are particularly useful for demonstrating systemic to pulmonary shunts as well as potential complications, including stenosis or occlusion of the shunt. Although long term results after surgical repair of tetral ogy of Fallot are good, most patients have some degree of pulmonary regurgitation. Long standing pulmonary regur gitation may result in severe arrhythmia and sudden death. Moreover, exercise capacity is often diminished in patients with severe dilation of the RV. Recent studies showed a 30% decrease in RV dilation and improved systolic func tion after valve replacement in adults who had undergone repair of tetralogy of Fallot.64 These authors suggest the use of axial cine CMR to evaluate RV function and double oblique velocity encoded CMR perpendicular to the vessel to assess flow measurement. Through plane flow is encoded up to 200 cm/sec. Recurrence of pulmonary regurgitation after surgery seems to predict reduced recov ery of RV systolic function.65 Repair of pulmonary atresia may also be accomplished by placing a valve conduit between the RV and the main pulmonary artery (Rastelli conduit). Sagittal CMR images are the most effective way to visualize the proximal anasto mosis to the RV and the distal anastomosis to the pulmo nary artery. Possible complications include false aneurisms at the anastomosis and stenosis or kinking of the conduit. Whereas spin echo CMR is used for morphologic evalu ation of postoperative CHD, CMR techniques, such as cine steady state free precession CMR and velocity encoded cine CMR, allow functional assessment of surgical baffles, con duits, and valvular function. In patients who had under gone a Mustard or Senning procedure for TGA, cine CMR was able to evaluate pulmonary and systemic venous con nections as well as RV systolic function and tricuspid and mitral regurgitation.66–69 Flow quantification with velocity encoded cine CMR improved the evaluation of venoatrial connections after a Mustard or Senning procedure.70,71 In addition, CMR velocity mapping has been used successfully to assess tricuspid volume flow in patients after Mustard or

EVALUATION OF FUNCTION IN CONGENITAL HEART DISEASE Cine gradient echo CMR and velocity encoded cine CMR are attractive methods for quantifying ventricular function and volumetric flow, respectively, in patients with untreated or repaired CHD. Standard or breath hold (segmented k space) cine CMR provides sequential images through the cardiac cycle, and the images can be viewed as a cine loop. Typically, cine CMR uses short repetition times (20 to 35 msec), short echo times (4 to 20 msec), and low flip angles (35 to 60 ) to acquire 16 phases spaced evenly throughout the cardiac cycle.75 Cine CMR images show high signal intensity in areas of normal blood flow. How ever, turbulent flow, which may occur in stenosis, regurgi tation, or shunt lesions, causes a signal loss within the blood pool, rendering these lesions readily visible on cine CMR. In addition, LV and RV mass and systolic function can be measured precisely with 3D cine CMR images (see Fig. 30 4). Unlike transthoracic echocardiography or inva sive left ventriculography, CMR does not rely on geometric assumptions or calculations based on partial sampling of the cardiac volume. This 3D dataset from end diastolic and end systolic tomograms encompassing both ventricles allows measurement of LV and RV volume, mass, stroke volume, and ejection fraction. In contrast to typical adult cardiologic studies, the RV is often of particular interest in CHD. Double oblique short axis tomograms are used to quantify biventricular volume and global function, with inclusion of trabeculations and papillary muscles in the RV cavity suggested for enhanced reproducibility.69 End diastolic and end systolic measure ments acquired through the entire stack of images provide end diastolic (EDV) and end systolic volume (ESV), stroke volume (SV ¼ EDV  ESV), and ejection fraction (EF ¼ SV/EDV) for both the LV and the RV. In normal individuals, LV stroke volume and RV stroke volume are the same.76 Therefore, differences in ventricular stroke volume can be used to quantify valvular regurgitation and shunt lesions.77 For example, in pulmonary or tricuspid regurgitation, the difference between RV stroke volume and LV stroke volume corresponds to regurgitant volume. LV stroke volume is greater than RV stroke volume in patients with patent ductus arteriosus, aortic regurgitation, and mitral regurgitation. Measurements of volume and function of the ventricles have been shown to be highly reproducible on sequential studies in patients with morphologically normal and abnormal LV.78 Thus, cine CMR is a highly attractive method for detecting changes in ventricular volume and function over time.

Velocity encoded cine CMR provides direct measure ment of aortic and pulmonary artery flow and therefore measures the effective stroke volume of both ventricles. In the absence of valvular regurgitation, this method can be used to determine intracardiac shunt volume. For example, in ASD, partial anomalous pulmonary venous connection, or VSD, the difference between RV stroke volume and LV stroke volume is the left to right shunt volume. Likewise, velocity encoded CMR can be used to quantify shunt vol ume in patent ductus arteriosus by calculating the differ ence between net forward flow in the ascending aorta and that in the main pulmonary artery. Velocity encoded cine CMR has been used successfully in patients with various congenital shunt lesions.79 The accuracy and reproducibil ity of this method for measuring Qp:Qs in left to right shunts have been assessed previously.80,81 Velocity encoded CMR can measure blood flow in the main pulmonary artery as well as separately in the left and right pul monary arteries. This is one of the few techniques with the capability to quantify left and right pulmonary flow sepa rately.81,82 This method can be used to determine the percent age of blood flow to each lung and assess the hemodynamic significance of stenoses of the left or right pulmonary artery. Tricuspid regurgitation and shunts may be detected and quantified during the same examination.83 In patients who have undergone a Mustard or Senning procedure, abnor mal tricuspid flow patterns have been found to precede RV systolic dysfunction.84 Velocity encoded cine CMR can also be used to deter mine the severity of coarctation by quantifying collateral flow in the descending aorta. Measurements of volume flow are performed in the proximal descending aorta just below the site of the coarctation and in the distal descending aorta near the diaphragm. In healthy subjects, flow volume decreases steadily from the proximal to the more distal parts of the descending aorta because of antegrade flow into the intercostal arteries. In patients with hemodynami cally significant coarctation of the aorta, the normal flow pattern in branches of the descending aorta is reversed, resulting in an increase in flow from the proximal to the distal descending aorta because of retrograde collateral flow mainly through intercostal arteries.85 Velocity encoded CMR can also be used for identification of patients with a mismatch between the severity of anatomic obstruction and collateral flow, which may be of importance for planning surgery.86,87 Velocity encoded cine CMR has been used successfully to demonstrate abnormalities in volume flow in the descending aorta in adult patients after repair of coarctation. These abnormalities are probably related to resistance to flow imposed by the coarctation segment and may represent an additional index for monitoring the hemodynamic significance of coarctation of recoarctation in patients before and after intervention.87,88 Thus, cine CMR allows for a comprehensive evaluation of aortic coarc tation by determining the location and severity of stenosis and pressure gradients across the coarctation segment.85–88

CONCLUSION Because of the progress in surgery in recent decades, an increasing number of patients with complex CHD survive to adulthood after having undergone previous palliative Cardiovascular Magnetic Resonance 417

30 CARDIOVASCULAR MAGNETIC RESONANCE IN COMPLEX CONGENITAL HEART DISEASE

Senning procedure, and it often showed abnormal tricus pid flow patterns.72 Velocity encoded cine CMR also pro vides accurate information about pulmonary flow volume and velocity after Fontan procedure. Velocity encoded cine CMR can be used to assess the volume of retrograde flow in patients after a Fontan procedure.73 Velocity encoded CMR has also been used to quantify the volume of pulmonary regurgitation after patch repair of tetralogy of Fallot74 and to estimate pressure gradients across ventriculopulmonary (Rastelli) conduits.

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

or corrective surgery. Other patients with milder form of CHD may become symptomatic only as adults. Patients with CHD often require numerous follow up studies for evaluation of morphology and function. CMR imaging is a noninvasive method that provides excellent anatomic detail of cardiac structures and the great vessels, even when echo cardiography access to the area of interest is limited. With cine CMR, it is possible to quantify valvular and shunt lesions and to measure ventricular volume without making any geometric assumptions. Velocity encoded CMR can be

used to measure flow velocity and volume in the heart and the great arteries. CMR is particularly useful for post operative follow up in patients with repaired CHD, espe cially after placement of intra atrial baffles or extracardiac conduits. Furthermore, CMR can assess RV function, which is of particular interest in many patients with CHD and which may be difficult for other imaging modalities. Thus, CMR is emerging as the noninvasive imaging technique of choice that allows comprehensive assessment of cardiovas cular anatomy and function in patients with complex CHD.

References 1. Higgins CB, Caputo GR. Role of MR imaging in acquired and congenital cardiovascular disease. AJR Am J Roentgenol. 1993;161:13. 2. Gomes AS, Lois JF, Williams RG. Pulmonary arteries: MR imaging in patients with congenital obstruction of the right ventricular outflow tract. Radiology. 1990;174:51. 3. Prasad SK, Soukias N, Hornung T, et al. Role of magnetic resonance angiography in the diagnosis of major aortopulmonary collateral arteries and patrial anomalous pulmonary venous drainage. Circulation. 2004;109:207–214. 4. Gomes AS, Lois JF, George B, et al. Congenital abnormalities of the aortic arch: MR imaging. Radiology. 1987;165:691. 5. Kersting-Sommerhoff BA, Diethelm L, Teitel DF, et al. Magnetic resonance imaging of congenital heart disease: sensitivity and specificity using receiver operating characteristic curve analysis. Am Heart J. 1989;118:155. 6. Muthurangu V, Lurz P, Critchely JD, Deanfield JE, Taylor AM, Hansen MS. Real-time assessment of right and left ventricular volumes and function in patients with congenital heart disease by using high spatiotemporal resolution radial k-t SENSE. Radiology. 2008;248: 782–791. 7. Srensen TS, Ko¨rperich H, Greil GF, et al. Operator-independent isotropic three-dimensional magnetic resonance imaging for morphology in congenital heart disease: a validation study. Circulation. 2004; 110:163–169. 8. Knauth Meadows A, Ordovas K, Higgins CB, Reddy GP. Magnetic resonance imaging in the adult with congenital heart disease. Semin Roentgenol. 2008;246–258. 9. Norton KI, Tong C, Glass RBJ, Nielsen JC. Cardiac MR imaging assessment following tetralogy of Fallot repair. Radiographics. 2006; 26:197–211. 10. Prakash A, Torres AJ, Printz BF, et al. Usefulness of magnetic resonance angiography in the evaluation of complex congenital heart disease in newborns and infants. Am J Cardiol. 2007;100:715–721. 11. Sonnet S, Buitrago-Tellez CH, Scheffler K, Strecker R, Bongartz G, Bremerich J. Dynamic time-resolved contrast-enhanced two-dimensional MR projection angiography of the pulmonary circulation: standard technique and clinical applications. AJR Am J Roentgenol. 2002;179:159–165. 12. Oosterhof T, Mulder BJ, Vliegen HW, de Roos A. Corrected tetralogy of fallot: delayed enhancement in right ventricular outflow tract. Radiology. 2005;237:868–871. 13. Harris MA, Johnson TR, Weinberg PM, Fogel MA. Delayed-enhancement cardiovascular magnetic resonance identifies fibrous tissue in children after surgery for congenital heart disease. J Thorac Cardiovasc Surg. 2007;133:676–681. 14. Warnes CA, Williams RG, Bashore TM, et al. ACC/AHA 2008 guidelines for the management of adults with congenital heart disease. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Develop Guidelines on the Management of Adults with Congenital Heart Disease). Circulation. 2008;118:e714–833. 15. Shinebourne EA, Macartney FJ, Anderson RH. Sequential chamber localization: logical approach to diagnosis in congenital heart disease. Br Heart J. 1976;41:327. 16. Van Praagh R. The importance of segmental situs in the diagnosis of congenital heart disease. Semin Roentgenol. 1985;20:254. 17. Didier D, Higgins CB, Fisher M, et al. Congenital heart disease: gated MR imaging in 72 patients. Radiology. 1986;158:227. 18. Anderson RH, Ho SY. The tomographic anatomy of the normal and malformed heart. In: Higgins CB, Silvermann NH, KerstingSommerhoff BA, Schmidt K, eds. Congenital Heart Disease: Echocardiography and Magnetic Resonance Imaging. New York: Raven Press; 1990:1.

418 Cardiovascular Magnetic Resonance

19. Guit GL, Bluemm R, Rohmer J, et al. Levotransposition of the aorta: identification of segmental cardiac anatomy using MR imaging. Radiology. 1986;161:376. 20. Geva T, Wesley Vick III G, Wendt RE, Rockey R. Role of spin echo and cine magnetic resonance imaging in presurgical planning of heterotaxy syndrome. Circulation. 1994;90:348. 21. Fletcher BD, Jacobstein MD, Abramowsky CR, et al. Right atrioventricular valve atresia: anatomic evaluation with MR imaging. Am J Roentgenol. 1987;148:671. 22. Higgins CB, Byrd III BF, Farmer DW, et al. Magnetic resonance imaging in patients with congenital heart disease. Circulation. 1984;70:851. 23. Kersting-Sommerhoff BA, Sechtem UP, Higgins CB. Evaluation of pulmonary artery supply by nuclear magnetic resonance imaging in patients with pulmonary atresia. J Am Coll Cardiol. 1989;11:166. 24. Lev M, Bharati S, Meng CCL, et al. A concept of double-outlet right ventricle. J Thorac Cardiovasc Surg. 1972;64:271. 25. Wilcox BR, Ho SY, Macartney FJ, et al. Surgical anatomy of doubleoutlet right ventricle with situs solitus and atrioventricular concordance. J Thorac Cardiovasc Surg. 1981;82:405. 26. Niezen RA, Beekman RP, Helbing WA, van der Wall EE, de Roos A. Double outlet right ventricle assessed with magnetic resonance imaging. Int J Card Imaging. 1999;15:323–329. 27. Mayo JR, Roberson D, Sommerhoff B, Higgins CB. MR imaging of double outlet right ventricle. J Comput Assist Tomogr. 1990;14:336. 28. Patrick DL, McGoon DC. An operation for double outlet right ventricle with transposition of the great arteries. J Cardiovasc Surg. 1968;9:537. 29. Sridaromont S, Feldt RH, Ritter DG, et al. Double outlet right ventricle: hemodynamic and anatomic correlations. Am J Cardiol. 1976;38:85. 30. Higgins CB. Congenital heart disease. In: Higgins CB, Hricak H, Helms CA, eds. Magnetic Resonance Imaging of the Body. 3rd ed. Philadelphia: Lippincott-Raven; 1996:461. 31. Smith JRWL, Stanford W, Skorton DJ, Wolf GL. Assessment of congenital heart disease by nuclear magnetic resonance imaging. In: Skorton DJ, ed. Marcus Cardiac Imaging: A Companion to Braunwald’s Heart Disease. 2nd ed., vol. 2. Philadelphia: W.B. Saunders Company; 1996:886. 32. Donnelly LF, Higgins CB. MR imaging of conotruncal abnormalities. AJR Am J Roentgenol. 1996;166:925. 33. Murashita T, Hatta E, Imamura M, Yasuda K. Giant pseudoaneurysm of the right ventricular outflow tract after repair of truncus arteriosus: evaluation by MR imaging and surgical approach. Eur J Cardiothorac Surg. 2002;22:849–851. 34. Kersting-Sommerhoff BA, Seelos KC, Hardy C, et al. Evaluation of surgical procedures for cyanotic congenital heart disease by using MR imaging. AJR Am J Roentgenol. 1990;155:259. 35. Berry BE, McGoon DC, Ritter DG, Davis GD. Absence of anatomic origin from heart of pulmonary arterial supply: clinical application of classification. J Thorac Cardiovasc Surg. 1974;68:119. 36. van Straten A, Vliegen HW, Hazekamp MG, de Roos A. Right ventricular function late after total repair of tetralogy of Fallot. Eur Radiol. 2005;15:702–707. 37. Niezen RA, Helbing WA, van der Wall EE, et al. Biventricular systolic function and mass studied with MR imaging in children with pulmonary regurgitation after repair for tetralogy of Fallot. Radiology. 1996;201:135. 38. Becker AE, Becker MJ, Edwards JE. Pathologic spectrum of dysplasia of the tricuspid valve: features in common with Ebstein’s malformation. Arch Pathol. 1971;91:167. 39. Anderson KR, Lie JT. Pathologic anatomy of Ebstein’s anomaly of the heart revisited. Am J Cardiol. 1978;41:739.

66. Fontan F, Deville C, Quaegebeur J, et al. Repair of tricuspid atresia in 100 patients. J Thoracic Cardiovasc Surg. 1983;85:647. 67. Rees S, Sommerville J, Warnes C, et al. Comparison of magnetic resonance imaging with echocardiography and radionuclide angiography in assessing cardiac function and anatomy following Mustard’s operation for transposition of the great arteries. Am J Cardiol. 1988;61: 1316. 68. Chung KJ, Simpson IA, Glass RF, et al. Cine magnetic resonance imaging after surgical repair in patients with transposition of the great arteries. Circulation. 1988;77:104. 69. Winter MM, Bernink JP, Groenink M, et al. Evaluating the systemic right ventricle by CMR: the importance of consistent and reproducible delineation of the cavity. J Cardiovasc Magn Reson. 2008;10:40. 70. Varaprasathan GA, Araoz PA, Higgins CB, Reddy GP. Quantification of flow dynamics in congenital heart disease: applications of velocityencoded cine MR imaging. Radiographics. 2002;22:895–905. 71. Sampson C, Kilner PJ, Hirsch R, et al. Venoatrial pathways after the Mustard operation for transposition of the great arteries: anatomic and functional MR imaging. Radiology. 1994;193:211. 72. Rebergen SA, Helbing WA, van der Wall EE, et al. MR velocity mapping of tricuspid flow in healthy children and in patients who have undergone Mustard or Senning repair. Radiology. 1995;194:505. 73. Rebergen SA, Ottenkamp J, Doornbos J, et al. Postoperative pulmonary flow dynamics after Fontan surgery assessment with nuclear magnetic resonance velocity mapping. J Am Coll Cardiol. 1993;21:123. 74. Rebergen SA, Chin JGJ, Ottenkamp J, et al. Pulmonary regurgitation in the late postoperative follow-up of tetralogy of Fallot: volumetric quantitation by nuclear magnetic resonance velocity mapping. Circulation. 1993;88:2257. 75. Sechtem U, Pflugfelder PW, White RD, et al. Cine MR imaging: potential for the evaluation of cardiovascular function. AJR Am J Roentgenol. 1987;148:239–246. 76. Wang ZJ, Reddy GP, Gotway MB, Yeh BM, Higgins CB. Cardiovascular shunts: MR imaging evaluation. Radiographics. 2003;23:S181–S194. 77. Sechtem U, Pflugfelder PW, Gould RG, et al. Measurement of right and left ventricular volumes in healthy individuals with cine MR imaging. Radiology. 1987;163:697. 78. Semelka RC, Tomei E, Wagner S, et al. Interstudy reproducibility of dimensional and functional measurements between cine magnetic resonance studies in the morphologically abnormal left ventricle. Am Heart J. 1990;119:1376. 79. Rees S, Firmin D, Mohiaddin R, et al. Application of flow measurements by magnetic resonance velocity mapping to congenital heart disease. Am J Cardiol. 1989;64:953. 80. Brenner LD, Caputo GR, Mostbeck GH, et al. Quantification of left-toright atrial shunts with velocity-encoded cine nuclear magnetic resonance imaging. J Am Coll Cardiol. 1992;20:1246. 81. Roman KS, Kellenberger CJ, Farooq S, MacGowan CK, Gilday DL, Yoo SJ. Comparative imaging of differential pulmonary blood flow in patients with congenital heart disease: magnetic resonance imaging versus lung perfusion scintigraphy. Pediatr Radiol. 2005;35:295–301. 82. Sieverding L, Jung WI, Klose U, Apirz J. Noninvasive blood flow measurement and quantification of shunt volume by cine magnetic resonance in congenital heart disease. Pediatr Radiol. 1992;22:48. 83. Theissen P, Kaemmerer H, Sechtem U, et al. Magnetic resonance imaging of cardiac function and morphology in patients with transposition of the great arteries following Mustard procedure. Thorac Cardiovasc Surg Suppl. 1991;39:221. 84. Araoz PA, Reddy GP, Tarnoff H, Roge CL, Higgins CB. MR findings of collateral circulation are more accurate measures of hemodynamic significance than arm-leg blood pressure gradient after repair of coarctation of the aorta. J Magn Reson Imaging. 2003;17:177–183. 85. Nishimura RA, Housmans PR, Hatle LK, Tajik AJ. Assessment of diastolic function of the heart: background and current applications of Doppler echocardiography. I. Physiologic and pathophysiologic features. Mayo Clin Proc. 1989;64:71. 86. Steffens JC, Bourne MW, Sakuma H. Quantification of collateral blood flow in coarctation of the aorta by velocity encoded cine magnetic resonance imaging. Circulation. 1994;90:937. 87. Mohiaddin RH, Kilner PJ, Rees S, Longmore DB. Magnetic resonance volume flow and jet velocity mapping in aortic coarctation. J Am Coll Cardiol. 1993;22:1515. 88. Oshinski JN, Parks WJ, Markou CP, et al. Improved measurement of pressure gradients in aortic coarctation by magnetic resonance imaging. J Am Coll Cardiol. 1996;28:1818.

Cardiovascular Magnetic Resonance 419

30 CARDIOVASCULAR MAGNETIC RESONANCE IN COMPLEX CONGENITAL HEART DISEASE

40. Link KM, Herrera MA, D’Souza VJ, Formanek AG. MR imaging of Ebstein anomaly: results in four cases. AJR Am J Roentgenol. 1988;150:363. 41. Markiewicz W, Sechtem U, Higgins CB. Evaluation of the right ventricle by magnetic resonance imaging. Am Heart J. 1987;113:8. 42. Choi YH, Park JH, Choe YH, Yoo SJ. MR imaging of Ebstein’s anomaly of the tricuspid valve. AJR Am J Roentgenol. 1994;163:539. 43. Lev M, Liberthson RR, Joseph RH, et al. The pathologic anatomy of Ebstein’s disease. Arch Pathol. 1970;90:334. 44. Carpentier A, Chauvaud S, Mace L, et al. A new reconstructive operation for Ebstein’s anomaly of the tricuspid valve. J Thorac Cardiovasc Surg. 1988;96:92. 45. Fogel MA. Is routine cardiac catheterization necessary in the management of patients with single ventricles across staged Fontan resonstruction? No! Pediatr Cardiol. 2005;26:154–158. 46. Kersting-Sommerhoff BA, Diethelm L, Stanger P, et al. Evaluation of complex ventricular anomalies with magnetic resonance imaging. Am Heart J. 1990;120:133. 47. Higgins CB, Byrd BF, Farmer DW, et al. Magnetic resonance imaging in patients with congenital heart disease. Circulation. 1984;70:851. 48. Fogel MA, Hubbard A, Weinberg PM. A simplified approach for assessment of intracardiac baffles and extracardiac conduits in congenital heart surgery with two- and three-dimensional magnetic resonance imaging. Am Heart J. 2001;142:1028–1036. 49. Higgins CB, Byrd BF, McNamara MT, et al. Magnetic resonance imaging of the heart: a review of the experience in 172 subjects. Radiology. 1985;155:671. 50. Jacobstein MD, Fletcher BD, Nelson AD, et al. Magnetic resonance imaging: evaluation of palliative systemic-pulmonary artery shunts. Circulation. 1984;70:650. 51. Soulen RL, Donner RM, Capitanio M. Postoperative evaluation of complex congenital heart disease by magnetic resonance imaging. Radiographics. 1987;7:975. 52. Sampson C, Martinez J, Rees S, et al. Evaluation of Fontan’s operation by magnetic resonance imaging. Am J Cardiol. 1990;65:819. 53. Fogel MA, Donofrio MT, Ramaciotti C, et al. Magnetic resonance and echocardiographic imaging of pulmonary artery size throughout stages of Fontan reconstruction. Circulation. 1994;90:2927. 54. Duerinckx AJ, Wexler L, Banerjee A, et al. Postoperative evaluation of pulmonary arteries in congenital heart surgery by magnetic resonance imaging: comparison with echocardiography. Am Heart J. 1994;128:1139. 55. Hornung TS, Kilner PJ, Davlouros PA, Grothues F, Li W, Gatzoulis MA. Excessive right ventricular hypertrophic response in adults with the Mustard procedure for transposition of the great arteries. Am J Cardiol. 2002;90:800–803. 56. Laffon E, Jimenez M, Latrabe V, et al. Quantitative MRI comparison of systemic hemodynamics in Mustard/Senning repaired patients and healthy volunteers at rest. Eur Radiol. 2004;14:875–880. 57. Mee RB. Severe right ventricular failure after Mustard or Senning operation. Two stage repair: pulmonary artery banding and switch. J Thorac Cardiovasc Surg. 1986;92:385. 58. Taylor AM, Dymarkowski S, Hamaekers P, et al. MR coronary angiography and late-enhancement myocardial MR in children who underwent arterial switch surgery for transposition of great arteries. Radiology. 2005;234:542–547. 59. Blankenberg F, Rhee J, Hardy C, et al. MRI vs echocardiography in the evaluation of the Jatene procedure. J Comput Assist Tomogr. 1994;18:749. 60. Fontan F, Baudet E. Surgical repair of tricuspid atresia. Thorax. 1971;26:240. 61. Fratz S, Hess J, Schwaiger M, Marinoff S, Stern HC. More accurate quantification of pulmonary blood flow by magnetic resonance imaging than by lung perfusion scintigraphy in patients with fontan circulation. Circulation. 2002;106:1510–1513. 62. Weir RA, Steedman T, Hillis WS, Swan L. Relief of Fontan obstruction demonstrated non-invasively by cardiac magnetic resonance imaging. Int J Cardiol. 2008;127:e167–e169. 63. Casolo G, Rega L, Gensini GF. Detection of right atrial and pulmonary artery thrombosis after the Fontan procedure by magnetic resonance imaging. Heart. 2004;90:825. 64. Vliegen HW, Van Straten A, de Roos A, et al. Magnetic resonance imaging to assess the hemodynamic effects of pulmonary valve replacement in adults late after repair of tetralogy of Fallot. Circulation. 2002;106:1703–1707. 65. Van Straten A, Vliegen HW, Hazekamp MG, et al. Right ventricular function after pulmonary valve replacement in patients with tetralogy of Fallot. Radiology. 2004;233:824–829.

RIGHT VENTRICULAR AND CONGENITAL HEART DISEASE

CHAPTER 31

Complex Congenital Heart Disease: Infant and Pediatric Patients Tal Geva, Adam L. Dorfman, and Andrew J. Powell

The role of cardiovascular magnetic resonance (CMR) in the evaluation of infants and children with complex con genital heart disease (CHD) is increasing as a result of continued technologic advances. Improvements in hard ware and the development of new, highly efficient imaging techniques have allowed for sufficient spatial and temporal resolution to evaluate cardiac anatomy and function in pediatric patients comprehensively, despite these patients’ small body size and rapid heart rate. Although transtho racic echocardiography provides the necessary diagnostic information in most patients in this age group, CMR offers an important alternative in certain circumstances: (1) when transthoracic echocardiography is incapable of providing the required diagnostic information; (2) when clinical assessment and other diagnostic tests are inconsistent; (3) as an alternative to diagnostic cardiac catheterization, with its associated risks and higher cost; and (4) to obtain diagnostic information for which CMR offers unique advantages. It is helpful to note that complex CHD has no precise accepted definition. For the purposes of this chapter, the definition of complex CHD includes conotruncal anomalies and single ventricle heart disease. Conotruncal anomalies refer to a group of congenital defects involving the outflow tracts of the heart and the great vessels. Single ventricle heart disease refers to a heterogeneous group of anomalies in which one of the two ventricular sinuses is absent (anatomic single ventricle) or to hearts with complex anatomy in which biventricular physiology cannot be attained (functional sin gle ventricle). The special considerations that pertain to patient preparation, sedation, and monitoring in pediatric patients undergoing CMR are discussed in Chapter 9.

TETRALOGY OF FALLOT Tetralogy of Fallot (TOF) is the most common type of cya notic CHD, with an incidence of 356 per million live births.1 Although TOF involves several anatomic components, the anomaly is believed to result from a single developmental anomaly, underdevelopment of the subpulmonary infundib ulum (conus).2,3 The anatomy is characterized by infundib ular and valvar pulmonary stenosis associated with anterior, superior, and leftward deviation of the infundibular (conal) septum; hypoplasia of the pulmonary valve annulus and 420 Cardiovascular Magnetic Resonance

thickened leaflets; ventricular septal defect (VSD); and overriding of the aortic valve above the ventricular septum. The degree of right ventricular outflow tract (RVOT) obstruction varies from mild to complete (i.e., TOF with pulmonary atresia). The size of the mediastinal pulmonary arteries (PAs) varies considerably. Although in some patients they can be dilated (e.g., TOF with absent pulmonary valve syndrome), more commonly, their diameter ranges from nor mal to hypoplastic. In some patients, the mediastinal PAs are discontinuous or absent. In patients with pulmonary atresia or diminutive or absent branch PAs, pulmonary blood flow may come from a patent ductus arteriosus, from collateral vessels arising from the aorta or its branches, or from both sources. The VSD in TOF is usually located between the malaligned conal septum superiorly and the muscular sep tum inferiorly (termed conoventricular septal defect4). The aor tic valve is rotated clockwise (as viewed from the apex) and is positioned above the ventricular septal crest, committing to both the left ventricle (LV) and the right ventricle (RV). In 5% to 6% of patients with TOF, a major coronary artery crosses the RVOT.5 Most commonly, the left anterior des cending (LAD) coronary artery originates from the right coronary artery (RCA) and traverses the infundibular free wall before reaching the anterior interventricular groove. Preoperative identification of a major coronary artery crossing the RVOT is important to avoid inadvertent dam age to the coronary artery during surgical relief of RVOT obstruction. Additional cardiovascular and noncardiac anomalies can be associated with TOF.6 Although the clinical presentation and course of patients with TOF can vary, most patients have cyanosis during the first year of life. Some patients with mild or no RVOT obstruction are not cyanotic at birth (“pink TOF”) and may exhibit signs and symptoms of pul monary overcirculation similar to patients with a large VSD. As these patients grow, the subpulmonary infundibulum becomes progressively obstructive and cyanosis ensues.7 Surgical repair of TOF is usually performed during the first year of life, often during the first 6 months.8 A typical repair includes patch closure of the VSD and relief of the RVOT obstruction using a combination of resection of obstructive muscle bundles and an overlay patch. When the pulmonary valve annulus is moderately or severely hypoplastic, the RVOT patch is extended across the pulmo nary valve annulus into the main pulmonary artery (MPA),

Cardiovascular Magnetic Resonance Evaluations Tetralogy of Fallot is the most frequent diagnosis among patients referred for CMR evaluation at Children’s Hospital Boston. Unlike infants in whom transthoracic echocardiog raphy generally provides all of the necessary diagnostic information for surgical repair,5,16 CMR assumes an increasing role in adolescents and adults with TOF, in whom the acoustic windows are frequently limited. CMR is useful in both pre and postoperative assessment of TOF, but the focus of the examination is different.

Preoperative Cardiovascular Magnetic Resonance Assessment In most patients with unrepaired TOF, the central question for the CMR examination is to delineate all sources of pulmonary blood flow, including the PAs, aortopulmonary

collaterals, and ductus arteriosus. Several studies have shown that black blood spin echo and two dimensional (2D) gradient recalled echo (GRE) cine CMR techniques provide excellent imaging of the central PAs and major aortopulmonary collaterals.17–20 However, these CMR tech niques require relatively long scan times for complete ana tomic coverage, and small vessels (50%) Doppler inflow velocities, diastolic collapse of the right atrium, left atrium, or RV, and dilation of the infe rior vena cava.35,47,50–52 However, as noted earlier, subop timal views may limit the usefulness of echocardiography in certain individuals, particularly those with certain locu lated pericardial effusions and hematomas that may be dif ficult to visualize.4,5 On CT images, common transudative effusions have attenuation similar to that of water. Attenuation greater than that of water suggests a more complex, generally more proteinaceous fluid, as in malignancy, hemopericardium, purulent exudates, or myxedema.8 Using CMR, transudative effusions have longer T1 and T2 relaxation times and tend to appear dark on T1 weighted imaging, bright (high signal intensity) on T2 weighted

Figure 36 5 Pericardial effusion. Transudative pericardial effusion (arrows) shows greater signal intensity than epicardial fat (*) steady state free precession CMR.

imaging, and bright on gradient echo cine images.2,5,8 Trans udative pericardial effusions are often even brighter than epicardial fat on gradient echo images (Fig. 36 5).18 More proteinaceous effusions, such as exudates, have shorter T1 and T2 relaxation times and tend to show intermediate signal intensity on both T1 and T2 weighted images (although sig nal may be reduced by fluid mobility) and lower signal inten sity than blood in the ventricular cavities on cine imaging.8,30,32 In hemorrhagic pericardial effusions, the appearance of blood will depend on the age of the collection.2 Fresh hematomas appear homogenously bright on T1 and T2 weighted images.2,8,53,54 Older collections show the effects of T1 and T2 shortening as a result of methemoglobin formation and generally show heterogeneous signal inten sity, with areas of high signal intensity on both T1 and T2 weighted images.2,8,53–56 A chronic organized thrombus may have a dark rim and low signal intensity foci that may represent fibrosis, calcification, or hemosiderin deposi tion.2,8,54,55 The size of pericardial effusions can be readily assessed quantitatively by CMR using volumetric methods.33

Constrictive Pericarditis Constrictive pericarditis as a result of chronic fibrosis or calcification of the pericardial sac is associated with loss of compliance that impedes diastolic cardiac filling.19,36 Pericardial constriction has a wide variety of potential causes, including cardiac surgery, radiation, trauma, infec tion (tuberculosis; bacterial, particularly purulent; or viral), neoplasia, uremia, connective diseases, Dressler syndrome, sarcoidosis, and drugs. The majority of cases are of uncertain etiology.3,19 Constriction is typically a chronic process, although it can present within days to months after the initial injury, as has been described after cardiac surgery.19 Constrictive pericarditis presents with

Cardiovascular Magnetic Resonance 493

36 THE PERICARDIUM: NORMAL ANATOMY AND SPECTRUM OF DISEASE

*

symptoms and signs of right sided heart failure, including edema or anasarca, and fatigue.3,19 Clinically, it may be difficult to distinguish constrictive pericarditis from other causes of heart failure and edema, such as restrictive car diomyopathy, although current imaging techniques facili tate differentiation of constriction. In both constrictive pericarditis and restrictive cardiomyopathy, ventricular filling is restricted, leading to an increase in diastolic pres sure in the cardiac chambers. In restrictive cardiomyop athy, ventricular filling is limited by abnormal myocardial compliance and relaxation, whereas in pure constriction, myocardial relaxation is preserved.57 Although constric tive pericarditis is occasionally transient, it is generally a chronic, persistent condition that necessitates pericar diectomy for relief. However, pericardiectomy may not be indicated in patients with very mild disease or with severe advanced disease because the operative risk is excessively high.49,58,59 Effusive constrictive pericarditis is an uncommon syn drome in which individuals with pericardial effusion and tamponade show clinical and hemodynamic evidence of pericardial constriction after normalization of intrapericar dial pressure by drainage of the effusion.60,61 Effusive constrictive disease may be caused by any of the many causes of pericarditis, particularly radiation therapy.62 It is best shown with right heart catheterization at the time of pericardiocentesis to assess for residual elevation in right atrial and ventricular diastolic pressure after normalization of intrapericardial pressure.60 Although this syndrome fre quently leads to persistent pericardial constriction that may require pericardiectomy, spontaneous resolution of idiopathic cases has been reported.62 Chest X ray may show a small cardiac silhouette. Peri cardial calcification may be detected in constriction (27% of cases in the 1985 to 1995 Mayo Clinic series3), although this finding is not diagnostic of clinical constriction. Peri cardial calcification is often associated with idiopathic dis ease (67% of cases in the Mayo Clinic series) and is an independent predictor of increased perioperative mortality with pericardiectomy.3 As noted earlier, pericardial thickness is not reliably assessed by TTE, but can be determined using CT or CMR. However, the presence of pericardial thickening by itself does not indicate constriction.33 Also, pericardial thickness is not increased (by imaging and histologic examination) in a significant minority of patients with sur gically proven constrictive pericarditis.63 When pericardial thickening is present in constriction, it may be localized. Nonspecific findings that may suggest constriction include ascites, pleural effusion, and occasionally some pericardial effusion. Often, there is dilation of the atria, coronary sinus, inferior vena cava, and hepatic veins.7 Echocardiography may identify flattening of diastolic LV inferolateral wall motion, abnormal septal motion (septal “bounce”; discussed later), and premature opening of the pulmonic valve.64 In addition, TTE may identify the non specific findings suggestive of elevated atrial pressure and venous congestion. Conventional Doppler imaging may show a restrictive LV inflow pattern, although this pattern may also be seen in restrictive cardiomyopathy and other conditions associated with high atrial pressure. Respiratory variation in transvalvular inflow patterns suggests constriction.65

VASCULATURE AND PERICARDIUM

Figure 36 6 Constrictive pericarditis. Axial contrast enhanced computed tomography image (A) and electrocardiographic gated spin echo T1 weighted cardiovascular magnetic resonance image (B) showing focal pericardial calcification and thickening (arrows). The dilated right atrium (RA) and coronary sinus (CS) are suggestive of pericardial constriction. LV, left ventricle; RV, right ventricle.

RV RA LV CS

CS

A

B

Assessment of pericardial thickening with CT shows good agreement with histopathologic identification of peri cardial thickening.63,66 As discussed earlier, CT is a sensi tive test for the detection of pericardial calcifications. Because calcium produces high attenuation on CT, but a signal deficit on CMR (Fig. 36 6), CT can detect minute amounts of pericardial calcium, whereas CMR may miss significant deposits.7 As discussed earlier, on CMR, normal pericardium has low signal intensity on T1 weighted spin echo imaging, typ ically seen as a darker stripe between the bright layers of epicardial fat and fat around the pericardium.2,30 Thickened pericardium may have moderate to high signal intensity on spin echo images.8,30 The appearance on T2 weighted imag ing is variable, but the signal intensity is usually lower than that of transudative fluid. Thickened pericardium shows intermediate signal intensity that is lower than that of trans udative fluid on gradient echo (including steady state free precession) sequences.30 Normal pericardial thickness seen with CMR is less than 3 mm.27 Thickness of greater than 4 mm indicates pericardial thickening and is strongly sugges tive of constrictive pericarditis in the proper clinical setting. Constriction is frequently localized to the right side of the heart and may even be localized to the region of the right

atrioventricular groove.18,33,67 Larger pericardial calcifica tions may be visualized as regions of low signal intensity by CMR (see Fig. 36 6). The presence of late gadolinium enhancement CMR may suggest pericardial inflammation in effusive constrictive pericarditis.40 Early diastolic septal flattening, giving the appearance of a septal “bounce,” is suggestive of constrictive pericarditis. This finding, originally noted on TTE, may also be detected by real time CMR, and can be helpful in distinguishing constriction from restrictive cardiomyopathy.68 This appear ance can be visualized on short or long axis cine CMR sequences.68 CMR tagging methods, such as spatial modulation of magnetization, may also be useful in diagnos ing constriction. In normal subjects, when tagging stripes laid down in end diastole are successively imaged during ventricular systole, the normal slippage between myo cardium and pericardium results in the appearance of discontinuities, or breaks, in the stripes at the myocar dium pericardium interface. In patients with constrictive pericarditis with adhesion of the parietal pericardium, this slippage is lost in the affected regions. As a result of this “tethering,” tag lines passing through the myocardium pericardium interface maintain continuity during systolic deformation (Fig. 36 7).2,69

RV LV

RA

LA

A

B

C

Figure 36 7 Constrictive pericarditis. Axial spin echo T1 weighted image showing a moderate sized partially organized pericardial effusion (A; arrows), particularly prominent around the lateral wall of the left ventricle (LV). Four chamber spatial modulation of magnetization images (B, end diastole; C, end systole) obtained at a similar level, but slightly oblique to A, show pericardial tethering (a lack of sliding at the myocardium pericardium interface, as indicated by preservation of continuity in each stripe at arrows) with systolic motion of the lateral wall of the LV. LA, left atrium; RA, right atrium; RV, right ventricle. 494 Cardiovascular Magnetic Resonance

PERICARDIAL TUMORS Primary Pericardial Tumors Primary pericardial tumors are rare. Benign pericardial tumors include lipoma, teratoma, fibroma, neuroma, and hemangioma; malignant tumors include mesothelioma, lymphoma, thymoma (may be benign or malignant), sar coma, and liposarcoma.8,19 Benign pericardial tumors found in children are often associated with large pericardial effusions.19 Primary malig nant mesothelioma of the pericardium may cause pericar dial effusion or pericardial plaques70 and may lead to pericardial constriction.19 Lymphoma, sarcoma, and lipo sarcoma typically appear as large, irregular masses, often associated with pericardial effusion.8 Lipomas can be readily recognized by their typical signal characteristics with CMR or CT.2 On CT, lipomas generally have low attenuation. On CMR T1 weighted spin echo images, they have characteristic high signal intensity8 that is not generally altered by contrast administration. Confir mation of the presence of fat signal is achieved by a decrease in signal intensity after application of a fat presaturation technique.34 Depiction of regions of calcium or fat in a peri cardial mass on CT or CMR suggests teratoma.8 Fibromas more commonly arise from the pleura, but may arise from the pericardium. They are usually homogenous in appear ance and appear isointense to hypointense compared with myocardium on T1 weighted images and hypointense on T2 weighted images.34,68 Fibromas may or may not show contrast enhancement.34,71–73 Hemangiomas are generally bright on T1 and T2 weighted images because of their con tent of slow moving blood, and they show strong enhance ment after contrast administration.34

Secondary Malignant Pericardial Tumors Secondary malignant pericardial tumors are much more com mon than primary pericardial tumors and have been found in 10% to 12% of all patients dying with malignancy.73,74 In 110 patients with cardiac metastases, at autopsy, the pericardium was involved in more than 70% of cases and pericardial effu sion was found in approximately one third of those with peri cardial involvement.74 Malignant pericardial involvement is often clinically silent and may be found incidentally, although in a significant minority of cases, it causes symptoms of

pericarditis, tamponade, or even constriction. Patients with large pericardial effusions who present with tamponade with out signs of pericarditis (e.g., chest pain, rub, fever, electrocar diographic changes) may be more likely to have a malignant effusion than other patients with large pericardial effusions.75 Pericardial effusions caused by either malignancy or treat ment of malignancy are more likely than other effusions to require repeat pericardiocentesis or surgical management.76 Malignant involvement of the pericardium may result from local invasion (as for lung, breast, esophageal and gastric can cers; lymphoma; thymoma; and pleural mesothelioma) or metastatic spread (as for breast, lung, melanoma, renal, and others).2,19,71,74 Lung cancer, breast cancer, esophageal can cers; melanoma, leukemia, and lymphoma are the malignan cies most likely to metastasize to the pericardium.42,73–77 Patients with symptomatic malignant pericardial effusion generally have a poor prognosis, although those with breast cancer, leukemia, and lymphoma may have a better prognosis than others.77–79 Metastatic involvement of the pericardium may be suggested by echo, CT, or CMR. Findings include pericar dial effusion and nodular pericardial thickening or peri cardial mass.18 However, pericardial effusion and pericardial thickening in patients with malignancy may not be caused by malignant involvement of the pericardium because radiation, drugs, and idiopathic etiologies may all cause pericardial disease in this population.68,77 Hemorrhagic pericardial effusions are strongly sugges tive of metastatic pericardial disease.7,75,77 Acute hemor rhagic effusions may be identified by their high signal intensity on T1 and T2 weighted spin echo images. Exten sion of local tumor to the pericardium may be confirmed by focal loss of the pericardial line, with or without asso ciated effusion.7,30 An intact pericardial line may be seen if an adjacent tumor extends up to the pericardium, but not through it.8 Cine gradient echo CMR may help deter mine whether the tumor is adherent to the pericardium.71 Most malignant tumors enhance after gadolinium administration.8,72 On noncontrast imaging, most neo plasms have low signal intensity on T1 weighted images and high signal intensity on T2 weighted images. However, metastatic melanoma may have high signal intensity on T1 and T2 weighted images because of the paramagnetic metals bound by melanin.79 Lymphomas appear iso to hypointense to the myocardium on T1 and T2 weighted images.34 They may show lesser contrast enhancement in central regions that may be necrotic.8,80

CONCLUSION The spectrum of pericardial disease produces a variety of clinical presentations, ranging from asymptomatic disor ders to severe hemodynamic compromise. CMR can serve as an important tool to characterize pericardial disorders to assist in their diagnosis and management.

Cardiovascular Magnetic Resonance 495

36 THE PERICARDIUM: NORMAL ANATOMY AND SPECTRUM OF DISEASE

Ventricular filling patterns can be assessed by CMR using flow velocity encoding (phase contrast) sequences. In con strictive pericarditis, increased mitral level E wave (early fill ing) may be observed as a consequence of increased diastolic pressure, whereas A wave (atrial filling) may be of reduced height because of reduced late diastolic filling.16

VASCULATURE AND PERICARDIUM

References 1. Eisenberg MJ, Dunn MM, Kanth N, et al. Diagnostic value of chest radiography for pericardial effusion. J Am Coll Cardiol. 1993;22:588–593. 2. Axel L. Assessment of pericardial disease by magnetic resonance and computed tomography. J Magn Reson Imaging. 2004;19:816–826. 3. Ling LH, Oh JK, Breen JF, et al. Calcific constrictive pericarditis: is it still with us? Ann Intern Med. 2000;132:444–450. 4. Yousem D, Traill TT, Wheeler PS, Fishman EK. Illustrative cases in pericardial effusion misdetection: correlation of echocardiography and CT. Cardiovasc Intervent Radiol. 1987;10:162–167. 5. Mulvagh SL, Rokey R, Vick 3rd GW, Johnston DL. Usefulness of nuclear magnetic resonance imaging for evaluation of pericardial effusions, and comparison with two-dimensional echocardiography. Am J Cardiol. 1989;64:1002–1009. 6. Ling LH, Oh JK, Tei C, et al. Pericardial thickness measured with transesophageal echocardiography: feasibility and potential clinical usefulness. J Am Coll Cardiol. 1997;29:1317–1323. 7. Breen JF. Imaging of the pericardium. J Thorac Imaging. 2001;16:47–54. 8. Wang ZJ, Reddy GP, Gotway MB, et al. CT and MR imaging of pericardial disease. Radiographics. 2003;23 Spec No:S167–S180. 9. Delille JP, Hernigou A, Sene V, et al. Maximal thickness of the normal human pericardium assessed by electron-beam computed tomography. Eur Radiol. 1999;9:1183–1189. 10. Silverman PM, Harell GS. Computed tomography of the normal pericardium. Invest Radiol. 1983;18:141–144. 11. Groell R, Schaffler GJ, Rienmueller R. Pericardial sinuses and recesses: findings at electrocardiographically triggered electron-beam CT. Radiology. 1999;212:69–73. 12. Ropers D, Regenfus M, Wasmeier G, Achenbach S. Non-interventional cardiac diagnostics: computed tomography, magnetic resonance and real-time three-dimensional echocardiography: techniques and clinical applications. Minerva Cardioangiol. 2004;52:407–417. 13. Kodama F, Fultz PJ, Wandtke JC. Comparing thin-section and thicksection CT of pericardial sinuses and recesses. AJR Am J Roentgenol. 2003;181:1101–1108. 14. Truong MT, Erasmus JJ, Gladish GW, et al. Anatomy of pericardial recesses on multidetector CT: implications for oncologic imaging. AJR Am J Roentgenol. 2003;181:1109–1113. 15. Goldstein JA. Cardiac tamponade, constrictive pericarditis, and restrictive cardiomyopathy. Curr Probl Cardiol. 2004;29:503–567. 16. Francone M, Dymarkowski S, Kalantzi M, Bogaert J. Magnetic resonance imaging in the evaluation of the pericardium: a pictorial essay. Radiol Med (Torino). 2005;109:64–74; quiz 75–66. 17. Ellis H. Gray’s Anatomy: The Anatomical Basis of Clinical Practice. NY: Elsevier Churchill Livingstone; 2005. 18. Oyama N, Oyama N, Komuro K, et al. Computed tomography and magnetic resonance imaging of the pericardium: anatomy and pathology. Magn Reson Med Sci. 2004;3:145–152. 19. Spodick D. The Pericardium: A Comprehensive Textbook. New York: Marcel Dekker, Inc; 1997. 20. Vesely TM, Cahill DR. Cross-sectional anatomy of the pericardial sinuses, recesses, and adjacent structures. Surg Radiol Anat. 1986;8: 221–227. 21. Ellis H. Heart and Mediastinum: Heart and Great Vessels. NY: Elsevier Churchill Livingstone; 2005. 22. Iacobellis G, Ribaudo MC, Assael F, et al. Echocardiographic epicardial adipose tissue is related to anthropometric and clinical parameters of metabolic syndrome: a new indicator of cardiovascular risk. J Clin Endocrinol Metab. 2003;88:5163–5168. 23. Iacobellis G, Leonetti F. Epicardial adipose tissue and insulin resistance in obese subjects. J Clin Endocrinol Metab. 2005;90:6300–6302. 24. Batra P, Bigoni B, Manning J, et al. Pitfalls in the diagnosis of thoracic aortic dissection at CT angiography. Radiographics. 2000;20:309–320. 25. Levy-Ravetch M, Auh YH, Rubenstein WA, et al. CT of the pericardial recesses. AJR Am J Roentgenol. 1985;144:707–714. 26. Bull RK, Edwards PD, Dixon AK. CT dimensions of the normal pericardium. Br J Radiol. 1998;71:923–925. 27. Sechtem U, Tscholakoff D, Higgins CB. MRI of the normal pericardium. AJR Am J Roentgenol. 1986;147:239–244. 28. Hort W, Braeun H. [Studies on the dimensions, wall thickness and microscopic structure of the pericardium in normal and pathological conditions.]. Arch Kreislaufforsch. 1962;38:1–22.

496 Cardiovascular Magnetic Resonance

29. Ferrans VJ, Ishihara T, Roberts WC. Anatomy of the Pericardium. New York: Raven Press; 1982. 30. Glockner JF. Imaging of pericardial disease. Magn Reson Imaging Clin N Am. 2003;11:149–162, vii. 31. Hartnell GG, Hughes LA, Ko JP, Cohen MC. Magnetic resonance imaging of pericardial constriction: comparison of cine MR angiography and spin-echo techniques. Clin Radiol. 1996;51:268–272. 32. Brulotte S, Roy L, Larose E. Congenital absence of the pericardium presenting as acute myocardial necrosis. Can J Cardiol. 2007;23: 909–912. 33. Smith WH, Beacock DJ, Goddard AJ, et al. Magnetic resonance evaluation of the pericardium. Br J Radiol. 2001;74:384–392. 34. Frank H, Globits S. Magnetic resonance imaging evaluation of myocardial and pericardial disease. J Magn Reson Imaging. 1999; 10:617–626. 35. Hoffmann U, Globits S, Frank H. Cardiac and paracardiac masses: current opinion on diagnostic evaluation by magnetic resonance imaging. Eur Heart J. 1998;19:553–563. 36. Troughton RW, Asher CR, Klein AL. Pericarditis. Lancet. 2004; 363:717–727. 37. Shaebetai RIM. Evaluation and Management of Acute Pericarditis. Waltham, MA: UpToDate; 2005. 38. Permanyer-Miralda G, Sagrista-Sauleda J, Soler-Soler J. Primary acute pericardial disease: a prospective series of 231 consecutive patients. Am J Cardiol. 1985;56:623–630. 39. Zayas R, Anguita M, Torres F, et al. Incidence of specific etiology and role of methods for specific etiologic diagnosis of primary acute pericarditis. Am J Cardiol. 1995;75:378–382. 40. Watanabe A, Hara Y, Hamada M, et al. A case of effusive-constructive pericarditis: an efficacy of GD-DTPA enhanced magnetic resonance imaging to detect a pericardial thickening. Magn Reson Imaging. 1998;16:347–350. 41. Taylor AM, Dymarkowski S, Verbeken EK, Bogaert J. Detection of pericardial inflammation with late-enhancement cardiac magnetic resonance imaging: initial results. Eur Radiol. 2006;16:569–574. 42. Chiles C, Woodard PK, Gutierrez FR, Link KM. Metastatic involvement of the heart and pericardium: CT and MR imaging. Radiographics. 2001;21:439–449. 43. Lam KY, Dickens P, Chan AC. Tumors of the heart: a 20-year experience with a review of 12,485 consecutive autopsies. Arch Pathol Lab Med. 1993;117:1027–1031. 44. Sexton DJCG. Purulent Pericarditis. Waltham, MA: UpToDate; 2005. 45. Rubin RH, Moellering Jr RC. Clinical, microbiologic and therapeutic aspects of purulent pericarditis. Am J Med. 1975;59:68–78. 46. Klacsmann PG, Bulkley BH, Hutchins GM. The changed spectrum of purulent pericarditis: an 86 year autopsy experience in 200 patients. Am J Med. 1977;63:666–673. 47. Maisch B, Seferovic PM, Ristic AD, et al. Guidelines on the diagnosis and management of pericardial diseases executive summary; the task force on the diagnosis and management of pericardial diseases of the European Society of Cardiology. Eur Heart J. 2004;25:587–610. 48. Spodick DH. Acute cardiac tamponade. N Engl J Med. 2003; 349:684–690. 49. Hoit BD. Management of effusive and constrictive pericardial heart disease. Circulation. 2002;105:2939–2942. 50. Cheitlin MD, Armstrong WF, Aurigemma GP, et al. ACC/AHA/ASE 2003 guideline update for the clinical application of echocardiography: summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASE Committee to Update the 1997 Guidelines for the Clinical Application of Echocardiography). Circulation. 2003;108: 1146–1162. 51. Leeman DE, Levine MJ, Come PC. Doppler echocardiography in cardiac tamponade: exaggerated respiratory variation in transvalvular blood flow velocity integrals. J Am Coll Cardiol. 1988;11:572–578. 52. Levine MJ, Lorell BH, Diver DJ, Come PC. Implications of echocardiographically assisted diagnosis of pericardial tamponade in contemporary medical patients: detection before hemodynamic embarrassment. J Am Coll Cardiol. 1991;17:59–65. 53. Seelos KC, Funari M, Chang JM, Higgins CB. Magnetic resonance imaging in acute and subacute mediastinal bleeding. Am Heart J. 1992;123:1269–1272. 54. Vilacosta I, Gomez J, Dominguez J, et al. Massive pericardiac hematoma with severe constrictive pathophysiologic complications after

56. 57.

58. 59. 60. 61. 62. 63. 64. 65. 66. 67.

68. Giorgi B, Mollet NR, Dymarkowski S, et al. Clinically suspected constrictive pericarditis: MR imaging assessment of ventricular septal motion and configuration in patients and healthy subjects. Radiology. 2003;228:417–424. 69. Kojima S, Yamada N, Goto Y. Diagnosis of constrictive pericarditis by tagged cine magnetic resonance imaging. N Engl J Med. 1999;341: 373–374. 70. Song H, Choi YW, Jang IS, et al. Pericardium: anatomy and spectrum of disease on computed tomography. Curr Probl Diagn Radiol. 2002;31:198–209. 71. Gilkeson RC, Chiles C. MR evaluation of cardiac and pericardial malignancy. Magn Reson Imaging Clin N Am. 2003;11:173–186, viii. 72. Funari M, Fujita N, Peck WW, Higgins CB. Cardiac tumors: assessment with Gd-DTPA enhanced MR imaging. J Comput Assist Tomogr. 1991;15:953–958. 73. Abraham KP, Reddy V, Gattuso P. Neoplasms metastatic to the heart: review of 3314 consecutive autopsies. Am J Cardiovasc Pathol. 1990;3:195–198. 74. Klatt EC, Heitz DR. Cardiac metastases. Cancer. 1990;65:1456–1459. 75. Sagrista-Sauleda J, Merce J, Permanyer-Miralda G, Soler-Soler J. Clinical clues to the causes of large pericardial effusions. Am J Med. 2000;109:95–101. 76. Gornik HL, Gerhard-Herman M, Beckman JA. Abnormal cytology predicts poor prognosis in cancer patients with pericardial effusion. J Clin Oncol. 2005;23:5211–5216. 77. Wilkes JD, Fidias P, Vaickus L, Perez RP. Malignancy-related pericardial effusion: 127 cases from the Roswell Park Cancer Institute. Cancer. 1995;76:1377–1387. 78. Tsang TS, Seward JB, Barnes ME, et al. Outcomes of primary and secondary treatment of pericardial effusion in patients with malignancy. Mayo Clin Proc. 2000;75:248–253. 79. Mousseaux E, Meunier P, Azancott S, et al. Cardiac metastatic melanoma investigated by magnetic resonance imaging. Magn Reson Imaging. 1998;16:91–95. 80. Dorsay TA, Ho VB, Rovira MJ, et al. Primary cardiac lymphoma: CT and MR findings. J Comput Assist Tomogr. 1993;17:978–981.

Cardiovascular Magnetic Resonance 497

36 THE PERICARDIUM: NORMAL ANATOMY AND SPECTRUM OF DISEASE

55.

insertion of an epicardial pacemaker. Am Heart J. 1995;130: 1298–1300. Jungehulsing M, Sechtem U, Theissen P, et al. Left ventricular thrombi: evaluation with spin-echo and gradient-echo MR imaging. Radiology. 1992;182:225–229. Ferguson ER, Blackwell GG, Murrah CP, Holman WL. Evaluation of complex mediastinal masses by magnetic resonance imaging. J Cardiovasc Surg (Torino). 1998;39:117–119. Rajagopalan N, Garcia MJ, Rodriguez L, et al. Comparison of new Doppler echocardiographic methods to differentiate constrictive pericardial heart disease and restrictive cardiomyopathy. Am J Cardiol. 2001;87:86–94. Seifert FC, Miller DC, Oesterle SN, et al. Surgical treatment of constrictive pericarditis: analysis of outcome and diagnostic error. Circulation. 1985;72:II264–II273. Ling LH, Oh JK, Schaff HV, et al. Constrictive pericarditis in the modern era: evolving clinical spectrum and impact on outcome after pericardiectomy. Circulation. 1999;100:1380–1386. Hancock EW. A clearer view of effusive-constrictive pericarditis. N Engl J Med. 2004;350:435–437. Zagol B, Minderman D, Munir A, D’Cruz I. Effusive constrictive pericarditis: 2D, 3D echocardiography and MRI imaging. Echocardiography. 2007;24:1110–1114. Sagrista-Sauleda J, Angel J, Sanchez A, et al. Effusive-constrictive pericarditis. N Engl J Med. 2004;350:469–475. Talreja DR, Edwards WD, Danielson GK, et al. Constrictive pericarditis in 26 patients with histologically normal pericardial thickness. Circulation. 2003;108:1852–1857. Engel PJ, Fowler NO, Tei CW, et al. M-mode echocardiography in constrictive pericarditis. J Am Coll Cardiol. 1985;6:471–474. Hatle LK, Appleton CP, Popp RL. Differentiation of constrictive pericarditis and restrictive cardiomyopathy by Doppler echocardiography. Circulation. 1989;79:357–370. Oren RM, Grover-McKay M, Stanford W, Weiss RM. Accurate preoperative diagnosis of pericardial constriction using cine computed tomography. J Am Coll Cardiol. 1993;22:832–838. Masui T, Finck S, Higgins CB. Constrictive pericarditis and restrictive cardiomyopathy: evaluation with MR imaging. Radiology. 1992;182: 369–373.

Valvular Heart Disease Philip J. Kilner and Raad H. Mohiaddin

As an imaging modality, cardiovascular magnetic resonance (CMR) offers unrivaled versatility and freedom of anatomic access. In relation to heart valve disease, its relative strengths include the following:
Depiction by cine imaging of valve movements and jet flow in planes, or stacks of planes, of any orientation
Measurement of right as well as left ventricular volumes and mass by multislice cine imaging
Measurement of volume flow and regurgitant fraction (pulmonary and aortic, at least) by phase contrast veloc ity mapping
Assessment of the context and consequences of heart valve disease using the wide fields of view, multiple image slices, and the versatility of tissue characterization available to magnetic resonance. So although CMR is generally regarded as a second line imaging modality after echocardiography for the assessment of heart valve disease,1 it can have important contributions to make toward decision making in regard to the timing and nature of surgical intervention, particu larly in cases in which there have been inconclusive or conflicting findings, perhaps owing to inadequate echocar diographic access, or in which heart valve dysfunction is one aspect of more complex congenital or acquired pathol ogy.2 And potentially, at least, CMR offers several possible methods for the measurement of valve regurgitation,3–6 although work is still needed to optimize and standardize acquisition protocols and to fully validate the techniques used. If CMR is to become established as a reliable second line modality, several potential weaknesses or pitfalls need to be recognized and avoided or corrected:
The slice thickness (typically 6 mm) and the dimen sions of voxels (typically about 6  1.5  1.2 mm) of cine and velocity map acquisitions need to be borne in mind.
The images are not usually acquired in real time but are reconstructed from data gathered over several heart cycles during a breath hold. For these reasons, valve leaflets are not always well seen, especially if there is arrhythmia. Nor is CMR effective for visualizing the smaller, more mobile vegetations of endocarditis, although it can be useful for identifying an abscess or false aneurysm.
It is important to attempt to depict valve and jet struc ture by cine imaging in several planes and orientations, not just one; appearances and, potentially, interpretation can differ considerably between images of a particular valve acquired in different planes (Fig. 37 1).

Contiguous stacks of cine images are valuable for cover ing all parts of the mitral or tricuspid valves.
The accuracy of phase velocity mapping cannot be taken for granted.7 For jet velocity mapping, the dimensions of the voxels can be large in relation to the size and shape of a narrow or fragmented jet, leading to possible inac curacies due to signal loss, partial volume averaging, and other artifacts. For the measurement of volume flow, particularly for the calculation of regurgitant fractions, surprisingly large inaccuracies have been found to occur as CMR systems have been “improved” to allow more rapid acquisition in the last 10 years, due to eddy cur rents and concomitant gradients.7,8 The severity of the problem can vary considerably according to the hard ware and software that are used and can change between the upgrades of a system. The uncertainties do not end there, however. The sequence parameters and imaging plane that are needed to be chosen appro priately relative to the characteristics of flow under investigation if artifacts are to be minimized.
Measurements of biventricular volume and function are not necessarily as reliable and straightforward as has often been implied, particularly if there is arrhythmia or the right ventricle is structurally abnormal owing to congenital heart disease or previous surgery, and volume analysis remains time consuming and to some extent subjective. The methods of acquisition and analysis need to be specified appropriately for comparisons over time or between patients.
In general, the unrivaled versatility of CMR is a strength, but it is also a challenge. There is an ongoing dilemma between continuing development on the one hand and standardization on the other. With these cautionary remarks in mind, the aim of this chapter is to describe some important underlying princi ples and to guide users to the CMR techniques that we and others have found the most informative. Stenotic and regurgitant lesions of each heart valve are considered below, and an overview of measures of the severity of valve dysfunction is given in Table 37 1, which is adapted from the 2006 ACC/AHA Guidelines for the Management of Patients with Valvular Heart Disease.1 It is important to realize that right sided valve lesions, particu larly pulmonary or tricuspid regurgitation, differ from their counterparts on the left. While echocardiography has become well established in the assessment of valvular lesions of the left heart, CMR, with phase velocity mapping, has particular strengths in relation to the assessment of val vular and perivalvular lesions of the right heart.

Cardiovascular Magnetic Resonance 501

37 VALVULAR HEART DISEASE

CHAPTER 37

FUNCTIONAL CARDIOVASCULAR DISEASE

Figure 37 1 Top left, A systolic frame of an SSFP cine acquisition shows the bright core and dark edges of the jet through the slitlike orifice of a stenosed bicuspid aortic valve. Top right, A phase contrast velocity map, encoded through plane, shows the dark jet, recording a peak velocity of 4.2 m/sec. Bottom left, Steady state free precession (SSFP) cine imaging aligned with the jet direction and orientated at right angles to the slit shows the narrower dimension of the jet, which appears bright and clearly delineated. Bottom right, The jet appears broad, diffuse, and dark in the SSFP cine, which is aligned with the length of the slit, its voxels spanning the width of the narrow jet.

Table 37-1 Classification of the Severity of Heart Valve Disease in Adults, Based Partly on the 2006 ACC/AHA Guidelines,1 Which Refer Mainly to Echocardiographic Indices Aortic Stenosis

Peak jet velocity (m/sec) Orifice area (cm2) Orifice area index (cm2/m2) Additional features

Mild

Moderate

Severe

1.5

3 4 1.0 1.5

>4 140/90 mmHg), hypertrophic hearts (assessed by two dimensional echocardiography), and no coronary artery stenosis and 9 healthy subjects with no heart dis ease. The results showed that the change in T2* between basal and dipyridamole infused states (DT2*) was approximately threefold lower in hypertrophic patients than in healthy controls. This observation is consistent with previous reports32 that myocardial perfusion reserves are reduced in people with hypertrophic hearts (refer to Fig. 42 1). From the perspective of myocardial oxygena tion, in hypertrophic patients, vasodilatory mechanisms are compromised depending on the severity of the disease and are not fully effective at altering the oxygenation of the myocardium as in a healthy heart. This inability to alter blood oxygenation between the basal and dipyrida mole infused states in hypertrophic patients is likely the reason for the substantially different T2* between healthy subjects and hypertrophic patients. Cardiovascular Magnetic Resonance 573

42 MAGNETIC RESONANCE ASSESSMENT OF MYOCARDIAL OXYGENATION

Pat. #1 No significant stenosis

FUNCTIONAL CARDIOVASCULAR DISEASE

EMERGING TECHNIQUES FOR OXYGEN-SENSITIVE MYOCARDIAL CARDIOVASCULAR MAGNETIC RESONANCE IMAGING More recently, systematic studies have demonstrated that SSFP imaging can be used in the assessment of oxygen sensitive imaging. Over the past decade, SSFP imaging has become invaluable in CMR. Superior temporal resolution and SNR are the distinguishing features of SSFP. Using this method, controlled in vitro studies with blood samples oxygenated to various levels revealed that when the off resonance effects are minimized through the appropriate choice of flip angle and phase cycling scheme, oxygen sensitive contrast can be realized in whole blood in a TR dependent manner.59 The results showed that relatively long TRs (compared to those that are used in conventional SSFP imaging) are necessary to estab lish oxygen sensitive contrast in SSFP images in blood. These studies were extended to intravascular 3D peripheral angio graphic methods aimed at discriminating arteries and veins based on %O2 differences between the vessels.60 The advan tages of the SSFP method over other oxygen sensitive CMR methods are the short scans, significant increase in SNR, reduced heat deposition, and improved oxygen sensitivity.

A

B

Studies demonstrating the feasibility of detecting oxygen sensitive signal changes in microcirculations, specifically in kidneys and dorsal muscles of rabbits, have also been achieved by altering the systemic %O2. The results showed that BOLD SSFP contrast is directly linked to field strength, blood volume, and baseline microcirculatory oxygen satura tion levels.61 The mechanisms that determine oxygen contrast were connected to a fast exchange of spins between the plasma and red blood cells, as well as the perivascular gradi ents due to the susceptibility shifts between the intravascular and extravascular pools of spins. To understand the relation ship between oxygen sensitivity and CMR scan parameters for tissue, ischemic leg cuff studies have been performed. The results show that flip angle and TR are important parameters that determine oxygen contrast in SSFP images of skeletal muscle.62 In particular, results indicate that relatively long TRs are necessary for ensuring realizing oxygen contrast. These studies were followed up with SSFP based myocar dial BOLD CMR in canine models. Using a similar animal model as was described previously,30 controlled studies, which allow for the occlusion of the LCX in the presence of adenosine infusion were performed. Two dimensional (2D) cine SSFP imaging with a relatively long TR (6.3 msec) was used to demonstrate that occlusion of the LCX leads to regional myocardial oxygen deficit in the coronary territory supplied by LCX. The results were compared to the first pass perfusion technique and validated with microsphere based perfusion analysis (Fig. 42 5). Two dimensional BOLD SSFP

C –5%

D

E

100%

F

Figure 42 5 End systolic short axis images obtained from an instrumented canine showing regional myocardial BOLD contrast obtained with two dimensional steady state free precession (SSFP) imaging at baseline (no adenosine and no stenosis) (A), prestenosis (with adenosine and no stenosis) (B), mild stenosis (C), and severe stenosis (D). First pass perfusion image (E) at severe stenosis and percent microsphere flow difference between stenosis and nonstenosis states (F) are shown for reference. All stenosis studies were performed with adenosine. Note the discriminating signal loss in the LCX territories (subtended by arrows) in image D during stenosis of the LCX and the close correspondence between the first pass perfusion and microsphere based flow difference map. Source: Adapted from Dharmakumar R, Mangalathu Arumana J, Larson AC, Chung Y, Wright D, Li D. Cardiac phase resolved blood oxygen sensitive steady state free precession MRI for evaluating the functional significance of coronary artery stenosis. Invest Radiol 2007;42(3):180 188. 574 Cardiovascular Magnetic Resonance

MS

LS

LD

Severe stenosis

Mild stenosis

Pre-stenosis

Baseline

ES

Recently, 2D SSFP BOLD imaging has also been used in a feasibility study in patients (N ¼ 9) suspected of coronary artery disease.64 Patients were identified on the basis of positive thallium SPECT and coronary angiograms. End systolic frames obtained before and during administration of adenosine were analyzed in a segmental fashion by using the six segment model recommended by the American Heart Association. On the basis of thallium SPECT, segments were

A Figure 42 6 Cardiac phase resolved BOLD CMR with two dimensional steady state free precession (SSFP) imaging in canines. On the left panel (A), typical cardiac phase resolved blood oxygen level dependent (BOLD) images (early systole (ES), midsystole (MS), late systole (LS), and late diastole (LD)) showing regional myocardial oxygen deficits in the left circumflex artery (LCX) territory during mild and severe stenosis of the LCX (with adenosine). Note that the extent of signal loss in the LCX territory is related to the extent of LCX stenosis. Baseline and prestenosis images are also shown at the same cardiac phases for reference. (Continued) Cardiovascular Magnetic Resonance 575

42 MAGNETIC RESONANCE ASSESSMENT OF MYOCARDIAL OXYGENATION

method accurately predicted the regional myocardial flow deficit region identified by the first pass technique employ ing an exogenous contrast media.63 SSFP method is expected to provide additional benefits over T2 prepared methods, because it has the capacity to allow for cardiac phase resolved BOLD imaging, permitting increased confidence for evaluating myocardial BOLD signal changes in the pres ence of a coronary artery stenosis (Fig. 42 6).

% change in microsphere-based ? ? severe stenosis relative to prestenosis

30

80

25

70 60

20

50

15

40 30

10

20

5

10 0

0 1

2

3

4

5

6

7

% change in SSFP signal in severe stenosis relative to prestenosis

8

Myocardial sector

B

Figure 42 6 cont’d The plot on the right panel (B) shows the percent change in SSFP based BOLD contrast at midsystole (open circles), end systole (open triangles), and late diastole (open squares) and the associated microsphere based flow changes (closed squares) observed relative to prestenosis over all studies. Note the close correspondence between the magnetic resonance (MR) and microsphere data throughout the myocardium (sectors 1 through 8) at all the cardiac phases analyzed. The MR and microsphere measures of relative signal changes are plotted as mean  standard error. The dotted black curves (MR) and solid gray curves (microsphere) are provided for visual guidance. Source: Adapted from Dharmakumar R, Mangalathu Arumana J, Larson AC, Chung Y, Wright D, Li D. Cardiac phase resolved blood oxygen sensitive steady state free precession MRI for evaluating the functional significance of coronary artery stenosis. Invest Radiol 2007;42(3):180 188.

classified as healthy, mildly affected, or severely affected. Seg mental SSFP signal intensities at rest and stress were measured, and the signal intensity ratio (stress/rest) was cal culated. Figure 42 7A shows representative examples of

SSFP BOLD

FPP

mid LV short axis SSFP based myocardial BOLD CMR images obtained under rest and stress from a patient with 70% narrowing of the LAD. For reference, the corresponding mid LV short axis first pass perfusion and thallium SPECT images are also shown. Note the discriminating signal loss in the stress images (relative to rest) in the anterior zones of the myocardium and its close correspondence to first pass and thallium SPECT images. Statistical results (Fig. 42 7B) show that there are significant differences in stress/rest values computed from healthy, mildly affected, and severely affected segments (p < 0.05). Further studies are warranted to establish the sensitivity and specificity of the technique. Although the advantage of using high field CMR for enhancing the sensitivity of myocardial BOLD imaging is well established in animals,35–36 it has not been validated in humans. To examine whether 3.0 T BOLD imaging can provide increased sensitivity for detecting myocardial oxy genation changes, theoretical and experimental canine studies were performed at 1.5 T and 3.0 T65 (Fig. 42 8). Theoretical results showed that at 3.0 T, a 3.0 fold increase in oxygen sensitivity could be expected in comparison to 1.5 T. Experimental canine studies showed that a 2.5 fold  0.2 fold increase in BOLD sensitivity is possible at 3.0 T relative to 1.5 T. On the basis of the relationship between BOLD signal changes and microsphere perfusion, it was found that the minimum regional perfusion differ ence that can be detected with SSFP based myocardial BOLD imaging at 1.5 T and 3.0 T were 2.9 and 1.6, respec tively. These findings suggest that SSFP based myocardial BOLD imaging at 3.0 T may have the necessary sensitivity to detect the clinically required minimum flow difference of 2.0 that is achievable with first pass perfusion methods. However, additional improvements in shimming techni ques are expected to be necessary prior to successful clini cal translation.

Thallium 1.20

Stress/rest

Rest

1.15 1.10 1.05 1.00

Stress

FUNCTIONAL CARDIOVASCULAR DISEASE

90

0.95 Severely affected Mildly affected

A

B

Healthy

Myocardial segments

Figure 42 7 Patient studies with two dimensional steady state free precession (SSFP) blood oxygen level dependent (BOLD) at 1.5 T. Midventricular short axis SSFP BOLD, first pass perfusion (FPP), and thallium SPECT images obtained from a patient with 70% stenosis of the left anterior descending (LAD) coronary artery at rest and stress (A). The windowing and leveling of images obtained at rest and stress are the same. Myocardial signal in the rest BOLD, FPP, and SPECT images are relatively homogenous. However, under stress, the territory supplied by the LAD (larger arc subtended by arrows) does not increase in the BOLD images as expected. This pattern of regional signal differences is also evident in the FPP and SPECT images. Statistical results from the myocardial BOLD signal analysis showed that significant differences in stress/rest values exist between healthy and affected regions (B). In comparison to healthy segments, the stress/rest values of affected regions are lower, consistent with previous findings in animals that SSFP signals obtained under pharmacologic stress are significantly reduced in myocardial territories supplied by stenotic arteries. Source: Adapted from Dharmakumar R, Green JD, Flewitt J, Voehringer M, Filipchuk NG, Li D, Friedrich MG. Blood oxygen sensitive SSFP imaging for probing the myocardial perfusion reserves of patients with coronary artery disease: A feasibility study. SCMR 2008 (Los Angeles, USA). 576 Cardiovascular Magnetic Resonance

A

B

C

D

A′

B′

C′

D′

100%

Figure 42 8 Short axis two dimensional steady state free precession (SSFP) magnetic resonance images obtained at 1.5 T (top row, A C) and at 3 T (bottom row, A0 C0 ) in canines. Images A and A0 are SSFP images obtained without stenosis, images B and B0 are SSFP images at systole under severe stenosis of the LCX, and images C and C0 are the corresponding first pass images acquired under stenosis of similar extent as in B and B0 . Images D and D0 represent the spatial map (scale provided by the gray scale bar) of percent difference in microsphere based regional flow between prestenosis and severe stenosis in the presence of adenosine infusion at 1.5 T and 3.0 T, respectively. The arrows subtend the suspected regions (left circumflex artery territory) where the perfusion deficits are expected to develop as a result of left circumflex artery stenosis in dogs. Note the discriminating signal loss in these regions in images B and B0 and the close correspondence between the first pass perfusion (C and C0 ) and microsphere based flow difference maps (D and D0 ). Source: Adapted from Dharmakumar R, Mangalathu Arumana J, Tang R, Harris K, Zhang Z, Li D. Assessment of regional myocardial oxygenation changes in the presence of coronary artery stenosis with balanced SSFP imaging at 3.0 T: theory and experimental evaluation in canines. J Magn Reson Imaging 2008;27(5):1037 1045.

FUTURE OF MYOCARDIAL BOLD CARDIOVASCULAR MAGNETIC RESONANCE IMAGING While recent advances in cardiac BOLD imaging in animal studies are promising, the newer techniques need to be validated and extended through prospective clinical studies in patients. Next, all myocardial BOLD imaging methods currently require multiple breath holds for full myocardial coverage. Since typical myocardial oxygenation changes are assessed in the presence of pharmacologic vasodilation, there are pragmatic limitations with 2D approaches that reduce the achievable myocardial coverage within the typi cal 4 to 6 minute adenosine infusion protocol. Moreover,

multiple breath holds require multiple recovery periods for patients during adenosine infusion, further reducing the time available for imaging. It is expected that free breathing methods, such as self gating,66 combined with parallel imag ing strategies67–70 can significantly enhance the myocardial coverage and reduce patient discomfort associated with breath holding during the adenosine protocol. Finally, the preliminary findings by Wacker and colleagues19 that chronic coronary artery disease may be assessed without pro vocative stress needs to be systematically investigated with newer BOLD MRI methods that provide substantial improve ment in image quality. Successful adoption of recent advances through clinical studies and additional technical improvements can propel myocardial BOLD MRI to becom ing a powerful noninvasive diagnostic method in the early detection and post treatment monitoring of ischemic heart disease.

References 1. Brown MA, Marshall DR, Sobel BE. Delineation of myocardial oxygen utilization with carbon-11-labeled acetate. Circulation. 1987; 76:687–696. 2. Brown MA, Myears DW, Bergmann SR. Noninvasive assessment of canine myocardial oxidative metabolism with carbon-11 acetate and positron emission tomography. J Am Coll Cardiol. 1988;12: 1054–1063. 3. Brown MA, Myears DW, Bergmann SR. Validity of estimates of myocardial oxidative metabolism with carbon-11 acetate in positron emission

tomography despite altered patterns of substrate utilization. J Nucl Med. 1989;30:187–193. 4. Buxton DB, Nienaber CA, Luxen A, et al. Noninvasive quantitation of regional myocardial oxygen consumption in vivo with [1–11C] acetate in dynamic positron emission tomography. Circulation. 1989;79:134–142. 5. Henes CG, Bergmann SR, Walsh MN. Assessment of myocardial oxidative metabolic reserve with positron emission tomography and carbon-11 acetate. J Nucl Med. 1989;30:1489–1499.

Cardiovascular Magnetic Resonance 577

42 MAGNETIC RESONANCE ASSESSMENT OF MYOCARDIAL OXYGENATION

0%

FUNCTIONAL CARDIOVASCULAR DISEASE

6. Gropler RJ, Siegel BA, Sampathkumaran KS. Dependence of recovery of contractile function on maintenance of oxidative metabolism after myocardial infarction. J Am Coll Cardiol. 1992;19:989–997. 7. Gropler RJ, Geltman EM, Sampathkumaran KS. Functional recovery after revascularization for chronic coronary artery disease is dependent on maintenance of oxidative metabolism. J Am Coll Cardiol. 1992;20:569–577. 8. Gropler RJ, Geltman EM, Sampathkumaran KS. Comparison of C-11 acetate with F-18 fluorodeoxyglucose for delineating viable myocardium by positron emission tomography. J Am Coll Cardiol. 1993;22:1587–1597. 9. Wendland MF, Saeed M, Lauerma K, de Crespigny A, Moseley ME, Higgins CB. Endogenous susceptibility contrast in myocardium during apnea measured using gradient recalled echo planar imaging. Magn Reson Med. 1993;29:273–276. 10. Atalay MK, Forder JR, Chacko VP, Kawamoto S, Zerhouni EA. Oxygenation in the Rabbit Myocardium: Assessment with SusceptibilityDependent MR imaging. Radiology. 1993;189:759. 11. Balaban RS, Taylor JF, Turner R. Effect of cardiac flow on gradient recalled echo images of the canine heart. NMR Biomed. 1994;7:89–95. 12. Stillman AE, Wilke N, Jerosch-Herold M. BOLD contrast of the heart during occlusion and reperfusion. In: Works in Progress Supplement, SMR 1st Meeting (Dallas); 1994:S24. 13. Atalay M, Reeder SB, Zerhouni E, Forder JR. Blood oxygenation dependence of T1 and T2 in the isolated, perfused rabbit heart at 4.7T. Magn Reson Med. 1995;34:623–627. 14. Li D, Dhawale P, Rubin PJ, Haacke EM, Gropler RJ. Myocardial signal response to dipyridamole and dobutamine: demonstration of the BOLD effect using a double-echo gradient-echo sequence. Magn Reson Med. 1996;36:16–20. 15. Niemi P, Poncelet BP, Kwong K, et al. Myocardial intensity changes associated with flow stimulation in blood oxygenation sensitive magnetic resonance imaging. Magn Reson Med. 1996;36:78–82. 16. Li D, Oellerich WF, Beck G, Gropler RJ. Assessment of myocardial response to pharmacologic interventions using an improved MR imaging technique to estimate T2* values. AJR Am J Roentgenol. 1999;172: 141–145. 17. Wacker CM, Bock M, Hartlep AM, et al. Changes in myocardial oxygenation and perfusion under pharmacological stress with dipyridamole: assessment using T2* and T1 measurements. Magn Reson Med. 1999;41:686–695. 18. Friedrich MG, Niendorf T, Schulz-Menger J, Gross CM, Dietz R. Blood oxygen level-dependent magnetic resonance imaging in patients with stress-induced angina. Circulation. 2003;108(18):2219–2223. 19. Wacker CM, Hartlep AW, Pfleger S, Schad LR, Ertl G, Bauer WR. Susceptibility-sensitive magnetic resonance imaging detects human myocardium supplied by a stenotic coronary artery without a contrast agent. J Am Coll Cardiol. 2003;41(5):834–840. 20. Beache GM, Herzka DA, Boxerman JL, et al. Attenuated myocardial vasodilator response in patients with hypertensive hypertrophy revealed by oxygenation-dependent magnetic resonance imaging. Circulation. 2001;104:1214–1217. 21. Wright KB, Klocke FJ, Deshpande VS, et al. Assessment of regional differences in myocardial blood flow using T2-weighted 3D BOLD imaging. Magn Reson Med. 2001;46:573–578. 22. Foltz WD, Huang H, Fort S, Wright GA. Vasodilator response assessment in porcine myocardium with magnetic resonance relaxometry. Circulation. 2002;106:2714–2719. 23. Foltz WD, Al-Kwifi O, Sussman MS, Stainsby JA, Wright GA. Optimized spiral imaging for measurement of myocardial T2 relaxation. Magn Reson Med. 2003;49:1089–1097. 24. Zheng J, Wang J, Rowold FE, Gropler RJ, Woodard PK. Relationship of apparent myocardial t2 and oxygenation: towards quantification of myocardial oxygen extraction fraction. J Magn Reson Imaging. 2004;20(2):233–241. 25. Zheng J, Wang J, Nolte M, Li D, Gropler RJ, Woodard PK. Dynamic estimation of the myocardial oxygen extraction ratio during dipyridamole stress by MRI: a preliminary study in canines. Magn Reson Med. 2004;51(4):718–726. 26. Zhang H, Gropler RJ, Li D, Zheng J. Assessment of myocardial oxygen extraction fraction and perfusion reserve with BOLD imaging in a canine model with coronary artery stenosis. J Magn Reson Imaging. 2007;26(1):72–79. 27. McCommis KS, Goldstein TA, Zhang H, Misselwitz B, Gropler RD, Zheng J. Quantification of myocardial blood volume during dipyridamole and dobutamine stress: a perfusion CMR study. J Cardiovasc Magn Reson. 2007;9(5):785–792.

578 Cardiovascular Magnetic Resonance

28. McCommis KS, Zhang H, Herrero O, Gropler RD, Zheng J. Feasibility study of myocardial perfusion and oxygenation by noncontrast MRI: comparison with PET study in a canine model. J Magn Reson Imaging. 2008;26(1):11–19. 29. Fieno DS, Shea SM, Li Y, Finn JP, Li D. Myocardial perfusion imaging based on the blood oxygen level-dependent effect using T2-prepared steady-state free precession magnetic resonance imaging. Circulation. 2004;110:1284–1290. 30. Shea SM, Fieno DS, Schirf BE, et al. T2-prepared steady-state free precession blood oxygen level-dependent MR imaging of myocardial perfusion in a dog stenosis model. Radiology. 2005;236(2):503–509. 31. Wilke N, Simm C, Zhang J, et al. Contrast-enhanced first pass myocardial perfusion imaging: correlation between myocardial blood flow in dogs at rest and during hyperemia. Magn Reson Med. 1993;29:485–497. 32. Klocke FJ. Measurements of coronary flow reserve: defining pathophysiology versus making decisions about patient care. Circulation. 1987;76(6):1183–1189. 33. Foltz WD, Merchant N, Downar E, Stainsby JA, Wright GA. Coronary venous oximetry using MRI. Magn Reson Med. 1999;42:837–848. 34. Wendland MF, Saeed M, Mausi T, Derugin N, Higgins CB. First pass of an MR susceptibility contrast agent through normal and ischemic heart: gradient-recalled echo-planar imaging. J Magn Reson Imaging. 1993;3:755–760. 35. Edelman RR, Li W. Contrast-enhanced echo-planar MR imaging of myocardial perfusion. preliminary study in humans. Radiology. 1994;190:771–777. 36. Wilke N, Jerosch-Herold M, Wang Y, et al. Myocardial perfusion reserve: assessment with multisection, quantitative, first-pass MR imaging. Radiology. 1997;204:373–384. 37. Pauling L, Coryell CD. The magnetic properties and structure of hemoglobin, oxyhemoglobin and carbonmonoxyhemoglobin. Proc Natl Acad Sci U S A. 1936;22:210–216. 38. Thulborn K, Waterton J, Matthews P, Radda G. Oxygenation dependence of the transverse relaxation time of water protons in whole blood at high field. Biochim Biophys Acta. 1982;714:265–270. 39. Li D, Wang Y, Waight DJ. Blood oxygen saturation assessment in vivo using T2* estimation. Magn Reson Med. 1998;39:685–690. 40. Wright GA, Hu BS, Macovski A. Estimating oxygen saturation of blood in vivo with MR imaging at 1.5T. J Magn Reson Imaging. 1991;1:275–283. 41. Brittain JH, Olcott EW, Szuba A, et al. Three-dimensional flowindependent peripheral angiography. Magn Reson Med. 1997;38 (3):343–354. Erratum in: Magn Reson Med. 1998;40(6):948–951. 42. Li KC, Dalman RL, Ch’en IY, et al. Chronic mesenteric ischemia: use of in vivo MR imaging measurements of blood oxygen saturation in the superior mesenteric vein for diagnosis. Radiology. 1997;204:71–77. 43. Nield LE, Qi XL, Valsangiacomo ER, et al. In vivo MRI measurement of blood oxygen saturation in children with congenital heart disease. Pediatr Radiol. 2005;35(2):179–185. 44. Kaul S, Jayaweera AR. Coronary and myocardial blood volumes: noninvasive tools to assess the coronary microcirculation? Circulation. 1997;96:719–724. 45. Bauer WR, Nadler W, Bock M, et al. Theory of the BOLD effect in the capillary region: an analytical approach for the determination of T2 in the capillary network of myocardium. Magn Reson Med. 1999;41:51–62. 46. Chien D, Levin DL, Anderson CM. MR gradient echo imaging of intravascular blood oxygenation: T2* determination in the presence of flow. Magn Reson Med. 1994;32:540–545. 47. Kim SG, Ugurbil K. Comparison of blood oxygenation and cerebral blood flow effects in fMRI: estimation of relative oxygen consumption change. Magn Reson Med. 1997;38:59–65. 48. Kennan RP, Scanley BE, Gore JC. Physiologic basis for BOLD MR signal changes due to hypoxia/hyperoxia: separation of blood volume and magnetic susceptibility effects. Magn Reson Med. 1997;37:953–956. 49. Boxerman JL, Hamberg LM, Rosen BR, Weisskoff RM. MR contrast due to intravascular magnetic susceptibility perturbations. Magn Reson Med. 1995;34(4):555–566. 50. Bauer WR, Nadler W, Bock M, et al. The relationship between the BOLD-induced T2 and T2*: a theoretical approach for the vasculature of myocardium. Magn Reson Med. 1999;42:1004–1010. 51. Belliveau JW, Kennedy DN, Mckinstry RC, et al. Functional mapping of human visual cortex by magnetic resonance imaging. Science. 1991;254:716–719. 52. Kwong KK, Belliveau JW, Chesler DA, et al. Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. Proc Natl Acad Sci U S A. 1992;89:5675–5679.

62. Mangalathu Arumana J, Li D, Dharmakumar R. Deriving bloodoxygen-level-dependent contrast in MRI: an evaluation of T2*weighted, T2-prepared and phase-cycled SSFP methods at 1.5T and 3.0T. Magn Reson Med. 2008;59:561–570. 63. Dharmakumar R, Mangalathu Arumana J, Larson AC, Chung Y, Wright GA, Li D. Cardiac phase-resolved blood oxygen-sensitive steady-state free precession MRI for evaluating the functional significance of coronary artery stenosis. Invest Radiol. 2007;42(3):180–188. 64. Dharmakumar R, Green JD, Flewitt J, et al. Imaging for Probing the Myocardial Perfusion Reserves of Patients with Coronary Artery Disease: A Feasibility Study. Los Angeles, USA: SCMR; 2008. 65. Dharmakumar R, Mangalathu Arumana J, Tang R, Harris K, Zhang Z, Li D. Assessment of regional myocardial oxygenation changes in the presence of coronary artery stenosis with balanced SSFP imaging at 3.0T: theory and experimental evaluation in canines. J Magn Reson Imaging. 2008;27(5):1037–1045. 66. Larson AC, White RD, Laub G, McVeigh ER, Li D, Simonetti OP. Selfgated cardiac cine MRI. Magn Reson Med. 2004;51(1):93–102. 67. Griswold MA, Jakob PM, Heidemann RM, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med. 2002;47:1202–1210. 68. Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: sensitivity encoding for fast MRI. Magn Reson Med. 1999;42:952–962. 69. Sodickson DK, Manning WJ. Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays. Magn Reson Med. 1997;38:591–603. 70. Sodickson DK, McKenzie CA. A generalized approach to parallel magnetic resonance imaging. Med Phys. 2001;28:1629–1643.

Cardiovascular Magnetic Resonance 579

42 MAGNETIC RESONANCE ASSESSMENT OF MYOCARDIAL OXYGENATION

53. Ogawa S, Menon RS, Tank DW, et al. Functional brain mapping by blood oxygenation level-dependent contrast magnetic resonance imaging: a comparison of signal characteristics with a biophysical model. Biophys J. 1993;64:803–812. 54. Lai S, Hopkins AL, Haacke EM, et al. Identification of vascular structures as a major source of signal contrast in high resolution 2D and 3D functional activation imaging of the motor cortex as 1.5T: preliminary results. Magn Reson Med. 1993;30:387–392. 55. McGuinness ME, Talbert RL. Pharmacologic stress testing: experience with dipyridamole, adenosine, and dobutamine. Am J Hosp Pharm. 1994;51:328–346. 56. Massie BM, Schwartz GG, Garcia J, Wisneski JA, Weiner MW, Owens T. Myocardial metabolism during increased work states in the porcine left ventricle in vivo. Circ Res. 1994;74:64–73. 57. Klocke FJ, Li D. Testing coronary flow reserve without a provocative stress: a “BOLD” approach. J Am Coll Cardiol. 2003;41(5):841–842. 58. Westwood M, Anderson LJ, Firmin DN, et al. A single breath-hold multiecho T2* cardiovascular magnetic resonance technique for diagnosis of myocardial iron overload. J Magn Reson Imaging. 2003;18 (1):33–39. 59. Dharmakumar R, Hong J, Brittain J, Plewes DB, Wright GA. Oxygensensitive contrast in blood for steady-state free precession imaging. Magn Reson Med. 2005;53:574–583. 60. Brittain JH, Reeder SB, Shimakawa A, et al. Non-contrast-enhanced angiography at 3T using SSFP and “Dixon” fat-water separation. In: 2004 ISMRM Conference Proceedings (Kyoto, Japan); p. 11. 61. Dharmakumar R, Qi X, Hong J, Wright GA. Detecting microcirculatory changes in blood oxygen state with steady-state free precession imaging. Magn Reson Med. 2006;55:1372–1380.

FUNCTIONAL CARDIOVASCULAR DISEASE

CHAPTER 43

Interventional Cardiovascular Magnetic Resonance Amish N. Raval and Robert J. Lederman

An ideal visual guidance system for cardiovascular, catheter based interventional procedures would offer real time, high resolution, three dimensional (3D) imaging of important ana tomic tissues and chambers, irrespective of respiratory, cardiac, or patient motion. Such tools would quickly enable novel minimally invasive alternatives to open surgical procedures. X ray fluoroscopy (XRF) guides most contemporary cath eter based procedures. However, XRF has important limita tions (Table 43 1). Iodinated radiocontrast, which provides the ability to outline chamber and vascular lumina, is injected only periodically. Tissue detail is minimal. Addition ally, XRF provides only two dimensional (2D) “projection” imaging, with limited depth perception. Iodinated radiocon trast is nephrotoxic in susceptible individuals. Ionizing radiation increases lifetime cancer risk, particularly in chil dren.1–9 Both the patient and the operator as well as in room personnel are exposed. Finally, operators and personnel risk disabling orthopedic injuries from the use of heavy and bulky protective lead apparel.10,11 Ultrasound is sometimes used in combination with certain XRF procedures. For example, transesophageal echocardiogra phy (TEE) provides adjunctive imaging during XRF guided atrial septal defect closure, by assisting with device sizing, posi tioning, and post deployment assessment. Ultrasound offers limited acoustic windows and has “shadowing” artifacts caused by devices. Advances in real time 3D ultrasound may offer interventional imaging guidance in the future,12 although restricted acoustic windows and shadow artifacts will likely remain problematic. Cardiovascular magnetic resonance (CMR) imaging more closely approaches the idealized visual guidance system described earlier. CMR offers superior tissue imaging and the opportunity to acquire images directly in any arbitrary orienta tion (rather than reconstructions), without ionizing radiation or nephrotoxic contrast. Advances in rapid imaging techniques and modified CMR visible interventional devices can now per mit real time CMR guided applications. In the future, the ability to visualize tissue space may enable catheter based procedures that currently require open surgical exposure.

INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE LABORATORY Early real time CMR (RT CMR) systems used low magnetic field strength magnets (0.2 to 0.5 Tesla), open or closed 580 Cardiovascular Magnetic Resonance

bore configurations, and sometimes incorporated XRF units.13,14 Major limitations included a relatively low sig nal to noise ratio (SNR); an inhomogeneous magnetic field, interfering with rapid imaging; and limited patient access when using long, closed bores. Short bore, 1.5 Tesla sys tems with high performance gradients provide excellent field homogeneity and a higher SNR. These scanners also include a large array of radiofrequency (RF) receivers (32 or more), intended for parallel imaging techniques (discussed later), to improve imaging speed. These addi tional receiver channels can also be used to attach “active” catheter devices to improve their visibility (discussed later). In addition, commercial CMR scanners now include in teractive scan user interfaces that increasingly resemble echo cardiography. These interfaces drive sophisticated image reconstruction hardware and software that permit images to be acquired with ease and to be displayed to operators with minimal delay. For investigational laboratories, newer CMR sys tems support better, highly versatile software environments that facilitate rapid development of investigational pulse sequence, image reconstruction, and user interface improvements. Combined XRF and RT CMR laboratories, or XMR labora tories, are probably best for investigational CMR interven tions. In these laboratories, conventional XRF guidance is immediately available for unanticipated CMR system failures or situations requiring emergency bailout. Additionally, early stage CMR guided procedures can be used as an adjunct to established XRF procedures. XMR laboratories are now commercially available. These systems are separated by RF shielded doors and can be used independently or together for combined procedures (Fig. 43 1).15 An auto mated transport table moves patients between the two modal ities. Scan plane and pulse sequences are modified either inside the scanner room or in the external control room. Wave guides and penetration panels are strategically posi tioned in the room so that new electronic communication and monitoring hardware can be installed without disrupting the RF shield system. Images can be displayed inside the scan room with liquid crystal display projectors or shielded liquid crystal display monitors (see Fig. 43 1B), and XRF and CMR datasets can be fused to guide interventions in both arenas.16

Communication and Monitoring Rapid CMR requires rapid gradient switching that causes sub stantial acoustic noise. This noise prevents verbal communi cation between operators, the patient, and staff. RF filtered

Guidance System

Advantages

Disadvantages

RT CMR

Excellent soft tissue imaging Simultaneous display of multiple arbitrary imaging perspectives Real time 3D device tracking Image based physiology assessment No ionizing radiation No lead aprons No nephrotoxic contrast Uninterrupted imaging to detect complications more efficiently

Clinical devices currently unavailable and require hardware attachments Acoustic noise Claustrophobia Rapid emergency response more difficult ECG monitoring for cardiac procedures more difficult Peripheral nerve stimulation from rapidly switching gradients Inferior spatial and temporal resolution in real time mode

XRF

Excellent temporal resolution Widely available in most centers Clinical devices available Simple ECG monitoring Floating table permits remote imaging (i.e., groin site) Greater physician and nurse access to patient

Poor soft tissue imaging 2D “projection” imaging offers limited depth perception Cancer risk from ionizing radiation to patient and room personnel Nephrotoxic contrast required Orthopedic injuries to room personnel from long term use of lead apron apparel Angiography interrupted during device deployment, resulting in detection of complications after contrast injection

Ultrasound

Good soft tissue imaging Excellent temporal resolution Widely available Imaging based physiology assessment (i.e., Doppler)

Limited acoustic windows Device related “shadow” artifacts No tip tracking capability Transesophageal and intracardiac echo procedures are invasive Limited orientation independent imaging

2D, two-dimensional; 3D, three-dimensional; ECG, electrocardiogram; RT-CMR, real-time cardiovascular magnetic resonance imaging; XRF, x-ray fluoroscopy.

Figure 43 1 A, National Heart, Lung and Blood Institute combined X ray/cardiovascular magnetic resonance suite. B, Interventional cardiovascular magnetic resonance suite.

A

headsets (Magnacoustics, Atlantic Beach, NY) with directional fiberoptic microphones (Phone Or, Yahuda, Israel) are used at the U.S. National Heart, Lung and Blood Institute (NHLBI). This versatile system offers multiple communication modes, such as operator control room only mode (the patient cannot hear), operator patient mode, and universal communication mode. Gradient systems are now commercially available that are quieter by 20 dB or more.17 Some laboratories use CMR to create intermittently updated road map image sets and then use nearly silent pulse sequences for tracking catheter position.18 These per mit open verbal communication among staff. Gradients, RF interference, and magnetohydrodynamic effects severely distort the electrocardiogram. Electronic

B

filters enable heart rate monitoring; however, electrocar diographic (ECG) waveform morphology is not interpret able for ST segment monitoring of myocardial ischemia or injury. Multichannel invasive pressure waveform monitor ing, temperature monitoring, and oxygen saturation moni toring are implemented using commercial systems, without modification from XRF laboratories. Interventional catheterization suites require continuous operator monitoring of multiple information streams, includ ing hemodynamics, imaging control, and interventional images. This is also true during interventional CMR proce dures, especially with active devices. At the NHLBI, hemody namic data, the scanner interface, and 3D rendered images are shown separately on a single large projector screen. Cardiovascular Magnetic Resonance 581

43 INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

Table 43-1 Advantages and Disadvantages of Interventional Image Guidance Systems

FUNCTIONAL CARDIOVASCULAR DISEASE

Interventional Cardiovascular Magnetic Resonance Scanner Interface The interventional CMR user interface has several custom features that are making their way into commercial packages. One of the most useful features is the colorized display of individual receiver channels that are attached to “active” catheter receiver coil devices. This makes cathe ter devices readily apparent and has proven extremely use ful in vivo. Other custom features include interactive “projection mode” imaging wherein the slice select feature is disabled and display of catheter devices as they would appear under conventional XRF. Operationally, it is useful to display multiple slices dur ing interventional procedures, both separately and com bined in a 3D rendered image (Fig. 43 2).19 Scanning parameters, including temporal resolution, spatial resolu tion, and field of view, are altered interactively to suit the particular stage of the procedure. Other useful interactive features include digital subtraction during selective

arteriography and ECG gating to “freeze” motion. Other laboratories are experimenting with voice control, auto matic adjustment of slice location to correspond to catheter position, and automatic adaptation of imaging parameters to the speed of catheter manipulation.

Safety Considerations All patient care staff must undergo formal training in CMR safety. For example, the hospital emergency medical response team is unlikely to be familiar with CMR operations, and team members may unwittingly jeopardize themselves, their collea gues, or the patient by rushing into the room with a steel oxy gen tank or other ferromagnetic objects. Rapid patient evacuation from the magnet room must be practiced repeat edly during mock drills. Ferromagnetic objects that cannot be removed from the room must be firmly secured to the wall at all times. Oxygen, inhalational anesthetic, and vacuum for suction can be fed through RF protective wall ports, eliminat ing the need for in room tanks. The RF energy from CMR transmission coils will concen trate on long conductive devices, causing local heating,

Figure 43 2 Commercial interventional magnetic resonance imaging user interface design to accommodate interleaved multi slice real time cardiovascular magnetic resonance image acquisition (three panels on left), a volume rendering of the slices indicating their three dimensional relationship (center panel), “postage stamps” to store and recall important graphic slice prescriptions (bottom row), and interactive scanner parameter control (right panels). (Courtesy of Christine H. Lorenz, PhD, Siemens Medical Solutions, with permission.) 582 Cardiovascular Magnetic Resonance

Interventional Device Factors Length of conductive elements Geometric shape Orientation in the magnet Distance from the radiofrequency transmitter Physical proximity to tissues Insulation Patient Factors Body mass and surface area Convective cooling of intravascular devices by local blood flow General body temperature Implanted conductive devices Tissue thermosensitivity Scanner Factors Field strength Pulse sequence Flip angle Scanning duty cycle Position relative to bore isocenter (closer is better)

potentially leading to burns.20 Table 43 2 lists factors asso ciated with heating, especially long conductive devices and cables. This complex problem is described later. Interven tionists must also take care that connections to intravascular coils as well as surface coils do not inadvertently form loops, which can lead to patient burns.

Real-Time CMR Imaging Rapid CMR is required for invasive procedures, catheter visu alization, and imaging of anatomic structures. Efficient image data sampling methods,21–31 parallel imaging,32–34 and coherent steady state techniques35–40 have enabled real time CMR without significant degradation of image quality in the field of interest. Frame rates of 10 to 15 images/sec or greater are now possible using multichannel (32 or more) RF systems that use parallel imaging techniques to achieve acceleration factors of 3 or more. In addition, interactive, real time color flow imaging may supplement anatomic detail with critical physiologic features, such as leaks or gradients, during thera peutic procedures or interventions.41,42 Slice orientation can automatically follow the tip of catheter devices as they move, known as adaptive imaging, so that the catheter features are kept in view during manipulation. These methods also can be applied automatically to alter scanning parameters such as field of view and temporal resolution.43

Catheter Devices Interventional devices must be clearly and distinctly visible to conduct therapeutic procedures. Conventional XRF devices are generally unsuitable because most incorporate steel braids to increase X ray attenuation for visibility and to enhance catheter performance characteristics, such as steerability, pushability, and trackability. The steel causes severe “blooming” or susceptibility artifacts that lead to artifacts and distort the CMR image (Fig. 43 3). Removing

Figure 43 3 A, Stainless steel braided Kumpe catheter. B, Cardiovascular magnetic resonance (CMR) of the catheter in a water phantom. Note the severe “blooming” signal void artifact in the image, rendering it useless for CMR guided interventional procedures.

these ferrous components usually renders the catheter devices virtually invisible under CMR and usually renders them mechanically unsuitable (floppy) as catheters. Several approaches have been fielded to overcome these problems. Interventional CMR catheter devices are generally classified as passive or active. Passive devices have elements that cause discrete susceptibility imaging artifacts that “darken” images or T1 shortening elements that “brighten” images. Alterna tively, active devices have built in microcoils and electrical cir cuitry that allow the device to act as an RF receiver or transmitter.

Passive Devices Devices That Create Dark Signals Stents44 and guidewires45–48 made of copper, dysprosium, cobalt chromium alloy, nitinol, titanium, and platinum are associated with less severe susceptibility artifacts than those made of iron alloys. They can be coated to improve biocom patibility. Carbon dioxide creates a dark CMR signal by excluding proton spins; carbon dioxide gas has been injected into humans for selective CMR angiography and has been used to fill balloon catheters for diagnostic CMR guided cath eter tracking in patients.49 A passive catheter tracking system based on an optically detuned resonance marker installed on the catheter tip has also been described in a canine model.50 Unfortunately, volume averaging makes it difficult to distin guish passive devices from neighboring anatomic features.

Devices That Create Bright Signals Devices that appear bright are less vulnerable to volume averag ing effects and are more easily visualized against the anatomic background. Dilute gadolinium chelates, such as gadopen tate dimeglumine (Gd DTPA), have been used to fill51 and coat Cardiovascular Magnetic Resonance 583

43 INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

Table 43-2 Factors Influencing Heating-Related Injury by Cardiovascular Magnetic Resonance Interventional Devices

FUNCTIONAL CARDIOVASCULAR DISEASE

catheters52,53 and balloons,54 offering bright device imaging. We have found this passive approach to be inferior in vivo com pared with “active” catheter device techniques (Fig. 43 4). Investigational, off resonance, chemical selective visuali zation using alcohol, F 19,55 or C 13,56 and other hyper polarizing agents has been proposed for catheter tracking. Catheters are filled with compounds (F 19, C 13) that res onate at a frequency other than proton frequency. These novel contrast technologies provide a very high SNR that can be used to enhance catheter visibility with short acting intravascular contrast agents (Fig. 43 5).

Active Devices Highly sensitive, ultra small receiver coils can be incorpora ted into devices to locate them (catheter tracking17,57), visua lize them,58–60 or both.61 (Figs. 43 6 and 43 7) Typically,

these devices incorporate conductive wires along the length of the device. Signals from these devices are displayed on top of previously acquired anatomic road maps (device local ization only) or combined into the array of receivers used to update real time CMR images. The signal receiver can be color coded to distinguish the device position from greyscale background images of the anatomy. Devices displayed in this way are readily visible inside thick slab projections that res emble projection X ray images showing devices in profile. The tip of these devices can be tracked in 3D, which is ideal for intracardiac applications. Alternatively, direct current applied to conductive elements along the device induces magnetic field inhomogeneities, disrupts local signal, and creates dark impressions.62 Unfortunately, the long conductive transmission wires used to connect catheter coils to the CMR transmitter or receiver system are susceptible to RF heating, which may damage neighboring tissue. Multiple approaches to reduce

Figure 43 4 Passive and active device imaging during aortic coarctation repair. Left panel, Nitinol guide wire is positioned retrograde through the aortic valve into the left ventricular (LV) chamber with real time, multi slice and rendered steady state free precession imaging. A partially filled balloon containing dilute gadolinium is positioned across the coarctation (yellow arrow). The wire and the balloon catheter shaft are poorly visualized against the anatomic background. However, the balloon itself is readily apparent. Right panel, An “active” guidewire coil is attached to a separate cardiovascular magnetic resonance receiver channel and the resulting signal is red. Both the gadolinium filled balloon and the anatomic features are more conspicuous compared with solely “passive” catheter devices.

Figure 43 5 Passive device visualization using multispectral (chemical selective) imaging of carbon 13 contrast filled catheters combined with standard proton cardiovascular magnetic resonance in swine (A and B) and for selective renal arteriography (C). (Courtesy of S. Mansson et al., Malmo University Hospital, Sweden. Reproduced with permission of Springer Verlag, GmBH.) 584 Cardiovascular Magnetic Resonance

A

43 INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

Figure 43 6 In vivo, real time active tracking of an electrophysiologically mapping catheter in a three dimensional surface rendered cardiovascular magnetic resonance image of the left ventricle (St. Jude). A shows catheter within the ventricular from “endocardial” perspective; B shows epicardial perspective. Blue dots are individual microcoils along the catheter. (Courtesy of Charles Dumoulin, PhD, General Electric Medical Systems.)

B

Figure 43 7 Active transseptal puncture needle photograph (A) and cardiovascular magnetic resonance image in a water phantom (B) and in vivo across the atrial septum (C).

heating can be combined to make catheter devices safe, such as the use of circuitry to decouple and detune the transmission lines, intermittent chokes and transformers in the transmission lines, and insulation. Another hybrid approach is to incorporate closed loop receiver coils into stents or catheter devices without con necting via transmission lines to the CMR system. These “inductively coupled” devices resonate at predetermined geometric shapes and thereby amplify the local RF sig nal.63,64 Unfortunately, inductively coupled catheter devices cannot easily be displayed in color during RT CMR, as can other actively visualized catheter devices.

Device Solutions for Cardiovascular Applications It is useful to use both active and passive devices during com plex interventional procedures, albeit in animal models. Real time, “active” visualization of the device tip in 3D is desirable during certain procedures, such as transseptal atrial puncture or myocardial wall injection. Similarly, active approaches

prove useful while surveying devices for common failure modes, such as buckling or kinking. Passive approaches unen cumbered by electrical hardware attachments are sufficient for simple procedures, such as placing introducer sheaths that typically do not move once positioned. More importantly, combining passive devices (e.g., balloons filled with dilute Gd DTPA) with active devices (e.g., active guidewire coils) generates wonderful and useful images (see Fig. 43 4).

APPLICATIONS Cardiac Applications Targeted Local Delivery of Cellular Agents to the Myocardium Investigational cell based and other biologic therapies have attracted great interest in the treatment of myocardial and vascular disease. Intravenous, surgical, and catheter based approaches to cell delivery have been used in animal as well as human studies in subjects with myocardial Cardiovascular Magnetic Resonance 585

FUNCTIONAL CARDIOVASCULAR DISEASE

infarction and heart failure. Some investigators believe that precise anatomic targeting, to infarct border zones, for example, may improve the biologic response to cell therapy. If so, CMR guided cell delivery may offer advantages over other imaging approaches, such as electromechanical mapping and intracardiac and 3D surface ultrasound. Minimally invasive procedures involving the myocar dium are particularly amenable to CMR guidance because of the high contrast between myocardium and blood and because of the readily obtained contrast between normal and pathologic myocardial tissue. RT CMR guided endo myocardial cell delivery has achieved millimeter scale pre cision using modified CMR needle catheters in animal models (Fig. 43 8).65–71 Major advantages with this tech nique include high anatomic detail of the myocardium combined with sufficiently rapid imaging to resolve non periodic cardiac and respiratory motion.

Electrophysiology: Atrial and Ventricular Mapping, Ablation, and Transseptal Puncture Therapeutic endomyocardial catheter ablation is widely performed using endomyocardial mapping systems and XRF to abolish atrial and ventricular tachyarrhythmia. In these procedures, a mapping catheter is advanced into the cardiac chambers, guided by endocardial electrogram pat terns to localize the arrhythmia. Key targets are subjected to radiofrequency or cryoablation to create nonconductive zones to abolish the arrhythmia. Because available imaging modalities afford poor visualization of tissue and anatomic structures, these procedures can be challenging and time consuming. Road maps created using previous electromag netic maps, CMR, or computed tomography can be used to “fuse” with updated catheter images72,73; however, these road maps are subject to intrinsic registration errors, non periodic cardiac and respiratory motion,74 alterations in volume as loading conditions change, and catheter induced geometric distortion. One particularly attractive electrophysiologic procedure is catheter treatment of atrial fibrillation by creating lines of ablation to isolate all four pulmonary veins. Even in

experienced hands, these procedures, guided by XRF and electromagnetic mapping, usually require hours of radiation exposure. “Image guided” treatment of atrial fibrillation, conducted under direct surgical exposure, can take minutes. It is tantalizing to speculate that comparable image guided treatment of atrial fibrillation might be afforded by RT CMR guidance without surgical exposure. The complex architec ture of the pulmonary veins, atria, atrial appendices, and ventricles can clearly be visualized with CMR, permitting precise anatomic targeting. RT CMR guidance systems using actively tracked catheters and filtered local electrograms75 are currently under development for use in these procedures.76 Interventional CMR in the electrophysiologic environment would also facilitate early detection of complications.77

Atrial Transseptal Procedures Atrial transseptal puncture is usually conducted as the first step in numerous cardiac procedures, such as pulmonary vein ablation. A needle is advanced from a vein through the right atrium (RA) and into left atrium (LA) across the interatrial sep tum. Currently, this procedure is conducted using subtle XRF visual cues and tactile feedback from sharp catheter devices, with or without adjunctive TEE or intracardiac echocardiogra phy. Poor tissue visualization and limited acquisition win dows, combined with unusual atrial anatomy, can lead to life threatening perforation and pericardial tamponade in as many as 1% to 6% of cases, even in experienced laboratories. Using custom active needles, RT CMR guided atrial trans septal puncture has been performed successfully in swine models with modified transseptal puncture needles and with laser catheters (see Fig. 43 7).78–81 Related therapeutic proce dures, such as closure of atrial septal defects and patent fora men ovale, have been reported using passively visualized, nitinol devices delivered with catheters in swine.82,83

Invasive Coronary Artery Imaging and Intervention Considered by some the “holy grail” of RT CMR guided inter ventions, percutaneous coronary selective angiography,84 Figure 43 8 A, Active endomyocardial injection guide catheter with deployed and undeployed Stiletto (Boston Scientific, Natick, MA) needles. B, Real time cardiovascular magnetic resonance guided endomyocardial delivery of iron labeled cells (yellow arrow) into the border zone of a swine infarct model.

A

586 Cardiovascular Magnetic Resonance

B

Valve Replacement and Repair In aortic or mitral valve replacement, surgical incisions, along with cardiac arrest and cardiopulmonary bypass, are required to resect the native valve structures and implant a prosthetic or homograft valve. Alternative percu taneous approaches to valve replacement and repair are under intense investigation. Early investigational clinical percutaneous aortic valve replacement has been performed with XRF and TEE guidance in adult calcific aortic steno sis.87 Similarly, percutaneous pulmonary valve implantation has been performed in children with acquired and congen ital pulmonary valve disease who previously would have required high risk corrective surgery.88 XRF combined with TEE cannot fully resolve complex valve structure and mechanical device interactions, once again caused by poor tissue characterization, limited imaging windows, and pro blems with device related acoustic shadowing. CMR can fully delineate all important structures in multiple imaging planes and can provide physiologic assessment before and after treatment. Successful RT CMR guided transcatheter prosthetic aortic valve implantation has been demonstrated successfully in swine using passive susceptibility markers89 and active elements.90

Extracardiac Vascular Applications Invasive Arterial Imaging Endovascular “imaging” coils for arterial plaque characteri zation have also been developed and tested in both phan toms and animal models.91–93 The Intercept Internal CMR guidewire coil (Surgi Vision, Gaithersburg, MD) is approved for marketing in the United States. This loopless coil design wire functions as a local receiver antenna, sensitive to excited spins within a radius of only a few millimeters. The feasibility of intravascular MRI has been tested in diseased human arteries, with mixed results.15,94,95 Dick and colleagues found in vivo that, although the procedure was feasible and safe, intravascular CMR with this simple device offered a poor SNR, poor spa tial resolution, and imaging quality that was not improved over that obtained with surface coil imaging and was far inferior to intravascular ultrasound.15 Imaging coil motion

and unbalanced proximity to adjacent tissues resulted in poor and unpredictable contrast. Larger deployable loop, bird cage, or even opposed solenoid96 intravascular designs are under development and are expected to improve the performance of such devices.

Aortic Aneurysm and Aortic Dissection Repair Percutaneous endograft repair of thoracoabdominal aneur ysms and aortic dissection is performed in patients with suitable anatomy who are considered to be at high risk for surgery. Aorta size, proximal and distal landing zones for stents and grafts, and vicinity to crucial arterial branches are vital measurements required for these proce dures. These procedures are typically performed using XRF with adjunctive intravascular ultrasound. Bulky stent or graft devices may distort the native anatomy, preventing operator confidence in preacquired XRF road maps. Ultra sound scatter within stents or endograft offer limited exter nal visualization. RT CMR guided endograft repair of abdominal aortic aneurysm97 and aortic dissection98 has been performed successfully in swine models using active and passive nitinol stents.99 Post procedure assessment using phase contrast flow within and adjacent to the endo graft showed the versatility of CMR guided endograft ther apy (Figs. 43 9 and 43 10).

Aortic Coarctation Stent Repair Advances in transcatheter therapy for many congenital car diovascular conditions have reduced the need for invasive open surgery. Children are particularly sensitive to the harm ful effects of X ray radiation.2–6,100–102 Children with com plex congenital cardiovascular disease often undergo multiple XRF procedures and therefore have greater cumula tive radiation exposure. Radiation sparing procedural guid ance with CMR is attractive and forms the basis for a number of RT CMR guided research applications. Stent repair of aortic coarctation under RT CMR guid ance has successfully been performed in a swine coarcta tion model using commercially available clinical grade devices (see Fig. 43 4).46,103

Transjugular Intrahepatic Portosystemic Shunt The transjugular intrahepatic portosystemic shunt (TIPS) procedure is performed in patients with liver cirrhosis and refractory portal hypertension. A needle/catheter is introduced via the internal jugular vein, traverses the liver parenchyma, and enters the portal vein. Stents are implanted to bridge a connection between the inferior vena cava and the portal vein, thereby relieving portal pressure. Major complications, such as liver capsule perforation lead ing to life threatening hemorrhage, can occur. CMR is ide ally suited to guide these procedures because all tissue structures involved are easily visualized. The TIPS procedure has been performed successfully using combination X ray and MRI and wholly RT CMR in healthy animals104 and humans.105 A hybrid open CMR unit combined with XRF was successfully used to perform TIPS in patients with cirrhosis.105 The investigators reported reduced fluoroscopy time and fewer needle passes compared with conventional XRF guided techniques. Cardiovascular Magnetic Resonance 587

43 INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

angioplasty,85 and stent insertion86 has been reported in large animals. Although this work is impressive, clinical translation of these coronary artery therapeutic procedures is hindered by seemingly insurmountable obstacles. Cur rently, XRF provides spatial resolution of 100 mm at a usual working temporal resolution of 66 msec to manipulate guidewires that are 350 mm wide and stent devices that are 600 mm across. It seems unlikely, barring an unforeseen tech nical breakthrough, that RT CMR can provide comparable spatial and temporal resolution to that required for safe guidewire and catheter manipulation through delicate dis eased human coronary arteries. Similarly, very low profile, distinctly conspicuous, CMR compatible catheter devices would be required for clinical implementation, and such devices are not currently available.

FUNCTIONAL CARDIOVASCULAR DISEASE

0.5 0.4 0.3 0.2 0.2 0 cm/s –0.1 –0.2 –0.3 –0.3 –0.4

Figure 43 9 Real time cardiovascular magnetic resonance guided abdominal aortic aneurysm repair. Before (top left) and after (top center) repair of infrarenal abdominal aortic aneurysms in a swine model using an active endograft (top right) guided by real time, multi slice, and three dimensional rendered imaging. Arrows reflect the distal and proximal stent limits. Figure 43 10 Real time cardiovascular magnetic resonance guided aortic dissection repair. Before (A) and after (C) repair of thoracic aortic dissection in a swine model using a passively visualized stent. B shows the stent delivery system being positioned. Fully deployed stent resolves the dissection (C). (Courtesy of Harold H. Quick, PhD, University Essen Duisberg.)

588 Cardiovascular Magnetic Resonance

through the “invisible” occluded segment may cause perfo ration and hemorrhage. These procedures are often long and require excessive iodinated contrast. Successful RT CMR guided chronic occlusion recanalization and subsequent balloon angioplasty was performed successfully in a swine model of peripheral artery chronic occlusion, using modified “active” wires and support catheters (Fig. 43 11).118,119

Peripheral Artery Disease

Inferior Vena Cava Filter

Several groups have shown the feasibility of CMR guided balloon angioplasty in healthy animals,107,108 animal models of arterial stenosis,109–112 and humans with obstructive peripheral artery disease.113 Dilute Gd DTPA (bright sig nal), undiluted Gd DTPA (dark signal), and carbon diox ide gas have been used to inflate balloons and provide balloon tissue contrast, ensuring full inflation. Active receiver coils, either embedded on the balloon catheter or inserted through the wire port of the balloon filled with dilute Gd DTPA, provide added contrast to adjacent tissue by enhancing further the bright signal within the inflated balloon. Radiopaque markers are typically added to balloon catheters to indicate the “shoulder” points of the balloon to assist with assessment of lesion length before and during deployment. These markers act as small susceptibility markers to assist with positioning and balloon deployment under CMR. Both balloon expandable and self expanding stents are implanted to prevent arterial recoil and alleviate flow limit ing dissection. Both stent designs have been successfully deployed under CMR guidance in animals114–116 and humans.117 Local susceptibility and shielding effects result in imaging voids within and adjacent to stents. Inductively coupled stents (described earlier) may ameliorate this problem.63 Chronic total arterial occlusion recanalization is particu larly challenging under XRF. Only the patent inflow, occluded artery, and patent outflow distal to the artery beyond the obstruction can be visualized with conventional X ray angiography. Traversal of the guidewires and catheter

Inferior vena cava filters are implanted scaffolds, usually self expanding nitinol, designed to entrap migratory venous thromboemboli. They are placed in patients with lower extremity venous thrombosis in whom systemic anticoagulation is contraindicated or unsuccessful in pre venting pulmonary embolism. Accurate positioning of these devices requires adequate visualization of important inflow branches, such as the renal and mesenteric veins, but is straightforward under XRF or ultrasound guidance. Successful RT CMR guided deployment of inferior vena cava filters in animals has been shown with passive imaging techniques.120–122 Concomitant CMR venography and thrombus imaging might have clinical value, especially in follow up assessment.

Figure 43 11 Real time cardiovascular magnetic resonance guided carotid chronic total occlusion recanalization in a swine model. A, Active total occlusion wire and catheter. B, Sagittal real time cardiovascular magnetic resonance showing total occlusion devices (red, wire; green, catheter) traversing the chronic occlusion. Simultaneous axial slice shows the wire tip confined within the arterial wall (yellow arrow). C, The contralateral carotid artery, which is patent, is shown (white arrow). The green arrow shows the occluded artery segment distal to the guidewire tip, and the yellow arrow shows the guidewire tip within the artery lumen.

A

B

CONCLUSION The combination of RT CMR and CMR visible devices may offer a complete imaging solution for therapeutic cardiovas cular interventions. The many advantages over existing guidance modalities include superior tissue imaging, no need for ionizing radiation or iodinated contrast, imaging based physiology assessment, and 3D perspective. Impor tant challenges remain to the clinical translation of RT CMR, especially the requirement for conspicuous, commer cial grade catheter devices. Minimally invasive and novel therapeutic interventions, once considered impossible with traditional imaging, may now be possible using this rapidly evolving technology.

C

Cardiovascular Magnetic Resonance 589

43 INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

Mesocaval shunt is a variant of the TIPS procedure in which systemic veins are connected directly to the portal vein without traversing the liver. Guided entirely by RT CMR, Arepally and associates showed successful application of mesocaval shunt in healthy swine using modified “active” transseptal puncture needles and self expanding stents.106 The investigators hope to conduct a range of novel proce dures using RT CMR access to the portal circulation.

FUNCTIONAL CARDIOVASCULAR DISEASE

References 1. American Academy of Pediatrics Committee on Environmental Health. Risk of ionizing radiation exposure to children: a subject review. Pediatrics. 1998;101(4 Pt 1): 717–719. 2. Raissaki M. Pediatric radiation protection. Eur Radiol Suppl. 2004;14 (suppl 1):74–83. 3. Committee on the Biological Effects of Ionizing Radiation BoRER. Commission on Life Sciences, National Research Council. Health effects of exposure to low levels of ionizing radiation, BEIR V. Washington, DC: National Academy Press; 1990. 4. Brenner D, Elliston C, Hall E, Berdon W. Estimated risks of radiation-induced fatal cancer from pediatric CT. AJR Am J Roentgenol. 2001;176(2):289–296. 5. Infante-Rivarde C, Mathonnet G, Sinnet D. Risk of childhood leukemia associated with diagnostic irradiation and polymorphisms in DNA repair genes. Environ Health Perspect. 2000;108:495–498. 6. Modan B, Keinan L, Blumstein T, Sadetzki S. Cancer following cardiac catheterization in childhood. Int J Epidemiol. 2000;29(3):424–428. 7. Martin EC, Olson AP, Steeg CN, Casarella WJ. Radiation exposure to the pediatric patient during cardiac catheterization and angiocardiography: emphasis on the thyroid gland. Circulation. 1981;64(1): 153–158. 8. Wu JR, Huang TY, Wu DK, Hsu PC, Weng PS. Radiation exposure of pediatric patients and physicians during cardiac catheterization and balloon pulmonary valvuloplasty. Am J Cardiol. 1991;68(2):221–225. 9. Berrington de Gonzalez A, Darby S. Risk of cancer from diagnostic X-rays: estimates for the UK and 14 other countries. Lancet. 2004;363(9406):345–351. 10. Ross AM, Segal J, Borenstein D, Jenkins E, Cho S. Prevalence of spinal disc disease among interventional cardiologists. Am J Cardiol. 1997;79(1):68–70. 11. Goldstein JA, Balter S, Cowley M, Hodgson J, Klein LW. Occupational hazards of interventional cardiologists: prevalence of orthopedic health problems in contemporary practice. Catheter Cardiovasc Interv. 2004;63(4):407–411. 12. Roman KS, Nii M, Golding F, Benson LN, Smallhorn JF. Images in cardiovascular medicine: real-time subcostal 3-dimensional echocardiography for guided percutaneous atrial septal defect closure. Circulation. 2004;109(24):e320–e321. 13. Adam G, Neuerburg J, Bucker A, et al. Interventional magnetic resonance: initial clinical experience with a 1.5-tesla magnetic resonance system combined with c-arm fluoroscopy. Invest Radiol. 1997;32 (4):191–197. 14. Silverman SG, Jolesz FA, Newman RW, et al. Design and implementation of an interventional MR imaging suite. AJR Am J Roentgenol. 1997;168(6):1465–1471. 15. Dick AJ, Raman VK, Raval AN, et al. Invasive human magnetic resonance imaging: feasibility during revascularization in a combined XMR suite. Catheter Cardiovasc Interv. 2005;64(3):265–274. 16. Gutierrez LF, Silva R, Ozturk C, et al. Technology preview: x-ray fused with magnetic resonance during invasive cardiovascular procedures. Catheter Cardiovasc Interv. 2007;70:773–782. 17. Hennel F, Girard F, Loenneker T. “Silent” MRI with soft gradient pulses. Magn Reson Med. 1999;42(1):6–10. 18. Dumoulin CL, Souza SP, Darrow RD. Real-time position monitoring of invasive devices using magnetic resonance. Magn Reson Med. 1993;29(3):411–415. 19. Guttman MA, Ozturk C, Raval AN, et al. Interventional cardiovascular procedures guided by real-time MR imaging: an interactive interface using multiple slices, adaptive projection modes and live 3D renderings. J Magn Reson Imaging. 2007;26(6):1429–1435. 20. Atalar E. Radiofrequency safety for interventional MRI procedures(1). Acad Radiol. 2005;12(9):1149–1157. 21. Guttman MA, Kellman P, Dick AJ, Lederman RJ, McVeigh ER. Realtime accelerated interactive MRI with adaptive TSENSE and UNFOLD. Magn Reson Med. 2003;50(2):315–321. 22. Madore B, Glover GH, Pelc NJ. Unaliasing by fourier-encoding the overlaps using the temporal dimension (UNFOLD), applied to cardiac imaging and fMRI. Magn Reson Med. 1999;42(5):813–828. 23. van Vaals JJ, Brummer ME, Dixon WT, et al. “Keyhole” method for accelerating imaging of contrast agent uptake. J Magn Reson Imaging. 1993;3(4):671–675.

590 Cardiovascular Magnetic Resonance

24. Peters DC, Lederman RJ, Dick AJ, et al. Undersampled projection reconstruction for active catheter imaging with adaptable temporal resolution and catheter-only views. Magn Reson Med. 2003;49(2):216–222. 25. Rasche V, de Boer RW, Holz D, Proksa R. Continuous radial data acquisition for dynamic MRI. Magn Reson Med. 1995;34(5):754–761. 26. Schaeffter T, Rasche V, Bornert P, Mens G. Interactive reduced FOV imaging for projection reconstruction and spiral acquisition. Magn Reson Imaging. 2001;19(5):677–684. 27. Boll DT, Merkle EM, Seaman DM, et al. Comparison of ECG-gated rectilinear vs. real-time radial K-space sampling schemes in cine True-FISP cardiac MRI. J Cardiovasc Magn Reson. 2004;6(4):793–802. 28. Rasche V, Holz D, Schepper W. Radial turbo spin echo imaging. Magn Reson Med. 1994;32(5):629–638. 29. Duerk JL, Lewin JS, Wu DH. Application of keyhole imaging to interventional MRI: a simulation study to predict sequence requirements. J Magn Reson Imaging. 1996;6(6):918–924. 30. Harvey PR. A keyhole echo planar hybrid imaging technique. MAGMA. 1999;9(1–2):16–20. 31. Hwang KP, Lim J, Wendt M, Merkle E, Lewin JS, Duerk JL. Improved device definition in interventional magnetic resonance imaging using a rotated stripes keyhole acquisition. Magn Reson Med. 1999;42 (3):554–560. 32. Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: sensitivity encoding for fast MRI. Magn Reson Med. 1999;42(5):952–962. 33. Sodickson DK, Manning WJ. Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays. Magn Reson Med. 1997;38(4):591–603. 34. Griswold MA, Jakob PM, Heidemann RM, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med. 2002;47(6):1202–1210. 35. Oppelt A. FISP—a new fast MRI sequence. Electromedica. 1986;54:15–18. 36. Busse RF, Kruger DG, Debbins JP, Fain SB, Riederer SJ. A flexible view ordering technique for high-quality real-time 2DFT MR fluoroscopy. Magn Reson Med. 1999;42(1):69–81. 37. Debbins JP, Riederer SJ, Rossman PJ, et al. Cardiac magnetic resonance fluoroscopy. Magn Reson Med. 1996;36(4):588–595. 38. Farzaneh F, Riederer SJ, Lee JN, Tasciyan T, Wright RC, Spritzer CE. MR fluoroscopy: initial clinical studies. Radiology. 1989;171 (2):545–549. 39. Rasche V, Holz D, Proksa R. MR fluoroscopy using projection reconstruction multi-gradient-echo (prMGE) MRI. Magn Reson Med. 1999;42(2):324–334. 40. Riederer SJ, Tasciyan T, Farzaneh F, Lee JN, Wright RC, Herfkens RJ. MR fluoroscopy: technical feasibility. Magn Reson Med. 1988;8(1):1–15. 41. Riederer SJ, Wright RC, Ehman RL, et al. Real-time interactive color flow MR imaging. Radiology. 1991;181(1):33–39. 42. Nayak KS, Pauly JM, Kerr AB, Hu BS, Nishimura DG. Real-time color flow MRI. Magn Reson Med. 2000;43(2):251–258. 43. Elgort DR, Wong EY, Hillenbrand CM, Wacker FK, Lewin JS, Duerk JL. Real-time catheter tracking and adaptive imaging. J Magn Reson Imaging. 2003;18(5):621–626. 44. Wang Y, Truong TN, Yen C, et al. Quantitative evaluation of susceptibility and shielding effects of nitinol, platinum, cobalt-alloy, and stainless steel stents. Magn Reson Med. 2003;49(5):972–976. 45. Buecker A, Spuentrup E, Schmitz-Rode T, et al. Use of a nonmetallic guide wire for magnetic resonance-guided coronary artery catheterization. Invest Radiol. 2004;39(11):656–660. 46. Raval AN, Telep JD, Guttman MA, et al. Real-time magnetic resonance imaging-guided stenting of aortic coarctation with commercially available catheter devices in swine. Circulation. 2005;. 47. Wacker FK, Reither K, Branding G, Wendt M, Wolf KJ. Magnetic resonance-guided vascular catheterization: feasibility using a passive tracking technique at 0.2 Tesla in a pig model. J Magn Reson Imaging. 1999;10(5):841–844. 48. Kochli VD, McKinnon GC, Hofmann E, von Schulthess GK. Vascular interventions guided by ultrafast MR imaging: evaluation of different materials. Magn Reson Med. 1994;31(3):309–314. 49. Razavi R, Hill DL, Keevil SF, et al. Cardiac catheterisation guided by MRI in children and adults with congenital heart disease. Lancet. 2003;362(9399):1877–1882.

74. Noseworthy PA, Malchano ZJ, Ahmed J, Holmvang G, Ruskin JN, Reddy VY. The impact of respiration on left atrial and pulmonary venous anatomy: implications for image-guided intervention. Heart Rhythm. 2005;2(11):1173–1178. 75. Susil RC, Yeung CJ, Halperin HR, Lardo AC, Atalar E. Multifunctional interventional devices for MRI: a combined electrophysiology/MRI catheter. Magn Reson Med. 2002;47(3):594–600. 76. Nazarian S, Kolandaivelu A, Zviman MM, et al. Feasibility of realtime magnetic resonance imaging for catheter gjuidance in electrophysiology studies. Circulation. 2008;118:223–229. 77. Ritter O, Nordbeck P, Fidler F, et al. Interventional electrophysiology studies and ablation therapy under real time MRI guidance: early detection of complications [abstract]. Circulation. 2008;118:S693. 78. Arepally A, Karmarkar PV, Weiss C, Rodriguez ER, Lederman RJ, Atalar E. Magnetic resonance image-guided trans-septal puncture in a swine heart. J Magn Reson Imaging. 2005;21(4):463–467. 79. Raval AN, Karmarkar P, Ozturk C, et al. Real-time MRI guided atrial septal puncture and balloon septostomy in swine. Catheter Cardiovasc Interv. 2006;67(4):637–643. 80. Elagha AA, Kocaturk O, Guttman MA, et al. Real-time MR imagingguided laser atrial septal puncture in swine. J Vasc Interv Radiol. 2008;19:1347–1353. 81. Elagha AA, Kocaturk O, Guttman MA, Ozturk C, Kim AH, Burton GW, Kim JH, Raman V, Raval AN, Wright VJ, Schenke WH, McVeigh ER, Lederman RJ. Real time MR imaging-guided laser atrial septal puncture in swine. J Vasc Interv Radiol. 2008;19(9):1347–1353. 82. Rickers C, Jerosch-Herold M, Hu X, et al. Magnetic resonance imageguided transcatheter closure of atrial septal defects. Circulation. 2003;107(1):132–138. 83. Buecker A, Spuentrup E, Grabitz R, et al. Magnetic resonance-guided placement of atrial septal closure device in animal model of patent foramen ovale. Circulation. 2002;106(4):511–515. 84. Omary RA, Green JD, Schirf BE, Li Y, Finn JP, Li D. Real-time magnetic resonance imaging-guided coronary catheterization in swine. Circulation. 2003;107(21):2656–2659. 85. Serfaty JM, Yang X, Foo TK, Kumar A, Derbyshire A, Atalar E. MRIguided coronary catheterization and PTCA: a feasibility study on a dog model. Magn Reson Med. 2003;49(2):258–263. 86. Spuentrup E, Ruebben A, Schaeffter T, Manning WJ, Gunther RW, Buecker A. Magnetic resonance–guided coronary artery stent placement in a swine model. Circulation. 2002;105(7):874–879. 87. Cribier A, Eltchaninoff H, Bash A, et al. Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description. Circulation. 2002;106(24):3006–3008. 88. Bonhoeffer P, Boudjemline Y, Qureshi SA, et al. Percutaneous insertion of the pulmonary valve. J Am Coll Cardiol. 2002;39(10):1664–1669. 89. Kuehne T, Yilmaz S, Meinus C, et al. Magnetic resonance imagingguided transcatheter implantation of a prosthetic valve in aortic valve position: feasibility study in swine. J Am Coll Cardiol. 2004;44 (11):2247–2249. 90. McVeigh ER, Guttman MA, Lederman RJ, et al. Real-time interactive MRI-guided cardiac surgery: aortic valve replacement using a direct apical approach. Magn Reson Med. 2006;56(5):958–964. 91. Kandarpa K, Jakab P, Patz S, Shoen F, Jolesz F. Prototype miniature endoluminal MR imaging catheter. J Vasc Interv Radiol. 1993; (4):419–427. 92. Botnar RM, Bucker A, Kim WY, Viohl I, Gunther RW, Spuentrup E. Initial experiences with in vivo intravascular coronary vessel wall imaging. J Magn Reson Imaging. 2003;17(5):615–619. 93. Martin AJ, Henkelman RM. Intravascular MR imaging in a porcine animal model. Magn Reson Med. 1994;32(2):224–229. 94. Hofmann LV, Liddell RP, Arepally A, Montague B, Yang X, Bluemke DA. In vivo intravascular MR imaging: transvenous technique for arterial wall imaging. J Vasc Interv Radiol. 2003;14(10):1317–1327. 95. Larose E, Yeghiazarians Y, Libby P, et al. Characterization of human atherosclerotic plaques by intravascular magnetic resonance imaging. Circulation. 2005;112(15):2324–2331. 96. Hillenbrand CM, Elgort DR, Wong EY, et al. Active device tracking and high-resolution intravascular MRI using a novel catheter-based, opposed-solenoid phased array coil. Magn Reson Med. 2004;51 (4):668–675. 97. Raman VK, Karmarkar PV, Guttman MA, et al. Real-time magnetic resonance-guided endovascular repair of experimental abdominal aortic aneurysm in swine. J Am Coll Cardiol. 2005;45(12):2069–2077.

Cardiovascular Magnetic Resonance 591

43 INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

50. Weiss S, Kuehne T, Brinkert F, et al. In vivo safe catheter visualization and slice tracing using an optically detonable resonant marker. Magn Reson Med. 2004;52:860–868. 51. Omary RA, Unal O, Koscielski DS, et al. Real-time MR imagingguided passive catheter tracking with use of gadolinium-filled catheters. J Vasc Interv Radiol. 2000;11(8):1079–1085. 52. Unal O, Korosec FR, Frayne R, Strother CM, Mistretta CA. A rapid 2D time-resolved variable-rate k-space sampling MR technique for passive catheter tracking during endovascular procedures. Magn Reson Med. 1998;40(3):356–362. 53. Strother CM, Unal O, Frayne R, et al. Endovascular treatment of experimental canine aneurysms: feasibility with MR imaging guidance. Radiology. 2000;215(2):516–519. 54. Bakker CJ, Hoogeveen RM, Weber J, van Vaals JJ, Viergever MA, Mali WP. Visualization of dedicated catheters using fast scanning techniques with potential for MR-guided vascular interventions. Magn Reson Med. 1996;36(6):816–820. 55. Kozerke S, Hegde S, Schaeffter T, Lamerichs R, Razavi R, Hill DL. Catheter tracking and visualization using 19F nuclear magnetic resonance. Magn Reson Med. 2004;52(3):693–697. 56. Mansson S, Johansson E, Magnusson P, et al. 13C imaging: a new diagnostic platform. Eur Radiol. 2006;16(1):57–67. 57. Ackerman JL, Offutt MC, Buxton RB, Brady TJ. ISMRM. Rapid 3D tracking of small RF coils. In: Proc 5th Annual Meeting, ISMRM; 1986:1131. 58. McKinnon GC, Debatin JF, Leung DA, Wildermuth S, Holtz DJ, von Schulthess GK. Towards active guidewire visualization in interventional magnetic resonance imaging. MAGMA. 1996;4(1):13–18. 59. Ocali O, Atalar E. Intravascular magnetic resonance imaging using a loopless catheter antenna. Magn Reson Med. 1997;37(1):112–118. 60. Ladd ME, Zimmermann GG, Quick HH, et al. Active MR visualization of a vascular guidewire in vivo. J Magn Reson Imaging. 1998;8 (1):220–225. 61. Kocaturk O, Saikus CE, Guttman MA, Faranesh AZ, Ratnayaka K, Ozturk C, McVeigh ER, Lederman RJ. Whole shaft visibility and mechanical performance for active MR catheters using copper-nitinol braided polymer tubes. J Cardiovasc Magn Reson. 2009;11(1):29. 62. Glowinski A, Adam G, Bucker A, Neuerburg J, Vanvaals JJ, Gunther RW. Catheter visualization using locally induced actively controlled field inhomogeneities. Magn Reson Med. 1997; (38):253–258. 63. Quick HH, Kuehl H, Kaiser G, Bosk S, Debatin JF, Ladd ME. Inductively coupled stent antennas in MRI. Magn Reson Med. 2002;48 (5):781–790. 64. Quick HH, Zenge MO, Kuehl H, et al. Interventional magnetic resonance angiography with no strings attached: wireless active catheter visualization. Magn Reson Med. 2005;53(2):446–455. 65. Dick AJ, Guttman MA, Raman VK, et al. Magnetic resonance fluoroscopy allows targeted delivery of mesenchymal stem cells to infarct borders in Swine. Circulation. 2003;108(23):2899–2904. 66. Lederman RJ, Guttman MA, Peters DC, et al. Catheter-based endomyocardial injection with real-time magnetic resonance imaging. Circulation. 2002;105(11):1282–1284. 67. Kraitchman DL, Heldman AW, Atalar E, et al. In vivo magnetic resonance imaging of mesenchymal stem cells in myocardial infarction. Circulation. 2003;107(18):2290–2293. 68. Corti R, Badimon J, Mizsei G, et al. Real time magnetic resonance guided endomyocardial local delivery. Heart. 2005;91(3):348–353. 69. Krombach GA, Pfeffer JG, Kinzel S, Katoh M, Gunther RW, Buecker A. MR-guided percutaneous intramyocardial injection with an MR-compatible catheter: feasibility and changes in T1 values after injection of extracellular contrast medium in pigs. Radiology. 2005;235(2):487–494. 70. Saeed M, Lee R, Martin A, et al. Transendocardial delivery of extracellular myocardial markers by using combination X-ray/MR fluoroscopic guidance: feasibility study in dogs. Radiology. 2004;231(3):689–696. 71. Rickers C, Gallegos R, Seethamraju RT, et al. Applications of magnetic resonance imaging for cardiac stem cell therapy. J Interv Cardiol. 2004;17(1):37–46. 72. Sermesant M, Rhode K, Sanchez-Ortiz GI, et al. Simulation of cardiac pathologies using an electromechanical biventricular model and XMR interventional imaging. Med Image Anal. 2005;9(5):467–480. 73. Reddy VY, Malchano ZJ, Holmvang G, et al. Integration of cardiac magnetic resonance imaging with three-dimensional electroanatomic mapping to guide left ventricular catheter manipulation: feasibility in a porcine model of healed myocardial infarction. J Am Coll Cardiol. 2004;44(11):2202–2213.

FUNCTIONAL CARDIOVASCULAR DISEASE

98. Eggebrecht H, Kuhl H, Kaiser GM, et al. Feasibility of real-time magnetic resonance-guided stent-graft placement in a swine model of descending aortic dissection. Eur Heart J. 2006;27(5):613–620. 99. Eggebrecht H, Heusch G, Erbel R, Ladd ME, Quick HH. Real-time vascular interventional magnetic resonance imaging: the future of aortic stent-graft placement? Basic Res Cardiol. 2007;102:1–8. 100. American Academy of Pediatrics. Committee on Environmental Health. Risk of ionizing radiation exposure to children: a subject review. Pediatrics. 1998;101(4 Pt 1):717–719. 101. McLaughlin JR, Kreiger N, Sloan MP, Benson LN, Hilditch S, Clarke EA. An historical cohort study of cardiac catheterization during childhood and the risk of cancer. Int J Epidemiol. 1993;22(4):584–591. 102. Spengler RF, Cook DH, Clarke EA, Olley PM, Newman AM. Cancer mortality following cardiac catheterization: a preliminary follow-up study on 4,891 irradiated children. Pediatrics. 1983;71(2):235–239. 103. Krueger JJ, Ewert P, Yilmaz S, et al. Magnetic resonance imagingguided balloon angioplasty of coarctation of the aorta: a pilot study. Circulation. 2006;113:1093–1100. 104. Kee ST, Rhee JS, Butts K, et al. Becker Young Investigator Award. MR-guided transjugular portosystemic shunt placement in a swine model. J Vasc Interv Radiol. 1999;10(5):529–535. 105. Kee ST, Ganguly A, Daniel BL, et al. MR-guided transjugular intrahepatic portosystemic shunt creation with use of a hybrid radiography/ MR system. J Vasc Interv Radiol. 2005;16(2):227–234. 106. Arepally A, Karmarkar PV, Weiss C, Atalar E. Percutaneous MR imaging-guided transvascular access of mesenteric venous system: study in swine model. Radiology. 2006;238(1):113–118. 107. Wildermuth S, Dumoulin CL, Pfammatter T, Maier SE, Hofmann E, Debatin JF. MR-guided percutaneous angioplasty: assessment of tracking safety, catheter handling and functionality. Cardiovasc Intervent Radiol. 1998;21(5):404–410. 108. Feng L, Dumoulin CL, Dashnaw S, et al. Feasibility of stent placement in carotid arteries with real-time MR imaging guidance in pigs. Radiology. 2005;234(2):558–562. 109. Yang X, Bolster Jr BD, Kraitchman DL, Atalar E. Intravascular MRmonitored balloon angioplasty: an in vivo feasibility study. J Vasc Interv Radiol. 1998;9(6):953–959. 110. Buecker A, Adam GB, Neuerburg JM, et al. Simultaneous real-time visualization of the catheter tip and vascular anatomy for MR-guided PTA of iliac arteries in an animal model. J Magn Reson Imaging. 2002;16(2):201–208.

592 Cardiovascular Magnetic Resonance

111. Godart F, Beregi JP, Nicol L, et al. MR-guided balloon angioplasty of stenosed aorta: in vivo evaluation using near-standard instruments and a passive tracking technique. J Magn Reson Imaging. 2000;12 (4):639–644. 112. Omary RA, Frayne R, Unal O, et al. MR-guided angioplasty of renal artery stenosis in a pig model: a feasibility study. J Vasc Interv Radiol. 2000;11(3):373–381. 113. Paetzel C, Zorger N, Bachthaler M, et al. Feasibility of MR-guided angioplasty of femoral artery stenoses using real-time imaging and intraarterial contrast-enhanced MR angiography. Rofo. 2004;176 (9):1232–1236. 114. Buecker A, Neuerburg JM, Adam GB, et al. Real-time MR fluoroscopy for MR-guided iliac artery stent placement. J Magn Reson Imaging. 2000;12(4):616–622. 115. Dion YM, Ben El Kadi H, Boudoux C, et al. Endovascular procedures under near-real-time magnetic resonance imaging guidance: an experimental feasibility study. J Vasc Surg. 2000;32(5):1006–1014. 116. Mahnken AH, Gunther RW, Tacke J. Radiofrequency ablation of renal tumors. Eur Radiol. 2004;14(8):1449–1455. 117. Manke C, Nitz WR, Djavidani B, et al. MR imaging-guided stent placement in iliac arterial stenoses: a feasibility study. Radiology. 2001;219(2):527–534. 118. Raval AN, Karmarkar PV, Guttman MA, et al. Real-time MRI guided endovascular recanalization of chronic total arterial occlusion in a swine model [In Press]. Circulation. 2006;. 119. Sampath S, Raval AN, Lederman RJ, McVeigh ER. High-resolution 3D arteriography of chronic total peripheral occlusions using a T1W turbo spin-echo sequence with inner-volume imaging. Magn Reson Med. 2007;57:40–49. 120. Frahm C, Gehl HB, Lorch H, et al. MR-guided placement of a temporary vena cava filter: technique and feasibility. J Magn Reson Imaging. 1998;8(1):105–109. 121. Bartels LW, Bos C, van Der Weide R, Smits HF, Bakker CJ, Viergever MA. Placement of an inferior vena cava filter in a pig guided by high-resolution MR fluoroscopy at 1.5 T. J Magn Reson Imaging. 2000;12(4):599–605. 122. Bucker A, Neuerburg JM, Adam GB, et al. Real-time MR guidance for inferior vena cava filter placement in an animal model. J Vasc Interv Radiol. 2001;12(6):753–756.

Pediatric Interventional Cardiovascular Magnetic Resonance Sanjeet R. Hegde and Reza S. Razavi

The last two decades have seen phenomenal advances made in the field of cardiovascular magnetic resonance (CMR), and these advances have fueled research into inter ventional applications for this remarkable imaging modal ity.1,2 Conventional X ray fluoroscopically guided cardiac catheterization and interventions carry a substantial risk of exposure to ionizing radiation for both patients and staff. This is particularly relevant in younger patients, who are often required to undergo multiple procedures. The need for an imaging modality that offers multiplanar imaging, superior structural delineation of complex cardiac anatomy, and additional physiologic information, without the risk of ionizing radiation, has brought CMR guidance to the fore. In the last 4 years, clinical programs using CMR guided cardiac catheterization have started and show promise.3 After the first MR images showing live human anatomy were produced,4–6 this technique evolved to enable a variety of clinical applications of MR.7,8 Over the years, improvements in signal detection, fast data handling, advanced understand ing of spin systems, pulse sequences, and artifact suppression have resulted in much faster scan times and considerable improvements in image resolution.9–17 These ultrafast imaging techniques form the basis of real time imaging, used for CMR guided cardiac catheterization. However, the first important step in making CMR cardiac catheterization a clinical reality is the design of a suitable interventional CMR system.18,19

INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE SYSTEMS In the design of an interventional CMR suite, it is important to retain the full capabilities of a state of the art diagnostic scan ner without encumbering the interventionalist or creating a risk of high radiofrequency (RF) or switched magnetic field exposure. Open magnet designs allow easier access to the patient, but typically are not available in field strengths higher than 1 Tesla (T). The cylindrical horizontal bore systems offer higher field strengths and gradient slew rates, allowing higher resolution imaging, shorter scan times, higher signal to noise ratio, reduced image distortion, and improved functionality with real time imaging, all of which are of paramount impor tance when endovascular interventions are considered.20

A trade off with the traditional cylindrical magnet design is access to the patient. More recently, magnets with shorter bores and flared margins have been introduced and offer bet ter patient access, especially for cardiovascular interventions, without compromising the advanced CMR features of diag nostic scanners. Rapid improvements in the processing power of computers, along with the use of powerful and intuitive software, have allowed researchers to develop novel strategies for image data acquisition and reconstruction. It is now possi ble to achieve frame rates of as high as 20 images/sec with the aid of new parallel imaging techniques while maintaining suit able spatial resolution for interventional applications.21–24 Despite the inherent potential and promise of CMR guided interventions and operations, there are still major obstacles associated with performing the complete procedure in the CMR scanner, particularly because of the lack of CMR compatible catheters and devices. Therefore, the immediate future of interventional CMR lies in exploiting multi modality imaging, such as X ray and CMR (XMR) or XMR and ultra sound. Such hybrid units already in existence allow the use of separate modalities or a combination of them when needed. Cross modality image integration, with spatial and temporal information about the anatomy, pathology, and therapy devices, can be provided to the users of these systems. A good example is the XMR system, which combines X ray and CMR by having both modalities in the same room, with a tabletop design that allows patients to be moved from one modality to the other in less than 1 minute (Fig. 44 1).25–28

MERITS OF CARDIOVASCULAR MAGNETIC RESONANCE GUIDANCE Improved Visualization of Cardiac Anatomy A problem with X ray guided cardiac catheterization is the inherent poor contrast of soft tissues, such as the heart and great vessels. This makes it difficult for the cardiologist to manipulate or position guidewires, catheters, balloons, or interventional devices within the heart and surrounding Cardiovascular Magnetic Resonance 593

44 PEDIATRIC INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

CHAPTER 44

FUNCTIONAL CARDIOVASCULAR DISEASE

Staff

Patient

A

B Figure 44 1 A, Schematic room plan of a typical X ray and magnetic resonance (XMR) suite. B, XMR room with the X ray and MR equipment joined by a movable tabletop. The c arm of the X ray unit is seen in the foreground, ceiling mounted MR monitor and controls are seen in the distance, and the 5 gauss area is demarcated by a change in the floor coloring from the MR to the X ray end of the room.

vessels. A skilled operator usually relies on recognizing anatomic structures from previous experience or on con trast angiographic images acquired earlier in the procedure. The lack of adequate visualization increases the risk of per forating the heart or great vessels, especially when performing complex interventional procedures. Certain interventional cardiac procedures involve selec tion of an appropriate cardiac device and its successful deployment within the heart, which requires accurate mea surement of the size of defects and nearby anatomic struc tures. Such measurements are possible under XRF, but can be difficult. A successful interventional cardiac procedure therefore relies heavily on adequate visualization of the heart or ves sel. This implies the need for superior imaging methods that provide excellent visualization without increasing the risk to the patient. This role fits CMR very well because it provides exceptional structural delineation of both the heart and its surrounding vasculature and therefore allows safe guidance of interventional procedures. 594 Cardiovascular Magnetic Resonance

Reduced Ionizing Radiation There is a pressing need for pediatric cardiac catheteriza tion procedures to be made safer, especially in terms of ionizing radiation. According to the U.K. National Radia tion Protection Board, the mean risk that a solid tumor will develop as a result of a single cardiac catheterization procedure is approximately 1 in 2500 in adults. This risk increases to 1 in 1000 in children if exposure occurs at 5 years of age.29–32 Also, the proportion of the body that is irradiated increases as the size of the patient decreases, and some procedures in patients with congenital heart disease often require much longer X ray exposure. These risks are multiplied in children in particular because they often undergo multiple cardiac catheter procedures. In addition to the patients, there is also a significant risk from ionizing radiation to the staff in the catheter lab oratory during these procedures, despite the use of pro tective shields.33,34

Cardiac catheterization is used not only to provide ana tomic information and perform intervention but also to obtain functional information. Invasive pressures and blood gases are used to calculate systemic and pulmonary blood flow and resistance with the Fick principle. Cine angiography is also used to assess global ventricular func tion as well as regional wall motion abnormalities. The functional information obtained at cardiac catheterization is used alongside anatomic information to assess patient suitability for surgery or interventional cardiac catheteriza tion or the need for long term vasodilator therapy in patients with pulmonary vascular disease. The Fick principle to quantify flow is dependent on multiple measurements (hemoglobin, aortic/pulmonary artery oxygen saturation, partial pressure, oxygen con sumption), which can be a considerable source of inaccu racy. In addition, in patients with large intracardiac shunts and high pulmonary blood flow, accuracy is further reduced.35–40 Therefore, there is a need for a method of flow quantification that allows accurate and reproducible measurement of pulmonary vascular resistance (PVR). Velocity encoded phase contrast CMR enables noninvasive quantification of blood flow in major vessels. Cardiac out put and the pulmonary to systemic flow ratio (Qp:Qs) measured using this technique have been shown to be accurate.41–47 In addition, phase CMR has been validated in numerous phantom experiments, allowing for a novel method of quantification of PVR in patients with pulmo nary hypertension by using invasive pressure measure ments and MR flow data.48–50 Assessment of global and regional ventricular function can also be carried out much more accurately with cine steady state free precession (SSFP) cardiovascular MR than with X ray angiography. When using cardiovascular MR for assessing global ventricular function, there is no need to make assumptions about cardiac geometry, unlike with XRF or even echocardiography. This is particularly impor tant when assessing right ventricular function (RV) and regional wall motion. Finally, combining invasive pressure measurement with CMR derived blood flow and ventricular volumes also opens up interesting new ways of looking at pathophysiol ogy. It allows for the study of pulmonary vascular compli ance, derived ventricular pressure volume loops, and assessment of load independent ventricular function.51,52

MAGNETIC INSTRUMENTATION AND VISUALIZATION STRATEGIES Crucial to the success of interventional CMR is real time tracking and visualization of catheters, guidewires, and devices in the CMR environment. Several groups around the world are putting considerable effort into developing CMR suitable catheters and devices. Device localization under CMR is made possible by a variety of approaches that can be broadly classified as either electrically passive or electrically active.53

Passive Catheter Tracking and Visualization The passive tracking technique is commonly based on visu alization of susceptibility artifacts or signal voids caused by the interventional device under CMR imaging. This is a well studied technique and to date it is the most clinically feasible (see Fig. 44 3).54–58 Passive visualization often does not require any special hardware or software and therefore it can be performed on any commercial CMR system. The ideal passive tracking catheter or guidewire must be made of a material that provides adequate torque and allows tracking, but does not obscure the underlying anatomy. Ferromagnetic materials cause large susceptibility artifacts and therefore are not generally suitable for CMR guided procedures. This rules out most metals used for making cardiac devices. However, certain alloys, such as nitinol (nickel and titanium), have magnetic susceptibility close to that of tissue. Therefore, they are best suited for making guidewires and braided catheters that are MR com patible but not necessarily CMR safe. The polymeric mate rials used for making catheters typically have low magnetic susceptibility and therefore cannot be easily localized on CMR images.59 This implies that, if materials with higher susceptibility can be incorporated into the wall of the catheters or sheaths or the lumen filled with a suitable con trast agent, then improved visualization can be achieved. One approach to generating susceptibility artifacts is locally impregnating the catheter wall with gadolinium like compounds, such as dysprosium oxide, in the form of rings or along the length of the catheter during the extrusion process (Fig. 44 2C).58 Another approach is to use gadolin ium contrast agents in varying concentrations within cath eter lumens60 or impregnated into catheter walls to create either a positive or negative signal on CMR imaging.61 Metallic devices and guidewires produce susceptibility arti facts that aid visualization by way of the artifacts, but differ ent metals behave differently under CMR. Titanium alloys produce narrower artifacts compared with ferromagnetic or even certain other nonferromagnetic alloys, such as nickel chromium, which can produce large RF and suscep tibility artifacts. Guidewires with a fiberglass core and non metallic guidewires made of resin microparticle compound covered by polytetrafluoroethylene have been used for MR guided interventions in animals.62,63 In the case of balloon angiographic catheters, if the bal loon is inflated with carbon dioxide, as is done convention ally with X ray, then the inflated balloon creates a signal void in the CMR image, thus enabling visualization (see Fig. 44 2A and B). This method has been used successfully to guide catheters in patients under CMR (Fig. 44 3).3,64 Although this technique allows easy visualization of the tip, the length is impossible to visualize because the signal void from the catheter length is masked by volume averag ing and dephasing effects of thicker slices.65 The success of passive visualization also relies on dedi cated scan techniques. A dynamic gradient echo sequence, such as SSFP, has been shown to be ideal for passive cathe ter tracking, especially when signal voids or susceptibility artifacts are used for visualization.64,66 Cardiac catheteriza tion under XRF guidance is usually performed at imaging Cardiovascular Magnetic Resonance 595

44 PEDIATRIC INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

Physiologic Information

FUNCTIONAL CARDIOVASCULAR DISEASE

A

B

C

Figure 44 2 Passive tracking. A, Inflated balloon angiographic Bermann catheter filled with 0.8 mL CO2. B, CO2 filled balloon catheter manipulated in a phantom. C, Dysprosium catheter: a catheter impregnated with dysprosium oxide is manipulated in an in vitro set up mimicking endovascular intervention. The catheter is clearly visualized along its full length, despite being orientated along B0. (A, Courtesy of Arrow International, Reading, PA.)

speeds of 25 to 30 frames/sec. The frame rates available for CMR guided interventions are not comparable because of the post processing of CMR images and their subsequent display, allowing a maximum of 10 to 14 frames routinely. Some of the proposed passive catheter tracking techniques require image subtraction or positive contrast to improve visualization of markers on the catheter, which means that, along with faster scan techniques, faster image processing algorithms are required.57,67–70

Active Catheter Tracking and Visualization The active catheter tracking and visualization method uses an electrical connection to the CMR scanner, and localiza tion or tracking of the device requires the device itself, along with any additional hardware or software that comes with it. Typically, the device is equipped with a coil or an antenna that functions in either receive only mode or trans mit/receive mode. Active catheters that are used as receivers have a coil or an antenna that receives signal from tissue in its immediate vicinity.71 These devices do not transmit signal into the patient, but rely on the body coil to transmit into the patient. The signal received by these coils can then be used to pinpoint their position, for imaging of local tissue, or both. There are two important types of active catheters: those based on small coils positioned, for example, at the end of a catheter, and those based on a loopless antenna that can run along a catheter or can be made into a guidewire (Fig. 44 4).72–76 In addition, active designs in 596 Cardiovascular Magnetic Resonance

which signal voids along the catheter are created by electri cally controlled magnetic field inhomogeneities have also been investigated.77 A small resonant coil at the tip of a catheter can be iden tified by a series of three one dimensional projections along each axis.71 This can be done quickly (in three repetition times) and so could be repeated for very fast update of the catheter position, allowing real time tracking of the catheter. The position of the catheter could then be pro jected over a previously acquired road map. Recently, simi lar techniques have been combined with fast/real time sequences, imaging the heart or vessels using surface coils, and the combined (interleaved) sequence has allowed simultaneous localization of the catheter and imaging of the surrounding tissue. Further adaptation of these sequences has allowed automatic changing of the imaging plane to match the change in the position of the catheter. Another recent development of active catheter tracking by the group at the U.S. National Institutes of Health allows the visualization of two simultaneously acquired planes as well as visualization of the catheter or device positions in real time, thus reducing the major problem of the catheter moving through the plane when only one imaging plane is visualized.78–80 The great advantage of these active systems is that loca tion of the catheter is unambiguous. Active visualization has great potential because it allows the whole length of the catheter or guidewire to be visualized and the imaging plane to be adapted to the moving catheter automatically. It may even allow high resolution imaging of a small area of interest, such as a plaque in the vessel, when the coil or antenna is used in its imaging mode.81 However, the main disadvantage is concern with safety.82–87

44 PEDIATRIC INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

Figure 44 3 Manipulation of carbon dioxide filled balloon catheter (arrows) from the inferior vena cava to the right pulmonary artery using solely magnetic resonance guidance. Real time interactive images: repetition time 2.9 msec, echo time 1.45 msec, flip angle 45
, matrix 128  128, field of view 250 to 350, and temporal resolution 10 to 14 frames/sec. Arrows show the signal void of the catheter tip as it traverses the inferior vena cava, right atrium, tricuspid valve, and right ventricular outflow tract and enters the pulmonary artery.

These devices use intravascular coils as RF antennas, and the connection to the external circuits via a long wire in the strong magnetic field makes induction of an electrical cur rent and heating possible. There have been recent develop ments to overcome this risk, such as electrical decoupling of loopless antennas and the use of optical coupling and long fiber optic connections.88 An innovative active catheter design that uses miniaturized transformers showed no

significant RF heating and holds promise for a safe transmis sion line for interventional applications (Fig. 44 5).89 Another promising approach to device localization is what some authors refer to as semi active catheter tracking, implying passive localization of an electrically isolated res onant coil.53 These resonant coils locally enhance B1 and signal reception so that, for very low global flip angles, the signal from the fiducial is prominent.90–95 Cardiovascular Magnetic Resonance 597

FUNCTIONAL CARDIOVASCULAR DISEASE

Figure 44 4 Active catheter designed for intramyocardial injection. (Courtesy of Dr. Parag Karmarkar, Johns Hopkins University, Baltimore, MD.)

The resonant coils can be interrogated by gradient echo sequences, such as SSFP with low flip angles. Catheters with multiple resonant coils can be tracked easily compared with passive catheters and have a relatively better safety profile compared with some of the active catheter designs (Fig. 44 6). Catheter visualization and localization using 19 F CMR in conjunction with proton imaging appears to be a promising alternative to existing methods that either are associated with safety concerns if active markers are used or have insufficient direction dependent contrast if passive visualization is used (Fig. 44 7).96 Other multispectral CMR methods under investigation are catheter tracking and angiography using hyperpolarized gases.97,98

SAFETY ISSUES Bioeffects of Magnetic Fields The patient undergoing a CMR scan typically is exposed to three forms of electromagnetic radiation: static magnetic field, gradient magnetic field, and RF electromagnetic field. These can cause bioeffects at significantly high exposure levels. A health care worker in such a setting can also be exposed to electromagnetic fields, although exposure is more chronic and intermittent. However, numerous studies have shown no substantial risks to patients from the elec tromagnetic fields used in clinical CMR scanners.99–102 The risks to the health care worker, especially in a CMR setting, are fiercely debated, but the consensus is that more work needs to be carried out before occupational electro magnetic field exposure limits can be set.103,104 Further more, the bioeffects specifically related to the use of interventional CMR have not yet been fully investigated. Many reports in the literature regarding the bioeffects of static magnetic fields are conflicting. There is no strong evi dence to suggest that there are any significant cardiac or neurologic effects from static magnetic fields of less than 2 T. In addition, several studies have shown that high static magnetic fields do not significantly alter skin and body temperature.105–110 598 Cardiovascular Magnetic Resonance

Gradient magnetic fields can induce electrical fields and current in conductive media, including biologic tis sue, according to Faraday law of induction. The thermal effects of switched magnetic fields are considered negligi ble and are not believed to be clinically significant. Elec trical stimulation of the retina is believed to cause magnetophosphenes, which are completely reversible, with no known residual side effects. Some volunteers have also reported experiencing a metallic taste and vertigo while undergoing imaging within 4 T magnets. These bioeffects caused by gradient fields are unusual in fields of less than 2 T.111 The exposure limits for RF radiation are set in terms of specific absorption rate in Wkg 1, which is the mass nor malized rate at which RF power is coupled with biologic tissue. The main bioeffects associated with exposure to RF radiation relate to the generation of heat in tissues. Contro versially, some researchers have reported that electromag netic fields cause cancer and developmental abnormalities in animal models. However, the efficiency and absorption pattern of RF radiation is mainly determined by the physi cal dimensions of the tissue in relation to the incident wavelength, which implies that laboratory animal experi ments cannot be simply scaled or extrapolated to humans.112–114

Heating and Electrical Safety of Interventional Equipment The heating of wires, devices, implants, and other instru ments is an important safety issue that is holding back the rapid advance of interventional CMR. Heating as a result of RF radiation occurs by three mechanisms, accord ing to Maxwell’s theory of electromagnetism.82 When a conductive device or instrument is moved through a magnetic field, small “rings” of current are induced that are called eddy currents and create internal magnetic fields opposing the change. The kinetic energy that goes into driving the eddy currents inside the metal will give off that energy as heat. Therefore, intravascular guidewires or device delivery systems with a metal core are unsafe in the CMR environment, with documented heating up to 74
C (165
F) of the tip.82,83,115,116 Electromagnetic induction heating has often been blamed for thermal injuries caused by monitoring cables used in CMR. RF electromagnetic fields and time varying gradient magnetic fields can induce voltage in conductive media and cause current to flow. The circulating currents cause power loss by heating that is referred to as induction heating. A loop in a monitoring cable would increase the inductance of the circuit; therefore, larger currents would be induced, resulting in greater heating of the cable.87,117,118 If a circuit is in a resonant state, then there is maximum current induction such that significant electromagnetic induction heating occurs. Lengths of wire, for example, can behave as RF antennas that capture electromagnetic waves to extract power from them. The electromagnetic waves that enter the antennas have electrical charges and corresponding currents associated with them. When the antenna is approximately half a wavelength long, resonance

Coaxial Transformer

Transformer loop

A

Passive markers

Cross at tip coil

B Figure 44 5 A, Safe transmission line for active catheter tracking created with integrated miniaturized transformers. B, To evaluate the transformer concept for active tracking in vivo, a 6 Fr catheter was built for catheterization of the arterial and venous system of a swine. The catheter is seen being manipulated in the heart by the active tracking method. (Courtesy of Dr. Steffen Weiss, Philips Research, Hamburg, Germany.)

occurs and the electrical energy remains confined to the immediate vicinity of a given antinode. Hence, the highest electrical field of the antennas is believed to be at the tip. The electrical properties of the media surrounding the

antennas and the operating frequency also determine the wavelength.85,86 Newer designs of wires and cables aimed at reducing heating are currently being investigated, along with novel RF shielding technologies.119,120 Cardiovascular Magnetic Resonance 599

44 PEDIATRIC INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

6F = 2 mm

FUNCTIONAL CARDIOVASCULAR DISEASE

A

B

C Figure 44 6 A, A 5 Fr balloon angiographic catheter with six prewound fiducial markers mounted onto the surface was manipulated in a 20 mm polyethylene tube taped to the chest of a volunteer. A real time spoiled gradient echo sequence (fast field echo [FFE]: repetition time 2.3 msec, echo time 1.2 msec, flip angle 50
, slice thickness 20 mm) followed by an interactive FFE sequence with interleaving of scans with flip angles of 2
and 50
and a frame rate of 4 frames/sec was used. All six markers are visualized along the length of the catheter. B, Distal end of a 6 Fr catheter with an integrated self resonant radiofrequency circuit. C, The active wireless catheter is shown being guided with real time projection reconstruction steady state free procession imaging into the celiac trunk. (A, Courtesy of Arrow International, Reading, PA. B and C, Courtesy of Dr. Harald H. Quick, University of Essen, Germany.)

Magnetic Force and Torque In addition to the bioeffects of CMR and heating and elec trical safety of interventional devices, a significant risk to interventional procedures is magnetic force and torque exerted by the magnetic field on metallic devices.121,122 Conventional guidewires made of ferromagnetic materi als, such as stainless steel, and catheters with metallic braid ing, are inherently unsafe for use in the CMR environment. Interventional devices that are ferromagnetic will be subject 600 Cardiovascular Magnetic Resonance

to both deflection force (translational movement) and torque (rotational movement); therefore, they cannot be used for procedures within a CMR scanner. Hence, all CMR imaging facilities must have safeguards to ensure that ferromagnetic objects are not brought into the vicinity of the magnet. However, there are certain other metallic alloys, such as nitinol, that are CMR compatible. They produce minimal susceptibility artifacts and are not affected by the magnetic field in terms of deflection force and torque. This is an important consideration in developing suitable catheters and guidewires for use in interventional CMR procedures.

44 PEDIATRIC INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

Figure 44 7 Corresponding 1H (top left) and 19F (top right) images of a 7 Fr catheter containing PFOB. With a simple peak search algorithm in the image space, the catheter tip position was extracted and two orthogonal 19F projections were used to determine the position of the catheter tip (þ), as shown. Source: Kozerke S, Hedge S, Schaeffter T, Lamerichs R, Razavi R, Hill DL. Catheter tracking and visualization using 19F nuclear magnetic resonance. Magn Reson Med. 2004;52(3):693 697.

X-RAY AND CARDIOVASCULAR MAGNETIC RESONANCE GUIDANCE X-Ray and Cardiovascular Magnetic Resonance Facility Design The room design of a typical XMR facility is shown in Figure 44 1. There are many design features that make this room different from standard CMR facilities. The XMR suite is designed so that half of the room is out side the 5 gauss line of the magnet, permitting the use of tra ditional instruments and devices as well as echocardiography and RF ablation equipment when required. A movable

tabletop allows patients to be moved easily between modal ities in less than 60 seconds. The paramount consideration in the design, construction, and operation of an XMR facility is safety, and a comprehensive safety protocol must be drawn up to minimize possible hazards (Table 44 1). Traditionally, CMR scans are planned and conducted from the control room, away from the magnet and the patient. However, during CMR guided cardiac catheteriza tion, there is a need for real time changes to the scanning sequence parameters to follow catheter manipulation in the heart and great vessels. Also, the imager needs to have a clear view of the CMR images while performing the procedure. Therefore, it is useful to have a fully functional set of ceiling mounted, movable screens and scanner controls within the CMR scanner room that can be placed at either end of the bore of the scanner, in close proximity to the patient. The XMR suite includes appropriate CMR compatible anesthetic equipment and monitoring equipment for inva sive pressure monitoring via the catheter. A great deal of thought has been given to the safety of patients under Cardiovascular Magnetic Resonance 601

FUNCTIONAL CARDIOVASCULAR DISEASE

Table 44-1 X-Ray and Cardiovascular Magnetic Resonance (CMR) Facility: Safety Features Compulsory safety training of all CMR interventional staff Specially designed clothes without pockets Safety officer restricting entry to the main room during XMR intervention Clear demarcation of ferromagnetic safe and unsafe areas within the room CMR compatible anesthetic and monitoring equipment Noiseproof headphone systems for all staff within the room X ray and radiofrequency shielded scrub room Positive pressure air handling and filtration system Tethering of all ferromagnetic equipment to the wall or floor Safety checks whenever a patient is transferred between X ray and CMR to ensure that metallic instruments used for catheterization are not taken across to the CMR end of the room Written log of all safety infringements and regular review of safety procedures

anesthesia, especially during the transfer between the X ray and CMR tables. All of the anesthetic and monitoring tubing and lines are designed with extra length and are secured to the movable tabletop to ensure smooth patient transfer. The electrocardiogram (ECG) and invasive pressure data are sent from the MR compatible monitoring equipment via an optical network to a computer in the control room, where the cardiac technician is stationed. The appropriate measure ment and recording of the data is made in the usual way. The technician has access to monitors that show the appropriate X ray or CMR images of the procedure. The imagers in the room can view the CMR images and any monitoring data (e.g., ECG, invasive pressure data). Blood samples taken dur ing the procedure are labeled in the room and passed to the technician in the control room via a wave guide. Reliable and accurate ECG synchronization is essential for CMR and in particular CMR guided cardiac catheterization. When catheters are manipulated in the heart, there is the potential to cause arrhythmias (tachyarrhythmia or heart block). It is therefore important to perform accurate monitor ing of the cardiac rhythm at all times during XMR catheteriza tion. Obtaining a reliable ECG in the magnet, particularly during some CMR sequences, can be difficult. The magneto hydrodynamic effect and gradient noise can seriously disturb the ECG signal.123,124 This interference can reduce trigger signal reliability to less than 40%. Vector electrocardiogram (VCG) is a QRS detection algorithm that automatically adjusts to the actual electrical axis of the patient’s heart and the spe cific multidimensional QRS waveform. In our experience, this greatly improves the reliability of R wave detection to nearly 100%. A reliable R wave, with the P and T waves that are also always clearly seen with VCG, allows detection of nearly all arrhythmias. Unfortunately, there are no ECG systems that can reliably provide ST segment or T wave morphologic infor mation. In the future, using signal processing techniques, it may be possible to obtain ECG during CMR scanning that provides ST and T wave information reliably. Another complication of performing cardiac catheteriza tion under CMR guidance is the noise generated during scan ning. There is a headphone and microphone system in the room that reduces the noise, but allows staff to communicate with each other in both the scanner and control rooms. Some CMR coils have X ray visible components and would need to be removed between CMR imaging and 602 Cardiovascular Magnetic Resonance

X ray imaging of patients. It is therefore necessary to have specifically designed coils that are sufficiently radiotranslu cent to be left in place during XRF without deterioration of image quality. We use these coils in our procedures so that patients do not have to be disturbed when moving from one imaging modality to the other.125 The XMR suite has positive pressure air handling and filtration appropriate for a catheterization laboratory. There is a scrub room that is also RF and X ray shielded and can be accessed both from the XMR suite and control room. This room acts as an RF lock, allowing access to the XMR suite during CMR scanning.

Performing X-Ray and Cardiovascular Magnetic Resonance Interventions In a typical XMR interventional procedure, after the induc tion of anesthesia, the patient is transferred from an MR compatible trolley to the CMR end of the XMR facility and positioned on the CMR scanner tabletop (Fig. 44 8A). The monitoring and anesthetic equipment are attached. A three lead ECG, separate from the VCG, is used for cardiac monitoring during MR scanning. The VCG electrodes are placed on the subcostal margin, outside the X ray field of view, and the VCG is used for triggering CMR scans. An MR compatible pulse oximeter and nonin vasive blood pressure monitoring equipment are also attached. The exhaled anesthetic gases are monitored for end tidal carbon dioxide as well as the concentration of the volatile anesthetic agents. Flexible phase array RF coils are used. These coils are relatively X ray lucent and thus do not need to be removed between MR and X ray imaging. The patient is then placed in the CMR scanner, and a multi breath hold three dimensional (3D) SSFP scan of the heart and great vessels (echo time 2, repetition time 4, flip angle 50, 80 to 120 slices reconstructed to 1 mm cubic vox els) is obtained.126 Using an interactive SSFP sequence (8 to 10 frames/sec), with real time manipulation of scan para meters, the likely imaging planes needed for subsequent catheter tracking, ventricular function, and flow quantifica tion are stored. The patient is then transferred to the X ray end of the room. Draping and vascular access are carried out as for routine cardiac catheterization; in addition, a sec ond large drape is placed over the patient (see Fig. 44 8B). The patient is transferred back to the MR scanner after safety checks are performed, including an operating the ater style check of all metallic objects used under X ray. The second drape is then lifted up and taped to the top of the magnet, which in effect provides sterile draping of the bore and sides of the magnet (see Fig. 44 8C). An end hole or side hole balloon angiographic catheter (4 to 7 Fr) is placed in the sheath, and with the balloon inflated with CO2 (see Fig. 44 3), the catheter tip is passively visua lized using the interactive sequences described earlier. The previously stored imaging planes are used, along with inter active slice selection, to track the catheter. Because only the tip of the catheter is visualized, care is taken not to push the catheter too fast and thus beyond the CMR imaging plane. This also ensures that the catheter does not acciden tally form loops and possible knots.

Early Experience in Humans

B

C Figure 44 8 X ray and magnetic resonance intervention. A, Patient is placed on the magnetic resonance tabletop. B, Patient is slid across to the X ray half of the room for sheath insertion. C, Passive catheter manipulation is performed under magnetic resonance guidance.

A duplicate CMR control console is positioned next to the bore of the magnet so that the interactive window can be easily visualized while the catheter is being manipulated. Therefore, this procedure requires two experienced opera tors, one to move the catheter and one to alter the CMR imaging planes to ensure that the catheter tip is tracked, using the real time interactive sequence.

In our center, we have performed 53 diagnostic CMR guided cardiac catheterizations. In most patients, the majority of the procedure was carried out under MR guidance, which allowed for a significant reduction of overall X ray dose.3,50 We used CMR to assess pulmonary vascular resistance in the patients because it allowed for simulta neous measurement of pulmonary arterial flow and inva sive pressures. We found moderate to good agreement between the Fick method and the CMR method of deriv ing PVR at baseline conditions. However, in the presence of nitric oxide, which is used to assess pulmonary vasor eactivity, there was less agreement between the two meth ods. There was not only worsening in agreement but also a large bias when PVR was measured in the presence of 100% oxygen and nitric oxide. We believe that this is the result of errors in the Fick method rather than the XMR method, which has important implications for patient management. This novel MR technique may prove to be a more accurate method to quantify PVR in humans; it also offers reduced exposure to ionizing radiation.50,127 Currently, RF ablation is used to treat some patients with symptomatic atrial or ventricular tachyarrhythmia. This is con ventionally carried out under XRF or ultrasound guidance using electrical and electromagnetic mapping. XRF guidance is conventionally used to guide such procedures because it offers excellent temporal resolution and good visualization of catheters. However, as a projection imaging modality, more than one view is necessary to gain an appreciation of the 3D location and path of catheters. This implies moving the X ray c arm to obtain different projections. A few centers use a biplane X ray system for the same purpose. The anatomic con text of the acquired images can be difficult to interpret because soft tissues, such as the heart and blood vessels, are not visible during X ray exposure. Therefore, we developed a real time XMR guidance system for cardiovascular interventions that allows the use of both CMR and X ray imaging for guidance, thereby overcoming some of the failings of exclusive XRF guidance.128 Cardiovascular Magnetic Resonance 603

44 PEDIATRIC INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

A

Once the catheter is positioned in the desired vessel or chamber, appropriate pressure data and saturation/blood gas samples are obtained, as for routine cardiac catheteriza tion. In addition, ventricular function (short axis balanced SSFP) and flow (phase contrast) scans can be performed using the appropriate previously stored imaging planes. If catheter manipulation into a particular heart chamber or vessel using CMR guidance alone is difficult, the patient is transferred back to the X ray end of the room, where cathe terization can be continued under XRF (e.g., to use a guide wire or a braided catheter). The patient can be transferred back to the CMR scanner for further CMR measurements once the catheter is positioned satisfactorily. An interventional procedure or RF ablation of arrhyth mias requires part of the procedure to be performed under X ray fluoroscopy because the ablation catheters and deliv ery devices are not MR compatible. Therefore, MR imaging is performed at the beginning of the procedure for planning purposes, used in guiding the procedure, and performed again at the end of the procedure for evaluation.

FUNCTIONAL CARDIOVASCULAR DISEASE

A combination of calibration and real time tracking is used to achieve X ray and CMR image registration. A Northern Digital Optotrak 3020 (NDI, Ontario, Canada) using infrared emitting diodes optically tracks the X ray c arm and the X ray table. The sliding table is tracked by the CMR system software while docked to the CMR scanner and becomes part of the X ray table when docked to the X ray system. Once calibrated, the system allows registration of 3D CMR image acquisitions to X ray image acquisitions. During intervention, the guidance system can provide a real time MR anatomy overlay onto X ray images. One monitor is used to display the control interface and the

A

second monitor shows the image overlay. During fluoros copy, the system acquires X ray images and computes the registration matrix from the tracking data at 10 frames/sec and updates the overlay display at 3 frames/sec. Using this unique XMR technology, we have carried out RF ablation in pulmonary veins, atria, and ventricles to treat arrhythmias successfully in 30 patients (Fig. 44 9).128 This CMR to X ray registration method also allows us to relate the posi tion of measured electrophysiology data to cardiac motion data from 3D CMR images. The XMR technology is also being used to perform stent implantation in patients with coarctation of the aorta (Fig. 44 10).

B

C

Figure 44 9 A and B, Biplane X ray views of the linear ablating catheter in the left atrial roof position. C, Posterior three dimensional view of the left atrium derived from a gadolinium cardiovascular magnetic resonance angiography scan. The green dots show the mapped locations of the linear ablating catheter in three positions: (1) left atrial roof position; (2) left upper pulmonary vein to mitral valve annulus position; and (3) right upper pulmonary vein to mitral valve annulus position.

A

Prestent

B

Poststent

Figure 44 10 Cardiovascular magnetic resonance angiography image superimposed onto the X ray cardiac catheter image during stent implantation. A, Undilated stent and guidewire across the coarctation site. B, The combined images show that the implanted open stent lies in a satisfactory position, distal to the origin of the left subclavian artery and across the coarctation narrowing. Stent implantation was performed in the X ray half of the X ray and magnetic resonance facility. Magnetic resonance imaging was used before stent insertion to acquire the three dimensional cardiovascular magnetic resonance angiography images and after the procedure (guidewires removed) to confirm satisfactory position of the stent and relief of aortic obstruction. 604 Cardiovascular Magnetic Resonance

Recent Progress Several groups have shown the immense potential of inter ventional CMR in animal models. The interventions that were shown to be feasible with passive and active catheter techniques include balloon angioplasty of arterial ste noses,129–134 stenting of vessels,92,135–138 and atrial septal puncture/septostomy.139,140 Device closure of atrial septal defects is another application that has been explored.141–144 CMR guided percutaneous pulmonary and aortic valve stent implantation have also been performed successfully (Fig. 44 11).135,145 More complex interventions, such as percutaneous coro nary catheterization and intervention, have also been demonstrated in healthy animals using CMR.146–149 Recently, the group in Berlin performed balloon dilation of aortic coarctation in patients under CMR guidance.134 Although this required some modification of the standard technique, with the likely arrival of new CMR compatible guidewires and catheters, it should soon be possible to per form a number of interventions under CMR guidance in patients with congenital heart disease.

Future Directions Novel catheters and guidewires have made possible tar geted intramyocardial injection of progenitor stem cells in myocardial infarction in animal models.80,150,151 Using real time CMR and direct apical access in porcine hearts, prosthetic aortic valves were implanted in the beating heart.152 This breakthrough application may allow CMR guidance of minimally invasive extra anatomic bypass and beating heart valve repair. MR guidance of intra myocardial gene therapy is another exciting field.153

A

Three dimensional electromechanical models of the heart have been created that allow simulation of cardiovascular pathologies to test therapeutic strategies and plan inter ventions (Fig. 44 12).154,155 Newer catheter tracking techniques using inductively coupled RF coils or hyper polarized 13C and visualization strategies using novel k space sampling also hold promise.156–158

Figure 44 12 Patient undergoing X ray and cardiovascular magnetic resonance guided biventricular pacing. Composite image showing one slice of a cardiovascular magnetic resonance cardiac anatomic scan with a superimposed surface model of the left ventricle. Cardiac electrical modeling was used to estimate myocardial conductivity for the left ventricle. The conductivity is represented by the color coding, with blue showing areas of low conductivity and yellow showing areas of normal conductivity. The white region shows the area of scarring segmented from late enhancement magnetic resonance imaging. There is good correspondence with predicted low conductivity and the region of the scar.

B

Figure 44 11 Percutaneous aortic valve stent implantation in a swine before (A) and after (B) valve stent implantation under cardiovascular magnetic resonance. (Courtesy of Dr. Titus Kuehne, German Heart Institute, Berlin, Germany.) Cardiovascular Magnetic Resonance 605

44 PEDIATRIC INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

INTERVENTIONAL CARDIAC APPLICATIONS

FUNCTIONAL CARDIOVASCULAR DISEASE

CONCLUSION As outlined in this chapter, CMR guidance of cardiac cath eterization is feasible and has been shown to be safe. The potential benefits of this new technique include reduction of X ray dose, accurate assessment of pulmonary vascular resistance, and better visualization of complex anatomy for both diagnostic cardiac catheterization and interven tional cardiac catheterization, such as RF ablation of arrhythmias, using XMR images. However, a number of issues must be resolved before exclusive interventional CMR becomes routine clinical practice. These include practical issues, such as reduction of noise in the CMR environment, improved access to the patient, and better CMR compatible patient monitoring equipment. There is also a pressing need for catheter and device manufacturers to produce tools specifically designed for CMR guided cardiac catheterization. This need, combined with the cost associated with installing expensive XMR suites, is holding back the rapid advance of interventional CMR. However, the potential benefits of 3D anatomic guidance for interventional cardiologists, radiologists, and surgeons, including the useful additional

physiologic information and the ability to assess tissue response to therapy with CMR, makes this remarkable imaging modality unique and one that offers great prom ise for safe guidance of complex cardiovascular interventions.79

ACKNOWLEDGMENTS Some of the work described in this chapter was performed at Guy’s Hospital, London, United Kingdom, by a team of academic and clinical staff. The authors thank Tobias Schaeffter, Vivek Muthurangu, Marc Miquel, Kawal Rhode, Redha Boubertakh, Andrew Taylor, Aaron Bell, Caroline Kehoe, Victoria Parish, Maxime Sermesant, Derek Hill, Ste phen Keevil, David Hawkes, Jas Gill, Shakeel Qureshi, Eric Rosenthal, Gerald Greil, Philipp Beerbaum, and Edward Baker in the Departments of Imaging Sciences, Pediatric and Adult Cardiology. We would also like to acknowledge Michael Barnet and other members of the Anesthetic Department; and John Spence, Stephen Sinclair, and Rebecca Lund and other staff from the Radiology Depart ment who have provided considerable support.

References 1. Jolesz FA, Blumenfeld SM. Interventional use of magnetic resonance imaging. Magn Reson Q. 1994;10(2):85–96. 2. Lufkin RB. Interventional MR imaging. Radiology. 1995;197(1): 16–18. 3. Razavi R, Hill DL, Keevil SF, Miquel ME, Muthurangu V, Hegde S, et al. Cardiac catheterisation guided by MRI in children and adults with congenital heart disease. Lancet. 2003;362 (9399):1877–1882. 4. Mansfield P, Maudsley AA. Line scan proton spin imaging in biological structures by NMR. Phys Med Biol. 1976;21(5):847–852. 5. Mansfield P, Maudsley AA. Medical imaging by NMR. Br J Radiol. 1977;50(591):188–194. 6. Damadian R. Tumor detection by nuclear magnetic resonance. Science. 1971;171(976):1151–1153. 7. Pykett IL, Newhouse JH, Buonanno FS, Brady TJ, Goldman MR, Kistler JP, et al. Principles of nuclear magnetic resonance imaging. Radiology. 1982;143(1):157–168. 8. Henderson RG. Nuclear magnetic resonance imaging: a review. J R Soc Med. 1983;76(3):206–212. 9. Riederer SJ, Tasciyan T, Farzaneh F, Lee JN, Wright RC, Herfkens RJ. MR fluoroscopy: technical feasibility. Magn Reson Med. 1988;8(1): 1–15. 10. Farzaneh F, Riederer SJ, Lee JN, Tasciyan T, Wright RC, Spritzer CE. MR fluoroscopy: initial clinical studies. Radiology. 1989;171(2): 545–549. 11. Wright RC, Riederer SJ, Farzaneh F, Rossman PJ, Liu Y. Real-time MR fluoroscopic data acquisition and image reconstruction. Magn Reson Med. 1989;12(3):407–415. 12. Holsinger AE, Wright RC, Riederer SJ, Farzaneh F, Grimm RC, Maier JK. Real-time interactive magnetic resonance imaging. Magn Reson Med. 1990;14(3):547–553. 13. Riederer SJ, Wright RC, Ehman RL, Rossman PJ, HolsingerBampton AE, Hangiandreou NJ, et al. Real-time interactive color flow MR imaging. Radiology. 1991;181(1):33–39. 14. Cox RW, Jesmanowicz A, Hyde JS. Real-time functional magnetic resonance imaging. Magn Reson Med. 1995;33(2):230–236. 15. Debbins JP, Riederer SJ, Rossman PJ, Grimm RC, Felmlee JP, Breen JF, et al. Cardiac magnetic resonance fluoroscopy. Magn Reson Med. 1996;36(4):588–595. 16. Gmitro AF, Ehsani AR, Berchem TA, Snell RJ. A real-time reconstruction system for magnetic resonance imaging. Magn Reson Med. 1996;35(5):734–740. 17. Kerr AB, Pauly JM, Hu BS, Hu BS, Li KC, Meyer CH, et al. Real-time interactive MRI on a conventional scanner. Magn Reson Med. 1997;38 (3):355–367. 606 Cardiovascular Magnetic Resonance

18. Schenck JF, Jolesz FA, Roemer PB, Cline HE, Lorensen WE, Kikinis R, et al. Superconducting open-configuration MR imaging system for image-guided therapy. Radiology. 1995;195(3):805–814. 19. Silverman SG, Collick BD, Figueira MR, Khorasani R, Adams DF, Newman RW, et al. Interactive MR-guided biopsy in an openconfiguration MR imaging system. Radiology. 1995;197(1):175–181. 20. Elgort DR, Duerk JL. A review of technical advances in interventional magnetic resonance imaging. Acad Radiol. 2005;12 (9):1089–1099. 21. Bucker A, Neuerburg JM, Adam GB, Glowinski A, Schaeffter T, Rasche V, et al. Real-time MR guidance for inferior vena cava filter placement in an animal model. J Vasc Interv Radiol. 2001;12(6):753–756. 22. Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: sensitivity encoding for fast MRI. Magn Reson Med. 1999;42(5):952–962. 23. Pruessmann KP, Weiger M, Boesiger P. Sensitivity encoded cardiac MRI. J Cardiovasc Magn Reson. 2001;3(1):1–9. 24. McVeigh ER, Guttman MA, Kellman P, Raval AN, Lederman RJ. Realtime, interactive MRI for cardiovascular interventions. Acad Radiol. 2005;12(9):1121–1127. 25. Fahrig R, Butts K, Rowlands JA, Saunders R, Stanton J, Stevens GM, et al. A truly hybrid interventional MR/X-ray system: feasibility demonstration. J Magn Reson Imaging. 2001;13(2):294–300. 26. Fahrig R, Butts K, Wen Z, Saunders R, Kee ST, Sze DY, et al. Truly hybrid interventional MR/X-ray system: investigation of in vivo applications. Acad Radiol. 2001;8(12):1200–1207. 27. Adam G, Neuerburg J, Bucker A, Glowinski A, Vorwerk D, Stargardt A, et al. Interventional magnetic resonance: initial clinical experience with a 1.5-tesla magnetic resonance system combined with c-arm fluoroscopy. Invest Radiol. 1997;32(4):191–197. 28. Rhode KS, Hill DL, Edwards PJ, Hipwell J, Rueckert D, Sanchez-Ortiz G, et al. Registration and tracking to integrate X-ray and MR images in an XMR facility. IEEE Trans Med Imaging. 2003;22(11): 1369–1378. 29. (NRPB) NRPB. Guidelines on patient dose to promote the optimisation of protection for diagnostic medical exposures: report of an advisory group on ionising. radiation. 1999;10(1). 30. Little MP, De Vathaire F, Charles MW, Hawkins MM, Muirhead CR. Variations with time and age in the risks of solid cancer incidence after radiation exposure in childhood. Stat Med. 1998;17(12):1341–1355. 31. Modan B, Keinan L, Blumstein T, Sadetzki S. Cancer following cardiac catheterization in childhood. Int J Epidemiol. 2000;29(3):424–428. 32. Risk of ionizing radiation exposure to children: a subject review. American Academy of Pediatrics. Committee on Environmental Health. Pediatrics. 1998;101(4 Pt 1):717–719.

53. 54. 55.

56.

57. 58.

59. 60.

61. 62.

63. 64.

65.

66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76.

volume loops: in vivo validation and clinical application in patients with pulmonary hypertension. Circulation. 2004;110(14):2010–2016. Duerk JL, Wong EY, Lewin JS. A brief review of hardware for catheter tracking in magnetic resonance imaging. MAGMA. 2002;13(3):199–208. Rubin DL, Ratner AV, Young SW. Magnetic susceptibility effects and their application in the development of new ferromagnetic catheters for magnetic resonance imaging. Invest Radiol. 1990;25(12):1325–1332. Bakker CJ, Hoogeveen RM, Weber J, van Vaals JJ, Viergever MA, Mali WP. Visualization of dedicated catheters using fast scanning techniques with potential for MR-guided vascular interventions. Magn Reson Med. 1996;36(6):816–820. Bakker CJ, Smits HF, Bos C, van der Weide R, Zuiderveld KJ, van Vaals JJ, et al. MR-guided balloon angioplasty: in vitro demonstration of the potential of MRI for guiding, monitoring, and evaluating endovascular interventions. J Magn Reson Imaging. 1998;8(1):245–250. Bakker CJ, Bos C, Weinmann HJ. Passive tracking of catheters and guidewires by contrast-enhanced MR fluoroscopy. Magn Reson Med. 2001;45(1):17–23. Bakker CJ, Hoogeveen RM, Hurtak WF, van Vaals JJ, Viergever MA, Mali WP. MR-guided endovascular interventions: susceptibility-based catheter and near-real-time imaging technique. Radiology. 1997;202 (1):273–276. Lardo AC. Real-time magnetic resonance imaging: diagnostic and interventional applications. Pediatr Cardiol. 2000;21(1):80–98. Omary RA, Unal O, Koscielski DS, Frayne R, Korosec FR, Mistretta CA, et al. Real-time MR imaging-guided passive catheter tracking with use of gadolinium-filled catheters. J Vasc Interv Radiol. 2000;11(8):1079–1085. Unal O, Li J, Cheng W, Yu H, Strother CM. MR-visible coatings for endovascular device visualization. J Magn Reson Imaging. 2006;23(5): 763–769. Mekle R, Hofmann E, Scheffler K, Bilecen D. A polymer-based MRcompatible guidewire: a study to explore new prospects for interventional peripheral magnetic resonance angiography (ipMRA). J Magn Reson Imaging. 2006;23(2):145–155. Buecker A, Spuentrup E, Schmitz-Rode T, Kinzel S, Pfeffer J, Hohl C, et al. Use of a nonmetallic guide wire for magnetic resonance-guided coronary artery catheterization. Invest Radiol. 2004;39(11):656–660. Miquel ME, Hegde S, Muthurangu V, Corcoran BJ, Keevil SF, Hill DL, et al. Visualization and tracking of an inflatable balloon catheter using SSFP in a flow phantom and in the heart and great vessels of patients. Magn Reson Med. 2004;51(5):988–995. Zimmermann-Paul GG, Ladd ME, Pfammatter T, Hilfiker PR, Quick JF, Debatin JF. MR versus fluoroscopic guidance of a catheter/guidewire system: in vitro comparison of steerability. J Magn Reson Imaging. 1998;8(5):1177–1181. Martin AJ, Weber OM, Saeed M, Roberts TP. Steady-state imaging for visualization of endovascular interventions. Magn Reson Med. 2003;50(2):434–438. Bakker CJ, Seppenwoolde JH, Bartels LW, van der Weide R. Adaptive subtraction as an aid in MR-guided placement of catheters and guidewires. J Magn Reson Imaging. 2004;20(3):470–474. Smits HF, Bos C, van der Weide R, Bakker CJ. Interventional MR: vascular applications. Eur Radiol. 1999;9(8):1488–1495. van der Weide R, Bakker CJ, Viergever MA. Localization of intravascular devices with paramagnetic markers in MR images. IEEE Trans Med Imaging. 2001;20(10):1061–1071. Duerk JL, Butts K, Hwang KP, Lewin JS. Pulse sequences for interventional magnetic resonance imaging. Top Magn Reson Imaging. 2000;11(3):147–162. Dumoulin CL, Souza SP, Darrow RD. Real-time position monitoring of invasive devices using magnetic resonance. Magn Reson Med. 1993;29(3):411–415. Ladd ME, Zimmermann GG, McKinnon GC, von Schulthess GK, Dumoulin CL, Darrow RD, et al. Visualization of vascular guidewires using MR tracking. J Magn Reson Imaging. 1998;8(1):251–253. Ladd ME, Zimmermann GG, Quick HH, Debatin JF, Boesiger P, von Schulthess GK, et al. Active MR visualization of a vascular guidewire in vivo. J Magn Reson Imaging. 1998;8(1):220–225. Ladd ME, Quick HH. Reduction of resonant RF heating in intravascular catheters using coaxial chokes. Magn Reson Med. 2000;43(4):615–619. Zhang Q, Wendt M, Aschoff AJ, Zheng L, Lewin JS, Duerk JL. Active MR guidance of interventional devices with target-navigation. Magn Reson Med. 2000;44(1):56–65. Hillenbrand CM, Elgort DR, Wong EY, Reykowski A, Wacker FK, Lewin JS, et al. Active device tracking and high-resolution

Cardiovascular Magnetic Resonance 607

44 PEDIATRIC INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

33. Bashore TM, Bates ER, Berger PB, Clark DA, Cusma JT, Dehmer GJ, et al. American College of Cardiology/Society for Cardiac Angiography and Interventions Clinical Expert Consensus Document on Cardiac Catheterization Laboratory Standards: A report of the American College of Cardiology Task Force on Clinical Expert Consensus Documents endorsed by the American Heart Association and the Diagnostic and Interventional Catheterization Committee of the Council on Clinical Cardiology of the AHA. J Am Coll Cardiol. 2001;37(8):2170–2214. 34. McCullough PA, Wolyn R, Rocher LL, Levin RN, O’Neill WW. Acute renal failure after coronary intervention: incidence, risk factors, and relationship to mortality. Am J Med. 1997;103(5):368–375. 35. Cigarroa RG, Lange RA, Hillis LD. Oximetric quantitation of intracardiac left-to-right shunting: limitations of the Qp/Qs ratio. Am J Cardiol. 1989;64(3):246–247. 36. Dhingra VK, Fenwick JC, Walley KR, Chittock DR, Ronco JJ. Lack of agreement between thermodilution and fick cardiac output in critically ill patients. Chest. 2002;122(3):990–997. 37. Hillis LD, Firth BG, Winniford MD. Variability of right-sided cardiac oxygen saturations in adults with and without left-to-right intracardiac shunting. Am J Cardiol. 1986;58(1):129–132. 38. van den Berg Jr E, Pacifico A, Lange RA, Wheelan KR, Winniford MD, Hillis LD. Measurement of cardiac output without right heart catheterization: reliability, advantages, and limitations of a left-sided indicator dilution technique. Cathet Cardiovasc Diagn. 1986;12(3):205–208. 39. Hillis LD, Firth BG, Winniford MD. Analysis of factors affecting the variability of Fick versus indicator dilution measurements of cardiac output. Am J Cardiol. 1985;56(12):764–768. 40. Dehmer GJ, Firth BG, Hillis LD. Oxygen consumption in adult patients during cardiac catheterization. Clin Cardiol. 1982;5(8): 436–440. 41. Beerbaum P, Korperich H, Barth P, Esdorn H, Gieseke J, Meyer H. Noninvasive quantification of left-to-right shunt in pediatric patients: phase-contrast cine magnetic resonance imaging compared with invasive oximetry. Circulation. 2001;103(20):2476–2482. 42. Beerbaum P, Korperich H, Gieseke J, Barth P, Peuster M, Meyer H. Rapid left-to-right shunt quantification in children by phase-contrast magnetic resonance imaging combined with sensitivity encoding (SENSE). Circulation. 2003;108(11):1355–1361. 43. Beerbaum P, Korperich H, Gieseke J, Barth P, Peuster M, Meyer H. Blood flow quantification in adults by phase-contrast MRI combined with SENSE–a validation study. J Cardiovasc Magn Reson. 2005;7(2): 361–369. 44. Firmin DN, Nayler GL, Klipstein RH, Underwood SR, Rees RS, Longmore DB. In vivo validation of MR velocity imaging. J Comput Assist Tomogr. 1987;11(5):751–756. 45. Kilner PJ, Manzara CC, Mohiaddin RH, Pennell DJ, Sutton MG, Firmin DN, et al. Magnetic resonance jet velocity mapping in mitral and aortic valve stenosis. Circulation. 1993;87(4):1239–1248. 46. Hundley WG, Li HF, Hillis LD, Meshack BM, Lange RA, Willard JE, et al. Quantitation of cardiac output with velocity-encoded, phasedifference magnetic resonance imaging. Am J Cardiol. 1995;75(17): 1250–1255. 47. Hundley WG, Li HF, Lange RA, Pfeifer DP, Meshack BM, Willard JE, et al. Assessment of left-to-right intracardiac shunting by velocityencoded, phase-difference magnetic resonance imaging: a comparison with oximetric and indicator dilution techniques. Circulation. 1995;91(12):2955–2960. 48. Mousseaux E, Tasu JP, Jolivet O, Simonneau G, Bittoun J, Gaux JC. Pulmonary arterial resistance: noninvasive measurement with indexes of pulmonary flow estimated at velocity-encoded MR imaging–preliminary experience. Radiology. 1999;212(3):896–902. 49. Kondo C, Caputo GR, Masui T, Foster E, O’Sullivan M, Stulbarg MS, et al. Pulmonary hypertension: pulmonary flow quantification and flow profile analysis with velocity-encoded cine MR imaging. Radiology. 1992;183(3):751–758. 50. Muthurangu V, Taylor A, Andriantsimiavona R, Hegde S, Miquel ME, Tulloh R, et al. Novel method of quantifying pulmonary vascular resistance by use of simultaneous invasive pressure monitoring and phase-contrast magnetic resonance flow. Circulation. 2004;110 (7):826–834. 51. Muthurangu V, Atkinson D, Sermesant M, Miquel ME, Hegde S, Johnson R, et al. Measurement of total pulmonary arterial compliance using invasive pressure monitoring and MR flow quantification during MR-guided cardiac catheterization. Am J Physiol Heart Circ Physiol. 2005;289(3):H1301–H1306. 52. Kuehne T, Yilmaz S, Steendijk P, Moore P, Groenink M, Saaed M, et al. Magnetic resonance imaging analysis of right ventricular pressure-

FUNCTIONAL CARDIOVASCULAR DISEASE

77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90.

91. 92.

93.

94. 95. 96. 97. 98. 99. 100. 101. 102.

intravascular MRI using a novel catheter-based, opposed-solenoid phased array coil. Magn Reson Med. 2004;51(4):668–675. Glowinski A, Adam G, Bucker A, Neuerburg J, van Vaals JJ, Gunther RW. Catheter visualization using locally induced, actively controlled field inhomogeneities. Magn Reson Med. 1997;38(2): 253–258. Guttman MA, Lederman RJ, Sorger JM, McVeigh ER. Real-time volume rendered MRI for interventional guidance. J Cardiovasc Magn Reson. 2002;4(4):431–442. Lederman RJ. Cardiovascular interventional magnetic resonance imaging. Circulation. 2005;112(19):3009–3017. Lederman RJ, Guttman MA, Peters DC, Thompson RB, Sorger JM, Dick AJ, et al. Catheter-based endomyocardial injection with real-time magnetic resonance imaging. Circulation. 2002;105(11):1282–1284. Atalar E, Bottomley PA, Ocali O, Correia LC, Kelemen MD, Lima JA, et al. High resolution intravascular MRI and MRS by using a catheter receiver coil. Magn Reson Med. 1996;36(4):596–605. Konings MK, Bartels LW, Smits HF, Bakker CJ. Heating around intravascular guidewires by resonating RF waves. J Magn Reson Imaging. 2000;12(1):79–85. Liu CY, Farahani K, Lu DS, Duckwiler G, Oppelt A. Safety of MRIguided endovascular guidewire applications. J Magn Reson Imaging. 2000;12(1):75–78. Yeung CJ, Atalar E. RF transmit power limit for the barewire loopless catheter antenna. J Magn Reson Imaging. 2000;12(1):86–91. Yeung CJ, Atalar E. A Green’s function approach to local rf heating in interventional MRI. Med Phys. 2001;28(5):826–832. Yeung CJ, Susil RC, Atalar E. RF heating due to conductive wires during MRI depends on the phase distribution of the transmit field. Magn Reson Med. 2002;48(6):1096–1098. Yeung CJ, Susil RC, Atalar E. RF safety of wires in interventional MRI: using a safety index. Magn Reson Med. 2002;47(1):187–193. Konings MK, Bartels LW, van Swol CF, Bakker CJ. Development of an MR-safe tracking catheter with a laser-driven tip coil. J Magn Reson Imaging. 2001;13(1):131–135. Weiss S, Vernickel P, Schaeffter T, Schulz V, Gleich B. Transmission line for improved RF safety of interventional devices. Magn Reson Med. 2005;54(1):182–189. Hegde S, Miquel ME, Boubertakh R, Gilderdale D, Muthurangu V, Keevil SF, et al. Interactive MR imaging and tracking of catheters with multiple tuned fiducial markers. J Vasc Interv Radiol. 2006;17(7): 1175–1179. Kuehne T, Fahrig R, Butts K. Pair of resonant fiducial markers for localization of endovascular catheters at all catheter orientations. J Magn Reson Imaging. 2003;17(5):620–624. Kuehne T, Weiss S, Brinkert F, Weil J, Yilmaz S, Schmitt B, et al. Catheter visualization with resonant markers at MR imaging-guided deployment of endovascular stents in swine. Radiology. 2004;233 (3):774–780. Quick HH, Zenge MO, Kuehl H, Kaiser G, Aker S, Massing S, et al. Interventional magnetic resonance angiography with no strings attached: wireless active catheter visualization. Magn Reson Med. 2005;53(2):446–455. Wong EY, Zhang Q, Duerk JL, Lewin JS, Wendt M. An optical system for wireless detuning of parallel resonant circuits. J Magn Reson Imaging. 2000;12(4):632–638. Weiss S, Kuehne T, Brinkert F, Krombach G, Katoh M, Schaeffter T, et al. In vivo safe catheter visualization and slice tracking using an optically detunable resonant marker. Magn Reson Med. 2004;52(4): 860–868. Kozerke S, Hegde S, Schaeffter T, Lamerichs R, Razavi R, Hill DL. Catheter tracking and visualization using 19F nuclear magnetic resonance. Magn Reson Med. 2004;52(3):693–697. Mansson S, Johansson E, Magnusson P, Chai CM, Hansson G, Petersson JS, et al. 13C imaging: a new diagnostic platform. Eur Radiol. 2006;16(1):57–67. Svensson J, Mansson S, Johansson E, Petersson JS, Olsson LE. Hyperpolarized 13C MR angiography using trueFISP. Magn Reson Med. 2003;50(2):256–262. McRobbie D, Foster MA. Thresholds for biological effects of timevarying magnetic fields. Clin Phys Physiol Meas. 1984;5(2):67–78. McRobbie D, Foster MA. Cardiac response to pulsed magnetic fields with regard to safety in NMR imaging. Phys Med Biol. 1985;30 (7):695–702. Budinger TF. MR safety: past, present, and future from a historical perspective. Magn Reson Imaging Clin N Am. 1998;6(4):701–714. Budinger TF. Emerging nuclear magnetic resonance technologies: health and safety. Ann N Y Acad Sci. 1992;649:1–18.

608 Cardiovascular Magnetic Resonance

103. Hill DL, McLeish K, Keevil SF. Impact of electromagnetic field exposure limits in Europe: is the future of interventional MRI safe? Acad Radiol. 2005;12(9):1135–1142. 104. Keevil SF, Gedroyc W, Gowland P, Hill DL, Leach MO, Ludman CN, et al. Electromagnetic field exposure limitation and the future of MRI. Br J Radiol. 2005;78(935):973–975. 105. Tenforde TS. Biological interactions and potential health effects of extremely-low-frequency magnetic fields from power lines and other common sources. Annu Rev Public Health. 1992;13:173–196. 106. Schenck JF, Dumoulin CL, Redington RW, Kressel HY, Elliott RT, McDougall IL. Human exposure to 4.0-Tesla magnetic fields in a whole-body scanner. Med Phys. 1992;19(4):1089–1098. 107. Schenck JF. MR safety at high magnetic fields. Magn Reson Imaging Clin N Am. 1998;6(4):715–730. 108. Kelley DA, Schenck JF. Very-high-field magnetic resonance imaging: instrumentation and safety issues. Top Magn Reson Imaging. 1999;10(1):79–89. 109. Schenck JF. Safety of strong, static magnetic fields. J Magn Reson Imaging. 2000;12(1):2–19. 110. Schenck JF. Physical interactions of static magnetic fields with living tissues. Prog Biophys Mol Biol. 2005;87(2–3):185–204. 111. Budinger TF, Fischer H, Hentschel D, Reinfelder HE, Schmitt F. Physiological effects of fast oscillating magnetic field gradients. J Comput Assist Tomogr. 1991;15(6):909–914. 112. Guidelines on limits of exposure to static magnetic fields. International Commission on Non-Ionizing Radiation Protection. Health Phys. 1994;66(1):100–106. 113. Kanal E, Borgstede JP, Barkovich AJ, Bell C, Bradley WG, Felmlee JP, et al. American College of Radiology White Paper on MR Safety. Am J Roentgenol. 2002;178(6):1335–1347. 114. Medical magnetic resonance (MR) procedures: protection of patients. Health Phys. 2004;87(2):197–216. 115. Shellock FG, Shellock VJ. Cardiovascular catheters and accessories: ex vivo testing of ferromagnetism, heating, and artifacts associated with MRI. J Magn Reson Imaging. 1998;8(6):1338–1342. 116. Nitz WR, Oppelt A, Renz W, Manke C, Lenhart M, Link J. On the heating of linear conductive structures as guide wires and catheters in interventional MRI. J Magn Reson Imaging. 2001;13(1): 105–114. 117. Armenean C, Perrin E, Armenean M, Beuf O, Pilleul F, SaintJalmes H. RF-induced temperature elevation along metallic wires in clinical magnetic resonance imaging: influence of diameter and length. Magn Reson Med. 2004;52(5):1200–1206. 118. Dempsey MF, Condon B, Hadley DM. Investigation of the factors responsible for burns during MRI. J Magn Reson Imaging. 2001;13(4): 627–631. 119. Gray RW, Bibens WT, Shellock FG. Simple design changes to wires to substantially reduce MRI-induced heating at 1.5 T: implications for implanted leads. Magn Reson Imaging. 2005;23(8):887–891. 120. Helfer JL, Gray RW, MacDonald SG, Bibens TW. Can pacemakers, neurostimulators, leads, or guide wires be MRI safe? Technological concerns and possible resolutions. Minim Invasive Ther Allied Technol. 2006;15(2):114–120. 121. Luechinger R, Duru F, Scheidegger MB, Boesiger P, Candinas R. Force and torque effects of a 1.5-Tesla MRI scanner on cardiac pacemakers and ICDs. Pacing Clin Electrophysiol. 2001;24(2):199–205. 122. Buecker A, Spuentrup E, Grabitz R, Freudenthal F, Schaeffter T, van Vaals JJ, et al. Real-time-MR guidance for placement of a selfmade fully MR-compatible atrial septal occluder: in vitro test. Rofo. 2002;174(3):283–285. 123. Dimick RN, Hedlund LW, Herfkens RJ, Fram EK, Utz J. Optimizing electrocardiograph electrode placement for cardiac-gated magnetic resonance imaging. Invest Radiol. 1987;22(1):17–22. 124. Tenforde TS. Magnetically induced electric fields and currents in the circulatory system. Prog Biophys Mol Biol. 2005;87(2–3):279–288. 125. Rieke V, Ganguly A, Daniel BL, Scott G, Pauly JM, Fahrig R, et al. X-ray compatible radiofrequency coil for magnetic resonance imaging. Magn Reson Med. 2005;53(6):1409–1414. 126. Razavi RS, Hill DL, Muthurangu V, Miquel ME, Taylor AM, Kozerke S, et al. Three-dimensional magnetic resonance imaging of congenital cardiac anomalies. Cardiol Young. 2003;13(5):461–465. 127. Kuehne T, Yilmaz S, Schulze-Neick I, Wellnhofer E, Ewert P, Nagel E, et al. Magnetic resonance imaging guided catheterisation for assessment of pulmonary vascular resistance: in vivo validation and clinical application in patients with pulmonary hypertension. Heart. 2005;91(8):1064–1069.

144. Schalla S, Saeed M, Higgins CB, Weber O, Martin A, Moore P. Balloon sizing and transcatheter closure of acute atrial septal defects guided by magnetic resonance fluoroscopy: assessment and validation in a large animal model. J Magn Reson Imaging. 2005;21(3): 204–211. 145. Kuehne T, Yilmaz S, Meinus C, Moore P, Saeed M, Weber O, et al. Magnetic resonance imaging-guided transcatheter implantation of a prosthetic valve in aortic valve position: feasibility study in swine. J Am Coll Cardiol. 2004;44(11):2247–2249. 146. Spuentrup E, Ruebben A, Schaeffter T, Manning WJ, Gunther RW, Buecker A. Magnetic resonance–guided coronary artery stent placement in a swine model. Circulation. 2002;105(7):874–879. 147. Zhang S, Rafie S, Chen Y, Hillenbrand CM, Wacker FK, Duerk JL, et al. In vivo cardiovascular catheterization under real-time MRI guidance. J Magn Reson Imaging. 2006;24(4):914–917. 148. Serfaty JM, Yang X, Aksit P, Quick HH, Solaiyappan M, Atalar E. Toward MRI-guided coronary catheterization: visualization of guiding catheters, guidewires, and anatomy in real time. J Magn Reson Imaging. 2000;12(4):590–594. 149. Serfaty JM, Yang X, Foo TK, Kumar A, Derbyshire A, Atalar E. MRIguided coronary catheterization and PTCA: a feasibility study on a dog model. Magn Reson Med. 2003;49(2):258–263. 150. Dick AJ, Guttman MA, Raman VK, Peters DC, Pessanha BS, Hill JM, et al. Magnetic resonance fluoroscopy allows targeted delivery of mesenchymal stem cells to infarct borders in swine. Circulation. 2003;108(23):2899–2904. 151. Dick AJ, Lederman RJ. MRI-guided myocardial cell therapy. Int J Cardiovasc Intervent. 2005;7(4):165–170. 152. McVeigh ER, Guttman MA, Lederman RJ, Li M, Kocaturk O, Hunt T, et al. Real-time interactive MRI-guided cardiac surgery: aortic valve replacement using a direct apical approach. Magn Reson Med. 2006;56(5):958–964. 153. Yang X, Atalar E. MRI-guided gene therapy. FEBS Lett. 2006;580(12): 2958–2961. 154. Sermesant M, Coudiere Y, Moreau-Villeger V, Rhode KS, Hill DL, Razavi RS. A fast-marching approach to cardiac electrophysiology simulation for XMR interventional imaging. Med Image Comput Comput Assist Interv Int Conf Med Image Comput Comput Assist Interv. 2005;8(Pt 2):607–615. 155. Sermesant M, Rhode K, Sanchez-Ortiz GI, Camara O, Andriantsimiavona R, Hegde S, et al. Simulation of cardiac pathologies using an electromechanical biventricular model and XMR interventional imaging. Med Image Anal. 2005;9(5):467–480. 156. Celik H, Uluturk A, Tali T, Atalar EA. Catheter tracking method using reverse polarization for MR-guided interventions. Magn Reson Med. 2007;58(6):1224–1231. 157. Magnusson P, Johansson E, Mansson S, et al. Passive catheter tracking during interventional MRI using hyperpolarized 13C. Magn Reson Med. 2007;57(6):1140–1147. 158. Peng H, Draper JN, Frayne R. Rapid passive MR catheter visualization for endovascular therapy using nonsymmetric truncated k-space sampling strategies. Magn Reson Imaging. 2008;26(3):293–303.

Cardiovascular Magnetic Resonance 609

44 PEDIATRIC INTERVENTIONAL CARDIOVASCULAR MAGNETIC RESONANCE

128. Rhode KS, Sermesant M, Brogan D, Hegde S, Hipwell J, Lambiase P, et al. A system for real-time XMR guided cardiovascular intervention. IEEE Trans Med Imaging. 2005;24(11):1428–1440. 129. Wildermuth S, Dumoulin CL, Pfammatter T, Maier SE, Hofmann E, Debatin JF. MR-guided percutaneous angioplasty: assessment of tracking safety, catheter handling and functionality. Cardiovasc Intervent Radiol. 1998;21(5):404–410. 130. Yang X, Atalar E. Intravascular MR imaging-guided balloon angioplasty with an MR imaging guide wire: feasibility study in rabbits. Radiology. 2000;217(2):501–506. 131. Yang X, Bolster BD, Kraitchman DL, Atalar E. Intravascular MR-monitored balloon angioplasty: an in vivo feasibility study. J Vasc Interv Radiol. 1998;9(6):953–959. 132. Godart F, Beregi JP, Nicol L, Occelli B, Vincentelli A, Daanen V, et al. MR-guided balloon angioplasty of stenosed aorta: in vivo evaluation using near-standard instruments and a passive tracking technique. J Magn Reson Imaging. 2000;12(4):639–644. 133. Omary RA, Frayne R, Unal O, Warner T, Korosec FR, Mistretta CA, et al. MR-guided angioplasty of renal artery stenosis in a pig model: a feasibility study. J Vasc Interv Radiol. 2000;11(3):373–381. 134. Krueger JJ, Ewert P, Yilmaz S, Gelernter D, Peters B, Pietzner K, et al. Magnetic resonance imaging-guided balloon angioplasty of coarctation of the aorta: a pilot study. Circulation. 2006;113(8):1093–1100. 135. Kuehne T, Saeed M, Higgins CB, Gleason K, Krombach GA, Weber OM, et al. Endovascular stents in pulmonary valve and artery in swine: feasibility study of MR imaging-guided deployment and postinterventional assessment. Radiology. 2003;226(2):475–481. 136. Raval AN, Telep JD, Guttman MA, Ozturk C, Jones M, Thompson RB, et al. Real-time magnetic resonance imaging-guided stenting of aortic coarctation with commercially available catheter devices in Swine. Circulation. 2005;112(5):699–706. 137. Buecker A, Neuerburg JM, Adam GB, Glowinski A, Schaeffter T, Rasche V, et al. Real-time MR fluoroscopy for MR-guided iliac artery stent placement. J Magn Reson Imaging. 2000;12(4):616–622. 138. Quick HH, Kuehl H, Kaiser G, Bosk S, Debatin JF, Ladd ME. Inductively coupled stent antennas in MRI. Magn Reson Med. 2002;48(5): 781–790. 139. Arepally A, Karmarkar PV, Weiss C, Rodriguez ER, Lederman RJ, Atalar E. Magnetic resonance image-guided trans-septal puncture in a swine heart. J Magn Reson Imaging. 2005;21(4):463–467. 140. Raval AN, Karmarkar PV, Guttman MA, Ozturk C, Desilva R, Aviles RJ, et al. Real-time MRI guided atrial septal puncture and balloon septostomy in swine. Catheter Cardiovasc Interv. 2006;67(4): 637–643. 141. Buecker A, Spuentrup E, Grabitz R, Freudenthal F, Muehler EG, Schaeffter T, et al. Magnetic resonance-guided placement of atrial septal closure device in animal model of patent foramen ovale. Circulation. 2002;106(4):511–515. 142. Rickers C, Jerosch-Herold M, Hu X, Murthy N, Wang X, Kong H, et al. Magnetic resonance image-guided transcatheter closure of atrial septal defects. Circulation. 2003;107(1):132–138. 143. Schalla S, Saeed M, Higgins CB, Martin A, Weber O, Moore P. Magnetic resonance–guided cardiac catheterization in a swine model of atrial septal defect. Circulation. 2003;108(15):1865–1870.

Analogous CMR Terminology Used by Various Vendors Sequence Type

Philips

Siemens

GE

Hitachi

Toshiba

Spin echo Multispin echo Fast spin echo Ultrafast spin echo

SE Multi SE TSE (Turbo SE) SSH TSE

SE Multi echo/MS Turbo SE SSTSE/HASTE

SE SE FSE (Fast SE) SS FSE

SE SE Fast SE FSE ADA

Inversion recovery Fast inversion recovery STIR FLAIR Gradient recalled echo Spoiled gradient echo Ultrafast gradient echo

IR IR TSE STIR FLAIR FFE T1 FFE T1 TFE T2 TFE BLISS, THRIVE THRIVE, eTHRIVE

IR/IRM Turbo IR/TIRM STIR FLAIR GRE FLASH TurboFLASH

IR FIR STIR FLAIR GE RSSG SARGE

Ultrafast 3D gradient echo

3D TFE

MPRAGE

Ultrafast gradient echo with magnetization preparation Volume interpolated gradient echo Steady state gradient echo Contrast enhanced steady state gradient echo Balanced gradient echo Fast balanced gradient echo Spin echo echo planar Gradient echo echo planar Hybrid echo Multi echo Spin echo black blood (cardiac)

IR TFE

T1/T2 Turbo FLASH

THRIVE, eTHRIVE

VIBE/LAVA/LAVA XV

IR FSE IR STIR FLAIR GRE SPGR/MPSPGR FGRE Fast SPGR VIBRANT LAVA, LAVA XV 3D FGRE, 3D fast SPGR IR prepped/ DE SPGR FAME

SE Multi echo Fast SE (Super) FASE DIET IR Fast IR STIR FLAIR FE RF spoiled/FE Fast FE

FFE T2 FFE

FISP PSIF

MPGR/GRE SSFP

TRSG

FE FE

BFFE BTFE SE EPI FFE EPI, TFE EPI GRASE mFFE Black blood prepulse

TrueFISP

FIESTA

BASG

True SSFP

EPI SE EPIFl TGSE

EPI SE GRE EPI

SE EPI SG EPI

SE EPI FE EPI

Spin echo black blood null fat (cardiac) Single shot black blood

Black blood prepulse with SPIR or SPAIR BB SSh

TOF (time of flight) MR angiography Time resolved time of flight with contrast Contrast enhanced MR angiography Noncontrast angiography Contrast enhanced MR angiography with moving table Real time interactive scan Susceptibility weighted imaging

Inflow MRA

Dark blood prepared TSE, HASTE TRIM

QUICK 3D MPRAGE Fast FE

Merge/Cosmic Double IR FSE with blood suppression Double IR FSE with blood suppression

Single shot 2D TrueFISP TOF

TRACS

TWIST

Bolus Trak

Care Bolus

TRANCE MobiTrak, MobiFlex

NATIVE

Interactive Venous BOLD

CARE Susceptibility weighted

TRICKS

SmartStep iDRIVE SWAN

Cardiovascular Magnetic Resonance 611

I ANALOGOUS CMR TERMINOLOGY USED BY VARIOUS VENDORS

APPENDIX I

CMR SCREENING FORM—BETH ISRAEL DEACONESS MEDICAL CENTER—CMR CENTER

APPENDIX II

CMR Screening Form—Beth Israel Deaconess Medical Center—CMR Center 1/6/2010-V5

Cardiac MR Center East Campus/Gryzmish 4 Telephone – 617-667-8555 Fax – 617-975-5480

Date of Birth ___/____/19___

Date __/___/2010 Name_____________________________ Height_________m_ Weight_______Kg An MRI involves the use of a very strong magnet. For your safety, the presence of certain metallic objects must be determined before you enter the exam room. Please place a check in the appropriate column for each item below. Yes 1.

No

Have you had an MRI/CMR before?

No

17. Do you have a Port-a-cath or Hickman device? If yes, is it accessed?

If yes, did you receive a contrast injection? 2.

Yes

Pacemaker/Pacer wires/Implantable defibrillator

18. Are you on dialysis? If yes, how often:___________________________

3.

Metallic heart valve or any metallic stents

19. Please list all surgeries:

4.

Intracranial or brain aneurysm clip/brain surgery

_________________________________________________________

5.

Bio or neurostimulator, electronic device, or implant

_________________________________________________________

6.

Tattoo(s), Tattooed eyeliner

_________________________________________________________ _________________________________________________________

If yes, location(s):_____________________________ 7.

20. Please circle if you have any of the following medical conditions:

Body piercing

If yes, location(s):____________________________

Asthma/Hay fever

Heart Disease

8.

Metal injury to the eye requiring medical attention

Thyroid Disease

Pheochromocytoma Sickle Cell Disease

9.

Shrapnel/gunshot (metal in body)

FEMALES

10. Eye prosthesis/surgery on eye

Y E S

Multiple Myeloma

MALES

N O

11. Ear prosthesis/surgery on ear

21. Possibility of

24. Do you have a

12. Limb or joint replacement or pinning

pregnancy?

penile implant?

13. Tissue expander (e.g. breast implant)

22. IUD (Intra

14. Implanted pump (insulin, pain med, chemotherapy)

Uterine Device) or

Y E S

N O

If yes, make and model:

Diaphragm

___________________________

15. Are you wearing a patch that delivers medication?

23. Pessary (in

___________________________

16. Do you have a history of difficult IV starts?

pelvis)?

CMR Staff will speak to you about the need for removing the following items: Removable dental work

Eyeglasses

Wallet/keys

Watch/Jewelry

Credit and ATM cards

Hearing aids

Wigs/hairpieces or bobby pins

How do you describe your ethnic background? (Answering this question is optional.) American Indian or Alaskan Native

Hispanic

Asian or Pacific Islander

White (not of Hispanic origin)

Black (not of Hispanic origin)

Prefer not to answer

Patient Signature _________________________________________ Relationship (if not the patient)_________ Date __/__/___ Signature of Nurse or Technologist__________________________________________________________________Date __/__/___

612 Cardiovascular Magnetic Resonance

East Campus/Gryzmish 4 Telephone – 617-667-8555 Fax – 617-975-5480

Date of Birth ___/____/19___

Yes

No

26. Have you ever been told you have kidney problems? 27. Have you ever been told you have protein in your urine? 28. Have you ever had high blood pressure? 29. Do you have diabetes? 30. Have you ever had gout? 31. Have you ever had kidney surgery? 32. Please list below any allergies to medications, food, or latex:

NONE

Reaction

33. Please list below all prescription and over-the-counter medications you take:

Last Dose Date & Time /

/

:

/

/

:

/

/

:

/

/

:

/

/

:

/

/

:

/

/

:

/

/

:

/

/

:

/

/

:

/

/

:

34. Are you taking any of the following medications? sildenafil

tadalafil

vardenafil

Date of Last Dose: ___/__/10

These drugs can interfere with certain aspects of some Cardiac MRI Examinations. If you take any of these drugs for erectile dysfunction you should refrain from taking these medications for 48 hours (2 days) prior to your Cardiac MRI. If you take any of these drugs for pulmonary hypertension you should NOT stop taking your medication but MUST inform the Cardiac MR Center staff before your examination.

Patient Signature_______________________________________ Relationship (if not the patient)__________ Date ___/___/_____ Signature of Nurse or Technologist_____________________________________________________________ Date ___/___/_____

Discharge instructions to patient:

Resume your usual medications

Special instructions:_____________________________________________________________________________________________________ RN/MD/RT Signature______________________________________________________________________________________ Date ___/___/_____

CMR Screening Form in use (2010) at the Beth Israel Deaconess Medical Center (BIDMC), Cardiac MR Center, Boston, MA. (Provided by Kraig V. Kissinger, RT)

Cardiovascular Magnetic Resonance 613

II CMR SCREENING FORM—BETH ISRAEL DEACONESS MEDICAL CENTER—CMR CENTER

Cardiac MR Center

CMR WORKSHEET AND SEQUENCE PROTOCOL DATAFORM IN USE (2010) AT THE BETH ISRAEL DEACONESS MEDICAL CENTER (BIDMC)—CMR CENTER

APPENDIX III

CMR Worksheet and Sequence Protocol Dataform in use (2010) at the Beth Israel Deaconess Medical Center (BIDMC)—CMR Center Beth Israel Deaconess Medical Center - Cardiac MRI 330 Brookline Avenue, Boston, MA 02215 Patient Name:

(617) 667-8555

fax: (617) 975-5480

BIDMC MRN:

Date of Birth:

/

/19

Height:

m

M F

Weight:

kg

Indication/History:

Scan Date:

BSA:

/

/201

m2

Referring MD: Heart Rate:

Analysis of Left Ventricular Function

bpm

Blood pressure: LV Cavity Dimension (diastole):

mm

LV Cavity Dimension (systole):

mm

Anteroseptal wall thickness:

mm

Inferolateral wall thickness:

mm

LV End Diastolic Volume:

ml

LV End Systolic Volume:

ml

LV Stroke Volume: LV ED Mass:

Rhythm SR AF Other:_____

/

mm/m2

mmHg

Comments:

7 2

3

g

g/m2

ml

16 11

5

4

Comments: ml/m2

ml RVEF:

RV Stroke Volume:

15

RCA

Analysis of Right Ventricular Function ml

17

9

6

12

10

%

RV End Systolic Volume:

LCX

13

8 14

ml/m2

ml LVEF:

RV End Diastolic Volume:

1

LAD

RV fatty infiltrate

%

Analysis of Aorta/Pulmonary Artery Flow PA total flow PA stroke vol PR TR

(ml) (ml) (ml) (ml) Qp/Qs

Ao total flow (ml) Ao stroke vol (ml) % AR (ml) % MR (ml) Cardiac output

Heart rate bpm Effective LVEF: % % l/min Cardiac index

%

l/min/m2

Measurements Ascending Aorta

Diameter=

mm

Transverse Aorta

Diameter=

mm

Descending Aorta

Diameter=

mm

Abdominal Aorta

Diameter=

mm

Pulmonary Artery

Diameter=

mm

PLA LA Dimension=

mm

4Chamber LA Length=

mm

4Chamber RA Length=

mm

2 Pulmonary veins dimension (mm) Area (m )

mm/m2

Plaque

2

Plaque

Plaque mm/m

Plaque mm/m2 cm2______ cm2/m2

Aortic valve area= pericardium thickness=

mm

Coronary sinus=

mm

Coronary origins/lengths

Left lower

RCA

Norm.

Abn.

Not Seen

Left upper

LM

Norm.

Abn.

Not Seen

Right lower

RCA Length (mm) _______

Disease: _________________

Right upper

LM Length (mm) _______

Disease: _________________

LGE:

LAD Length (mm) _______

Disease: _________________

LCX Length (mm) _______

Disease: _________________

1-24%:

ⱖ=50%:

25-49%:

mid/epi:

7

Dobutamine viability Stress Findings

2

614 Cardiovascular Magnetic Resonance

Pericardial effusion

Pleural effusion

R/L

9

17 15 10

RCA

LCX

13

8 14

3

Additional Findings

1

LAD

4

12

6

16 11

5

A. LV cines (2Ch, HLA, SA stack, 4CH)

H1. 2D Late Gd-enhancement (short axis stack, 2Ch, 4Ch)

A1. LA cine (SAX extension)

H2. 3D Hi-res late Gd-enhancement (LA views)

A2. LV cine (4Chamber Stack)

H3. 3D Hi-res late Gd-enhancement (LV)

A3. Real time-LV cines (2Ch, 4Ch, SA mid LV) if Afib

H4. Early (5min) 3D Hi-res late Gd-enhancement (LV)

A4. C-SPAMM – 4Ch, 2Ch, mid-ventricular 3 SA

I1. Cor MRI (navigator, 3D targeted TFE w/ Isordil 5 mg)

B. LVOT cines, Aortic valve cine

I2. Low res Cor (nav, 3D targeted axial SSFP w/ Isordil 5 mg)

B2. LVOT stack

I3. Low res Cor (wide nav, 3D whole H, SSFP w/ Isordil 5 mg)

B3. RVOT/main PA sagittal series, pulm valve cine

J1. Thoracic Axial T1w TSE without fat saturation

C. Aortic Q-flow (axial at level of PA bifurcation)

J2. Thoracic Axial T1w TSE with fat saturation

D. Pulmonary artery Q-flow (oblique coronal)

J3. Thoracic axial T2w TSE

E1. Resting myocardial perfusion (SA stack)

J4. Thoracic axial T1w TSE without fat sat post Gd

E2. Stress myocardial perfusion (SA stack)

J5. Thoracic Sagittal T1w TSE without fat saturation (aorta)

F. Low dose dobutamine viability 5mcg/kg/min

K. T2* (hemachromatosis)

G1. Pulmonary vein 3D BH CE-MRA

L. Coronary vein imaging

G2. Aorta 3D BH CE-MRA Comments/Notes: Signature

Allergies:

Medications: Gd-DTPA Stress

Isordil

2.5mg SL

Magnevist

Multihance ____ ml _____mmol/kg

Dobut 0.05mmol/kg/min

5mg SL Atropine ___ mg

Injection Site:

IV Fluid

Cr ____mg/dl on ___/___/___

eGFR____

Dobut max ____ mmol/kg/min

Persantine _____

Catheter Size:

Problems/Reactions: Nurse: Technologist:

Protocols: 1. LV/RV function only

A, B, C, D, J1

2. ARVC

A, B, C, D, J1, J2 (H3 if at least mod probability)

3a. Mitral Valve (Prolapse)

A, B, B2, C, D, J1

3b. Pulmonic Valve

A, B, B3, C, D, J1

3c. Tricuspid/Aortic Valve

A, B, C, D, J1

4. Pericardial Disease

A, A4, A3 (4Ch, mid-SA), B, C, D, J1, J3, H1, H3

5a. Cardiomyopathy-dilated

(if 1st CMR - I1) A, B, C, D, J1, H1, H3 (if preCRT - A4,L)

5b. Cardiomyopathy-HCM, amyloid

A, A4, B, C, D, H1, (5min H1 if Amyloid), (H3 if HCM), J1

5c. Cardiomyopathy-sarcoid, myocarditis, Hemo A, B, C, D, H1, J1, J4 (sarcoid-J3, H3; myocarditis J3, J4, H4; iron depo-K) 5d. Cardiomyopathy-VT

A, B, C, D, J1, J3, H1, H3

6. CAD Viability - delayed enhancement

A, B, C, D, (E if CAD), H1, H3, J1, if WMA -->Low dose Dobut x 5 min --> A

7. CAD wall motion stress (dobutamine)

A, B, C, D, J1; Dobutamine (stages - A3, A)

8. CAD perfusion stress (dipyridamole)

A, B, C, D, J1, E2, E1

7a. Pulm vein isolation (pre-ablation)

A, (Afib-A3), B, (Sinus-C, D), G1, H2, J1

7b. Pulm vein isolation (F/u)

A, (AFib-A3), (Sinus-C, D), G1, H2.

8. Coronaries

CAD - I1; Anomalous - I2; , CABG grafts - I3, A, B, C, D, J1

9. Congenital

A, (ASD-A1), A2, B, C, D, J1, special views

10. Cardiac mass/tumor

A, (A1), A2, B, C, D, E1, J1, (J2), J3, J4, (H1) special views

11. Aorta

A, B, C, D, J1, J5, G2

CMR Worksheet and Sequence Protocol dataform in use at the Beth Israel Deaconess Medical Center (BIDMC), Cardiac MR Center, Boston, MA. (Provided by Warren J. Manning, MD) 2D, two dimensional; 3D, three dimensional; AR, aortic regurgitation; ARVC, arrhythmogenic right ventricular cardiomyopathy; BH, breath hold; BSA, body surface area; CAD, coronary artery disease; CE, contrast enhanced; Ch, chamber; ED, end diastolic; eGFR, estimated glomerular filtration rate; F, female; Gd, gadolinium; HCM, hypertrophic cardiomyopathy; LA, left atrium; LAD, left anterior descending; LCX, left circumflex; LGE, late gadolinium enhancement, LM, left main; LV, left ventricular; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; M, male; MR, mitral regurgitation; MRA, magnetic resonance angiography; MRN, medical record number; PA, pulmonary artery; PLA, parasternal long axis; PR, pulmonary regurgitation; RA, right atrium; RCA, right coronary artery; RV, right ventricular; RVOT, right ventricular outflow tract; T1w, T1 weighted; T2w, T2 weighted; TR, tricuspid regurgitation; TSE, fast spin echo; VT, ventricular tachycardia.

Cardiovascular Magnetic Resonance 615

III CMR WORKSHEET AND SEQUENCE PROTOCOL DATAFORM IN USE (2010) AT THE BETH ISRAEL DEACONESS MEDICAL CENTER (BIDMC)—CMR CENTER

12/28/09 v10

CMR Sequences

INDEX

Index Note: Page numbers followed by f indicate figure and those followed by t indicate table.

A

Abdominal aorta, MRA of, 468–470 Abdominal ischemia, 472 Ablation, interventional CMR for, 586 Ablavar. See Gadofosveset trisodium (MS-325, Vasovist, Ablavar). Absolute tissue perfusion, in stress myocardial perfusion imaging, 221–222 Accept-reject algorithm, for navigator echoes, 131, 131t, 132f ACE-I (angiotensin-converting enzyme inhibitor), for ventricular remodeling animal studies of, 260, 261f human studies of, 262 Acorn cardiac support device, for ventricular remodeling, 261 Active catheter tracking and visualization, 584–585, 585f, 596–598 advantage of, 596–597 for aortic coarctation repair, 584f for endomyocardial injection, 586f 19 F, 598, 601f for intramyocardial injection, 598f with multiple resonant coils, 597–598, 600f safe transmission line for, 599f semi-, 597–598 for transseptal puncture, 585f Acute myocardial infarction (AMI), 241–252 appropriateness of indications for CMR for, 250t cine CMR for, 241–242, 242f complications of, 248–249, 249f coronary artery CMR for, 248 late gadolinium enhancement for, 242, 243f validation of, 243–244, 244t microvascular obstruction after pathophysiology of, 253 prognostic significance of, 246 and regional recovery of function, 246–247 residual coronary occlusion vs., 245–246, 245f, 246f myocardial viability after, 269–274 contrast-enhanced studies for, 269–274 late gadolinium enhancement for, 269–271 adjustment of T1 in, 271, 273f ischemic bed at risk in, 271, 272f nulling of signal intensity of normal myocardium in, 269, 270f and recovery of function, 273–274 T1-weighted segmented inversion recovery pulse sequence for, 269, 269f time between contrast injection and imaging in, 269–271, 271f no-reflow phenomenon and, 271–272, 274f T2-weighted images for, 269 wall thickness and, 274 stress perfusion imaging after, 247, 247f T2-weighted CMR of, 247–248, 248f ventricular remodeling after, 253–266

Acute myocardial infarction (AMI) (Continued) CMR spectroscopy of energetics during, 256–257, 257f contrast-enhanced CMR and predictors of, 257–259, 258f, 259f early phase of, 253 late phase of, 253, 254f pathophysiology of, 253 regional left ventricular function during, 255–256 therapy for animal studies of, 259–262, 261f human studies of, 262–264, 263f ventricular anatomy during, 253–255, 254f Adenosine contraindications and termination criteria for, 236, 236t, 237t drug interactions with, 236 pharmacologic effects of, 231–232, 231t route and duration of administration of, 231, 238 safety aspects of, 232–234 stress-inducible perfusion abnormalities with, 232 stress-inducible wall motion abnormalities with, 231–232 Adenosine stress CMR, 198t, 208–209 abnormalities induced by, 231–232 duration of, 237 Adenosine stress myocardial perfusion imaging, 29–30, 214, 216f abnormalities induced by, 232 combined dobutamine wall motion CMR with diagnostic performance of, 238–239, 238f protocol for, 232, 233f, 234f in comprehensive CMR assessment of coronary artery disease, 159–160 safety of, 104 Adenosine triphosphate (ATP) transfer, in 31 P-CMRS, 557–558, 560 Advanced CMR technique(s), 37–56 for increased speed, 37, 52f, 53–54 parallel imaging as, 42–53 applications of, 49–53 for assessment of global and regional cardiac function, 47f, 49–51 for coronary artery, 50f, 51–53 for detection of myocardial infarction and assessment of myocardial viability, 50f, 51 for first-pass myocardial perfusion imaging, 50f, 51 for imaging of cardiac anatomy and structure, 49, 50f artifacts in, 48–49, 48f coil arrays for, 46 coil sensitivity calibration strategies in, 46, 47f data acquisition and image reconstruction in, 45–46, 45f dynamic, 47–48 multi-detector-row CT vs., 42–45, 44f

Advanced CMR technique(s) (Continued) principles of, 42–49 signal-to-noise ratio in, 46–47, 47f undersampling in, 46 radial imaging as, 40–42 applications of, 42, 43f principles of, 39f, 40–42, 40f undersampling in, 41, 41f spiral imaging as, 37–40 applications of, 38–40, 39f off-resonance effects in, 38, 39f principles of, 37–38, 38f Adventitia, of artery, 362–363 Aliasing artifacts in parallel imaging, 46 in radial imaging, 41, 41f Allograft rejection, 548–555 CMR of experimental animal models of, 548–549, 549f patient studies of, 549–550 coronary artery CMR in, 550 CMR spectroscopy of animal studies of, 550–551, 551f patient studies of, 551–553, 552f, 553f, 562–563 comparison of diagnostic modalities for, 553, 554t Alpha pulse, in gradient echo imaging, 14 AMI. See Acute myocardial infarction (AMI). AMI-227 (ferumoxtran), 83 Amyloidosis, cardiac, 523–524, 523f, 524f morphology and function in, 523 tissue characterization in, 523–524, 524f Anatomic imaging of coarctation of the aorta, 458–459, 458f of congenital heart disease, 112–113 double inversion recovery dark-blood imaging for, 113, 114f, 115f gadolinium-based CMR for, 116–118 of arrhythmogenic right ventricular cardiomyopathy, 118 of coronary artery, 117, 119f late gadolinium enhancement imaging as, 117, 118f myocardial and blood tagging in, 116, 119f myocardial perfusion imaging as, 116, 118f, 120 saturation band in, 116, 119f static, 116, 117f technical considerations with, 120 time-resolved, 116, 118f of tumor or mass, 117 multiplanar reconstruction for, 113, 114f steady-state free precession for, 112–113, 112f of coronary artery bypass graft, 330–332 conventional spin echo and gradient echo imaging as, 330, 331f, 332f imaging strategy for, 332–333, 337f three-dimensional contrast-enhanced breath hold MRA as, 332, 335f, 336f Cardiovascular Magnetic Resonance 617

INDEX

Anatomic imaging (Continued) three-dimensional respiratory gated MRA as, 331 two-dimensional breath hold MRA as, 330–331, 333f, 334f parallel imaging for, 49, 50f in pediatric CMR, 112–113 double inversion recovery dark-blood imaging for, 113, 114f, 115f gadolinium-based, 116–118 of arrhythmogenic right ventricular cardiomyopathy, 118 of coronary artery, 117, 119f late gadolinium enhancement imaging as, 117, 118f myocardial and blood tagging in, 116, 119f myocardial perfusion imaging as, 116, 118f, 120 saturation band in, 116, 119f static, 116, 117f technical considerations with, 120 time-resolved, 116, 118f of tumor or mass, 117 multiplanar reconstruction for, 113, 114f steady-state free precession for, 112–113, 112f Anesthesia, for pediatric CMR, 395–396 Aneurysm(s) aortic, 374f, 375f causes of, 469 classification of, 469 clinical features of, 469 interventional CMR for, 587, 588f MRA of, 470 mycotic, 469–470 pseudo-, 469 thoracic, 456–457, 456f, 457f coronary artery, 299–301, 300f in Kawasaki disease, 353, 354f pseudoof thoracic aorta, 456 ventricular, due to acute myocardial infarction, 249, 249f ventricular, due to acute myocardial infarction, 249 Angiogenesis, in atherosclerotic plaques of aorta and carotid artery, 345 of coronary artery, 358–359 Angiography CT magnetic resonance vs., 466–467 of pulmonary embolism, 480 of renal artery stenosis, 471 magnetic resonance (See Magnetic resonance angiography [MRA]) phase contrast, 94 radionuclide, for right ventricular assessment, 382 X-ray of mesenteric arteries, 472 of peripheral vascular disease, 473–474 of pulmonary embolism, 480 of renal artery stenosis, 471 Angiosarcoma, 537, 540f, 544t Angiotensin type 1 receptor (AT1R), in ventricular remodeling, 260 Angiotensin type 1 receptor blocker (ARB), for ventricular remodeling, 261f Angiotensin type 2 receptor (AT2R), in ventricular remodeling, 260 Angiotensin-converting enzyme inhibitor (ACE-I), for ventricular remodeling animal studies of, 260, 261f human studies of, 262

618 Cardiovascular Magnetic Resonance

Anomalous coronary artery disease, 299, 300f, 300t Anxiolytic, in CMR stress tests, 198t Aorta abdominal, MRA of, 468–470 ascending, morphology of, 409 atherosclerotic plaque imaging in, 341–350 future directions for, 347 molecular, 344–347 angiogenesis in, 345 endothelial dysfunction in, 344–345 extracellular matrix in, 345–346 plaque inflammation in, 346–347, 347f thrombus in, 345, 346f multicontrast CMR for, 342–344, 342f with automatic segmentation, 342, 344f with complications, 343–344 imaging sequences for, 342, 342t with pharmacologic therapy, 343 in subclinical disease, 342–343 validation of, 342 rationale for, 342–347 coarctation of (See Aortic coarctation) MRA of, 468–470 after repair of transposition of the great vessels, 485, 486f thoracic, 450–462 aneurysm of, 456–457, 456f, 457f aortitis of, 459, 459f coarctation of, 458–459, 458f dissection of, 452–454, 452f, 453f, 454f flow mapping of, 450–451, 451f gradient echo CMR imaging of, 450–451 interventional CMR imaging of, 459–460 intramural hematoma of, 454–455, 455f MRA of, 451–452, 452f, 468–470 penetrating ulcer of, 455–456, 455f spin echo CMR imaging of, 450, 451f trauma to, 457–459, 457f trauma to, 457–459, 457f Aortic aneurysm(s), 374f, 375f causes of, 469 classification of, 469 clinical features of, 469 interventional CMR for, 587, 588f magnetic resonance angiography of, 470 mycotic, 469–470 pseudo-, 469 thoracic, 456–457, 456f, 457f Aortic arch, interrupted, 428–430 vs. aortic arch atresia, 428 epidemiology of, 428 in infant and pediatric patients classification of, 428, 429f evaluation of, 429 postoperative assessment of, 430, 430f preoperative assessment of, 429, 429f with truncus arteriosus, 426, 427f, 428f surgical repair of, 428–429 Aortic arch atresia, 428 Aortic arch hypoplasia, truncus arteriosus with, 426, 427f Aortic arch imaging, with single ventricle, 123–124, 123f, 124f Aortic coarctation, 124–125, 126f, 400–403, 458–459 adult, 458 anatomic imaging for, 458–459, 458f cine CMR of, 113, 458–459 clinical manifestations of, 400 CMR-guided intervention for, 604, 604f, 605 contrast-enhanced MRA of, 400, 401f dark-blood imaging of, 115f defined, 124–125 epidemiology of, 458

Aortic coarctation (Continued) etiology of, 400, 458 fast spin echo image of, 400f flow mapping of, 400–402, 402f, 458–459 gross morphologic features of, 400 infantile, 458 postductal, 458 preductal, 458 repair of, 402, 459 imaging after, 402 passive and active catheter devices during, 584f, 587 truncus arteriosus with, 426, 427f velocity encoded cine CMR of, 417 Aortic compliance, 362 CMR of regional, 364, 364f and coronary artery disease, 368, 368f flow wave velocity and, 368, 368f fluvastatin and, 368–369 menotropin and, 369 Aortic dissection, 452–454 acute phase of, 452–453 characteristics of, 452–453 choice of imaging modality for, 453 chronic, 469–470 classification of, 452–453, 452f, 453f CMR vs. TEE for, 453–454 contrast-enhanced MRA of, 454, 454f etiology of, 470 imaging sequence for, 452–453 interventional CMR for, 587, 588f pathophysiology of, 470 therapeutic management of, 470 Aortic distensibility cilazapril and, 369 in Marfan syndrome, 369 Aortic elastic modulus, 369 Aortic flow, 152, 153f Aortic flow wave velocity, 366, 366f Aortic recesses, normal anatomy of, 489–490, 490f Aortic regurgitation, 403–404, 509 measurement of, 509, 510f severity of, 502t Aortic root, total coronary flow reserve from measurements in, 313–314 Aortic stenosis, 502t, 507 Aortic stiffness. See Aortic wall stiffness. Aortic ulcer, penetrating, 455–456, 455f Aortic valve bicuspid, 403–404, 403f cine CMR of, 113, 116f normal function of, 152–154, 154f papillary fibroelastoma of, 534, 536f, 544t in tetralogy of Fallot, 420 Aortic valve replacement, interventional CMR for, 587, 605, 605f Aortic valve stenosis, 502t, 507 Aortic wall stiffness and future cardiac events, 370, 372f in Marfan syndrome, 369 Aortitis, 459, 459f Aortocoronary bypass graft. See Coronary artery bypass graft (CABG). Aortography, for aortic dissection, 453 Aortopulmonary collateral vessels, in tetralogy of Fallot, 421f Apical rotation, 71f, 73–74, 73f, 73t ARB (angiotensin type 1 receptor blocker), for ventricular remodeling, 260, 261f Array spatial sensitivity encoding technique (ASSET), 45f, 46 Arrhythmogenic right ventricular cardiomyopathy (ARVC), 118, 520–521, 520f diagnostic criteria for, 520, 521, 521t

Atherosclerotic plaque imaging (Continued) challenge(s) in, 351 cardiac motion as, 351–352 respiratory motion as, 352–353, 353f clinical studies of, 359 contrast-enhanced, 354–356, 355f molecular, 356, 356f angiogenesis in, 358–359 inflammation in, 358 thrombosis in, 356–358, 357f, 358f noncontrast, 353–354, 354f, 355f outlook for, 359 future directions for, 347 rationale for, 342–347 Atherothrombosis of aorta and carotid artery, 345, 346f of coronary artery, 356–358, 357f, 358f Athlete’s heart, CMR spectroscopy of, 560–561 ATP (adenosine triphosphate) transfer, in 31 P-CMRS, 557–558, 560 Atrial fibrillation epidemiology of, 445 interventional CMR for, 586 pulmonary veins in pathophysiology of, 445 Atrial fibrillation ablation, pulmonary vein imaging before and after, 445–446, 445f, 446f Atrial isomerism, 409 Atrial mapping, interventional CMR for, 586 Atrial morphology, 408–409 Atrial septal defect (ASD), 116, 398–400 advantages of CMR for, 399 clinical manifestations of, 398 CMR-guided catheterization for, 399 with Ebstein anomaly, 414f location of, 398, 398f, 399f post-closure evaluation of, 400 shunt quantification in, 399, 399f types of, 398, 398f, 399f uses of CMR for, 398–399 Atrial septum, lipomatous hypertrophy of, 142, 146f, 533–534, 535f, 544t Atrial situs inversus, 408 Atrial situs solitus, 408, 409 Atrial switch procedure, 415–416 postoperative assessment of, in infant and pediatric patients, 423–425 contrast-enhanced CMR of, 425f ECG-gated SSFP imaging of, 423, 423f navigator-gated imaging of, 423, 424f Atrial tachyarrhythmia, CMR-guided RF ablation for, 603 Atrial transseptal procedures, interventional CMR for, 585f, 586 Atrial-esophageal fistula, 445–446 Atriopulmonary anastomosis, 432, 432f Atriopulmonary Fontan connection, steadystate free precession images of, 112f Atrioventricular (AV) connection abnormalities of, 409 concordant, 409 discordant, 409 double-inlet, 409 Atrioventricular plane descent (AVPD), normal values for, 385–388, 388t in females, 385–388, 387t in males, 385–388, 387t Atrioventricular (AV) septal defect, 398, 398f Atrioventricular (AV) valve(s), straddling, 409 Atrioventricular (AV) valve atresia, 409 Atropine, in dobutamine CMR stress test, 198, 199f, 202–203 Atropine stress CMR pharmacokinetics of, 196 safety of, 196–197, 198t

Auditory considerations, with CMR, 103 Automatic segmentation, for atherosclerotic plaques, of aorta and carotid arteries, 342, 344f Automatic triggering, for contrast-enhanced MRA, 465 AV. See Atrioventricular (AV). AVMs (arteriovenous malformations), pulmonary, 486 Axial plane scout image in, 140, 141f uses for, 140–142, 143f Axial scout image, 20, 21f

B

B0 field, 3–5, 4f B1 field, 3–5, 4f B22956. See Gadocoletic acid (B22956). Background phase offset, 94–95 BACSPIN (breathing autocorrection with spiral interleaves), 136–137 Balanced steady-state free precession, 15, 15f BAR (brachial artery reactivity) testing, 370–371, 372f, 373f Becker muscular dystrophy, CMR spectroscopy for, 565 Benign cardiac tumor(s), 532–536 fibroma as, 534–536, 537f, 544t hemangioma as, 536, 538f, 544t leiomyomatosis with intracardiac extension as, 536, 539f, 544t lipoma as, 533–534, 535f, 544t myxoma as, 533, 534f, 544t other, 536, 539f papillary fibroelastoma as, 534, 536f, 544t rhabdomyoma as, 536, 538f, 544t Beta-blockers, for ventricular remodeling, 262 Bicuspid valve, 403 aortic, 403–404, 403f cine CMR of, 113, 116f pulmonary, 404, 404f Bing-Taussig anomaly, 412, 425, 426 Biological effects, safety of, 100 Biplanar approach, for cardiomyopathy, 516 Biventricular pacing, X-ray and CMR-guided, 605f Biventricular volume and function, in valvular heart disease, 501 Black blood imaging for cardiac allograft rejection, 550 for congenital heart disease, 396 for coronary artery CMR, 289t with atherosclerotic plaques, 353–354, 355f fast spin echo, 13, 13f in morphology scanning cardiac gating for, 26f goal of, 21–22, 23f physiology of, 22, 26f for right ventricular assessment, 383, 383f of thoracic aorta, 450, 451f, 468–469 uses for, 140–142, 143f Blalock-Taussig shunt, 433f BLAST. See Broad-use linear acquisition speed-up technique (BLAST). Blood flow velocity assessment, 91–99 history of, 91 phase flow imaging methods for, 91–93 Fourier flow imaging as, 91, 93 velocity phase encoding in, 93, 93f visualizing flow data in, 96, 96f phase contrast velocity mapping as, 91, 92–93 flow vector map in, 96, 97f flow velocity images in, 92–93, 93f and Fourier velocity imaging, 94

Cardiovascular Magnetic Resonance 619

INDEX

Arrhythmogenic right ventricular cardiomyopathy (ARVC) (Continued) function and morphology in, 520, 521 right ventricular assessment in, 391 tissue characterization in, 521 Arterial blood, T1 and T2 values for, 7t Arterial compliance, 363 Arterial imaging, invasive, 587 Arterial input function, in quantitative evaluation of myocardial perfusion, 65–66, 65f Arterial spin labeling, for stress myocardial perfusion imaging, 215 Arterial stenosis, interventional CMR for, 589, 589f Arterial stiffness. See Arterial wall stiffness. Arterial structure, 362–363 Arterial switch procedure, 120–121, 121f for transposition of the great arteries, 410–411, 416 Arterial wall biophysical mechanical properties of, 362–378 structure of, 362–363 Arterial wall compliance, 363 Arterial wall shear stress, 371–374, 374f, 375f Arterial wall stiffness clinical use of CMR for assessing, 368–370, 368f, 370f, 371f, 372f defined, 363 measurement of, 363 Arteriovenous malformations (AVMs), pulmonary, 486 Arteritis, Takayasu, 459, 459f Artifacts, 142–149 cardiac motion, 142–145, 146, 147f chemical shift, 142–145, 147f, 148–149 with high field CMR, 170, 171f metal, 142–145, 146–148, 147f in parallel imaging, 48–49, 48f respiratory motion, 142–145, 146, 147f ARVC. See Arrhythmogenic right ventricular cardiomyopathy (ARVC). ASD. See Atrial septal defect (ASD). ASSET (Array spatial sensitivity encoding technique), 45f, 46 AT1R (angiotensin type 1 receptor), in ventricular remodeling, 260 AT2R (angiotensin type 2 receptor), in ventricular remodeling, 260 Atherogenesis, 213, 214f Atherosclerosis defined, 341, 351 development of, 213, 214f epidemiology of, 341 natural history of, 341 pathobiology of, 341–342 risk factors for, 341 subclinical, 342–343 Atherosclerotic plaque, rupture of, 341–342 Atherosclerotic plaque imaging of aorta and carotid artery, 341–350 molecular, 344–347 angiogenesis in, 345 endothelial dysfunction in, 344–345 extracellular matrix in, 345–346 plaque inflammation in, 346–347, 347f thrombus in, 345, 346f multicontrast CMR for, 342–344, 342f, 343f with automatic segmentation, 342, 344f with complications, 343–344 imaging sequences for, 342, 342t with pharmacologic therapy, 343 in subclinical disease, 342–343 validation of, 342 of coronary arteries, 351–361

INDEX

Blood flow velocity assessment (Continued) improving accuracy of, 93–97, 94f, 95f rapid, 95–96 validation of, 95 principles of, 91–93, 92f rapid, 95–96, 96f time-of-flight methods for, 91, 92f visualizing flow data in, 96–97 flow pressure maps in, 96–97, 97f flow vector map in, 96, 97f Fourier velocity imaging in, 96, 96f phase contrast velocity images in, 96, 97f three-dimensional, 97 wall shear stress in, 96 Blood oxygen level dependent (BOLD) imaging in clinical setting, 572–573, 573f emerging techniques for, 574–576, 574f, 575f, 576f, 577f future of, 577 high field, 576, 577f for myocardial perfusion imaging, 62, 215–216 myocardial perfusion reserve in, 573 without pharmacologic stress, 572–573, 573f in preclinical setting, 571–572, 572f rationale for, 570 vs. SPECT, 572, 573f SSFP-based, 574–576, 574f, 575f, 576f Blood pool contrast agents, 80–81, 80f, 81t for three-dimensional magnetic resonance angiography, 464, 465f Blood tagging, for congenital heart disease, 116, 119f BOLD. See Blood oxygen level dependent (BOLD). Bolus chase technique, for peripheral vascular disease, 474, 475f Bolus tagging, for coronary artery velocity measurement, 314 Bolus timing, for contrast-enhanced magnetic resonance angiography, 465–466 Bolus tracking approach in pediatric CMR, 120 in stress myocardial perfusion imaging, 221 BOOST trial, 263 Brachial artery reactivity (BAR) testing, 370–371, 372f, 373f Breath holding alternative to, 129 in coronary artery CMR, 286, 286t, 289–290, 289f with atherosclerotic plaques, 352 with coronary artery bypass grafts, 289–290, 290f three-dimensional contrast-enhanced, 332, 335f, 336f two-dimensional, 330–331, 333f, 334f for coronary artery velocity measurement, 315–316 with native vessel stenosis, 301–302, 302t for coronary sinus flow assessment, 311 limitations of, 129 multiple, 129–130 mean diaphragm displacement in, 130f respiratory trace data for, 130f in pediatric CMR, 120 in phase contrast velocity mapping, 95 for ventricular function assessment, 149 Breathing autocorrection with spiral interleaves (BACSPIN), 136–137 Bright blood imaging of aorta, 468–469 for congenital heart disease, 396 for coronary artery CMR, 289t in morphology scanning, 21–22, 24f of myocardial function, 141f, 142, 143f for right ventricular assessment, 383, 384f

620 Cardiovascular Magnetic Resonance

Bright signals, catheter devices that create, 583–584, 584f Broad-use linear acquisition speed-up technique (BLAST), k-t applications of, 50f, 51 to assess cardiac function, 185 for myocardial perfusion imaging, 61 principles of, 47–48 Bulboventricular foramen, 430 Bypass graft. See Coronary artery bypass graft (CABG).

C

CABG. See Coronary artery bypass graft (CABG). CAD. See Coronary artery disease (CAD). Calcification, in atherosclerosis, 341 Calcium, epicardial, 304–305, 305f Captopril, for ventricular remodeling, 260 Captopril renography, 471 Carbon dioxide (CO2)-filled balloon, in interventional CMR, 595, 596f, 597f Carbon-13 (13C), hyperpolarized, 87 Carbon-13 (13C) CMR spectroscopy, 87 experimental studies with, 559 Cardiac allograft rejection, 548–555 CMR of experimental animal models of, 548–549, 549f patient studies of, 549–550 coronary artery CMR in, 550 CMR spectroscopy of animal studies of, 550–551, 551f patient studies of, 551–553, 552f, 553f, 562–563 comparison of diagnostic modalities for, 553, 554t Cardiac amyloidosis, 523–524, 523f, 524f morphology and function in, 523 tissue characterization in, 523–524, 524f Cardiac anatomy anatomic variants of, 142, 146f artifacts in imaging of, 142–149 cardiac motion, 142–145, 146, 147f chemical shift, 142–145, 147f, 148–149 metal, 142–145, 146–148, 147f respiratory motion, 142–145, 146, 147f breath hold scout image for, 140, 141f imaging planes and, 140, 141f, 143f main structures in, 140, 145f parallel imaging of, 49, 50f preparatory prepulses for, 140 Cardiac catheterization and intervention, CMR-guided for atrial septal defect, 399 for congenital heart disease, 396, 415–417 for patent ductus arteriosus, 400 for ventricular septal defect, 397 Cardiac disease, congenital. See Congenital heart disease (CHD). Cardiac dysfunction, population impact of, 181 Cardiac function aortic flow in, 152, 153f assessment of, 179–195 CMR for, 183–185 accuracy and reproducibility of, 185–186, 185f, 186f advantages of, 183 of diastolic function, 190, 192f ECG gating in, 188 end-diastolic and end-systolic volumes in, 183–184, 183f, 184f FISP vs. FLASH in, 185, 187

Cardiac function (Continued) future of, 193 left ventricular, 183–185, 183f, 184f practical guide to, 186–189, 187f reference ranges for, 189, 189t, 190t, 191f of regional function, 190–192 right ventricular, 185 scanning time for, 185 short axis slices in, 183–184, 184f, 188 Simpson’s rule in, 183–184, 184f of systolic function, 189–190 computed tomography for, 182–183 echocardiography for, 181–182, 182f importance of, 181 nuclear cardiology for, 182 slice-thickness for, 188 in coronary artery disease, 158 left ventricular, 141f, 142, 149, 150f left ventricular mass in, 142, 150 effect of imaging sequence and magnetic field strength on, 150–152, 151t, 152t myocardial, 141f, 142, 143f parallel imaging of, 49–51, 50f pulmonary artery flow in, 152 right ventricular, 150 stroke volume in, 150 effect of imaging sequence and magnetic field strength on, 150–152, 151t, 152t systolic and diastolic, 149–150 valvular, 152–155, 154f Cardiac gating for black-blood imaging, 26f for cine CMR, 24–25, 27f for coronary artery CMR, 286–287, 287f with slice tracking, 287–288 for morphology scanning, 22, 25f for scout scanning, 21, 23f for vascular angiography, 34–35 for velocity-encoded CMR imaging, 32 for viability imaging, 31, 32f Cardiac hemangioma, 536, 538f, 544t Cardiac masses, 532–547 contrast agents for, 532 due to intracardiac thrombus formation, 540–542, 542f, 544t pediatric, 117 technical considerations with, 532, 533t tumors as (See Cardiac tumor[s]) Cardiac morphology, in coronary artery disease, 158 Cardiac motion, 69–70 assessment of CMR methods of, 69–70 non-CMR methods of, 69 in coronary artery and vein CMR, 284–285, 285f with coronary artery atherosclerotic plaque imaging, 351–352 Cardiac motion artifacts, 142–145, 146, 147f Cardiac output, 150, 189–190 Cardiac pacemakers, safety of CMR with, 101, 102, 107–108, 108f Cardiac phase to order reconstruction (CAPTOR), 74 Cardiac rejection, 548–555 CMR of experimental animal models of, 548–549, 549f patient studies of, 549–550 coronary artery CMR in, 550 CMR spectroscopy of animal studies of, 550–551, 551f

Cardiomyopathy(ies) (CMPs) (Continued) tissue characterization in, 523–524, 524f myocardial siderosis as, 524 morphology and function in, 524 tissue characterization in, 524 sarcoidosis as, 523 morphology and function in, 523 tissue characterization in, 523 metabolism in, 516 morphology and function in, 515–516 myocarditis as, 517–518, 517f combined protocols for, 517f, 518 early enhancement in, 517f, 518 edema in, 517f, 518 follow-up for, 518 function and morphology in, 517f, 518 late gadolinium enhancement in, 517f, 518 tissue characterization in, 517f, 518 noncompaction, 522, 522f characteristics of, 522 function and morphology in, 522 tissue characterization in, 522 primary, 515 restrictive, 521–522 vs. constrictive pericarditis, 493 morphology and function in, 522 tissue characterization in, 522 stress-induced (Tako-Tsubo), 525f, 526 function and morphology in, 525f, 526 tissue characterization in, 526 tissue characterization in, 516 transthoracic echocardiography for, 515 Cardiovascular magnetic resonance (CMR) alignment with main magnetic field in, 3 balanced steady-state free precession in, 15, 15f basic principles of, 1–18 contraindications to, 105–108 with coronary stents, 106 general, 105–108 with pacemakers and implantable cardioverter defibrillators, 101, 102, 107–108, 108f with valvular prosthesis, 106 echo planar imaging, spiral, and radial in, 16–17, 16f frequency encoding: position in X in, 9–10, 9f, 10f gradient echo imaging in, 13–14, 14f gradients in, 7–8 image creation in, 7, 7f inversion recovery fast gradient recalled echo: late gadolinium enhancement in, 15 phase encoding: position in Y in, 10–11, 10f pulse sequences and contrast in, 12–13, 12f radiofrequency and magnet strength in, 3–5, 4f raw k-space data and fast Fourier transform in, 11–12, 11f screening form for, 612–613 selective excitation: position in Z in, 8–9, 8f, 9f signal detection in, 3–17 spin echo imaging in, 6–7, 6f, 7t fast (turbo), 6–7, 13 double inversion recovery (black-blood), 13, 13f T1 relaxation, 5 T2 relaxation and spin phase in, 5–6, 5f terminology used by various vendors for, 611 three-dimensional fast gradient echo: magnetic resonance angiography in, 15

Cardiovascular magnetic resonance (CMR) (Continued) worksheet and sequence protocol dataform for, 614–615 Cardiovascular magnetic resonance spectroscopy (CMRS), 556–568 atomic nucleic used in, 556, 557t for cardiac allograft rejection animal studies of, 550–551, 551f vs. other imaging modalities, 554t patient studies of, 551–553, 552f, 553f, 562–563 for cardiomyopathy, 516 dilated, 517 hypertrophic, 519 clinical studies of, 559–565 in athlete’s heart and hypertension, 560–561 in diabetes and obesity, 561 in healthy subjects, 560, 561f with heart failure and cardiac transplantation, 561–563, 562f in ischemic heart disease, 563–565 for myocardial viability assessment, 564–565, 564f for stress testing, 563–564, 563f methodologic considerations in, 559–560, 560f, 561f with specific gene defects with cardiac pathology, 565 in valvular heart disease, 563 of energetics during left ventricular remodeling, 256–257, 257f experimental foundations of, 557–559 other nuclei in, 559 31 P-CMRS in, 557–558, 558f high field, 175 for myocardial viability, 268, 278–280 perspectives on, 565–566 physical principles of, 556–557, 557f at 3-Tesla, 565, 565f Cardiovascular magnetic resonance (CMR) tagging assessment, of left ventricular function, 69–75 apical rotation in, 71f, 73–74, 73f, 73t cardiac motion and, 69–70 complementary spatial modulation of magnetization for, 70–71 echo planar imaging in, 70–71 grid-tagged images in, 70, 71f, 72f limitations of, 74 slice thickness in, 70 slice-following principle in, 70, 70f spiral imaging in, 70–71 steady-state free precession in, 70–71 with strain measurements, 74 3D, 71 in dobutamine stress CMR, 208f, 209 evaluation of motion data from, 71–72, 73f, 73t limitations of, 74 methods for, 70–72 during remodeling, 255–256 results of, 72–74 strain measurement in, 74 Carotid arteries atherosclerotic plaque imaging in, 341–350 future directions for, 347 molecular, 344–347 angiogenesis in, 345 endothelial dysfunction in, 344–345 extracellular matrix in, 345–346 plaque inflammation in, 346–347, 347f thrombus in, 345, 346f multicontrast CMR for, 342–344, 343f

Cardiovascular Magnetic Resonance 621

INDEX

Cardiac rejection (Continued) patient studies of, 551–553, 552f, 553f, 562–563 comparison of diagnostic modalities for, 553, 554t Cardiac rotation, 69–70 assessment of CMR methods of, 69–70 non-CMR methods of, 69 Cardiac structure, parallel imaging of, 49, 50f Cardiac tamponade, 492 Cardiac thrombus, 540–542, 542f, 544t Cardiac transplantation, 548–555 CMR of experimental animal models of, 548–549, 549f patient studies of, 549–550 coronary artery CMR in, 550 CMR spectroscopy of animal studies of, 550–551, 551f patient studies of, 551–553, 552f, 553f, 562–563 comparison of diagnostic modalities for, 553, 554t Cardiac triggering. See Cardiac gating. Cardiac tumor(s), 532–547 benign, 532–536 fibroma as, 534–536, 537f, 544t hemangioma as, 536, 538f, 544t leiomyomatosis with intracardiac extension as, 536, 539f, 544t lipoma as, 533–534, 535f, 544t myxoma as, 533, 534f, 544t other, 536, 539f papillary fibroelastoma as, 534, 536f, 544t rhabdomyoma as, 536, 538f, 544t contrast agents for, 532 malignant, 537–540 lymphoma as, 540, 541f, 544t metastatic, 540 sarcoma as, 537–540 angio-, 537, 540f, 544t leiomyo-, 537, 544t lipo-, 540, 541f, 544t pediatric, 117 prognosis of, 545 step-by-step procedure for assessment of, 545 technical considerations with, 532, 533t tissue characterization for, 543–544, 544t Cardiomyopathy(ies) (CMPs), 515–531 arrhythmogenic right ventricular, 118, 515, 520f diagnostic criteria for, 520, 521, 521t function and morphology in, 520, 521 right ventricular assessment in, 391 tissue characterization in, 521 classification of, 515 clinical presentation of, 515 CMR-derived information in, 515, 516t dilated, 516–517 CMR spectroscopy in, 561–563, 562f function and morphology in, 516–517 metabolic CMR in, 517 endomyocardial diseases as, 524–526, 525f morphology and function in, 526 tissue characterization in, 526 hypertrophic, 518–520, 519f CMR spectroscopy for, 565 follow-up for, 520 function and morphology in, 519 LVOT obstruction in, 518, 519–520 tissue characterization in, 519 infiltrative secondary, 515, 523 cardiac amyloidosis as, 523–524, 523f, 524f morphology and function in, 523

INDEX

Carotid arteries (Continued) with automatic segmentation, 342, 344f with complications, 343–344 imaging sequences for, 342, 342t with pharmacologic therapy, 343 in subclinical disease, 342–343 validation of, 342 rationale for, 342–347 extracranial, magnetic resonance angiography of, 467–468, 467f, 468f Carotid artery stenosis, 467f, 468, 468f Cartesian pattern, 37 Carvedilol, for ventricular remodeling, 262, 263f Catheter tracking and visualization, 583, 595–598 active, 584–585, 585f, 596–598 advantage of, 596–597 for aortic coarctation repair, 584f for endomyocardial injection, 586f 19 F, 598, 601f for intramyocardial injection, 598f with multiple resonant coils, 597–598, 600f safe transmission line for, 599f semi-, 597–598 for transseptal puncture, 585f application(s) of for aortic coarctation repair, 584f for atrial septal defect, 399 for congenital heart disease, 396, 415–417 for endomyocardial injection, 586f for intramyocardial injection, 598f for patent ductus arteriosus, 400 for transseptal puncture, 585f for ventricular septal defect, 397 vs. conventional XRF devices, 583, 583f passive, 583–584, 595–596 chemical-selective visualization of, 584, 584f CO2-filled balloon in, 595, 596f, 597f dysprosium oxide, 595, 596f ideal material for, 595 that create bright signals, 583–584, 584f that create dark signals, 583 strategy for, 585 Cavity dilation, after acute myocardial infarction, 253 CE-CTA (contrast-enhanced computed tomography angiography) CE-MRA vs., 466–467 of pulmonary embolism, 480 Cellular agents, endomyocardial delivery of, 585–587, 586f CE-MRA. See Contrast-enhanced magnetic resonance angiography (CE-MRA). Central volume principle, in quantitative evaluation of myocardial perfusion, 62, 63 Cerebral aneurysm clips, CMR with, 105–106 CFD (computational fluid dynamic) simulation, for arterial wall shear stress, 371–372 CFR. See Coronary flow reserve (CFR). CHD. See Congenital heart disease (CHD). Chemical shift, in CMR spectroscopy, 557, 560 Chemical shift artifacts, 142–145, 147f, 148–149 Chemical-selective visualization, of passive catheter device, 584, 584f Chest pain cine CMR for, 241 in CMR stress tests, 198t Chest x-ray of constrictive pericarditis, 493 of pericardial disease, 488 Children, CMR in. See Pediatric CMR. Chordae tendineae, 381

622 Cardiovascular Magnetic Resonance

CHRISTMAS trial, 262, 263f Chronic myocardial infarction (CMI), 242, 244t myocardial viability in, 274–280 contractile reserve and, 275–278 late gadolinium enhancement for, 276, 277f, 278f, 279f vs. dobutamine CMR, 277–278 vs. other imaging modalities, 276–277, 281f thickness of epicardial rim and recovery of function in, 276, 280f wall thickness and, 275, 275f Cilazapril and aortic distensibility, 369 for ventricular remodeling, 260 Cine CMR, 22–25 acquisition time in, 24 for acute myocardial infarction, 241–242, 242f of aortic coarctation, 113, 458–459 cardiac gating for, 24–25, 27f of congenital heart disease, 113, 115f, 408 for evaluation of function, 417 after repair, 416–417 due to transposition of the great arteries, 120–122, 121f valvular, 113, 116f of coronary artery disease for disease detection, 160, 161, 161f for viability studies, 165f, 166f goal of, 23–24, 27f of myocardial function, 141f, 142 for right ventricular assessment, 383, 384f with single ventricle, 124, 124f for stress tests, 198, 199t of thoracic aorta, 450–451, 451f of valvular heart disease, 501, 502f in infants and children, 113, 116f visualization and planimetry of jets by, 504–505, 505f, 510f of ventricular remodeling, 255, 256 CK (creatine kinase) flux, in 31P-CMRS, 557–558, 560 CK/PCr (creatine kinase/phosphocreatine) energy shuttle, in 31P-CMRS, 557–558, 558f Claustrophobia, during CMR, 103 Clinical technique(s), 19–36 cine CMR as, 22–25 acquisition time in, 24 cardiac gating for, 24–25, 27f goal of, 23–24, 27f combination of, 35, 35f morphology scanning as, 21–22 black-blood imaging in cardiac gating for, 26f goal of, 21–22, 23f physiology of, 22, 26f bright-blood imaging in, 21–22, 24f cardiac gating for, 22, 25f goal of, 21–22, 23f, 24f HASTE in, 21–22, 23f pulse sequence in, 21–22, 25f steady-state free precession imaging in, 21–22, 24f myocardial perfusion scanning as, 25–31 goal of, 25–27, 28f image acquisition in, 27–28, 29f magnetization recovery in, 28, 29f pulse sequence in, 28–29, 30f rest, 31 stress adenosine for, 29–30

Clinical technique(s) (Continued) specific steps in, 31 timeline for, 30–31, 30f scout scanning as, 19–21 cardiac gating for, 21, 23f goal of, 19–20 image acquisition in, 20, 21f k-space filling in, 20–21, 22f pulse sequence in, 20, 22f vascular angiography as, 34–35 cardiac gating for, 34–35 goal of, 34, 34f pulse sequence in, 35 timing of image acquisition in, 34, 35f velocity-encoded imaging as, 32–34 cardiac gating for, 32 goal of, 32, 33f pulse sequence in, 33–34, 33f viability imaging as, 31 cardiac gating for, 31, 32f goal of, 31, 31f inversion recovery in, 31, 32f CMI. See Chronic myocardial infarction (CMI). CMRS. See Cardiovascular magnetic resonance spectroscopy. CNR (contrast-to-noise ratio), in stress myocardial perfusion imaging, 218, 222 CO2 (carbon dioxide)-filled balloon, in interventional CMR, 595, 596f, 597f Coarctation of the aorta, 124–125, 126f, 400–403, 458–459 adult, 458 anatomic imaging for, 458–459, 458f cine CMR of, 113, 458–459 velocity encoded, 417 clinical manifestations of, 400 contrast-enhanced MRA of, 400, 401f dark-blood imaging of, 115f defined, 124–125 epidemiology of, 458 etiology of, 400, 458 fast spin echo image of, 400f flow mapping of, 400–402, 402f, 458–459 gross morphologic features of, 400 infantile, 458 postductal, 458 preductal, 458 repair of, 402, 459 CMR-guided, 604, 604f, 605 imaging after, 402 passive and active catheter devices during, 584f, 587 truncus arteriosus with, 426, 427f Cochlear implants, as contraindication to CMR, 105–106 Coil arrays, in parallel imaging, 46 Coil sensitivity calibration strategies, in parallel imaging, 46, 47f Collagen, type I, contrast agent specific to, 85 Collagen fibers, in arterial wall, 362–363 Collateral vessels, in tetralogy of Fallot, 421f Column orientations, multiple, for navigator echoes, 133–136, 135f Column positioning, for navigator echoes, 133, 135f, 135t Column selection, for navigator echoes, 131–132 Comb-excited Fourier velocity-encoded measurement, of aortic flow wave velocity, 366–367, 367f Combidex (ferumoxtran), 83 Communication, in interventional CMR laboratory, 580–581 Compartmental model, in quantitative evaluation of myocardial perfusion, 63–65, 64f

Congenital heart disease (CHD) (Continued) late gadolinium enhancement imaging as, 117, 118f myocardial and blood tagging in, 116, 119f myocardial perfusion imaging as, 116, 118f, 120 saturation band in, 116, 119f static, 116, 117f technical considerations with, 120 time-resolved, 116, 118f of tumor or mass, 117 multiplanar reconstruction for, 113, 114f steady-state free precession for, 112–113, 112f atrial morphology and determination of situs in, 408–409 atrial septal defect as, 398–400 advantages of CMR for, 399 clinical manifestations of, 398 CMR-guided catheterization for, 399 location of, 398, 398f post-closure evaluation of, 400 shunt quantification in, 399, 399f types of, 398, 399f uses of CMR for, 398–399 atrial switch for, 423–425 contrast-enhanced CMR of, 425f ECG-gated SSFP for, 423, 423f navigator-gated imaging of, 423, 424f cardiac catheterization and intervention for, 395, 415–417 cine CMR of, 113, 115f, 116f gradient echo, 408 CMR vs. transesophageal echocardiography for, 408, 415 coarctation of the aorta as, 124–125, 126f, 400–403 cine CMR of, 113 velocity encoded, 417 clinical manifestations of, 400 contrast-enhanced MRA of, 400, 401f dark-blood imaging of, 115f defined, 124–125 etiology of, 400 fast spin echo image of, 400f flow mapping of, 400–402, 402f gross morphologic features of, 400 post-repair imaging of, 402 repair of, 402 complex, 408–419 in infant and pediatric patients, 420–438 ventricular, 413–415 contrast-enhanced 3D MRA for, 408 after correction, 395 double-outlet right ventricle as, 410f, 412 in infant and pediatric patients, 425–426 postoperative assessment of, 426 preoperative assessment of, 426, 426f Ebstein anomaly of tricuspid valve as, 413, 414f, 415f ECG-gated spin echo CMR of, 408 epidemiology of, 395, 396t evaluation of function in, 417 functional fetal CMR for, 125–127, 126f future of, 125–127, 126f general protocol for, 112–118, 112f imaging techniques in, 396–404 in infant and pediatric patients, 420–438 double-outlet right ventricle as, 425–426 for postoperative assessment, 426 for preoperative assessment, 426, 426f interrupted aortic arch as, 428–430 classification of, 428, 429f for evaluation, 429 for postoperative assessment, 430, 430f

Congenital heart disease (CHD) (Continued) for preoperative assessment, 429, 429f postoperative atrial switch, 423–425 contrast-enhanced, 425f ECG-gated SSFP, 423, 423f navigator-gated, 423, 424f single ventricle as, 430–435 for evaluation, 432 Fontan procedure for, 432, 432f left, 430, 431f post-Fontan, 434–435, 434f right, 430, 431f during staged palliation, 433, 433f, 434f tetralogy of Fallot as, 420–422 for evaluation, 421 for postoperative assessment, 421–422, 422f for preoperative assessment, 421, 421f transposition of the great arteries as, 422–423 truncus arteriosus as, 426–428 classification of, 426, 427f contrast-enhanced, 428f for postoperative assessment, 427–428 for preoperative assessment, 427, 428f interrupted aortic arch as, in infant and pediatric patients, 428–430 classification of, 428, 429f evaluation of, 429 postoperative assessment of, 430, 430f preoperative assessment of, 429, 429f late gadolinium enhancement for, 408 limitations and challenges of, 111 patent ductus arteriosus as, 400 postoperative evaluation of, 415–417 right ventricular assessment in, 391 for shunt evaluation, 395 simple, 395–407 single ventricle as, 122–124, 413–415 aortic arch imaging for, 123–124, 123f, 124f cine CMR for, 124, 124f with Fontan baffle, 123, 123f, 125f in infant and pediatric patients, 430–435 evaluation of, 432 Fontan procedure for, 432, 432f left, 430, 431f post-Fontan, 434–435, 434f right, 430, 431f during staged palliation, 433, 433f, 434f late gadolinium enhancement for, 124f perfusion imaging for, 124f pulmonary artery imaging for, 123, 123f, 125f time-resolved gadolinium imaging for, 118f velocity mapping for, 123–124 venous pathway imaging for, 123f, 124 with ventricular outflow obstruction, 125f surgical procedures for, 411t technical consideration(s) in, 118–120 with gadolinium-based techniques, 120 inability to hold breath as, 120 spatial and temporal resolution as, 118–120 tetralogy of Fallot as, 413, 420–422 in infant and pediatric patients, 420–422 evaluation of, 421 postoperative assessment of, 421–422, 422f preoperative assessment of, 421, 421f surgical repair of, 416 transposition of the great arteries as, 120–122, 409–411, 410f, 411f cine SSFP of, 120–122, 121f dark-blood CMR of, 120–122, 121f

Cardiovascular Magnetic Resonance 623

INDEX

Compartmentalization, and contrast-enhanced tissue relaxation, 84–85 Complementary spatial modulation of magnetization (CSPAMM), 70–71 echo planar imaging in, 70–71 grid-tagged images in, 70, 71f, 72f limitations of, 74 slice thickness in, 70 slice-following principle in, 70, 70f spiral imaging in, 70–71 steady-state free precession in, 70–71 with strain measurements, 74 3D, 71 Comprehensive CMR assessment, of coronary artery disease, 159–166 analysis of studies with, 164–166 for detection of disease, 164, 167f for viability studies, 164–166, 167f contrast agent delivery for, 160 defined, 159 detection of disease in analysis of studies with, 164, 167f protocols for, 160–162, 161f, 162f, 163f future directions for, 167–168 historical background of, 158 protocols for, 160–164, 160t, 161f, 162f sensitivity and specificity of, 161, 163f suggested, 161, 165f selection of methods for, 159–160 viability studies in analysis of studies with, 164–166, 167f protocols for, 163–164, 165f, 166f, 167f Computational fluid dynamic (CFD) simulation, for arterial wall shear stress, 371–372 Computed tomography (CT) of aortic dissection, 453 of aortic intramural hematoma, 454–455 to assess cardiac function, 182–183 of constrictive pericarditis, 494, 494f multidetector to assess cardiac function, 182–183 vs. coronary artery CMR, 304–305, 305f, 305t vs. parallel MRI, 42–45, 44f of pericardial disease, 488 multislice, for right ventricular assessment, 382 of pericardial disease, 488 single photon emission (See Single photon emission computed tomography [SPECT]) Computed tomography angiography (CTA) MRA vs., 466–467 of pulmonary embolism, 480 of renal artery stenosis, 471 Computer architecture, for navigator echoes, 137 Concordant atrioventricular connection, 409 Congenital anomalies of coronary arteries, 299, 300f, 300t of pulmonary arteries, 485–486, 485f, 486f Congenital heart disease (CHD), 111–128 abnormalities of atrioventricular connection as, 409 abnormalities of ventriculoarterial connections as, 409–413 advantages of CMR for, 111 anatomic imaging of, 112–113 double inversion recovery dark-blood imaging for, 113, 114f, 115f gadolinium-based CMR for, 116–118 of arrhythmogenic right ventricular cardiomyopathy, 118 of coronary artery, 117, 119f

INDEX

Congenital heart disease (CHD) (Continued) in infant and pediatric patients, 422–423 three-dimensional contrast-enhanced MRA of, 117f, 120–122, 122f truncus arteriosus as, 412–413, 412f in infant and pediatric patients, 426–428 classification of, 426, 427f contrast-enhanced CMR of, 428f postoperative assessment of, 427–428 preoperative assessment of, 427, 428f valvular, 403–404 with atresia, 402f, 403 bicuspid valve as, 403 aortic, 403–404, 403f pulmonary, 404, 404f gradient recalled echo CMR of, 403f, 404f pressure and volume overload in, 404 quantification of regurgitation volume in, 403f, 405f after repair, 403, 405f transverse spin echo CMR of, 402f velocity mapping for, 113–115 ventricular morphology and isomerism in, 409 ventricular septal defect as, 396–397 anatomic delineation of, 397, 397f clinical manifestations of, 396–397 CMR-guided catheterization and intervention for, 397 location of, 396, 396f shunt quantification in, 397, 398f spontaneous closure of, 396–397 surgical management of, 396–397, 397f Congenital pericardial defects, 490 Conotruncal anomalies, 420 Conoventricular septal defect, 420 Constrictive pericarditis, 493–495 chest x-ray of, 493 clinical presentation of, 493 CMR of, 494, 494f CT of, 494, 494f effusive-, 493 etiology of, 493 pericardial thickening in, 493 vs. restrictive cardiomyopathy, 493 transthoracic echocardiography of, 493 ventricular filling pattern in, 495 Contractile function, cine CMR of, 22–25 acquisition time in, 24 cardiac gating for, 24–25, 27f goal of, 23–24, 27f Contractile reserve in chronic myocardial infarction, 275–278 in dobutamine stress CMR studies, 204f, 206f of viable myocardium, 267–268 Contraindications, 105–108 with coronary stents, 106 general, 105–108 with pacemakers and implantable cardioverter defibrillators, 101, 102, 107–108, 108f with valvular prosthesis, 106 Contrast, 12–13, 12f endogenous, for assessment of myocardial perfusion, 61–62 Contrast agents, 76–90 biophysics of, 76–81, 77f, 78f blood pool, 80–81, 80f, 81t for cardiac and paracardiac masses, 532 for comprehensive CMR assessment of coronary artery disease, 160 contrast-enhanced tissue relaxation with, 84–85

624 Cardiovascular Magnetic Resonance

Contrast agents (Continued) in development, 85–87, 86f effects on signal intensity of, 77, 78f extracellular, 78–79, 79f, 79t gadolinium, 78–79, 79f, 79t history of, 76, 77–78 hyperpolarized, 87 iron oxide–based cross-linked, 86–87 in development, 86–87 relaxivity with, 83 structure of, 83 uses of, 81 for myocardial oxygenation assessment, 570 newer, 85 positive and negative, 76 relaxation rate with, 76–77 relaxivity of, 81–83 correlation time in, 82 effect and definition of, 76, 77f effect of correlation time and field strength on, 82, 83, 83f electronic relaxation in, 82 inner- and outer-sphere, 81–82 iron oxide–based, 83 longitudinal and transverse, 82, 83f magnetic field dependence on, 82, 82f magnetic moment in, 82 molecule size and, 82 for selected media, 83, 83t safety of, 87–88, 103–104 with spoiled gradient recalled echo, 77, 78f for stress myocardial perfusion imaging endogenous, 215–216 exogenous, 216–218 extravascular, 216–217, 217f hyperpolarized, 217–218 intravascular, 217 T1 and T2, 76–81, 77f, 78f Contrast echocardiography, to assess cardiac function, 181–182 Contrast enhancement, water exchange and its effects on myocardial, 61 Contrast-enhanced CMR of acute myocardial infarction, 269–274 of coarctation of the aorta, 400, 401f of coronary artery, 292–294, 293f with atherosclerosis, 354–356, 355f with native vessel stenosis, 304, 304f of postoperative atrial switch, 425f and predictors of left ventricular remodeling, 257–259, 258f, 259f Contrast-enhanced computed tomography angiography (CE-CTA) CE-MRA vs., 466–467 of pulmonary embolism, 480 Contrast-enhanced magnetic resonance angiography (CE-MRA), 34–35 advantages of, 463 of aorta, 468–470 of aortic aneurysms, 470 of aortic dissection, 452–453, 454, 454f, 469–470 bolus timing for, 465–466 cardiac gating for, 34–35 of congenital heart disease, 116, 117f, 396, 408 due to transposition of the great arteries, 117f, 120–122, 122f contrast agents for, 80–81, 80f, 81t of coronary artery bypass graft, 332, 335f, 336f vs. CT angiography, 466–467 of extracranial carotid arteries, 467–468, 468f goal of, 34, 34f of mesenteric arteries, 473

Contrast-enhanced magnetic resonance angiography (CE-MRA) (Continued) parallel imaging in, 466 of peripheral vascular disease, 474, 475f for assessment of bypass graft patency, 474 bolus chase technique in, 474, 475f of hand and wrist, 475 sensitivities and specificities with, 474 time-resolved technique in, 474 for pulmonary artery hypertension, 482–483, 482f, 483f of pulmonary embolism, 481, 481f, 482f pulse sequence in, 35 of renal artery stenosis, 472 technique of, 463 of thoracic aortic aneurysm, 456–457, 457f three-dimensional, 464–465 gadolinium chelates for, 464, 465f pulse sequences for, 464 time-resolved, 466 vessel brightness (T1) vs. gadolinium dose in, 464–465 timing of image acquisition in, 34, 35f Contrast-enhanced tissue relaxation, 84–85 Conus arteriosus, 381 Coronal plane scout image in, 140, 141f uses for, 140–142, 143f Coronal scout image, 20, 21f, 140, 141f Coronary arterial pressure, in myocardial oxygenation assessment, 569, 570f Coronary artery(ies) atherosclerotic plaque imaging of, 351–361 challenge(s) in, 351 cardiac motion as, 351–352 respiratory motion as, 352–353, 353f clinical studies of, 359 contrast-enhanced, 354–356, 355f molecular, 356, 356f angiogenesis in, 358–359 inflammation in, 358 thrombosis in, 356–358, 357f, 358f noncontrast, 353–354, 354f, 355f outlook for, 359 congenital anomalies of, 299, 300f, 300t physiology and pathophysiology of, 229 in tetralogy of Fallot, 420 Coronary artery aneurysms, 299–301, 300f Coronary artery blood flow. See Coronary artery flow. Coronary artery bypass graft (CABG), 329–340 coronary artery CMR for, 305–306, 329–330, 330t anatomic imaging techniques of, 330–332 conventional spin echo and gradient echo imaging as, 330, 331f, 332f imaging strategy for, 332–333, 337f three-dimensional contrast-enhanced breath hold MRA as, 332, 335f, 336f three-dimensional respiratory gated MRA as, 331 two-dimensional breath hold MRA as, 330–331, 333f, 334f breath holding for, 289–290, 290f diagnostic accuracy of, 306, 307f, 307t fast spin echo for, 305–306, 306f indications for, 338 limitations of, 306, 307f, 337–338 for quantification of flow and flow reserve, 333–337, 337f, 338f sensitivity and specificity of, 305–306, 306t demographics of, 329 occlusion of, 329 other imaging modalities for, 329

Coronary artery CMR (Continued) suppression of signal from surrounding tissue as, 288, 288f three-dimensional acquisition sequence for, 290–292 of native vessel stenosis with navigator gating, 302–303, 303t targeted, 301–302, 301f whole heart, 302f, 303, 303t targeted acquisition sequence for, 291, 291f of native vessel stenosis, 301–302, 301f thin-slab, 291–292 whole heart acquisition sequence for, 291, 291f, 292f of native vessel stenosis, 302f, 303, 303t Coronary artery disease (CAD), 158–169 aortic compliance and, 368, 368f CMR in, 158–159, 159t for coronary artery imaging, 159, 159t of ischemia, 158–159, 159t of morphology and function, 158, 159t of viability, 159, 159t comprehensive CMR assessment of, 159–166 analysis of studies with, 164–166 for detection of disease, 164, 167f for viability studies, 164–166, 167f contrast agent delivery for, 160 defined, 159 for detection of disease analysis of studies with, 164, 167f protocols for, 160–162, 161f, 162f, 163f future directions for, 167–168 historical background of, 158 protocols for, 160–164, 160t, 161f, 162f sensitivity and specificity of, 161, 163f suggested, 161, 165f selection of methods for, 159–160 for viability studies analysis of studies with, 164–166, 167f protocols for, 163–164, 165f, 166f, 167f development of, 213, 214f dobutamine stress CMR of, 198, 201f contemporary studies of, 202–203, 202t, 203f, 203t early studies of, 201–202, 202f 3-T, 203 epidemiology of, 284 Coronary artery flow, 310–328 with coronary artery bypass graft, 333–337, 337f, 338f direct assessment of, 320 future developments in, 324–327 patient studies of, 324, 325f, 326f validation and feasibility studies of, 320–324, 322f, 323f effect of nitrates on, 285, 286f indirect assessment of, 310–324, 312f, 313f in myocardial oxygenation assessment, 569, 570f Coronary artery flow velocity mapping. See Coronary artery velocity mapping. Coronary artery flow velocity reserve, 322–323, 323f Coronary artery MRA, for coronary artery disease, 159 Coronary artery stenosis coronary CMR of, 301–304 contrast-enhanced, 304, 304f three-dimensional with navigator gating, 302–303, 303t targeted, 301–302, 301f

Coronary artery stenosis (Continued) whole heart, 302f, 303, 303t two-dimensional, 301–302, 302f, 302t and plaque rupture, 213, 214f after stent implantation, 324, 326f Coronary artery velocity mapping, 314 bolus tagging for, 314 with coronary artery bypass graft, 333–337, 337f, 338f echo planar time-of-flight technique for, 314–315, 315f future developments in, 324–327 gradient echo phase, 315–317 breath holding techniques for, 315–316 in-plane, 316–317, 317f navigator techniques for, 316–317, 317f, 318f through-plane, 317, 317f, 318f interleaved spiral phase, 317–320, 319f, 320f, 321f patient studies of, 324, 325f, 326f validation and feasibility studies of, 320–324, 322f, 323f Coronary artery wall, high field CMR of, 174, 174f Coronary autoregulation, 229 Coronary blood flow. See Coronary artery flow. Coronary flow. See Coronary artery flow. Coronary flow reserve (CFR), 213–214, 224, 230 with coronary artery bypass graft, 333–337, 337f, 338f defined, 310 direct assessment of, 320 future developments in, 324–327 patient studies of, 324, 325f, 326f validation and feasibility studies of, 320–324, 322f, 323f indirect assessment of from coronary sinus flow, 310–313, 313f from measurements in aortic root, 313–314 Coronary flow velocity measurement. See Coronary artery velocity mapping. Coronary sinus defect, 398f Coronary sinus flow, 310–313, 312f, 313f Coronary stenosis. See Coronary artery stenosis. Coronary stents restenosis after implantation of, 324, 326f safety of CMR with, 106 Coronary vein CMR, 295, 295f Coronary venous outflow, velocity mapping of, 310–313, 312f, 313f Correction factors, for navigator echoes, 132–133, 133f, 134f Creatine kinase (CK) flux, in 31P-CMRS, 557–558, 560 Creatine kinase/phosphocreatine (CK/PCr) energy shuttle, in 31P-CMRS, 557–558, 558f Creatine phosphate to adenosine triphosphate (CP-to-ATP) ratio, during left ventricular remodeling, 256, 257, 257f Crusher gradient, 14, 14f CSPAMM. See Complementary spatial modulation of magnetization (CSPAMM). CT. See Computed tomography (CT). CTA (computed tomography angiography) MRA vs., 466–467 of pulmonary embolism, 480 of renal artery stenosis, 471 Cyst(s) hydatid, 536, 539f pericardial, 490–491, 491f, 543, 543f, 544t

Cardiovascular Magnetic Resonance 625

INDEX

Coronary artery CMR, 284–298 acquisition sequence for, 289–292, 289f, 289t breath hold two-dimensional segmented k-space gradient echo, 289–290, 289f with coronary artery bypass grafts, 289–290, 290f free breathing spin echo, 289 three-dimensional, 290–292 targeted, 291, 291f thin-slab, 291–292 whole heart, 291, 291f, 292f for acute myocardial infarction, 248 advanced methods for, 292–296 contrast-enhanced, 292–294, 293f with intracoronary stents, 294–295, 294f spiral and radial, 292, 292f, 293f 3-Tesla, 294, 294f of aneurysms and Kawasaki disease, 299–301, 300f for cardiac allograft rejection, 550 clinical results of, 299–309 of congenital heart disease, 299, 300f, 300t navigator-gated, ECG-gated, 117, 119f perfusion imaging in, 116 of coronary artery bypass grafts, 305–306, 329–330, 330t anatomic imaging techniques of, 330–332 conventional spin echo and gradient echo imaging as, 330, 331f, 332f imaging strategy for, 332–333, 337f three-dimensional contrast-enhanced breath hold MRA as, 332, 335f, 336f three-dimensional respiratory gated MRA as, 331 two-dimensional breath hold MRA as, 330–331, 333f, 334f breath holding for, 289–290, 290f diagnostic accuracy of, 306, 307f, 307t fast spin echo for, 305–306, 306f indications for, 338 limitations of, 306, 307f, 337–338 for quantification of flow and flow reserve, 333–337, 337f, 338f sensitivity and specificity of, 305–306, 306t for coronary artery disease, 159 future technical developments in, 295–296 high field, 172–173, 173f history of, 284 imaging planes for, 142 invasive and interventional, 586–587 vs. multidetector CT, 304–305, 305f, 305t of native vessel stenosis, 301–304 contrast-enhanced, 304, 304f three-dimensional with navigator gating, 302–303, 303t targeted, 301–302, 301f whole heart, 302f, 303, 303t two-dimensional, 301–302, 302f, 302t parallel imaging for, 50f, 51–53 rationale for, 284 technical challenge(s) of, 284–288 cardiac motion as, 284–285, 285f effect of nitrates on coronary artery blood flow as, 285, 286f respiratory motion as, 285–286, 286t breath hold methods for, 286, 286t free breathing methods for, 286, 286t navigators for gating and slice tracking for, 287–288 navigators for triggering alone for, 286–287, 287f spatial resolution as, 288, 288f

INDEX

D

DANTE schemes, for cardiac allograft rejection, 548–549 Dark signals, catheter devices that create, 583 Dark-blood imaging, of congenital heart disease, 113, 114f, 115f due to transposition of the great arteries, 120–122, 121f DCM. See Dilated cardiomyopathy (DCM). DCMR. See Dobutamine stress CMR (DCMR). DCMRC (Duke Cardiovascular Magnetic Resonance Center) clinical volume at, 19, 20f overall makeup of, 19, 20f exam menu for, 19, 20f DeBakey classification, of aortic dissection, 452–453, 452f Deconvolution analysis, in quantitative evaluation of myocardial perfusion, 63 Defibrillators, implantable cardioverter, safety of CMR with, 102, 107–108 Delayed enhancement. See Late gadolinium enhancement. Delayed gadolinium enhancement. See Late gadolinium enhancement. DENSE (displacement encoding with stimulated echoes), for left ventricular systolic function, 149 Deoxyhemoglobin, in BOLD technique, 62 Dephasing, 5f, 6 Dephasing gradient, 10 Diabetes, CMR spectroscopy in, 560–561 Diastolic function, 149–150 assessment of, 190, 191f Diastolic strain, 74 Diffusion tensor magnetic resonance imaging (DTMRI), after acute myocardial infarction, 255 Dilated cardiomyopathy (DCM), 516–517 CMR spectroscopy in, 561–563, 562f function and morphology in, 516–517 metabolic CMR in, 517 Diminishing variance algorithm, for navigator echoes, 131, 131t, 132f Diphenhydramine, in CMR stress tests, 198t 2,3-Diphosphoglycerate (2,3-DPG), in 31PCMRS, 559, 561f Dipyridamole for coronary sinus flow assessment, 311, 312f for myocardial oxygenation assessment, 570–571 safety considerations with, 104 during stress CMR, 208–209 for stress myocardial perfusion studies, 214, 216f Discordant atrioventricular connection, 409 Displacement encoding with stimulated echoes (DENSE), for left ventricular systolic function, 149 Distensibility, of blood vessel, 363 D-loop transposition of the great arteries, 409–410, 410f, 422–423 Dobutamine contraindications and termination criteria for, 236, 236t, 237t drug interactions with, 236 for myocardial oxygenation assessment, 570–571 pharmacologic effects of, 231t, 232 route and duration of administration of, 238 safety considerations with, 104, 234–236 stress-inducible perfusion abnormalities with, 232 stress-inducible wall motion abnormalities with, 232 626 Cardiovascular Magnetic Resonance

Dobutamine stress CMR (DCMR) abnormalities induced by, 232 apical and short axis views in, 198, 200f atropine in, 198, 199f, 202–203 cine GRE or SSFP bright-blood images in, 198, 199t combined adenosine perfusion and diagnostic performance of, 238–239, 238f protocol for, 232, 233f, 234f of coronary artery disease, 158–159 in comprehensive CMR assessment for disease detection, 159–160, 160t for viability studies, 164, 165f delineation of orthogonal left ventricular myocardial segments in, 198, 200f vs. dobutamine stress echocardiography, 202–203, 202t accuracy of, 197 safety of, 196–197, 197t technique of, 197 uses of, 197 for viability studies, 204 dopamine infusion protocol for, 198, 199f duration of, 237 facilities for, 198, 199f inducible ischemia during, 198, 201f contemporary studies of, 202–203, 202t, 203f, 203t early studies of, 201–202, 202f late gadolinium enhancement for, 205, 207f for myocardial infarction acute, 241 chronic, 275–278 in myocardial perfusion studies, 214, 216f pharmacokinetics of, 196 for prognosis, 205, 207f vs. radionuclide studies, 204 safety of, 196–197, 197t, 198t, 234–236 sensitivity and specificity of, 203, 203f, 203t vs. stress myocardial perfusion imaging, 229–240 technique for, 197–200 3-T, 203 tissue tagging during, 206–208, 208f for viability studies, 204–205, 207f in viability studies, 204–205 contractile reserve in, 204f, 206f end-systolic wall thickening in, 204, 204f intramyocardial segment shortening in, 205 vs. late gadolinium enhancement, 277–278 low-dose, 204–205, 237–238 short-axis basal views in, 205f tissue tagging in, 204–205, 207f Dobutamine stress echocardiography (DSE) accuracy of, 197 vs. dobutamine stress CMR, 202–203, 202t safety of, 196–197, 197t technique of, 197 uses of, 197 for viability studies, 204 Doppler ultrasonography for monitoring during CMR, 105 of peripheral vascular disease, 473 of renal artery stenosis, 471–472 DORV. See Double-outlet right ventricle (DORV). Dotarem (gadoterate), 79f, 79t compartmentalization and relaxivity with, 84 safety of, 87–88 Double inversion recovery dark-blood imaging, for congenital heart disease, 113, 114f, 115f Double inversion recovery fast spin echo imaging, 13, 13f

Double oblique planes, 19–20 Double-chambered right ventricle, 508 jet flow in, 504–505, 505f Double-outlet right ventricle (DORV), 410f, 412 defined, 425 epidemiology of, 425 features of, 425 in infant and pediatric patients, 425–426 postoperative assessment of, 426 preoperative assessment of, 426, 426f surgical management of, 425 late complications after, 425–426 ventricular septal defect in, 412, 425 2,3-DPG (2,3-diphosphoglycerate), in 31 P-CMRS, 559, 561f Dressler syndrome, 491–492 DSE. See Dobutamine stress echocardiography (DSE). DTMRI (diffusion tensor magnetic resonance imaging), after acute myocardial infarction, 255 D-transposition of the great arteries, 409–410, 410f, 422–423 Dual-bolus approach, in stress myocardial perfusion imaging, 222 Dual-T1-sensitivity method, in stress myocardial perfusion imaging, 222 Ductus arteriosus, 400 patent, 400 CMR evaluation of, 400 dark-blood imaging of, 115f interventional closure of, 400 pathogenesis of, 400 pulmonary arteries in, 485, 485f Duke Cardiovascular Magnetic Resonance Center (DCMRC) clinical volume at, 19, 20f overall makeup of, 19, 20f exam menu for, 19, 20f D-ventricular loop, 409 Dysprosium oxide catheter, for interventional CMR, 595, 596f

E

Early gadolinium hypoenhancement, and myocardial viability, 268 Earplugs, for CMR, 103 Ebstein anomaly, of tricuspid valve, 413, 414f, 415f ECF (extracellular fluid) contrast agents, 78–79, 79f, 79t ECG. See Electrocardiographic (ECG). Echinococcus, 536, 539f Echo(es), 6–7, 6f gradient recalled (See Gradient recalled echo [GRE]) navigator (See Navigator echoes) spin (See Spin echo imaging) Echo planar imaging (EPI), 16, 16f for chemical shift artifact, 147f, 148–149 for coronary artery velocity measurement, 314–315, 315f in CSPAMM, 70–71 for myocardial perfusion imaging, 58–59 for stress myocardial perfusion imaging, 218 Echo time (TE), 11f, 12 effect on signal of, 12, 12f for myocardial perfusion imaging, 58 Echocardiography to assess cardiac function, 181–182, 182f for cardiac allograft rejection, 554t dobutamine stress accuracy of, 197 vs. dobutamine stress CMR, 202–203, 202t safety of, 196–197, 197t

End-diastolic volume (EDV) (Continued) measurement of, 183–184, 183f, 184f intraobserver, interobserver, and interstudy variability in, 187f normal values for, 385–388, 388t in females, 385–388, 387t, 390f in males, 385–388, 387t, 389f for right ventricular assessment, 383–385, 386f Endocardial cushion defects, 398 Endocardial trabeculae, in left ventricular mass, 188 Endocarditis, Loeffler’s eosinophilic, 524–526, 525f morphology and function in, 526 tissue characterization in, 526 Endogenous contrast, for assessment of myocardial perfusion, 61–62 Endomyocardial biopsy, for cardiac allograft rejection, 554t Endomyocardial catheter ablation, 586 Endomyocardial delivery, of cellular agents, 585–587, 586f Endomyocardial diseases, 524–526, 525f morphology and function in, 526 tissue characterization in, 526 Endomyocardial fibrosis, 524–526, 525f morphology and function in, 526 tissue characterization in, 526 Endomyocardial mapping, 586 Endorem (ferumoxide), 83 Endothelial dysfunction, in atherosclerosis, 341 of aorta and carotid artery, 344–345 Endothelial function, assessment of, 370–371, 372f, 373f End-systolic volume (ESV) after acute myocardial infarction, 253–254, 255 measurement of, 183–184, 183f, 184f intraobserver, interobserver, and interstudy variability in, 187f normal values for, 385–388, 388t in females, 385–388, 387t, 390f in males, 385–388, 386t, 389f for right ventricular assessment, 383–385, 386f Eovist (gadoxetic acid), 85 EP-1242, for thrombus formation in aorta and carotid arteries, 345 EP-1873, for thrombus formation in coronary artery, 356–357, 357f EP-2104R, 85, 86f for thrombus formation in aorta and carotid arteries, 345, 346f in coronary artery, 356–357, 358f EPI. See Echo planar imaging (EPI). Epicardial calcium, 304–305, 305f Epicardial rim, thickness of viable, 276, 280f Ernst angle, in gradient echo imaging, 13–14, 59 E-selectin, in atherosclerosis of aorta and carotid artery, 345 of coronary artery, 358 ESV. See End-systolic volume (ESV). Exchange time, in contrast-enhanced tissue relaxation, 84 Excitation field, 3–5, 4f Exercise, for ventricular remodeling, 264 Exercise stress CMR, left ventricular wall motion during, 209–210, 210f Extracellular fluid (ECF) contrast agents, 78–79, 79f, 79t Extracellular matrix, in atherosclerotic plaques of aorta and carotid artery, 345–346

Extravascular contrast media, for stress myocardial perfusion imaging, 216–217, 217f

F 19

F (fluorine-19) CMR, catheter visualization and localization using, 598, 601f Fast acquisition relaxation mapping (FARM), T1, for myocardial perfusion imaging, 59–60, 60f Fast exchange, in contrast-enhanced tissue relaxation, 84 Fast Fourier transform, 11–12 Fast gradient recalled echo, 14, 14f inversion recovery, 15 three-dimensional, 15 Fast imaging with steady-state precession (FISP), to assess cardiac function, 185, 187 Fast low angle shot (FLASH) imaging to assess cardiac function, 185 for coronary artery velocity mapping, 315–316, 317–320, 319f, 320f turbo, 14, 14f for valvular heart disease, 505 Fast spin echo (FSE) imaging, 6–7, 13 of aorta, 468 of aortic coarctation, 400, 400f of coronary artery atherosclerotic plaques, 353 of coronary artery bypass graft, 300t, 305–306, 306f double inversion recovery (black-blood), 13, 13f pulse sequence diagram for, 13, 13f Fat, T1 and T2 values for, 7t Fat saturation, in coronary artery CMR, 288, 288f Fat-excitation acquisition, for coronary artery velocity mapping, 317–320, 321f Fatty streak, in atherosclerosis, 341 FDG. See Fluorodeoxyglucose (FDG). Ferromagnetic materials, 83 Ferromagnetism, safety of, 101, 101f, 102f Ferucarbotran (Resovist), relaxivity with, 83 Ferumoxide (Endorem, Feridex), 83 Ferumoxtran (AMI-227, Sinerem, Combidex), 83 Ferumoxytol (Feraheme), 85 for 3D MRA, 465f Fetal CMR, functional real-time, 125–127, 126f 18 F-fluorodeoxyglucose positron emission tomography, to assess cardiac function, 182 FFR (fractional flow reserve), 230 FHS (Framingham Heart Study), aortic atherosclerosis in, 342–343 Fibrin-targeted contrast agent, 85, 86f Fibroelastoma, papillary, 534, 536f, 544t Fibroma(s), 534–536, 537f, 544t pediatric, 117, 118f pericardial, 495 Fibromuscular dysplasia, renal artery stenosis due to, 470 Fibrous cap, in atherosclerosis, 341 ruptured, 341–342 CMR imaging of, 343–344 Fick principle, 595 Fick’s law, 569 FID (free induction decay), 4f, 5 in CMR spectroscopy, 556–557 Field of view (FOV), 10 Field strength high (See High field CMR)

Cardiovascular Magnetic Resonance 627

INDEX

Echocardiography (Continued) technique of, 197 uses of, 197 for viability studies, 204 for pulmonary artery hypertension, 482t for right ventricular assessment, 382 transesophageal for aortic dissection, 453–454 of aortic intramural hematoma, 454–455 of pericardial disease, 488 of thoracic aortic aneurysm, 457 transthoracic of atrial septal defect, 398–399 of cardiomyopathy, 515 of congenital heart disease, 408, 415 of constrictive pericarditis, 493 of pericardial disease, 488 of pericardial effusions, 492 Eddy currents, 94–95 with interventional CMR, 598 Edema in acute myocardial infarction, 269 and myocardial viability, 268 in myocarditis, 517f, 518 EDV. See End-diastolic volume (EDV). Effusive-restrictive pericarditis, 493 Ejection fraction (EF) intraobserver, interobserver, and interstudy variability for measurement of, 187f normal values for, 385–388, 388t in females, 385–388, 387t, 390f in males, 385–388, 386t, 389f reference ranges for, 190t for right ventricular assessment, 383–385 Elastic limit, 363 Elastic modulus aortic, 369 of vascular wall, 363 Elasticity, of blood vessel, 363 Elastin fibers, in arterial wall, 362–363 Electrical safety, with interventional CMR, 598–599 Electrocardiographic (ECG) electrodes, safety of, 103, 104–105, 105f Electrocardiographic (ECG) gating for atherosclerotic plaques of coronary artery, 351–352 for congenital heart disease, 408 for coronary sinus flow assessment, 311 in pediatric CMR, 117, 119f in PET, 182 for postoperative atrial switch, 423, 423f prospective vs. retrospective, 188 for right ventricular assessment, 382–383 in SPECT, 182 for thoracic aorta, 450, 451f, 468 Electrocardiographic (ECG) setup, for patient monitoring, 104–105, 105f Electrocardiographic (ECG) synchronization, for interventional CMR, 602 Electronic implants, as contraindication to CMR, 105–106 Electrophysiology, interventional CMR for, 586 Embolism, pulmonary, 480–481 catheter-based X-ray pulmonary angiography of, 480 CE-CTA for, 480 CE-MRA of, 481, 481f, 482f CMR lung perfusion imaging of, 481, 481f current workup for, 480 incidence of, 480 pulmonary artery hypertension with, 482f ventilation/perfusion scanning for, 480 Enalaprilat, for ventricular remodeling, 262 End-diastolic volume (EDV) after acute myocardial infarction, 255

INDEX

Field strength (Continued) for stress myocardial perfusion imaging, 219 FISP (fast imaging with steady-state precession), to assess cardiac function, 185, 187 FLASH imaging. See Fast low angle shot (FLASH) imaging. Flip angle, in gradient echo imaging, 13–14, 59 Flow data, visualizing, 96–97 Flow enhancement methods, 91, 92f Flow mapping of coarctation of the aorta, 400–402, 402f, 458–459 of congenital heart disease, 396 of thoracic aorta, 450–451, 451f Flow pressure maps, 96–97, 97f Flow vector map, 96, 97f Flow velocity, phase of signal and, 92 Flow velocity images, in phase contrast velocity mapping, 92–93, 93f Flow wave, foot of, 366, 366f Flow wave velocity aortic compliance and, 368, 368f CMR of, 364–367, 365f, 366f, 367f defined, 364–366 Flow-related enhancement, 463 Flow-related signal loss, 94, 95f Flow/velocity imaging, 32–34 cardiac gating for, 32 goal of, 32, 33f pulse sequence in, 33–34, 33f Fluorine-19 (19F) CMR, catheter visualization and localization using, 598, 601f Fluorodeoxyglucose positron emission tomography (FDG PET), to assess cardiac function, 182 Fluorodeoxyglucose (FDG) uptake, in chronic myocardial infarction vs. late gadolinium enhancement, 276–277, 281f and myocardial wall thickness, 275, 275f and viable epicardial rim thickness, 276, 280f Fluoroscopy, X-ray, vs. interventional CMR, 580, 581t Fluvastatin, and aortic compliance, 368–369 Fontan baffle, with single ventricle, 123, 123f, 125f Fontan procedure, 416 aortopulmonary, 432, 432f classic, 432, 432f extracardiac, 432, 432f fenestrated, 432, 432f for single ventricle, 125f, 432, 432f postoperative assessment of, 434–435, 434f variations on, 432, 432f Fossa ovalis defect, 398, 398f, 399f Fourier flow imaging, 91, 93 velocity phase encoding in, 93, 93f visualizing flow data in, 96, 96f Fourier velocity imaging, 96, 96f phase contrast velocity mapping and, 94 for valvular heart disease, 507 Fourier velocity-encoded measurement, of aortic flow wave velocity, 366–367, 367f FOV (field of view), 10 Fractional flow reserve (FFR), 230 Frame rate, for ventricular function, 149 Framingham Heart Study (FHS), aortic atherosclerosis in, 342–343 Free breathing methods, 129–131 in coronary artery CMR, 286, 286t, 289 with native vessel stenosis, 302–303, 303t for coronary sinus flow assessment, 311–312 mean diaphragm displacement in, 130f respiratory trace data for, 130f

628 Cardiovascular Magnetic Resonance

Free induction decay (FID), 4f, 5 in CMR spectroscopy, 556–557 Frequency encoding, 7, 7f, 9–10, 9f pulse sequence diagram with, 10, 10f Frequency encoding direction, 9–10 Frequency encoding gradients, 7–8, 9–10, 9f Friedreich ataxia, CMR spectroscopy for, 562 FSE imaging. See Fast spin echo (FSE) imaging. Functional imaging, high field CMR for, 170, 171f Functional real-time fetal CMR, 125–127, 126f Fundamental law of magnetostimulation, 102

G

Gadobenate dimeglumine (Gd-BOPTA, MultiHance), 81, 85 binding and relaxivity features of, 81t, 83t chemical structure of, 80f for coronary artery CMR, 293, 293f safety of, 87–88 Gadobutrol (Gd-DO3A-butrol, Gadovist), 79f, 79t safety of, 87–88 Gadocoletic acid (B22956), 81 binding and relaxivity features of, 81, 81t chemical structure of, 80f factors affecting relaxivity of, 84 Gadodiamide (Gd-DTPA-BMA, Omniscan), 79f, 79t relaxivity with, 83t safety of, 87–88 Gadofluorine(s), 85, 86f Gadofluorine M, for atherosclerotic plaques of aorta and carotid arteries, 345–346 Gadofosveset trisodium (MS-325, Vasovist, Ablavar), 80, 85 binding of, 81, 81t chemical structure of, 80f for coronary artery CMR, 293–294 relaxivity of, 81, 81t, 83t factors affecting, 84 magnetic field dependence on, 82, 82f safety of, 87 for stress myocardial perfusion imaging, 217 for three-dimensional MRA, 464 Gadolinium (Gd)-based CMR, for pediatric imaging, 116–118 of arrhythmogenic right ventricular cardiomyopathy, 118 of coronary artery, 117, 119f late gadolinium enhancement, 117, 118f myocardial and blood tagging in, 116, 119f myocardial perfusion imaging as, 116, 118f, 120 saturation band in, 116, 119f static, 116, 117f technical considerations with, 120 time-resolved, 116, 118f of tumor or mass, 117 Gadolinium (Gd) chelates for stress myocardial perfusion imaging, 216–217, 217f for three-dimensional MRA, 464 Gadolinium (Gd) contrast agents, 78–79, 79f, 79t safety of, 87–88, 103–104 Gadolinium diethylenetriamine pentaacetic acid (Gd-DTPA, gadopentetate dimeglumine, Magnevist), 79f, 79t for atherosclerotic plaques of aorta and carotid arteries, 346

Gadolinium diethylenetriamine pentaacetic acid (Gd-DTPA, gadopentetate dimeglumine, Magnevist) (Continued) for cardiac allograft rejection, 548, 550 relaxivity with, 83t compartmentalization and, 84 magnetic field dependence on, 82, 82f safety of, 87–88, 103 for stress myocardial perfusion imaging, 217f for three-dimensional MRA, 464 for ventricular remodeling, 257–258 Gadolinium enhancement, late. See Late gadolinium enhancement (LGE). Gadolinium (Gd)-mesoporphyrin, after acute myocardial infarction, 259 Gadolinium-enhanced T1-weighted CMR, for cardiomyopathy, 516 Gadolinium-enhanced three-dimensional cardiovascular MRA, 464–465 gadolinium chelates for, 464, 465f pulse sequences for, 464 vessel brightness (T1) vs. gadolinium dose in, 464–465 Gadomer-17 binding of, 81, 81t chemical structure of, 80f relaxivity of, 81, 81t, 83t molecule size and, 82 Gadopentetate dimeglumine. See Gadolinium diethylenetriamine pentaacetic acid (GdDTPA, gadopentetate dimeglumine, Magnevist). Gadoterate (Gd-DOTA, Dotarem), 79f, 79t compartmentalization and relaxivity with, 84 safety of, 87–88 Gadoteridol (Gd-HPDO3A, ProHance), 79f, 79t relaxivity with, 83t safety of, 87 Gadoversetamide (Gd-DTPA-BMEA, Optimark), 79f, 79t safety of, 87–88 Gadovist (gadobutrol), 79f, 79t safety of, 87–88 Gadoxetic acid (Gd-EOB-DTPA, Primovist, Eovist), 85 Gating, cardiac. See Cardiac gating. Gd. See Gadolinium (Gd). Gd-BOPTA. See Gadobenate dimeglumine (Gd-BOPTA, MultiHance). Gd-DO3A-butrol (gadobutrol), 79f, 79t safety of, 87–88 Gd-DOTA (gadoterate), 79f, 79t compartmentalization and relaxivity with, 84 safety of, 87–88 Gd-DTPA. See Gadolinium diethylenetriamine pentaacetic acid (Gd-DTPA, gadopentetate dimeglumine, Magnevist). Gd-DTPA-BMA (gadodiamide), 79f, 79t relaxivity with, 83t safety of, 87–88 Gd-DTPA-BMEA (gadoversetamide), 79f, 79t safety of, 87–88 Gd-EOB-DTPA (gadoxetic acid), 85 Gd-HPDO3A (gadoteridol), 79f, 79t relaxivity with, 83t safety of, 87 General anesthesia, for pediatric CMR, 120 General Electric (GE), CMR terminology used by, 611 Generalized autocalibrating partially parallel acquisition (GRAPPA), 45–46, 45f

H

Half-Fourier single-shot fast spin echo (HASTE) for coronary artery bypass graft, 330–331, 334f in morphology scanning goal of, 21–22, 23f physiology of, 22, 26f Hand, peripheral vascular disease of, 475 Harmonic phase (HARP) technique for dobutamine stress CMR, 207–208 for left ventricular systolic function, 149 HCM. See Hypertrophic cardiomyopathy (HCM). 1 H-CMRS. See Proton CMR spectroscopy (1H-CMRS). HDL (high-density lipoprotein), in atherosclerotic plaques of aorta and carotid arteries, 346–347, 347f HDL (high-density lipoprotein)-based contrast agent, 85 Headphones, for CMR, 103 Hearing impairment, due to CMR, 103 Heart disease congenital (See Congenital heart disease [CHD]) valvular (See Valvular heart disease)

Heart failure CMR spectroscopy in, 561–563, 562f right ventricular assessment in, 388–391 Heart rates, in pediatric CMR, 120 Heart transplantation. See Cardiac transplantation. Heart valve(s) disease of (See Valvular heart disease) mechanical (prosthetic), 512–513 safety of CMR with, 101, 101f, 106 Heating-related injury from CMR imaging, 102 from interventional CMR, 582–583, 583t, 598–599 Hemangioma(s) cardiac, 536, 538f, 544t pericardial, 495 Hematoma, aortic intramural, 454–455, 455f Hemi-Fontan procedure, for single ventricle, 124, 124f Hemopericardium, 491–492 Hemorrhagic infarcts, 259 Hemorrhagic pericardial effusions, 495 Hibernating myocardium, 267 High field CMR, 170–178 of coronary artery, 172–173, 173f, 294, 294f of coronary artery wall, 174, 174f for dobutamine stress CMR, 203 for functional imaging, 170, 171f for late gadolinium enhancement imaging, 174 limitations of, 170, 171f for magnetic resonance angiography, 466, 466f for magnetic resonance spectroscopy, 175, 565, 565f for myocardial stress perfusion imaging, 174 for myocardial tagging, 170–171, 172f for oxygen-sensitive myocardial MRI, 576, 577f for parallel imaging, 175 rationale for, 170 High-density lipoprotein (HDL), in atherosclerotic plaques of aorta and carotid arteries, 346–347, 347f High-density lipoprotein (HDL)-based contrast agent, 85 High-energy phosphates, and myocardial viability, 268 Highly constrained back projection (HYPR), 52f, 53 Hitachi, CMR terminology used by, 611 HMG-CoA (3-Hydroxy-methylglutaryl coenzyme A) reductase inhibition, in ventricular remodeling, 261 Hooke’s law, 363 Horizontal long axis (HLA) image, 140, 141f, 186–187 for left ventricular function and size, 142 Hydatid cysts, 536, 539f Hydrogen spins, 3, 4f 3-Hydroxy-methylglutaryl coenzyme A (HMGCoA) reductase inhibition, in ventricular remodeling, 261 Hyperpolarized contrast media, for stress myocardial perfusion imaging, 217–218 Hypertension CMR spectroscopy in, 560–561 pulmonary artery (See Pulmonary artery [PA] hypertension) Hypertrophic cardiomyopathy (HCM), 518–520, 519f CMR spectroscopy for, 565 follow-up for, 520 function and morphology in, 519

Hypertrophic cardiomyopathy (HCM) (Continued) LVOT obstruction in, 518, 519–520 tissue characterization in, 519 Hypoplastic left heart syndrome aortic arch, pulmonary artery, and venous pathway imaging with, 123–124, 123f myocardial and blood tagging for, 119f staged palliation of, 433, 433f, 434f HYPR (highly constrained back projection), 52f, 53

I

ICAM-1 (intercellular adhesion molecule 1), in atherosclerosis of aorta and carotid artery, 345 of coronary artery, 358 IMA (internal mammary artery) graft, 329 Image-selected in vivo spectroscopy (ISIS), for cardiac allograft rejection, 551–552, 552f Imaging planes, and cardiac anatomy, 140, 141f, 143f IMH (intramural hematoma), aortic, 454–455, 455f Implantable cardioverter defibrillators (ICDs), safety of CMR with, 102, 107–108 Impulse response, in quantitative evaluation of myocardial perfusion, 62–63 Inducible nitric oxide synthase (iNOS), in ventricular remodeling, 260 Induction heating, with interventional CMR, 598 Infants, CMR in. See Pediatric CMR. Infarct(s), hemorrhagic vs. nonhemorrhagic, 259 Infarct expansion, after acute myocardial infarction, 253, 274 Infarct resorption, after acute myocardial infarction, 258, 258f Inferior vena cava filter, interventional CMR for, 589 Inflammation, with atherosclerotic plaques of aorta and carotid arteries, 346–347, 347f of coronary artery, 358 Infundibulum, of right ventricle, 381 Inner-sphere relaxivity, 81–82 Inorganic phosphate (Pi), in 31P-CMRS, 557, 558, 558f iNOS (inducible nitric oxide synthase), in ventricular remodeling, 260 Inotropic stress, for comprehensive CMR assessment of coronary artery disease, 159–160 In-plane velocity mapping of congenital heart disease, 113 of coronary artery, 316–317, 317f Interatrial septum, lipomatous hypertrophy of, 142, 146f, 533–534, 535f, 544t Intercellular adhesion molecule 1 (ICAM-1), in atherosclerosis of aorta and carotid artery, 345 of coronary artery, 358 Intercept Internal CMR guidewire coil, 587 Interleaved spiral imaging, for coronary artery velocity mapping, 317–320, 319f, 320f, 321f Intermediate exchange, in contrast-enhanced tissue relaxation, 84 Internal mammary artery (IMA) graft, 329 Interrupted aortic arch, 428–430 vs. aortic arch atresia, 428 epidemiology of, 428 in infant and pediatric patients classification of, 428, 429f evaluation of, 429

Cardiovascular Magnetic Resonance 629

INDEX

Generalized encoding matrix (GEM), 45f, 46 Glagov effect, in atherosclerosis, 341, 353–354 Glenn shunt, 416 for single ventricle, 123, 124, 433, 433f Gradient(s), 7–8 Gradient coils, 3, 7–8 Gradient magnetic fields, bioeffects of, 598 Gradient recalled echo (GRE), 10, 13–14, 14f of aorta, 468–469 thoracic, 450–451 for cardiac and paracardiac masses, 532 for congenital heart disease, 396, 408 of coronary artery, 289–290 with coronary artery bypass graft, 305–306, 306t conventional, 330, 332f three-dimensional respiratory gated, 331 for coronary artery velocity mapping, 315–317 breath hold technique for, 315–316 navigator techniques for, 316–317, 317f, 318f with native vessel stenosis, 301–302, 301f, 303t fast, 14, 14f inversion recovery, 15 for stress myocardial perfusion imaging, 218 three-dimensional, 15 for myocardial perfusion imaging single-shot, 58, 59f with steady-state free precession, 59 for right ventricular assessment, 383, 384f spoiled, with contrast agents, 77, 78f for stress tests, 198, 199t for ventricular volumes, 150–151, 151t Gradient strengths, 7–8 GRAPPA (generalized autocalibrating partially parallel acquisition), 45–46, 45f GRE. See Gradient recalled echo (GRE) Great arteries relationship of, 409 transposition of (See Transposition of the great arteries [TGA]) Grid-tagged images, in CSPAMM, 70, 71f, 72f Guidewires, in interventional CMR, 600 Gyromagnetic ratio, 3

INDEX

Interrupted aortic arch (Continued) postoperative assessment of, 430, 430f preoperative assessment of, 429, 429f with truncus arteriosus, 426, 427f, 428f surgical repair of, 428–429 Interventional CMR, 580–592 application(s) of cardiac, 585–587, 605 for atrial and ventricular mapping, 586 for atrial septal defect, 399 for atrial transseptal procedures, 585f, 586 for congenital heart disease, 396, 415–417 future directions in, 605, 605f for invasive coronary artery imaging and intervention, 586–587 for patent ductus arteriosus, 400 recent progress in, 605, 605f for RF ablation, 586, 603, 604, 604f for targeted local delivery of cellular agents to myocardium, 585–586, 586f for valve replacement and repair, 587, 605f for ventricular septal defect, 397 extracardiac, 587–589 for aortic aneurysm and aortic dissection repair, 587, 588f for aortic coarctation stent repair, 584f, 587 for inferior vena cava filter, 589 for invasive arterial imaging, 587 for peripheral artery disease, 589, 589f for transjugular intrahepatic portosystemic shunt, 587–589 catheter devices for, 583, 595–598 active, 584–585, 585f, 596–598 advantage of, 596–597 for aortic coarctation repair, 584f for endomyocardial injection, 586f 19 F, 598, 601f for intramyocardial injection, 598f with multiple resonant coils, 597–598, 600f safe transmission line for, 599f semi-, 597–598 for transseptal puncture, 585f vs. conventional XRF devices, 583, 583f passive, 583–584, 595–596 chemical-selective visualization of, 584, 584f CO2-filled balloon in, 595, 596f, 597f dysprosium oxide, 595, 596f ideal material for, 595 that create bright signals, 583–584, 584f that create dark signals, 583 strategy for, 585 communication and monitoring with, 580–581 ECG synchronization for, 602 laboratory for, 580–585, 581f magnetic instrumentation and visualization strategies for, 595–598 merit(s) of, 593–595 improved visualization of cardiac anatomy as, 593–594 physiologic information as, 595 reduced ionizing radiation as, 594 pediatric, 593–610 reduced ionizing radiation in, 594 real-time, 581t, 583

630 Cardiovascular Magnetic Resonance

Interventional CMR (Continued) safety consideration(s) with, 582–583, 598–600 bioeffects of magnetic fields as, 598 heating and electrical safety as, 465, 583t, 598–599 magnetic force and torque as, 600 scanner interface for, 582, 582f system design for, 593, 594f in thoracic aorta, 459–460 vs. ultrasound, 580, 581t in XMR system, 593, 601–604 for biventricular pacing, 605f early experience in humans with, 603–604, 604f facility design for, 594f, 601–602 image registration in, 604 performance of, 602–603, 603f safety features for, 594f, 601 vs. X-ray fluoroscopy, 580, 581t Interventricular septum, rhabdomyoma of, 538f Intestinal ischemia, 472 Intima, of artery, 362–363 Intracardiac thrombus, 540–542, 542f, 544t Intracoronary stents coronary artery CMR with, 294–295, 294f safety of CMR with, 106 Intramural hematoma (IMH), aortic, 454–455, 455f Intramyocardial injection, active catheter tracking and visualization for, 598f Intramyocardial segment shortening, in dobutamine stress CMR, 205 Intravascular contrast media, for stress myocardial perfusion imaging, 217 Inversion recovery curve, in viability imaging, 31, 32f Inversion recovery fast gradient recalled echo, 15 Inversion recovery pulse sequence, 14f, 15 for acute myocardial infarction, 269, 269f Inversion recovery technique for acute myocardial infarction, 242 for cardiac and paracardiac masses, 532 Inversion time (TI), 14f, 15 Iron oxide–based contrast agents cross-linked, 86–87 in development, 86–87 relaxivity with, 83 structure of, 83 uses of, 81 Ischemia myocardial (See Myocardial ischemia) subendocardial vulnerability to, 244 Ischemic bed at risk, in acute myocardial infarction, 271, 272f Ischemic cascade, 230 Ischemic heart disease. See also Myocardial ischemia. CMR spectroscopy in, 563–565 for myocardial viability assessment, 564–565, 564f for stress testing, 563–564, 563f right ventricular assessment in, 391 ISIS (image-selected in vivo spectroscopy), for cardiac allograft rejection, 551–552, 552f Isolated ventricular inversion, 409 Isomerism, 409

J

Jatene procedure, for transposition of the great arteries, 410–411, 416 Jet flow, 504, 505f Jet velocity mapping, for stenotic valvular heart disease, 501, 506

K

Kawasaki disease, 299–301, 300f mural thrombosis in, 353, 354f Killer gradient, 14, 14f k-space, 11–12, 11f k-space data, raw, 11–12, 11f in cine CMR, 24–25, 27f in scout scanning, 20–21, 22f k-space gradient echo imaging, of coronary artery, 289–290 k-space ordering, for navigator echoes, 131, 131t, 132f k-space signal, 11f k-t broad-use linear acquisition speed-up technique (k-t BLAST) applications of, 50f, 51 to assess cardiac function, 185 for myocardial perfusion imaging, 61 principles of, 47–48 k-t sensitivity encoding (k-t SENSE) applications of, 51 to assess cardiac function, 185 for stress myocardial perfusion imaging, 218

L

LA (left atrium), morphology of, 408 Larmor equation, 3 Late gadolinium enhancement (LGE) imaging, 31 cardiac gating for, 31, 32f of cardiomyopathy dilated, 516–517 hypertrophic, 519 of congenital heart disease, 117, 118f, 408 of coronary artery atherosclerotic plaques, 354–355, 355f in coronary artery disease for detection of disease analysis of, 164, 167f protocols for, 160–164, 160t, 161f, 162f, 163f for viability assessment analysis of, 164–166, 167f protocols for, 163–164, 165f, 166f, 167f for dobutamine stress CMR, 205, 207f goal of, 31, 31f high field CMR for, 174 in infarcted tissue, 268 inversion recovery in, 31, 32f of myocardial infarction acute, 242, 243f, 269–271 adjustment of T1 in, 271, 273f ischemic bed at risk in, 271, 272f and left ventricular remodeling, 258, 258f, 259f nulling of signal intensity of normal myocardium in, 269, 270f and recovery of function, 273–274 T1-weighted segmented inversion recovery pulse sequence for, 269, 269f time between contrast injection and imaging in, 269–271, 271f validation of, 243–244, 244t chronic, 276, 277f, 278f, 279f of myocarditis, 517f, 518 in parallel imaging, 50f, 51 for pulmonary artery hypertension, 484–485, 485f of pulmonary veins, 447, 447f with single ventricle, 124f LDL (low-density lipoprotein), in atherosclerosis, 341 Lecompte maneuver, 120–121, 121f Left atrium (LA), morphology of, 408

Left ventricular (LV) stroke volume, validation of, 186f Left ventricular (LV) systolic function, reference ranges for, 189t, 191f Left ventricular (LV) thrombosis, due to acute myocardial infarction, 248, 249f Left ventricular (LV) volume after acute myocardial infarction, 253–255 measurement of, 185f reference ranges for, 189t, 191f Left ventricular (LV) wall motion assessment cine CMR of, 22–25 acquisition time in, 24 in acute myocardial infarction, 241, 242f cardiac gating for, 24–25, 27f goal of, 23–24, 27f wall motion stress CMR for (See Wall motion stress CMR) Left ventricular (LV) wall thickness, and myocardial viability, 267 Left-sided isomerism, 409 Leiomyomatosis, with intracardiac extension, 536, 539f, 544t Leiomyosarcoma, 537, 544t Levo-transposition of the great arteries, 410–411, 411f, 423 LGE. See Late gadolinium enhancement (LGE). Lipoma(s), 533–534 atrial, 535f endocardial, 533–534 epidemiology of, 533–534 pediatric, 117 pericardial, 495 subepicardial, 533–534 tissue characterization of, 544t Lipomatous hypertrophy, of atrial septum, 142, 146f, 533–534, 535f, 544t Liposarcoma, 540, 541f, 544t L-loop transposition of the great arteries, 410–411, 411f, 423 L-NAME (N-methyl-L-arginine methyl ester), for ventricular remodeling, 261 Loeffler’s eosinophilic endocarditis, 524–526, 525f morphology and function in, 526 tissue characterization in, 526 Longitudinal direction, 3 Low-density lipoprotein (LDL), in atherosclerosis, 341 L-transposition of the great arteries, 410–411, 411f, 423 Lung, T1 and T2 values for, 7t Lung perfusion imaging for pulmonary artery hypertension, 483–484, 483f for pulmonary embolism, 481, 481f Lung transplantation, right ventricular assessment in, 391–392 LV. See Left ventricle (LV). LVOT. See Left ventricular outflow tract (LVOT). Lymphoma, 540, 541f, 544t

M

Macrophage(s) in atherosclerotic plaques of aorta and carotid arteries, 346 in cardiac allograft rejection, 548 MPIO-labeled, 549, 549f USPIO-labeled, 548–549, 549f Macrophage Scavenger Receptor-A (MSR-A), for atherosclerotic plaques of aorta and carotid arteries, 346 Magnet strength, 3–5, 4f

Magnetic field(s) alignment with main, 3 bioeffects of, 598 in CMR spectroscopy, 556–557 safety of, 100, 598 radiofrequency, 102–103 rapidly switched, 101–102 Magnetic force, in interventional CMR, 600 Magnetic resonance angiography (MRA), 34–35, 463–479 of aorta, 468–470 thoracic, 451–452, 452f of aortic aneurysms, 470 thoracic, 456–457, 457f of aortic dissection, 452–453, 454, 454f, 470 basic principles and techniques of, 463–466 for cardiac allograft rejection, 554t cardiac gating for, 34–35 contrast-enhanced (See Contrast-enhanced magnetic resonance angiography [CE-MRA]) of coronary artery bypass graft three-dimensional contrast-enhanced breath hold, 332, 335f, 336f three-dimensional respiratory gated, 331 two-dimensional breath hold, 330–331, 333f, 334f of coronary artery disease, 159, 160t, 161, 162f of extracranial carotid arteries, 467–468, 467f, 468f fast gradient echo in, 15 gadolinium-enhanced 3D, 464–465 gadolinium chelates for, 464, 465f pulse sequences for, 464 vessel brightness (T1) vs. gadolinium dose in, 464–465 goal of, 34, 34f of mesenteric arteries, 472–473 non-contrast approaches to, 463 of peripheral vessels, 473–475, 475f phase contrast, 94, 464 pulse sequence in, 35 of renal arteries, 470–472 of renal artery stenosis, 472 at 3-Tesla, 466 time-of-flight, 463 of extracranial carotid arteries, 467–468, 467f timing of image acquisition in, 34, 35f Magnetic resonance imaging (MRI) alignment with main magnetic field in, 3 balanced steady-state free precession in, 15, 15f basic principles of, 1–18 echo planar imaging, spiral, and radial in, 16–17, 16f frequency encoding: position in X in, 9–10, 9f, 10f gradient echo imaging in, 13–14, 14f gradients in, 7–8 image creation in, 7, 7f inversion recovery fast gradient recalled echo: late gadolinium enhancement in, 15 phase encoding: position in Y in, 10–11, 10f pulse sequences and contrast in, 12–13, 12f radiofrequency and magnet strength in, 3–5, 4f raw k-space data and fast Fourier transform in, 11–12, 11f selective excitation: position in Z in, 8–9, 8f, 9f

Cardiovascular Magnetic Resonance 631

INDEX

Left ventricle (LV) morphology of, 409 single, 430, 431f Left ventricular (LV) anatomy, during remodeling, 253–255, 254f Left ventricular (LV) diastolic function CMR tagging assessment of, 69–75 apical rotation in, 71f, 73–74, 73f, 73t cardiac motion and, 69–70 complementary spatial modulation of magnetization for, 70–71 echo planar imaging in, 70–71 grid-tagged images in, 70, 71f, 72f limitations of, 74 slice thickness in, 70 slice-following principle in, 70, 70f spiral imaging in, 70–71 steady-state free precession in, 70–71 with strain measurements, 74 3D, 71 in dobutamine stress CMR, 208f, 209 evaluation of motion data from, 71–72, 73f, 73t limitations of, 74 methods for, 70–72 during remodeling, 255–256 results of, 72–74 strain measurement in, 74 normal values for, 190, 192f Left ventricular ejection fraction (LVEF), after acute myocardial infarction, 255 Left ventricular (LV) function, 149 in acute myocardial infarction, 241–242, 242f CMR assessment of, 183–185, 183f short axis slices in, 183–184, 184f, 188 Simpson’s rule method for, 181–182, 184f global, 149 imaging planes for, 141f, 142 regional systolic, 149, 150f during remodeling, 255–256 Left ventricular (LV) hypertrophy, CMR spectroscopy for, 563 Left ventricular (LV) mass, 142, 150 after acute myocardial infarction, 254–255 effect of imaging sequence and magnetic field strength on, 150–152, 151t papillary muscles and endocardial trabeculae in, 188 reference ranges for, 189t, 191f validation of, 186f Left ventricular outflow tract (LVOT), volume flow through, in aortic stenosis, 507 Left ventricular outflow tract (LVOT) obstruction, in hypertrophic cardiomyopathy, 518, 519–520 Left ventricular outflow tract (LVOT) view, 140, 141f Left ventricular (LV) remodeling, after acute myocardial infarction, 253–266 CMR spectroscopy of energetics during, 256–257, 257f contrast-enhanced CMR and predictors of, 257–259, 258f, 259f early phase of, 253 late phase of, 253, 254f pathophysiology of, 253 regional left ventricular function during, 255–256 therapy for animal studies of, 259–262, 261f human studies of, 262–264, 263f ventricular anatomy during, 253–255, 254f

INDEX

Magnetic resonance imaging (MRI) (Continued) signal detection in, 3–17 spin echo imaging in, 6–7, 6f, 7t fast (turbo), 13 double inversion recovery (black-blood), 13, 13f T1 relaxation, 5 T2 relaxation and spin phase in, 5–6, 5f three-dimensional fast gradient echo: magnetic resonance angiography in, 15 Magnetic resonance (MR) signal, detection of, 3–17 Magnetic resonance spectroscopy (MRS), cardiovascular. See Cardiovascular magnetic resonance spectroscopy (CMRS). Magnetization preparation, for stress myocardial perfusion imaging, 218–219, 219f Magnetization recovery, in myocardial perfusion scanning, 28, 29f Magnetization transfer contrast for assessment of myocardial perfusion, 61–62 in coronary artery CMR, 288, 288f Magnetohydrodynamic effect, 104–105, 105f Magnetostimulation, fundamental law of, 102 Magnevist. See Gadolinium diethylenetriamine pentaacetic acid (Gd-DTPA, gadopentetate dimeglumine, Magnevist). Malignant cardiac tumor(s), 537–540 lymphoma as, 540, 541f, 544t metastatic, 540 sarcoma as, 537–540 angio-, 537, 540f, 544t leiomyo-, 537, 544t lipo-, 540, 541f, 544t Marfan syndrome, aortic distensibility and stiffness in, 369 Matrix metalloproteinases (MMPs), in atherosclerotic plaques of aorta and carotid arteries, 346 Maximal intensity projection (MIP), for pulmonary veins, 441, 442f Maxwell gradients, 94–95 MBF (myocardial blood flow), 57 measurement of, 311, 313f MBF (myocardial blood flow) reserves, 311–312, 313f MDCT. See Multidetector computed tomography (MDCT). Mean transit time (MTT), in stress myocardial perfusion imaging, 221 Mechanical heart valves, 512–513 safety of CMR with, 101, 101f, 106 Mechanical restraint devices, for ventricular remodeling, 261 Media, of artery, 362–363 Menotropin, and aortic compliance, 369 MERIT-HF study, 262 MESA (Multiethnic Study of Atherosclerosis), aortic atherosclerosis in, 342–343 Mesenteric arteries, MRA of, 472–473 Mesenteric artery stenosis, 472–473 intestinal ischemia due to, 472 MRA of, 472–473 contrast-enhanced, 473 noncontrast CMR of, 473 X-ray angiography of, 472 Mesenteric ischemia, 472 Mesocaval shunt, interventional CMR for, 589 Mesoscopic inhomogeneities, in contrastenhanced tissue relaxation, 84–85 Metal artifacts, 142–145, 146–148, 147f

632 Cardiovascular Magnetic Resonance

Metallic shard injuries, as contraindication to CMR, 105–106 Metastasis, pericardial, 542–543 Metastatic cardiac tumors, 540 Metoprolol for ventricular remodeling, 260, 262 MI. See Myocardial infarction. Micrometer-sized paramagnetic iron oxide (MPIO), for cardiac allograft rejection, 549, 549f Microvascular obstruction (MO), after acute myocardial infarction contrast-enhanced CMR of, 257–258 late gadolinium enhancement of, 258–259, 259f no-reflow phenomenon and, 271–272, 274f pathophysiology of, 253 prognostic significance of, 246 and regional recovery of function, 246–247 residual coronary occlusion vs., 245–246, 245f, 246f Microvessels, in coronary flow reserve, 230 MIP (maximal intensity projection), for pulmonary veins, 441, 442f Missile effect, with ferromagnetic objects, 101 Mitral regurgitant fraction, 512 Mitral regurgitant volume, 512 Mitral regurgitation, 510–512 asymmetric, 510, 512f central, 510, 511f quantification of, 512 severity of, 502t Mitral stenosis, 502t, 508 Mitral valve replacement, interventional CMR for, 587 Mitral valve stenosis, 502t, 508 M-mode echocardiography, to assess cardiac function, 181–182, 182f MMPs (matrix metalloproteinases), in atherosclerotic plaques of aorta and carotid arteries, 346 MO. See Microvascular obstruction (MO). Moderator band, 381 Modified Simpson’s rule, 149 Molecular imaging, of atherosclerotic plaques in aorta and carotid arteries, 344–347 angiogenesis in, 345 endothelial dysfunction in, 344–345 extracellular matrix in, 345–346 plaque inflammation in, 346–347, 347f thrombus in, 345, 346f in coronary arteries, 356, 356f angiogenesis in, 358–359 inflammation in, 358 thrombosis in, 356–358, 357f, 358f Monitoring during CMR, 104–105, 105f in interventional CMR laboratory, 580–581 Monophosphate esters (MPE), in 31P-CMRS, 557 Morphology scanning, 21–22 black-blood imaging in cardiac gating for, 26f goal of, 21–22, 23f physiology of, 22, 26f bright-blood imaging in, 21–22, 24f cardiac gating for, 22, 25f goal of, 21–22, 23f, 24f HASTE in, 21–22, 23f pulse sequence in, 21–22, 25f steady-state free precession imaging in, 21–22, 24f Motion artifacts cardiac, 142–145, 146, 147f respiratory, 142–145, 146, 147f navigator echoes for (See Navigator echoes) in pediatric CMR, 120 Motion models, for navigator echoes, 137

MPEs (monophosphate esters), in 31P-CMRS, 557 MPIO (micrometer-sized paramagnetic iron oxide), for cardiac allograft rejection, 549, 549f MPO (myeloperoxidase), contrast agent sensitive to, 85–86 MPR (multiplanar reconstruction), for congenital heart disease, 113, 114f MR (magnetic resonance) signal, detection of, 3–17 MRA. See Magnetic resonance angiography (MRA). MRI. See Magnetic resonance imaging (MRI). MR-Imaging for Myocardial Perfusion Assessment in Coronary Artery Disease Trial (MR-IMPACT), 224–226 MRS (magnetic resonance spectroscopy), cardiovascular. See Cardiovascular magnetic resonance spectroscopy (CMRS). MS-325. See Gadofosveset trisodium (MS-325, Vasovist, Ablavar). MSCT (multislice computed tomography), for right ventricular assessment, 382 MSR-A (Macrophage Scavenger Receptor-A), for atherosclerotic plaques of aorta and carotid arteries, 346 MTT (mean transit time), in stress myocardial perfusion imaging, 221 Multicontrast CMR, of atherosclerotic plaques of aorta and carotid artery, 342–344, 342f, 343f with automatic segmentation, 342, 344f with complications, 343–344 imaging sequences for, 342, 342t with pharmacologic therapy, 343 in subclinical disease, 342–343 validation of, 342 Multidetector computed tomography (MDCT) to assess cardiac function, 182–183 vs. coronary artery CMR, 304–305, 305f, 305t vs. parallel MRI, 42–45, 44f of pericardial disease, 488 Multi-echo imaging, 16, 16f Multiethnic Study of Atherosclerosis (MESA), aortic atherosclerosis in, 342–343 MultiHance. See Gadobenate dimeglumine (GdBOPTA, MultiHance). Multiplanar reconstruction (MPR), for congenital heart disease, 113, 114f Multiple breath hold methods, 129–130 mean diaphragm displacement in, 130f respiratory trace data for, 130f Multiple column orientations, for navigator echoes, 133–136, 135f Multiple excitations, in pediatric CMR, 120 Multislice computed tomography (MSCT), for right ventricular assessment, 382 Muscular dystrophy, Becker, CMR spectroscopy for, 565 Mustard procedure, 415, 416–417 MVO2. See Myocardial oxygen consumption (MVO2). Mycotic aortic aneurysms, 469–470 Myeloperoxidase (MPO), contrast agent sensitive to, 85–86 Myocardial blood flow (MBF), 57 measurement of, 311, 313f Myocardial blood flow (MBF) reserves, 311–312, 313f Myocardial BOLD MRI in clinical setting, 572–573, 573f emerging techniques for, 574–576, 574f, 575f, 576f, 577f

Myocardial oxygenation assessment (Continued) high field, 576, 577f myocardial perfusion reserve in, 573 without pharmacologic stress, 572–573, 573f in preclinical setting, 571–572, 572f rationale for, 570 vs. SPECT, 572, 573f SSFP-based, 574–576, 574f, 575f, 576f myocardial perfusion reserve in, 569, 570f vasodilators in, 570–571 Myocardial perfusion imaging, 25–31, 57–68 acceleration techniques for, 60–61 applications of, 57 of congenital heart disease, 116, 118f, 120 contrast agent injection in, 66 contrast residue detection in, 66 echo planar imaging for, 58–59 endogenous contrast for, 61–62 first pass imaging with exogenous tracers in, 58–60 goal of, 25–27, 28f gradient echo imaging for single-shot, 58, 59f with steady-state free precession, 59 image acquisition in, 27–28, 29f magnetization recovery in, 28, 29f measurement of arterial input in, 66 vs. other imaging modalities, 57–58, 60f parallel imaging for, 50f, 51, 60–61, 66 physiologic basis for, 57–58 practical aspects of, 66 pulse sequence in, 28–29, 30f quantitative evaluation of, 62–65, 63f, 64f arterial input function in, 65–66, 65f rest, 31 with single ventricle, 124f spatial resolution in, 66 stress (See Stress myocardial perfusion imaging) T1 fast acquisition relaxation mapping for, 59–60, 60f T1-weighted techniques for, 58 T2*-weighted techniques for, 58–59 temporal resolution of measurements in, 66 up-slope method in, 66 water exchange and its effect on contrast enhancement in, 61 Myocardial perfusion reserve, for myocardial oxygenation assessment, 569, 570f, 573 Myocardial siderosis, 524 morphology and function in, 524 tissue characterization in, 524 Myocardial stunning, 267 Myocardial tagging for congenital heart disease, 116, 119f high field CMR for, 170–171, 172f for left ventricular diastolic function, 69–75 apical rotation in, 71f, 73–74, 73f, 73t cardiac motion and, 69–70 complementary spatial modulation of magnetization for, 70–71 echo planar imaging in, 70–71 grid-tagged images in, 70, 71f, 72f limitations of, 74 slice thickness in, 70 slice-following principle in, 70, 70f spiral imaging in, 70–71 steady-state free precession in, 70–71 with strain measurements, 74 3D, 71 evaluation of motion data from, 71–72, 73f, 73t limitations of, 74 methods for, 70–72

Myocardial tagging (Continued) during remodeling, 255–256 results of, 72–74 strain measurement in, 74 for left ventricular systolic function, 149 during remodeling, 255–256 for right ventricular assessment, 388 Myocardial viability, 267–283 in acute myocardial infarction, 269–274 contrast-enhanced studies for, 269–274 late gadolinium enhancement for, 269–271 adjustment of T1 in, 271, 273f ischemic bed at risk in, 271, 272f nulling of signal intensity of normal myocardium in, 269, 270f and recovery of function, 273–274 T1-weighted segmented inversion recovery pulse sequence for, 269, 269f time between contrast injection and imaging in, 269–271, 271f no-reflow phenomenon and, 271–272, 274f T2-weighted images for, 269 wall thickness and, 274 assessment of, 31 cardiac gating for, 31, 32f CMR spectroscopy for, 564–565, 564f goal of, 31, 31f inversion recovery in, 31, 32f parallel imaging for, 50f, 51 in chronic myocardial infarction, 274–280 contractile reserve and, 275–278 late gadolinium enhancement for, 276, 277f, 278f, 279f vs. dobutamine CMR, 277–278 vs. other imaging modalities, 276–277, 281f thickness of epicardial rim and recovery of function in, 276, 280f wall thickness and, 275, 275f CMR spectroscopy for, 278–280 contractile reserve and, 267–268 defined, 267 feature(s) of, 267–269 contractile reserve as, 267–268 early hypoenhancement with gadolinium as, 268 late gadolinium enhancement in infarcted tissue as, 268 left ventricular wall thickness as, 267 no-reflow phenomenon as, 268 scar formation as, 267 tissue edema as, 268 high-energy phosphates and, 268–269 sodium and potassium CMR for, 268–269 Myocardial wall thickness, and myocardial viability, in chronic myocardial infarction, 275, 275f Myocarditis, 517–518, 517f combined protocols for, 517f, 518 early enhancement in, 517f, 518 edema in, 517f, 518 follow-up for, 518 function and morphology in, 517f, 518 late gadolinium enhancement in, 517f, 518 tissue characterization in, 517f, 518 Myocardium hibernating, 267 T1 and T2 values for, 7t targeted local delivery of cellular agents to, 585–587, 586f vulnerable, 351 Myxoma, 533, 534f, 544t

Cardiovascular Magnetic Resonance 633

INDEX

Myocardial BOLD MRI (Continued) future of, 577 high field, 576, 577f myocardial perfusion reserve in, 573 without pharmacologic stress, 572–573, 573f in preclinical setting, 571–572, 572f rationale for, 570 vs. SPECT, 572, 573f SSFP-based, 574–576, 574f, 575f, 576f for stress myocardial perfusion imaging, 215–216 Myocardial contrast enhancement, water exchange and its effects on, 61 Myocardial edema and myocardial viability, 268 in acute myocardial infarction, 269 in myocarditis, 517f, 518 Myocardial function, 141f, 142, 143f Myocardial infarction acute (See Acute myocardial infarction [AMI]) chronic (See Chronic myocardial infarction [CMI]) parallel imaging of, 50f, 51 viability imaging for, 31 cardiac gating for, 31, 32f goal of, 31, 31f inversion recovery in, 31, 32f Myocardial ischemia, 57–58 absolute vs. relative, 230–231 in coronary artery disease, 158–159 during dobutamine stress CMR, 198, 201f contemporary studies of, 202–203, 202t, 203f, 203t early studies of, 201–202, 202f ischemic cascade in, 230 noninvasive imaging of, 230–231 spatiotemporal disparity in, 230 stress testing for, 231–239 contraindications and termination criteria for, 236, 236t, 237t cost of, 238 coverage with, 237 diagnostic performance of, 238–239, 238f drug interactions in, 236 duration of examination with, 237 functional assessment of viable myocardium with, 237–238 image display and analysis for, 237 imaging protocols for, 232, 233f, 234f monitoring during, 236–237 patient evacuation and emergency equipment for, 237 pharmacologic effects of, 231–232, 231t pitfalls and advanced issues with, 237–239 practicability of, 236–237 route and duration of administration in, 238 safety aspects of, 232–236 Myocardial oxygen consumption (MVO2) estimation of, 569 maximal, 229 measurement of, 569 resting, 229 Myocardial oxygenation assessment, 569–579 contrast for, 570 coronary flow and coronary arterial pressure in, 569, 570f historical background of, 569 myocardial BOLD MRI for in clinical setting, 572–573, 573f emerging techniques for, 574–576, 574f, 575f, 576f, 577f future of, 577

INDEX

N 23

Na CMR spectroscopy. See Sodium-23 (23Na) CMR spectroscopy. Navigator(s), in coronary artery CMR with atherosclerotic plaques, 352–353, 353f for gating and slice tracking, 287–288 of native vessel stenosis, 302–303, 303t other forms of, 136–137 for triggering alone, 286–287, 287f for velocity mapping, 316–317, 317f, 318f Navigator acceptance window, 129–130 Navigator echoes, 129–139 accept-reject algorithm for, 131, 131t, 132f column positioning for, 133, 135f, 135t column selection for, 131–132 computer architecture for, 137 correction factors for, 132–133, 133f, 134f diminishing variance algorithm for, 131, 131t, 132f free breathing methods for, 129–131 mean diaphragm displacement in, 130f respiratory trace data for, 130f history of, 129 implementation of, 131–133 k-space ordering for, 131, 131t, 132f more recent approaches to, 136–137 motion models for, 137 multiple breath hold methods for, 129–130 mean diaphragm displacement in, 130f respiratory trace data for, 130f multiple column orientations for, 133–136, 135f navigator timing for, 133–136, 136f other forms of navigators for, 136–137 precision of measurement with, 136 uses for, 129–131, 130f Navigator timing, 133–136, 136f Navigator-based respiratory gating, in pediatric CMR, 117, 119f, 120 Navigator-gated imaging, of postoperative atrial switch, 423, 424f Negative contrast agents, 76 Neovascularization, in atherosclerosis, 341 of aorta and carotid artery, 345 of coronary artery, 358–359 Nephrogenic systemic fibrosis (NSF), gadolinium contrast agents and, 87 Neurostimulation systems, as contraindication to CMR, 105–106 Nicorandil, for ventricular remodeling, 260–261 90 pulse, 3–5 Nitrates effect on coronary artery blood flow of, 285, 286f for ventricular remodeling, 260 Nitric oxide synthase (NOS), inducible, in ventricular remodeling, 260 Nitric oxide synthase (NOS) inhibitor, for ventricular remodeling, 260, 261 N-methyl-L-arginine methyl ester (L-NAME), for ventricular remodeling, 261 Noise pixels, 95 Noise reduction, during CMR, 103 Non-Cartesian paths, 37 Noncompaction cardiomyopathy, 522, 522f characteristics of, 522 function and morphology in, 522 tissue characterization in, 522 Nonselective radiofrequency pulse, 8 No-reflow phenomenon after acute myocardial infarction, 257–258 and myocardial viability, 271–272, 274f defined, 268 and myocardial viability, 268

634 Cardiovascular Magnetic Resonance

No-reflow phenomenon (Continued) after acute myocardial infarction, 271–272, 274f NOS (nitric oxide synthase), inducible, in ventricular remodeling, 260 NOS (nitric oxide synthase) inhibitor, for ventricular remodeling, 260, 261 NSF (nephrogenic systemic fibrosis), gadolinium contrast agents and, 87 Nuclear cardiology, to assess cardiac function, 182 Nuclear spins, 3, 4f Nulling, of signal intensity of normal myocardium, in acute myocardial infarction, 269, 270f

O

Obesity, CMR spectroscopy with, 560–561 Oblique sagittal planes, uses for, 140–142, 143f Off-resonance effects, in spiral imaging, 38, 39f Off-resonance spins, 6 Omniscan (gadodiamide), 79f, 79t relaxivity with, 83t safety of, 87–88 One-dimensional chemical shift imaging (1DCSI), for cardiac allograft rejection, 551–552 Optical pumping, with contrast agents, 87 Optimark (gadoversetamide), 79f, 79t safety of, 87–88 Ostium primum defect, 398, 398f Ostium secundum defect, 398, 398f, 399f Outer-sphere relaxivity, 81–82 Oxygen consumption, myocardial maximal, 229 resting, 229 Oxygen-sensitive myocardial imaging, 569–579 contrast for, 570 coronary flow and coronary arterial pressure in, 569, 570f historical background of, 569 myocardial BOLD MRI for in clinical setting, 572–573, 573f emerging techniques for, 574–576, 574f, 575f, 576f, 577f future of, 577 high field, 576, 577f myocardial perfusion reserve in, 573 without pharmacologic stress, 572–573, 573f in preclinical setting, 571–572, 572f rationale for, 570 vs. SPECT, 572, 573f SSFP-based, 574–576, 574f, 575f, 576f myocardial perfusion reserve in, 569, 570f vasodilators in, 570–571 Oxyhemoglobin, in BOLD technique, 62

P 31

P CMR spectroscopy. See Phosphorus-31 (31P) CMR spectroscopy. P947, for atherosclerotic plaques of aorta and carotid arteries, 346 PA. See Pulmonary artery(ies) (PA). Pacemakers, safety of CMR with, 101, 102, 107–108, 108f Papillary fibroelastoma, 534, 536f, 544t Papillary muscles, in left ventricular mass, 188 Parallel imaging, 42–53 applications of, 49–53 for assessment of global and regional cardiac function, 47f, 49–51, 185 for coronary artery, 50f, 51–53

Parallel imaging (Continued) for detection of myocardial infarction and assessment of myocardial viability, 50f, 51 for imaging of cardiac anatomy and structure, 49, 50f for myocardial perfusion imaging, 50f, 51, 60–61, 66 for ventricular function, 149 artifacts in, 48–49, 48f coil arrays for, 46 coil sensitivity calibration strategies in, 46, 47f for contrast-enhanced MRA, 466 data acquisition and image reconstruction in, 45–46, 45f dynamic, 47–48 high field, 175 multi-detector-row CT vs., 42–45, 44f principles of, 42–49 signal-to-noise ratio in, 46–47, 47f undersampling in, 46 Parallel imaging with augmented radius in k-space (PARS), 45f, 46 Parasagittal planes, uses for, 140–142, 143f Partial saturation, in CMR spectroscopy, 556–557 Passive catheter tracking and visualization, 583–584, 595–596 chemical-selective, 584, 584f CO2-filled balloon in, 595, 596f, 597f dysprosium oxide, 595, 596f ideal material for, 595 that create bright signals, 583–584, 584f that create dark signals, 583 Patent ductus arteriosus (PDA), 400 CMR evaluation of, 400 dark-blood imaging of, 115f interventional closure of, 400 pathogenesis of, 400 pulmonary arteries in, 485, 485f Patient monitoring during CMR, 104–105, 105f in interventional CMR laboratory, 580–581 PC. See Phase contrast (PC). PCI (percutaneous coronary intervention) CMR-guided, 605 for ventricular remodeling, 262–263 PCr (phosphocreatine), in 31P-CMRS, 557–558, 558f PCr/ATP (phosphocreatine/adenosine triphosphate) ratio in 31P-CMRS, 557, 558, 559 in cardiac allograft rejection, 551–553, 553f PDA. See Patent ductus arteriosus (PDA). PDE (phosphodiesters), in 31P-CMRS, 559, 561f Peak filling rate (PFR), normal values for, 385–388, 388t in females, 385–388, 387t, 390f in males, 385–388, 386t, 387t, 389f Peak flow velocity, 189–190 Pediatric CMR, 111–128, 395–396 adult vs., 111, 395–396 advantages of, 111 anatomic imaging in, 112–113 double inversion recovery dark-blood imaging for, 113, 114f, 115f gadolinium-based, 116–118 of arrhythmogenic right ventricular cardiomyopathy, 118 of coronary artery, 117, 119f late gadolinium enhancement imaging as, 117, 118f myocardial and blood tagging in, 116, 119f

Pediatric CMR (Continued) spatial and temporal resolution as, 118–120 velocity mapping in, 113–115 Percutaneous coronary intervention (PCI) CMR-guided, 605 for ventricular remodeling, 262–263 Perfluorocarbon-based contrast agents, 85, 86f Perfusion defect, after acute myocardial infarction prognostic significance of, 246 and regional recovery of function, 246–247 with residual coronary occlusion vs. microvascular obstruction, 245–246, 245f, 246f Perfusion imaging. See Myocardial perfusion imaging. Perfusion status, after acute myocardial infarction prognostic significance of, 246 and regional recovery of function, 246–247 with residual coronary occlusion vs. microvascular obstruction, 245–246, 245f, 246f Perfusion-related parameters, in stress myocardial perfusion imaging, 220, 221f Pericardial agenesis, 490 Pericardial cavity, 489 Pericardial cysts, 490–491, 491f, 543, 543f, 544t Pericardial defects, 490 Pericardial effusions, 491–493, 493f hemorrhagic, 495 malignant, 495 Pericardial fluid, 489 Pericardial lesions, 542–543 contrast agents for, 532 cysts as, 490–491, 491f, 543, 543f, 544t metastases as, 542–543 technical considerations with, 532, 533t tumors as, 542 Pericardial metastasis, 542–543 Pericardial recess, superior, 142, 146f Pericardial sinuses, 489 Pericardial thickness, 490 Pericardial tubes, 489 Pericardial tumors, 495, 542 primary, 495 secondary malignant, 495 Pericarditis, 491, 492f constrictive, 493–495 chest x-ray of, 493 clinical presentation of, 493 CMR of, 494, 494f CT of, 494, 494f effusive-, 493 etiology of, 493 pericardial thickening in, 493 vs. restrictive cardiomyopathy, 493 transthoracic echocardiography of, 493 ventricular filling pattern in, 495 Pericardium, 488–498 in cardiac homeostasis, 489 fibrous, 489 imaging modalities for, 488–489 normal anatomy of, 489–490, 489f aortic recesses in, 489–490, 490f serous, 489 Peripheral artery disease, interventional CMR for, 589, 589f Peripheral bypass graft patency, 474 Peripheral vascular disease, 473–475 diagnosis of Doppler ultrasonography for, 473 noninvasive techniques for, 473 3D contrast-enhanced MRA for, 474, 475f

Peripheral vascular disease (Continued) for assessment of bypass graft patency, 474 bolus chase technique in, 474, 475f of hand and wrist, 475 sensitivities and specificities with, 474 time-resolved technique in, 474 2D time-of-flight MRA for, 474 X-ray angiography for, 473–474 etiology of, 473 interventional CMR for, 589, 589f risk factors for, 473 Peripheral vessels, MRA of, 473–475, 475f PET. See Positron emission tomography (PET). PFR (peak filling rate), normal values for, 385–388, 388t in females, 385–388, 387t, 390f in males, 385–388, 386t, 387t, 389f Phase of signal, and flow velocity, 92 of spin, 5–6, 5f Phase contrast (PC) MRA, 94, 464 of aorta, 468 of renal artery stenosis, 472 Phase contrast (PC) velocity mapping, 91, 92–93 flow vector map in, 96, 97f flow velocity images in, 92–93, 93f and Fourier velocity imaging, 94 improving accuracy of, 93–97, 94f, 95f rapid, 95–96 of thoracic aorta, 450–451, 451f validation of, 95 for valvular heart disease, 501, 502f, 506 Phase diagram, 5–6, 5f Phase encoding, 7, 7f, 10–11, 10f pulse sequence diagram with, 11–12, 11f Phase encoding gradients, 7–8 Phase flow imaging methods, 91–93 Fourier flow imaging as, 91, 93 velocity phase encoding in, 91, 93 visualizing flow data in, 96, 96f phase contrast velocity mapping as, 91, 92–93 flow vector map in, 96, 97f flow velocity images in, 92–93, 93f and Fourier velocity imaging, 94 improving accuracy of, 93–97, 94f, 95f rapid, 95–96 validation of, 95 principles of, 91–93, 92f rapid, 95–96, 96f Phase ordering, for navigator echoes, 131, 131t, 132f Phase reconstruction, 92 Phase sensitive reconstruction of inversion recovery (PSIR), in parallel imaging, 50f, 51 Phase unwrapping, 92, 93–94, 94f Phase velocity encoding, principles of, 91–93, 92f Phase velocity mapping, for coronary artery velocity, 315–317 breath hold technique for, 315–316 navigator techniques for, 316–317, 317f, 318f Phased-array coils, in stress myocardial perfusion imaging, 220 Philips, CMR terminology used by, 611–611 Phosphates, high-energy, and myocardial viability, 268 Phosphocreatine (PCr), in 31P-CMRS, 557–558, 558f Phosphocreatine/adenosine triphosphate (PCr/ATP) ratio in 31P-CMRS, 557, 558, 559 in cardiac allograft rejection, 551–553, 553f

Cardiovascular Magnetic Resonance 635

INDEX

Pediatric CMR (Continued) myocardial perfusion imaging as, 116, 118f, 120 saturation band in, 116, 119f static, 116, 117f technical considerations with, 120 time-resolved, 116, 118f of tumor or mass, 117 multiplanar reconstruction for, 113, 114f steady-state free precession for, 112–113, 112f anesthesia for, 395–396 cine, 113, 115f, 116f for coarctation of the aorta, 115f, 124–125, 126f of congenital heart disease, 420–438 double-outlet right ventricle as, 425–426 for postoperative assessment, 426 for preoperative assessment, 426, 426f interrupted aortic arch as, 428–430 classification of, 428, 429f for evaluation, 429 for postoperative assessment, 430, 430f for preoperative assessment, 429, 429f postoperative atrial switch, 423–425 contrast-enhanced, 425f ECG-gated SSFP, 423, 423f navigator-gated, 423, 424f single ventricle as, 122–124, 430–435 aortic arch imaging for, 123–124, 123f, 124f cine CMR for, 124, 124f for evaluation, 432 with Fontan baffle, 123, 123f, 125f Fontan procedure for, 432, 432f late gadolinium enhancement for, 124f left, 430, 431f perfusion imaging for, 124f post-Fontan, 434–435, 434f pulmonary artery imaging for, 123, 123f, 125f right, 430, 431f during staged palliation, 433, 433f, 434f time-resolved gadolinium imaging for, 118f velocity mapping for, 123–124 venous pathway imaging for, 123f, 124 with ventricular outflow obstruction, 125f tetralogy of Fallot as, 420–422 for evaluation, 421 for postoperative assessment, 421–422, 422f for preoperative assessment, 421, 421f transposition of the great arteries as, 120–122, 422–423 cine SSFP as, 120–122, 121f dark-blood CMR as, 120–122, 121f three-dimensional contrast-enhanced MRA as, 117f, 120–122, 122f truncus arteriosus as, 426–428 classification of, 426, 427f contrast-enhanced, 428f for postoperative assessment, 427–428 for preoperative assessment, 427, 428f duration of, 395 functional fetal, 125–127, 126f future of, 125–127, 126f general protocol for, 112–118, 112f interventional, 593–610 reduced ionizing radiation in, 594 limitations and challenges of, 111 technical consideration(s) in, 118–120 with gadolinium-based techniques, 120 inability to hold breath as, 120

INDEX

Phosphodiesters (PDE), in 31P-CMRS, 559, 561f Phosphorus-31 (31P) CMR spectroscopy, 556, 557t for cardiac allograft rejection, 551–553, 552f for cardiomyopathy, 516 dilated, 517 clinical studies of, 559–565 in athlete’s heart and hypertension, 560–561 in diabetes and obesity, 561 in healthy subjects, 560, 561f with heart failure and cardiac transplantation, 561–563, 562f in ischemic heart disease, 563–565 for myocardial viability assessment, 564 for stress testing, 563–564, 563f methodologic considerations in, 559–560, 560f, 561f with specific gene defects with cardiac pathology, 565 in valvular heart disease, 563 experimental foundations of, 557–558, 558f during left ventricular remodeling, 257 and myocardial viability, 268, 278–279 physical principles of, 556–557, 557f Pi (inorganic phosphate), in 31P-CMRS, 557, 558, 558f Plaque inflammation, in atherosclerotic plaques of aorta and carotid artery, 346–347, 347f Plaque rupture, coronary stenosis and, 213, 214f Poly-lysine-Gd compounds, for stress myocardial perfusion imaging, 217 Positive contrast agents, 76 Positive remodeling, in atherosclerosis, 341 Positron emission tomography (PET) to assess cardiac function, 182 for cardiac allograft rejection, 554t of myocardial infarction acute, 243–244 chronic vs. late gadolinium enhancement, 276–277, 281f and myocardial wall thickness, 275, 275f and viable epicardial rim thickness, 276, 280f Postoperative atrial switch, 423–425 contrast-enhanced CMR of, 425f ECG-gated SSFP imaging of, 423, 423f navigator-gated imaging of, 423, 424f Post-pericardiotomy syndrome, 491–492 Postprocessing adaptive motion correction technique, 133 Potassium CMR, for myocardial viability, 268–269 PR (projection-reconstruction) imaging, 40, 40f undersampled, 41–42, 41f Precession, steady-state free. See Steady-state free precession (SSFP). Precessional frequency, 3 Primovist (gadoxetic acid), 85 ProHance (gadoteridol), 79f, 79t relaxivity with, 83t safety of, 87 Projectile effect, with ferromagnetic objects, 101 Projection-reconstruction (PR) imaging, 40, 40f undersampled, 41–42, 41f Prosthetic heart valves, 512–513 safety of CMR with, 101, 101f, 106 Proton CMR spectroscopy (1H-CMRS), 556, 557t for cardiomyopathy, 516 dilated, 517

636 Cardiovascular Magnetic Resonance

Proton CMR spectroscopy (1H-CMRS) (Continued) experimental studies with, 559 during left ventricular remodeling, 257 for myocardial viability assessment, 279–280, 564 P-selectin, in atherosclerosis of coronary artery, 358 Pseudoaneurysm of thoracic aorta, 456, 469 ventricular, due to acute myocardial infarction, 249, 249f PSIR (phase sensitive reconstruction of inversion recovery), in parallel imaging, 50f, 51 Psychological effects, of CMR, 103 Pulmonary arteriovenous malformations, 486 Pulmonary artery(ies) (PA), 480–487 in arterial switch procedure, 120–121, 121f CE-MRA of, 481, 481f congenital anomalies of, 485–486, 485f, 486f morphology of, 409 pulse wave velocity of, 370 in tetralogy of Fallot, 420 in truncus arteriosus, 426, 427f, 428f Pulmonary artery (PA) compliance, 362 Pulmonary artery (PA) distensibility, 369, 370f, 371f Pulmonary artery (PA) flow, 152 Pulmonary artery (PA) hypertension, 481–485 CE-MRA for, 482–483, 482f, 483f clinical features of, 481–482 CMR perfusion imaging for, 483–484, 483f CMR vs. other diagnostic techniques for, 481–482, 482t defined, 481–482 early diagnosis of, 481–482 echocardiography for, 482t etiology of, 482 late gadolinium enhancement for, 484–485, 485f with pulmonary embolism, 482f right heart catheterization for, 482t right ventricular assessment in, 391–392 in tetralogy of Fallot, 483f thromboembolic vs. nonthromboembolic, 482 velocity-encoded CMR for, 484–485, 484f Pulmonary artery (PA) imaging, with single ventricle, 123, 123f, 125f Pulmonary artery (PA) stenosis, 410f after repair of transposition of the great vessels, 485, 486f in tetralogy of Fallot, 413 Pulmonary atresia repair of, 416 tetralogy of Fallot with, 420–421, 421f Pulmonary embolism, 480–481 catheter-based X-ray pulmonary angiography of, 480 CE-CTA for, 480 CMR lung perfusion imaging of, 481, 481f contrast-enhanced MRA of, 481, 481f, 482f current workup for, 480 incidence of, 480 pulmonary artery hypertension with, 482f ventilation/perfusion scanning for, 480 Pulmonary regurgitation, 405f, 509–510 etiology of, 509–510 free, 509–510, 511f measurement of, 509–510, 511f severity of, 502t Pulmonary valve, bicuspid, 404, 404f Pulmonary valve implantation, interventional CMR for, 587

Pulmonary valve stenosis, 502t, 508 Pulmonary vascular resistance (PVR), CMR measurement of, 595, 603 Pulmonary vein(s) anatomy of normal, 442–443, 442f variant, 443, 443f, 444f congenital anomalies of, 443–445, 444f embryology of, 441–442 and pathophysiology of atrial fibrillation, 445 quantification of size of, 442f, 446–447, 446f Pulmonary vein atresia, congenital, 444 Pulmonary vein imaging, 439–449 before and after atrial fibrillation ablation, 445–446, 445f, 446f of congenital anomalies, 443–445, 444f image display for, 441, 442f imaging method for, 441 late gadolinium enhancement, 447, 447f of normal anatomy, 442–443, 442f for quantification of size, 442f, 446–447, 446f of variant anatomy, 443, 443f, 444f Pulmonary vein stenosis after atrial fibrillation ablation, 446, 446f congenital, 444 Pulmonary venous return, partial anomalous, 444f, 445 Pulmonary-to-systemic flow ratio (Qp/Qs) with atrial septal defect, 399, 399f with single ventricle, 123–124 with ventricular septal defect, 113, 397, 398f Pulse sequence(s), 12–13, 12f for CMR spectroscopy, 559, 560f for three-dimensional MRA, 464 for vascular angiography, 35 Pulse sequence diagram for fast spin echo imaging, 13, 13f with frequency encoding, 10, 10f for gradient recalled echo imaging, 14, 14f for morphology scanning, 21–22, 25f for myocardial perfusion scanning, 28–29, 30f with phase encoding, 11–12, 11f for scout scanning, 20, 22f with slice selection, 8, 9f for velocity-encoded CMR imaging, 33–34, 33f Pulse wave velocity (PWV) age-related increase in, 368 assessment of, 365f, 366 of pulmonary artery, 370 PVR (pulmonary vascular resistance), CMR measurement of, 595, 603

Q

Qp/Qs (pulmonary-to-systemic flow ratio) with atrial septal defect, 399, 399f with single ventricle, 123–124 with ventricular septal defect, 113, 397, 398f Quantitative coronary angiography (QCA), vs. stress myocardial perfusion imaging, 222, 223t, 224 Quantitative evaluation, of myocardial perfusion, 62–65, 63f, 64f

R

RA (right atrium), morphology of, 408 RA (right atrial) ridge, 142, 146f RACE (real-time acquisition and velocity evaluation), 96 Radial imaging, 16–17, 16f, 40–42 applications of, 42, 43f of coronary artery, 292

Renal arteries, magnetic resonance angiography of, 470–472 Renal artery stenosis, 470–472 diagnosis of, 471 captopril test for, 471 Doppler ultrasonography for, 471 MRA for, 472 contrast-enhanced, 472 phase contrast, 472 radionuclide renography for, 471 X-ray angiography for, 471 epidemiology of, 471 etiology of, 470 renovascular hypertension due to, 470 treatment of, 471 Renography captopril, 471 radionuclide, 471 Renovascular hypertension, 470, 471 Reperfusion, after acute myocardial infarction prognostic significance of, 246 and regional recovery of function, 246–247 with residual coronary occlusion vs. microvascular obstruction, 245–246, 245f, 246f Repetition, 11 Repetition time (TR), 11, 11f, 12 effect on signal of, 12–13, 12f for myocardial perfusion imaging, 58 in pediatric CMR, 120 Rephasing, 6–7, 6f Rephasing gradient, 8–9, 9f Resistance vessels, in coronary flow reserve, 230 Resovist (ferucarbotran), relaxivity with, 83 Respiratory drift, 130 Respiratory efficiency, with navigator echoes, 129–130 Respiratory feedback monitor, 130 Respiratory gating for coronary artery bypass graft, 331 retrospective, 130–131, 131t, 132f for coronary artery velocity mapping, 317, 318f Respiratory motion artifacts, 142–145, 146, 147f in coronary artery CMR, 285–286, 286t with atherosclerotic plaques, 352–353, 353f navigator echoes for (See Navigator echoes) in pediatric CMR, 120 Rest function, in comprehensive CMR assessment of coronary artery disease, 160t Rest myocardial perfusion imaging, 31 Rest perfusion imaging, of coronary artery disease for disease detection, 160t, 161 for viability studies, 166f Restrictive cardiomyopathy (RCM), 521–522 vs. constrictive pericarditis, 493 morphology and function in, 522 tissue characterization in, 522 Retrospective respiratory gating (RRG), 130–131, 131t, 132f for coronary artery velocity mapping, 317, 318f Revascularization, predicting myocardial response to, 276, 277f, 278f, 279f RF. See Radiofrequency (RF). Rhabdomyoma, 536, 538f, 544t Right atrial (RA) ridge, 142, 146f Right atrium (RA), morphology of, 408 Right heart catheterization, for pulmonary artery hypertension, 482t Right ventricle (RV) in congenital heart disease, 417 double-chambered, 508 jet flow in, 504–505, 505f

Right ventricle (RV) (Continued) double-outlet, 410f, 412 defined, 425 epidemiology of, 425 features of, 425 in infant and pediatric patients, 425–426 postoperative assessment of, 426 preoperative assessment of, 426, 426f surgical management of, 425 late complications after, 425–426 ventricular septal defect in, 412, 425 morphology of, 409 single, 430, 431f Right ventricular (RV) anatomy assessment of (See Right ventricular [RV] assessment) normal, 381–385 during remodeling, 253–254, 255 Right ventricular (RV) assessment advantages of CMR for, 382 in arrhythmogenic RV cardiomyopathy, 391 in congenital heart disease, 391 in heart failure, 388–391 imaging strategies for, 382–385 ECG gating in, 382–383 gradient echo (white-blood) cine imaging as, 383, 384f myocardial tagging as, 388 other techniques in, 385 Simpson’s rule in, 383–385, 386f spin echo (black-blood) sequences in, 383, 383f transaxial plane in, 383, 384f importance of, 381 in ischemic heart disease, 391 normal values in, 385–388 for RV diastolic function and atrioventricular plane descent, 385–388 for females, 385–388, 387t, 390f for males, 385–388, 387t, 389f for RV volumes, systolic function, and mass, 385–388 for females, 385–388, 387t, 390f for males, 385–388, 386t, 389f summary data of, 385–388, 388t in pulmonary hypertension and lung transplantation, 391–392 techniques for, 382 Right ventricular (RV) dimensions, assessment of. See Right ventricular (RV) assessment. Right ventricular ejection fraction (RVEF) after acute myocardial infarction, 255 normal values for, 385–388, 388t in females, 385–388, 387t, 390f in males, 385–388, 386t, 389f Right ventricular (RV) failure, causes of, 381 Right ventricular (RV) function, 150, 185 assessment of (See Right ventricular [RV] assessment) prognostic value of, 381 Right ventricular (RV) infarction, 391 Right ventricular (RV) mass normal values for, 385–388, 388t in females, 385–388, 387t, 390f in males, 385–388, 386t, 389f reference ranges for, 190t validation of, 186f Right ventricular outflow tract (RVOT) aneurysm, in tetralogy of Fallot, 421–422, 422f Right ventricular outflow tract (RVOT) obstruction, in tetralogy of Fallot, 420–421 Right ventricular outflow tract (RVOT) stenosis, 508

Cardiovascular Magnetic Resonance 637

INDEX

Radial imaging (Continued) principles of, 39f, 40–42, 40f undersampling in, 41–42, 41f Radiofrequency (RF) ablation, CMR guidance for, 603, 604, 604f Radiofrequency (RF) excitation, 3–5, 4f Radiofrequency (RF) magnetic fields, safety of, 102–103 Radiofrequency (RF) pulse, nonselective, 8 Radiofrequency (RF) radiation, bioeffects of, 598 Radionuclide angiography, for right ventricular assessment, 382 Radionuclide renography, 471 Radionuclide ventriculography, to assess cardiac function, 182 Rapid phase flow imaging methods, 95–96, 96f Rapid volumetric imaging, 52f, 54 RCM. See Restrictive cardiomyopathy (RCM). Readout gradient, 9–10, 9f Real-time acquisition and velocity evaluation (RACE), 96 Real-time CMR imaging, 581t, 583 Real-time prospective slice following, 132–133, 134f Receiver coil, 3 Reflected waves, 367–368 Refocusing, 6–7, 6f Refocusing echo, 7, 7f Regional function, assessment of, 190–192 Regional left ventricular function, during remodeling, 255–256 Regurgitant valvular heart disease, 509–512 aortic, 502t, 509, 510f classification of severity of, 502t flow measurements for, 506–507 general principles for, 509 mitral, 502t, 510–512, 511f, 512f pulmonary, 502t, 509–510, 511f quantification of regurgitation volume in, 403f, 405f for multiple valves, 509 for single valve, 509 surgical intervention for, 509 tricuspid, 502t, 512, 513f Relaxation, 5 Relaxation rate, with contrast agents, 76–77 Relaxivity, of contrast agents, 81–83 correlation time in, 82 effect and definition of, 76, 77f effect of correlation time and field strength on, 82, 83, 83f electronic relaxation in, 82 inner- and outer-sphere, 81–82 iron oxide–based, 83 longitudinal and transverse, 82, 83f magnetic field dependence on, 82, 82f magnetic moment in, 82 molecule size and, 82 for selected media, 83, 83t Remodeling, after acute myocardial infarction, 253–266 CMR spectroscopy of energetics during, 256–257, 257f contrast-enhanced CMR and predictors of, 257–259, 258f, 259f early phase of, 253 late phase of, 253, 254f pathophysiology of, 253 regional left ventricular function during, 255–256 therapy for animal studies of, 259–262, 261f human studies of, 262–264, 263f ventricular anatomy during, 253–255, 254f

INDEX

Right ventricular (RV) stroke volume after acute myocardial infarction, 255 measurement of, 189–190 validation of, 186f Right ventricular (RV) systolic function, reference ranges for, 190t Right ventricular (RV) volume measurement of, 150–151, 152t normal values for, 190t, 385–388 in females, 385–388, 387t, 390f in males, 385–388, 386t, 389f Right ventricular (RV) wall motion assessment, cine CMR for, 22–25 acquisition time in, 24 cardiac gating for, 24–25, 27f goal of, 23–24, 27f Right-sided isomerism, 409 R-R interval, in pediatric CMR, 120 RRG (retrospective respiratory gating), 130–131, 131t, 132f for coronary artery velocity mapping, 317, 318f RV. See Right ventricle (RV). RVEF. See Right ventricular ejection fraction (RVEF). RVOT (right ventricular outflow tract) aneurysm, in tetralogy of Fallot, 421–422, 422f RVOT (right ventricular outflow tract) obstruction, in tetralogy of Fallot, 420–421 RVOT (right ventricular outflow tract) stenosis, 508

S

Safety consideration(s), 100–103 auditory, 103 with biologic effects, 100 in interventional CMR, 598 with contrast agents, 87–88, 103–104 with ferromagnetism, 101, 101f, 102f general, 100 with interventional CMR, 582–584, 583t bioeffects of magnetic fields as, 598 heating and electrical safety as, 465, 583t, 598–599 magnetic force and torque as, 600 with pacemakers and implantable cardioverter defibrillators, 101, 102, 107–108, 108f with psychological effects, 103 with radiofrequency time varying field, 102–103 with rapidly switched magnetic fields, 101–102 during stress conditions, 104 with superconducting system, 103 Sagittal plane scout image in, 140, 141f uses for, 140–142, 143f Sagittal scout image, 20, 21f Saphenous vein graft (SVG), 306, 307f, 307t, 329 Sarcoidosis, myocardial involvement in, 523 morphology and function in, 523 tissue characterization in, 523 Sarcoma, 537–540 angio-, 537, 540f, 544t leiomyo-, 537, 544t lipo-, 540, 541f, 544t Saturation, in time-of-flight methods, 91 Saturation band, for congenital heart disease, 116, 119f Saturation factors, in CMR spectroscopy, 556–557

638 Cardiovascular Magnetic Resonance

Saturation pulse, in myocardial perfusion scanning, 28, 29f Scan efficiency, with navigator echoes, 129–130 Scar formation after acute myocardial infarction, 253 and myocardial viability, 267 Scimitar syndrome, 444f, 445 Scout images, for cardiac anatomy, 140, 141f Scout scanning, 19–21 cardiac gating for, 21, 23f goal of, 19–20 image acquisition in, 20, 21f k-space filling in, 20–21, 22f pulse sequence in, 20, 22f Screening form, for CMR, 612–613 Segmentation, automatic, for atherosclerotic plaques of aorta and carotid arteries, 342, 344f Selectins, in atherosclerosis of aorta and carotid artery, 345 of coronary artery, 358 Selective excitation, 8–9, 8f, 9f Semi-active catheter tracking, 597–598 Senning procedure, 415, 416–417 Sensitivity encoding (SENSE) applications of, 50f, 51 high field, 175 k-t applications of, 51 to assess cardiac function, 185 for stress myocardial perfusion imaging, 218 for myocardial perfusion imaging, 60 stress, 218 principle of, 45f, 46 for stress tests, 199t time-adaptive applications of, 51 for stress myocardial perfusion imaging, 218 Sensitivity profiles from an array of coils for encoding and reconstruction in parallel space (SPACE RIP), 46 Septal leaflets, 409 Sequence protocol dataform, for CMR, 614–616 Sequence type, terminology used by various vendors for, 611–611 Shim gradients, in CMR spectroscopy, 556–557 Short axis slices, to assess cardiac function, 183–184, 184f, 188 Short T1 inversion recovery techniques, for cardiomyopathy, 516 SHU555C, relaxivity with, 83 Shunt quantification, for ventricular septal defect, 397, 398f Siderosis, myocardial, 524 morphology and function in, 524 tissue characterization in, 524 Siemens, CMR terminology used by, 611 Signal intensity, effect of contrast agents on, 77, 78f Signal intensity–contrast media concentration relationship, 221 Signal intensity–time curves, in stress myocardial perfusion imaging, 220, 221f Signal loss, flow-related, 94, 95f Signal misregistration, in time-of-flight methods, 91 Signal-to-noise ratio (SNR) in CMR spectroscopy, 556–557 with high field CMR, 170 in parallel imaging, 46–47, 47f in pediatric CMR, 118–120 in stress myocardial perfusion imaging, 218

Simpson’s rule method, 149, 183–184, 184f for right ventricular assessment, 383–385, 386f Simultaneous acquisition of spatial harmonics (SMASH), 45–46, 45f high field, 175 for myocardial perfusion imaging, 60 sinc function, 8, 8f Sinerem (ferumoxtran), 83 Single photon emission computed tomography (SPECT) of acute myocardial infarction, 243–244 to assess cardiac function, 182 for cardiac allograft rejection, 554t vs. dobutamine stress CMR, 201 vs. myocardial BOLD MRI, 572, 573f vs. stress myocardial perfusion imaging, 222, 224–226 Single ventricle, 122–124, 413–415 aortic arch imaging for, 123–124, 123f, 124f cine CMR for, 124, 124f defined, 430 epidemiology of, 430 with Fontan baffle, 123, 123f, 125f in infant and pediatric patients, 430–435 evaluation of, 432 Fontan procedure for, 432, 432f left, 430, 431f post-Fontan, 434–435, 434f right, 430, 431f during staged palliation, 433, 433f, 434f late gadolinium enhancement for, 124f perfusion imaging for, 124f pulmonary artery imaging for, 123, 123f, 125f time-resolved gadolinium imaging for, 118f velocity mapping for, 123–124 venous pathway imaging for, 123f, 124 with ventricular outflow obstruction, 125f Single-ventricle heart disease, 420 Sinus venosus defect, 398, 398f, 399f Skeletal muscle, T1 and T2 values for, 7t Slice following, real-time prospective, 132–133, 134f Slice following principle, in CSPAMM, 70, 70f Slice selection, 7, 7f, 8–9, 8f pulse sequence diagram with, 8, 9f Slice selection gradients, 7–8, 8f in pulse sequence diagram, 8, 9f Slice thickness, 8, 8f to assess cardiac function, 188 in CSPAMM, 70 Slice tracking, 132–133, 134f for coronary artery CMR, 287–288 SLOOP (spectral localization with optimum point spread function), in 31P-CMRS, 560 Slow exchange, in contrast-enhanced tissue relaxation, 84 Small particle iron oxide (SPIO) contrast agents relaxivity with, 83 structure of, 83 uses of, 81 SMASH (simultaneous acquisition of spatial harmonics), 45–46, 45f high field, 175 for myocardial perfusion imaging, 60 SNR. See Signal-to-noise ratio (SNR). Sodium CMR, for myocardial viability, 268–269 Sodium-23 (23Na) CMR spectroscopy, 556, 557t experimental studies with, 559 during left ventricular remodeling, 257 for myocardial viability assessment, 564, 564f SPACE RIP (sensitivity profiles from an array of coils for encoding and reconstruction in parallel space), 46

Steady-state free precession (SSFP) (Continued) in coronary artery stenosis, 301f, 302–303, 302f in CSPAMM, 70–71 in morphology scanning, 21–22, 24f of myocardial function, 141f, 142 for myocardial oxygenation assessment, 574–576, 575f, 576f, 577f for myocardial perfusion imaging, 59 stress, 218 of postoperative atrial switch, 423, 423f for right ventricular assessment, 384f in scout scanning, 20, 22f for stress tests, 198, 199t of valvular heart disease, 502f, 504–505, 505f for ventricular volumes, 150–151, 151t Stem cell transplantation, for ventricular remodeling, 261–262, 263 Stenotic valvular heart disease, 507–508 aortic, 502t, 507 classification of severity of, 502t jet velocity mapping for, 501, 506 mitral and tricuspid, 502t, 508 pulmonary, 502t, 508 of right ventricular outflow tract, 508 subaortic, 508 Stents, safety of CMR with, 106 Sternal wires, safety of CMR with, 101, 102f, 106 Strain measurements, 74 Strain-encoded CMR imaging, for myocardial tagging, 171 Stress agents, 231–239 for comprehensive CMR assessment of coronary artery disease, 159–160 contraindications and termination criteria for, 236, 236t, 237t cost of, 238 coverage with, 237 diagnostic performance of, 238–239, 238f drug interactions with, 236 duration of examination with, 237 functional assessment of viable myocardium with, 237–238 image display and analysis for, 237 imaging protocols for, 232, 233f, 234f monitoring during, 236–237 patient evacuation and emergency equipment for, 237 pharmacologic effects of, 231–232, 231t pitfalls and advanced issues with, 237–239 practicability of, 236–237 route and duration of administration in, 238 safety aspects of, 232–236 Stress CMR, 196–212 adenosine, 198t, 208–209 atropine pharmacokinetics of, 196 safety of, 196–197, 198t dipyridamole, 208–209 dobutamine for acute myocardial infarction, 241 apical and short axis views in, 198, 200f atropine in, 198, 199f, 202–203 cine GRE or SSFP bright-blood images in, 198, 199t delineation of orthogonal left ventricular myocardial segments in, 198, 200f vs. dobutamine stress echocardiography, 202–203, 202t accuracy of, 197 safety of, 196–197, 197t technique of, 197 uses of, 197 for viability studies, 204

Stress CMR (Continued) dopamine infusion protocol for, 198, 199f facilities for, 198, 199f inducible ischemia during, 198, 201f contemporary studies of, 202–203, 202t, 203f, 203t early studies of, 201–202, 202f late gadolinium enhancement for, 205, 207f pharmacokinetics of, 196 for prognosis, 205, 207f vs. radionuclide studies, 204 safety of, 196–197, 197t, 198t sensitivity and specificity of, 203, 203f, 203t technique for, 197–200 3-T, 203 tissue tagging during, 206–208, 208f for viability studies, 204–205, 207f in viability studies, 204–205 contractile reserve in, 204f, 206f end-systolic wall thickening in, 204, 204f intramyocardial segment shortening in, 205 low-dose, 204–205, 237–238 short-axis basal views in, 200f tissue tagging in, 204–205, 207f during exercise, 209–210, 210f on-site medications for, 198t Stress conditions, patient safety during, 104 Stress echocardiography, dobutamine accuracy of, 197 vs. dobutamine stress CMR, 202–203, 202t safety of, 196–197, 197t technique of, 197 uses of, 197 for viability studies, 204 Stress myocardial perfusion imaging, 213–228 after acute myocardial infarction, 247, 247f adenosine for, 29–30 analysis of data from, 219–222, 220f quantitative approach for, 220 absolute tissue perfusion in, 221–222 clinical performance of, 224, 224f perfusion-related parameters in, 220, 221f visual assessment for, 219–220 clinical performance of, 222–226, 223t clinical performance of, 222–226 multicenter studies of, 223t, 224–226, 225f single-center studies of quantitative semiautomatic analysis of, 224, 224f visual interpretation of, 222–223, 223t CMR data readout from, 218 CMR spectroscopy for, 563–564, 563f combined dobutamine wall motion CMR with diagnostic performance of, 238–239, 238f protocol for, 232, 233f, 234f contrast media for endogenous, 215–216 exogenous, 216–218 extravascular, 216–217, 217f hyperpolarized, 217–218 intravascular, 217 of coronary artery disease, 160, 160t, 161 field strength for, 219 high field CMR for, 174 magnetization preparation for, 218–219, 219f options for inducing stress in, 214, 215f, 216f perspectives on, 226 protocol for, 213–214

Cardiovascular Magnetic Resonance 639

INDEX

Spatial encoding, 7–8 Spatial modulation of magnetization (SPAMM), 69–70 for cardiac allograft rejection, 548–549 complementary, 70–71 echo planar imaging in, 70–71 grid-tagged images in, 70, 71f, 72f limitations of, 74 slice thickness in, 70 slice-following principle in, 70, 70f spiral imaging in, 70–71 steady-state free precession in, 70–71 with strain measurements, 74 3D, 71 Spatial resolution in coronary artery CMR, 288, 288f in pediatric CMR, 118–120 Spatiotemporal disparity, in myocardial ischemia, 230 SPECT. See Single photon emission computed tomography (SPECT). Spectral localization with optimum point spread function (SLOOP), in 31P-CMRS, 560 Spectroscopy, CMR. See Cardiovascular magnetic resonance spectroscopy (CMRS). Speed, increased, 37, 52f, 53–54 Spin(s), 3, 4f off-resonance, 6 Spin echo imaging, 6–7, 6f, 7t of acute myocardial infarction, 269–274 of cardiac anatomy, 140, 143f for congenital heart disease, 396, 408 of coronary artery, 12–13 of coronary artery bypass graft, 330, 331f fast (turbo), 6–7, 13 double inversion recovery (black-blood), 13, 13f pulse sequence diagram for, 13, 13f for right ventricular assessment, 383, 383f of thoracic aorta, 450, 451f Spin echo sequence, 11f, 12 Spin exchange, with contrast agents, 87 Spin labeling, for assessment of myocardial perfusion, 62 Spin phase, 5–6, 5f Spin-lattice relaxation, 5 Spin-spin relaxation, 6 SPIO (small particle iron oxide) contrast agents relaxivity with, 83 structure of, 83 uses of, 81 Spiral imaging, 16–17, 16f, 37–40 applications of, 38–40, 39f of coronary artery, 292, 292f, 293f for coronary artery velocity mapping, 317–320, 319f, 320f, 321f in CSPAMM, 70–71 off-resonance effects in, 38, 39f principles of, 37–38, 38f Spoiler gradient, 14, 14f SSFP. See Steady-state free precession (SSFP). Stainless steel implants, safety of, 101 Stanford classification, of aortic dissection, 452–453, 453f Starr-Edwards prosthetic valve, safety of CMR with, 101f, 106 Static magnetic fields, bioeffects of, 598 Steady-state free precession (SSFP) in acute myocardial infarction, 241 balanced, 15, 15f for cardiomyopathy, 515–516 for congenital heart disease, 112–113, 112f, 396 due to transposition of the great arteries, 120–122, 121f

INDEX

Stress myocardial perfusion imaging (Continued) rationale for, 213, 214f specific steps in, 31 stress-only vs. stress-rest, 213–214, 215t subendocardial, 224, 224f timeline for, 30–31, 30f vs. wall motion CMR, 229–240 Stress-induced cardiomyopathy, 525f, 526 function and morphology in, 525f, 526 tissue characterization in, 526 Stroke volume(s) (SV), 150 calculation of, 150 defined, 150 effect of imaging sequence and magnetic field strength on, 150–152, 151t, 152t normal values for, 385–388, 388t in females, 385–388, 387t, 390f in males, 385–388, 386t, 389f for right ventricular assessment, 383–385 validation of, 186f Subaortic stenosis, 508 Subendocardial perfusion data, 224, 224f Subendocardial vulnerability, to ischemia, 244 Subinfundibular stenosis, 505f, 508 Superconducting system, safety issues with, 103 Superior pericardial recess, 142, 146f Superparamagnetism, 83 Suppression of signal, from surrounding tissues, in coronary artery CMR, 288, 288f Susceptibility contrast, 84–85 SV. See Stroke volume(s) (SV). SVG (saphenous vein graft), 306, 307f, 307t, 329 Swan-Ganz catheters, as contraindication to CMR, 106 Systolic function, 149–150 assessment of, 189–190 Systolic strain, 74

T

3-T (3-Tesla) systems. See High field CMR. T1 contrast agents, 76–81, 77f, 78f T1 fast acquisition relaxation mapping (T1-FARM), for myocardial perfusion imaging, 59–60, 60f T1 relaxation, 5 effects of contrast agent on, 61 T1 values, 7, 7t T1-weighted segmented inversion recovery pulse sequence, for acute myocardial infarction, 269, 269f T1-weighted techniques, for myocardial perfusion imaging, 58 T2 contrast agents, 76–81, 77f, 78f T2 preparation prepulses, in coronary artery CMR, 288, 288f T2 relaxation, 5–6 T2 values, 7, 7t T2-weighted CMR of acute myocardial infarction, 247–248, 248f and left ventricular remodeling, 259–260 signal intensity changes on, 269 for cardiomyopathy, 516 T2*, 6–7 T2*-weighted techniques, for myocardial perfusion imaging, 58–59 Tagged time-of-flight methods, 91, 92f Takayasu arteritis, 459, 459f Tako-Tsubo cardiomyopathy, 525f, 526 function and morphology in, 525f, 526 tissue characterization in, 526 Tantalum stents, safety of CMR with, 106

640 Cardiovascular Magnetic Resonance

Targeted 3D coronary artery CMR acquisition sequence for, 291, 291f of atherosclerotic plaques, 353–354, 354f, 355f of native vessel stenosis, 301–302, 301f Target-specific imaging, of atherosclerotic plaques in aorta and carotid artery, 344–347 angiogenesis in, 345 endothelial dysfunction in, 344–345 extracellular matrix in, 345–346 plaque inflammation in, 346–347, 347f thrombus in, 345, 346f in coronary arteries, 356, 356f angiogenesis in, 358–359 inflammation in, 358 thrombosis in, 356–358, 357f, 358f Taussig-Bing anomaly, 412, 425, 426 TE (time to echo), 11f, 12 effect on signal of, 12, 12f for myocardial perfusion imaging, 58 TEE. See Transesophageal echocardiography (TEE). Telemetric ECG, for monitoring during CMR, 105 Temporal resolution in pediatric CMR, 118–120 for ventricular function, 149 Temporary pacing wires, as contraindication to CMR, 106 Terminology, used by various vendors, 611–611 3-Tesla (3-T) systems. See High field CMR. Tetralogy of Fallot (TOF), 408–409 characteristics of, 409 cine CMR of, 115f clinical presentation of, 420 dark-blood imaging of, 115f epidemiology of, 409 etiology of, 409 in infant and pediatric patients, 420–422 evaluation of, 421 postoperative assessment of, 421–422, 422f preoperative assessment of, 421, 421f “pink,” 420 pulmonary artery hypertension in, 483f with pulmonary atresia, 420–421, 421f surgical repair of, 416, 420–421 pulmonary regurgitation after, 509–510, 511f TFE (turbo field echo), 14, 14f TGA. See Transposition of the great arteries (TGA). Thallium-201 uptake, and myocardial viability, 276–277 The Open Artery Trial (TOAT), 263 Thermal effects of CMR imaging, 102 of interventional CMR, 582–583, 583t, 598–599 Thin-slab 3D coronary artery CMR, 291–292 Thoracic aorta, 450–462 aneurysm of, 456–457, 456f, 457f aortitis of, 459, 459f coarctation of, 458–459, 458f dissection of, 452–454, 452f, 453f, 454f flow mapping of, 450–451, 451f gradient echo CMR imaging of, 450–451 interventional CMR imaging of, 459–460 intramural hematoma of, 454–455, 455f MRA of, 451–452, 452f penetrating ulcer of, 455–456, 455f spin echo CMR imaging of, 450, 451f trauma to, 457–459, 457f 3-Tesla (3-T) systems. See High field CMR.

Three-dimensional (3D) contrast-enhanced MRA of coronary artery bypass graft, 332, 335f, 336f of transposition of the great arteries, 117f, 120–122, 122f Three-dimensional (3D) coronary artery CMR, 290–292 acquisition sequence for, 290–292 of native vessel stenosis with navigator gating, 302–303, 303t targeted, 301–302, 301f whole heart, 302f, 303, 303t targeted acquisition sequence for, 291, 291f of atherosclerotic plaques, 353–354, 354f, 355f of native vessel stenosis, 301–302, 301f thin-slab, 291–292 whole heart acquisition sequence for, 291, 291f, 292f of native vessel stenosis, 302f, 303, 303t Three-dimensional dobutamine stress CMR, 208 Thromboembolic pulmonary artery hypertension, 482 Thrombosis, left ventricular, due to acute myocardial infarction, 248, 249f Thrombus in atherosclerotic plaques of aorta and carotid artery, 345, 346f of coronary artery, 356–358, 357f, 358f intracardiac, 540–542, 542f, 544t Through-plane velocity mapping of congenital heart disease, 113–115 of coronary artery, 317, 317f, 318f of valvular heart disease, 506–507 TI (time to inversion), 14f, 15 Time to echo (TE), 11f, 12 effect on signal of, 12, 12f for myocardial perfusion imaging, 58 Time to inversion (TI), 14f, 15 Time to repetition (TR), 11, 11f, 12 effect on signal of, 12–13, 12f for myocardial perfusion imaging, 58 in pediatric CMR, 120 Time-adaptive sensitivity encoding (TSENSE) applications of, 51 for stress myocardial perfusion imaging, 218 Time-of-flight (TOF) methods, 91, 92f for coronary artery velocity measurement, 314–315, 315f for MRA, 463 of extracranial carotid arteries, 467–468, 467f for peripheral vascular disease, 474 Time-resolved gadolinium imaging of congenital heart disease, 116, 118f with single ventricle, 118f Time-resolved imaging of contrast kinetics (TRICKS) technique, for extracranial carotid arteries, 467 Time-resolved technique, for peripheral vascular disease, 474 TIPS (transjugular intrahepatic portosystemic shunt), interventional CMR for, 587 Tissue edema, and myocardial viability, 268 in acute myocardial infarction, 269 Tissue relaxation, contrast-enhanced, 84–85 Tissue tagging, during dobutamine stress CMR studies, 206–208, 208f of viability, 204–205, 207f TMLR (transmyocardial laser revascularization), for ventricular remodeling, 261 TOAT (The Open Artery Trial), 263

TSE imaging. See Turbo spin echo (TSE) imaging. TSENSE (time-adaptive sensitivity encoding) applications of, 51 for stress myocardial perfusion imaging, 218 TTE. See Transthoracic echocardiography (TTE). Tumor(s) cardiac (See Cardiac tumor(s)) pericardial, 495, 542 primary, 495 secondary malignant, 495 Turbo field echo (TFE), 14, 14f Turbo FLASH, 14, 14f Turbo spin echo (TSE) imaging, 6–7, 13 of cardiac and paracardiac masses, 532 double inversion recovery (black-blood), 13, 13f of transposition of the great arteries, 410f Two-chamber view, 140, 141f

U

Uhl anomaly cine CMR of, 115f myocardial and blood tagging for, 119f Ulcer, penetrating aortic, 455–456, 455f Ultra-fast flow imaging techniques, 95–96 Ultrasmall particle iron oxide (USPIO) contrast agents for atherosclerotic plaques, of aorta and carotid arteries, 346 for cardiac allograft rejection, 548–549, 549f relaxivity with, 83 structure of, 83 uses of, 81 Ultrasonography Doppler for monitoring during CMR, 105 of peripheral vascular disease, 473 of renal artery stenosis, 471–472 vs. interventional CMR, 580, 581t Unaliasing by Fourier encoding the overlaps using the temporal dimension (UNFOLD) applications of, 51 to assess cardiac function, 185 for myocardial perfusion imaging, 61 principles of, 47–48 Undersampling in parallel imaging, 46 in radial imaging, 41–42, 41f Untwisting time, 71–72, 73t Untwisting velocity, 71–72, 73t Upslope parameter, in stress myocardial perfusion imaging, 220, 221f USPIO contrast agents. See Ultrasmall particle iron oxide (USPIO) contrast agents.

V

Valve replacement and repair, invasive and interventional CMR for, 587 Valvular anatomy cine CMR of, 113, 116f velocity mapping for, 113 Valvular atresia, 402f, 403 Valvular function, normal, 152–155, 154f Valvular heart disease, 403–404, 499–514 advantages of CMR for, 501 with atresia, 402f, 403 bicuspid valve as, 403 aortic, 403–404, 403f pulmonary, 404, 404f biventricular volume and function in, 501 cine imaging of, 501, 502f

Valvular heart disease (Continued) in infants and children, 113, 116f visualization and planimetry of jets by, 504–505, 505f, 510f classification of severity of, 502t CMR spectroscopy for, 563 FLASH imaging for, 505 Fourier CMR velocity traces for, 507 gradient recalled echo CMR of, 403f, 404f invasive and interventional CMR for, 587 mechanical heart valves for, 512–513 phase contrast velocity mapping for, 501, 502f, 506 pressure and volume overload in, 404 regurgitant, 509–512 aortic, 502t, 509, 510f classification of severity of, 502t flow measurements for, 506–507 general principles for, 509 mitral, 502t, 510–512, 511f, 512f pulmonary, 502t, 509–510, 511f quantification of regurgitation volume in, 403f, 405f for multiple valves, 509 for single valve, 509 surgical intervention for, 509 tricuspid, 502t, 512, 513f after repair, 403, 405f slice thickness and visualization of thin structures in, 503, 504f SSFP imaging of, 502f, 504–505, 505f stenotic, 507–508 aortic, 502t, 507 classification of severity of, 502t jet velocity mapping for, 501, 506 mitral and tricuspid, 502t, 508 pulmonary, 502t, 508 of right ventricular outflow tract, 508 subaortic, 508 transverse spin echo CMR of, 402f weaknesses of CMR for, 501 Valvular insufficiency, 403 Valvular prostheses, safety of CMR with, 101, 101f, 106 Valvular regurgitation, 403 Valvular stenosis, 403 Vascular angiography, 34–35 cardiac gating for, 34–35 goal of, 34, 34f pulse sequence in, 35 timing of image acquisition in, 34, 35f Vascular cell adhesion molecule (VCAM), in atherosclerosis of aorta and carotid artery, 344–345 of coronary artery, 358 Vascular Doppler, for monitoring during CMR, 105 Vascular smooth muscle cells (VSMCs), in atherosclerosis, 341–342 Vascular wall stiffness, defined, 363 Vasodilator(s), for myocardial oxygenation assessment, 570–571 Vasodilator stress for comprehensive CMR assessment of coronary artery disease, 159–160 for myocardial perfusion studies, 214, 216f Vasovist. See Gadofosveset trisodium (MS-325, Vasovist, Ablavar). Vastly undersampled isotropic projection reconstruction (VIPR), 41–42 VCAM (vascular cell adhesion molecule), in atherosclerosis of aorta and carotid artery, 344–345 of coronary artery, 358 Vectorcardiography, 105, 146

Cardiovascular Magnetic Resonance 641

INDEX

TOF (tetralogy of Fallot). See Tetralogy of Fallot (TOF). TOF methods. See Time-of-flight (TOF) methods. Torque, in interventional CMR, 600 Toshiba, CMR terminology used by, 611–611 TR. See Time to repetition (TR). Trabeculae carneae, 381 Tracer kinetic model, in quantitative evaluation of myocardial perfusion, 63–65, 64f Transesophageal echocardiography (TEE) of aortic dissection, 453–454 of aortic intramural hematoma, 454–455 dobutamine, of chronic myocardial infarction, 276 of pericardial disease, 488 of thoracic aortic aneurysm, 457 Transjugular intrahepatic portosystemic shunt (TIPS), interventional CMR for, 587 Transmyocardial laser revascularization (TMLR), for ventricular remodeling, 261 Transposition of the great arteries (TGA), 120–122, 409–411 arterial switch procedure for, 410–411, 416 atrial switch procedure for, 415–416 postoperative assessment of, 423–425 contrast-enhanced CMR for, 425f ECG-gated SSFP imaging for, 423, 423f navigator-gated imaging for, 423, 424f cine SSFP of, 120–122, 121f congenitally (physiologically) corrected, 410–411, 411f, 423 D- (complete), 409–410, 410f, 422–423 dark-blood CMR of, 120–122, 121f defined, 422 in infant and pediatric patients, 422–423 L- (levo-, L-loop), 410–411, 411f, 423 pulmonary arteries after repair of, 485, 486f steady-state free precession images of, 411f three-dimensional contrast-enhanced MRA of, 117f, 120–122, 122f types of, 422 Transseptal needle puncture, interventional CMR for, 585f, 586 Transthoracic echocardiography (TTE) of atrial septal defect, 398–399 of cardiomyopathy, 515 of congenital heart disease, 408, 415 of constrictive pericarditis, 493 of pericardial disease, 488 of pericardial effusions, 492 Transverse plane, 3 Transverse relaxation, 6 TRICKS (time-resolved imaging of contrast kinetics) technique, for extracranial carotid arteries, 467 Tricuspid regurgitation, 182f, 417, 512 etiology of, 512 measurement of, 512, 513f severity of, 502t Tricuspid stenosis, 502t, 508 Tricuspid valve, Ebstein anomaly of, 413, 414f, 415f Tricuspid valve atresia, 402f Tricuspid valve stenosis, 502t, 508 Triggering, cardiac. See Cardiac gating. Truncus arteriosus, 412–413, 412f associated anomalies with, 426–427 cine CMR of, 116f defined, 426 epidemiology of, 426 in infant and pediatric patients, 426–428 classification of, 426, 427f contrast-enhanced CMR of, 428f postoperative assessment of, 427–428 preoperative assessment of, 427, 428f surgical repair of, 427

INDEX

Velocity encoded CMR of congenital heart disease for evaluation of function, 417 after repair, 416–417 for pulmonary artery hypertension, 484–485, 484f for valvular heart disease, 403 Velocity encoding value (VENC), for aortic flow, 152 Velocity mapping of congenital heart disease, 113–115 due to single ventricle, 123–124 due to transposition of the great arteries, 121–122 of coronary artery velocity, 314 bolus tagging for, 314 with coronary artery bypass graft, 333–337, 337f, 338f echo planar time-of-flight technique for, 314–315, 315f gradient echo phase, 315–317 breath holding techniques for, 315–316 navigator techniques for, 316–317, 317f, 318f in-plane, 316–317, 317f interleaved spiral phase, 317–320, 319f, 320f, 321f through-plane, 317, 317f, 318f of diastolic function, 190, 192f phase contrast, 91, 92–93 flow vector map in, 96, 97f flow velocity images in, 92–93, 93f and Fourier velocity imaging, 94 improving accuracy of, 93–97, 94f, 95f rapid, 95–96 of thoracic aorta, 450–451, 451f validation of, 95 for valvular heart disease, 501, 502f, 506 with single ventricle, 123–124 of systolic function, 189–190 Velocity phase encoding, in Fourier flow imaging, 91, 93 Velocity phase sensitivity, 92 Velocity sensitivity, in phase contrast velocity mapping, 93–94 Velocity-encoded CMR imaging (VENC), 32–34 cardiac gating for, 32 goal of, 32, 33f pulse sequence in, 33–34, 33f of thoracic aorta, 450–451, 451f VENC. See Velocity-encoded CMR imaging. VENC (velocity encoding value), for aortic flow, 152 Venous pathway imaging, with single ventricle, 123f, 124 Ventilation/perfusion scanning, for pulmonary embolism, 480 Ventricle left (See Left ventricle (LV)) right (See Right ventricle (RV)) single, 122–124, 413–415 aortic arch imaging for, 123–124, 123f, 124f cine CMR for, 124, 124f defined, 430 epidemiology of, 430 with Fontan baffle, 123, 123f, 125f in infant and pediatric patients, 430–435 evaluation of, 432 Fontan procedure for, 432, 432f left, 430, 431f post-Fontan, 434–435, 434f right, 430, 431f during staged palliation, 433, 433f, 434f late gadolinium enhancement for, 124f perfusion imaging for, 124f

642 Cardiovascular Magnetic Resonance

Ventricle (Continued) pulmonary artery imaging for, 123, 123f, 125f time-resolved gadolinium imaging for, 118f velocity mapping for, 123–124 venous pathway imaging for, 123f, 124 with ventricular outflow obstruction, 125f Ventricular abnormalities, complex, 413–415 Ventricular anatomy, during remodeling, 253–255, 254f Ventricular aneurysm, due to acute myocardial infarction, 249 Ventricular fibroma, 118f Ventricular filling patterns, in constrictive pericarditis, 495 Ventricular function, 149–150 left, 149, 150f right, 150 Ventricular inversion, isolated, 409 Ventricular loop, 409 Ventricular mapping, interventional CMR for, 586 Ventricular mass, for right ventricular assessment, 383–385, 386f Ventricular morphology, 409 Ventricular noncompaction (VNC), 522, 522f characteristics of, 522 function and morphology in, 522 tissue characterization in, 522 Ventricular outflow obstruction with single ventricle, 125f with transposition of the great arteries, 121f, 122, 122f Ventricular pseudoaneurysm, due to acute myocardial infarction, 249, 249f Ventricular remodeling, after acute myocardial infarction, 253–266 CMR spectroscopy of energetics during, 256–257, 257f contrast-enhanced CMR and predictors of, 257–259, 258f, 259f early phase of, 253 late phase of, 253, 254f pathophysiology of, 253 regional left ventricular function during, 255–256 therapy for animal studies of, 259–262, 261f human studies of, 262–264, 263f ventricular anatomy during, 253–255, 254f Ventricular septal defect (VSD), 396–397 due to acute myocardial infarction, 249, 249f anatomic delineation of, 397, 397f cine CMR of, 113 clinical manifestations of, 396–397 CMR-guided catheterization and intervention for, 397 with double-outlet right ventricle, 412, 425 location of, 396, 396f saturation band for, 116 shunt quantification in, 397, 398f spontaneous closure of, 396–397 surgical management of, 396–397, 397f in tetralogy of Fallot, 420 in truncus arteriosus, 426 velocity mapping of, 113 Ventricular tachyarrhythmia, CMR-guided RF ablation for, 603 Ventricular volumes effect of imaging sequence and magnetic field strength on, 150–152, 151t, 152t for right ventricular assessment, 383–385, 386f

Ventriculoarterial connection disorder(s), 409–413 double-outlet right ventricle as, 410f, 412 tetralogy of Fallot as, 413 transposition of the great arteries as, 409–411, 410f, 411f truncus arteriosus as, 412–413, 412f Vertical long axis (VLA) image, 140, 141f, 186–187 Viability imaging, 31, 267–283 in acute myocardial infarction, 269–274 contrast-enhanced studies for, 269–274 late gadolinium enhancement for, 269–271 adjustment of T1 in, 271, 273f ischemic bed at risk in, 271, 272f nulling of signal intensity of normal myocardium in, 269, 270f and recovery of function, 273–274 T1-weighted segmented inversion recovery pulse sequence for, 269, 269f time between contrast injection and imaging in, 269–271, 271f no-reflow phenomenon and, 271–272, 274f T2-weighted images for, 269 wall thickness and, 274 cardiac gating for, 31, 32f in chronic myocardial infarction, 274–280 contractile reserve and, 275–278 late gadolinium enhancement for, 276, 277f, 278f, 279f vs. dobutamine CMR, 277–278 vs. other imaging modalities, 276–277, 281f thickness of epicardial rim and recovery of function in, 276, 280f wall thickness and, 275, 275f CMR spectroscopy for, 564–565, 564f for coronary artery disease CMR for, 159 comprehensive CMR assessment for analysis of studies with, 164–166, 167f protocols for, 163–164, 165f, 166f, 167f dobutamine stress CMR for, 204–205 contractile reserve in, 204f, 206f end-systolic wall thickening in, 204, 204f intramyocardial segment shortening in, 205 low-dose, 204–205, 237–238 short-axis basal views in, 205f tissue tagging in, 204–205, 207f feature(s) of, 267–269 contractile reserve as, 267–268 early hypoenhancement with gadolinium as, 268 late gadolinium enhancement in infarcted tissues as, 268 left ventricular wall thickness as, 267 no-reflow phenomenon as, 268 scar formation as, 267 tissue edema as, 268 goal of, 31, 31f high-energy phosphates and, 268–269 inversion recovery in, 31, 32f magnetic resonance spectroscopy for, 278–280 parallel imaging for, 50f, 51 sodium and potassium CMR for, 268–269 Viable myocardium, 267–283 in acute myocardial infarction, 269–274 contrast-enhanced studies for, 269–274 late gadolinium enhancement for, 269–271

W

Wall motion stress CMR, 196–212 adenosine, 198t, 208–209 atropine pharmacokinetics of, 196 safety of, 196–197, 198t dipyridamole, 208–209 dobutamine apical and short axis views in, 198, 200f atropine in, 198, 199f, 202–203 cine GRE or SSFP bright-blood images in, 198, 199t combined adenosine perfusion and diagnostic performance of, 238–239, 238f protocol for, 232, 233f, 234f delineation of orthogonal left ventricular myocardial segments in, 198, 200f vs. dobutamine stress echocardiography, 202–203, 202t accuracy of, 197 safety of, 196–197, 197t technique of, 197 uses of, 197 for viability studies, 204 dopamine infusion protocol for, 198, 199f facilities for, 198, 199f inducible ischemia during, 198, 201f contemporary studies of, 202–203, 202t, 203f, 203t early studies of, 201–202, 202f late gadolinium enhancement for, 205, 207f pharmacokinetics of, 196 for prognosis, 205, 207f vs. radionuclide studies, 204 safety of, 196–197, 197t, 198t sensitivity and specificity of, 203, 203f, 203t technique for, 197–200 3-T, 203 tissue tagging during, 206–208, 208f for viability studies, 204–205, 207f in viability studies, 204–205 contractile reserve in, 204f, 206f end-systolic wall thickening in, 204, 204f intramyocardial segment shortening in, 205 low-dose, 204–205, 237–238 short-axis basal views in, 200f tissue tagging in, 204–205, 207f during exercise, 209–210, 210f on-site medications for, 198t vs. stress myocardial perfusion CMR, 229–240 Wall shear stress (WSS) arterial, 371–374, 374f, 375f in blood flow velocity assessment, 96 Wall thickness, and myocardial viability, 267 after acute myocardial infarction, 274

Wall thickness, and myocardial viability (Continued) in chronic myocardial infarction, 275, 275f Washin/washout methods, 91, 92f Water exchange in contrast-enhanced tissue relaxation, 84 effects on myocardial contrast enhancement of, 61 in stress myocardial perfusion imaging, 222 Water-excitation acquisition, for coronary artery velocity mapping, 317–320, 321f Whole heart 3D coronary artery CMR acquisition sequence for, 291, 291f, 292f of native vessel stenosis, 302f, 303, 303t Windowing, for navigator echoes, 131, 131t, 132f Worksheet, for CMR, 614–616 Wrist, peripheral vascular disease of, 475 WSS (wall shear stress) arterial, 371–374, 374f, 375f in blood flow velocity assessment, 96

X

X position, 7, 7f selection of, 9–10, 9f, 10f X-ray and CMR (XMR) guidance system, 593, 601–604 for biventricular pacing, 605f early experience in humans with, 603–604, 604f facility design for, 594f, 601–602 image registration in, 604 laboratory for, 580 performance of intervention using, 602–603, 603f safety features for, 594f, 601 X-ray angiography of mesenteric arteries, 472 of peripheral vascular disease, 473–474 of pulmonary embolism, 480 of renal artery stenosis, 471 X-ray coronary angiography, vs. dobutamine stress echocardiography and dobutamine stress CMR, 202t, 203 X-ray fluoroscopy (XRF), vs. interventional CMR, 580, 581t

Y

Y position, 7, 7f selection of, 10–11, 10f

Z

Z position, 7, 7f selection of, 8–9, 8f, 9f Z-axis direction, 3

Cardiovascular Magnetic Resonance 643

INDEX

Viable myocardium (Continued) adjustment of T1 in, 271, 273f ischemic bed at risk in, 271, 272f nulling of signal intensity of normal myocardium in, 269, 270f and recovery of function, 273–274 T1-weighted segmented inversion recovery pulse sequence for, 269, 269f time between contrast injection and imaging in, 269–271, 271f no-reflow phenomenon and, 271–272, 274f T2-weighted images for, 269 wall thickness and, 274 assessment of, 31 cardiac gating for, 31, 32f CMR spectroscopy for, 564–565, 564f goal of, 31, 31f inversion recovery in, 31, 32f parallel imaging for, 50f, 51 in chronic myocardial infarction, 274–280 contractile reserve and, 275–278 late gadolinium enhancement for, 276, 277f, 278f, 279f vs. dobutamine CMR, 277–278 vs. other imaging modalities, 276–277, 281f thickness of epicardial rim and recovery of function in, 276, 280f wall thickness and, 275, 275f CMR spectroscopy for, 278–280 contractile reserve of, 267–268 defined, 267 feature(s) of, 267–269 contractile reserve as, 267–268 early hypoenhancement with gadolinium as, 268 late gadolinium enhancement in infarcted tissues as, 268 left ventricular wall thickness as, 267 no-reflow phenomenon as, 268 scar formation as, 267 tissue edema as, 268 high-energy phosphates and, 268–269 sodium and potassium CMR for, 268–269 VIPR (vastly undersampled isotropic projection reconstruction), 41–42 Visceral-atrial rule, 409 Viscoelasticity, 363 VLA (vertical long axis) image, 140, 141f, 186–187 VNC. See Ventricular noncompaction (VNC). VSD. See Ventricular septal defect (VSD). VSMCs (vascular smooth muscle cells), in atherosclerosis, 341–342 Vulnerable myocardium, 351 Vulnerable patient, 351 Vulnerable plaques defined, 351 markers of, 351, 352t noninvasive diagnosis of, 351, 352t