Evidence based cardiology

Evidence-based Cardiology Second edition Evidence-based Cardiology Second edition Edited by Salim Yusuf Ernest L Fa...

Author: Salim Yusuf | John A. Caitns | A. John Camm | Ernest L. Fallen | Bernard J. Gersh

137 downloads 2298 Views 13MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Evidence-based Cardiology Second edition

Evidence-based Cardiology Second edition

Edited by Salim Yusuf

Ernest L Fallen

Heart and Stroke Foundation of Ontario Research Chair, Senior Scientist of the Canadian Institute of Health Research Director of Cardiology and Professor of Medicine, McMaster University, Hamilton Health Sciences, Hamilton, Canada

Professor Emeritus, McMaster University, Faculty of Health Sciences, Hamilton, Canada

Bernard J Gersh John A Cairns Dean, Faculty of Medicine, University of British Columbia, Vancouver, Canada

A John Camm Professor of Clinical Cardiology and Chief, Department of Cardiological Sciences, St George’s Hospital Medical School, London, UK

Consultant in Cardiovascular Diseases and Internal Medicine, Mayo Clinic; Professor of Medicine, Mayo Medical School, Rochester, Minnesota, USA

©BMJ Books 1998, 2003 BMJ Books is an imprint of the BMJ Publishing Group Chapter 27 (Rihal) All figures are © Mayo Foundation All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording and/or otherwise, without the prior written permission of the publishers. Second edition first published in 2003 First edition published in 1998 Second impression 1999 by BMJ Books, BMA House, Tavistock Square, London WC1H 9JR www.bmjbooks.com British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 0 7279 1699 8 Typeset by Newgen Imaging Systems (P) Ltd. Printed and bound by MPG Books, Bodmin, Cornwall

Contents

Contributors Preface to the Second edition Preface to the First edition Glossary Part I: General concepts and critical appraisal Salim Yusuf, Editor 1 What is evidence-based cardiology? PJ Devereaux, R Brian Haynes, Salim Yusuf

xi xvii xix xxi 1 3

2 A critical appraisal of the cardiovascular history and physical examination Akbar Panju, Brenda Hemmelgarn, Jim Nishikawa, Deborah Cook, Allan Kitching

14

3 Obtaining incremental information from diagnostic tests Raymond J Gibbons

23

4 Clinical trials and meta-analysis Colin Baigent

34

5 Finding current best evidence to practice evidence-based cardiology Dereck L Hunt, K Ann McKibbon, R Brian Haynes

40

6 Understanding concepts related to health economics Mark Hlatky

46

7 Introduction to decision analysis Kevin A Schulman, Henry A Glick, Allan S Detsky

56

8 Assessing and changing cardiovascular clinical practices C David Naylor, David A Alter

71

Part II: Prevention of cardiovascular diseases Salim Yusuf, Editor 9 Global perspective on cardiovascular disease K Srinath Reddy

89 91

10 Tobacco: global burden and community solutions Terry F Pechacek, Samira Asma, Nicole Blair, Michael P Eriksen

103

11 Tobacco and cardiovascular disease: achieving smoking cessation Godfrey H Fowler

114

12 Lipids and cardiovascular disease Malcolm Law

121

13 Use of lipid lowering agents in the prevention of cardiovascular disease Jeffrey L Probstfield

130

v

Evidence-based Cardiology

14 Blood pressure and cardiovascular disease Curt D Furberg, Bruce M Psaty

146

15 Glucose abnormalities and cardiovascular disease: “dysglycemia” as an emerging cardiovascular risk factor Sarah E Capes, Hertzel C Gerstein

161

16 Physical activity and exercise in cardiovascular disease prevention and rehabilitation Erika S Froelicher, Roberta K Oka, Gerald F Fletcher

170

17 Psychosocial factors in the primary and secondary prevention of coronary heart disease: an updated systematic review of prospective cohort studies Harry Hemingway, Hannah Kuper, Michael Marmot

181

18 Emerging approaches in cardiovascular prevention Eva M Lonn, Marek Smieja, Salim Yusuf

219

19 Obesity Arya M Sharma

231

20 Postmenopausal hormone therapy and cardiovascular disease Jacques E Rossouw

244

21 Ethnicity and cardiovascular disease Sonia S Anand, Stephanie Ounpuu, Salim Yusuf

259

22 The fetal origins of coronary heart disease David JP Barker

279

23 Molecular genetics of cardiovascular disorders AJ Marian, Robert Roberts

287

24 Cost effectiveness of prevention of cardiovascular disease Daniel B Mark

300

25 Diet and cardiovascular disease K Srinath Reddy

309

Part IIIa: Specific cardiovascular disorders: Stable coronary artery disease Bernard J Gersh and John A Cairns, Editors

vi

327

26 Anti-ischemic drugs Lionel H Opie

329

27 Impact of revascularization procedures in chronic coronary artery disease on clinical outcomes: a critical review of the evidence Charanjit S Rihal, Dominic Raco, Bernard J Gersh, Salim Yusuf

339

28 Adjunctive medical therapy in percutaneous coronary intervention James L Velianou, Ronald R van der Wieken, Maarten M Simoons

360

29 Restenosis: etiologies and prevention Giuseppe Sangiorgi, David R Holmes, Robert S Schwartz

371

Contents

Part IIIb: Specific cardiovascular disorders: Acute ischemic syndromes and acute myocardial infarction John A Cairns and Bernard J Gersh, Editors

395

30 Acute non-ST-segment elevation coronary syndromes: unstable angina and non-ST-segment elevation myocardial infarction Pierre Theroux, John A Cairns

397

31 Fibrinolytic therapy James S Zebrack, Jeffrey L Anderson

426

32 Mechanical reperfusion strategies in patients presenting with acute myocardial infarction Sanjaya Khanal, W Douglas Weaver

444

33 Adjunctive antithrombotic therapy for ST-elevation acute myocardial infarction John K French, Harvey D White

456

34 Pain relief, general management, and other adjunctive treatments Aldo P Maggioni, Roberto Latini, Gianni Tognoni, Peter Sleight

477

35 Complications after myocardial infarction Peter L Thompson, Barry McKeown

488

36 An integrated approach to the management of patients after the early phase of the acute coronary syndromes Desmond G Julian

507

Part IIIc: Specific cardiovascular disorders: Atrial fibrillation and supraventricular tachycardia A John Camm and John A Cairns, Editors

517

37 Atrial fibrillation: antiarrhythmic therapy Harry JGM Crijns, Isabelle C Van Gelder, Irina Savelieva, A John Camm

519

38 Atrial fibrillation: antithrombotic therapy John A Cairns

548

39 Atrial fibrillation: non-pharmacologic therapies Sanjeev Saksena, Andrew J Einstein

556

40 Supraventricular tachycardia: drugs v ablation Neil R Grubb, Peter Kowey

567

Part IIId: Specific cardiovascular disorders: Ventricular arrhythmias, bradyarrhythmias and cardiac arrest A John Camm, Editor

575

41 Prevention and treatment of life-threatening ventricular arrhythmia and sudden death Eugene Crystal, Stuart J Connolly, Paul Dorian

577

42 Impact of pacemakers: when and what kind? William D Toff, A John Camm

587

43 Syncope David G Benditt, Cengiz Ermis, Keith G Lurie, Scott Sakaguchi

619

vii


44 Cardiopulmonary resuscitation Nicola E Schiebel, Roger D White Part IIIe: Specific cardiovascular disorders: Left ventricular dysfunction Salim Yusuf, Editor

641

45 Prevention of congestive heart failure and treatment of asymptomatic left ventricular dysfunction RS McKelvie, CR Benedict, Salim Yusuf

643

46 Management of overt heart failure Bert Andersson, Karl Swedberg

659

47 Acute myocarditis and dilated cardiomyopathy Barbara A Pisani, John F Carlquist

681

48 Hypertrophic cardiomyopathy Perry M Elliott, Rajesh Thaman, William J McKenna

703

49 Other cardiomyopathies José A Marin-Neto, Marcus Vinícius Simões, Benedito Carlos Maciel

718

Part IIIf: Specific cardiovascular disorders: Pericardial disease Bernard J Gersh, Editor 50 Pericardial disease: an evidence-based approach to diagnosis and treatment Bongani M Mayosi, James A Volmink, Patrick J Commerford Part IIIg: Specific cardiovascular disorders: Valvular heart disease Bernard J Gersh, Editor

viii

634

733

735

749

51 Rheumatic heart disease: prevention and acute treatment Edmund AW Brice, Patrick J Commerford

751

52 Mitral valve disease: indications for surgery Blasé A Carabello

758

53 Indications for surgery in aortic valve disease Heidi M Connolly, Shahbudin H Rahimtoola

767

54 Balloon valvuloplasty: aortic valve Daniel J Diver, Jeffrey A Breall

782

55 Balloon valvuloplasty: mitral valve Zoltan G Turi

796

56 Valve repair and choice of valves Paul J Pearson, Hartzell V Schaff

809

57 Diagnosis and management of infective endocarditis David T Durack, Michael L Towns

817

58 Antithrombotic therapy after heart valve replacement Alexander GG Turpie, Walter Ageno

832

Contents

Part IIIh: Specific cardiovascular disorders: Other conditions Bernard J Gersh and Salim Yusuf, Editors

837

59 Treatment of patients with stroke Craig S Anderson

839

60 Heart disease and pregnancy Samuel C Siu, Jack M Colman

853

61 Venous thromboembolic disease Clive Kearon, Jeffrey S Ginsberg, Jack Hirsh

864

62 Peripheral vascular disease Jesper Swedenborg, Jan Östergren

877

Part IV: Clinical applications Ernest L Fallen, Editor

887

63 Clinical applications of external evidence Ernest L Fallen, Salim Yusuf

889

64 Stable angina: choice of PCI v CABG v drugs Douglas A Holder

892

65 Acute coronary syndromes George J Philippides

896

66 Acute myocardial infarction Bryan F Dias, Ernest L Fallen

902

67 Postmyocardial infarction: preventive measures Ernest L Fallen

906

68 Metabolic risk and secondary prevention of coronary disease Jacques Genest Jr

909

69 Peripheral vascular disease with suspect coronary artery disease Peter C Spittell

912

70 Heart failure Michael M Givertz

915

71 Atrial fibrillation Michael Klein

921

72 Ventricular dysrhythmias: pharmacologic v non-pharmacologic treatment L Brent Mitchell

925

73 Bradyarrhythmias: choice of pacemaker John A Boone

931

74 Valvular heart disease: timing of surgery Adrian P Banning, Brian B Gribbin

934

Index

939

ix

Evidence Based Cardiology CD Rom Features Evidence-based Cardiology PDF eBook ● bookmarks and hyperlinks for instant access to all chapters and topics ● searchable ● requires Adobe Acrobat Reader, free download available on the CD Rom, or from http://www.adobe.com/products/acrobat/readstep2.html Evidence-based Cardiology PDA edition – sample chapter ● a free sample chapter from the forthcoming PDA edition. Works on all Portable Digital Assistants – Palm/Pocket PC/Psion etc. ● Requires Mobipocket Reader, free download available on the CD Rom, or from http://www.mobipocket.com/en/DownloadSoft/ DownLoadReaderStep1.asp BMJ Books Catalogue ● Instant access to BMJ Books full catalogue, including an order form Also included – instant access to the Evidence-based Cardiology update website and regularly updated information on all BMJ Books Cardiology titles. Note: the Evidence-based Cardiology PDF eBook is for search and reference only and cannot be printed. A printable PDF version and the full PDA edition can be purchased from http://www.bmjbookshop.com. Instructions for use: The CD Rom included with Evidence-based Cardiology should start automatically upon insertion into any PC running Microsoft Windows. The menu screen will appear and you can then navigate by clicking on the headings and icons. If the CD does not start automatically, or if you wish to use again after previously quitting, browse the CD Rom using Windows Explorer and double click the heart-shaped icon. Troubleshooting: Some users may receive the following error message: “A required .DLL file, MSUBVM60.DLL, was not found”. If this or any other error message is received, please update your system by browsing the CD Rom with Windows Explorer and double-clicking the file “setup.exe”. Once this installation process is completed, restart your PC and try again. If you continue to experience difficulties, or if you are using an alternative operating system, we can give you access to an identical electronic version of the text as well as the sample PDA Edition chapter via the internet. Please send proof of purchase to the following address, with a letter advising your email address and the problem you have encountered: Evidence-based Cardiology eBook access BMJ Bookshop BMA House Tavistock Square London WC1H 9JR

Contributors

Walter Ageno Department of Medicine University of Insubria Varese, Italy David A Alter Institute for Clinical Evaluation Sciences Deparment of Medicine University of Toronto Toronto, Canada Sonia S Anand Department of Medicine and Population Health Research Institute McMaster University Hamilton, Canada Craig S Anderson Clinical Trials Research Unit University of Auckland Auckland, New Zealand Jeffrey L Anderson Department of Medicine University of Utah Cardiology Division, LDH Hospital Salt Lake City, USA Bert Andersson Department of Cardiology Sahlgrenska University Hospital Göteborg, Sweden Samira Asma Office on Smoking and Health Centres for Disease Control and Prevention Atlanta, USA Colin Baigent Clinical Trial Service Unit and Epidemiological Studies Unit University of Oxford Oxford, UK Adrian P Banning Department of Cardiology John Radcliffe Hospital Oxford, UK David JP Barker University of Southampton MRC Environmental and Epidemiology Unit Southampton General Hospital Southampton, UK

David G Benditt Professor of Medicine Co-Director Cardiac Arrhythmia Center University of Minnesota Medical School Minneapolis, Minnesota, USA CR Benedict Professor of Medicine Department of Internal Medicine Division of Cardiology The University of Texas Medical School Houston, Texas, USA Nicole Blair Office on Smoking and Health Centres for Disease Control and Prevention Atlanta, USA John A Boone Burrard Medical Building University of British Columbia Vancouver, Canada Jeffery A Breall University of Indiana Indianapolis, USA Edmund A W Brice Cardiology Department Tygerberg Academic Hospital Cape Town, South Africa Sarah E Capes Division of Endocrinology and Metabolism Department of Medicine and Population Health Research Institute McMaster University Hamilton, Canada Blasé A Carabello Department of Medicine Medical University of South Carolina Charleston, USA John F Carlquist LDS Hospital Division of Cardiology and University of Utah Department of Medicine Salt Lake City Utah, USA

xi


Jack M Colman Associated Professor Toronto Congenital Cardiac Center for Adults University Health Network and Mount Sinai Hospital University of Toronto Toronto, Canada Patrick J Commerford Helen and Morris Mauerberger Professor of Cardiology Department of Medicine University of Cape Town and Cardiac Clinic Groote Schuur Hospital Cape Town, South Africa Heidi M Connolly Consultant, Cardiovascular Diseases and Internal Medicine Associate Professor of Medicine Mayo Medical School Mayo Clinic Rochester, USA

Paul Dorian St Michael’s Hospital University of Toronto Toronto, Canada David T Durack Becton Dickinson Microbiological Systems Sparks, USA Andrew J Einstein Department of Medicine University of Medicine and Dentistry of New Jersey Robert Wood Johnson Medical School New Brunswick, USA Perry M Elliot St George’s Hospital Medical School London, UK Michael P Eriksen Centres for Disease Control and Prevention Atlanta, USA

Stuart J Connolly Division of Cardiology and Population Health Research Institute McMaster University Hamilton, Canada

Cengiz Ermis Fellow in Clinical Cardiac Electrophysiology Cardiac Arrhythmia Center University of Minnesota Medical School Minneapolis, Minnesota, USA

Deborah Cook Professor, Department of Medicine McMaster University Hamilton, Canada

Gerald F Fletcher Emory School of Medicine Center for Rehabilitation Medicine Atlanta, USA

Harry JGM Crijns Academische Ziekenhuis Groningen, Groningen, the Netherlands

Godfrey H Fowler Emeritus Professor of General Practice Institute of Health Science University of Oxford, Oxford, UK

Eugene Crystal Division of Cardiology and Population Health Research Institute McMaster University Hamilton, Canada Allan S Detsky Departments of Health Administration and Medicine University of Toronto Toronto, Canada PJ Devereaux Department of Medicine McMaster University Hamilton, Canada Bryan F Dias University of Western Ontario London, Canada Daniel J Diver Cardiac Catheterization Laboratory Georgetown University Medical Center, Washington, USA

xii

John K French Cardiology Department Green Lane Hospital Auckland, New Zealand Erika S Froelicher University of California San Francisco School of Nursing Department of Psychological Nursing San Francisco, USA Curt D Furberg Department of Public Health Services Bowman Gray School of Medicine Winston-Salem, USA Jacques Genest Jr Professor, Faculty of Medicine Novartis Chair in Medicine and Director, Division of Cardiology McGill University Montreal, Canada

Contributors

Hertzel C Gerstein Division of Endocrinology and Metabolism and Population Health Research Institute Department of Medicine McMaster University Hamilton, Canada Raymond J Gibbons Nuclear Cardiology Laboratory Mayo Medical School Mayo Clinic Rochester, USA Jeffrey S Ginsberg Department of Medicine McMaster University Hamilton, Canada Michael M Givertz Cardiovascular Division Brigham and Women’s Hospital Harvard Medical School Boston, USA Henry A Glick Assistant Professor University of Pennsylvania School of Medicine Philadelphia, Pennsylvania, USA Brian B Gribbin John Radcliffe Hospital Oxford, UK Neil R Grubb Department of Cardiology Royal Infirmary of Edinburgh Edinburgh, UK R Brian Haynes Health Information Research Unit Department of Clinical Epidemiology and Biostatistics McMaster University Faculty of Health Sciences Hamilton, Canada Harry Hemingway Department of Research and Development International Centre for Health and Society University College London Medical School and Kensington, Chelsea and Westminster Health Authority London, UK

Mark Hlatky Department of Health Research and Policy Department of Medicine Stanford University School of Medicine Stanford CA, USA Douglas A Holder Division of Cardiology Hamilton Health Sciences McMaster University Hamilton, Canada David R Holmes Division of Cardiovascular Diseases Department of Internal Medicine Mayo Clinic and Mayo Foundation Rochester, USA Dereck L Hunt Department of Clinical Epidemiology and Biostatistics McMaster University Hamilton, Canada Desmond G Julian Department of Cardiology University of Newcastle-upon-Tyne Newcastle-upon-Tyne, UK Clive Kearon Department of Medicine McMaster University Hamilton, Canada Sanjaya Khanal Cardiac Catheterization Laboratory Henry Ford Heart and Vascular Institute Detroit, USA Allan Kitching Assistant Clinical Professor Department of Medicine McMaster University Hamilton, Canada Michael Klein Cardiology Department University Hospital Boston, USA

Brenda Hemmelgarn Nephrology, Department of Medicine University of Calgary Calgary, Alberta, Canada

Peter Kowey Professor of Medicine Jefferson Medical College Philadelphia and Chief of Electrophysiology Mainline Arrhythmia Philadelphia, USA

Jack Hirsh Department of Medicine Hamilton Health Sciences, Research Center McMaster University Hamilton, Canada

Hannah Kuper International Centre for Health and Society Department of Epidemiology and Public Health University College London Medical School London, UK

xiii


Roberto Latini Department of Cardiovascular Research Mario Negri Institute Milano, Italy Malcolm Law St Bartholomew’s Medical College London, UK Eva M Lonn Division of Cardiology and Population Health Research Institute McMaster University Hamilton, Canada Keith G Lurie Professor of Medicine Co-Director Cardiac Arrhythmia Center University of Minnesota Medical School Minneapolis, Minnesota, USA Benedito Carlos Maciel Associate Professor of Medicine Cardiology Division Internal Medicine Department Medical School of Ribeirão Preto University of São Paulo Brazil Aldo P Maggioni ANMCO Research Centre Florence, Italy AJ Marian Department of Medicine Baylor College of Medicine Houston, USA José A Marin-Neto Full Professor and Head Cardiology Division Internal Medicine Department Medical School of Ribeirão Preto, University of São Paulo Brazil

RS McKelvie Division of Cardiology and Population Health Research Institute McMaster University Hamilton, Canada William J McKenna St George’s Hospital Medical School London, UK Barry McKeown Advanced Trainee in Cardiology The Heart Research Institute of Western Australia Sir Charles Gairdner Hospital Perth, Western Australia K Ann McKibbon Health Information Research Unit McMaster University Hamilton, Canada L Brent Mitchell Foothills Hospital Division of Cardiology University of Calgary Calgary, Canada C David Naylor Sunnybrook HSC University of Toronto Toronto, Canada Jim Nishikawa Associate Professor Department of Medicine University of Ottawa Ottawa, Ontario, Canada Roberta K Oka School of Nursing University of California San Francisco, USA

Daniel B Mark Professor of Medicine and Director, Outcomes Group Duke University Medical Center Duke Clinical Research Institute Durham, USA

Lionel H Opie Heart and Research Unit and Hypertension Clinic Department of Medicine Medical School Observatory Cape Town, South Africa

Michael Marmot International Centre for Health and Society Department of Epidemiology and Public Health University College London Medical School London, UK

Stephanie Ounpuu Population Health Research Institute McMaster University Hamilton Civic Hospitals Research Centre Hamilton, Canada

Bongani M Mayosi Cardiac Clinic University of Cape Town Cape Town, South Africa

Jan Östergren Department of Medicine, Karolinska Hospital Stockholm, Sweden

xiv

Contributors

Akbar Panju Professor, Department of Medicine McMaster University Chief, Department of Medicine Hamilton Health Sciences Hamilton, Canada

Jacques E Rossouw Women’s Health Initiative National Heart, Lung, and Blood Institute Bethesda, USA

Paul J Pearson Michigan Heart and Vascular Institute Minnesota, USA

Scott Sakaguchi Associate Professor of Medicine Cardiac Arrhythmia Center University of Minnesota Medical School Minneapolis, Minnesota, USA

Terry F Pechacek Office on Smoking and Health National Center for Chronic Disease Prevention and Health Promotion Atlanta, USA

Sanjeev Saksena Director, Cardiovascular Institute, AHS (East) Clinical Professor of Medicine RWJ Medical School New Brunswick, USA

George J Philippides Coronary Care Unit Boston Medical Center Boston, USA

Giuseppe Sangiorgi Department of Cardiovascular Diseases Cardiac Catheterization Laboratory Istituto Policlinico San Donato Milan, Italy

Barbara A Pisani Loyola University Medical Center Maywood, USA Jeffrey L Probstfield University of Washington School of Medicine and University of Washington School of Public Health USA Bruce M Psaty Cardiovascular Research Unit Metropolitan Park Seattle, USA Dominic Raco Division of Cardiology McMaster University Hamilton, Canada Shahbudin H Rahimtoola University of Southern California and Keck School of Medicine at USC Los Angeles, California, USA K Srinath Reddy Initiative for Cardiovascular Health Research in the Developing Countries New Delhi India Charanjit S Rihal Division of Cardiovascular Diseases and Internal Medicine Mayo Clinic Rochester, USA Robert Roberts Department of Medicine Baylor College of Medicine Houston, USA

Irina Savelieva St George’s Hospital Medical School Department of Cardiology Cranmer Terrace, Tooting London, UK Hartzell V Schaff Division of Thoracic and Cardiovascular Surgery Mayo Clinic and Mayo Foundation Rochester, Minnesota, USA Nicola E Schiebel Department of Emergency Medicine Mayo Clinic and Mayo Foundation Rochester, USA Kevin A Schulman Center for Clinical and Genetic Economics Duke Clinical Research Institute Duke University Medical Center Durham, USA Robert S Schwartz Department of Cardiovascular Diseases Minneapolis Heart Institute and Foundation Minneapolis, USA Arya M Sharma Department of Medicine and Population Health Research Institute McMaster University Hamilton, Canada Marcus Vinícius Simões Associate Professor of Medicine Cardiology Division Internal Medicine Department Medical School of Ribeirão Preto University of São Paulo Brazil

xv


Maarten M Simoons Thoraxcenter Erasmus University Rotterdam, the Netherlands

Gianni Tognoni Department of Cardiovascular Research Mario Negri Institute Milano, Italy

Samuel C Siu Toronto Congenital Cardiac Center for Adults University Health Network and Mount Sinai Hospital Toronto, Canada

Michael L Towns Becton Dickinson Microbiology Systems Sparks, USA

Peter Sleight University Department of Cardiovascular Medicine John Radcliffe Hospital Oxford, UK Maek Smieja Department of Pathology and Molecular Medicine McMaster University Hamilton, Canada Peter C Spittell Mayo Clinic Rochester Minnesota, USA Karl Swedberg Department of Medicine Sahlgrenska University Hospital/Östra University of Göteborg Göteberg, Sweden Jesper Swedenborg Department of Surgery Division of Vascular Surgery Karolinska Hospital Stockholm, Sweden Rajesh Thaman St George’s Hospital Medical School London, UK Pierre Theroux Department of Medicine Montreal Heart Institute Montreal, Canada Peter L Thompson Clinical Professor of Medicine and Public Health University of Western Australia and Cardiologist, Sir Charles Gairdner Hospital Perth, Western Australia William D Toff Division of Cardiology University of Leicester Leicester, UK

xvi

Zoltan G Turi MCP Hahnemann University Medical School Philadelphia, USA Alexander GG Turpie Department of Medicine McMaster University Hamilton, Canada Isabelle C Van Gelder Thorax Center Department of Cardiology University Hospital Groningen, the Netherlands James L Velianou Division of Cardiology McMaster University Hamilton, Canada James A Volmink Global Health Council Washington, USA W Douglas Weaver Division of Cardiovascular Medicine Heart and Vascular Institute Detroit, USA Harvey D White Cardiology Department Green Lane Hospital Auckland, New Zealand Roger D White Department of Anesthesiology Mayo Clinic Rochester, USA Ronald R van der Wieken Ouze Lieve Vrouwe Gasthuss Amsterdam, the Netherlands James S Zebrack Cardiology Division Salt Lake Regional Hospital Salt Lake City, USA

Preface to the Second edition

“Where is the knowledge in all that information? Where is the wisdom in all that knowledge?” W H AUDEN The recent proliferation of carefully controlled large scale clinical trials, their meta-analyses and selective observational studies has contributed to the remarkable strides made in the management of cardiovascular disease. One of the prophesies stated in the first edition of this textbook has come to pass – namely, that management guided by external evidence is an evolving process as newer and more effective treatment modalities come to light. While successful as a critical approach for managing patients, evidence-based medicine is nevertheless a work in progress which, if allowed to rest on its laurels, will “by nature be threatened with impending obsolescence”. In addition to keeping abreast of new information, there is a need to integrate and distill the information into coherent recommendations. Authors were therefore instructed to provide their recommendations including those based on qualitative judgments. The recognition of new developments in a rapidly changing dynamic field combined with the overwhelmingly positive worldwide response to the first edition have prompted the publication of this second edition. This edition is again dedicated to providing a comprehensive compendium of best evidence for the diagnosis and management of a wide variety of cardiovascular disorders. To avoid critical information gaps as meaningful new data emerge, the text contains several new features. Because our concepts of what constitutes

evidence-based medicine is subject to change we have included a completely revised introductory chapter. Appended to the printed text is a CD Rom that permits ready access to new information and periodic updates by way of a dedicated and active website. In addition, there will be available a compact hand-held (PDA) version of the text. There are new chapters on clinical trials and metaanalysis; fetal origins of cardiovascular disease; genetics; diet and cardiovascular disease; obesity; and cardiopulmonary resuscitation. Several chapters have been completely rewritten and most have undergone substantial revision. Finally, the layout of the text has been reformatted for better handling, portability, readability and affordability. In preparing this edition the editors and contributors have subscribed to the principle that the best external evidence found in these pages are not to be considered as hierarchical choices but rather should be used judiciously with other forms of evidence be they pathophysiologic, observational or experiential. No effort has been spared in the preparation of this edition and to this end invaluable assistance has been accorded us by Judy Lindeman at McMaster University and Mary Banks and Christina Karaviotis at BMJ Books. Salim Yusuf John A Cairns A John Camm Ernest L Fallen Bernard J Gersh

xvii

Preface to the First edition

“… if a man declares to you that he has found facts that he has observed and confirmed with his own experience, be cautious in accepting what he says. Rather, investigate and weigh this opinion or hypothesis according to requirements of pure logic, without paying attention to this contention that he affirms empirically.” MOSES MAIMONIDES. ca. 1195 Thus did the great physician Maimonides make a plea for an evidence-based approach to medicine by admonishing his followers to seek common ground between objectivism and empiricism. If Maimonides had lived in the year 1785, he would likely have read William Withering’s An Account of the Foxglove; a compendium of Withering’s personal observations on the clinical effect of the digitalis leaf. At first blush, Maimonides would cry foul at such flagrant empiricism, demanding to know the whole of the inception cohort. It turns out that Withering, instead of selecting specific cases which would have “… spoken strong in favour of the medicine, and perhaps been flattering to my own reputation” went on to say in his Preface “I have therefore mentioned every case in which I have prescribed the foxglove, proper or improper, successful or otherwise …” thus heralding a genuine, albeit retrospective, cohort study. It took 212 years before Withering was ultimately vindicated by the results of the first large scale randomized placebo controlled trial of digoxin (N Engl J Med 1997; 336: 526). Sixtyeight hundred patients with congestive heart failure, in sinus rhythm, were randomized to receive digoxin (avg dose 0·25 mg/ day) or placebo in addition to ACE inhibitors and diuretics. Over a three-year period there was no statistical difference in overall mortality but digoxin proved to be effective in reducing hospitalizations due to worsening heart failure. The advent of large scale prospective randomized clinical trials has strengthened the external evidence upon which management decisions can be made with some confidence. We have come to rely on so-called external best evidence as critical guideposts for establishing minimal criteria for treatment of many cardiovascular disorders. In the process, some myths based on putative mechanisms have been dispelled while insights into the efficacy of new treatments have been more rapidly facilitated. On the other hand there is a danger of righteous complacency which, if unchecked, could lead to a slavish dependency on statistical bottom lines and, ultimately, to “cook book” medicine. It is the intent of this textbook to present a proper balance between “objectivism and empiricism”. In this regard, the very first chapter begins by defining the practice of evidence-based cardiology as “… integrating individual clinical expertise with the best available external clinical evidence from systematic research”. The textbook has four principal components. An introductory general section addresses important topics in clinical epidemiology, as applied both to the bedside and to a population. This section includes: critical appraisal of data; clinical trials methodology;

quality of life measurements; health economics; and methods of decision analysis, all in the context of current clinical practice. Next follows a section on preventive strategies based on evidence that should enable the practicing physician to advise, with confidence, on risk factor modification and quality of life issues for selected patients. There follows a section on a broad range of specific cardiovascular disorders that highlight management issues based on current best evidence. Finally, the section on clinical applications is an attempt to put a clinical face on evidence derived from population statistics through the use of “live” clinical cases. Here, an attempt is made judiciously to couple external evidence with clinical expertise and a sound knowledge of cardiovascular pathophysiology. There is understandably a wide range of the kinds of evidence available to support different practices and treatments. The editors have chosen not to constrain the authors into rigid and uniform formats for each chapter. While several of the chapters have the level of evidence/recommendations graded, or key messages highlighted, a uniform format would not have been appropriate for every chapter. This textbook is designed for a wide audience. Since cardiovascular disease comprises more than fifty percent of adult medicine, there is something here for everyone in clinical practice and at all levels of medical undergraduate and postgraduate training. Its emphasis on practical applications of research methodology and critical appraisal of data covering a cross-section of clinical topics should invite interest among those engaged in population studies, biostatistics, clinical epidemiology and health economics as well as those involved in healthcare decision analysis, quality assurance committees and stakeholders responsible for healthcare planning. Because this textbook relies so heavily on current best evidence, it is by nature threatened with impending obsolescence. To ensure that this does not happen, the editors, in concert with the publisher, have agreed to issue up-dates periodically in the form of special supplements or updated editions, so that the text can be continually revised in accordance with emerging relevant data. In this context, it is well to bear in mind that good science always proceeds hesitantly through a series of tenuous conclusions. And so any recommendation made on the basis of available best evidence is subject to revision as we probe deeper into the mysterious nature of disease processes. One may ask of the large scale clinical trial “Why did it require more than 10,000 patients to show incontrovertible evidence that the experimental drug is effective?” Aye, there is the scientific question! The editors wish to acknowledge the herculean efforts of Catherine Wright and Karin Dearness who kept everyone on track and offer a special appreciation to Mary Banks for her editorial expertise, patience and support. Salim Yusuf John A Cairns A John Camm Ernest L Fallen Bernard J Gersh

xix

Glossary

Abbreviations commonly used in this book ABI ACC ACE AED AF AHA AMI APSAC

ankle brachial pressure index American College of Cardiology angiotensin-converting enzyme automated external defibrillator atrial fibrillation American Heart Association acute myocardial infarction anisoylated plasminogen streptokinase activator complex APTT activated partial thromboplastin time ARR associated risk reduction AS aortic stenosis ASD atrial septal defect ASMR age standardized mortality rate BBB bundle branch block BMI body mass index CABG coronary artery bypass grafting CAD coronary artery disease CBVD cerebrovascular disease CCB calcium-channel blockers CCU coronary care unit CEE conjugated equine estrogen CHD coronary heart disease CHF congestive heart failure CI confidence interval CK-MB creatinine kinase MB isoenzyme CPP coronary perfusion pressure CPR cardiopulmonary resuscitation CT computerized tomography CYA cyclophosphamide DA dopamine DALY disability adjusted life years DHP dihydropyridines DM diabetes mellitus DVT deep vein thrombosis ECG electrocardiogram EEG electroencephalogram EGF epidermal growth factor EMF endomyocardial fibrosis EOA effective orifice area EPS electrophysiologic studies FGF fibroblast growth factor FS fractional shortening GPI glycoprotein inhibitor HCM hypertrophic cardiomyopathy HDL high density lipoprotein (HDL2) HMG-CoA 3-hydroxy-3-methylglutaryl-coenzyme A HOCM hypertrophic obstructive cardiomyopathy HRT hormone replacement therapy

IC ICD ICH IDC IDL IE IFN- IGF IGT IL IM INR IQR IV LAE LBBB LDL LDL-C LMWH Lp(a) LQTS LV LVE LVEF LVH MCP MHC MHS MI MPA MRI MUFA NA NHLBI NINDS NNT NSAIDs NSTEMI NYHA OR P PAI PCI PCR PDGF PE PET PPCM PSVT PTA PTCA

intracoronary implantable cardioverter defibrillator intracerebral hemorrhage idiopathic dilated cardiomyopathy intermediate density lipoprotein infective endocarditis interferon gamma insulin-like growth factor impaired glucose tolerance interleukin intramuscular international normalization ratio interquartile range intravenous left atrial enlargement left bundle branch block low density lipoprotein low density lipoprotein cholesterol low molecular weight heparin lipoprotein long QT syndrome left ventricular left ventricular enlargement left ventricular ejection fraction left ventricular hypertrophy monocyte chemoattractant protein major histocompatibility complex Milan Hypertensive Strain myocardial infarction medroxyprogesterone acetate magnetic resonance imaging monounsaturated fatty acid not available National Heart Lung Blood Institute National Institute of Neurologic Disease and Stroke number needed to treat non-steroidal anti-inflammatory drugs non-ST-segment elevation myocardial infarction New York Heart Association odds ratio probability plasminogen activator inhibitor percutaneous coronary intervention polymerase chain reaction platelet derived growth factor pulmonary embolism positron emission tomography peripartum cardiomyopathy paroxysmal supraventricular tachycardia percutaneous transluminal angioplasty percutaneous transluminal coronary angioplasty

xxi


PUFA PVC RCT RFLP ROSC RRR rtPA RV RVEF RVF RVH SAECG SC SK SMC SFA SFA STEMI TEA

xxii

polyunsaturated fatty acid premature ventricular complex randomized controlled trial restriction fragment length polymorphisms return of spontaneous circulation relative risk reduction recombinant tissue plasminogen activator right ventricular right ventricular ejection fraction right ventricular enlargement right ventricular hypertrophy signal-averaged ECG subcutaneous streptokinase smooth muscle cells saturated fatty acid superficial femoral artery ST-segment elevation myocardial infarction thromboendarterectomy

TEE t-FA TGF TIA TIMI TMP TNF TNK tPA TTE UK v VF VPD VSD VT VTE VUI

transesophageal echocardiography trans fatty acid transforming growth factor transient ischemic attack Thrombolysis in Myocardial Infarction TIMI myocardial perfusion tumor necrosis factor tenecteplase tissue plasminogen activator transthoracic echocardiography urokinase versus ventricular fibrillation ventricular premature depolarization ventricular septal defect ventricular tachycardia venous thromboembolism venous ultrasound imaging

Grading of recommendations and levels of evidence used in Evidence-based Cardiology

GRADE A

GRADE C

Level 1a Evidence from large randomized clinical trials (RCTs) or systematic reviews (including meta-analyses) of multiple randomized trials which collectively has at least as much data as one single well-defined trial. Level 1b Evidence from at least one “All or None” high quality cohort study; in which ALL patients died/failed with conventional therapy and some survived/succeeded with the new therapy (for example, chemotherapy for tuberculosis, meningitis, or defibrillation for ventricular fibrillation); or in which many died/failed with conventional therapy and NONE died/failed with the new therapy (for example, penicillin for pneumococcal infections). Level 1c Evidence from at least one moderate-sized RCT or a meta-analysis of small trials which collectively only has a moderate number of patients. Level 1d Evidence from at least one RCT.

Level 5

GRADE B Level 2

Level 3 Level 4

Evidence from at least one high quality study of nonrandomized cohorts who did and did not receive the new therapy. Evidence from at least one high quality case–control study. Evidence from at least one high quality case series.

Opinions from experts without reference or access to any of the foregoing (for example, argument from physiology, bench research or first principles).

A comprehensive approach would incorporate many different types of evidence (for example, RCTs, non-RCTs, epidemiologic studies, and experimental data), and examine the architecture of the information for consistency, coherence and clarity. Occasionally the evidence does not completely fit into neat compartments. For example, there may not be an RCT that demonstrates a reduction in mortality in individuals with stable angina with the use of blockers, but there is overwhelming evidence that mortality is reduced following MI. In such cases, some may recommend use of blockers in angina patients with the expectation that some extrapolation from post-MI trials is warranted. This could be expressed as Grade A/C. In other instances (for example, smoking cessation or a pacemaker for complete heart block), the non-randomized data are so overwhelmingly clear and biologically plausible that it would be reasonable to consider these interventions as Grade A. Recommendation grades appear either within the text, for example, Grade A and Grade A1a or within a table in the chapter. The grading system clearly is only applicable to preventive or therapeutic interventions. It is not applicable to many other types of data such as descriptive, genetic or pathophysiologic.

xxiii

Part I General concepts and critical appraisal Salim Yusuf, Editor

1


GRADE A

GRADE C


Level 5

GRADE B Level 2

Level 3 Level 4

2




1

What is evidence-based cardiology? PJ Devereaux, R Brian Haynes, Salim Yusuf

Introduction In 1836 the editor of the American Journal of Medical Sciences, Elisha Bartlett, heralded a study as “one of the most important medical works of the present century, marking the start of a new era in science”.1 What evoked such praise and suggested a paradigm shift was Dr Pierre Louis’ systematic collection and numerical presentation of data on bloodletting. Louis adopted a baconian approach of collecting vast amounts of data on a large number of patients (by the standards of the early 1800s), which allowed him to systematically evaluate the efficacy of bloodletting. Louis argued that large numbers of patients and enumeration were necessary to equalize differences between treatment groups, as “by so doing, the errors (which are inevitable), being the same in two groups of patients subjected to different treatment, mutually compensate each other, and they may be disregarded without sensibly affecting the exactness of the results”.2 Louis subsequently went on to state: “a therapeutic agent cannot be employed with any discrimination or probability of success in a given case, unless its general efficacy, in analogous cases, has been previously ascertained”, and thus, “without the aid of statistics nothing like real medicine is possible”.3 The prevailing concept of illness at the time was that the sick were contaminated, whether by some toxin or contagion, or by an excess of one humour or another. This understanding of illness contained within it the idea that these states were improved by opening a vein and letting the sickness run out. Louis’ finding that bloodletting hastened the death of the ill was a bombshell. George Washington had 2·4 liters of blood drained from him in the 15 hours prior to his death: he had been suffering from a fever, sore throat and respiratory difficulties for 24 hours.4 Some have stated that in this way Washington was murdered.5–7 Although this is a relatively recent example, the plea for comparative evaluation was mentioned as early as the Old Testament. Throughout history there have been repeated exhortations to quantify medical or health problems and to compare outcomes in patient groups managed differently, with the goal of setting state policy or assisting individual physicians. In this chapter we will discuss what evidence-based medicine is, and then discuss an approach to evidence-based

decision making. We will use a clinical case to highlight the components of this approach, which include clinical state and circumstances, patients’ preferences and actions, research evidence, and clinical expertise. At the end of the chapter we will review the application of these components of evidence-based decision making as they apply to our patient, and provide a decision aid that can be used in such a case.

What is evidence-based medicine? Although the foundations for evidence-based medicine were laid over several centuries, an explicit philosophy, with its attendant concepts, definitions and models, has been largely developed as a formal doctrine over the last few decades. Evidence-based medicine is about solving clinical problems. Initially, the focus of evidence-based medicine was largely to find the best objective quantifiable research evidence relevant to the particular problem, and to apply that evidence in resolving the particular issue.8 This early focus de-emphasized “intuition, unsystematic clinical experience, and pathophysiologic rationale as sufficient grounds for clinical decision making” and stressed “the examination of evidence from clinical research”.9 Subsequent versions have emphasized that research evidence alone is never sufficient to make a clinical decision.10 Research evidence by itself rarely tells us what to do in individual situations, but rather it provides useful information that allows us to make more informed decisions. Clinicians must always view evidence in the context of the individual patient, and then weigh the potential benefits versus the risks, costs and inconveniences. Ideally the patient’s values and preferences take precedence10 (Figure 1.1). Figure 1.1 is based on the first edition of Evidence-based medicine11 and was published in an editorial that appeared in ACP Journal Club and Evidence-Based Medicine in 1996, along with the definition: “Evidence-based medicine is the conscientious and judicious use of current best evidence from clinical care research in the management of individual patients”.12 The editorial also included the caveat that the definition of evidence-based medicine would evolve as new types of information emerged, and would therefore be continuously refined. The concepts of evidence-based 3


Clinical expertise

Research evidence

Patient preferences

Figure 1.1 Early model of the key elements for evidencebased clinical decisions

medicine have evolved considerably and the initial model has recently been enhanced,8 especially for what is meant by clinical expertise and the additional consideration of clinical situation and circumstances. In the next section we use this new model of “evidence-based clinical decisions” to help resolve a common clinical scenario.

Approach to evidence-based clinical decision making

Clinical scenario

4

New model for evidence-based clinical decisions (Figure 1.2) Figure 1.2 depicts the evolution of the model for evidencebased clinical decisions,8 which has more recently been redefined as “the integration of best research evidence with clinical expertise and patient values”.13 This model represents a desirable approach as to how all clinical decisions should be made. However, we acknowledge that, at present, many clinical decisions are not made in this way. For instance, at present, clinicians’ individual preferences (as distinct from clinical expertise) often play a large role in their actions, leading to large “practice variations” in managing similar cases. When faced with critically ill patients with identical circumstances, different clinicians may, according to their preferences, institute aggressive lifeprolonging interventions or withdraw life support.14 Our model acknowledges that patients’ preferences should be considered first and foremost, rather than clinicians’ preferences, whenever it is possible to do so. In Figure 1.2, the “clinical state and circumstances” of the patient replace “clinical expertise” as one of the key elements in clinical decisions, “patient preferences” is expanded to include patients’ actions, and this element is reversed in position with “research evidence”, signifying its frequent precedence. Integrating all three aspects requires judgment and clinical expertise, thus constituting a fourth overarching element. We will describe each of the components, and the role of clinical expertise in integrating them. Clinical state and circumstances A patient’s clinical state and circumstances often play a dominant role in clinical decisions. Clinical trials provide us with results reflective of the average patient within the treatment groups of the trial, but rarely is a patient in

A family physician refers a patient requesting your input on the issue of antithrombotic therapy. The patient is an 80 year old man with a history of hypertension who 10 months ago, on routine examination, was diagnosed with atrial fibrillation. The patient suffered a major gastrointestinal bleed, requiring hospitalization, urgent endoscopy, and a transfusion the day after his atrial fibrillation was discovered (the patient had not started any antithrombotic therapy prior to his bleeding episode). He had, however, been receiving a non-steroidal anti-inflammatory drug (NSAID) for osteoarthritis. The patient has been free of any gastrointestinal symptoms since his bleed and has successfully avoided using an NSAID by using acetaminophen. Eight months earlier the patient’s echocardiogram demonstrated normal valvular and left ventricular function and a left atrial measurement of 6·5 cm. Based on the duration of atrial fibrillation and the size of his left atrium, you decide that cardioversion is not an option. The patient is very worried about having a stroke, as his wife was left dependent on him for 2 years prior to her death following a major stroke. The referring physician, who recently had a patient who suffered a serious gastrointestinal bleed while on warfarin, is very concerned about the risk of bleeding, given this patient’s age and recent history of gastrointestinal bleeding.

What is evidence-based cardiology?

Clinical expertise

Research evidence

Patient preferences

Clinical state and circumstances

Clinical expertise

Patients’ preferences and actions

Research evidence

Figure 1.2 Evolving model for evidence-based clinical decisions

clinical practice the same as the average patient from a clinical trial. Individual patients have unique characteristics that typically put them at lower or higher risk of the outcome or treatment side effect than the average patient in the trial. As

such, optimal clinical decisions should be individualized to the patient’s clinical state. A patient who is at very high risk of a future vascular event, but at low risk of any complication from a drug (for example, a patient with a low density lipoprotein value of 8·0 mmol/l post myocardial infarction and no contraindication to statin therapy), or conversely a patient who is at low risk of the outcome and high risk of a treatment’s complications (for example, a 40 year old man with atrial fibrillation without any associated stroke risk factors who has experienced a recent major gastrointestinal bleed), may find their clinical state dominating the clinical decision making process. It is notable that the circles of clinical state and circumstances and research evidence overlap. Frequently research evidence can inform us about the influence of the clinical state and circumstances. Considering our patient, the pooled data from five randomized controlled trials (RCTs) evaluating the efficacy of warfarin in patients with non-valvular atrial fibrillation (NVAF) demonstrated an average annual stroke rate of 4·5% and a major bleeding rate of 1% in patients not receiving antithrombotic therapy.15 The investigators who combined the five RCTs used the control patient data to develop a clinical prediction tool to estimate the annual risk of stroke. Independent risk factors that predicted stroke in control patients were increasing age, a history of hypertension, diabetes, and prior stroke or transient ischemic attack (TIA).15 Our patient’s annual risk of stroke is predicted to be about 8%, which is higher than that of the average control patient in the five RCTs, whose annual stroke rate was 4·5%.15 Similarly, a clinical prediction tool has been developed for predicting the risk of major bleeding (defined as the loss of two units of blood within 7 days, or life-threatening bleeding) while taking warfarin therapy.16 Independent risk factors that predict major bleeding in patients taking warfarin include age 65, history of stroke, history of gastrointestinal bleeding, recent myocardial infarction, anemia, renal failure and diabetes. (Note that many of the factors that predict a higher risk of stroke also increase the risk of bleeding.) Our patient’s annual risk of major bleeding of 8% also differs from that of the average patient receiving warfarin in the five RCTs, whose annual risk of major bleeding was 1·3%. We are unaware of any clinical prediction tool for predicting major bleeding while taking aspirin, and the atrial fibrillation trials had inadequate power to estimate this. However, based on the results of the meta-analysis by the antithrombotic trialists’ collaboration, we would expect aspirin to increase the risk of major bleeding from 1% to about 1·3% on average.17 The clinical circumstances in which you and your patient find yourselves (for example, your ability to administer and monitor a treatment) may be very different from those of an RCT. For example, the patient may not be able to obtain frequent tests of the intensity of anticoagulation. However, for a patient with the same clinical characteristics, we can frequently optimize clinical circumstances to decrease the risk 5


of an outcome or treatment side effect. For example, we can decrease the risk of bleeding due to warfarin therapy by more intensive monitoring. Thus, an “evidence-based” decision about anticoagulation for a patient with atrial fibrillation is not only determined by the demonstrated efficacy of anticoagulation and its potential adverse effects,18 but will vary based on the patient’s clinical state and according to individual clinical circumstances. Patients’ preferences and actions Patients may have no views or, alternatively, unshakable views, on their treatment options, depending on their condition, personal values and experiences, degree of aversion to risk, healthcare insurance and resources, family, willingness to take medicines, accurate or misleading information at hand, and so on.8 Accordingly, individuals with very similar clinical states and circumstances may choose very different courses of action, despite being presented with the same information about the benefits and risks of an intervention. For our patient with NVAF, research evidence informs us about the differing preferences of patients and their physicians for antithrombotic therapy in atrial fibrillation when they weigh the competing risks of stroke and bleeding.19 In this study,19 participants (that is both physicians and patients) reviewed flip charts describing in detail the acute and longterm consequences of a major and minor stroke and a major bleeding event. Participants were instructed that the likelihood of a minor or major stroke was equal. The participants then underwent a probability trade-off technique which determined the minimum number of strokes that needed to be prevented before the participant felt antithrombotic therapy was justified (this value was determined for both warfarin and aspirin), given the associated increased risk of bleeding, costs and inconveniences. The same technique was also used to determine the maximum number of excess bleeds the participant would consider to be acceptable with antithrombotic therapy (determined both for warfarin and aspirin), given the benefits in terms of stroke reduction with this therapy. This study demonstrates significant variability between physicians and patients in their weighing of the potential outcomes associated with atrial fibrillation and its treatment. Patients required less stroke reduction and were more tolerant of the risk of bleeding than physicians. For example, on average, patients were willing to accept the risk of 17 extra major bleeding events in 100 patients over a 2 year period if warfarin prevented eight strokes among these 100 patients. Physicians, however, were only willing to accept 10 major bleeding events for the same level of benefit. Furthermore, physicians varied significantly in how much bleeding risk they thought was acceptable for a given stroke reduction associated with an antithrombotic agent. Hence different physicians would make very different recommendations to the same patient with identical risks of bleeding and 6

stroke. This underscores the importance of having patient values and preferences drive clinical decision making. It is the patient who is at risk of the outcome and so, when willing and able, they should be the one to weigh the potential benefits versus the risks, costs and inconveniences. There is debate regarding the optimal way to elicit and incorporate patient preferences into clinical decision making. One method is to discuss the potential benefits and risks with a patient and then qualitatively incorporate your impression of the patient’s preferences into the clinical decision. Alternatively, at least two quantitative approaches exist: decision analytic modeling and probability trade-off technique. In a decision analytic model, a standard gamble, time trade-off or visual analog scale technique is used to determine the utility (patient value/preference) of the various outcomes. This information is then fed into a decision tree that includes the probabilities of the outcomes for all clinical decisions being considered. Using the decision tree, calculations are undertaken to determine what course of action optimally fits the patient’s preferences. The probability trade-off technique presents patients with the probabilities for the various interventions being considered and then asks them to make a decision based on this information. This allows a direct and quantitative incorporation of the patient’s preferences. Proponents of decision analytic modeling question whether patients can understand probabilities to allow the appropriate incorporation of their preferences. Proponents of probability trade-off techniques wonder if a measure of utility (that is preference) in the absence of probabilities is meaningful. Only one study has directly compared decision analytic modeling with a probability trade-off technique.20 This study focused on the primary prevention of stroke and myocardial infarction with aspirin therapy in elderly patients. Both methods (that is decision analysis and probability trade-off) were performed on all patients at separate times. This study demonstrated that treatment recommendations varied significantly, depending on which method was used. After patients were presented with their individual treatment thresholds as determined by both methods, over twice as many stated they would base their preferences on the results of the probability trade-off as opposed to the decision analysis.20 Further research is needed to determine which of the models better represents patients’ self-interests. Regardless of what their preferences may be, patients’ actions may differ from both their preferences and their clinicians’ advice.21 For example, a patient may prefer to lose weight, quit smoking and take their medications as prescribed, but their actions may fall short of achieving any of these objectives. Alternatively, they may follow the treatment as prescribed, even if they resent its imposition, adverse effects and costs. Unfortunately, clinicians’ estimates of their patients’ adherence to prescribed treatments


have no better than chance accuracy.22 Thus, physicians’ decisions for care will better meet the model’s specifications if they are able to assess whether their patients will follow, or are following, their prescriptions.22 We recognize that at present patients’ preferences are rarely formally incorporated in clinical practice. This may be related to lack of physician training in these approaches, a reluctance to tread unfamiliar ground, and also in many circumstances the lack of accurate quantitative information on risk and benefits, as well as clinical risk prediction tools. However, this is likely to change rapidly as clinical models can be derived from large databases and handheld computers can be utilized to quantify risks and benefits at the bedside. Research evidence We support a very broad definition of research evidence, namely, “any empirical observation about the apparent relation between events”.23 In keeping with this definition, research evidence includes everything from the unsystematic observation of a single physician to a systematic review of large RCTs. Not all evidence is created equal, and hence there is a hierarchy of evidence that varies depending on whether one is addressing a diagnostic, prognostic or therapeutic decision. We will focus on the hierarchy of evidence for therapeutic decisions (Box 1.1).23 Box 1.1 Hierarchy of evidence for treatment decisions* Coherence of evidence from multiple sources Systematic review of several well designed, large randomized controlled trials Single large randomized controlled trial Systematic review of several well designed small randomized controlled trials Single small randomized controlled trial Systematic review of several well designed observational studies Single observational study Physiologic studies Unsystematic observation from a physician * This hierarchy cannot be rigidly adhered to. At times a single observation may be very powerful (for example, defibrillation for ventricular fibrillation), or observational studies may provide unequivocal evidence (for example, smoking cessation and lung cancer). However, in most cases where treatment effects may be moderate, outcomes variable or the clinical course unpredictable, the proposed hierarchy is useful.

All evidence has value, and the best evidence available in the hierarchy should be given appropriate consideration, even if not at the top of the hierarchy. Therefore, the unsystematic observations of colleagues should not be dismissed when no higher level evidence exists. Indeed, unsystematic observations can lead to many important insights, and experienced clinicians usually develop a respect for the insights

of their astute colleagues. However, it is equally important to recognize that unsystematic observations are commonly limited by the small number of observations, variability in outcomes, lack of objectivity, and the difficulties in integrating (for example, taking into account the natural history of a disorder, placebo effect, and a patient’s desire to please) and drawing inferences from observations.24 All evidence has limitations. Although the majority of advances in medicine are initially uncovered through individual observations, physiologic studies, observational studies or randomized controlled trials evaluating surrogate endpoints, there have also been several extremely misleading findings that have, at times, resulted in harm. It is important to remember that contradictory results across studies on the hierarchy of evidence table are not isolated to one or two instances (Table 1.1). Perhaps the most powerful example is the story of antiarrhythmic therapy. Despite encouraging evidence that encainide and flecainide could prevent premature ventricular beats, a large RCT demonstrated a higher mortality rate with these drugs than with placebo, such that these drugs resulted in an extra death for every 20 patients treated with encainide or Flecainide.39 It is estimated that more Americans were killed by these drugs than died in the Vietnam War.40 Ideally, we would have evidence from all levels of the hierarchy and the evidence would be coherent across all levels. This would represent the most persuasive evidence. However, this rarely happens, as even RCTs may by chance frequently demonstrate contradictory findings, especially when they are small. Therefore, physicians should always aim for the highest level of evidence for clinical decision making. Clinicians can still make strong inferences, particularly when there is evidence from a systematic review of several well designed large RCTs, or simply a large single pragmatic RCT. The RCT is such a powerful tool because randomization is our only means to reduce bias in treatment comparisons by controlling for unknown prognostic factors.41 Therefore, RCTs have the potential to provide the most valid (that is likelihood that the trial results are unbiased) estimates of treatment effect.42 Furthermore, large RCTs with broad eligibility criteria enhance the generalizability of their findings. An n of 1 randomized controlled trial is an RCT where individual patients are randomized to pairs of treatment periods, such that they receive the experimental treatment during one period and a placebo during the other.43 Both patients and healthcare providers are blind to which period is the experimental and which the placebo. Patients continue undergoing pairs of treatment periods until they and the healthcare providers become convinced that the experimental intervention either does or does not work.43 The advantage of an n of 1 RCT is that it provides evidence directly from the patient. However, this method is applicable only in a disease state that has limited fluctuation, and 7


Table 1.1 Some examples of contradictory results across studies at various positions in the hierarchy of evidence Results from lower level evidence

Results from higher level evidence

Milrinone demonstrated improvement in left ventricular function during exercise25

A large RCT26 and meta-analysis of several RCTs27 demonstrated a 28% relative increase in mortality with milrinone compared to placebo

An observational study of extracranial to intracranial bypass surgery suggested a “dramatic improvement in the symptomatology of virtually all patients” undergoing the procedure28

A large RCT demonstrated a 14% relative increase in the risk of fatal and non-fatal stroke in patients undergoing this procedure compared to medical management29

A meta-analysis of 16 cohort studies and 3 cross-sectional angiographic studies (including studies of women with known coronary artery disease) demonstrated a relative risk of 0·5 (95% CI 0·44–0·57) for coronary artery disease among women taking estrogen30

A moderate-sized secondary prevention RCT did not demonstrate any reduction in coronary heart disease events but did demonstrate an increase in thromboembolic events in patients receiving estrogen.31 Preliminary reports from an ongoing very large RCT (Women’s Health Initiative) indicate an increased risk of MI and strokes in the first 2 years of estrogen therapy32

A secondary analysis of an RCT suggested that lower doses of ASA were associated with a higher risk of perioperative stroke and death in patients undergoing carotid endarterectomy33

A large prospective RCT showed a higher risk of perioperative stroke, myocardial infarction or death with high-dose ASA33

A physiologic study demonstrated that blockers result in a decline in ejection fraction and increases in end-diastolic volume in patients with prior myocardial infarction34

A meta-analysis of 18 RCTs35 and 3 large trials (CIBIS2,36 MERIT-HF37 and COPERNICUS38) in patients with heart failure found a 32% relative risk reduction in death in patients receiving blockers

for treatments that can be crossed over (for example, shortacting medical treatments rather than surgery) and which are targeted at symptom relief and quality of life, as opposed to serious outcomes such as myocardial infarction and death. Even then, n of 1 RCTs are not feasible for many patients because of lack of infrastructure to support them, such as a pharmacy that is able and willing to provide matching placebos. Also, short-term symptomatic effects of treatments may differ from their long-term effects, so that n of 1 trials may provide misleading answers. Similarly, if side effects occur only after prolonged treatment (for example, during to drug accumulation, as with amiodarone), then short-term crossover studies (which is what n of 1 trials are) may not identify the full risks associated with a treatment. As such, there has been limited implementation of n of 1 RCTs in cardiology, but they represent a unique opportunity (when possible and applicable) to obtain individual patient level evidence. Considering our case of the patient with NVAF, the highest level of evidence comes from a systematic review of all the RCTs that have evaluated antithrombotic therapy in patients with atrial fibrillation.18 This study demonstrates that warfarin reduces the relative of stroke (ischemic and hemorrhagic) by 62%, and aspirin by 22%. Considering the risk of bleeding associated with warfarin therapy, there is an RCT that demonstrates a 50% decrease in the risk of bleeding if a patient is willing to undergo education, training and self-monitoring of prothrombin time.44 8

Clinical expertise Evidence-based decision making requires clinical expertise to establish and balance the patient’s clinical state and circumstances, preferences and actions, and the best research evidence. Before a therapeutic decision can be considered, clinical expertise is required to get the diagnosis and prognosis right. As shown above, clinical prediction tools can be extremely helpful in determining a patient’s prognosis, but they are unlikely to eliminate the need for sound clinical judgment acquired through clinical experience. Sizing up the clinical circumstances has never been more challenging, as commonly there exist several potential interventions, some of which require technical expertise for their effective and safe delivery. Getting the evidence right requires the skill to identify, evaluate and apply the evidence appropriately. Communicating with patients has always been considered important. This takes on greater importance as there is a growing desire on the part of patients to be involved in decisions relating to their health. Expertise is required to provide patients with the information they need, to elicit their preferences, and to incorporate those preferences into the decision. Currently there is no consensus on how this information should be presented to patients and how their preferences should be incorporated. However, we know that information should not be presented in relative terms (for example, warfarin will decrease your risk of stroke by 62%) because


patients assume their baseline risk is 100% even when they are instructed it is not.45 A recent systematic review of RCTs that compared decision aids (that is interventions designed to help people make specific choices among options by providing information on those options and outcomes relevant to the patient’s health) to traditional ways of involving/informing patients in decision making46 demonstrated that decision aids, as opposed to usual care, improved the average knowledge scores of patients for the options and outcomes by 20% (95% CI 13–25), reduced decisional conflict scores (that is patients felt more certain, informed, and clear about values in their decision), and increased patient participation in decision making.46 Where available, decision aids provide a potential means to facilitate information presentation, incorporation of preferences, and participation in the decision-making process. The varying roles of the components of evidence-based clinical decisions Depending on the circumstances, any of the circles in the new model could predominate. Varying the size of the circles to reflect their actual contribution to the clinical decision could portray this visually. Sometimes the clinical state or circumstance dominates the clinical decision. For example, a patient who is at very high risk of an outcome and low risk of a complication may have their clinical state dominate the decisionmaking process. A patient living in a remote area may not have access to anticoagulation monitoring, and this would probably dominate the decision-making process. Patient’s preferences can be so strong that they act as the driving factor in the decision-making process. For example, some patients will not take blood products regardless of the clinical situation. Research evidence can be the main factor in decision making when the benefit of an intervention is moderate to large in size and the risk of treatment small, as with blocker therapy in patients post myocardial infarction, ACE inhibitors in coronary

artery disease or heart failure, or cholesterol lowering with statins. Finally, clinical expertise can predominate, especially when it is related to technical capabilities. Application to our patient For our patient the evidence would suggest an 8% annual risk of stroke and 1% risk of major bleeding without any antithrombotic therapy. With warfarin therapy we would expect the annual risk of stroke to decrease to 3% and the risk of major bleeding to increase to 8%. This latter could be reduced to 4% if the patient were willing to undergo selfmonitoring of their prothrombin time and an education program, as discussed above.44 With aspirin therapy we would expect the annual risk of stroke to decrease to 6% and the risk of major bleeding to increase to 1·3%. As discussed above, there is no consensus on how to present this information to our patient or how to incorporate his preferences. We have provided a decision aid for patients that describes atrial fibrillation (Table 1.2), a major and minor stroke (Table 1.3), a severe bleed (Table 1.4), and a probability trade-off for no treatment, aspirin and warfarin therapies (Figure 1.3). The descriptions of major and minor stroke and a severe bleed are slight modifications of the descriptions developed and tested by Man-Son-Hing and colleagues.47 We have also individualized the probability tradeoff for our patient, with the knowledge that he would undergo self-monitoring of his prothrombin time if he decided to take warfarin therapy (Figure 1.4). Once this evidence-based clinical decision is reached our job is not over. The patient will need monitoring to ensure he is able to follow through on his clinical decision. One advantage of the decision aid provided (including his individualized probability trade-off) is that the patient can take the information home and does not have to rely on his memory to recall the facts discussed during your meeting.

Table 1.2 Atrial fibrillation: the most common disorder of the heartbeat Risk

Chances of developing atrial fibrillation increase with age and it occurs in approximately 10% of all people above the age of 75

Physical symptoms

Irregular and usually rapid beating of the heart, sensed as a fluttering in the chest. Some patients feel no symptoms and are unaware that they have atrial fibrillation

Complications

Stroke ● Atrial fibrillation increases the risk of a clot developing in the heart. This clot can be swept up towards the brain, causing a stroke ● The chance of developing a stroke with atrial fibrillation increases with either age greater than 65 years, high blood pressure, diabetes, heart failure, or a history of strokes or “mini-strokes” ● The risk of developing a stroke with atrial fibrillation varies, depending on how many of these risk factors you have ● There are medications that thin the blood, which help to prevent clots and therefore stroke ● Because the blood is thinned there is an increased risk of bleeding

Treatment

9


Table 1.3 Strokes can be minor or major in severity. If you have a stroke as a result of atrial fibrillation, your chance of having a minor or major stroke are equal Minor stroke

Major stroke

Physical symptoms

You suddenly cannot move or feel one arm and one leg

You suddenly are unable to move one arm and one leg You cannot swallow

Mental symptoms

You are unable to fully understand what is being said to you You have difficulty expressing yourself

You are unable to understand what is being said You are unable to speak

Pain

You feel no physical pain

You feel no physical pain

Recovery

You are admitted to hospital Your weakness, numbness and problem with understanding improve, but you still feel slightly weak or numb in one arm and one leg You are able to do almost all of the activities you did before the stroke You can function independently You leave the hospital after 1 week

You are admitted to hospital You cannot dress The nurses feed you You cannot walk After 1 month of physiotherapy you are able to wiggle your toes and lift your arm off the bed

Further risk

You have an increased risk of having more strokes

Another illness will probably cause your death

Table 1.4

Severe

You remain this way for the rest of your life

bleeding while taking warfarin or ASA: an example of a stomach bleed

Physical

You feel unwell for 2 days, then suddenly you vomit blood

Treatment

You are admitted to hospital You stop taking warfarin or ASA A doctor puts a tube down your throat to see where you are bleeding from You receive sedation to ease the discomfort of the test You do not need an operation You receive blood transfusions to replace the blood you lost

Recovery

You stay in hospital for 1 week You feel well at the end of your hospital stay You need to take pills for the next 6 months to prevent further bleeding After that you are back to normal

Bleeding from the stomach is the most common type of serious bleeding while taking warfarin or ASA; however, rarely other serious forms of bleeding can occur, such as bleeding within the head after a fall. Warfarin or ASA can also cause minor bleeding, including bruising and nose bleeds. Taking warfarin can mean costs and inconvenience to yourself and family. For example: need for blood tests; parking/transportation; cost of warfarin. Taking ASA can mean costs to yourself. For example: cost of ASA.

Limitations of evidence-based clinical decision model This model does not consider the important roles that society, governments or healthcare organizations can play in decision making. We deliberately restricted ourselves to decisions made by patients and their healthcare providers to allow a focused exploration of the issues involved in their immediate decision making process. However, a healthcare 10

organization may pre-empt these decisions. For example, not funding primary percutaneous transluminal coronary angioplasty in acute myocardial infarction can have an enormous impact on health outcomes, and will impose a clinical decision on all patients and physicians by eliminating this option. Physicians will have to factor in such issues when considering their patient’s clinical circumstances.


Without any blood thinning medication Chance of stroke over next 2 years is out of 100 Chance of severe bleeding over next 2 years is out of 100

ASA Chance of stroke over next 2 years is out of 100 Chance of severe bleeding over next 2 years is out of 100

Warfarin Chance of stroke over next 2 years is out of 100 Chance of severe bleeding over next 2 years out of 100 is

Figure 1.3 Without any blood thinning medication Chance of stroke over next 2 years is 8 out of 100 Chance of severe bleeding over next 2 years is 1 out of 100

ASA Chance of stroke over next 2 years is 6 out of 100 Chance of severe bleeding over next 2 years is 1·3 out of 100 (i.e.13 out of 1000)

Warfarin Chance of stroke over next 2 years is 3 out of 100 Chance of severe bleeding over next 2 years out of 100 is 4

Figure 1.4

11


Conclusions The foundations for evidence-based medicine have been established over the centuries but the specific philosophies, concepts, definitions and models have essentially evolved over the past few decades. Evidence-based medicine is about solving clinical problems. Evidence-based decision making depends upon utilizing clinical expertise to integrate information about a patient’s clinical setting and circumstances with the best research evidence while incorporating the patient’s preferences and actions.

References 1.Louis PCA. Researches on the effects of blood-letting in some inflammatory diseases, and on the influence of tartarised antimony and vesication in pneumonitis. Am J Med Sci 1836;18:102–11. 2.Louis PCA. Researches on the Effects of Bloodletting in Some Inflammatory Diseases and on the Influence of Tartarised Antimony and Vesication in Pneumonitis. Translated by CG Putnam. Boston: Hilliard, Gray, 1836. 3.Louis PCA. Medical statistics. Am J Med Sci 1837;21:525–8. 4.Morens DM. Death of a president. N Engl J Med 1999;341:1845–9. 5.Lloyd JU. Who killed George Washington? Eclectic Med J 1923;83:353–6, 403–8, 453–6. 6.Marx R. A medical profile of George Washington. Am Heritage 1955;6:43–7, 106–7. 7.Pirrucello F. How the doctors killed George Washington. Chicago Tribune Magazine 20 February 1977. 8.Haynes RB, Devereaux PJ, Guyatt GH. Clinical expertise in the era of evidence-based medicine and patient choice. ACP Journal Club 2002;136:A11–A13. 9.Evidence-based medicine working group. Evidence-based medicine, a new approach to teaching the practice of medicine. JAMA 1992;268:2420–5. 10.Haynes RB, Sackett DL, Gray JMA, Cook DC, Guyatt GH. Transferring evidence from research into practice: 1. The role of clinical care research evidence in clinical decisions. ACP Journal Club 1996;125:A-14. Evidence-Based Medicine 1996;1:196. 11.Sackett DL, Richardson SR, Rosenberg W, Haynes RB. Evidence-Based Medicine: how to practice and teach EBM. London: Churchill Livingstone, 1997. 12.Sackett DL, Rosenberg WMC, Gray JA, Haynes RB, Richardson WS. Evidence-Based Medicine: What it is and what it isn’t. BMJ 1996;312:71–2. 13.Sackett DL, Straus S, Richardson SR, Rosenberg W, Haynes RB. Evidence-Based Medicine: how to practice and teach EBM, 2nd edn. London: Churchill Livingstone, 2000. 14.Cook DJ, Guyatt GH, Jaeschke R. Determinants in Canadian health care workers of the decision to withdraw life support from the critically ill. JAMA 1995;273:703–8. 15.Atrial Fibrillation Investigators. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Arch Intern Med 1994;154:1449–57.

12

16.Beyth RJ, Quinn LM, Landefeld S. Prospective evaluation of an index for predicting the risk of major bleeding in outpatients treated with warfarin. Am J Med 1998;105:91–9. 17.Antithrombotic Trialists’ Collaboration. Collaborative metaanalysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ 2002;324:71–86. 18.Hart RG, Benavente O, McBride R, Pearce LA. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis. Ann Intern Med 1999;131: 492–501. 19.Devereaux PJ, Anderson DR, Gardner MJ et al. Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: observational study. BMJ 2001;323:1218–22. 20.Man-Son-Hing M, Laupacis A, O’Connor AM, Coyle D, Berquist R, McAlister F. Patient preference-based treatment thresholds and recommendations: a comparison of decisionanalytic modeling with the probability-tradeoff technique. Med Decis Making 2000;20:394–403. 21.Haynes RB. Improving patient adherence: State of the art, with a special focus on medication taking for cardiovascular disorders. In: Burke LE, Okene IS, eds. Patient Compliance in Healthcare and Research. American Heart Association Monograph Series. Armonk, NY: Futura Publishing Co, 2001. 22.Stephenson BJ, Rowe BH, Macharia WM, Leon G, Haynes RB. Is this patient taking their medication? JAMA 1993; 269:2779–81. 23.Guyatt G, Haynes B, Jaeschke R et al. Introduction: the philosophy of evidence-based medicine. In: Guyatt G, Rennie DR, eds. Users’ guides to the medical literature. AMA Press, 2002. 24.Nisbett R, Ross L. Human Inference. Englewood Cliffs, NJ: Prentice-Hall, 1980. 25.Timmis AD, Smyth P, Jewith DE. Milrinone in heart failure: effects on exercise haemodynamics during short term treatment. Br Heart J 1985;54:42–7. 26.Packer M, Carver JR, Rodeheffer RJ et al. Effect of oral milrinone on mortality in severe chronic heart failure. The PROMISE Study Research Group. N Engl J Med 1991; 325:1468–75. 27.Yusuf S, Teo KK. Inotropic agents increase mortality in patients with congestive heart failure. American Heart Association 63rd Scientific Sessions. Dallas (Texas), 12–15 November 1990. Circulation 1990;82(SIII):673. 28.Popp AJ, Chater N. Extracranial to intracranial vascular anastomosis for occlusive cerebrovascular disease: experience in 110 patients. Surgery 1977;82:648–54. 29.Failure of extracranial–intracranial arterial bypass to reduce the risk of ischemic stroke: results of an international randomized trial. The EC/IC Bypass Study Group. N Engl J Med 1985;313:1191–200. 30.Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med 1991;20:47–63. 31.Hulley S, Grady D, Bush T et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary artery disease in postmenopausal women. Heart and Estrogen/progestin Replacement Study (HERS) Research Group. JAMA 1998;280:605–13. 32.http://www.nhlbi.nih.gov/whi/hrt.htm


33.Taylor DW, Barnett HJ, Haynes RB et al. Low-dose and highdose acetylsalicylic acid for patients undergoing carotid endarterectomy: a randomised controlled trial. ASA and Carotid Endarterectomy (ACE) Trial Collaborators. Lancet 1999;353:2179–84. 34.Coltart J, Alderman EL, Robison SC, Harrison DC. Effect of propranolol on left ventricular function, segmental wall motion, and diastolic pressure-volume relation in man. Br Heart J 1975;37:357–64. 35.Lechat P, Packer M, Chalon S, Cucherat M, Arab T, Boissel JP. Clinical effects of beta-adrenergic blockade in chronic heart failure: a meta-analysis of double-blind, placebo-controlled, randomized trials. Circulation 1998;98:1184–91. 36.CIBIS-II Investigators and Committees. The Cardiac Insufficiency Bisoprolol Study II (CIBIS-II): a randomised trial. Lancet 1999;353:9–13. 37.The MERIT-HF Study Group. Effect of metoprolol CR/XL in chronic heart failure: Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF). Lancet 1999;353:2001–7. 38.Packer M, Coats AJ, Fowler MB et al. Effect of carvedilol on survival in severe chronic heart failure. N Engl J Med 2001;344:1651–8. 39.Echt DS, Liebson PR, Mitchell LB. Mortality and morbidity in patients receiving encainide, flecainide, or placebo: The

Cardiac Arrhythmia Suppression Trial. N Engl J Med 1991; 324:781–8. 40.Moore TJ. Excess mortality estimates. Deadly medicine: why tens of thousands of heart patients died in America’s worst drug disaster. New York: Simon & Schuster, 1995. 41.Kunz R, Oxman AD. The unpredictability paradox: review of empirical comparisons of randomised and non-randomised clinical trials. BMJ 1998;317:1185–90. 42.Chalmers I. Unbiased, relevant, and reliable assessments in health care. BMJ 1998;317:1167–8. 43.Guyatt GH, Sackett DL, Taylor DW et al. Determining optimal therapy: randomized trials in individual patients. N Engl J Med 1986;314:889–92. 44.Beyth RJ, Quinn L, Landefeld CS. A multicomponent intervention to prevent major bleeding complications in older patients receiving warfarin. Ann Intern Med 2000; 133:687–95. 45.Malenka DJ, Baron JA, Johansen S, Wahrenberger JW, Ross JM. The framing effect of relative and absolute risk. J Gen Intern Med 1993;8:543–8. 46.O’Connor AM, Rostom A, Fiset V et al. Decision aids for patients facing health treatment or screening decisions: a systematic review. BMJ 1999;319:731–4. 47.Man-Son-Hing M, Laupacis A, O’Connor A et al. Warfarin for atrial fibrillation: The patient’s perspective. Arch Intern Med 1996;156:1841–8.

13

2

A critical appraisal of the cardiovascular history and physical examination Akbar Panju, Brenda Hemmelgarn, Jim Nishikawa, Deborah Cook, Allan Kitching

There have been numerous technological advances made in the diagnosis and treatment of cardiovascular disease. In spite of this, a carefully conducted clinical examination remains the cornerstone in the initial assessment of the patient with known or suspected cardiovascular disease. Before conducting further laboratory or radiologic diagnostic tests, clinicians implicitly consider each piece of historical information and each finding from the physical examination as a diagnostic test that increases or decreases the probability of the possible diagnoses. The competency and accuracy of the clinical examination is therefore crucial, for it serves as the basis for our judgment regarding not only diagnosis, but prognosis and therapy as well. This chapter is not intended to provide details of how to perform a cardiovascular history and physical examination, and should be read in conjunction with standard textbooks on cardiology to obtain such information. Instead, we will provide the reader with the tools to identify those features of the history and physical examination reported in the literature which are the most reliable and valid in assessing a patient with cardiovascular disease. We will focus on strategies to locate literature on the clinical examination, as well as guidelines to assess the quality of those studies. These techniques will then be applied to three common features of the cardiovascular history, namely chest pain, dyspnea and syncope, as well as common features of the physical examination, including assessment of the apical impulse, the third heart sound, central venous pressure, systolic murmurs, blood pressure and arterial pulse. We would also encourage the reader to access The Rational Clinical Examination series published in the Journal of the American Medical Association for further reviews on various aspects of the cardiovascular physical examination.1–9 The following topics will be covered in this chapter: ● ● ● ●

14

Strategies used to locate literature on clinical examination How to critically appraise this literature Application of the above in the cardiovascular history (chest pain, dyspnea, syncope) Application of the above in the cardiovascular physical examination (apical impulse, central venous pressure, systolic murmurs, blood pressure, arterial pulse).

Strategies used to locate literature on clinical examination There are no validated strategies for locating precise and accurate information on obtaining a cardiovascular history and conducting a physical examination. A proposed strategy for searching the MEDLINE database is summarized in Box 2.1. This is the method suggested for authors of the Rational Clinical Examination series appearing in the Journal of the American Medical Association.10 The first terms capture the clinical topic of interest by specifying the disease or presentation or function/dysfunction being sought. The second group of terms seeks clinical skills articles. The third group of terms is intended to find articles of high methodologic quality. An efficient strategy to locate high-quality articles would be to combine the first two groups of terms with “diagnosis (pre-exploded)” to maximize sensitivity, or with “sensitivity (textword)” to maximize specificity. This is an extension of the method suggested by the ACP Journal Club for finding high-quality articles on diagnostic tests in general.11 Box 2.1 Search strategy for clinical skills articles using MEDLINE Group 1 terms Term(s) for clinical entity of interest (for example, syncope, myocardial infarction) combined with (AND)* Group 2 terms ● Physical examination (exploded; in title, abstract or subject heading) ● Medical history-taking (exploded) ● Professional competence (exploded) ● Diagnostic tests, routine ● Combined with (OR)* Group 3 terms ● Sensitivity and specificity (textword; exploded) ● Reproducibility of results ● Observer variation ● Decision support techniques ● Bayes’ theorem *AND and OR represent Boolean terms (symbolic representation of relationships between sets) for combining items.

Any comprehensive search for relevant articles should include a review of reference lists from the articles found and

Cardiovascular history and physical examination

review articles on the topic, as well as textbooks on clinical examination, and advice from clinicians interested in clinical examination. How to critically appraise the literature on clinical examination studies Having located articles on the cardiovascular clinical examination, one must carefully review each study to establish its validity, or accuracy, prior to deciding whether the results obtained will aid in establishing or ruling out a particular diagnosis. We propose a strategy for evaluating the literature on clinical examination based on a framework developed for the Users’ Guides to the Medical Literature series.12 In assessing the validity of the study, and interpreting the results, the following points should be considered. ●

Are the results of the clinical examination study valid? 1. 2.

3. 4. 5. ●

Was there an independent blind comparison with a reference (gold) standard of diagnosis? Was the clinical feature evaluated in an appropriate spectrum of patients (like those in whom it would be used in clinical practice)? Was the reference standard applied regardless of the result of the clinical feature? Were the methods of performing the clinical features described in sufficient detail to permit replication? Was there a description of the experience of the individuals doing the examination?

What were the results? 1. 2.

Are likelihood ratios for the results presented, or data necessary for their calculation provided? Has there been consideration given to reproducibility, precision, and disagreement?

The application of the initial five guides will help the reader determine whether the results of the study are likely to be valid. If the results are deemed to be valid, the reader can then go on to interpret the results presented, of which the likelihood ratio (LR) is the most important index in determining how good a particular diagnostic test is. The likelihood ratio is the probability that the results of a test would be expected in a patient with, as opposed to one without, the target disorder. The application of these techniques for critically appraising the cardiovascular history and physical examination will now be described. Clinical features in the cardiovascular history Chest pain There are many causes of chest pain, including both cardiac and non-cardiac conditions, as outlined in Figure 2.1.

Chest pain

Cardiac

Ischemic

Non-cardiac

Non-ischemic

Angina Unstable MI Pericarditis Valvular angina

Gastroesophageal

Non-gastroesophageal

GastroEsophageal Peptic esophageal spasm ulcer reflux disease disease

Aortic Pneumothorax Pulmonary Musculo- Somatoform dissection embolism skeletal disorder: panic attack

Figure 2.1 Cardiac and non-cardiac conditions presenting with chest pain

Elucidating the cause of the pain is important for both management purposes and prognosis. To ensure that the appropriate intervention is undertaken in the clinical setting, it is useful to classify patients presenting with chest pain into three categories: 1. Patients with myocardial infarction 2. Patients with myocardial ischemia but no infarction 3. Patients with non-cardiac chest pain. The characteristics of the chest pain may help differentiate patients into the appropriate category. To identify features of the pain that might aid in classifying patients into category 1, myocardial infarction, we undertook a review of the literature using a search strategy similar to that outlined in the first section above. Relevant articles identified from this search were critically appraised using criteria outlined in the previous section. For the sake of relevance and clarity we have chosen to present only the results of those features in which a likelihood ratio of at least 2·0 or greater, or 0·5 or less, was obtained. The five studies that meet this criterion provide the best available evidence for identifying features of chest pain which aid in the diagnosis of myocardial infarction. As outlined in Table 2.1, the features of the pain that increased the probability of a myocardial infarction included radiation, pain in the chest or left arm, and chest pain described as the most important symptom. Chest pain radiation was the clinical feature which increased the probability of a myocardial infarction the most, with a widespread distribution of pain being associated with the highest likelihood ratios. In particular, chest pain radiating to the left arm was twice as likely to occur in patients with rather than without an acute myocardial infarction, whereas radiation to the right shoulder was three times, and radiation to both the left and right arm seven times, as likely to occur in such patients. The quality of the pain, including pain described as squeezing or pressure, added little to establishing a diagnosis of myocardial infarction, with likelihood ratios of less than 2. 15


Table 2.1 Features of chest pain that increase the probability of a myocardial infarction Clinical feature

References

LR (95% CI)

Chest pain radiation: (R) shoulder (L) arm both (L) and (R) arm Pain in chest or (L) arm Chest pain most important symptom

Tierney et al14 Berger et al13 Berger et al13 Pozen et al15 Pozen et al15

2·9 (1·4–6·0) 2·3 (1·7–3·1) 7·1 (3·6–14·2) 2·7* 2·0*

Abbreviations: CI, confidence interval; LR, likelihood ratio * Data not available to calculate CIs.

Features of the chest pain that decrease the probability of myocardial infarction, and which therefore would be useful in ruling out a myocardial infarction, are outlined in Table 2.2. Pleuritic or positional chest pain, as well as chest pain described as sharp or stabbing, decrease the likelihood of a myocardial infarction. In addition, chest pain reproduced by palpation on physical examination was also associated with a low probability of myocardial infarction. Table 2.2 Features of chest pain that decrease the probability of a myocardial infarction Clinical feature Pleuritic chest pain

Chest pain sharp or stabbing Positional chest pain Chest pain reproduced by palpation

References 14

Tierney et al, Lee et al,16 Solomon et al17 Tierney et al,14 Lee et al16 Lee et al,16 Solomon et al17 Tierney et al,14 Lee et al,16 Solomon et al17

LR (95% CI)

Dyspnea

0·2 (0·2–0·3)

Dyspnea, defined as an uncomfortable awareness of breathing, is a common complaint of both in- and outpatients. Cardiac and pulmonary causes of dyspnea are most common, with congestive heart failure, asthma and chronic obstructive pulmonary disease accounting for most complaints.21 However, standard textbooks of internal medicine list over 30 different etiologies for dyspnea,22 often with multiple etiologies explaining a patient’s symptoms. It is often taught that the cause of dyspnea, of either the heart or the lungs, can be differentiated at the bedside by thorough history-taking. Unfortunately, such strategies to diagnose a cardiac cause for the breathless patient have been incompletely studied. Zema and coworkers23 looked at the value of symptoms as predictors of left ventricular systolic dysfunction in 37 patients with a clinical diagnosis of chronic obstructive pulmonary disease (COPD). Eliciting a symptom of dyspnea on exertion predicted depressed left ventricular systolic function with a sensitivity of 100% and a specificity of 20%. The symptom of orthopnea generated a sensitivity and specificity of 71% and 65%, paroxysmal nocturnal dyspnea 47% and 75%, and ankle edema 41% and 75%, respectively. All features were associated with a likelihood ratio of 2 or less. In general the study was well conducted, but the value of the results to the practicing clinician must be questioned. First, the symptoms of shortness of breath attributed to the heart

0·3 (0·2–0·5) 0·3 (0·2–0·4) 0·2–0·4*

Abbreviations: CI, confidence interval; LR, likelihood ratio * In heterogenous studies the likelihood ratios are reported as ranges.

The precision in obtaining a chest pain history was addressed by Hickman and colleagues,18 who assessed the interobserver agreement in chest pain histories obtained by general internists, nurse practitioners, and self-administered questionnaires for 197 inpatients and 112 outpatients with chest pain. The agreement between two internists for seven of the 10 items, including location and description of the pain, as well as aggravating and relieving factors, was substantial (, a measure of chance-corrected agreement, was 0·50–0·89). Agreement was slightly lower between internist and questionnaire, and between the nurse practitioners and 16

internists, with the lowest level of agreement between nurse and questionnaire. Features of the chest pain associated with a lower probability of myocardial infarction, namely pleuritic, positional and sharp chest pain, were typically associated with a modest level of agreement for all comparisons ( 0·26–0·62). Although cardiac catheterization remains the definitive diagnostic procedure for allocating patients to category 2 – that is, the presence of myocardial ischemia or coronary artery disease – the character of the chest pain has also been identified as one of the most important clinical features in establishing the diagnosis of coronary artery disease.19 The combination of typical angina and a long duration of symptoms was particularly predictive of severe disease. Although this study was undertaken in a very select group of patients (those who underwent cardiac catheterization), similar results were obtained from outpatients referred for non-invasive testing.20 After smoking, typical angina was the variable most strongly associated with significant coronary disease (defined as 75% luminal narrowing of at least one major coronary artery). Subjects with typical angina were 13 times more likely to have significant coronary disease than those without. There are many causes of non-cardiac chest pain, as outlined in Figure 2.1, and each condition has its own characteristic features and associated symptoms. It is beyond the scope of this chapter to identify all these conditions.


were only considered in the context of impaired left ventricular (LV) systolic function. It is now generally agreed that abnormalities in LV diastolic function also cause symptoms of dyspnea. A better gold standard would perhaps have been radionuclide ventriculographic evidence of both LV systolic and diastolic dysfunction. The generalizability of the results is also lessened by the fact that their definition of heart failure was a left ventricular ejection fraction (LVEF) 50%, when in fact the target for treatment of patients with heart failure is most often an LVEF of 40%. Finally, the study was performed in patients who first had a clinical diagnosis of COPD, when patients present with many causes of shortness of breath, not just COPD. In summary, therefore, specific features when elicited in a patient presenting with a complaint of dyspnea are of limited usefulness in making a definitive diagnosis of impaired LV function. Syncope Little detailed evidence exists for either individual or clusters of clinical examination findings in the evaluation of syncope. In a prospective study of 433 syncopal patients presenting in a university setting (emergency, in- and outpatients), the history and physical examination were found to identify 55% (140) of the 254 causes ultimately found.24 Many of the noncardiac causes of syncope in this study were defined in clinical terms, and so provided the “diagnostic standard” for classification. The three most common non-cardiac causes were “orthostatic hypotension” (systolic drop of more than 25 mmHg, or drop of more than 10 mmHg to less than 90 mmHg with symptoms), “situational” (situations included cough, micturition and defecation, and required appropriate timing and no other identifiable cause) and “vasovagal” (requiring a precipitating event and premonitory symptoms), representing 31%, 26% and 25%, respectively, of identifiable causes of syncope overall. Follow-up of the cohort demonstrated a 5 year mortality of 50·5% for cardiac versus 30% for non-cardiac or 24% for unknown causes. This provides some independent validation for the clinical classification criteria. There is a need for further work in this area, particularly in developing and validating practical clinical tools to screen for psychiatric causes, to distinguish patients who will benefit from electrophysiologic testing, and to predict those who will have a positive tilt-table test. Clinical features in the cardiovascular physical examination Apical impulse The apical impulse was first described by William Harvey in 192825 and is one of a number of palpable precordial

pulsations reflecting the underlying movement of the heart and great vessels. Many criteria exist defining the normal location, size and character of the apical impulse, and many generations of medical students have been taught that an “abnormal” apical impulse may assist with the diagnosis of left ventricular enlargement and/or hypertrophy. It is only recently that evidence has been published to support these claims. The relationship between the location and size of the apical impulse and LV size, as determined by two-dimensional echocardiography (gold standard), was evaluated by Eilen and colleagues.26 An apical impulse lateral to the midclavicular line, defined as half the distance between the tip of the acromion process and the sternal notch, was a sensitive (100%) but not specific (18%) indicator for LV enlargement, with a likelihood ratio of only 1·2. Identification of the apical impulse 10 cm from the midsternal line was just as sensitive (100%) but only marginally more specific (33%). An apical diameter of 3 cm was a good indicator of LV enlargement, with a sensitivity of 92% and a specificity of 75%, and was almost four times as likely to occur in patients with, as opposed to those without, LV enlargement (LR3·7). O’Neill and coworkers27 examined the relationship between the location of the apical impulse and the presence or absence of cardiomegaly on chest x-ray (defined as a cardiothoracic ratio greater than 50%). An apical impulse lateral to the midclavicular line had a sensitivity of 57%, a specificity of 76%, and a likelihood ratio of 2·4 for identifying cardiomegaly. Identification of the apical impulse 10 cm from the midsternal line was slightly more sensitive (78%) but considerably less specific (28%), and added little to establishing the diagnosis (LR1·1). The results of this investigation must be accepted with caution, as the gold standard used in this case was chest x-ray, which is not a sensitive or specific marker of LV enlargement. Therefore, the validity of this gold standard must be questioned. This was, however, one of the few studies that also evaluated the variation between observers (interobserver variation) in the clinical assessment of the apical impulse, and reported good agreement on apex palpability (0·72) and moderate agreement on degree of apex displacement (0·56) between two physicians. Eagle and coworkers28 examined several clinical features in 125 inpatients with a variety of cardiac and non-cardiac diagnoses in an attempt to determine which features best predicted LVEF. In general, physician estimates of LVEF were good, with 56% being accurate within 7·5% of measured value; 27% of physicians overestimated and 17% underestimated the LVEF. Multiple regression analysis identified three clinical features most predictive of LVEF, including S3 gallop, hypotension, and sustained LV apical impulse (defined as a palpable impulse greater than two thirds the ventricle systole). In summary, the location, size and character of the apical impulse may be used to assess LV size, LV function and 17


cardiomegaly, either alone or in combination with other clinical features or simple diagnostic tests. However, a number of limitations exist, including the fact that a palpable impulse may only be found in approximately 50% of patients. In addition, the high sensitivity but low specificity associated with determining the location and size of the apical impulse make it a better test for ruling out rather than ruling in LV enlargement, which is good for screening but has limited usefulness at the bedside.

Few studies have assessed the reliability and validity of detecting a third heart sound on physical examination. The studies that have been conducted suggest that the agreement between observers with respect to the presence of a third heart sound is low or moderate at best.29–31 In one study, cardiologists, internists and residents in internal medicine examined 46 patients for the presence or absence of a third heart sound.30 The overall interobserver agreement was poor, with a of only 0·18. A somewhat better agreement for the presence of a third heart sound was achieved in an earlier study by two internists and two cardiologists, with a of 0·40.31 The evidence regarding the validity of the third heart sound is even more limited. Using a computerized phonocardiogram as a gold standard for the presence of a third heart sound, Lok et al 30 report positive and negative predictive values for identifying a third heart sound of 71% and 64%, respectively. Although the reliability and validity of this physical examination finding may be limited, the detection of a third heart sound on physical examination may have important prognostic implications. Drazner and colleagues32 performed a retrospective analysis of 2569 patients with symptomatic heart failure enrolled in the Studies of Left Ventricular Dysfunction treatment trial. In multivariate analyses adjusted for other markers of severity of heart failure, a third heart sound was associated with an almost 50% increased risk of hospitalization for heart failure, or death from pump failure.

physicians was moderate ( 0·56), and agreement between residents and staff was modest ( 0·30). Possible causes for disagreement include positioning of patients, poor lighting, difficulty in distinguishing carotid from venous pulsations, and variation in pressure with respiration. As regards the relation between clinical assessments of CVP and the gold standard of simultaneous pressure measurements through a central venous catheter, one study34 used an attending physician, a fellow, a medical resident, an intern and a student to predict whether four hemodynamic variables, including CVP, were low, normal, high or very high. The sensitivity of the clinical examination at identifying low (0 mmHg), normal (0–7 mmHg) or high (7 mmHg) CVP was 33%, 33% and 49%, respectively. The specificity of the clinical examination at identifying low, normal or high CVP was 73%, 62% and 76%, respectively. In another study, Eisenberg and colleagues35 compared clinical assessments with pulmonary artery catheter readings in 97 critically ill patients. Physicians predicted CVP correctly only 55% of the time, more frequently (27%) underestimating than overestimating (17%). Clinical assessments of a high CVP increase the likelihood that the measured CVP will be high by about fourfold; conversely, clinical assessments of a low CVP make the probability of finding a high measured CVP extremely unlikely (LR 0·2).33 The data demonstrate that clinical assessments of a normal CVP are truly indeterminate, with likelihood ratios approaching 1; such estimates provide no information because they neither increase nor decrease the probability of an abnormal CVP. Apart from less observer variation, CVP estimates are most accurate in patients breathing spontaneously. The precision of the abdominojugular reflux test has not been reported, but its results will vary with the force of abdominal compression. Although this is an insensitive way to diagnose congestive heart failure, the specificity of the test is high.36,37 Moreover, the positive likelihood ratios (6·4 when diagnosis was based on a clinical–radiographic score, and 6·0 when diagnosis was based on emergency room physician judgment) indicate that this is a useful bedside test.1

Central venous pressure

Systolic murmurs

The right internal jugular vein lies directly in line with the right atrium and acts as a manometer, displaying changes in blood flow and pressure caused by right atrial filling, contraction and emptying. Elevated jugular venous pressure reflects an increase in central venous pressure (CVP). The reliability and validity of the clinical assessment of CVP have been assessed in a limited number of studies. In one study, medical students, residents and attending physicians examined the same 50 ICU patients and estimated their CVP as low (5 cm), normal (5–10 cm) or high (10 cm).33 Agreement between students and residents was substantial ( 0·65), agreement between students and attending

Etchells and colleagues2 have published a thorough review of the clinical examination for systolic murmurs. This included a systematic review of the literature and grading of the quality of the original articles. Quality was assessed by the sample size and recruitment (consecutive versus convenience) and whether comparison with the diagnostic standard was done independently and blindly. Useful data for ruling aortic stenosis in or out are given in Tables 2.3 and 2.4. The reliability of the examination by cardiologists for late peaking murmur shape is good ( 0·74), for the presence of murmurs is fair to moderate ( 0·29–0·48),2 but for other maneuvers may be poorer.38

Third heart sound

18


Table 2.3 Features of the clinical examination that increase the probability of aortic stenosis Clinical feature

LR*

Slow rate of rise of carotid pulse Late peaking murmur Soft or absent second heart sound

2·8–130 8–101 3·1–50

* LR, likelihood ratio: range of point estimates from original studies cited Data from Etchells et al 2 Table 2.4 Features of the clinical examination that decrease the probability of aortic stenosis Clinical feature

LR*

Absence of a murmur No radiation to right carotid artery

0 0·05–0·10

* LR, likelihood ratio: range of point estimates from original studies cited Data from Etchells et al 2

Studies of the clinical examination for other etiologies of systolic murmur were also reviewed but tended to be of lesser quality than those addressing aortic stenosis. Subsequent to their original work,2 Etchells and colleagues have gone on to develop a two-stage prediction rule for moderate–severe aortic stenosis (defined as an average valve area of less than or equal to 1·2 cm2 or a peak gradient at or above 25 mmHg).39 In this rule a murmur not radiating to the right clavicle was associated with a likelihood ratio of 0·1 (95% CI 0·02–0·44), significantly reducing the likelihood of aortic stenosis. If the murmur did radiate to the clavicle, the presence of 0–2 associated findings increased the likelihood ratio to 1·76 (95% CI 0·9–2·9), and 3–4 associated findings resulted in a likelihood ratio of 40 (95% CI 6·6–239), suggesting that the diagnosis of aortic stenosis is supported by a greater number of associated findings. The associated findings were reduced carotid volume, slow carotid upstroke, reduced second heart sound intensity, and murmur intensity in the second right intercostal space as loud as or louder than in the fifth left intercostal space. Etchells and colleagues2 point out that the majority of studies of this topic have used cardiologists as observers. The performance of non-cardiologists appears to be less accurate when studied. Further work, like their own, using a broader range of clinicians and patients, is needed to discover the value of the clinical examination in more general settings. Blood pressure An extensive review of the technique, reliability and validity of blood pressure (BP) measurement has been provided by

Reeves.3 As outlined in the review, two important sources of variation in BP measurement include the patient and the examiner. Random fluctuation in BP over time has been documented by the SD of readings, with a minute-to-minute variation of about 4 mmHg systolic and 2–3 mmHg diastolic, and day to day variation of 5–12 mmHg systolic and 6–8 mmHg diastolic. With respect to the examiner as the source of variability, differences of 10–8 mmHg by both physicians and nurses in routine medical practices have been noted. Intra-arterial blood pressure measurement has been used as the gold standard to assess the accuracy of indirect BP measurement. With the indirect BP the phase I Korotkoff, or first audible sound, appears 15–4 mmHg below the direct systolic BP, whereas phase V, or disappearance of all sounds, appears 3–6 mmHg above the true diastolic BP in adults. Other factors that affect the accuracy of the indirect BP measurement, resulting in both an increase and a decrease in systolic and/or diastolic measurements, are outlined in Tables 2.5 and 2.6. Table 2.5 pressure

Factors associated with an increase in blood

Factor Examinee Pseudohypertension “White coat reaction” to physician “White coat reaction” to non-physician Paretic arm (due to stroke) Pain, anxiety Acute smoking Acute caffeine Acute ethanol ingestion Distended bladder Talking, sighing Setting, equipment Leaky bulb valve Blocked manometer vents Examination Cuff too narrow Cuff not centered Cuff over clothing Elbow too low Back unsupported Arm unsupported Too slow deflation Too fast deflation Parallax error Using phase IV (adult) Too rapid remeasure Cold season (v warm)

Magnitude, SBP/ DBP (mmHg) 2–98/3–49 11–28/3–15 1–12/2–7 2/5 May be large 6/5 11/5 8/8 15/10 7/8 2 DBP 2 to 10 8–10/2–8 4/3 5–50 6 6–10 1–7/5–11 1–2/5–6 DBP only 2–4 6 DBP 1/1 6/3–10

Abbreviations: DBP, diastolic blood pressure; SBP, systolic blood pressure Data from Reeves et al 3

19


Table 2.6 Factors associated with a decrease in blood pressure Factor

Magnitude, SBP/ DBP (mmHg)

Examinee Recent meal Missed auscultatory gap High stroke volume Habituation Shock (additional pseudohypotension) Setting, equipment Faulty aneroid device Leaky bulb

1–1/1–4 10–50 SBP Phase V can0 0–7/2–12 33 SBP Can be 10 2 SBP

Examiner Reading to next lowest 5 or 10 mmHg or expectation bias Impaired hearing

Probably 10 SBP only

Examination Left v right arm Resting for too long (25 min) Elbow too high Too rapid deflation Excess bell pressure Parallax error (aneroid)

1/1 10/0 5/5 SBP only 9 DBP 2–4

Abbreviations: DBP, diastolic blood pressure; SBP, systolic blood pressure Data from Reeves et al 3

Arterial pulse Few studies have been undertaken to assess the reliability and validity of features of the arterial pulse in the cardiovascular examination, despite numerous descriptive accounts of its variability in different clinical conditions. Case series indicate that details regarding the presence and quality of the arterial pulse are more sensitive markers of coarctation of the aorta than aortic dissection. Absent femoral pulses or a femoral/brachial pulse discrepancy in patients was associated with a sensitivity of 88% in the diagnosis of coarctation of the aorta in patients less than 6 months of age.40 Similar results were obtained for patients diagnosed with coarctation after 1 year of age, where weak or absent femoral pulses were associated with a sensitivity of 85%.41 The sensitivity of the presence and quality of the carotid, subclavian and femoral pulses in establishing a diagnosis of both proximal (primary tear in the ascending aorta with or without involvement of the arch, De Bakey classification type I and II) and distal (primary tear in the descending thoracic aorta, De Bakey classification type III) aortic dissections are outlined in Table 2.7. Proximal dissections were primarily associated with an absence or decrease in the brachiocephalic vessels, whereas distal dissections almost exclusively involved the femoral arteries. 20

Table 2.7 Sensitivity of the arterial pulse in the diagnosis of aortic dissection Aortic dissection (%) References

Proximal*

Distal†

Lindsay and Hurst42§ Slater and De Sanctis43§ Spittell et al 44**

62·5 50·9 9·0

10·5 15·5 2·4

* De Bakey classification type I and II. De Bakey classification type III. § Absence or decrease in amplitude of carotid, subclavian or femoral pulse(s). **Absence of palpable carotid, subclavian or femoral pulse(s). †

Features of the arterial pulse may also be used to determine the presence of valvular heart disease. As reported by Etchells et al,2 features of the arterial pulse, including rate of rise of the carotid pulse, apical carotid delay and brachioradial delay, all increase the likelihood of establishing the diagnosis of aortic stenosis (Table 2.8). Table 2.8 Features of the arterial pulse that increase the probability of aortic stenosis Clinical feature

LR*

Slow rate of rise of carotid pulse Apical carotid delay Brachioradial delay

2·8–130 ∞ 6·8

* LR, likelihood ratio: range of point estimates from original studies cited. Data from Etchells et al 2

The diagnostic value of the pedal pulse examination, as an aid to establishing the diagnosis of peripheral arterial disease, has also been studied.45 In this review the absence of both the dorsalis pedis and posterior tibial pulses was a powerful predictor for the presence of vascular disease (defined as an ankle-to-arm systolic pressure index of 0·9), with likelihood ratios ranging from 9·0 to 44·6. The presence of a femoral arterial bruit was also a strong indicator of disease, with likelihood ratios of 4·7–5·7. Heart rate is another important component of the cardiovascular examination. The accuracy of the assessment of heart rate may be affected by both the site (apical or radial) as well as the counting interval (15, 30 or 60 seconds). With a regular rhythm, radial 15 second counts were the least accurate for both resting and rapid heart rates, whereas the 30 second counts were found to be the most accurate and efficient for rapid rates.46 With the irregularly irregular rhythm of atrial fibrillation, however, the apical method and 60 second count have been reported to be the most accurate, with site being a more important source of error than


counting interval.47 Using the ECG as the measure of true heart rate, the mean radial error for all counting intervals was 19·5 beats per minute, which was significantly higher than the mean apical error of 9·7 beats per minute. Although the pulse in atrial fibrillation is typically described as “irregularly irregular”, Rawles and Rowland,48 using computerized analysis of R–R intervals and pulse volumes in patients with atrial fibrillation, disputed this assumption. In an assessment of 74 patients with atrial fibrillation they reported a non-random sequence of R–R intervals in 30%, and the presence of pulsus alternans in less than half (46%). The authors concluded that patterns of regularity of the pulse are common in patients with atrial fibrillation.

aortic dissection has been limited to case series, therefore estimates of sensitivity only are available. Features of the arterial pulse have been shown to be relatively sensitive markers for coarctation of the aorta and for chronic lower extremity ischemia, but less so for aortic dissection. Finally, both counting interval and site (radial versus apical) have important implications on the accuracy of heart rate assessment. As is evident from the information presented, unfortunately, for a variety of reasons, research on clinical examination has lagged behind basic science and therapeutic research. So far, clinical examination is identified as the “art” of medicine, and by incorporating an evidence-based approach one can make clinical examination the “art and science” of medicine.

Summary Despite the frequency with which details of the history and physical examination are used to establish or rule out a particular cardiovascular condition, there is a very limited amount of data available to support the reliability and validity of these features. The one component of the cardiovascular history which has been studied is that of chest pain in the diagnosis of myocardial infarction. Features of chest pain, particularly pain that has a wide distribution of radiation, increase the probability of myocardial infarction, whereas chest pain that is pleuritic, sharp or stabbing, positional or reproduced by palpation, decreases the probability of myocardial infarction. The reliability and validity of various features of the cardiovascular physical examination have also received little attention in the literature. Of those that have been studied, the apical impulse has been shown to be a sensitive but nonspecific marker of LV size, which makes it useful for ruling out, rather than ruling in, LV enlargement. Clinical assessment of elevated CVP has been shown to be associated with a fourfold likelihood that the measured CVP will be high, with the abdominojugular reflex being a useful bedside test to assist in the diagnosis of congestive heart failure. Of the cardiac murmurs, aortic stenosis has been studied the most thoroughly. Features of the clinical examination that increase the probability of diagnosing aortic stenosis include slow rate of rise of the carotid pulse, late peaking murmur, and soft or absent second heart sound. Conversely, absence of a murmur or no radiation to the right carotid artery or clavicle were features associated with a decreased probability of aortic stenosis. Recent work would suggest that the presence of an increased number of associated findings increases the likelihood of aortic stenosis. A number of features have been shown to influence the accuracy of the indirect assessment of BP, including those related to the examinee, the examiner, the setting and equipment, and the examination itself. Assessment of the arterial pulse in diagnosing coarctation of the aorta and

References 1.Cook DJ, Simel DL. Does this patient have abnormal central venous pressure? JAMA 1996;275:630–4. 2.Etchells E, Bell C, Robb K. Does this patient have an abnormal systolic murmur? JAMA 1997;277:564–71. 3.Reeves RA. Does this patient have hypertension? JAMA 1995; 273:1211–18. 4.Choudhry NK, Etchells EE. Does this patient have aortic regurgitation? JAMA 1999;281:2231–8. 5.Panju AA, Hemmelgarn BR, Guyatt GH, Simel DL. Is this patient having a myocardial infarction? JAMA 1998;280: 1256–63. 6.Turnbull JM. Is listening for abdominal bruits useful in the evaluation of hypertension? JAMA 1995;274:1299–301. 7.Badgett RG, Lucey CR, Mulrow CD et al. Can the clinical examination diagnose left-sided heart failure? JAMA 1997; 277:1712–19. 8.Lederle FA, Simel DL. Does this patient have abdominal aortic aneurysm? JAMA 1999;281:77–82. 9.McGee S, Abernathy WB, Simel DL. Is this patient hypovolemic? JAMA 1999;281:1022–9. 10.Simel D (Section editor, Rational Clinical Examination, JAMA). Personal communication, December 1996. 11.McKibbon KA, Walker-Dilks CJ. Beyond ACP Journal Club: How to harness MEDLINE for diagnostic problems (Editorial). ACP J Club 1994;121:A10–A12. 12.Oxman AD, Sackett DL, Guyatt GH. Users’ Guides to the Medical Literature: 1. How to get started. JAMA 1993;270: 2093–5. 13.Berger JP, Buclin R, Haller E, Van Melle G, Yersin B. Right arm involvement and pain extension can help to differentiate coronary diseases from chest pain of other origin: a prospective emergency ward study of 278 consecutive patients admitted for chest pain. J Intern Med 1990;227:165–72. 14.Tierney WM, Fitzgerald D, McHenry R et al. Physicians’ estimates of the probability of myocardial infarction in emergency room patients with chest pain. Med Decis Making 1986;6:12–17. 15.Pozen MW, D’Agostino RB, Selker HP, Sytkowski PA, Hood WB. A predictive instrument to improve coronary-care-unit

21


admission practices in acute ischemic heart disease. N Engl J Med 1984;310:1273–8. 16.Lee TH, Cook EF, Weisberg M et al. Acute chest pain in the emergency room. Arch Intern Med 1985;145:65–9. 17.Solomon CG, Lee TH, Cook EF et al. Comparison of clinical presentation of acute myocardial infarction in patients older than 65 years of age to younger patients: the multicenter chest pain study experience. Am J Cardiol 1989;63:772–6. 18.Hickman DH, Sox HC, Sox CH. Systematic bias in recording the history in patients with chest pain. J Chron Dis 1985; 38:91–100. 19.Pryor DB, Shaw L, Harrell FE et al. Estimating the likelihood of severe coronary artery disease. Am J Med 1991;90:553–62. 20.Pryor DB, Shaw L, McCants CB. Value of the history and physical in identifying patients at increased risk for coronary artery disease. Ann Intern Med 1993;118:81–90. 21.Mulrow CD, Lucey CR, Farnett LE. Discriminating causes of dyspnea through clinical examination. J Gen Intern Med 1993;8:383–92. 22.Ingram RH Jr, Braunwald E. Dyspnea and pulmonary edema. In: Wilson JD et al., eds. Harrison’s principles of internal medicine, 12th edn. New York: McGraw-Hill, 1991. 23.Zema MJ, Masters AP, Malgouleff D. Dyspnea: the heart or the lungs? Differentiation at bedside by use of the simple valsalva maneuver. Chest 1984;85:59–64. 24.Kapoor WN. Evaluation and outcome of patients with syncope. Medicine 1990;69:160–75. 25.Harvey W. An anatomical disquisition on the motion of the heart and blood in animals. London, 1928. (Translated from the Latin by Robert Willis, Barnes, Surrey, England, 1847.) In: Willius FA, Key TE. Classics of cardiology, vol. 1. Malabar, Florida: Robert E. Krieger, 1983. 26.Eilen SD, Crawford MH, O’Rourke RA. Accuracy of precordial palpation for detecting increased left ventricular volume. Ann Intern Med 1983;99:628–30. 27.O’Neill TW, Barry M, Smith M, Graham IM. Diagnostic value of the apex beat. Lancet 1989;i:410–11. 28.Eagle KA, Quertermous T, Singer DE et al. Left ventricular ejection fraction. Physician estimates compared with gated blood pool scan measurements. Arch Intern Med 1988;148:882–5 29.Westman EC, Matchar DB, Samsa GP, Mulrow CD, Waugh RA, Feussner JR. Accuracy and reliability of apical S3 gallop detection. J Gen Intern Med 1995;10:455–7. 30.Lok CE, Morgan CD, Ranganathan N. The accuracy and interobserver agreement in detecting the “gallop sounds” by cardiac auscultation. Chest 1998;114:1283–8. 31.Ishmail AA, Wing S, Ferguson J, Hutchinson TA, Magder S, Flegel KM. Interobserver agreement by auscultation in the

22

presence of a third heart sound in patients with congestive heart failure. Chest 1987;91:870–3. 32.Drazner MH, Rame JE, Phil M, Stevenson LW, Dries DL. Prognostic importance of elevated jugular venous pressure and a third heart sound in patients with heart failure. N Engl J Med 2001;345:574–81. 33.Cook DJ. The clinical assessment of central venous pressure. Am J Med Sci 1990;299:175–8. 34.Connors AF, McCaffree DR, Gray BA. Evaluation of right heart catheterization in the critically ill patient without acute myocardial infarction. N Engl J Med 1983;308:263–7. 35.Eisenberg PR, Jaffe AS, Schuster DP. Clinical evaluation compared to pulmonary artery catheterization in the hemodynamic assessment of critically ill patients. Crit Care Med 1984;12:549–53. 36.Marantz PR, Kaplan MC, Alderman MH. Clinical diagnosis of congestive heart failure in patients with acute dyspnea. Chest 1990;97:776–81. 37.Maisel AS, Atwood JE, Goldberger AL. Hepatojugular reflux: useful in the bedside diagnosis of tricuspid regurgitation. Ann Intern Med 1984;101:781–2. 38.Spodick DH, Sugiura T, Doi Y, Paladion D, Jaffty BG. Rate of rise of the carotid pulse: an investigation of observer error in a common clinical measurement. Am J Cardiol 1982;49:159–62. 39.Etchells E, Glenns V, Shadowitz S, Bell C, Siu S. A bedside clinical prediction rule for detecting moderate or severe aortic stenosis. J Gen Intern Med 1998;13:699–704. 40.Ward KE, Pryor RW, Matson JR et al. Delayed detection of coarctation in infancy: implications for timing of newborn follow-up. Pediatrics 1990;86:972–6. 41.Strafford MA, Griffiths SP, Gersony WM. Coarctation of the aorta: a study in delayed detection. Pediatrics 1982;69:159–63. 42.Lindsay J, Hurst JW. Clinical features and prognosis in dissecting aneurysm of the aorta. Circulation 1967;35:880–8. 43.Slater EE, DeSanctis RW. The clinical recognition of dissecting aortic aneurysm. Am J Med 1976;60:625–33. 44.Spittell PC, Spittell JA, Joyce JW et al. Clinical features and differential diagnosis of aortic dissection: experience with 236 cases (1980 through 1990). Mayo Clin Proc 1993;68:642–51. 45.McGee SR, Boyko EJ. Physical examination and chronic lowerextremity ischemia. Arch Intern Med 1998;158:1357–64. 46.Hollerbach AD, Sneed NV. Accuracy of radial pulse assessment by length of counting interval. Heart Lung 1990;19: 258–64. 47.Sneed NV, Hollerbach AD. Accuracy of heart rate assessment in atrial fibrillation. Heart Lung 1992;21:427–33. 48.Rawles JM, Rowland E. Is the pulse in atrial fibrillation irregularly irregular? Br Heart J 1986;56:4–11.

3

Obtaining incremental information from diagnostic tests Raymond J Gibbons

Consider the following case history. A 75 year old male presents with a history of exertional chest pain. The patient describes substernal chest pain that he perceives as a “pressure sensation” occurring when he walks too fast, uphill, or in the cold. It is relieved by rest within a few minutes. On two recent occasions, he tried a friend’s nitroglycerin tablets, and obtained even more rapid relief of his symptoms. His symptoms have never occurred at rest. The patient has a history of diabetes mellitus, hypertension, and hypercholesterolemia. He smokes one pack of cigarettes a day. Several male family members died of coronary artery disease before the age of 60. The patient underwent carotid artery surgery a year ago for treatment of transient ischemic attacks. On the basis of his age, gender, chest pain description, and risk factors, this patient is highly likely to have significant obstructive coronary artery disease (CAD). The added, or incremental, value of any stress test for the diagnosis of the presence of disease in such a situation is very small. Out of 100 patients with this presentation, perhaps only one or two will not have obstructive CAD. The potential contribution of stress testing is therefore restricted to only these one or two patients. This example demonstrates the importance of the concept of incremental value for diagnostic tests. In the current era of healthcare reform, it is no longer sufficient that a test simply provide “more information”. The more appropriate current questions are: ● ●

how much information does the test provide, and at what cost?

Increasingly, tests are also required to have a demonstrable impact on critical nodal, or decision, points with respect to patient management. The demonstration of the incremental value of diagnostic tests requires rigorous methodology. The principles of the required methodology should be credited primarily to Dr George Diamond and his colleagues at Cedar Sinai Medical Center in Los Angeles.1–3 First and foremost, such an analysis should reflect clinical decision making. Since clinical assessment is performed before any diagnostic tests, and usually at lower cost, parameters available from this assessment should be considered separately without any information from subsequent testing. The analysis should

preferably focus on hard, demonstrable end points such as significant obstructive CAD, severe (three vessel or left main) coronary artery disease, myocardial infarction, or death. Although alternative end points, such as functional impairment, unstable angina, and the need for revascularization, are often included to increase statistical power, such end points have major limitations with respect to reversibility, subjectivity, and definite impact on patient outcome. The analysis should create appropriate models that include all available important variables. An experienced clinician always takes the patient’s age, gender, and history into account in making his or her clinical decision regarding patient management, even when testing results are available. These important clinical parameters must therefore be included in any final model that reflects the clinical decision making process. The analysis must demonstrate that the additional information is statistically significant in an appropriate patient population. Analyses that demonstrate additional information in older, “sicker” inpatient populations should not be casually extrapolated to younger, “less sick” outpatients in whom testing is customarily performed. Finally, the test must provide information that is clinically significant and cost effective. In very large patient samples, differences that have little, if any, clinical significance for individual patient management may emerge as statistically significant. The potential impact on patient management in some patients must compare favorably with the incremental cost of the test in all the patients who must be tested. This chapter will attempt to elucidate this methodology using the published data with respect to the diagnosis of significant obstructive CAD, non-invasive screening for severe CAD, and patient outcome. All of these examples are drawn from the arena of ischemic heart disease, because this entity is a predominant feature of clinical practice in cardiology, and the published literature is voluminous and extensive. However, the same principles apply to other disease entities, both cardiac and non-cardiac.

Clinical assessment As outlined above, the initial step in any analysis designed to demonstrate incremental value is the consideration of all 23


the information available prior to performance of the test. This will always include the results of the history and physical examination, and may sometimes include the results of other tests already performed. This section focuses on the information available from clinical assessment.

A more comprehensive attempt to consider all clinical characteristics, including risk factors for atherosclerosis, was published from the Duke University Medical Center databank.6 In addition to the three parameters previously discussed, this analysis found that evidence for previous infarction, smoking, hyperlipidemia, ST and T wave changes on the resting electrocardiogram (ECG), and diabetes were all highly significant predictors of the presence of coronary artery disease. Figure 3.1 shows a published nomogram for men that incorporates all of these parameters. Careful inspection of this figure demonstrates that the impact of the clinical parameters other than age, gender, and chest pain is variable. ECG and historical evidence of previous infarction have a major impact, diabetes and ECG ST-T changes have a modest impact, and lipids and smoking have a minimal impact. For example, a 50 year old male with atypical angina has a 46% pretest probability of disease in the absence of smoking, hyperlipidemia, or diabetes, a 48% pretest probability in the presence of both smoking and hyperlipidemia, and a 65% pretest probability if he has diabetes as well. In the presence of ECG Q waves and a history of MI, his pretest probability exceeds 90%.

Diagnosis of coronary disease As demonstrated by the earlier example, clinicians often encounter patients with chest pain and suspected CAD. The ability of clinical assessment to predict the likelihood of significant obstructive CAD has been demonstrated in numerous studies. The likelihood of significant disease based on clinical assessment is appropriately labeled the “pretest probability”, in statistical terms. Age, gender, and the patient’s chest pain description are the most important clinical parameters for estimating the likelihood of CAD.4 Older patients, men, and patients with chest pain that is typical, or classic, for angina pectoris are more likely to have coronary disease. Although multiple different systems have been used to classify chest pain, the simplest and easiest was proposed by Diamond.5 He suggested a classification based on three elements – substernal location, precipitation by exertion, and relief by rest or nitroglycerin. If all three elements are present, the chest pain is classified as “typical angina”. If two elements are present, the chest pain is classified as “atypical angina”. If only one or none is present, the chest pain is classified as “non-anginal chest pain”. Table 3.1 shows published estimates of pretest probability on the basis of age, gender, and chest pain description.4 It is obvious that there is a very wide range of pretest probability, ranging from 1% for a 35 year old woman with non-anginal chest pain to 94% for a 65 year old man with typical angina. Note that a 50 year old man with atypical angina has about a 50% probability of disease.

Non-invasive screening for severe coronary artery disease Not surprisingly, clinical parameters are also very important in estimating the likelihood of severe (three vessel or left main) CAD.7 The same parameters that are most important for predicting the presence of disease – age, gender, and chest pain description – remain important. In addition, diabetes mellitus and history or ECG evidence of myocardial infarction are also very important. The simplest approach for estimating the likelihood of severe disease was published by Hubbard et al.8 They assigned one point each for: male gender; typical angina; history and ECG evidence of myocardial infarction;

Table 3.1 Pretest probability of coronary artery disease Age (years)

30–39 40–49 50–59 60–69

Pretest probability (%) Asymptomatic

Non-anginal chest pain

Atypical angina

Typical angina

F

M

F

M

F

M

F

M

1 1 4 8

2 6 9 11

1 3 8 19

5 14 22 28

4 13 32 54

22 46 59 67

26 55 79 91

70 87 92 94

From Diamond and Forrester.4 Reprinted by permission of the New England Journal of Medicine, and Diamond GA. A clinically relevant classification of chest discomfort. J Am Coll Cardiol 1983;1:547–75

24

Obtaining incremental information from diagnostic tests

AGE No smoking or lipids Lipids only Smoker only History – ECG Point score

80

80

Angina – Typical – 26 Atypical – 10 Non-angina – 0 Previous MI History only – 11 ECG Q waves – 12 Both – 30

Both

80

70

Point score

80

70 70

70 60

60 70

ECG ST-T Changes – 6 Diabetes mellitus – 7

Probability of significant CAD

50 60

50

50 40 30

99 98 97

50

60

60

50

40 40

20

95 30

Directions for use Step 1: Calculate the history ECG score and locate point on left scale

40

Step 2: Using the appropriate age scale extrapolate to the reading line for age (right scale) Step 3: Place a ruler between the point score and age (moved to the reading line) Step 4: Read off the probability of significant CAD on the center scale

30

90 85 80 70 60 50 40 30 20 15 10 05

20

40

30 20 20

30

20

03 02 01

Reading line

10

0

Figure 3.1 Nomogram for predicting the probability of significant coronary artery disease (CAD) in men. ECG, electrocardiogram; MI, myocardial infarction. (After Pryor et al.6) Example: A 50 year old, white male with atypical angina and diabetes mellitus, but no ECG ST changes, previous MI, smoking, or hyperlipidemia. Point score on left scale 10 717. Appropriate reading line on right labeled “no smoking or lipids”. Connect age 50 on this reading line to point score of 17 with a straight edge. This intersects the middle line at 60, indicating that this is the percentage probability of significant CAD.

diabetes; and insulin use. Thus, the point score had a minimum value of 0 and a maximum value of 5. Figure 3.2 shows a nomogram for the probability of severe CAD based on age and this point score. It is quickly apparent that age is an extremely important parameter for predicting severe disease.

A more comprehensive analysis on a larger number of patients was published from the Duke University Medical Center databank.9 In addition to the five parameters already mentioned, these workers found that the duration of chest pain symptoms, other risk factors (blood pressure, 25


Predicted probability

1·0 5

0·8

4 3

0·6

2 1

0·4

0 0·2 0 30

35

40

45

50

55 60 Age, yr

65

70

75

80

Figure 3.2 Nomogram showing the probability of severe (three vessel or left main) coronary artery disease based on a 5 point score. One point is awarded for each of the following variables – male gender, typical angina, history, and electrocardiographic evidence of MI, diabetes, and use of insulin. Each curve shows the probability of severe coronary disease as a function of age. (From Hubbard et al,8 with permission.)

Approaches to the assessment of incremental value Once the information available from clinical assessment (and other tests already performed) has been considered, there are a variety of conceptual and statistical approaches that can be employed to assess the incremental value of the test in question. This section will present examples of three such approaches.

Diagnosis of CAD The application of multiple different stress tests for the diagnosis of coronary artery disease has been extensively studied. The most common approach used in this setting to demonstrate the incremental value of a new test employs Bayes’ theorem.17 This theorem indicates that the likelihood of disease following testing (post-test probability) can be calculated from the test characteristics (sensitivity and specificity) and the pretest probability. This calculated posttest probability is often plotted graphically as a function of pretest probability (Figure 3.3).

hyperlipidemia, and smoking), a carotid bruit, and chest pain frequency were also important determinants of the likelihood of severe CAD. However, the magnitude of their additional effect was modest.

1·0 Acceptable B

Prediction of patient outcome The ability of clinical assessment to predict patient outcome has been demonstrated in numerous previous studies. The largest and most important of these came from the Duke University databank10 and the Coronary Artery Surgical Study Registry.11 Many of the same parameters that predict the presence of disease and the presence of severe disease are also associated with adverse patient outcome. Age, gender, chest pain description, and previous myocardial infarction all have independent value in predicting patient outcome. In addition, history and physical examination evidence for congestive heart failure, history and physical examination evidence of vascular disease, unstable chest pain characteristics, and other ECG findings, such as ST-T wave changes, left bundle branch block, and intraventricular conduction delay, all have prognostic value. It is not generally appreciated how well clinical parameters perform in this regard. The Duke group reported that 37% of the patients undergoing stress testing at their institution had a predicted average annual mortality of 1% or less over the next 3 years, on the basis of clinical assessment.11 Several studies have shown that a normal resting ECG, and the absence of a history of prior infarction, predict a normal ejection fraction with 90% confidence,12,13 and therefore a favorable prognosis.14–16 26

Post-test probability

0·8 Positive test 0·6

C

Negative test

0·4

0·2 A Acceptable 0·2

0·4 0·6 Pretest probability

0·8

1·0

Figure 3.3 Relationship between pretest probability and post-test probability. The solid curves for positive and negative tests are plotted for a test with 80% sensitivity and a 90% specificity. Post-test probabilities that are acceptable for diagnosis (90% and 10%) are shown in the shaded zones. Line A represents a patient with a very low pretest probability; line B, a patient with a high pretest probability; line C, a patient with an intermediate probability. (Modified from Berman DS, Garcia EV, Maddahi J. Thallium-201 scintigraphy in the detection and evaluation of coronary artery disease. In: Berman DS, Mason DT, eds. Clinical nuclear cardiology. New York: Grune and Stratton, 1981, with permission.)


In Figure 3.3, the pretest probability is shown on the X-axis and the post-test probability is shown on the Y-axis. The dotted line represents the line of identity. The vertical distance from this line to the upper solid curve represents the increase in the probability of disease as a result of positive test. In analogous fashion, the vertical distance from this dotted line to the lower solid curve represents the decrease in probability as a result of a negative test. The solid vertical lines describe three different clinical situations. Line A represents a patient with a very low pretest probability, such as a 40 year old woman with non-anginal pain. A negative test changes probability very little. A positive test increases probability somewhat, but the post-test probability remains well under 50%, and the test is most likely a “false positive”. Line B represents a patient with a high pretest probability of disease, such as a 65 year old man with typical angina. A positive test will increase the probability only slightly. A negative test will decrease the probability of disease somewhat, but the post-test probability remains substantially greater than 50%, so that the test is most likely a “false negative”. The final situation (line C) represents a patient with an intermediate probability of disease, such as a 50 year old male with atypical angina. A positive test in such a patient would increase the probability of disease substantially to near 90%. On the other hand a negative test would decrease the probability of disease substantially to approximately 18%. Thus, it is evident that the incremental value of diagnostic testing is greater in patients with an intermediate probability of disease, a principle that is broadly recognized.17 However, it is also recognized that this kind of analysis has a number of limitations. The single curves for positive and negative tests do not take into account the degree of test abnormality. The test results are therefore better displayed for a whole range of values for a parameter that helps distinguish normal from abnormal. The best known example of this would be the magnitude of ST segment depression on treadmill exercise testing.18 In addition, multiple other parameters are reported during a treadmill exercise test, which help to distinguish severely abnormal tests from only mildly abnormal tests.19 Ideally, all of these parameters would be incorporated into a single “score” and a series of curves would be plotted. Next, construction of such curves relies on the premise that the sensitivity of tests will be identical for any population of patients with disease regardless of disease prevalence. This assumption is usually invalid. As demonstrated in the previous section, those parameters which help to identify the presence of disease also help to identify the presence of severe disease. In general, the sensitivity of most tests is greater in patients with more severe disease. It is therefore quickly evident that sensitivity would be expected to vary with the prevalence of disease. This point has been demonstrated by several investigators,20 and

provides justification for the use of logistic regression analysis for diagnostic purposes.21 Despite these limitations, bayesian analysis serves as a useful framework for understanding the potential incremental value of diagnostic tests. Post-test referral bias, also known as work up bias or verification bias, occurs whenever the results of the test in question influence the subsequent performance of the “reference” test (sometimes referred to as the “gold standard”). This bias has been recognized for more than 20 years.22 An early survey of the literature on exercise testing showed that only 2 of 33 studies avoided this bias.23 The recognition of the importance of this phenomenon was emphasized in a landmark paper in 1983, which described the “declining specificity” of radionuclide angiography as a result of this bias.24 More than 10 years ago, a monograph from the Institute of Medicine emphasized this well established concept.25 The key question to ascertain whether postreferral bias is present is “did the results of the test being evaluated influence the decision to perform reference standard?”.26 Although this bias potentially occurs for any diagnostic test, it is particularly important for non-invasive diagnostic tests for CAD. Patients with positive non-invasive tests are often referred to coronary angiography (the “reference” test). In contrast, patients with negative tests are often sent home without coronary angiography. The effects of this preferential referral to coronary angiography are to markedly decrease the observed specificity of the test in question and modestly increase its sensitivity. The clearest solution to the problem of post-test referral bias is to avoid it completely by studying patients in whom the decision to proceed with the “reference” test is made before the performance of the diagnostic test in question.25 For the diagnosis of CAD, this standard is incredibly difficult and rarely achieved. A more feasible alternative is the mathematical correction of sensitivity and specificity for post-test referral bias using one of two published formulae and information about all of the patients who were studied using the diagnostic test in question and did not proceed with coronary angiography.27,28 There are a number of published studies demonstrating the effect of these corrections on the observed test performance for exercise electrocardiographic testing,29 exercise echocardiography,30 and exercise Single Photon Computed Tomography (SPECT) perfusion imaging.31 Correction for referral bias markedly increases the specificity and modestly decreases the sensitivity of these tests. As a result, the predictive value of a positive test is improved, but the predictive value of a negative test decreases. It is generally difficult to confirm the validity of these corrections. However, a carefully designed prospective study of exercise echocardiography in women has now reported sensitivity and specificity values that are very close to those reported after correction for referral bias.32 27


Post-test referral bias has numerous important implications for the interpretation of the diagnostic literature.32 Many of the reported sensitivity and specificity values are very likely to be erroneous.32 Widespread misconceptions exist regarding gender differences in test performance. The post-test probability of CAD is higher for either a positive or negative test than that which would be calculated from Bayes’ theorem using the reported values of sensitivity and specificity.31 Non-invasive screening for severe CAD The incremental value of testing for the diagnosis of severe CAD has been studied using both bayesian analysis and logistic regression analysis. When the latter analysis is conducted properly, all of the previously discussed clinical parameters that are associated with severe coronary disease are incorporated into a model that is used to predict the probability of severe CAD. The output of such a model is a probability that ranges between 0 and 1. It is critically important that these candidate variables be “forced” into the model, even if they are statistically insignificant in the population under study. Most study populations are too

small to have adequate power to detect the true significance of these variables, which has been demonstrated in very large subsets. For example, age should always be forced into such models, even if it does not appear to be significant in the particular population in question, because there is abundant evidence that it should always be considered (and indeed usually is by clinicians). Using this approach, a second model should then be constructed which includes all of the clinical parameters, as well as pertinent new parameters from the test in question. If these parameters have statistical significance independent of the clinical parameters, the test has incremental value. This approach is demonstrated in Table 3.2, which shows the improvement in the logistic regression model for severe CAD reported by Christian et al,16 when the exercise test was added to clinical parameters, and when thallium imaging parameters were added to clinical and exercise parameters. An alternative approach is to construct the receiver operating characteristic (ROC) curves, which display sensitivity and specificity as a function of the predicted probability of severe disease (the output of a logistic regression model). The area under the ROC curve can then be compared between the model that incorporates clinical parameters,

Table 3.2 Logistic regression multivariate analysis: prediction of three vessel or left main (coronary artery) disease Model

Direction

Odds ratio (95% CI)

P value

Clinical Diabetes mellitus Typical angina Sex Agea

2 31·3

Present Present Male Older

2·0 (1·3–3·1) 2·3 (1·4–3·9) 3·2 (1·4–4·0) 1·4 (1·1–1·9)

0·001 0·001 0·007 0·01

Clinical and exercise Diabetes mellitus Typical angina Sex Agea Magnitude of ST depression Peak heart rate peak systolic blood pressureb 2 65·0

Present Present Male Older More Lower

1·9 (1·2–3·0) 1·9 (1·1–3·3) 2·3 (0·9–5·3) 1·2 (0·9–1·7) 1·5 (1·3–1·8) 0·9 (0·86–0·95)

0·005 0·02 0·07 0·16 0·001 0·001

Clinical, exercise, and thallium-201 Diabetes mellitus Typical angina Sex Agea Peak heart rate peak systolic blood pressureb Magnitude of ST depression Global T1-201 score (delayed – after exercise) 2 70·4

Present Present Male Older Lower More Higher

1·9 (1·2–3·0) 1·8 (1·1–3·2) 2·2 (0·9–5·3) 1·2 (0·9–1·7) 0·9 (0·86–0·95) 1·4 (1·2–1·7) 1·1 (1·0–1·1)

0·004 0·03 0·07 0·17 0·001 0·001 0.02

a

Increments of 10 years (each 10-year increase in age increases the odds of severe disease 1·4-fold). Increments of 1000 units. From Christian TF et al,16 with permission b

28


1·0

Sensitivity

0·8 0·6 0·4 Clinical and exercise thallium Clinical and exercise Clinical

0·2 0·0 1·0

0·8

0·6 0·4 Specificity

0·2

0·0

Figure 3.4 Receiver operator characteristic curves for three logistic regression multivariate models for the prediction of severe coronary disease. (From Christian et al,16 with permission.)

and the model that incorporates clinical parameters and the new test parameters. Methods are available for determining the statistical significance of changes in the area under these two ROC curves.33 An example of this approach is shown in Figure 3.4, taken from Christian et al 16 The clinical significance of these differences in the models (assessed by either

2 analysis or ROC curves) is discussed later. Prediction of patient outcome The demonstration of incremental prognostic value for diagnostic tests is obviously extremely important for clinical decision making. It requires strict adherence to the rigorous standards that were outlined previously. In general, very few of the published studies demonstrating prognostic value of diagnostic tests meet the strict criteria necessary to demonstrate incremental prognostic value for these tests. The statistical model most often used for this purpose is a linear proportional hazards, or Cox, model.34 When strictly applied, all the previous information available to the clinician, either from clinical assessment or previous testing, should be incorporated into a linear proportional hazards model that predicts time to an event. Once again, parameters that have been clearly demonstrated in larger populations to be significant must be “forced” into such models to make sure that their contribution is not neglected. The events in question should preferably be hard end points such as death and myocardial infarction. As previously mentioned, unstable angina and the need for revascularization are alternative end points that are often included to enhance statistical power, but these have major limitations. One of the best examples of a rigorously constructed analysis demonstrating incremental prognostic value was published by Pollock et al in 1992.35 They tested the association between various combinations of variables, and time to death or myocardial infarction, in a linear proportional

hazards model using the 2 statistic. Clinical and exercise variables were significantly better than clinical variables alone. Similarly, a model that added thallium redistribution to clinical and exercise variables was significantly better than the combination of clinical and exercise variables. Another example of such a rigorous analysis was that reported by Christian et al 6 in patients with a normal resting electrocardiogram. Using a similar approach, these investigators reported that a model adding thallium variables to clinical and exercise variables did not add significantly to the model using clinical and exercise variables. Thus, in the subset of patients with a normal resting ECG, Christian et al 16 were unable to confirm the findings of Pollock et al 24

Clinical significance and cost effectiveness Even when statistically significant incremental value has been demonstrated for a diagnostic test using appropriate rigorous methodology, the clinical significance of the findings must be equally rigorously examined. The two fundamental issues that should be addressed are the actual impact of this incremental value on clinical decision making and, where possible, cost effectiveness. The principles of decision analysis pertinent to the first criterion will be presented in much greater detail in Chapter 7. The available published data on diagnostic testing in coronary disease that will be presented here use only rudimentary concepts with respect to decision analysis. Formal cost analysis also requires understanding of a much greater body of published knowledge, which will not be presented here. The examples presented will again be very rudimentary, but demonstrate the principle. Diagnosis of CAD The clinical significance of diagnostic testing can best be understood in terms of decision making thresholds. From the standpoint of diagnosis, a test will be useful primarily if it moves a significant number of patients from an “uncertain” pretest probability to an “acceptably certain” post-test probability. The exact criteria, or threshold, to be used in these classifications are clearly a matter of judgment; many investigators have chosen post-test probabilities of less than 10% and greater than 90% as criteria for definitive diagnosis.36 Thus, non-invasive testing will be useful for diagnosis if it moves a reasonable number of patients into the shaded zone shown in Figure 3.3. Although treadmill testing has clear incremental value for diagnosis, particularly in patients with intermediate pretest probability, as discussed earlier, its ability to move patients across such thresholds of probability appears to be very limited. Goldman et al 37 examined the ability of treadmill 29


Table 3.3 Effect of treadmill exercise test results in moving patients across various diagnostic thresholds Threshold probability

Patients moved across (n)

Correctly moved

Incorrectly moved

Net increase in diagnoses (correct-incorrect)

0·10 0·90 Either 0·10 or 0·90

8 53 61

6 33 39

2 20 22

4 13 17 (5%)

From Goldman et al, 37 by permission of the American Heart Association, Inc.

Non-invasive screening for severe CAD The same threshold approach has been applied to the non-invasive identification of severe CAD. Here the ability of tests to move patients across somewhat different thresholds of probability, as assessed by logistic regression models, has 30

60

% of patients 40 classified with this probability 20

Post-test Pretest

90

85 80 Probability (%)

75

Figure 3.5 Percentage of patients classified with a given probability of coronary disease before and after exercise radionuclide angiography. The prospective study group of 76 patients excluded males of 40 years or older with typical angina. (From Gibbons et al,38 with permission.) 250 200 Patients

exercise variables to classify 329 patients with CAD. Their results are summarized in Table 3.3. The pretest model was very powerful, as it classified 84% of the patients correctly. Table 3.3 shows the number of additional patients classified correctly for given thresholds of probability. For example, if 10% was considered an acceptable threshold to “rule out” CAD, eight of 324 patients were moved across this threshold, but only six were moved correctly. Similarly, for a 90% threshold to “rule in” CAD, 53 patients were moved across this threshold but only 33 were moved across correctly. As a result, the net total number of patients who were correctly moved into the diagnostic zone in Figure 3.3 was only 17, or 5% of the patient population. Thus, the clinical significance of the incremental value provided by the treadmill test appears to be very limited. Similar rigorous analyses have been published for radionuclide angiography.38 The results of one of these are displayed in Figure 3.5. The study group excluded men with typical angina over the age of 40 in order to eliminate most patients with a high pretest probability. Logistic regression models developed on a retrospective population were applied prospectively to a group of 76 patients. As demonstrated in Figure 3.5, eight (11%) of the 76 patients could be classified with 90% certainty on the basis of clinical variables alone. Following radionuclide angiography, 24 patients (32%) could be classified directly. Thus, the incremental value of exercise radionuclide angiography in moving patients across clinically meaningful decision thresholds appeared to be much greater than for the treadmill exercise test, as 21% of the patients were correctly classified by the radionuclide angiogram. Similar findings have been reported for planar thallium imaging.39,40 Unfortunately, no rigorous analyses are available for either SPECT imaging or sestamibi imaging, primarily because post-test selection bias has greatly limited the feasibility of such studies in the current era.

189

Low probability High probability Intermediate probability

238

226 185

135

150

125

100 50 0

37 C I U Clinical

50

48

C I U Clinical and exercise

C I U Clinical, exercise and thallium

Figure 3.6 Correct (C), incorrect (I), and uncertain (U) classification of patients with three vessel or left main coronary artery disease by the use of logistic regression multivariate models. Low, intermediate, and high probability defined using: clinical variables only; clinical and exercise variables; clinical, exercise, and thallium-201 variables. (From Christian et al,16 with permission.)

been tested. Christian et al 16 defined a low-probability group for severe CAD as 0·15, a high-risk group as 0·35, and an intermediate group as 0·15–0·35. These thresholds were chosen to correspond to earlier work from Duke University reporting on the utility of the early positive treadmill test.41 Figure 3.6 shows the results that were


obtained using this approach. Using clinical parameters alone, 189 patients (46% of the study group) were correctly classified as low or high probability. Thirty-seven patients (9%) were incorrectly classified as low or high probability. The remaining 185 patients (45% of the study group) had an intermediate probability, and were therefore in an uncertain category. The addition of exercise parameters correctly classified an additional 37 patients at the expense of 13 additional incorrect classifications, for a net of 24 additional correct classifications (6% of the study group). The addition of thallium parameters led to 12 additional correct classifications, and two fewer incorrect classifications for a net increase of 14 correct classifications (3% of the study group). These workers then used Medicare reimbursement figures to calculate the cost per additional correct classification. For exercise testing, the cost per additional correct patient classification was $1524. For thallium scintigraphy, the cost was $20 550 per additional correct classification. Thus, this analysis demonstrated that the clinical impact was modest, and the cost was high, when thallium imaging was used in patients with a normal resting ECG to try to identify patients non-invasively with severe CAD. Although thallium scintigraphy clearly had statistically significant incremental value, it did not appear to be cost effective for this purpose. Prediction of patient outcome The issues of clinical significance and cost effectiveness are particularly pertinent to the application of diagnostic tests for the prediction of the patient outcome. These applications often involve relatively low-risk patient groups with few subsequent events. Tests applied to the entire population may identify a subset of patients who are at considerably increased risk.42,43 These results will be highly statistically significant, and generate very impressive P values and risk ratios. However, it must be recognized that the absolute rate of events often remains too low in the high-risk patient subgroup to be clinically meaningful, and the cost of this identification is often therefore prohibitive when viewed on a per event basis. This concept was nicely demonstrated in a study by Berman et al 44 on patients with a low clinical likelihood of CAD studied by SPECT sestamibi. During 20 months of follow up, only patients with an abnormal sestamibi study suffered death or myocardial infarction. This difference was statistically highly significant (P 0 · 007). However, this increment in prognostic value was clearly not cost effective, as noted by the authors. Although the cost analysis by the authors was quite detailed, the cost ineffectiveness of this approach is readily apparent with very simple analysis. The 107 patients in the high-risk group suffered only three events during 20 months of follow up. In order to identify the high-risk group, testing was required of 548 patients.

Using a Medicare reimbursement figure of $700 per test,16 more than $383 000 of testing would be required to identify the high-risk cohort. The cost of testing alone would therefore exceed $127 000 per possible event prevented. This simple analysis ignores the additional costs that would accrue from the subsequent cardiac catheterizations and coronary revascularizations that would be necessary in the high-risk group in order to attempt to prevent the three events. (There is obviously no certainty that the three events could actually be prevented by revascularization.) Similar analyses have been published for screening in asymptomatic individuals. As a general principle, it should be recognized that non-invasive testing for the assessment of prognosis is far less cost effective in subsets of patients at intrinsically low risk.

Conclusion Clinicians should recognize that an evidence-based approach to the evaluation of the incremental value of diagnostic tests is not simple or straightforward. Unfortunately, it is far easier for both clinicians and investigators to use simple, less rigorous, approaches that appear to demonstrate important incremental value for each new diagnostic test. Although convenient, such approaches lead to incorrect conclusions, and generally overestimate the added value of each new testing modality. The examples presented in this chapter should provide a framework for thinking clinicians to evaluate better new publications on new diagnostic tests. However difficult these analyses may be, and however disappointing the results, the escalating costs of healthcare demand an approach of rigorous methodology and thoughtful analysis to make certain that the incremental value of a diagnostic test is not only statistically significant, but clinically significant and cost effective.

References 1.Diamond G. Penny wise. Am J Cardiol 1988;62:806–8. 2.Bobbio M, Pollock BH, Cohen I, Diamond GA. Comparative accuracy of clinical tests for diagnosis and prognosis of coronary artery disease. Am J Cardiol 1988;62:896–900. 3.Ladenheim ML, Kotler TS, Pollock BH, Berman DS, Diamond GA. Incremental prognostic power of clinical history, exercise electrocardiography and myocardial perfusion scintigraphy in suspected coronary artery disease. Am J Cardiol 1987;59: 270–7. 4.Diamond GA, Forrester JS. Analysis of probability as an aid in the clinical diagnosis of coronary artery disease. N Engl J Med 1979;300:1350–8. 5.Diamond GA. Letter: a clinical relevant classification of chest discomfort. J Am Coll Cardiol 1983;1:574–5.

31


6.Pryor DB, Harrell FE Jr, Lee KL, Califf RM, Rosati RA. Estimating the likelihood of significant coronary artery disease. Am J Med 1983;75:771–80. 7.Weiner DA, McCabe CH, Ryan TJ. Identification of patients with left main and three vessel coronary disease with clinical and exercise test variables. Am J Cardiol 1980;46:21–7. 8.Hubbard BL, Gibbons RJ, Lapeyre AC, Zinsmeister AR, Clements IP. Identification of severe coronary artery disease using simple clinical parameters. Arch Intern Med 1992;152: 309–12. 9.Pryor DB, Shaw L, Harrell FE Jr et al. Estimating the likelihood of severe coronary artery disease. Am J Med 1991;90: 553–62. 10.Pryor DB, Shaw L, McCants CB et al. Value of the history and physical in identifying patients at increased risk for coronary artery disease. Ann Intern Med 1993;118:81–90. 11.Weiner DA, Ryan TJ, McCabe CH et al. The role of exercise testing in identifying patients with improved survival after coronary artery bypass surgery. J Am Coll Cardiol 1986;8: 741–8. 12.O’Keefe JH Jr, Zinsmeister AR, Gibbons RJ. Value of electrocardiographic findings in predicting rest left ventricular function in patients with chest pain and suspected coronary artery disease. Am J Med 1989;86:658–62. 13.Rihal CS, Davis KB, Kennedy JW, Gersh BJ. The utility of clinical, electrocardiographic, and roentgenographic variables in the prediction of left ventricular function. Am J Cardiol 1995;75:220–3. 14.Connolly DC, Elveback LR, Oxman HA. Coronary heart disease in residents of Rochester, Minnesota. IV. Prognostic value of the resting electrocardiogram at the time of initial diagnosis of angina pectoris. Mayo Clin Proc 1984;59:247–50. 15.Gibbons RJ, Zinsmeister AR, Miller TD, Clements IP. Supine exercise electrocardiography compared with exercise radionuclide angiography in noninvasive identification of severe coronary artery disease. Ann Intern Med 1990;112:743–9. 16.Christian TF, Miller TD, Bailey KR, Gibbons RJ. Exercise tomographic thallium-201 imaging in patients with severe coronary artery disease and normal electrocardiogram. Ann Intern Med 1994;121:825–32. 17.Epstein SE. Implications of probability analysis on the strategy used for noninvasive detection of coronary artery disease. Am J Cardiol 1980;46:491–9. 18.Rifkin RD, Hood WB Jr. Bayesian analysis of electrocardiographic exercise stress testing. N Engl J Med 1977;297: 681–6. 19.Cohn K, Kamm B, Feteih N, Brand R, Goldschlager N. Use of treadmill score to quantify ischemic response and predict extent of coronary disease. Circulation 1979;59:286–96. 20.Currie PJ, Kelly MJ, Harper RW et al. Incremental value of clinical assessment, supine exercise electrocardiography, and biplane exercise radionuclide ventriculography in the prediction of coronary artery disease in men with chest pain. Am J Cardiol 1983;52:927–35. 21.Morise AP, Detrano R, Bobbio M, Diamond GA. Development and validation of a logistic regression-derived algorithm for estimating the incremental probability of coronary artery disease before and after exercise testing. J Am Coll Cardiol 1992; 20:1187–96. 22.Ransohoff DF, Feinstein AR. Problems of spectrum and bias in evaluating the efficacy of diagnostic tests. N Engl J Med 1978; 299:926–30.

32

23.Philbrick JT, Horwitz RI, Feinstein AR. Methodologic problems of exercise testing for coronary artery disease: groups, analysis and bias. Am J Cardiol 1980;46:807–12. 24.Rozanski A, Diamond GA, Berman D, Forrester JS, Morris D, Swan HJC. The declining specificity of exercise radionuclide ventriculography. N Engl J Med 1983; 309:518–22. 25.Council on Health Care Technology, Institute of Medicine. Assessment of diagnostic technology in health care. Washington, DC: National Academy Press,1989. 26.Jaeschke R, Guyatt G, Sackett DL. Users’ guides to the medical literature. JAMA 1994; 271:389–91. 27.Diamond GA. An alternative factor affecting sensitivity and specificity of exercise electrocardiology (editorial). Am J Cardiol 1986; 57:1175–80. 28.Begg CB, Greenes RA. Assessment of diagnostic tests when disease verification is subject to selection bias. Biometrics 1983;39:207–15. 29.Morise AP, Diamond GA. Comparison of the sensitivity and specificity of exercise electrocardiography in biased and unbiased populations of men and women. Am Heart J 1995; 130:741–7. 30.Roger VL, Pellikka PA, Bell MR, Chow CWH, Bailey KR, Seward JB. Sex and test verification bias: Impact on the Diagnostic Value of Exercise Echocardiology. Circulation 1997; 95:405–10. 31.Miller TD, Hodge DO, Christian TF, Milavetz JJ, Bailey KR, Gibbons RJ. Effects on adjustment for referral bias on the sensitivity of specificity of single photon emission computed tomography for the diagnosis of coronary artery disease. Am J Med 2002;112:290–7. 32.Gibbons RJ, Chatterjee K, Daley J, et al. ACC/AHA/ACP-ASIM guidelines for the management of patients with chronic stable angina: a report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Chronic Stable Angina). J Am Coll Cardiol 1999;33:2092–7. 33.Wieand S, Gail M, James K, James B. A family of nonparametric statistics for comparing diagnostic tests with paired or unpaired data. Biometrika 1989;76:585–92. 34.Cox DR. Regression models and life tables. J R Stat Soc B 1972;34:197–220. 35.Pollock SG, Abbott RD, Boucher CA, Beller GA, Kaul S. Independent and incremental prognostic value of tests performed in hierarchical order to evaluate patients with suspected coronary artery disease. Validation of models based on these tests. Circulation 1992;85:237–48. 36.Diamond GA, Forrester JS, Hirsch M et al. Application of conditional probability analysis to the clinical diagnosis of coronary artery disease. J Clin Invest 1980;65:1210–21. 37.Goldman L, Cook EF, Mitchell N et al. Incremental value of the exercise test for diagnosing the presence or absence of coronary artery disease. Circulation 1982;66:945–53. 38.Gibbons RJ, Lee KL, Pryor DB et al. The use of radionuclide angiography in the diagnosis of coronary artery disease: a logistic regression analysis. Circulation 1983;68:740–6. 39.Detrano R, Yiannikas J, Salcedo EE et al. Bayesian probability analysis: a prospective demonstration of its clinical utility in diagnosing coronary disease. Circulation 1984;69: 541–7.


40.Melin JA, Wijns W, Vanbutsele RJ et al. Alternative diagnostic strategies for coronary artery disease in women: demonstration of the usefulness and efficiency of probability analysis. Circulation 1985;71:535–42. 41.McNeer JF, Margolis JR, Lee KL et al. The role of the exercise test in the evaluation of patients for ischemic heart disease. Circulation 1978;57:64–70. 42.Rautaharju PM, Prineas RJ, Eifler WJ et al. Prognostic value of exercise electrocardiogram in men at high risk of future coronary heart disease: multiple risk factor intervention trial experience. J Am Coll Cardiol 1986;8:1–10.

43.Giagnoni E, Secchi MB, Wu SC et al. Prognostic value of exercise EKG testing in asymptomatic normotensive subjects: a prospective matched study. N Engl J Med 1983;309: 1085–9. 44.Berman DS, Hachamovitch R, Hosen K et al. Incremental value of prognostic testing in patients with known or suspected ischemic heart disease: a basis for optimal utilization of exercise technetium-99 m sestamibi myocardial perfusion single-photon emission computed tomography. J Am Coll Cardiol 1995;26:639–47.

33

4

Clinical trials and meta-analysis Colin Baigent

Introduction Although large effects on survival arising from certain treatments may occasionally be obvious from simple observation (as, for example, when cardioversion for ventricular fibrillation avoids otherwise certain death), the vast majority of interventions have only moderate effects on major outcomes and hence are impossible to evaluate without careful study. Enthusiasm for the biologic foundations of a particular therapeutic approach often leads to exaggerated hopes for the effects of treatment on major clinical outcomes. These hopes may be based on dramatic laboratory measures of efficacy, or on the types of surrogate outcome that are commonly studied before drugs go into Phase III or IV studies: for example, a drug may almost completely prevent experimental ischemia progressing to infarction, or practically abolish experimental thrombosis. However, these large effects on surrogate end points very rarely translate into large effects on major clinical outcomes: the overwhelming message from two decades of clinical trials in cardiology is that the net effects of most treatments are typically moderate in size.* This chapter explains why largescale randomized evidence, either in a single “mega-trial” or in a meta-analysis of similar trials, is generally an absolute requirement if such moderate effects on major outcomes are to be characterized reliably. It is important to appreciate that progress in cardiologic practice, and in the prevention of cardiovascular disease, has been and remains dependent on the availability of largescale randomized trials and appropriately large-scale metaanalyses of such trials. In the management of acute myocardial infarction (MI), for example, these methods have helped to demonstrate that fibrinolytic therapy,1–3 aspirin,1,3,4

* For rare adverse effects, however, there may be large proportional differences between one treatment and another, or between treatment and control. For example, some non-steroidal antiinflammatory drugs may substantially increase the risk of gastrointestinal bleeding. Rare adverse effects with extreme relative risks can often be recognized reliably by careful clinical observation, or by other non-randomized methods, and such relative risks are sometimes best quantified in case–control or cohort studies.

34

angiotensin-converting-enzyme (ACE) inhibitors5–7 and blockers8 all produce net benefits which, although individually moderate in size, have together produced a substantial improvement in the prognosis of acute MI. Similarly, the demonstration that ACE inhibitors produce moderate reductions in the risk of death and in the rates of hospitalization for worsening heart failure,9 and that the addition of digoxin further reduces the need for recurrent hospitalization,10 has improved the prognosis of congestive heart failure.

Clinical trials: minimizing biases and random errors Any clinical study whose main objective is to assess moderate treatment effects must ensure that any biases and any random errors that are inherent in its design are both substantially smaller than the effect to be measured.11,12 Biases in the assessment of treatment can be produced by differences in factors other than the treatment under consideration. Observational (that is non-randomized) studies in which the outcome is compared between individuals who received the treatment of interest and those who did not, can be subject to large biases.13 Instead, the guaranteed avoidance of biases requires the proper randomized allocation of treatment and appropriate statistical analysis, with no unduly data-dependent emphasis on specific subsets of the overall evidence (Table 4.1).12 Avoidance of moderate biases Proper randomization The fundamental reason for random allocation of treatment in clinical trials is to maximize the likelihood that each type of patient will have been allocated in similar proportions to the different treatment strategies being investigated.14 Proper randomization requires that trial procedures are organized in a way that ensures that the decision to enter a patient is made irreversibly and without knowledge of the trial treatment to which a patient will be allocated. In situations where the next treatment allocation can be deduced by those entering patients, decisions about whether to enter a

Clinical trials and meta-analysis

Table 4.1 Requirements for reliable assessment of MODERATE treatment effects11,12 1. Negligible biases (that is guaranteed avoidance of MODERATE biases) ● Proper RANDOMIZATION (non-randomized methods cannot guarantee the avoidance of moderate biases) ● Analysis by ALLOCATED treatments (that is an “intention to treat” analysis) ● Chief emphasis on OVERALL results (with no unduly data-derived subgroup analysis) ● Systematic META-ANALYSIS of all the relevant randomized trials (with no unduly data-dependent emphasis on the results from particular studies) 2. Small random errors (that is guaranteed avoidance of MODERATE random errors) ● LARGE NUMBERS (with minimal data collection as detailed statistical analyses of masses of data on prognostic features generally add little to the effective size of a trial) ● Systematic META-ANALYSIS of all the relevant randomized trials

particular patient may be affected, and those allocated one treatment might then differ systematically from those allocated another.15 In the Captopril Prevention Project (CAPPP) trial,16 for example, envelopes containing the antihypertensive treatment allocation could be opened before patients were irreversibly entered into the study. Highly significant differences in pre-entry blood pressure between the treatment groups, which were too large to have been due to chance, may well have been the result of this design weakness.17 Intention to treat analysis Even when studies have been properly randomized and well conducted, moderate biases can still be introduced by inappropriate analysis or interpretation. One well recognized circumstance is when patients are excluded after randomization, particularly when the prognosis of the excluded patients in one treatment group differs from that in the other (such as might occur, for example, if non-compliers were excluded after randomization). This point is well illustrated by the Coronary Drug Project, which compared clofibrate versus placebo among around 5000 patients with a history of coronary heart disease. In this study, patients who took at least 80% of their allocated clofibrate (“good” compliers) had substantially lower 5 year mortality than “poor” compliers who did not (15·0% v 24·6% respectively; P 0·0001). However, there was a similar difference in outcome between “good” and “poor” compliers in the placebo group (15·1% v 28·3%, respectively; P 0·00001),

suggesting that “good” and “poor” compliers were prognostically different even after allowing for any benefits of actually taking clofibrate.18 If there is really no difference in outcome between two treatments, then the least biased assessment of the treatment effect is that which compares all those allocated to one treatment versus all those allocated to the other (that is an “intention to treat” analysis), irrespective of what treatment they actually received.19 Because some degree of non-compliance with allocated treatments is unavoidable in randomized trials, intention to treat analyses will obviously underestimate the effects produced by full compliance. However, “on treatment” analyses, which compare effects among compliant patients with those in non-compliant patients, are potentially biased, and it is more appropriate to calculate an “adjustment” based on the level of compliance and then to apply this to the estimate of the treatment effect provided by the intention to treat comparison.20 For example, in a meta-analysis of the randomized trials of prolonged use of antiplatelet therapy among patients with occlusive vascular disease, the average compliance 1 year after treatment allocation seemed to be around 80%.4 Application of this estimate of compliance to the proportional reduction of about 30% in non-fatal MI and stroke estimated from intention to treat analyses of these trials suggests that full compliance with antiplatelet therapy produces reductions in risk of about 35–40%. Dangers of data-dependent emphasis on particular results In the medical literature a particularly important source of bias is unduly data-dependent emphasis on particular trials or on particular subgroups of patients. Such emphasis is often entirely inadvertent, arising from a perfectly reasonable desire to understand the randomized trial results in terms of who to treat, which treatments to prefer, or disease mechanisms. However, whatever its origins, selective emphasis on particular parts of the evidence can often lead to seriously misleading conclusions. This is because the identification of categories of patients for whom treatment is particularly effective (or ineffective) requires surprisingly large quantities of data. Even if the real sizes of the treatment effects do vary substantially among subgroups of patients, subgroup analyses are so statistically insensitive that they may well fail to demonstrate these differences. On the other hand, if the real proportional risk reductions are about the same for everybody, subgroup analyses can vary so widely just by the play of chance that the results in selected subgroups may be exaggerated. Even when highly significant “interactions” are found, they may be a poor guide to the sizes (or even the directions) of any genuine differences, as the more extreme such results may still owe more to chance than to reality. This is particularly the case when such interactions have emerged after an overzealous examination of multiple 35


subgroups. For example, in the large Second International Study of Infarct Survival (ISIS-2), the 1 month survival advantage produced by aspirin was particularly clear (804 vascular deaths among 8587 patients allocated aspirin v 1016 among 8600 allocated placebo; proportional reduction of 23% (SD 4); P 0·000001).1 When these overall results were subdivided by the patients’ astrological birth signs, however, no fewer deaths were observed with aspirin than with placebo among patients born under Libra or Gemini (Table 4.2). Although few doctors would consider such analyses to be valid, similarly unreliable conclusions based on “exploratory” data-derived subgroup analyses are widely reported in medical journals and at scientific meetings, and may well have adverse consequences for patient care. An example of how such subgroup analyses resulted in inappropriate management of patients is provided by the early trials of aspirin for the secondary prevention of stroke. Here, emphasis on the results in men led to a situation where, for almost 20 years, the US Food and Drug Administration approved this use of aspirin only for males; more recent evidence shows this to have been mistaken.4 A further example is provided by the large Italian GISSI-1 trial comparing streptokinase versus control after acute MI. The overall results favoured streptokinase, but subgroup analyses suggested that streptokinase was beneficial only in patients without prior MI. Fortunately, the GISSI investigators were circumspect about this “finding”,2 and their caution turned out to have been wise, as a subsequent overview of all the large fibrinolytic trials showed that the proportional benefits were similar, irrespective of a history of MI.21 Many thousands of patients with a previous history of MI might well have been denied fibrinolytic therapy, however, if the apparent pattern of the results in the GISSI-1 subgroups had been believed. A similar bias may arise in a situation where several studies have addressed much the same therapeutic question but only a few of them are chosen for emphasis. This could be a source of serious bias, as chance fluctuations for or against treatment might affect this choice. It is therefore more appropriate to base inference on a meta-analysis of results from all relevant randomized trials (or, at least, on an unbiased subset of the relevant trials, such as all trials above a certain minimum sample size).22,23 One additional advantage of such an

approach is that such meta-analyses will also minimize random errors, because far more patients (and most importantly, more events) will be available for analysis. The separate trials might well be heterogeneous, but with careful interpretation of such heterogeneity it is often possible to enhance understanding of particular clinical questions.24 Occasionally, when detailed information on individual patients is available within a really large meta-analysis that includes several thousand major outcomes, such as death21 or cancer recurrence,25 it may be feasible to identify particular groups of individuals in whom the benefits or hazards of treatment really are especially great. (Where it has been possible to establish cooperation between trialists before any of the trial results are known, having just a few prespecified subgroup hypotheses can provide some protection against unduly data-dependent emphasis on particular results in a large meta-analysis.26) Avoidance of moderate random errors Small trials may produce false negative results Whereas the avoidance of moderate biases requires careful attention both to the randomization process and to the analysis and interpretation of the available trial evidence, the avoidance of moderate random errors requires large numbers of events. Because major outcomes such as death may affect only a small proportion of those randomized, very large numbers of patients often need to be studied before the results can be guaranteed to be statistically (and hence medically) convincing. For example, the early trials of intravenous fibrinolytic therapy for acute myocardial infarction were individually too small to provide reliable evidence about any moderate effects of this treatment on mortality, although several did identify an increased risk of serious bleeding. As a result, fibrinolytic therapy was not used routinely until the GISSI-12 and ISIS-21 “mega-trials” provided such definite evidence of benefit that treatment patterns changed rapidly.27 It is worth noting, however, that GISSI-1 and ISIS-2 both included more than 10 000 patients and 1000 deaths, but had they only been one tenth as large the observed reduction in mortality would not have been conventionally significant, and would therefore have had much less influence on medical practice.

Table 4.2 Unreliability of “data-dependent” subgroup analyses: ISIS-2 trial of aspirin among over 17 000 patients with suspected acute myocardial infarction1 Astrological birth sign

Vascular death by 1 month Aspirin

36

P value

Placebo

Libra or Gemini All other signs

150 (11·1%) 654 (9·0%)

147 (10·2%) 869 (12·1%)

0·5 0·0001

Any birth sign

804 (9·4%)

1016 (11·8%)

0·0001


Small-scale meta-analyses may be unreliable Because meta-analyses are appearing in medical journals with increasing frequency it is useful to be able to judge the reliability of such reviews – and, in particular, the extent to which confounding, biases or random errors could lead to mistaken conclusions. (In randomized trials, “confounding” exists when a comparison of some particular treatment in one group versus a control group involves the routine coadministration in one group, but not the other, of some cointervention that might affect the outcome.) To avoid any possibility of confounding, and to avoid any flexibility in the question of which trials to consider, meta-analyses should generally include only unconfounded properly randomized trials. The main problems that then remain are those of biases and random errors. Two types of bias could affect the reliability of a metaanalysis: those that occur within individual trials, and those that relate to the selection of trials. More empirical research into the numerous biases that can occur within randomized trials would be valuable. However, it is clear from existing studies that, for example, inadequate concealment of the likely treatment allocation does quite often result in exaggerated estimates of treatment effect,28 and that the inappropriate postrandomization exclusion of particular patients is common.29 Such defects have unpredictable consequences for particular trials, however, and no generalizations about the likely size, or even direction, of the resultant biases are possible. A further problem involves the process of identifying all relevant trials. Unfortunately, the subset of trials that are eventually published (and hence which are conveniently available) is often a biased sample of the trials that have been done. Trials may well be more likely to be submitted for publication if their results are strikingly positive than if they are negative or null.30–33 Such “publication bias” can, along with other sources of bias, produce surprisingly impressive looking evidence of effectiveness for treatments that are actually useless.34 The particular circumstances in which publication bias has contributed to producing misleading estimates of treatment are difficult to identify, and it is still more difficult to generalize about the exact size of any such bias when it does occur. The problem of incomplete ascertainment is likely to be particularly acute within small meta-analyses that contain no more than a few hundred major outcomes and which consist mainly of small published trials. This is because results from trials with only a limited number of end points are subject to large random errors, and such trials are therefore particularly likely to generate implausibly large effect estimates. If publication bias then results in emphasis on the more promising of these small trial results, the resulting summary odds ratios are likely to be unreliable.35 Hence, unless the particular circumstances of a small-scale meta-analysis suggest that publication bias is unlikely, it may be best to treat such results as no more

than “hypothesis generating”. On the other hand, a thoroughly conducted meta-analysis that in aggregate contains sufficient numbers of major outcomes to constitute “largescale” randomized evidence4,21,25 is unlikely to be materially affected by publication bias and, provided there are no serious uncontrolled biases (see above) within the individual component trials, is likely to be fairly trustworthy – although, even then, inappropriate subgroup analyses may generate mistaken conclusions.

Large-Scale Randomized Trials Trials of the effects of treatments on major outcomes can only be made large if they are kept as simple as possible. In particular, as many as possible of the main barriers to rapid recruitment need to be removed. An important way in which trial design can facilitate this is to limit the amount of information that is recorded. For example, data recorded at baseline can often be restricted to important clinical details, including at most only a few major prognostic factors and only a few variables that are thought likely to influence substantially the benefits or hazards of treatment. Similarly, the information recorded at follow up need not be extensive and can be limited largely to those major outcomes that such studies have been designed to assess, and to approximate measures of compliance. (Other outcomes that are of interest but which do not need to be studied on such a large scale may best be assessed in separate smaller studies, or in subsets of these large studies when this is practicable.) Likewise, complicated eligibility criteria, inappropriately detailed consent procedures36 and unnecessarily extensive auditing of data can all prevent the recruitment of large numbers of patients. Furthermore, if trials are complex they are likely to involve a high cost per patient, which again tends to limit their size. Either way, complexity is rarely a virtue in trials designed to assess major outcomes, whereas simplicity can sometimes lead to the rapid randomization of very large numbers of patients, and to results that change clinical practice within very short periods of time.1,27

The “uncertainty principle” For ethical reasons, randomization is appropriate only if both the doctor and the patient feel substantially uncertain as to which trial treatment is best. The “uncertainty principle” maximizes the potential for recruitment within this ethical constraint (see Box on p 38). If many hospitals are collaborating in a trial then wholehearted use of the uncertainty principle encourages clinically appropriate heterogeneity in the resulting trial population, and in large trials this may add substantially to the practical value of the results. Among the early trials of

37


The “uncertainty principle” A patient can be entered if, and only if, the responsible physician is substantially uncertain as to which of the trial treatments would be most appropriate for that particular patient. A patient should not be entered if the responsible physician or the patient is, for any medical or non-medical reason, reasonably certain that one of the treatments that might be allocated would be inappropriate for this particular individual (in comparison either with no treatment or with some other treatment that could be offered to the patient in or outside the trial).37

fibrinolytic therapy, for example, most of the studies had restrictive entry criteria that precluded the randomization of elderly patients, and so those trials contributed nothing of direct relevance to the important clinical question of whether treatment was useful in older patients. Other trials that did not impose an upper age limit, however, did include some elderly patients, and were therefore able to show that age alone is not a contraindication to fibrinolytic therapy.21 Thus, homogeneity of those randomized may be a serious defect in clinical trial design, whereas heterogeneity may be a scientific strength: after all, trials do need to be relevant to a very heterogeneous collection of future patients. The “uncertainty principle” not only ensures ethicality and clinically useful heterogeneity, but also is easily understood and remembered by busy collaborating clinicians, which in turn helps the randomization of large numbers of patients.

Can observational studies substitute for large-scale randomized trials? As the resources will never be available to design large, simple trials to address all the questions of clinical interest, there have been several recent suggestions that observational studies might be able to provide reliable estimates of the effects of particular treatments. Non-randomized studies do not necessarily provide inaccurate estimates of the effects of treatments, but the point is that they cannot be guaranteed to produce reliable estimates because of biases that are inherent in their design. It may well be difficult or impossible to avoid such biases, or to adjust fully for their effects.38 When non-randomized studies suggest that certain treatments have surprisingly large effects, such findings are often refuted when those treatments are assessed in large randomized trials.39 For example, the claims of hazards with digoxin in heart failure,40 based on non-randomized evidence, were not confirmed by the very large randomized DIG (Digitalis Investigation Group) trial.10 Even if non-randomized comparisons happen to get the right answer then nobody will really know that they have done so. Thus nonrandomized studies are of little practical value if the primary aim is to assess moderate treatment effects (whether beneficial or adverse) on major outcomes. 38

Summary Many interventions in cardiological practice produce only moderate effects on major outcomes such as death or serious disability. However, even a moderate effect of treatment, if demonstrated clearly enough for that treatment to be widely adopted, can prevent disabling events or death in substantial numbers of people. Moreover, if – as in the treatment of acute myocardial infarction – more than one moderately effective treatment can eventually be identified, then the combination of two or three individually moderate improvements in outcome may collectively result in substantial health gains. In some instances sufficient information is already available from large-scale randomized trials – or, better still, from meta-analyses of those trials – to allow the balance of risk and benefit of particular treatments to be defined for particular patients. But many important questions have still not been answered reliably, and there remains a need for many more large “streamlined” megatrials, and meta-analyses of such trials, to help resolve some of the outstanding clinical uncertainties in the management of cardiovascular disease. References 1.ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction: ISIS-2. Lancet 1988; ii:349–60. 2.GISSI (Gruppo Italiano per lo Studio della Streptochinasi nell’infarto miocardico). Effectiveness of intravenous thrombolytic treatment in acute myocardial infarction. Lancet 1986;i:397–402. 3.Collins R, Peto R, Baigent C, Sleight P. Aspirin, heparin, and fibrinolytic therapy in suspected acute myocardial infarction. N Engl J Med 1997;336:847–60. 4.Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy. I: Prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. BMJ 1994; 308:81–106. 5.ISIS-4 (Fourth International Study of Infarct Survival) Collaborative Group. ISIS-4: A randomised factorial trial assessing early oral captopril, oral mononitrate, and intravenous magnesium sulphate in 58 050 patients with suspected acute myocardial infarction. Lancet 1995;345:669–85. 6.Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico. GISSI-3: effects of lisinopril and transdermal glyceryl trinitrate singly and together on 6-week mortality and ventricular function after myocardial infarction. Lancet 1994; 343:1115–22. 7.Chinese Cardiac Study Collaborative Group. Oral captopril versus placebo among 13,634 patients with suspected acute myocardial infarction: interim report from the Chinese Cardiac Study (CCS-1). Lancet 1995;345:686–7.


8.ISIS-1 (First International Study of Infarct Survival) Collaborative Group. Randomised trial of intravenous atenolol among 16,027 cases of suspected acute myocardial infarction: ISIS-1. Lancet 1986;ii:57–66. 9.The SOLVD Investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med 1991;325:293–302. 10.The Digitalis Investigation Group. The effect of digoxin on mortality and morbidity in patients with heart failure. N Engl J Med 1997;336:525–33. 11.Collins R, Peto R, Gray R, Parish S. Large-scale randomized evidence: trials and overviews. In: Weatherall D, Ledingham JGG, Warrell DA, eds. Oxford Textbook of Medicine, Vol. 1 Oxford: Oxford University Press, 1996. 12.Collins R, MacMahon S. Reliable assessment of the effects of treatment on mortality and major morbidity, I: clinical trials. Lancet 2001;357:373–80. 13.MacMahon S, Collins R. Reliable assessment of the effects of treatment on mortality and major morbidity, II: observational studies. Lancet 2001;357:455–62. 14.Armitage P. The role of randomization in clinical trials. Stat Med 1982;1:345–52. 15.Kunz R, Oxman AD. The unpredictability paradox: review of empirical comparisons of randomised and non-randomised clinical trials. BMJ 1998;317:1185–90. 16.Hansson L, Lindholm LH, Niskanen L et al. for the Captopril Prevention Project (CAPPP) study group. Effect of angiotensinconverting-enzyme inhibition compared with conventional therapy on cardiovascular morbidity and mortality in hypertension: the Captopril Prevention Project (CAPPP) randomised trial. Lancet 1999;353:611–16. 17.Peto R. Failure of randomisation by “sealed” envelope. Lancet 1999;354:73. 18.The Coronary Drug Project Research Group. Influence of adherence to treatment and response of cholesterol on mortality in the Coronary Drug Project. N Engl J Med 1980;303: 1038–41. 19.Peto R, Pike MC, Armitage P et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. Part I: Introduction and design. Br J Cancer 1976;34:585–612. 20.Cuzick J, Edwards R, Segnan N. Adjusting for non-compliance and contamination in randomized clinical trials. Stat Med 1997;16:1017–29. 21.Fibrinolytic Therapy Trialists’ Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet 1994;343:311–22. 22.Collins R, Gray R, Godwin J, Peto R. Avoidance of large biases and large random errors in the assessment of moderate treatment effects: the need for systematic overviews. Stat Med 1987;6:245–50.

23.Clarke M, Chalmers I. Discussion sections in reports of controlled trials published in general medical journals: islands in search of continents? JAMA 1998;280:280–2. 24.Thompson SG. Why sources of heterogeneity in meta-analysis should be investigated. BMJ 1994;309:1351–5. 25.Early Breast Cancer Trialists’ Collaborative Group. Systemic treatment of early breast cancer by hormonal, cytotoxic, or immune therapy: 133 randomised trials involving 31 000 recurrences and 24 000 deaths among 75 000 women. Lancet 1992;339:1–15 (Part I) & 71–85 (Part II). 26.Cholesterol Treatment Trialists’ (CTT) Collaboration. Protocol for a prospective collaborative overview of all current and planned randomized trials of cholesterol treatment regimens. Am J Cardiol 1995;75:1130–4. 27.Collins R, Julian D. British Heart Foundation surveys (1987 and 1989) of United Kingdom treatment policies for acute myocardial infarction. Br Heart J 1991;66:250–5. 28.Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias: dimensions of methodologic quality associated with estimates of treatment effects in controlled trials. JAMA 1995;273:408–12. 29.Schulz KF, Grimes DA, Altman DG, Hayes RJ. Blinding and exclusions after allocation in randomized controlled trials: survey of published parallel group trials in obstetrics and gynaecology. BMJ 1996;312:742–4. 30.Dickersin K, Chan S, Chalmers TC et al. Publication bias and clinical trials. Contr Clin Trials 1987;8:343–53. 31.Easterbrook PJ, Berlin JA, Gopelan R, Matthews DR. Publication bias in clinical research. Lancet 1991;337:867–72. 32.Dickersin K, Min Y-I, Meinert CL. Factors influencing publication of research results. Follow-up of applications submitted to two institutional review boards. JAMA 1992;267:374–8. 33.Dickersin K, Min Y-I. Publication bias: the problem that won’t go away. Ann NY Acad Sci 1993;703:135–46. 34.Counsell CE, Clarke MJ, Slattery J, Sandercock PAG. The miracle of DICE therapy for acute stroke: fact or fictional product of subgroup analysis. BMJ 1994;309:1677–81. 35.Davey Smith G, Egger M. Misleading meta-analysis. BMJ 1995;310:742–54. 36.Doyal L. Journals should not publish research to which patients have not given fully informed consent – with three exceptions. BMJ 1997;314:1107–11. 37.Collins R, Doll R, Peto R. Ethics of clinical trials. In: Williams CJ, ed. Introducing New Treatments for Cancer: Practical, Ethical and Legal Problems. Chichester: John Wiley & Sons, 1992. 38.Sheldon TA. Please bypass the PORT. Observational studies of effectiveness run a poor second to randomised controlled trials. BMJ 1994;309:142–3. 39.Peto R. Clinical trial methodology. Biomedicine Special Issue 1978;28:24–36. 40.Yusuf S, Wittes J, Bailey K, Furberg C. Digitalis – a new controversy regarding an old drug: the pitfalls of inappropriate methods. Circulation 1986;73:13–18.

39

5

Finding current best evidence to practice evidence-based cardiology Dereck L Hunt, K Ann McKibbon, R Brian Haynes

Staying current with new diagnostic tests, treatments, and other clinically useful new knowledge in a rapidly evolving field such as cardiology requires effort. Fortunately, this once daunting task is becoming more feasible because of new evidence-based information resources and the steady advance of information technology into clinical settings. This chapter will review ways to find current best evidence for the care of patients with cardiac problems, including both solving patient problems as they arise, and keeping up with new evidence that is ready for application in clinical practice. The patients whom we see on a daily basis provide an excellent stimulus to staying current. They may have clinical problems that we are unfamiliar with, or that we have not recently reviewed. They may also present us with information from the media or friends to evaluate, or ask us questions that we need to research before answering. Depending on the type of center in which we work, our colleagues, teachers, and students may also ask questions or provide suggestions, making us realize that our knowledge may be “time-challenged”. To become proficient in responding to such challenges (also known as “learning opportunities”), we can make use of a growing array of specialized information resources, aided by information technology that can bring access to our fingertips, almost wherever we may be. To illustrate how patient contacts can provide us with the stimulus to keep up to date and be aware of new evidence, consider the following scenarios. 1.

40

During your outpatient clinic, you see a 56 year old woman who recently became your patient after she moved to your community. She has been diagnosed as having an idiopathic dilated cardiomyopathy and had an echocardiogram 6 months ago that revealed a diffusely enlarged left ventricle with no segmental abnormalities. The ejection fraction was estimated to be less than 30%. You review her current condition and note that her symptoms are controlled on an angiotensin converting enzyme (ACE) inhibitor, digoxin, and a diuretic. Still, she complains of fatigue and dyspnea on moderate exertion. Before leaving, she asks if you can recommend any other medications that would help.

2.

Later that day, you pass through the emergency department where an emergency physician happens to notice you. She has been working up a 65 year old man who presented with a swollen left calf. He had an ultrasound that confirmed the presence of a deep venous thrombosis. The patient is anxious to return home because his wife is ill and requires care. The emergency physician is interested in your opinion on the use of low molecular weight heparin for the treatment of deep venous thrombosis in outpatients.

These questions are consistent with the common information needs of physicians. For internists, questions arise at a rate of two questions for every three outpatients seen1 and five questions on average for every inpatient.2 No one knows whether cardiologists confront similar numbers of questions, but no professional discipline studied to date is immune from the need to address unanswered questions to keep up with the advance of knowledge. How would you address each of the questions raised by the clinical scenarios? The possibilities include using an electronic bibliographic database such as MEDLINE, a specialized abstract journal like ACP Journal Club or EvidenceBased Cardiovascular Medicine, a current textbook, or the Cochrane Library. We will consider the strengths and weaknesses of these resources and apply them to the clinical problems.

MEDLINE MEDLINE is a huge, multipurpose database of medical literature citations and abstracts produced by the US National Library of Medicine (NLM). It includes citations to almost all important clinical studies, and also a much larger volume of non-clinical studies and articles. Few other resources currently rival this scope. Accessing MEDLINE is relatively easy.3–5 CD Rom based systems, online systems, and, most importantly, internet access are all available. Examples of CD Rom systems include OVID, Aries, SilverPlatter, and DIALOG. Online access by modem is available through vendors such as PaperChase and HealthGate.6 Internet access is readily available (see Medical Matrix (http://www. medmatrix.org)

Finding current best evidence to practice evidence-based cardiology

and Dr Felix’s Free MEDLINE Page (http://www.beaker. iupui. edu/drfelix) sites for locations that offer MEDLINE access). Some of these MEDLINE systems have user fees but, at least at present, free access is available from many, led by the NLM PubMed (http://www.ncbi.nlm.nih. gov/PubMed/) internet site. This site features the earliest MEDLINE access to newly published articles and point-andclick search strategies that improve the yield of clinically useful studies on the cause, course, diagnosis, and treatment of clinical problems. If ready access is one of MEDLINE’s strengths, the skills needed to rapidly and dependably locate high quality articles that specifically address a clinical question are a weakness. A working knowledge of MEDLINE searching terminology and searching strategies is essential. Luckily, most hospital and university libraries offer training courses for MEDLINE. The NLM has also established a set of eight regional medical libraries that are charged with providing access and training for all US health personnel (1-800338-7657). Physicians in the UK can call the Health Care Information Service (44-207-412-7477) for similar information, while Canadians can call the Canada Institute for Scientific and Technical Information (1-800-668-1222). Turning to our initial scenario, we are interested in locating information about new medical therapies for patients with idiopathic dilated cardiomyopathy. Also, it would be wise to focus initially on treatments that already have been adequately tested in well-designed clinical trials.7 While it may be interesting to read about new medications that are being designed and tested at the laboratory level, or are undergoing early human testing, this information will not be immediately applicable in our clinical practices. Turning to MEDLINE for assistance, we might begin searching by using a medical subject heading (MeSH) for congestive heart failure. MEDLINE indexers choose appropriate terms from a thesaurus of 14 000 specific terms and 18 000 synonyms for content and methodology. Unfortunately, these terms are not always intuitive (for example, blockers indexed as adrenergic antagonists). Therefore, it is often necessary to search through the MeSH vocabulary before carrying out a search. The software for all search systems includes MeSH, so it is quite easy to search for appropriate terms. For our topic, a search for CONGESTIVE HEART FAILURE leads to HEART FAILURE, CONGESTIVE. Depending on the topic and the scope that you are interested in covering, you may also want to take advantage of two additional features of MeSH headings. Because many articles deal predominantly with two or three topics, the NLM will indicate these topics for each citation by designating them as major subjects of the article. Limiting your search to articles in which the search term has been designated as the major subject heading will be beneficial if you retrieve too many citations from using the search term without “majoring” it. Sometimes, though, you can miss

important studies this way. A trial and error approach may be needed to retrieve the best studies. “Exploding” is another useful feature of MEDLINE MeSH indexing. When articles are indexed, they are classified according to the most specific MeSH heading available. Thus, if you wish to identify all articles that deal with congestive heart failure, including those with more specific MeSH terms such as congestive cardiomyopathy, then you can do so by searching with the term EXPLODE HEART FAILURE, CONGESTIVE. If you are searching for a topic that has not been well indexed, you may want to take advantage of textword searching. Using this approach, you are simply asking MEDLINE to search the titles and abstracts of all the citations for any occurrences of a certain sequence of letters, such as “dilated”. This approach is particularly useful for new drugs or concepts that have not yet been incorporated into MeSH. MeSH is updated annually, but the lag can be considerably longer for certain terms. If several different endings to a word may have been used, and you wish to identify them all, you can use “truncation”, using the “*” symbol. For example, if you asked for RANDOM*, MEDLINE would search for RANDOM, RANDOMIZATION, RANDOMIZED, RANDOMISATION, RANDOMISED, and RANDOMLY. Be careful with truncation. The term “salmon*” retrieves not only the fish but salmonella as well! Some systems may use symbols other than “*”, such as “:” or “?”. Returning to identifying new therapies for patients with significant left ventricular dysfunction, EXPLODE HEART FAILURE, CONGESTIVE is a good start, but we need to narrow in on treatments that have been tested in well-designed studies. Luckily, a number of methodological search strategies have been tested and validated for retrieving sound studies for questions relating to therapy, prognosis, etiology or cause, and diagnosis (Box 5.1).8,9 Alternatively, you can search for a systematic review of studies. Research is currently ongoing to establish the best approach to identify systematic reviews and meta-analyses.10 For our current search, limiting the citations to systematic reviews and meta-analyses seemed like a reasonable first step. A simple but not fully validated strategy to identify systematic reviews and meta-analyses is to identify all citations in which the publication type is designated as meta-analysis (note that in addition to indexing articles according to subject, the NLM also indexes citations according to “publication type”). Over 40 publication types are recognized, including “meta-analysis”, “randomized controlled trial”, and “review”), as well as citations that include the phrase “meta-anal*” as a textword, and citations that are designated as reviews in the publication type section, but also have the textword “MEDLINE” in their abstract. Putting this all together yielded the search strategy in Box 5.2 using PubMed. 41


Box 5.1 Optimal search strategies for identifying studies relating to treatment, diagnosis, prognosis, or etiology using MEDLINE ● Treatment ● Best single term: clinical trial.pt. (“pt” indicates publication type) Combination of terms with best specificity: placebo:.tw. (“tw” indicates textword) OR double.tw. AND blind:.tw. Combination of terms with best sensitivity: randomized controlled trial.pt. OR random:.tw. OR drug therapy (as a subheading of the subject) OR therapeutic use (as a subheading of the subject) ● Diagnosis ● Best single term: explode diagnosis Combination of terms with best specificity: explode “sensitivity and specificity” OR predictive.tw. AND value:.tw. Combination of terms with best sensitivity: explode “sensitivity and specificity” OR explode diagnosis (as a subheading of the subject) OR sensitivity.tw. OR specificity.tw. OR diagnostic use (as a subheading of the subject) ● Prognosis ● Best single term: explode cohort studies Combination of terms with best specificity: prognosis OR survival analysis Combination of terms with best sensitivity: incidence OR explode mortality OR follow up studies OR prognos:.tw. OR predict:.tw. OR course:.tw. OR mortality (as a subheading of the subject) ● Etiology or cause ● Best single term: risk.tw. Combination of terms with best specificity: cohort studies OR case–control studies Combination of terms with best sensitivity: explode cohort studies OR explode risk OR odds.tw. AND ratio:.tw. OR relative.tw. AND risk:.tw. OR case.tw. AND control:.tw. Based on Haynes et al 8 and Wilczynski et al 9

Box 5.2 1. Heart failure, congestive 2. Meta-analysis[pt] 3. Meta-anal*[tw] 4. Review[pt] AND medline[tw] 5. #2 OR #3 OR #4 6. #1 AND #5

41 251 6123 9683 5156 16 637 141

(PubMed automatically explodes MeSH terms) (pt indicates publication type) (tw indicates textword) (the “OR” means that all citations in either #2 or #3 or #4 will be included) (the “AND” means that only citations that occur in both #1 and #5 will be identified)

Looking at the titles and abstracts of these articles, you find that one is a meta-analysis on blockers, and the abstract suggests that these medications are almost certainly beneficial. You decide to go to the library to retrieve this paper,11 and then to critically appraise it using the guidelines for a systematic review.12 42

Many alternative ways exist for conducting a MEDLINE search, including the one just displayed. Unfortunately, because no perfect recipe exists, what works well in one situation may not work as well in another. Combining an appropriate content term (HEART FAILURE, CONGESTIVE, in this case) with methods terms for reviews (as above) or


for sound study designs (as in Box 5.1) is a good place to start. It also has to be considered, however, that such searches are bound to take some time. This is because of the general nature of this huge biomedical research database: it is so large and comprehensive that even the extensive indexing and care that is taken in preparing it are insufficient to guarantee quick and accurate retrieval for clinical uses. Fortunately, many vendors have developed specialized subsets of MEDLINE for clinical use in cardiology. For example, Aries (http://www.kfinder.com) offers a cardiovascular disease subset (CardLine) on compact disc that you can subscribe to yourself. Instead of having 1 full year of MEDLINE on each CD Rom disc, these subsets provide journals and citations relating to a specific field for inclusion. For example, CardLine has cardiology citations from MEDLINE for the 10 most recent years on one disc.

Specialized clinical information resources While large electronic bibliographic databases such as MEDLINE can be very helpful, they can also be very frustrating or overwhelming because of the different ways that articles can be indexed and because of the vast array of preclinical and non-clinical literature that is included. MEDLINE serves many user groups besides clinicians (basic scientists and other researchers, educators, librarians, journalists, etc.). An alternative is to use a resource that includes only methodologically sound and clinically relevant articles, such as ACP Journal Club (American College of Physicians (ACP-ASIM)), for internal medicine and its subspecialties, Evidence-Based Medicine (for all major specialties; from ACP-ASIM and from the BMJ Publishing Group), and the cardiology journal Evidence-Based Cardiovascular Medicine (published by Churchill Livingstone). These are available in both paper and electronic versions. In addition to including only methodologically sound articles13 and presenting the results using a structured abstract format, these journals also include a commentary written by a clinical expert, designed to put the study findings into clinical context. Searching ACP Journal Club (www.acpjc.org) using the text phrase “low molecular weight heparin” locates several relevant references, including one directly on target.14 This report summarizes the findings of two randomized controlled trials comparing intravenous heparin administered in hospital with subcutaneous low molecular weight heparin administered at home, and both found that outpatient therapy was as safe and effective as inhospital management.

Other resources The Cochrane Library is an increasingly valuable source of evidence summaries and trials of healthcare interventions.

This new electronic database is updated quarterly and contains the collected work of the Cochrane Collaboration, an international voluntary organization that prepares, maintains, and disseminates systematic reviews of randomized trials of healthcare interventions. Available on CD Rom and via the internet (http://www.cochranelibrary.com), the Cochrane Library consists of three key sections for locating clinical evidence: the Cochrane Database of Systematic Reviews (CDSR), the Database of Abstracts of Reviews of Effectiveness (DARE), and the Cochrane Controlled Trials Registry (CCTR). The CDSR consists of the full reports of Cochrane Collaboration systematic reviews as well as protocols for ongoing systematic reviews. DARE is produced by the UK National Health Services Center for Reviews and Dissemination located at the University of York. It contains citations to many non-Cochrane systematic reviews, and includes structured abstracts for many of them. The CCTR is a growing collection of over 320 000 citations to therapeutic intervention trials. Searching the Cochrane Library is relatively easy and requires only entering a word or short phrase. The Library automatically searches all three sections for any relevant reviews or citations. Applying this to our scenarios, searching using the term “dilated cardiomyopathy” in Cochrane Library 2001, Issue 4, yields numerous citations: three citations to completed reviews in the CDSR, seven citations in the DARE, and 267 citations in the CCTR. The Cochrane reviews address the role of anticoagulation, antiplatelet agents, and digoxin in patients with a cardiomyopathy, and the structured abstracts within the DARE include a meta-analysis of blocker studies.15 Doing a similar search using the term “low molecular weight heparin” locates numerous references including a Cochrane review entitled “Home versus in-patient treatment for deep vein thrombosis”.16 This systematic review was updated in February 2001 and identified the two studies that were found earlier using ACP Journal Club.

Textbooks At this point, you may be thinking about your textbooks. What role do these have in clinical practice and in particular with respect to staying current? Textbooks remain an important resource for clinicians in terms of anatomy and pathophysiology, the basics of practice that usually do not change very quickly, except perhaps for molecular biology. They also provide descriptions of the classic presentations of numerous disease conditions and review important aspects of the history, physical examination, and diagnostic testing. By reviewing conditions that may present with similar findings, a good textbook can also help to broaden the differential diagnosis in more complex cases. These references may also describe medication adverse effects and pharmacokinetics, and 43


may include historical perspectives and practical suggestions to assist in patient management. Textbooks, however, are seldom explicit about the quality or currency of evidence used in recommendations for management. Also, there is often a passage of 3 or more years between updates of specialty textbooks, and new studies may have been published in the interval. Particularly for rapidly evolving aspects of management such as laboratory diagnosis and therapeutics, print textbooks simply cannot be trusted. Fortunately, we are now seeing the emergence of CD-ROM and Internet versions of textbooks with regular updates, such as UpToDate17 and Scientific American Medicine (SAM).18

of easy-access resources are available so that we can use this time effectively. MEDLINE is more readily available now than ever, and is seeding the development of specialtyspecific collections. Journals that abstract only high-quality, clinically relevant articles are appearing, and systematic reviews are becoming the norm. Internet-accessible textbooks that are regularly updated are also becoming available. Applying these resources to clinical care on an ongoing basis after appraising the quality of information and considering how it relates to our individual patient’s circumstances can lead to improvements in the quality of care we provide. Key points ●

The internet This brings us to the world wide web, an increasingly useful resource for locating current information, and one that our patients are accessing at an increasing rate. We have already mentioned how MEDLINE, ACP Journal Club and the Cochrane Library can be accessed over the web. A rapidly growing number of journals are also available online. A few examples include the New England Journal of Medicine (http://www.nejm.org), Annals of Internal Medicine (http://www.acponline.org), JAMA (http://jama.amaassn.org), BMJ (http://www.bmj.com), and The Lancet (www. thelancet. com). A number of cardiology textbooks are also available over the internet, as are many clinical practice guidelines. Two websites that have extensive cardiology sections are the Medscape (http://www.medscape.com) and Medical Matrix (http://www.medmatrix.org) sites. Journals and browsing to keep up to date We have focused to this point on looking for evidence when it is needed. If the search is successful, the evidence can be applied immediately and this can be a powerful learning experience. But what if we don’t search for evidence because we don’t know that we are out of date? A complementary strategy is needed, browsing the medical literature regularly in one way or another. The difficulty is that so many journals include articles relevant to cardiology that it is impracticable to review them all. One of the best solutions is to subscribe to a journal such as Evidence-Based Cardiovascular Medicine that continuously scans a wide range of journals in a systematic way, according to explicit criteria, and includes structured abstracts and commentaries on methodologically sound and clinically relevant studies. Conclusion In summary, while the time that we devote to updating ourselves with new developments is limited, a growing number 44

●

●

New resources are rapidly emerging that make keeping up to date with clinically significant developments in cardiology easier than ever. Large bibliographic databases, such as MEDLINE, are becoming more accessible to practicing physicians, and search strategies for locating high quality studies are now available. Specialty journals, such as Evidence-Based Cardiovascular Medicine, that identify and abstract methodologically sound and clinically relevant studies, also facilitate the ongoing process of staying current.

References 1.Covell DG, Uman GC, Manning PR. Information needs in office practice: are they being met? Ann Intern Med 1985; 103:596–9. 2.Osherof JA, Forsythe DE, Buchanan BG et al. Physicians’ information needs: analysis of questions posed during clinical teaching. Ann Intern Med 1991;114:576–81. 3.McKibbon KA, Walker-Dilks CJ, Beyond ACP Journal Club: how to harness MEDLINE to solve clinical problems (Editorial). ACP J Club 1994;120:A10–12. 4.Haynes RB, Walker CJ, McKibbon KA, Johnston M, Willan A. Performance of 27 MEDLINE systems tested by searches on clinical questions. J Am Med Informatics Assoc 1994;1: 285–95. 5.Engstrom P. MEDLINE free-for-all spurs questions about search value: who pays? Medicine on the NET 1996;2:1–5. 6.Haynes RB, McKibbon KA, Walker CJ et al. Online access to MEDLINE in clinical settings. A study of use and usefulness. Ann Intern Med 1990;112:78–84. 7.Sackett DL, Richardson SR, Rosenberg W, Haynes RB. Evidence-based medicine: how to practise and teach EBM. London: Churchill Livingstone, 1997. 8.Haynes RB, Wilczynski N, McKibbon KA, Walker CJ, Sinclair JC. Developing optimal search strategies for detecting clinically sound studies in MEDLINE. J Am Med Informatics Assoc 1994;1:447–58. 9.Wilczynski NL, MWalker CJ, McKibbon KA, Haynes RB. Assessment of methodological search filters in MEDLINE. Proc Ann Symp Comp Appl Med Care 1994;17:601–5.


10.Hunt DL, McKibbon KA. Locating and appraising systematic reviews. Ann Intern Med 1997;126:532–8. 11.Brophy JM, Joseph L, Rouleau JL. -blockers in congestive heart failure. A bayesian meta-analysis. Ann Intern Med 2001;134:550–60. 12.Oxman A, Cook D, Guyatt G. Users’ guides to the medical literature. VI. How to use an overview. JAMA 1994;272:1367–71. 13.Haynes RB. The origins and aspirations of ACP Journal Club (Editorial). ACP J Club 1991;114:A18. 14.Low-molecular-weight heparin at home was as effective as unfractionated heparin in the hospital in proximal DVT (Abstracts). ACP J Club 1996;125:2–3. [Abstracts of Koopman MM, Prandoni P, Piovella F et al. Treatment of venous thrombosis with intravenous unfractionated heparin administered in the hospital as compared with subcutaneous low-molecularweight heparin administered at home. N Engl J Med 1996;

334:682–7; and Levine M, Gent M, Hirsh J et al. A comparison of low-molecular-weight heparin administered primarily at home with unfractionated heparin administered in the hospital for proximal deep-vein thrombosis. N Engl J Med 1996;334:677–81.] 15.Zarembsk DG, Nolan PE Jr, Slack MK, Lui CY. Meta-analysis of the use of low-dose beta-adrenergic blocking therapy in idiopathic or ischemic dilated cardiomyopathy. Am J Cardiol 1996;77:1247–50. 16.Schraibman IG, Milne AA, Royle EM. Home versus in-patient treatment for deep vein thrombosis (Cochrane Review). In: The Cochrane Library, Issue 4, 2001. Oxford: Update Software. 17.Rose BD, ed. UpToDate. Wellesley, MA: UpToDate, Inc., 2001. 18.Dale DC, Federman DD, eds. Scientific American Medicine. New York: Scientific American Medicine, 1978–97.

45

6

Understanding concepts related to health economics Mark Hlatky

Introduction Economics is concerned with how to allocate scarce resources among alternative uses efficiently and effectively. It is a fundamental principle of economics that resources are limited relative to human wants, and that those resources have alternative uses.1 Consequently, when people say that the cost of health care has grown too high, they mean that the quantity of resources flowing toward medical care has grown to the point where additional funds cannot be spent on other things that society values, such as education, public safety, environmental protection, public works, pensions for the retired or disabled, or assistance to the poor. The fact that most people put a very high value on health does not mean that they are willing to provide limitless resources to medical care. Indeed, even the goal of improving health and longevity may also be served by non-medical expenditures on programs such as nutritional supplements, a safe and clean water supply, police and fire protection, or safety improvements to roads, as well as by medical expenditures. The cost of medical care has been rising steadily for the past 50 years, but it has only been in the past decade that the level of expenditures became so large as to cause alarm among

Table 6.1 US national healthcare expenditures, 2000 Category Hospital care Physician services Other professional services Drugs, supplies Nursing home care Home health care Other personal health care Administration Public health Research Construction Total

US$ (109) 412·1 286·4 99·0 171·5 92·2 32·4 36·7 80·9 44·2 25·3 18·6 1299·5

Source: Health Affairs 2002;21:207–18

46

Percentage 32 22 8 13 7 2 3 6 3 2 1 100

policy makers, payers, and the general public (Table 6.1). The steady expansion of health care has now begun to meet substantial resistance in the large industrial countries, and new policies and payment mechanisms have been introduced to contain the rising cost of medical care. As a consequence, physicians must now consider cost as they design programs to prevent, diagnose, and treat disease. Cardiovascular diseases consume a large share of health care resources (Table 6.2), so cardiovascular specialists must be particularly knowledgeable about health economics. This chapter will attempt to outline the major principles of health economics relevant to cardiovascular medicine. First, some general concepts of health economics will be presented. Second, methods to identify and compare the costs of cardiovascular interventions will be described. Finally, the principles of cost effectiveness analysis will be discussed.

Table 6.2 the USA

Resources devoted to cardiovascular care in

Category

n (103)

Deathsa Hospital admissionsb Myocardial infarction Heart failure Cerebrovascular disease Operations and proceduresb Cardiac catheterization Coronary bypass surgery Coronary angioplasty Pacemaker-related Physician office visitsc Electrocardiograms Prescriptionsc

934 6344 829 962 961 6133

Percentage (of total) 39 20

15

1271 355 599 336 59 996 22 596 176 839

8 16

Sources: a NCHS Monthly Vital Statistics Report 2001; 49:12 b NCHS Advance Data 2001 (No. 319) c NCHS Advance Data 2001 (No. 322)

Health economics

General concepts C Health outcomes

Various societies have adopted different systems to pay for health care, and these systems reflect societal values and the historical experience within each country. The United Kingdom has a national health service, Canada has national health insurance, France and Germany have public/private financing for health care, and the United States has a perplexing and rapidly evolving patchwork of public and private health insurance systems. These are very different systems to finance health care, and yet each is faced with the same issues of how to allocate the limited resources available to best provide health care. Each country is also facing the same steady rise of healthcare costs, despite the wide differences in the ways they finance health care. Provision of cardiovascular services requires resources in all societies, irrespective of the method of financing or delivering health care. Coronary bypass surgery, for example, is very resource-intensive, with the operation requiring cardiac surgeons, a cardiac anesthesiologist or anesthetist, a perfusionist, several nurses, and considerable quantities of specialized supplies and equipment. Postoperative care also requires skilled nurses and physicians, with support from specialized supplies, equipment, and facilities. Each health professional involved in cardiac surgery spends the scarce resource of time to care for the patient – time that could be put to other valuable uses, such as care for other patients. The drugs used, the disposable supplies, the operating room equipment, even the hospital building, all cost money. All of these are true costs to the system, even if the coronary bypass operation is performed “for free” – that is, without charge to the patient. The scene in the operating room, the postoperative recovery areas, and the hospital wards is much the same in the United Kingdom, Canada, France, Germany, and the United States despite the different ways these societies pay for medical care. The resources used in the care of patients, and the increasing sophistication of that care, drive healthcare costs up in each of these countries, irrespective of the way such care is paid for. Another basic concept of economics is the so-called “law of diminishing returns”. This concept is illustrated in Figure 6.1, in which the quantity of resources used in health care is plotted on the horizontal axis, and the resulting health benefits on the vertical axis. In the case of the patient with an acute myocardial infarction, for example, survival would be improved as more resources are applied, such as prehospital transportation, electrocardiographic monitoring, access to defibrillation, and a competent team to deliver coronary care. Outcomes might be further improved by reperfusion therapy, but with a greater increment in survival from using a cheaper, basic approach (streptokinase, for example) relative to no therapy, than from more expensive alternatives (such as tissue plasminogen activator (tPA) or PTCA). The extra benefit from adding even more aggressive care will be

B

D

A

Resources

Figure 6.1 General relationship between increasing healthcare resources (horizontal axis) and health outcomes (vertical axis). At point A, outcomes are improving rapidly with increased resources and treatment is cost effective. At point B, outcomes are still improving with increased resources, but at a rate that is not cost effective. At point C, increased resources are no longer improving outcome (that is, “flat of the curve”), and at point D increased resources actually lead to worse outcomes, through iatrogenic complications and overtreatment.

smaller still, and at some point the patient may be harmed by overly aggressive care. Helping physicians define the optimal point on this curve (Figure 6.1) is one of the goals of economic analysis. Determination of costs The cost of producing a particular healthcare service can be defined in a variety of ways. The cost of performing a coronary angiogram can be used as a specific example that will illustrate the various aspects of cost and how the cost might be measured. Performing a coronary angiogram requires a variety of resources, including radiographic equipment, trained personnel (including an angiographer and technical assistants), and specialized supplies such as catheters, radiographic contrast, and sterile drapes. The equipment needed is very expensive to purchase, and the healthcare facility where it is installed may require special renovations to assure proper radiation shielding and adequate electrical power. The capital cost for a coronary angiography laboratory will be considerable, perhaps $2–3 million, depending on the type of equipment purchased. The laboratory will have a physical lifespan of perhaps 7–10 years, although technologic innovations may lead to replacement of the equipment before the end of its physical lifespan. The cost of building an angiography suite represents a large fixed cost for coronary angiography, a cost that is roughly the same whether the laboratory performs 200 or 2000 angiograms per year. The cost per case is lower in the high volume laboratory, however, because the fixed equipment costs can be spread over more cases. Thus, if the equipment costs 47


$2 500 000 $1250/case Fixed costs/case (200 cases/yr)(10 years) whereas in the high volume laboratory (2000 cases per year) the prorated share of fixed costs per case would be $2 500 000 $125/case Fixed costs/case (2000 cases/yr)(10 years) Procedures that have high fixed costs will be performed with greater economic efficiency in centers that have sufficient volume to spread those fixed costs over a larger number of individual patients. (There may be additional advantages to larger procedure volumes as well, since the technical proficiency is higher and clinical outcomes of many procedures are usually better when performed in higher volume clinical centers.)2–4 Procedures with lower fixed costs will have a smaller effect of volume on costs. In contrast to the fixed equipment costs, the cost of supplies consumed in performing coronary angiography varies directly with the volume of cases performed, and the supply cost per case will be fairly constant irrespective of the volume of cases performed (apart from the small effect of discounts available to large volume purchasers). The cost of laboratory staff falls in between these two extremes, in that the hours worked in the catheterization laboratory by technical staff can be varied somewhat according to the volume of cases performed, but some staff effort is required regardless of patient volumes, such as supervisors. Hospital overhead is also a real cost, but one that is less directly linked to any one medical service or procedure. Hospitals must pay for admitting offices, the medical records department, central administration, the laundry service, the cafeteria, housekeeping and utilities, to name just a few areas. These costs cannot be tied easily to the coronary angiography procedure in the same way as the cost of the catheters or radiographic contrast. Most facilities assign a share of these costs to patient care services according to a formula such as the step down method. Discussion of specific methods to allocate hospital overhead is beyond the scope of this chapter, but can be found in several articles and books.5,6 The overall effect of procedure volume on the cost per case is illustrated in Figure 6.2. In general, the cost per case declines as more cases are performed, up to the limit of the facility’s capacity (for example, 2000 cases). If volume increases further, more facilities must be built, increasing the cost per case, as more fixed costs are spread over a few more patients. Figure 6.2 also illustrates the distinction between concepts of “marginal cost” and long run “average cost”. The marginal cost is the added cost of doing one more case. In an already equipped and staffed coronary angiography laboratory, the marginal cost of performing one more 48

1250

Cost per case ($)

$2·5 million and has a useful life of 10 years, the prorated share of fixed costs for each patient in the low volume laboratory performing 200 cases per year is

1000 750 500 250 1 lab 0

0

2 labs

3 labs

4 labs

1000 2000 3000 4000 5000 6000 7000 8000 Number of cases

Figure 6.2 Cost per case for coronary angiography as a function of clinical volume. Assumes fixed costs per laboratory of $250 000 per year, and marginal (that is, variable) costs of $250 per case. When volume reaches 2000 cases per year in a laboratory, the model assumes an additional laboratory will be built. The dotted line indicates the “long run average” cost per case of $375.

procedure is just the cost of the disposable supplies consumed in the case: the catheters, radiographic contrast, and other sterile supplies. In the example of Figure 6.2, the marginal cost is $250 per case. The marginal cost is lower than the average cost per case ($375 per case), which also includes a prorated share of the salaries of the laboratory staff, depreciation of the laboratory equipment costs, and the facility’s overhead costs. Costs v charges Costs and charges are related but distinct concepts. The cost of a medical service represents the value of the resources required to produce it. The charge for a service is a specific form of reimbursement to healthcare providers in a fee-forservice healthcare system. The cost and charge for providing a service should be quite close to one another in a competitive economic market. The reason is simple: if one provider charged an amount much higher than it actually cost to provide a service, a competitor could offer the same service at a lower price and still come out ahead. Conversely, if a provider charged less than the costs of production, the provider would lose money. These basic economic principles have not applied very well to medical care, at least until recently, because medical care has not had significant competition on prices. In regulated or non-price competitive healthcare systems, the charge (price) for a service need not bear a close relationship to the cost of producing a service. Hospitals might choose to set high prices for some services, such as coronary bypass surgery, and use the excess revenues to subsidize other services that were less well reimbursed, such as the emergency department or, in academic centers,

Health economics

medical education, and research. With greater price sensitivity on the part of healthcare payers, the subsidization of one medical service with the proceeds of another service has been sharply curtailed. While this trend has had the positive effect of bringing an element of economic reality to medicine, it has also caused dislocations and considerable harm as worthwhile endeavors have lost the funding they previously enjoyed from cross-subsidization. In the long run, medical education, clinical research, and services to uninsured or poorly insured patients should receive direct funding to replace the indirect funding by cross-subsidies, but in the transition these endeavors have been threatened due to loss of a traditional funding base.

Estimation of costs The cost of providing a specific service, such as coronary bypass surgery, can be established in several alternative ways. In principle, one valid way to measure cost would be to identify a competitive market for medical service, and note the charge (price) for coronary bypass surgery in that market. While competitive market pricing might work well for commodities such as consumer electronic devices or farm products, it is not well suited to medical care, where there are few competitive markets. An alternative method to measure cost is to take note of the charge for a service, and apply correction factors to estimate cost more accurately. A third method to estimate costs is to examine in detail the resources used to provide a service, and apply price weights to the resources used: Costj

(Quantity) (Price) ij

ij

i

The use of these different approaches to cost determination is illustrated by a study that estimated the costs savings achievable by substituting coronary angioplasty for coronary bypass surgery.7 In that study, hospital billing records were used to construct resource consumption profiles for patients

undergoing either angioplasty or bypass surgery for the treatment of stable angina. A microcosting approach was then applied to the resource consumption profiles, with the cost of a specific resource (for example, an electrocardiogram) defined either as the cost of supplies only (marginal cost or variable cost) or as the cost of supplies, personnel and equipment, but omitting overhead (average direct cost). Charges on the billing record for a service were also converted to costs by two different correction factors, also known as ratios of cost to charges. One cost–charge ratio included all direct costs of providing a service (supplies, personnel, and equipment), but omitted hospital overhead (for example, medical records, laundry, utilities, administration). The second cost–charge ratio allocated a share of hospital overhead to each service in addition to its direct costs. As shown in Table 6.3, the cost savings attributable to substitution of angioplasty for bypass surgery varied considerably according to how costs were defined. The lowest cost savings are evident when only marginal costs are included, and fixed cost and overhead excluded. The average direct cost difference is intermediate in value, and comparable estimates of this cost savings were obtained from the use of resource consumption profiles and cost weights (method 2) or the cost-to-charge ratio method that omitted overhead (method 3). Finally, the inclusion of overhead (method 4) led to the highest absolute difference in costs. The differences in cost as estimated using these methods is directly related to the issue of how the information is to be used or, put another way, depends on the answer to the question, “cost to whom?”. A hospital manager might be most interested in the marginal cost of procedures (method 1) in looking at the effect of adding or subtracting a small number of cases to the volume performed in an institution. Under a fixed budget for cardiac services, for instance, the effect of substituting angioplasty for bypass surgery may be small, given that the personnel, equipment, and overhead are largely fixed. Performing a few extra procedures with existing facilities adds very little cost from the perspective of the

Table 6.3 Effect of different definitions of cost on the savings possible by substitution of coronary angioplasty for bypass surgery Definition of cost

PTCA cost ($)

CABG cost ($)

Difference ($)

PTCA/CABG ratio (%)

Variable cost only Average direct cost By microcosting By ratio of cost-to-charges Average cost overhead Charges

2672

4607

1935

58

4073 4935 7530 9556

8666 10 281 15 367 19 644

4593 5346 7837 10 088

47 48 49 49

Source: adapted from Hlatky et al 7

49


head of a clinical service or a hospital manager. They may even be willing to perform a modest number of additional cases at a reimbursement level below their actual average cost, but above marginal cost, in order to increase volume and spread their fixed costs over more patients. Thus, marginal or variable costs are quite relevant to decision makers within the institution providing a service. The perspective of a policy maker or health planner includes a longer time horizon and the possibility of adding or subtracting substantial volumes of clinical services. From this perspective, no cost is truly fixed, for personnel needs can be adjusted, and the number of facilities providing a service can be altered. This perspective is a broader one, and the costs considered are therefore more inclusive. For most policy level discussions, long run average cost is the most relevant measure.

International perspectives With the advent of large multicenter clinical trials that enroll patients from several countries, interest has developed in cost comparisons between countries for the same service. Cost estimation as part of large randomized trials will enhance clinical decision making, for the randomized design is the strongest way to compare all outcomes of therapeutic alternatives, including cost. Extension of cost comparisons across national borders introduces a number of technical and conceptual issues that deserve discussion. Different countries measure cost in their respective national currencies, so that readers in another country need to convert between units (for example, pounds sterling or euros to US dollars). These conversions can be done using currency exchange rates, or the closely related purchasing power parity factors. The differences between countries in units of measurement are important, but this issue is fairly simple to address. A more thorny issue in international comparisons is differences in the relative prices of the resources used to provide a service and differences in resource profiles used to

provide a service. Thus, if the cost of service j is defined as Costj

(Quantity) (Price) ij

then cost may differ among countries due to either differences in the quantity of resources used to provide a service, price differentials for the same resources, or both. A specific example will help illustrate these concepts (Table 6.4). Care of a patient with acute myocardial infarction given thrombolysis includes the cost of the drug, the cost of basic hospital care, and the cost of additional tests and treatments in the convalescent phase. Table 6.4 presents hypothetical costs of basic care in two countries, with monetary values expressed in dollar units for simplicity. The costs of drugs in Country 1 are higher than in Country 2, where drug prices are strictly regulated. The time spent by the hospital staff to care for the patient are quite similar in Country 1 and Country 2 (50 hours per patient for Treatment A and 54 for Treatment B, a difference due to lower complication rates with Treatment A). The average hourly compensation for hospital staff is, however, higher in Country 2, so that total personnel costs are higher as well. Thus, both cost savings and the relative costs of Treatment A and B are different in these two healthcare systems, due to different prices for the same resources used to care for a myocardial infarct patient. There may also be differences in the level of resource use between countries, especially for discretionary procedures such as coronary angiography. Suppose that the use of Treatment A cuts the use of coronary angiography by one third, partially offsetting the higher cost of the drug. If, however, the baseline rates of angiography are different between countries, the cost implications of reducing angiography by one third in each country will be quite different (Table 6.5). A reduction by one third in the high rate of angiography in Country 1 (from 60% to 40%) provides a $200 cost offset per patient, whereas a reduction by one third in the low rate of angiography in Country 2 (from 20% to 15%) provides only a $50 cost offset per patient. International comparisons of the cost of therapies can thus be affected by (a) differences in resource use patterns that reflect differences in practice style and the organization

Table 6.4 Effect of differences in medical prices on costs of alternative treatments Country 1 Tx A Drug ($) 2000 Nursing hours 50 Nursing wages ($) 30 Total ($) 3500 Cost savings (AB) ($) Cost ratio B/A Abbreviation: Tx, treatment

50

1680 52%

ij

i

Country 2 Tx B

Tx A

Tx B

200 54 30 1820

1500 50 35 3250

150 54 35 2040 1210 63%

Health economics

Table 6.5 Effect of differences in resource utilization on costs of alternative treatments Country 1

Drug/nursing ($) Coronary angiography Angio cost ($) Total ($) Cost savings (AB) ($) Cost ratio (B/A)

Country 2

Tx A

Tx B

Tx A

Tx B

3500 40% 1000 3900

1820 60% 1000 2420

3250 10% 1000 3350

2040 15% 1000 2190

1480 62%

1160 65%

Abbreviation: Tx, treatment

of medical care, and (b) by differences between countries in the prices attached to specific resources, such as healthcare wages, drugs, and supplies. Data from cost studies can be most readily applied in different practice environments if the study provides information on both resource consumption patterns and price weights attached to the specific resources used, as well as a summary cost measure. This detail is needed for readers to understand the applicability of the cost findings to their own practice settings.

Cost effectiveness analysis The cost of providing a particular medical service can be measured, but determination of whether the service provides good value for the money spent is a more difficult judgment. Cost effectiveness analysis is a method of weighing the cost of a service in light of the health effects it confers in an attempt to facilitate the ultimate value judgment about whether the service is “worth” the cost. Cost effectiveness analysis is one of several alternative analytic methods, each with its own strengths and limitations.5 If two alternative therapies are either known to yield identical results or can be shown to be clinically equivalent, they can be compared on the basis of cost alone. This form of analysis, which is termed “cost-minimization analysis”, is particularly appropriate to commodities such as drugs, supplies, and equipment that can be expected to yield equivalent results when applied clinically. In such situations, the relative costs of the alternatives become the predominant consideration. Many alternative therapies are known to differ both in clinical outcomes and in cost. In this situation, both the difference in cost and the difference in effectiveness of the therapeutic alternatives must be measured and weighed against each other. When the effectiveness on intervention is measured in clinical terms (for example, lives saved, years of life added), the analysis is termed “cost effectiveness”. If the clinical measures of effectiveness are translated into monetary units, the term “cost–benefit analysis” is applied.

Cost–benefit analysis has been used to guide public policy in areas outside of medicine, such as in the construction of transportation systems or whether to remove or reduce environmental exposures. Cost–benefit analysis measures the effects of programs in monetary terms, so that net cost (in dollars) can be compared with net benefits (in dollars). Since there is great reluctance on the part of physicians and health policy makers to assign a dollar value to saving a life or improving a patient’s function, cost effectiveness analysis rather than cost–benefit analysis has been applied predominantly to medical problems. Cost effectiveness analysis was first applied to medical programs only 25 years ago8,9 and has since been widely used.10–12 The principles of cost effectiveness analysis for medical programs have recently been examined in detail by a Task Force convened by the United States Public Health Service.13–15 A group of experts attempted to establish consensus on a number of methodologic issues, with the goal of standardizing the technical aspects of cost effectiveness analysis among studies, thereby enhancing their comparability. The principles articulated by this group are reasonable, and should guide this important field in its next 20 years. A basic principle of cost effectiveness analysis is that the analysis should compare alternative programs, and not look at any single program in isolation. Thus, a drug to treat lifethreatening arrhythmias might be compared with placebo, or an implantable cardioverter defibrillator might be compared with a drug. In essence, cost effectiveness analysis must always answer the question “cost effective compared with what?”. Another principle is that the costs included in cost effectiveness analysis should be comprehensive. The cost of a specific therapy should include the cost of the intervention itself (for example, thrombolytic therapy for acute myocardial infarction) and the costs of any complications the therapy induced (for example, bleeding), less any cost savings due to reduction of complications (such as, heart failure). The need for other concomitant therapy should also be included, which is particularly important when assessing the cost effectiveness of screening programs or diagnostic testing strategies. 51


The length of follow up should be sufficient to include all relevant costs and benefits – such as readmissions to the hospital due to treatment failures. Non-monetary costs directly related to the medical intervention should also be included, such as the cost of home care by the patient’s family, since omission of these costs would bias assessments toward programs that rely on unpaid work by family members or volunteers. Other costs not directly related to the intervention, however, such as the patient’s lost wages or pension costs, are omitted by convention from the measured costs in a cost effectiveness analysis. Another important issue in cost effectiveness analysis is the perspective taken by the analysis. There is general agreement that the analysis should include all relevant costs, regardless of who pays them. This principle is known as “taking the societal perspective”, and it assures a complete accounting of costs in the analysis. A hospital, for instance, may not care about the out of pocket costs paid directly by the patient, but these are real costs and should be considered in the analysis. Medical costs may accrue over long periods of time, especially in preventive programs or the treatment of chronic disease. Time scales of more than a year or so bring up two related but distinct issues – inflation and discounting. The nominal value of any currency changes over time; a dollar in 1977 had more purchasing power than a dollar in 1997. Studies conducted over long time periods will need to correct for the changing value of currencies, typically by application of the Consumer Price Index (or the GDP deflator). Application of the Consumer Price Index removes the effect of inflation, but does not address the separate issue of time preference for money. Even in a country free of inflation, citizens would prefer to receive $100 today than a promise they will be paid $100 in a year. One might have to promise to pay more money in a year, say $103, to compensate for the delay. The same is true in health programs: we would prefer to be paid today instead of in the future, and we would also prefer to pay our obligations in the future rather than today. Use of a discount rate provides a way to correct for the lower value of future costs relative to current costs. The technical experts’ consensus is that future costs should be discounted at a rate equivalent to the interest paid on safe investments such as government bonds in an inflation-free environment, or about 3% per year. The effect of alternative discount rates between 0% and 5–10% per year should also be checked to document the sensitivity of the analysis to future costs. In summary, a cost effectiveness analysis should include all medical costs, including those of complications of therapy and adverse effects prevented. The study should be of sufficient duration to measure all relevant costs and benefits of the treatment. All costs and benefits should be included, regardless of who bears or receives them. In studies covering more than a year or so, corrections should be made for 52

inflation, and 3% per year discount rate should be applied to follow up costs.

Measuring effectiveness The effectiveness of an intervention in practice can be measured in a variety of ways, with different outcome measures most appropriate for specific applications. Physiologic end points are often used in clinical trials, with the result of therapy assessed by a laboratory measure such as millimeters of mercury for blood pressure or episodes of non-sustained ventricular tachycardia on an electrocardiographic monitor. Laboratory measures are useful in judging the physiologic effects of therapy and its mechanism of action, but these surrogate markers may not predict the ultimate effect of therapy on mortality and morbidity, as vividly illustrated by the results of the Cardiac Arrhythmia Suppression Trial (CAST).16 Physiologic end points are also tied closely to one specific disease, making comparisons against other benchmark therapies difficult. The patient and public are most concerned with the effect of therapy on survival and on their ability to function – that is, upon the length of life and the quality of life. A common denominator measure of effectiveness is thus the life years of expected survival, or the quality adjusted life years (QALYs). This measure is relevant to patients and to the public and can be applied to virtually any therapy. Mortality is a common end point in clinical trials, and leads directly to the measure of life years of survival. The mean life expectancy of a cohort of patients is equal to the area under a standard survival curve. The difference in life expectancy between two therapies is therefore equal to the difference in the areas under their respective survival curves. Since many clinical studies do not follow patient cohorts long enough to observe complete survival times for all patients, some assumptions and modeling of long-term survival may be needed to estimate the full survival benefit of therapy for a cost effectiveness analysis.17 Improvement in quality of life is often as important to patients as reducing mortality, and it is often the main goal of therapies, such as the relief of disabling angina or improvement in exercise tolerance. A quality of life measure can be translated into a scale that ranges from a low of 0·0 (the worst possible health state, usually taken as death) to 1·0 (perfect health). This quality of life measure is multiplied by the length of time a patient spends in the health state to yield a quality adjusted life year (QALY). Thus: QALY

Qt i

i

i

where QALY the quality adjusted life years, Qi the quality factor for follow up period “i” and ti the length of time spent in period “i”. This equation shows that the effectiveness of a treatment, as measured in QALYs, can be improved

Health economics

by either enhancing the patient’s quality of life (Qi) or the patient’s length of life (ti), or both.

Calculation of cost effectiveness After the costs of therapy and the medical effectiveness of therapy have been assessed, cost effectiveness (CE) can be calculated as: CE ratio

Cost2 Cost1 QALY2 QALY1

where Cost1 and Cost2 represent the costs of program 1 and program 2, respectively, and QALY1 and QALY2 represent the effectiveness of programs 1 and 2, respectively. There are several implications of using this formula. First, cost effectiveness ratios that are positive (that is, 0) result if and only if one alternative has both higher cost and greater effectiveness – that is, Cost2 Cost1 and QALY2 QALY1 (or the reverse: Cost2 Cost1 and QALY2 QALY1). Cost effectiveness ratios of 0 are not generally important for decision making, since they arise only when one alternative has both lower costs and greater clinical effectiveness than the other (for example, Cost2 Cost1, and QALY2 QALY1). In this case, program 1 is superior in all respects: it has better outcomes and lower cost than program 2, and thus is said to “dominate” the alternative. The decision of which program to recommend is therefore simple. Another important implication of the formula used to calculate cost effectiveness is that the ratio is undefined when the two alternatives provide equal outcomes, since when QALY2 QALY1 the denominator in the cost effectiveness ratio is equal to zero. The implication is that when the difference in outcomes between two programs is negligible, cost effectiveness analysis is unnecessary, and the choice between two alternatives can be based on cost alone (that is, cost minimization analysis is more appropriate than cost effectiveness analysis). Most commonly, one of two therapeutic alternatives has higher costs and greater effectiveness, and use of the formula yields a cost effectiveness ratio greater than zero. One treatment may have a cost effectiveness ratio of $5000 per year of life saved, and another might have a ratio of $75 000 per year of life saved. Since it is problematic to assign a dollar value to life, interpretation of these ratios is best made by consideration of benchmarks – other generally accepted therapies that serve as a rough gauge to an “acceptable” cost effectiveness ratio. Renal dialysis is a form of therapy that most people would consider expensive, and yet dialysis is an intervention that the USA and most other industrialized countries provide as a life saving therapy. The end stage renal disease program in the USA costs about $35 000 a year per patient, and if this therapy were withdrawn the

patient would die. Thus, renal dialysis has a cost effectiveness ratio of $35 000 per year of life saved (or if one considers the reduced quality of life for a dialysis patient, perhaps $50 000 per quality adjusted year of life saved). Therapies with cost effectiveness ratios considerably more favorable than renal dialysis (that is, $20 000) would be considered very cost effective, whereas therapies with cost effective ratios much higher (say $75 000) would be considered too expensive. Different societies may come to different conclusions about the level of cost effectiveness they consider good value. Wealthy countries with high per capita incomes are more willing to pay for expensive therapies than are poor countries. For instance, the percentage of gross domestic product and per capita health spending in Eastern Europe is much less than in Western Europe or North America, and these countries have not chosen to provide expensive services such as bypass surgery as readily or as frequently. Decisions about funding programs might be more equitable and rational when guided by the relative cost effectiveness of programs. When studies use similar methods to measure cost and effectiveness, cost effectiveness ratios can be compared to rank the economic attractiveness of alternatives. Tables comparing various treatments, such as Table 6.6, have been termed “league tables” because of their similarity to the athletic league standings published in newspapers. Given the uncertainty inherent in measuring cost and effectiveness of medical interventions, and the methodologic variations among studies, only relatively large differences in cost effectiveness ratios should be considered significant. Thus, a program with a cost effectiveness ratio of $5000 per life year added is much better than one with a ratio of $30 000. Programs with ratios of $25 000 and $30 000 are so close that no firm conclusion about the relative values should be drawn.

Patient selection and cost effectiveness Drugs and procedures in medicine are applied to different patient groups for different clinical indications. The medical effectiveness of therapies varies considerably according to patient selection. Cholesterol lowering therapy, for instance, will extend the life expectancy of a patient with multiple cardiac risk factors more than it will for a patient with the same cholesterol level and no other cardiac risk factors. Coronary bypass surgery provides greater life extension to a patient with left main coronary artery obstruction than it does to a patient with single vessel disease.18 The cost effectiveness ratio for these therapies will therefore vary among patient subgroups due to the impact of patient characteristics on the clinical effectiveness of therapy, which forms the denominator of the cost effectiveness ratio. Similarly, the cost of a particular therapy may also vary according to patient characteristics, since the therapy itself may be more 53


Table 6.6 Cost effectiveness of selected cardiovascular therapies Strategy

Patient group

Cost effectivenessa

Lovastatin

Post MI Men 45–54 Chol 250 CHF EF 3 5% WPW, post cardiac arrest Smoking Post-MI High-risk Left main CAD Severe angina Post-MI Low-risk Primary prevention Men 55–64 Chol 300 Three risk factors Acute MI Sustained VT Two vessel CAD Angina Primary prevention Men 55–64 Chol 300 No other risk factors Asymptomatic 40 year old men Single vessel CAD Mild angina Primary prevention 35–44 year old women Chol 300 No other risk factors

Saves dollars and lives

Enalapril Radio frequency ablation Physician counseling blocker CABG blocker Lovastatin

tPA ICD CABG Lovastatin

Exercise ECG CABG Lovastatin

Saves dollars and lives Saves dollars and lives $1300 $3600 $9200 $20 200 $20 200

$32 800 $35 000 $42 500 $78 300

$124 400 $1142 000 $2 024 800

a

$ values, dollars per year of life added. Abbreviations: CAD, coronary artery disease; CHF, congestive heart failure; Chol, Cholesterol; ICD implantable defibrillator; MI, myocardial infarction; VT, ventricular tachycardia; WPW, Wolff–Parkinson–White syndrome; See glossary for other abbreviations Source: adapted from Kupersmith et al.10–12

or less expensive according to different patient subgroups, or the likelihood of costly complications may be higher or lower in different groups of patients. The clinical effectiveness of a therapy is generally the most important factor determining cost effectiveness. The reason for this importance is (a) that clinical effectiveness of a therapy generally varies more among patients than does the cost of therapy, and (b) the value of the cost effectiveness ratio is more sensitive to changes in the denominator (effectiveness) than to changes in numerator (cost). In the last analysis, a therapy must be clinically effective before it can be cost effective. Cost 54

effectiveness analysis relies more on the assessment of medical effectiveness than it does on determination of cost. Diagnostic tests and cost effectiveness Cost effectiveness analysis has been applied primarily to assess specific therapies or therapeutic strategies, for which it is natural to measure effectiveness in terms of patient outcome. The principles of cost effectiveness can be extended to analyze screening tests and diagnostic strategies as well, but some additional factors must also be considered.

Health economics

Therapies are expected to improve patient outcome directly, by intervening in the pathophysiology of disease processes. In contrast, a diagnostic test is expected to provide the physician with information about the patient, which in turn is expected to improve management and thereby indirectly improve patient outcome. The value of a test is therefore linked closely with patient selection for therapy, and the value of testing may well change as new therapies are developed, or alternative tests become available. The information provided by a test may be used in different decisions, and the test may be more or less useful in these different settings. An exercise electrocardiogram, for example, can be used as a diagnostic test for coronary disease, a prognostic test for patients with recent myocardial infarction, a monitoring test to assess the effect of antiischemic therapy, or even as a way to establish target heart rates for an exercise training program. The efficacy and cost effectiveness of applying the exercise electrocardiogram will be different for these varied uses of the information provided by the test. The value of the test will depend on the indication for which it is used, much as the value of a blocker will vary whether it is used to treat hypertension or as secondary prevention after a myocardial infarction. The same test (for example, the exercise ECG) applied for the same purpose (such as diagnosis of coronary disease) will provide more information in some groups of patients than in others. As discussed elsewhere in this book, a diagnostic test provides more value if used when the pretest probability of disease is intermediate than when the pretest probability is either very high or very low. The test has the most value when the result is likely to change the estimated probability of disease such that clinical management is changed. Tests that never change patient management cannot change patient outcome, which is the “bottom line” in assessing cost effectiveness.

Conclusions Economic analysis is designed to assist decisions about the allocation of scarce resources. Physicians now must address the cost implications of clinical decisions, and be aware of the effects on scarce resources. Economic efficiency is but one of many goals, however, and issues of fairness and humaneness are also central to medical care, and must be considered as well.

References 1.Fuchs VR. Who shall live? Health, economics and social choice. New York: Basic Books, 1974.

2.Jollis JG, Peterson ED, DeLong ER et al. The relation between the volume of coronary angioplasty procedures at hospitals treating Medicare beneficiaries and short-term mortality. N Engl J Med 1994;331:1625–9. 3.Kimmel SE, Berlin JA, Laskey WK. The relationship between coronary angioplasty procedure volume and major complications. JAMA 1995;274:1137–42. 4.Hannan EL, Racz M, Ryan TJ et al. Coronary angioplasty volume–outcome relationships for hospitals and cardiologists. JAMA 1997;279:892–8. 5.Drummond MF, Stoddart GL, Torrance GW. Methods for the economic evaluation of health care programmes. Oxford: Oxford University Press, 1987. 6.Finkler SA. The distinction between costs and charges. Ann Intern Med 1982;96:102–10. 7.Hlatky MA, Lipscomb J, Nelson C et al. Resource use and cost of initial coronary revascularization. Coronary angioplasty versus coronary bypass surgery. Circulation 1990;82(Suppl. IV): IV-208–IV-213. 8.Weinstein MC, Stason WB. Foundations of cost-effectiveness analysis for health and medical practices. N Engl J Med 1977;296:716–21. 9.Detsky AS, Naglie IG. A clinician’s guide to cost-effectiveness analysis. Ann Intern Med 1990;113:147–54. 10.Kupersmith J, Holmes-Rovner M, Hogan A, Rovner D, Gardiner J. Cost-effectiveness analysis in heart disease, Part I: general principles. Prog Cardiovasc Dis 1994;37:161–84. 11.Kupersmith J, Holmes-Rovner M, Hogan A, Rovner D, Gardiner J. Cost effectiveness analysis in heart disease, Part II: preventive therapies. Prog Cardiovasc Dis 1995;37:243–71. 12.Kupersmith J, Holmes-Rovner M, Hogan A, Rovner D, Gardiner J. Cost effectiveness analysis in heart disease, Part III: ischemia, congestive heart failure, and arrhythmias. Prog Cardiovasc Dis 1995;37:307–46. 13.Russell LB, Gold MR, Siegel JE, Daniels N, Weinstein MC. The role of cost-effectiveness analysis in health and medicine. JAMA 1996;276:1172–7. 14.Weinstein MC, Siegel JE, Gold MR, Kamlet MS, Russell LB. Recommendations of the panel on cost effectiveness in health and medicine. JAMA 1996;276:1253–8. 15.Siegel JE, Weinstein MC, Russell LB, Gold MR. Recommendations for reporting cost-effectiveness analyses. JAMA 1996;276:1339–41. 16.Echt DS, Liebson PR, Mitchell LB et al. Mortality and morbidity in patients receiving encainide, flecainide, or placebo: the Cardiac Arrhythmia Suppression Trial. N Engl J Med 1991;324:781–8. 17.Mark DB, Hlatky MA, Califf RM et al. Cost effectiveness of thrombolytic therapy with tissue plasminogen activator as compared with streptokinase for acute myocardial infarction. N Engl J Med 1995;332:1418–24. 18.Yusuf S, Zucker D, Peduzzi P et al. Effect of coronary artery bypass graft surgery on survival: overview of 10-year results from randomized trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration. Lancet 1994;344:563–70.

55

7

Introduction to decision analysis Kevin A Schulman, Henry A Glick, Allan S Detsky

The concept of evidence-based medicine challenges physicians to improve their use of the medical literature to guide their decision making in specific clinical settings. The concept is discussed extensively throughout this book. However, there are circumstances in which clinical trials do not address all of the issues of interest to a clinician. This may be because the trials do not compare the risks and benefits of all relevant treatment alternatives, or because the trials lack important data on the outcomes and costs of therapy. In these cases, researchers and clinicians are developing analytical strategies to improve their ability to synthesize the available information from the clinical literature and to help resolve these unanswered questions. One method of achieving this synthesis is the use of decision analysis, a set of mathematical strategies for aggregating information, making issues related to clinical decisions explicit, and solving for an optimal strategy under the constraints of the analysis. This decision analysis is a framework that can be used in the analysis of clinical problems as well as in economic analysis (see Chapter 6). Decision analysis has been available to cardiologists for over 20 years.1–3 In that time, the techniques have become more sophisticated and begun to address a broader range of questions.4–6 The goal of this chapter is to introduce the reader to some of the basic concepts of decision analysis and to review its use in the cardiovascular literature. For more specific information about the concepts or methods of decision analysis, the reader is referred to several excellent summary articles2–11 or to one of the major texts in the field.12–14

Steps in decision analysis

Examples of decision analysis

Step 2: Draw the tree

In this section we present two examples of the use of decision analysis, a clinical example and an economic example. These are provided to demonstrate the steps involved in developing a decision analysis. As will be clearly demonstrated, decision analytic models must simplify reality in order to structure the problem and analysis. Although our examples are extremely simplified to illustrate the steps involved in decision analysis, many models in the clinical literature offer more complex depictions of clinical reality.16–22,26,27,36,40,41,53–58

Based on these facts, we can graphically depict the issue using a decision tree (Figure 7.1). The tree is displayed so that the decision of interest is on the left side of the diagram, while the strategies to be compared are in the center, and the outcomes of those strategies are on the right. There are several pieces of information included in this simple figure. In the decision tree a choice is represented by a square, also called a “decision node”. In this example, the decision node represents a choice between warfarin and aspirin. Once a decision is made, patients experience the potential

56

1. 2. 3. 4.

Identify the strategic options. Draw the tree (structure of outcomes). Determine the probabilities. Determine the relevant outcome measures (effects, utility, survival, costs). 5. Evaluate the tree. 6. Make a structured analysis of the problem. 7. Develop a conclusion. A comparative clinical analysis: warfarin v aspirin for atrial fibrillation For patients with non-valvular atrial fibrillation, both warfarin and aspirin have been shown to reduce the risk of stroke.28–35 However, the effectiveness and side effects of these two treatments can vary substantially. As there has been no randomized trial of aspirin and warfarin for stroke prevention, decision analysis has been used to identify the clinical outcomes resulting from treatment with each medication.36 Step 1: Identify the strategic options In terms of therapeutic benefit, patients who receive either warfarin or aspirin experience a reduction in the risk of stroke. However, patients receiving these therapies also experience risk of bleeding complications. Both stroke and hemorrhage can be either fatal or non-fatal.

Introduction to decision analysis

Stroke (p1) Hemorrhage w/o stroke

Warfarin (p2) 1 – (p1 + p2)

No stroke or hemorrhage

Outcomes

Outcomes

Outcomes

Atrial fibrillation patients

Stroke 1)

(p1 Aspirin

Hemorrhage w/o stroke (p21) 1 – (p11 + p21)

No stroke or hemorrhage

Outcomes

Outcomes

Outcomes

Figure 7.1 A decision tree for a comparative clinical analysis: warfarin v aspirin for atrial fibrillation

for different clinical events (stroke or hemorrhage). These decisions and their subsequent clinical events are represented by lines or “pathways” running through the tree diagram. Figure 7.1 contains six possible pathways: warfarin with stroke; warfarin with neither stroke nor hemorrhage; warfarin with hemorrhage but without stroke; aspirin with stroke; aspirin with neither stroke nor hemorrhage; and aspirin with hemorrhage but without stroke. After the initial treatment decision between warfarin and aspirin has been made, subsequent outcomes occur with a defined probability such that all of the potential treatment outcomes are represented by the treatment pathways. The individual patient’s achievement of a given treatment outcome (for example stroke or no stroke) is not a decision: it is, instead, a chance occurrence, where the “chance” event is represented by a circle in the decision tree. The final treatment outcomes for each pathway are represented by triangles. These figures represent the outcomes of each treatment strategy. One, two or more outcomes can be expressed for each pathway (survival, quality adjusted survival, or costs). Step 3: Determine the probabilities Once a tree has been developed to depict a clinical problem, the next step is to begin to develop the data required to complete the analysis. In our example, we must identify the probability of stroke for patients in our two treatment categories and identify the potential risk of bleeding complications associated with each therapy. Rates of stroke without therapy, outcomes of stroke, and stroke risk reduction with prophylaxis with aspirin or warfarin can be estimated from clinical trials or epidemiologic studies.36 Rates of major hemorrhage associated with warfarin and aspirin therapy, and the outcomes of such an event, can be estimated in the same fashion.36 However, in

pooling these various data sources, investigators are left with a degree of uncertainty about these estimates. Sensitivity analysis, a method for assessing the impact of uncertainty in data analysis of clinical problems, will be discussed later, but it is an integral component of most well constructed decision analytic models. Step 4: Determine the relevant outcome measures (effects, utility, survival, costs) For this analysis, quality adjusted survival will be the primary outcome measure. Other possible outcome measures include event-free survival or simple survival. Analysis of quality adjusted survival uses estimates of patient preferences for a variety of possible health states for patients with stroke. Patient preferences are a measure of health-related quality of life, or utility, as defined on a 0–1 scale, in which 0·0 represents the worst imaginable health state and 1·0 denotes the best imaginable health state. Quality adjusted survival is the product of the expected survival of patients and their preferences for the different health states resulting from a stroke or hemorrhage. These data can be estimated from expert opinion, as reported in the medical literature, or derived from patient interviews.37 Calculation of quality adjusted life years (QALYs) is described in greater detail elsewhere.9 Step 5: Evaluate the tree Once data have been compiled for the specified model parameters, the next step is to analyze the tree. This requires the calculation of the expected value for each pathway of the tree. For both warfarin and aspirin therapies, the expected value of the outcome (effects, utility, survival or costs) is a weighted average of all possible treatment outcomes. This weighted average is calculated as the product of 57


the value of each terminal node and the probabilities of the occurrence of that node (the product of the probabilities of achieving that node). The value of each node is then summed to result in the weighted average value for the treatment (for example, the outcome for warfarin would be the weighted average of the products of the probability of stroke while taking warfarin and the outcome for stroke (P1 O1), the probability of hemorrhage without stroke while taking warfarin and the outcome for hemorrhage without stroke (P 2 O2), and the probability of no stroke or hemorrhage and the outcome for no stroke or hemorrhage {(1 [P1 P 2]) O3}). A more complicated decision analysis proceeds in stepwise fashion for each set of probabilities and outcomes. This is called folding back the tree. The net result is an assessment of the outcomes for the two treatments, warfarin and aspirin. Other techniques can be used to solve more complicated problems, for which there are many branches of each tree – for example, when the risk of stroke or hemorrhage is related to the duration of treatment. (These methods are based on the probability of moving between health states over time. Analysis can also be based on “state transition models” or Markov models.) For clinical analyses, decision trees allow an incremental analysis of the treatment benefits of one medical therapy compared to another. They are used to compare the expected utility for each branch of the tree to pick the best treatment option. The best option is the one with the highest value in terms of clinical effects (survival or utility) or the one with the smallest value in terms of cost. An incremental analysis assesses the additional benefits gained from one treatment and, thus, differs from a calculation of the absolute benefit of a treatment. Step 6: Structured analysis of the problem Finally, the primary analysis having been completed, investigators should examine uncertainty in their estimates using a technique called sensitivity analysis. By recognizing that a decision tree can suffer from uncertainty in the probability of each treatment strategy, investigators can ask how the results might change were the possibilities of stroke or hemorrhage to increase or decrease by 10% for each treatment arm. In a sensitivity analysis, the investigator recalculates the results of the model to address the robustness of the analysis to changes in the model specification.

and the differences in hemorrhage resulting from the prophylactic treatment. The analysis would end with an estimate of the quality adjusted survival resulting from each treatment strategy. The results could reveal that warfarin is superior to aspirin, that aspirin is superior to warfarin, that the treatments are comparable, or that there is not enough information from which to draw a firm conclusion. The analysis would also address how sensitive the analysis was to differing model parameters. This could help define areas for further research to resolve outstanding issues in the clinical assessment. A cost effectiveness analysis: implantable cardiac defibrillators At present there is a great deal of debate about the most appropriate treatment of patients with arrhythmias, especially about whether implantable cardiac defibrillators (ICDs) will reduce cost and mortality for high-risk patients. Early clinical trial results address mortality issues related to the use of ICDs in high-risk populations.38 However, there remains a great deal of concern about the findings of the study and the robustness of its results.39 Although decision analysis cannot answer the clinical questions regarding ICD use, these techniques have been used to model the costs and effects of ICD insertion to estimate the potential cost effectiveness of this therapy, given current estimates of ICD clinical effectiveness.40,41 To understand the decision analysis approach to this question, we will review the clinical issues and then build a decision analytic model to formalize the question. High-risk patients experience an increased incidence of sudden cardiac death.42 One new technology, the ICD, has been proposed as a means of reducing the incidence of sudden death in cardiac patients.38,43–48 Patients who choose to receive this therapy must undergo a surgical procedure and maintain the device over the remainder of their lifetime. Step 1: Identify the strategic options In terms of treatment benefit, patients who receive an ICD have the potential for a different survival probability than patients who do not receive an ICD. From a cost perspective, patients receiving an ICD bear the additional cost of the device itself, as well as the future costs of maintaining it. Step 2: Draw the tree

Step 7: Conclusion This decision analysis was structured to compare the outcomes of two strategies for the treatment of stroke prophylaxis – warfarin and aspirin. Such an analysis allows for an assessment of the clinical benefits of the two strategies, incorporating both the differences in risk reduction of stroke 58

Based on this discussion, we can graphically depict the issue using a decision tree (Figure 7.2). There are four possible pathways in this figure: ICD with sudden death; ICD without sudden death; no ICD with sudden death; and no ICD without sudden death. In this simple model we consider only two health states: sudden death and no sudden death.


Sudden death

Outcomes

(P) ICD (1 – p)

No sudden death

Outcomes

High-risk patients Sudden death (p1)

Outcomes

No ICD (1 – p1)

No sudden death

Outcomes

Figure 7.2 Decision tree for a cost effectiveness analysis: ICD to reduce incidence of sudden death in cardiac patients

Step 3: Determine the probabilities Estimates of the possibility of sudden cardiac death for highrisk patients are available in the medical literature and in trials of ICDs.43–49 Estimates of the probability of sudden death for patients receiving an ICD are available from the MADIT Study38 or may be estimated based on clinical trial protocols for expected treatment benefits.41 The quality of the evidence from these data sources can vary. Data from the literature on non-ICD patients, the probability of sudden death without an ICD, come from observational studies, whereas data on ICD patients come from a controversial randomized controlled trial. Thus, there is some uncertainty about these estimates.10 Step 4: Determine the relevant outcome measures (effects, utility, survival, costs) Treatment benefits can be expressed in terms of survival (years of life gained) or in terms of quality adjusted survival (QALYs). Calculation of these benefits proceeds as outlined in the stroke example. Estimates of treatment costs often must be developed from primary sources (for example, hospital accounting departments), from standard price lists for specific costs,50 from literature reviews, or from expert opinion. Costs included in these models can include direct medical costs (the costs of medical care, such as hospital or physician costs), direct non-medical costs (the costs patients incur in receiving medical care services, such as the cost of transportation to a physician’s office), indirect costs (the costs of morbidity or mortality related to disease), or intangible costs (the costs of pain and suffering related to disease).51,52 Step 5: Evaluate the tree Once the data are available for all of these model parameters, the next step is to analyze the tree. For economic

analyses, decision trees allow an incremental analysis of the treatment costs and benefits of one medical therapy compared to another in a cost effectiveness analysis. The incremental cost effectiveness of therapy A compared to therapy B is defined by the following formula: Cost effectiveness of treatment A

(CostACostB) (EffectsAEffectsB)

where CostA is the cost of treatment A, CostB is the cost of treatment B, EffectsA are the effects of treatment A, and EffectsB are the effects of treatment B.52 Decision trees may also allow enumeration of the costs and consequences of different treatments without comparing the costs and effects of treatment in a cost effectiveness ratio. Step 6: Structured analysis of the problem Sensitivity analysis would be conducted to assess the impact of uncertain values on the model. For example, because there was uncertainty in the probability of each treatment strategy, how would the results change if the possibilities of sudden death were increased or decreased by 10% for each treatment arm? Similarly, how would the results differ if ICD costs were increased or decreased by 10%? In a sensitivity analysis, the investigator recalculates the results of the model to address the robustness of the analysis to changes in the model specification. Step 7: Conclusion This decision analysis was structured to assess the cost effectiveness of a new therapy for the treatment of patients at high risk for sudden cardiac death. It would conclude with an estimate of the incremental effects of ICD therapy in years of life gained per patient, the incremental costs of ICD treatment per patient, and an estimate of the cost effectiveness of 59


ICD therapy for patients evaluated in the model. The paper would also address how sensitive the analysis was to different model parameters. This sensitivity analysis could help define areas for further research to resolve outstanding issues in the clinical assessment. Applications of decision analysis to cardiology The above examples offer a simplified explanation of some of the basic components of decision analysis. They also illustrate the issues that must be addressed before using the results of a decision analysis to guide clinical decision making. As when reviewing clinical trials, clinicians must assess whether the population considered in the decision analysis model is relevant to their own population. The reader must consider the strength of the evidence available to the investigator in developing the model to understand the strength of the recommendations resulting from the model. This includes not only whether the evidence was based on randomized controlled trials or on observational studies, but also whether the original studies included detailed information required by the model (for example, in the stroke analysis, whether the clinical studies reported both hemorrhage

and stroke rates for the study’s patients). Finally, the reader should consider the model used by the investigator to determine whether it was constructed appropriately and considered all relevant comparisons.1,5 The models below that use lower-quality data or evidence should be considered exploratory analyses, not definitive evidence. As such, they also should be interpreted as potential rationale for future studies. Likewise, decision models that project results to new time periods, new populations and new interventions – even those that use A1a-grade evidence – should be viewed only as exploratory analyses. Decision analysis has been used extensively in cardiology over the past several years (Table 7.1). These examples include articles from a MEDLINE search of decision analysis and cardiology from 1993 to 2001. Issues addressed have included the use of specific technologies, such as ICDs for patients at risk for sudden death, as well as specific diagnostic or pharmacologic products for defined populations of patients (for example, treatment of high blood cholesterol), and the assessment of patient management strategies for defined populations of patients (for example the selection of patients for placement on a cardiac transplant list). Each of the analyses listed in Table 7.1 will be reviewed in this section.

Table 7.1 Use of decision analysis in the cardiovascular literature Clinical issue

Efficacy data

Cost data

Sensitivity analysis

Source

Evidence grade*

Observational study; utility not assessed

Hospital charges; literature review for resource use data

Yes

Kupersmith et al 16

B4

Outpatient ICD placement

Literature review for survival and utility estimates

Hospital and claims data; literature review for resource use data

Yes

Owens et al 41

A1c

Treatment strategies for WPW syndrome

Literature review; expert opinion; authors’ estimates for utility data

Cost-accounting data for 13 patients at one study center

Yes

Hogenhuis et al17

B4

Literature review; authors’ estimates for utility data; patient survey for intangible cost estimates

Resource use from a clinical trial; literature review; costs from Canadian hospital

Yes

Barrett et al 18

A1c

Clinical trial data; utility not assessed

Resource use from clinical trial; costs from hospitals in Sweden; employment status from clinical trial

Yes

Johannesson et al19

A1a

New technologies Inpatient ICD placement

Specific products Low v high-osmolality contrast media

Simvastatin, high cholesterol

60


Table 7.1

Continued

Clinical issue

Efficacy data

Cost data


Source

Evidence grade*

Pravastatin, high cholesterol

Clinical trial data from 2 studies; utility not assessed

Literature review and expert opinion for resource use data; costs from aggregate US hospital data

Yes

Ashraf et al 20

A1c

Captopril, acute MI

Clinical trial data; utility data from 82 patients

Resource use from subset of study patients; costs from US Medicare reimbursement rates

Yes

Tsevat et al 21

A1a

Estrogen replacement

Literature review; utility not assessed

Not assessed

Yes

Zubialde et al 22

B2

Streptokinase v tPA, suspected MI

Literature review, including utility data

Resource use estimated; drug and hospital cost data from Ireland

Yes

Kellett et al 26

A1a

Warfarin v quinidine v amiodarone, acute atrial fibrillation

Literature review and expert opinion, including utility data

Not assessed

Yes

Disch et al 27

A1a

Warfarin v aspirin, stroke prophylaxis

Literature review; utility data from study of 74 patients

Resource use estimated; costs from literature review, Medicare data, and survey of pharmacies

Yes

Gage et al 36

A1a

Preoperative coronary angiography and revascularization, non-cardiac vascular surgery

Literature review; utility not assessed

Literature review

Yes

Mason et al 55

A1a

CCU admission

Cohort study; utility not assessed

Hospital charges from the cohort adjusted to costs

Yes

Tosteson et al 56

B4

Emergency medical services

Literature review; Analysis of existing utility not assessed EMS program in Canada

Yes

Nichol et al 57

B4

Cardiac transplantation selection

Transplant registries

Yes

Stevenson et al 58

B4

Wong et al 59

B4

Treatment strategies

Not assessed

Aortic valve replacement Medical v surgical therapy for chronic stable angina

Expert guidelines, Not assessed randomized trials, and meta-analyses; utility not assessed

Yes

Kwok et al 61

A1a

Strategies for hypoplastic left heart syndrome

Literature review and data from 231 patients; utility not assessed

Yes

Jenkins et al 62

B3

Not assessed

61


Table 7.1

Continued

Clinical issue

Efficacy data

Cost data


Source

Evidence grade*

Electrocardiogram algorithm to predict myocardial infarction

Retrospective cohort study; utility not assessed

Not assessed

Yes

Shlipak et al 63

B2

*Evidence grades for decision analysis are complicated by the many different sources of data used in constructing the analysis. Evidence grades here are based on the data for the most important component of the analysis for the clinical portion of the decision tree. Where sources of evidence for the analysis were from a variety of sources, two grades were assigned to reflect the differing quality of data available for the analysis (see Owens et al.41 for an example of grades of evidence for data incorporated into a decision analysis). Grade A: Decision trees with the primary effect estimate from a large, high-quality study (a randomized controlled trial with more than 500 patients), or decision trees with a formal meta-analysis for the primary effect estimate. Grade B: Decision trees with the primary effect estimate based on literature review but without a formal meta-analysis for primary effect estimate; includes evidence from case series and randomized controlled trials with fewer than 500 patients. Grade C: Decision trees with the primary effect estimate based on expert opinion.

Decision analysis evaluating new technologies ICD placement Over the past several years, investigators have attempted to calculate the cost effectiveness of the ICD in patients at high risk for sudden cardiac death. Recent evidence from the Antiarrhythmics versus Implantable Defibrillators Trial indicates a decrease of 27% in 2 year mortality with ICD.15 Kupersmith et al 16 assessed ICD placement on an inpatient basis for patients with and without prior electrophysiologic (EP) studies. Grade A1c The investigators assumed an 84% improvement in life expectancy for patients undergoing ICD therapy based on a case series of 218 non-randomized patients who received an ICD when it was assumed that the patients would have died at the time of the first event (first shock or death). In this analysis, ICD patients had a mean life expectancy of 3·78 years, whereas EP-guided drug therapy patients had a mean life expectancy of 2·06 years. Total charges for these treatments were $146 797 for ICD patients and $93 340 for the EP-guided patients. The investigators found that the cost of ICD placement, including the cost of the device and the hospitalization, would range between $27 200 and $44 000 per year of life saved. The investigators conducted an extensive sensitivity analysis around their cost data and around the period of replacement of the ICD generator. They found that the cost effectiveness of the therapy was sensitive to the magnitude of the clinical benefit of the therapy (this included the efficacy of the therapy, as well as the estimated life expectancy for the underlying population, as represented by ejection fraction). The model was less sensitive to the cost of ICD therapy. The authors concluded that ICD use was economically attractive, 62

especially using endocardial lead placement (based on preliminary estimates of the cost of this new procedure).* Owens et al 41 assessed ICD implantation on an outpatient basis using a decision analytic model. In this analysis, the investigators modeled the potential cost effectiveness of therapy, assuming in their principal analysis that the ICD led to a 20–40% reduction in mortality. Grade A1c The investigators found that the cost of patients receiving ICD therapy would be $88 400, and the cost of patients receiving amiodarone therapy alone would be $51 000. For high-risk

* There are four possible outcomes of a cost effectiveness analysis: (1) the intervention will save money and be more effective than the comparison; (2) the intervention will cost money and be more effective than the comparison; (3) the intervention will save money and be less effective than the comparison; and (4) the intervention will cost money and be less effective than the comparison.51 The first outcome is the most preferred, and the intervention will always be adopted. The last outcome is never preferred, and the intervention will never be adopted. The second and third outcomes may be preferred at times, and the interventions may be adopted, depending on the relationship between the costs and effects of the intervention (the cost effectiveness ratio). The second outcome may be adopted if the intervention yields a great enough benefit for the additional cost (in the USA, an economically attractive intervention may be one that costs less than $50 000 per year of life gained, whereas some Canadian authors have suggested that therapies that cost less than CDN$100 000 might be economically attractive).35,53 The third outcome may be adopted if the intervention yields a small enough reduction in outcomes for the reduction in cost (for example, the same Canadian authors suggested an economically attractive intervention might be one that saves more than CDN$100 000 per year of life forgone).35,52


patients, the investigators reported that ICD patients would have an estimated survival of 4·18 QALYs, whereas patients treated with amiodarone alone could expect a survival of 3·68 QALYs. Investigators found that the cost effectiveness of therapy ranged from $37 300 per QALY saved for high-risk patients, assuming a 40% reduction in mortality for patients treated with the ICD compared to those treated with amiodarone alone, to $138 900 QALYs saved for intermediate-risk patients and assuming a 20% reduction in mortality for patients treated with the ICD compared to amiodarone alone. They concluded that the use of an ICD will not be economically attractive unless all-cause mortality is reduced by 30% or more compared to amiodarone. Alternative therapies for WPW syndrome Hogenhuis et al 17 determined which of five management strategies should be used for the treatment of patients with Wolff–Parkinson–White (WPW) syndrome: observation, observation until cardiac arrest-driven therapy, initial drug therapy guided by non-invasive monitoring, initial radiofrequency ablation, and initial surgical ablation. The model included the risks of cardiac arrest, arrhythmia, drug adverse effects, procedure-related complications and mortality, and assumed that radiofrequency ablation had an overall efficacy of 92% in preventing cardiac arrest and arrhythmia. Grade B4 For survivors of a cardiac arrest, radiofrequency ablation offered additional survival at reduced cost compared to all other treatment strategies. For patients with arrhythmia without hemodynamic compromise, radiofrequency ablation resulted in a cost of $6 600 per QALY gained in 20 year old patients and $19 000 per QALY gained in 60 year old patients without hemodynamic compromise. For asymptomatic patients, radiofrequency ablation costs from $33 000 per QALY gained in 20 year old patients to $540 000 per QALY gained for 60 year old patients. The authors conclude that their analysis supports the practice of radiofrequency ablation in patients with WPW syndrome who survive cardiac arrest. For asymptomatic patients, however, the analysis supports the current practice of mere observation, given that radiofrequency ablation was economically unattractive in this population of patients. Decision analysis in the evaluation of specific products Decision analysis has been used extensively in the evaluation of specific clinical products, including contrast media and pharmaceutical products. Contrast media Grade A1c Barrett et al 18 developed a decision analytic model to assess the economic impact of low- and high-osmolality contrast media for cardiac angiography. Investigators

assumed that low-osmolality contrast media reduced the risk of myocardial infarction and stroke. Reduction in the risk of specific clinical events with low-osmolality contrast media was assumed to be 0% in fatal events, 25% in severe events, 80% in moderate events and 10% in minor events. The investigators found that the incremental cost per QALY gained with these media was $17 264 in high-risk patients and $47 874 in low-risk patients for a third-party payer. From a societal perspective, the corresponding costs are $649 and $35 509. The authors report that these estimates were sensitive to cost of the contrast media and the total cost of contrast media used per patient. The authors also suggest that the model is extremely sensitive to changes in assumptions regarding the efficacy of low-osmolality contrast media for the prevention of severe reactions. To allow the reader to better understand the inputs of this model, the authors include a cost–consequence analysis of the program as a separate presentation in the results. The authors concluded that, in the context of restricted budgets, limiting the use of low-osmolality contrast media to high-risk patients is justifiable. The recommendation to limit use of this medium was also justified by the lack of clinical evidence that lowosmolality contrast media prevent severe or fatal reactions. Cholesterol reduction Several authors have used decision analysis to investigate the cost effectiveness of therapies designed to reduce high blood cholesterol.54 Two recent studies use clinical trial data to assess the cost effectiveness of cholesterol reduction in secondary prevention of coronary artery disease. Johanneson et al19 developed an analysis based on the Scandinavian Simvastatin Survival Study, which reported that, in patients with pre-existing coronary disease, reduction in blood cholesterol resulted in a 30% reduction in overall mortality based on a median follow-up of 5·4 years. Grade A1a The authors modeled the effects of 5 years of cholesterolreducing therapy on patients’ outcomes, using a model based on data reported from the trial. The costs of therapy were based on the assumption that the use of cholesterolreducing agents would not entail any additional costs for patients with pre-existing coronary disease other than the cost of medication itself, and then used data on hospitalizations to estimate the direct medical costs incurred for the treatment of cardiovascular disease. Interestingly, this model also included the indirect costs of medical care based on the employment status of patients in the trial. The investigators found that simvastatin treatment for 5 years in 59 year old patients with a history of heart disease and a pretreatment total cholesterol level of 261 mg/dl would have a net cost of $1 524, with 0·28 years of life gained, resulting in a cost per year of life gained of $5 400 for men, and a net cost of $1 685 with 0·61 years of life gained, resulting in a cost per year of life 63


gained of $10 500 for women. An analysis that included direct and indirect costs showed that cholesterol reduction leads to an additional $1 065 decrease in associated morbidity cost for men and an $876 reduction in associated morbidity cost for women. The analysis was somewhat sensitive to baseline cholesterol level and patient age at the initiation of treatment, to follow-up and screening costs and to the price of simvastatin. However, treatment remained economically attractive in all of these analyses. The model was somewhat sensitive to reduction in cardiovascular risk and the risk of mortality after coronary events. The authors concluded that in patients with coronary artery disease, simvastatin therapy is economically attractive among both men and women at the ages and cholesterol levels studied. Ashraf et al 20 assessed the cost effectiveness of cholesterol reduction based on 3 year data from the Pravastatin Limitation of Atherosclerosis in the Coronary Arteries (PLAC I) and Pravastatin, Lipids and Atherosclerosis in the Carotids (PLAC II) studies. Grade A1c These trials reported no statistically significant decrease in all-cause mortality, but did report a decrease in the number of coronary events in men in the group receiving drug therapy to reduce high blood cholesterol. Therapy was estimated using a Markov model based on data from the Framingham Heart Study to estimate subsequent annual morbidity and mortality rates for patients with non-fatal myocardial infarction. Costs of therapy were based on the costs of drug therapy, and hospitalization costs were derived from the cost of treatment of myocardial infarction and from expert opinion on the frequency of medical events. Investigators found that cost per year of life saved due to secondary prevention was sensitive to a number of risk factors, but ranged from $7124 per year of life saved for a male patient with three risk factors, to $12 665 per year of life saved for a male patient with one risk factor. The model was sensitive to assumptions about efficacy of therapy and cost of services. It was also sensitive to patient characteristics, such as the number of risk factors of patients receiving secondary prevention. The authors conclude that pravastatin is economically attractive compared to other widely accepted medical interventions. A potentially serious limitation of the Ashraf et al 20 analysis is its strategy of deriving costs for 3 years while projecting the effects over 10 years. Specifically, the authors project years of life saved by avoiding events in the first 3 years over the next 7 years. This potentially problematic practice of generating a differential time horizon should be avoided. Postmyocardial infarction treatment Tsevat et al 21 used decision analysis to assess the cost effectiveness of captopril therapy after acute myocardial infarction (MI). Grade A1a In this paper, the investigators used data from the Survival and Ventricular Enlargement (SAVE) trial, which demonstrated that captopril therapy reduced 64

mortality in patients who survived MI. The effectiveness of therapy was modeled using a decision analytic model based on all-cause mortality within the clinical trial observation period and the projected clinical benefits over a patient’s lifetime. This paper also incorporated data on quality of life from a subset of patients in the SAVE trial. Cost estimates for the model were based on a subset of 123 study patients for whom hospital data were obtained for all hospitalizations in the subset. The investigators used two projection methods, a limited-benefit model and a persistent-benefit model. The limited-benefit model was more conservative in that it assumed similar annual mortality rates between captopril and control patients beyond the clinical trial period. This analysis resulted in an estimated cost effectiveness for captopril therapy ranging from $60 800 per QALY for 50 year old patients to $3 600 per QALY for 80 year old patients. The persistent-benefit model was more optimistic in that it assumed that the clinical benefits observed in the trial persisted throughout each patient’s lifetime. In this analysis, the cost effectiveness ratios were similar to those in the limitedbenefit model for patients aged 60–80 years, but they were substantially better for 50 year old patients. In the sensitivity analysis, the models were most sensitive to the annual cost of captopril therapy. In addition, the persistent-benefit model appeared to be more “stable” than the limited-benefit analysis. That is, when the benefits persist, there are few changes to the values of other variables that would affect the resulting cost effectiveness ratios (owing to the magnitude of the benefit), whereas if the benefits do not persist, variations in other variables do have an effect. The investigators concluded that angiotensin converting enzyme inhibitor therapy with captopril was not only effective in improving survival after MI, but also moderately economically attractive. Hormone replacement therapy Zubialde et al 22 used a decision analytic model to assess gains in life expectancy resulting from the use of estrogen replacement therapy for postmenopausal women. Efficacy data for this analysis were obtained from a review of the literature which suggested that risk reduction with estrogen therapy for coronary artery disease was between 40% and 50%. Grade B2 The model did not assume an increased incidence in breast cancer in the principal analysis, but it did include an increased incidence of endometrial cancer. Results of the analysis suggested that the benefit of estrogen and progesterone therapy in average-risk women aged 50 years at the time of therapy initiation was 0·86 years, with a range of 0·41–1·19 years, whereas therapy in average-risk women aged 65 years at the time of therapy initiation was 0·47 years, with a range of 0·21–0·66 years. The authors reported that the benefits of estrogen and progesterone therapy were similar to the gains from cholesterol reduction to


200 mg/dl and smoking cessation. The authors concluded that significant potential benefits in life expectancy in coronary artery disease reduction, combined with the osteoporosis prevention in symptom relief, would point to greater emphasis on postmenopausal estrogen use in appropriate patients. Since the report by Zubialde et al 22 hormone replacement therapy has undergone additional study. A growing body of literature suggests that its predicted effects have not been fully realized,23,24 and the results of a recent polymorphism study have further complicated matters.25 It bears repeating here that the reliability of a decision analysis is related directly to the quality of the data on which the analysis is based. The Zubialde analysis was based on the best data of its time, but superior data from clinical trials have since called the findings into question. Thrombolytic therapy Kellett et al 26 presented a paper on the use of thrombolytic therapy for patients with suspected MI. This assessed the use of two types of thrombolytic therapy, streptokinase and accelerated tissue plasminogen activator (tPA), on patients with suspected MI. Grade A1a The efficacy of the two therapies was based on reports from the medical literature. The authors assessed the clinical benefits of thrombolytic therapies for patients presenting with different likelihoods of MI, given their clinical and ECG findings, different age groups, and different probabilities of death given MI. Data on clinical efficacy for the two strategies were based on the GISSI-2, ISIS-3 and GUSTO trials. The authors suggested that, for patients with a 26% probability of MI (a group with chest pain and a history of coronary artery disease but a normal ECG), thrombolytic therapy would only be beneficial if the probability of death given an MI was 20% or greater. In contrast, for patients presenting with a probability of MI of 78% (chest pain plus ST or T wave changes), thrombolytic therapy would be beneficial for all patients except those over 80 years of age who had a probability of death given an MI of 2·5% or less. The authors conclude that, for a typical 60 year old man presenting 4 hours after the onset of symptoms with definite acute MI, treatment with streptokinase in addition to aspirin would gain 150 quality adjusted life days, whereas treatment with aspirin and accelerated tPA would result in 255 quality adjusted life days, compared to no thrombolytic therapy. Thrombolytic therapy is preferred over no thrombolytic therapy as long as the probability of stroke is less than 5% for streptokinase and 8% for accelerated tPA. The cost per QALY was estimated based on the probability of acute MI, the extra days of quality adjusted life, and the probability of death given an MI. The analysis was sensitive to estimates of efficacy for both streptokinase and accelerated tPA, as well as the probability of death given thrombolytic therapy. The authors conclude that decision analysis can be a useful bedside tool to guide thrombolytic

therapy. It is important to bear in mind, however, that this decision model has not been tested on actual patients. Management of atrial fibrillation Dirsch et al 27 developed a decision analytic model to assess the outcomes of four treatment strategies for patients with acute atrial fibrillation undergoing cardioversion: warfarin therapy, quinidine therapy and low-dose amiodarone therapy. Grade A1a Efficacy was based on a review of the literature, including randomized controlled trials, observational studies, and expert clinical opinion when necessary. Investigators found that all four treatment strategies differed by 0·2 QALYs over patients’ lifetimes, with 4·55 expected QALYs for patients who undergo no treatment after cardioversion and 4·75 expected QALYs for patients who undergo cardioversion with amiodarone. Use of warfarin and quinidine therapies yielded expected quality adjusted life benefits between amiodarone and no treatment. The model was sensitive to the annual rate of bleeding on warfarin, the annual rate of stroke for patients on warfarin, the annual rate of stroke for patients with atrial fibrillation, the decrement in quality of life associated with taking warfarin, and the excess mortality of quinidine and amiodarone. The authors conclude that cardioversion followed by low-dose amiodarone to maintain normal sinus rhythm appears to be a relatively safe and effective treatment for a hypothetical cohort of patients with atrial fibrillation. Prophylaxis of stroke Gage et al 36 developed a decision analytic model to assess the cost effectiveness of warfarin and aspirin treatment for prophylaxis of stroke in patients with non-valvular atrial fibrillation. The clinical efficacy of the treatment strategies was obtained from the published literature. Grade A1a The quality-of-life estimates for this study were obtained by interviewing patients with atrial fibrillation. Costs were also estimated from a literature review and from a survey of national pharmacies and laboratories. The authors found that, for patients with non-valvular atrial fibrillation and no additional risk factors for stroke, warfarin would minimally affect quality adjusted survival but increase costs significantly. For patients with non-valvular atrial fibrillation and one additional risk factor, warfarin therapy resulted in a cost of $8 000 per QALY saved compared to aspirin. The model was most sensitive to the rate of stroke if no therapy was prescribed, the effectiveness of aspirin, the rates of major hemorrhage, and the disutility of taking warfarin. The authors conclude that treatment with warfarin is economically attractive (has a low cost effectiveness ratio) in patients with non-valvular atrial fibrillation and one or more additional risk factors for stroke. However, in patients with non-valvular atrial fibrillation without other risk factors for stroke, the use 65


of warfarin instead of aspirin would add significantly to costs with minimal additional clinical benefit. Preoperative cardiac revascularization Mason et al 55 developed an analysis to determine whether preoperative coronary angiography and revascularization improved short-term outcomes in patients undergoing noncardiac vascular surgery with three strategies. Grade A1a The first was to proceed directly to vascular surgery; the second was to perform coronary angiography followed by selective coronary revascularization prior to surgery and to cancel vascular surgery in patients with severe inoperable coronary disease; and the third was to perform coronary angiography followed by selective coronary revascularization, and to perform vascular surgery in patients with inoperable coronary artery disease. The literature was scrutinized for data on the efficacy of all three strategies. The authors found that proceeding directly to vascular surgery led to a lower morbidity and cost in the base-case analysis. The coronary angiography strategy led to a higher mortality of vascular surgery in patients with inoperable coronary disease, but to a lower mortality in operable patients who did not proceed to vascular surgery. The model was sensitive to the surgical mortality rates for both catheterization and the vascular surgical procedure. The authors concluded that decision analysis indicates that vascular surgery without preoperative angiography generally leads to better outcomes, and that preoperative coronary angiography should be reserved for patients whose estimated mortality for vascular surgery is substantially higher than average. Use of decision analysis in treatment strategies CCU admission Tosteson et al 56 used a decision analytic model to identify cost effective guidelines for admission to a coronary care unit (CCU) for uncomplicated patients without other indications for intensive care. The probabilities of death, and minor, major and life-threatening complications were based on 12 139 emergency department patients who were enrolled in a multicenter chest pain study. Cost data were available from a subset of patients in the study admitted to one study center. Under the assumption that there is a 15% relative increase in mortality when patients with acute MI are admitted to the intermediate care unit instead of an intensive CCU, the authors found that costs per year of life saved for triage to the CCU varied markedly depending on the age of the patient and the probability of MI. For 55–64 year old patients with an emergency department probability of infarction of 1%, the cost per year of life saved was $1·4 million; but when the probability of infarction was 99%, the cost per year of life saved was $15 000. Admission to the intensive care unit was generally more costly for 66

younger patients, and use of the CCU had a cost effectiveness ratio of less than $50 000 per year of life saved when the initial probability of acute MI was greater than 57% among patients 30–44 years of age, and greater than 21% among patients 65–74 years of age. The model was sensitive to the reduction of mortality associated with the use of the intensive care unit and to the costs of the intensive care unit. The authors conclude that the CCU should generally be reserved for patients with a moderate or high probability of acute MI, unless they need intensive care for other reasons. Emergency medical services Nichol et al 57 used a decision analytic model to assess the cost effectiveness of potential improvements to emergency medical services (EMS) for patients with out-of-hospital cardiac arrest. Grade B4 The authors developed their analysis based on a review of the effectiveness of various emergency systems from an extensive meta-analysis, costing of each component of the EMS, and community characteristics and response times for EMS. The authors also modeled a onetier system versus a two-tier system. In the one-tier system the response team is trained in advanced life support, and in the two-tier system the first response team is trained in basic life support and the second in advanced life support. The authors found that the fixed cost of the first tier of a two-tier EMS system was $651 129 for Hamilton, Ontario, with estimates of survival of 5·2% in the one-tier system and 10·5% in the two-tier system. They found that a 1 minute reduction in response time improved survival by 0·4% in a one-tier system and by 0·7% in a two-tier system. The authors found that a change from a one-tier system to a twotier system would result in 0·19 QALYs saved and an incremental cost of $7 700 per patient, or a cost per QALY of $40 000. Improvement in a one-tier EMS system by the addition of more basic life support providers in the first tier would result in an incremental survival benefit of 0·40 QALYs, with an incremental cost of $2 400 or cost per QALY of $53 000. An improvement in response time in a one-tier system by the addition of more providers and ambulances would achieve an incremental survival benefit of 0·2 QALYs for a cost per QALY of $368 000. The authors performed an extensive sensitivity analysis based on a combination of the model’s parameters. They concluded that the most attractive options in terms of incremental cost effectiveness ratios for an EMS program would be improved response time in a twotier EMS system, or a change from a one-tier EMS system to a two-tier system. However, the authors were concerned about the poor quality of the data available for their analysis. Heart transplantation Stevenson et al 58 used a decision analytic model to determine optimal strategies for selecting patients for cardiac


transplantation. Grade B4 The authors developed a model based on data from cardiac transplantation databases. The decision analytic model was developed to determine the size and outcomes of the waiting list population, depending upon different strategies for listing heart transplant candidates. They found that if current practices continued all hearts would be transplanted to hospitalized candidates and newly listed urgent candidates, and 3700 outpatient transplant candidates would be listed with virtually no transplantation unless they deteriorated to an urgent status. A decrease in the upper age limit for transplantation to 55 years would reduce the number listed each month by 30%. If this strategy were to be adopted, the waiting list would reduce to one third its current size, with 50% of all hearts being available for outpatient candidates. The authors conclude that immediate provisions should be made to limit candidate listing and revise expectations to reflect the diminishing likelihood of transplantation for outpatient candidates. Surgery for aortic stenosis Wong et al 59 used decision analysis to assess whether to recommend cardiac surgery for elderly women with aortic stenosis. Grade B4 This analysis was based on a specific case of assessing the treatment choice for an 87 year old patient with severe aortic stenosis, three vessel coronary disease, depressed left ventricular function and moderately severe heart failure. Data for the analysis were based on the medical literature. Specific data elements included in the analysis were life expectancy with and without surgery for an octogenarian, morbidity and mortality associated with surgery, and quality of life with congestive heart failure. Sensitivity analysis assessed the sensitivity of the model to assumptions used in developing the analysis and assessed the impact of patients’ risk preferences regarding treatment choice. The authors also modeled valvuloplasty compared to surgery. They found that life expectancy with surgery (5·0 QALYs) was greater than that for medical therapy (1·1 QALYs). (These gains in life expectancy are substantial. Most interventions reported in the medical literature yield incremental gains in life expectancy from 0·167 to 1·2 years of life.60) In sensitivity analysis, surgery still had the highest life expectancy until mortality from the procedure was greater than 70%. Valvuloplasty was the best strategy if the patient was not the best candidate for surgery or, perhaps, in cases in which the perioperative mortality rate was greater than 50%. They concluded that even in the later decades of life, aortic valve surgery is substantially preferable to medical therapy. Treatment strategies for chronic stable angina Kwok et al 61 used a decision analytic model to simulate a randomized controlled trial of coronary artery bypass graft surgery versus medical therapy for chronic stable angina.

Grade A1a The authors developed a Markov model that incorporated current American College of Cardiology/ American Heart Association guidelines, baseline data from a meta-analysis of randomized trials of the two therapies, and risk reduction data from randomized trials and meta-analyses. The outcome measures of interest were 5 and 10 year mortality, as well as incidence of non-fatal myocardial infarction. The authors conducted a base-case analysis of the two therapies, which they supplemented with annual fixed transition probabilities to account for a steady linear increase in mortality observed in the meta-analysis. They also conducted two subgroup analyses, one to examine 5 year mortality and infarction rates for patients with triple vessel disease, the other to examine the same outcomes for patients with impaired left ventricular function. In the base-case and subgroup analyses, the authors found that both therapies increased overall and infarction-free survival. The relative advantage of surgery over medical therapy found in this study mirrored the findings of previous trials. One-way and multiway sensitivity analyses yielded absolute differences of less than 2% for overall and infarction-free survival rates, except that use of the upper limit of aspirin therapy’s relative reduction of myocardial infarction yielded a 3% increase in infarction-free survival among patients receiving medical therapy. The authors concluded that therapeutic advances have improved outcomes for both medical and surgical patients, as well as preserving the advantages of surgery, thereby confirming that the conclusions of previous bypass trials remain valid. Treatment strategies for hypoplastic left heart syndrome Jenkins et al 62 used a decision analytic model to determine the optimal treatment strategy for maximizing 1 year survival among patients with hypoplastic left heart syndrome. Grade B3 Using data from the literature and from a data set of 231 patients treated at four US surgical centers, the authors obtained probabilities for the following treatment strategies: complete staged surgery; stage 1 surgery as an interim to transplantation; patient listing, then stage 1 surgery if no donor is found within 1, 2 or 3 months; and patient listing without surgery until transplantation. The authors conducted one- and two-way sensitivity analyses on all probabilities in the decision tree to determine the values at which the optimal treatment strategy would change. In the base-case analysis, transplantation within 1 month emerged as the preferred strategy, followed by staged surgery if no donor is available after listing the patient for 1 month. These results were sensitive to several probability thresholds, including stage 1 and stage 2 mortality rates, the surgical center’s 3 month organ donation rate, and the transplantation mortality rate. Centers with high organ donation rates are best served by a strategy of listing without surgery 67


until transplantation. Those with low donation rates, however, should perform staged surgery. In two-way sensitivity analyses, the authors found that the highest 1 year survival rates were achieved with staged surgery; patient listing, then stage 1 surgery at 1 month; and listing without surgery until transplantation. The authors concluded that each surgical center can determine its optimal treatment strategy with an algorithm that uses the center’s organ donation rates and stage 1 survival outcomes, as well as individual patients’ risk factors for mortality and organ availability.

when the relevant outcomes were not collected as part of the clinical trial; or when the decision maker is concerned with both clinical benefits and costs. Readers of a decision analysis paper should consider the strength of the evidence underlying the analysis, whether the model was constructed appropriately from a clinical perspective, and whether all relevant comparisons were included in the model.5,6 Key points ●

Use of electrocardiogram to predict myocardial infarction Shlipak et al 63 used a decision analytic model to assess the clinical utility of a previously reported electrocardiogram (ECG)-based algorithm to predict myocardial infarction in patients with left bundle branch block (LBBB). Grade B2 The authors developed probability data for their analysis by first conducting a retrospective cohort study of patients presenting with LBBB on their initial ECG. The subsequent decision analysis was performed to determine which of the following strategies would constitute optimal therapy: thrombolysis for all patients with LBBB; no treatment for these patients; or use of the ECG-based algorithm to screen patients for the appropriateness of thrombolysis. The authors found that the ECG algorithm had low sensitivity and would predict less than 10% of myocardial infarctions in patients presenting with LBBB and acute symptoms. As a screening test, the algorithm resulted in a survival rate less than that yielded by thrombolysis and similar to that yielded by no therapy. In one-way sensitivity analysis, thrombolysis was always the optimal strategy. In two-way sensitivity analyses thrombolysis was always preferred, unless the ECGbased algorithm had a sensitivity greater than 85%. If the ECG algorithm were used as a screening test for thrombolytic therapy, almost no patients with LBBB and myocardial infarction would receive the treatment. The authors conclude that the ECG algorithm is a poor predictor of myocardial infarction and that thrombolysis should be used for all patients with LBBB and symptoms of myocardial infarction.

Summary Decision analysis offers powerful techniques to better understand uncertain clinical decisions in cardiology. Increasing use of these techniques has already shown them to be very valuable in clinical and policy decision making in a variety of settings. Decision analysis may be most useful when clinical trial data do not clearly answer the clinical issue; when the clinical trial concludes that there are differences in risks and benefits across two treatment groups; 68

● ●

●

Decision analysis may be most useful when clinical trial data do not clearly answer the clinical issue; when the clinical trial concludes that there are differences in risks and benefits across two treatment groups; when the relevant outcomes were not collected as part of the clinical trial; or when the decision maker is concerned with both clinical benefits and costs. Sensitivity analysis is used to assess the impact of uncertainty on decision analytic models. In reviewing a decision analysis paper, the reader must assess whether the population considered in the model is relevant to the clinician’s population, the strength of the evidence available to the investigator in developing the model, and whether the model used by the investigator is constructed appropriately by including all relevant comparisons. Decision analysis has been used to assess a wide variety of clinical issues in cardiology.

Acknowledgment The authors are grateful to Damon Seils for research and editorial assistance.

References 1.Paulker SA. Coronary artery surgery: the use of decision analysis. Ann Intern Med 1976;85:8–18. 2.Detsky AS, Naglie G, Krahn MD, Naimark D, Redelmeier DA. Primer on medical decision analysis: Part 1 – getting started. Med Decis Making 1997;17:123–5. 3.Stason WB, Weinstein MC. Allocation of resources to manage hypertension. N Engl J Med 1977;296:732–9. 4.Kassirer JP, Moskowitz AJ, Lau J, Pauker SG. Decision analysis: a progress report. Ann Intern Med 1987;106:275–91. 5.Richardson WS, Detsky AS. User’s guide to the medical literature: VII. How to use a clinical decision analysis: A. Are the results of the study valid? JAMA 1995;273:1292–5. 6.Richardson WS, Detsky AS. User’s guide to the medical literature: VII. How to use a clinical decision analysis: B. What are the results and will they help me in caring for my patients? JAMA 1995;273:1610–13. 7.Greenberg ML, Malenka DJ, Disch DL. Therapeutic strategies for atrial fibrillation: the value of decision analysis. Cardiol Clin 1996;14:623–40.


8.Detsky AS, Naglie G, Krahn MD, Redelmeier DA, Naimark D. Primer on medical decision analysis: Part 2 – building a tree. Med Decis Making 1997;17:126–35. 9.Naglie G, Krahn MD, Naimark D, Redelmeier DA, Detsky AS. Primer on medical decision analysis: Part 3 – estimating probabilities and utilities. Med Decis Making 1997;17: 136–41. 10.Krahn MD, Naglie G, Naimark D, Redelmeier DA, Detsky AS. Primer on medical decision analysis: Part 4 – analyzing the model and interpreting the results. Med Decis Making 1997;17:142–51. 11.Naimark D, Krahn MD, Naglie G, Redelmeier DA, Detsky AS. Primer on medical decision analysis: Part 5 – working with Markov processes. Med Decis Making 1997;17:152–9. 12.Weinstein MC, Fineberg HV et al. Clinical decision analysis. Philadelphia: WB Saunders, 1980. 13.Sox HC, Blatt MA, Higgins MC, Marton KI. Medical decision making. Boston: Butterworth–Heinemann, 1988. 14.Petitti DB, Sidney S, Quesenberry CP Jr, Bernstein A. Incidence of stroke and myocardial infarction in women of reproductive age. Stroke 1997;28:280–3. 15.The Antiarrhythmics versus Implantable Defibrillators (AVID) Investigators. A comparison of antiarrhythmic-drug therapy with implantable defibrillators in patients resuscitated from near-fatal ventricular arrhythmias. N Engl J Med 1997;337: 1576–83. 16.Kupersmith J, Hogan A, Guerrero P et al. Evaluating and improving the cost-effectiveness of the implantable cardioverter-defibrillator. Am Heart J 1995;130:507–15. 17.Hogenhuis W, Stevens SK, Wang P et al. Cost-effectiveness of radiofrequency ablation compared with other strategies in Wolff–Parkinson–White syndrome. Circulation 1993;88: 437–46. 18.Barrett BJ, Parfrey PS, Foley RN, Detsky AS. An economic analysis of strategies for the use of contrast media for diagnostic cardiac catheterization. Med Decis Making 1994;14: 325–35. 19.Johannesson M, Jönsson B, Kjekshus J et al. Cost-effectiveness of simvastatin treatment to lower cholesterol levels in patients with coronary heart disease. N Engl J Med 1997;336:332–6. 20.Ashraf T, Hay JW, Pitt B et al. Cost-effectiveness of pravastatin in secondary prevention of coronary artery disease. Am J Cardiol 1996;78:409–14. 21.Tsevat J, Duke D, Goldman L et al. Cost-effectiveness of captopril therapy after myocardial infarction. J Am Coll Cardiol 1995;26:914–19. 22.Zubialde JP, Lawler F, Clemenson N. Estimated gains in life expectancy with use of postmenopausal estrogen therapy: a decision analysis. J Fam Pract 1993;36:271–80. 23.Hulley S, Grady D, Bush T et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMA 1998;280:605–13. 24.Herrington DM, Reboussin DM, Brosnihan KB et al. Effects of estrogen replacement on the progression of coronary-artery atherosclerosis. N Engl J Med 2000;343:522–9. 25.Herrington DM, Howard TD, Hawkins GA et al. Estrogenreceptor polymorphisms and effects of estrogen replacement on high-density lipoprotein cholesterol in women with coronary disease. N Engl J Med 2002;346:967–74.

26.Kellett J, Clarke J. Comparison of accelerated tissue plasminogen activator with streptokinase for treatment of suspected myocardial infarction. Med Decis Making 1995;15:297–310. 27.Dirsch DL, Greenberg ML, Holzberger PT, Malenka DJ, Birkmeyer J. Managing chronic atrial fibrillation: a Markov decision analysis comparing warfarin, quinidine, and low-dose amiodarone. Ann Intern Med 1994;120:449–57. 28.The European Atrial Fibrillation Trial Study Group. Secondary prevention in non-rheumatic atrial fibrillation after transient ischæmic attack or minor stroke. Lancet 1993;342:1255–62. 29.Connolly SJ. Canadian Atrial Fibrillation Anticoagulation (CAFA) Study. J Am Coll Cardiol 1991;18:349–55. 30.Ezekowitz MD, Bridgers SL, James KE et al. Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. N Engl J Med 1992;327:1406–12. 31.Stroke Prevention in Atrial Fibrillation Investigators. Stroke Prevention in Atrial Fibrillation (SPAF) Study: final results. Circulation 1991;84:527–39. 32.Stroke Prevention in Atrial Fibrillation Investigators. Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation: Stoke Prevention in Atrial Fibrillation II Study. Lancet 1994;343:687–91. 33.Petersen P, Boysen G, Godtfredsen J, Andersen ED, Andersen B. Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: the Copenhagen aFASAK Study. Lancet 1989; i:175–9. 34.The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators. The effect of low-dose warfarin on the risk of stroke in patients with nonrheumatic atrial fibrillation. N Engl J Med 1990;323:1505–11. 35.Laupacis A, Feeny D, Detsky AS, Tugwell PX. How attractive does a new technology have to be to warrant adoption and utilization? Tentative guidelines for using clinical and economic evaluations. Can Med Assoc J 1992;146:473–81. 36.Gage BF, Cardinalli AB, Albers GW, Owens DK. Cost-effectiveness of warfarin and aspirin for prophylaxis of stroke in patients with nonvalvular atrial fibrillation. JAMA 1995;274: 1839–45. 37.Solomon NA, Glick HA, Russo CJ, Schulman KA. Patient preferences for stroke outcomes. Stroke 1994;25:1721–5. 38.Moss AJ, Jackson Hall W, Cannom DS for the Multicenter Automatic Defibrillator Implantation Trial (MADIT) Investigators. Improved survival with an implanted defibrillator in patients with coronary disease at high risk for ventricular arrhythmia. N Engl J Med 1996;335:1933–40. 39.Friedman PL, Stevenson WG. Unsustained ventricular tachycardia – to treat or not to treat? N Engl J Med 1996; 335:1984–5. 40.Boyko W, Schulman KA, Tracy CM, Glick H, Solomon AJ. The economic impact of prophylactic defibrillators. J Am Coll Cardiol 1997;29(2 Suppl A):256A. 41.Owens DK, Sanders GD, Harris RA et al. Cost-effectiveness of implantable cardioverter defibrillators relative to amiodarone for prevention of sudden cardiac death. Ann Intern Med 1997;126:1–12. 42.Schatzkin A, Cupples LA, Heeren T et al. The epidemiology of sudden unexpected death: risk factors for men and women in the Framingham Heart Study. Am Heart J 1984;107:1300–6.

69


43.Pinski SL, Trohman RG. Implantable cardioverter-defibrillators: implications for the nonelectrophysiologist. Ann Intern Med 1995;122:770–7. 44.The Coronary Artery Bypass Graft (CABG) Patch Trial Investigators and Coordinators. The CABG Patch Trial. Prog Cardiovasc Dis 1993;36:97–114. 45.Cardiomyopathy Trial Investigators. The cardiomyopathy trial. Pacing Clin Electrophysiol 1993;16:576–81. 46.The DEFIBRILAT Study Group. Actuarial risk of sudden death while awaiting cardiac transplantation in patients with atherosclerotic heart disease. Am J Cardiol 1991;68:545–6. 47.AVID Trial Investigators. Antiarrhythmics Versus Implantable Defibrillators (AVID) – rationale, design, and methods. Am J Cardiol 1995;75:470–5. 48.Connolly SJ, Gent M, Roberts RS et al. Canadian Implantable Defibrillator Study (CIDS): study design and organization. Am J Cardiol 1993;72:103F–8F. 49.Hine LK, Laird NM, Hewitt P, Chalmers TC. Meta-analysis of empirical long-term antiarrhythmic therapy after myocardial infarction. JAMA 1989;262:3037–40. 50.Health Care Financing Administration. Revisions to payment policies and adjustments to the relative value units under the physician fee schedule for calendar year 1995; Final rule. Fed Reg 2 December 1995. 51.Eisenberg JM, Schulman KA, Glick H, Koffer H. Pharmacoeconomics: economic evaluation of pharmaceuticals. In: Strom BL, ed. Pharmacoepidemiology, 2nd edn. New York: John Wiley & Sons, 1994. 52.Detsky AS, Naglie IG. A clinician’s guide to cost-effectiveness analysis. Ann Intern Med 1990;113:147–54. 53.Naimark DM, Detsky AS. The meaning of life expectancy: what is a clinically significant gain? J Gen Intern Med 1994;9:702–7.

70

54.Glick H, Heyse JF, Thompson D et al. A model for evaluating the cost-effectiveness of cholesterol-lowering treatment. Int J Technol Assessment Hlth Care 1992;8:719–34. 55.Mason JJ, Owens DK, Harris RA, Cooke JP, Hlatky MA. The role of coronary angiography and coronary revascularization before noncardiac vascular surgery. JAMA 1995;273:1919–25. 56.Tosteson ANA, Goldman L, Udvarhelyi S, Lee TH. Costeffectiveness of a coronary care unit versus an intermediate care unit for emergency department patients with chest pain. Circulation 1996;94:143–50. 57.Nichol G, Laupacis A, Stiell IG et al. Cost-effectiveness analysis of potential improvements to emergency medical services for victims of out-of-hospital cardiac arrest. Ann Emerg Med 1996;27:711–20. 58.Stevenson LW, Warner SL, Steimle AE et al. The impending crisis awaiting cardiac transplantation: modeling a solution based on selection. Circulation 1994;89:450–7. 59.Wong JB, Salem DN, Paulke SG. You’re never too old. N Engl J Med 1993;328:971–5. 60.Naimark DM, Detsky AS. The meaning of life expectancy: what is a clinically significant gain? Med Decis Making 1992;12:344. 61.Kwok YS, Kim C, Heidenreich PA. Medical therapy or coronary artery bypass graft surgery for chronic stable angina: an update using decision analysis. Am J Med 2001;111:89–95. 62.Jenkins PC, Flanagan MF, Sargent JD et al. A comparison of treatment strategies for hypoplastic left heart syndrome using decision analysis. J Am Coll Cardiol 2001;38:1181–7. 63.Shlipak MG, Lyons WL, Go AS et al. Should the electrocardiogram be used to guide therapy for patients with left bundlebranch block and suspected myocardial infarction. JAMA 1999;281:714–9.

8

Assessing and changing cardiovascular clinical practices C David Naylor, David A Alter

Research into cardiovascular clinical practice has grown early enormously in volume and sophistication since the early twentieth century, driven by the worldwide prominence of atherosclerotic vascular diseases. The sheer volume of research literature has made it virtually impossible for even a subspecialist to stay abreast of her/his field. There is insufficient time for any evidence-oriented practitioner to critically appraise the full array of individual studies relevant to practice, and a real risk that, as the years go by, his/her filtering of the literature will prove misleading. One solution is for practitioners to rely increasingly on integrative reports. As documented throughout this volume, evidence on a particular clinical topic is often usefully compiled in published meta-analyses, decision analyses, or practice guidelines. These integrative reports synthesize the best evidence available from multiple research studies to help define what a practitioner ought to do when confronted with a particular clinical situation. While information uptake from integrative reports is necessary to ensure that clinical care evolves in evidence-driven directions, it may not be sufficient. For decades, researchers have shown that the rates of provision of various cardiovascular services vary inexplicably across regions and among nations. Some of this variation is random; some represents reasonable disagreement in the absence of definitive evidence about best practices. However, when practices are examined more closely using explicit criteria for appropriateness of care, it has become clear that actual practice sometimes differs sharply from what the evidence suggests ought to be done, raising concerns about quality of care. Quality concerns are further galvanized by evidence that technical skill and patient outcomes vary among procedural specialists. Not surprisingly, then, concerns with costs and quality of care have led a growing cadre of researchers, clinical leaders, facility managers, third party payers, and public policy makers to examine what clinicians do, and to seek ways to change clinical practice. Assessing and changing clinical practice is central to the discipline commonly known as health services research. This chapter accordingly provides an introduction to some of the key methods of health services research as applied to cardiovascular medicine and surgery.

By definition this chapter demands a different treatment than later chapters where it is possible to provide integrative summaries of evidence to inform contemporary practice or steer future research. Since our focus is on how evidence is translated into clinical action, it stands to reason that there will seldom be one “right answer”. Practice will instead be shaped not just by evidence, but by values and circumstances or context. Thus, it is important for the reader to suspend judgment as to whether there is necessarily one right health system, or one right profile of services for all populations with a given cardiovascular condition. A corollary of this point is that hundreds of descriptive and analytical studies have been published in cardiovascular health services research, many of which are context-specific. Our hope is to use a small number of these studies to heighten the reader’s understanding of analytical principles and general lessons. For consistency, the examples will relate to clinical management of coronary artery disease, not to primary and secondary prevention. However, the conceptual frameworks are applicable to all areas of cardiovascular care. It is hoped that the evidence-oriented reader will be able to generalize the methodological insights from this chapter to his/her particular clinical and research context. The specific objectives of this chapter are three: ● ●

●

to outline the challenges and opportunities in gathering evidence about how health care is delivered; to describe and provide illustrations of the various types of studies done to evaluate processes and outcomes of care; and to examine some of the interventions that can be undertaken to improve the quality of cardiovascular care.

Gathering evidence about health care: challenges and opportunities Study designs Randomized controlled clinical trials are the most rigorous tool for confirming causal relationships between a given outcome and intervention or factor. Most randomized clinical 71


trials are designed to test the efficacy of an intervention within a controlled and stable environment. In contrast, health services research focuses on assessing and improving the provision of care in usual practice settings. Observational studies of health services may be cross-sectional or cohort designs. Cross-sectional studies or survey designs offer convenient one-off snapshots of patient populations, providers, or practice settings. Disease- and procedure-specific inception cohorts have the advantage of delineating relationships of particular variables to outcomes over time. Cross-sectional and cohort methodologies can also be combined in a single study. For example, Payne and Saul1 undertook a mail survey of a random sample of 16 750 residents of the Sheffield (UK) region, and found that 4·0% of subjects had symptoms suggestive of angina pectoris. The prevalence of angina was significantly higher in neighborhoods with lower socioeconomic status, but these same areas had significantly lower rates of mechanical revascularization. In other words, variations in service profiles were inversely related to ecological markers of both population need and population deprivation – obvious grounds for concern about access or equity of services use. The authors went further, however, and used data linkage methods to determine procedures that were actually provided to individuals identified as having angina. In so doing, they effectively shifted from a cross-sectional study reliant on ecological inferences to a full-fledged cohort design. They found that among subjects reporting angina who lived in affluent neighborhoods, 11·2% had undergone procedures, as compared to 4·2% in less affluent areas (P 0·03). Similar socioeconomic-related disparities in cardiovascular processes of care have been well described in both private and publicly-funded healthcare systems.2,3 Intervention studies in health services research focus on effectiveness and efficiency rather than efficacy. Quasiexperimental designs and formal randomized clinical trials are brought into play to test interventions designed to improve care. However, for obvious reasons, it is often providers, clinics, hospitals, or regions that are randomized rather than patients.4 Whatever the internal validity of the design chosen, health services researchers face a recurrent challenge to prove the external validity of their work. Some of the published literature in health services research consists of local or regional quality assurance projects with uncertain generalizability, and evidence-oriented practitioners may not find these studies applicable in their own context. Data sources and collection Health services researchers use both primary and secondary data sources. Primary data are collected by design to answer specific research questions, whereas secondary data are used for multiple purposes and their use for research purposes 72

may be unplanned. Administrative databases designed for purposes of health service funding and administration are among the most common secondary data sources used in assessing clinical practice. Databases specifically constructed for ongoing epidemiologic surveillance of medical care, such as clinical registries, sit on the cusp between primary and secondary data, in that they are valuable for management and quality assurance, but are usually designed to meet specific research objectives as well. Prospective primary data collection is costly but crucial for complex variables that are poorly covered in most secondary data sources – for example, patients’ quality of life and psychosocial status. Retrospective primary data collection through chart reviews is also possible, but can be costly and time-consuming. It is best focused on routinely-recorded variables. For example, in charts of patients hospitalized with acute myocardial infarction (AMI), data on variables such as presenting symptoms, heart rate, blood pressure, ECGs, and cardiac enzymes are almost uniformly recorded. Absent primary data collection, there is always a risk that researchers will frame their questions around convenient access to data rather than addressing pressing issues. Researchers often combine primary and secondary data collection, or incorporate multiple data sources to address specific research questions. For instance, a study may assess patients’ short-term outcomes using self-administered health status questionnaires, and then track their subsequent use of health services and outcomes through administrative data. As an inexpensive solution to the limitations of single secondary databases, many researchers now link data across multiple administrative databases to provide better patient characterization and longitudinal follow up.5 Finally, linkage of samples from randomized clinical trials to administrative databases is becoming more common both to provide accurate and cost efficient follow up of clinical trial populations and to enable comparison of the characteristics and outcomes of trial participants to the broader populations from which they are drawn.6 Data quality Inaccurate measurement or recording is a particular concern when information comes from secondary data sources that are not designed for research or epidemiologic surveillance of medical care. For instance, Jollis et al 7 compared information about cardiac risk factors in an administrative database in patients undergoing angiography with information collected prospectively for a clinical database. A chance-corrected measure of agreement (kappa statistic) showed moderate to poor agreement as follows: hypertension (56%), heart failure (39%), and unstable angina (9%). Hannan et al 8 found similar discrepancies in comparing a cardiac surgery registry to an administrative database in New York State. While the accuracy of coding in

Assessing and changing cardiovascular clinical practices

administrative databases appears to be improving over time,9,10 significant undercoding of comorbidities still exists, especially among the elderly.10,11 As noted above, limited or inaccurate data in insurance databases or computerized hospital discharge abstracts may be supplemented or corrected by chart audits. A more efficient approach is to establish registries geared to measuring key patient characteristics, process-of-care elements, and relevant outcomes. Registries are proliferating in cardiovascular medicine and surgery, especially for acute ischemic syndromes and coronary surgery. This has led, however, to a new challenge – that is, agreement on a set of core data elements and definitions so that reliable comparisons can be drawn across registries from different jurisdictions.

studies are therefore useful indicators of quality of care for technically demanding services.19 Outcomes of interest, after Kerr White, can be conveniently remembered as the six “Ds”: death, disease, dysfunction, disability, distress, and dissatisfaction.20 The easiest outcomes for health services researchers to measure are those that are defined objectively and usually captured in large insurance databases or computerized hospital administrative data. These include death, routinely-coded complications following surgery, or hospital re-admissions. Linkage to vital status registries is also performed to track out-of-hospital deaths. Unfortunately, health services researchers have often failed to assess other outcomes, such as functional status, symptom relief, or overall quality of life, that are very important to patients and their physicians.21

Key measures Processes of care

Assessing processes and outcomes of care

Process of care is an umbrella term, encompassing all inputs into the clinical encounter that are relevant to the effectiveness and efficiency of the service provided. Process measures of particular interest for this chapter are the clinical decisionmaking patterns of physicians and other health professionals, as these reflect the uptake and use of evidence from the literature of medicine. Other inputs may also be relevant, such as hospital staffing ratios and qualifications of providers. Not infrequently, researchers use characteristics of the admitting hospital as ecologic proxies for processes of care that may affect individual patients.12–14 In this respect, hospital volumes for specific diagnoses or procedures are often taken as proxies for the expertise or experience of the relevant providers. Some measures are intermediate. Waiting times for services and lengths of stay, for example, are at once indicators of the process of care, and outcomes of interest to patients, professionals, and administrators alike.

Assessing processes of care

Outcomes The most important outcomes studies in cardiovascular care are conventional randomized trials used to test the efficacy of novel interventions, as described elsewhere in this volume. However, non-randomized outcomes studies have a role in assessing practice patterns. These studies allow for the evaluation of therapies and the natural history of disease in realworld settings.15,16 In some cases where randomization is simply not feasible (for example, socioeconomic status as a factor in prognosis), they also allow us to isolate patient characteristics from process-of-care factors to help elucidate pathophysiologic mechanisms of disease.17,18 Perhaps most importantly, a cardiovascular service may be provided to the right patient at the right time, and for the right reasons, but be delivered in a technically substandard fashion that leads to needlessly poor outcomes. Non-randomized outcomes

Descriptive studies Health services research gained considerable momentum in the 1970s and 1980s from studies pioneered by Wennberg and Gittelsohn,22,23 which documented unexplained geographic variations in rates of services. These early studies were a population-wide extension of research done in single hospitals or in public and private prepayment plans starting in the 1930s and showed variations in how different physicians managed apparently similar patients. However, Wennberg and coworkers coupled computerized systems of hospital discharge abstracts to census data and showed that citizens living in one area were significantly more or less likely to undergo certain procedures than those living in other areas. They also showed that greater variations were generally demonstrable when procedures were more discretionary or elective, or where there was uncertainty about the indications for the procedure or service of interest.24,25 In these latter instances, values and circumstances apparently interact strongly with evidence in driving decisions about service provision.26 Such descriptive studies continue to appear in the health services literature. They involve simple rates or proportions, with various numerators and denominators. Possible numerators include primary care visits or encounters, specialized diagnostic and therapeutic services, composite measures of use, such as overall numbers of hospital bed-days used per 10 000 residents, or even mean expenditures per capita on health care for all types of services. Denominators may tally patients according to the clinics or hospitals that they use, or by their residency in a given geographic area. These two denominators may be melded into hospital market shares – for example, the total population living in an area where a specified percentage of all patients receive their cardiac care at the hospital of interest. 73


Several statistical summary measures are used in variations analyses.27,28 Computational details and statistical properties of these measures are beyond the scope of this chapter. What matters is that the degree of variation should be both statistically significant and suggestive of meaningful differences from the standpoint of quality, accessibility, or efficiency of care provision. Thus, examination of the patterns of service and potential outcome implications is arguably more illuminating than the focusing on specific summary measures. The interpretive challenges of such descriptive studies are illustrated by evidence assembled with clinical and/or administrative data showing sex differences in treatments for patients hospitalized with acute myocardial infarction (AMI).29–31 Sex differences in care have been found in several nations, but the relationship between gender and service intensity is not consistent.32–34 The debate about the gender gap in service intensity is likely to continue until there is clearer evidence from randomized trials to delineate whether and how men and women with otherwise similar cardiovascular disease should be managed differently. Variations in processes of care have been well documented to extend beyond patient factors. As one example of this genre, Chen et al 35 documented significant interhospital variations in length of stay after AMI in Ontario. These variations persisted after adjustment for various factors such as coronary angiography on the index admission, patients’ age and sex, and comorbidity as inferred from secondary diagnoses on discharge abstracts. In almost any jurisdiction and for almost any cardiovascular service where interpractitioner, interinstitutional, or interregional variations in patterns of service provision have been sought, they are demonstrable. In sum, descriptive studies showing process-of-care variations are tantamount to screening tests in medical practice. They raise the possibility that there may be a problem with quality, efficiency, or accessibility. However, the finding of statistically significant variations is predicated on a null hypothesis that processes of care should vary no more than would be expected on the basis of the play of chance. Most such studies apply direct or indirect standardization to control for differences in the age–sex profile of the populations being compared, but may not consider myriad other sources of variation (Box 8.1). In response to that limitation, researchers may either develop evidence-oriented criteria to examine decision making at the level of the individual case, or try to link processes and outcomes of care in the same study as a means of inferring a causal connection. We examine both types of studies below. Criteria-based utilization analyses Given the limitations of descriptive studies that delineate variations in processes of care, health services researchers have developed other methods to determine whether the 74

Box 8.1 Sources of regional/institutional variation in service profiles ● Age and sex composition ● Age/sex specific disease incidence ● Random variation with time and place ● Availability and practice organization, such as ● primary care ● specialist services ● hospital services/bed provision ● overall funding levels ● methods of payment ● alternative services ● Referral patterns ● Practice styles of service providers ● Variations in patient expectations, demands, health education/behaviors ● Rates of previous service (for example, organ removal where relevant)

right service is provided to the right type of patient for the right reasons at the right time and place. One approach is implicit reviews of case records, drawing on the individualized judgments of expert clinicians. Unfortunately, lack of standardization renders implicit reviews unreliable.36,37 Explicit criteria, which form the basis for most processof-care analyses in the literature, have the advantages of standardization and consistency, as well as transparency. Where necessary, trained staff can apply them retrospectively to medical records without a major time commitment from clinicians. These studies are described in America as “utilization reviews” and in the UK as “clinical audits”.38 Process-of-care audits have the advantage of efficiency in comparison to outcomes studies as quality management tools. Bad outcomes caused by negligence and incompetence are (happily) rare. Technical competence does not necessarily equate with good judgment and appropriateness of service provision. Moreover, bad outcomes from undertreatment are hard to detect because the impact of modern cardiovascular care is often to make life only a little better on average for patients or to reduce their risk of otherwise rare events. For example, from overviews of randomized placebo-controlled trials we know that blockers confer about a 25% relative reduction in mortality in the first year after a myocardial infarction. For a cohort of medium-risk patients, this equates to an absolute reduction in cumulative postdischarge mortality from 4% to 3%. To show such a mortality difference on a comparative outcomes audit of two practices (80% power, 2-sided alpha of 0·05), we require over 5000 patients per practice; but a 1% mortality difference presumes absolutely no use of blockers in the practice with poorer outcomes. A more realistic assumption would be that about 70% of eligible patients receive blockers in the practice with worse outcomes versus over 95%


in the exemplary practice. Based on the randomized trials, this equates to perhaps a 0·2% increase in mortality. To detect such a small difference in mortality would require over 100 000 patients per practice! In contrast, one could simply examine charts to see whether patients were getting blocker prescriptions, versus 70% in the other practice, one would only need to examine about 75 charts in each practice for a reliable assessment. This latter audit is simple in another respect. We can basically use randomized trial inclusion and exclusion criteria to decide who should be getting the drug, make sure there are no obvious contraindications or medication intolerances documented on the medical record, and tally whether patients are getting the treatment that they ought to be getting. In general, however, audits require close attention to the validity, application, and applicability of the criteria chosen (Box 8.2).38 Box 8.2 User’s guide to appraising and applying the results of a process-of-care audit ● Are the criteria valid? ● Was an explicit and sensible process used to identify, select, and combine evidence for the criteria? ● What is the quality of the evidence used in framing the criteria? ● If necessary, was an explicit, systematic, and reliable process used to tap expert opinion? ● Was an explicit and sensible process used to consider the relative values of different outcomes? ● If the quality of the evidence used in originally framing the criteria was weak, have the criteria themselves been correlated with patient outcomes? ● Were the criteria applied appropriately? ● Was the process of applying the criteria reliable, unbiased, and likely to yield robust conclusions? ● What is the impact of uncertainty associated with evidence and values on the criteria-based ratings of process of care? ● Can you use the criteria in your own practice setting? ● Are the criteria relevant to your practice setting? ● Have the criteria been field-tested for feasibility of use in diverse settings, including settings similar to yours? Adapted from Naylor and Guyatt38

Validity of audit criteria To be valid, the criteria must have a direct link either to improving health (as is obvious with blockers for secondary prevention after AMI) or to lowering resource use without compromising health outcomes. There should be an explicit and sensible process to identify, select, and combine the relevant outcomes-based evidence. The hierarchy of evidence outlined above by Kitching, Sackett and Yusuf applies here. Evidence from randomized

trials is strongly preferred, but evidence from observational sources cannot be ignored. For example, from observational studies within trials, it is plain that the largest survival benefits with thrombolytic therapy are obtained when treatment is administered early.39 It would be unethical to randomize patients to receive thrombolysis on a delayed or urgent basis to determine how large these effects are. Thus, guidelines now recommend that thrombolytic therapy be administered, wherever possible, within 30 minutes of a pateint’s arrival to hospital.40 Studies from America,41 Canada,42 the UK,43 Italy,44 and New Zealand45 have all documented remediable problems with treatment delays in administering thrombolytic agents to eligible patients. All are classic examples of criteria-based audits. If only some of the indications for a particular service under audit will be covered by high quality evidence, then weaker sources of evidence, inference, and expert opinion must often be brought into play, usually through formal panel processes. Such panels should include an explicit process for selecting panelists, and a sensible, systematic method for collating their judgments. In this respect, the RAND group has pioneered multispecialty panel methods that are widely emulated.46–48 Scenarios are compiled that describe a potential indication for the procedure or clinical service in question. Each expert panelist independently rates hundreds of different case scenarios on a risk–benefit scale. Scenarios are re-rated at a panel meeting after patterns of interpanelist agreement and disagreement are shown anonymously and discussed. The final set of panelists’ ratings then determines whether a given indication is deemed potentially appropriate, uncertain, or inappropriate. With this method, it is not clear whether the appropriateness ratings for any given indication rest primarily on research evidence or inference, extrapolation, and opinion. The relative values placed on different outcomes are also unclear. For example, in randomized trials of CABG versus percutaneous transluminal coronary angioplasty (PTCA),49–52 PTCA has a slightly lower early mortality, along with lower initial costs and more rapid recovery from the procedure. Longer term mortality data are similar, but CABG patients appear to achieve better symptom relief, have decreased use of medication, and require fewer subsequent procedures.53 When an expert panel addresses the respective appropriateness of PTCA and CABG, the findings reflect these trade offs, but we cannot be sure that patients themselves would make the same choices. The conflation of facts and values in panel-based criteria is highlighted by studies showing that the nationality of a panel markedly affects the criteria and results of applying them to cardiovascular procedures (Table 8.1).26,54 Indeed, available evidence would also suggest that hospital practice settings and resource availability influence panel-based criteria.55 Nonetheless, the RAND methods compare very favorably with those used to create several utilization review tools now in widespread use.38 75


Table 8.1 Categorization of appropriateness of indications for cardiovascular procedures based on actual audits in the field: cross-national differences in expert panel assessments Procedure

Location/sample

Year

n

Panel nationality

Appropriate

Uncertain

Inappropriate

Coronary artery bypass graft

USA, 4 hospitals in Washington State

1979–80

386

American

62

25

13

British

41

24

35

American

67

26

7

British

57

27

16

American

88

9

3

Canadian

85

11

4

American

91

7

2

Canadian

85

10

6

American

50

23

27

British

11

29

60

American

74

9

17

British

39

19

42

American

71

12

17

British

49

30

21

American

77

18

5

Canadian

58

33

9

American

76

20

4

Canadian

51

39

10

1979–82 UK, 3 hospitals in Trent region Canada, 13 hospitals in Ontario and British Columbia USA, 15 hospitals in New York State Coronary angiography

USA, 4 hospitals in Washington State

1987–88

1989–90

1990

1979–80

319

556

1336

376

1979–82 USA, Medicare beneficiaries in 3 states UK, 3 hospitals in Trent region Canada, 20 hospitals in Ontario and British Columbia USA, 15 hospitals in New York State

1981

1987–88

1989–90

1990

1677

320

533

1333

Adapted from Naylor26 The data show the appropriateness ratings for sets of identical patient charts as described. Each set of charts was assessed according to criteria derived by expert panels based in the listed countries.

Application and applicability of the audit criteria Application of explicit process-of-care criteria often rests on data derived from retrospective chart reviews by professional auditors. The audit process must therefore be reliable. Biases can be introduced through skewed sampling of practitioners, hospitals, and patients. Even a meticulous audit, however, may miss mitigating factors. Thus, in many instances, if the explicit review shows potential problems 76

with the appropriateness of a service, the case is assessed by experienced clinicians to preclude “false positives”. It is also crucial that enough cases be reviewed to draw robust conclusions. For example, in one study, RAND researchers used explicit criteria to assess the appropriateness of PTCA in 1990 for 1306 randomly selected patients in 15 randomly selected New York State hospitals.54 The inappropriate utilization rate varied by hospital from 1% to

9% (P 0·12). Differences of this magnitude, if real, could be important to patients, payers, and policy makers. Thus, this sample size may have been insufficient for the investigators to confirm important differences in quality among hospitals. Although the task is subjective, end users must consider intangibles such as local medical culture and practice circumstances before accepting audit criteria that may not be relevant. The stronger the evidence on which the criteria are based, the less one needs to consider local factors; for example, few medical cultures would reject aspirin for AMI – a cheap and simple drug treatment that has been definitively proven to yield reductions in mortality. With weaker evidence and higher costs, however, the judgments are less straightforward. Last, even if criteria are sufficiently valid and relevant, training times and other costs must be considered. Special logistical problems arise when criteria are used for concurrent case management rather than retrospective utilization review. Any errors associated with concurrent care management will have immediate consequences for individual patients and physicians. Nonetheless, many American hospitals already do a range of concurrent reviews. The use of chart audits to infer appropriateness Table 8.1 shows the proportion of appropriate, inappropriate, and “uncertain” indications for cardiac procedures as randomly audited in the USA, UK, and Canada.26,56–58 Since all the procedures shown are used many times more often in the USA than in the UK, it seems almost paradoxical that the proportions of inappropriate cases are not much higher in the USA. The literature has suggested that relationships between appropriateness of care and cardiovascular service intensity are similarly weak within nations.25,58–60 However, two studies shed a slightly different light on this issue. The rates of all major coronary procedures in New York State, USA are about twice as high as in Ontario, Canada.61 Figure 8.1 shows the relative rate of isolated coronary artery bypass surgery (CABG) for the two jurisdictions by age and anatomy. Overall, only 6% of CABG patients in Ontario versus 30% of patients in New York had limited coronary artery disease – one or two vessel disease without proximal left anterior descending (PLAD) involvement. However, more patients in New York had left mainstem disease (23% v 16%, P 0·001). In relative terms, the differences are most dramatic among elderly persons. For example, New York brings 17 times as many persons over the age of 75 to surgery with anatomic patterns of coronary disease that are not associated with life expectancy gains after CABG. Nonetheless, much of this extra use could pass an appropriateness audit, since 90% of the persons with limited coronary anatomic disease in New York had moderate to severe angina before surgery.61

Relative CABG surgery rate (NY to ON)


20 16·75

Age 20–64 yr Age 65–74 yr Age 75 yr

15 10·78 10 7·28

4·55

5 0·85 1·18 0

2·17

2·01 2·52

One vessel or two Two vessel with Left main disease vessel without PLAD or PLAD disease three vessel disease Coronary anatomy

Figure 8.1 Relative rate of isolated CABG for New York State (NY) and Ontario (ON) according to age and disease anatomy. Adapted from Tu et al.61 PLAD, proximal left anterior descending.

A reasonable inference is that major increases in capacity, and expansion of population-based services rates, are associated with diminishing marginal returns. The Canadian approach – fixed budgets in a universal health system, and “managed delay” with organized waiting lists62 – seems to promote more efficient use of resources, with patients receiving surgery primarily if they are likely to have life expectancy gain. However, restricted use of coronary angiography leads to some implicit rationing that affects primarily the elderly, and a certain proportion of patients at all ages with left mainstem disease are not detected and/or do not undergo surgery. A second study63 of CABG develops this argument more strongly. Rather than using appropriateness criteria from an expert panel, Hux et al based their case-specific process assessments on a meta-analysis of randomized trials by Yusuf et al 64 Whereas the broad category of “appropriate” care as defined by expert panels includes a range of risk–benefit ratios, a trials-based assessment allowed estimation of the degree of potential 10 year survival benefit conferred by CABG surgery among patients for whom, by and large, it was appropriate. Hux et al found that only 6% of 5058 Ontario patients undergoing isolated CABG in 1992–93 fell in the low benefit category – that is, patients for whom there is no survival advantage from early CABG. However, the degree of anticipated benefit differed according to the center where surgery was provided. For instance, the proportion of patients in a high-benefit category ranged from 65·2 to 79·9% (P 0·001). Significantly more patients were in a high-benefit category in hospitals serving areas with lower population-based rates of CABG. Analyzing the data by site of residence, there was an inverse relationship between marginal degree of life expectancy gains and the surgical rates for each county.63 77


In sum, if one accepts that overtly inappropriate services are unlikely to be commonplace in any health system, the relationship between appropriateness of care and populationbased services rates can be redefined. Rather than seeking to relate the prevalence of bad judgment to high service intensity, or decrying health systems with low service intensity for rationing care, researchers might better assess whether the marginal returns of other forms of cardiovascular care are indeed smaller in areas where those services are used more frequently. The policy decision then becomes one of trade offs: given competing demands on scarce healthcare resources, at what point do the marginal returns of particular cardiac services become low enough that further investment in those services cannot be justified? Evidence-oriented clinicians must be positioned to contribute to these debates by marshaling comparative utilization data that help decision makers make explicit determination of the likely yields from funding different sets of cardiovascular and non-cardiovascular services. Arguably, they must also use these evaluative tools to safeguard their patients against inappropriate underuse of necessary services. Again, explicit process-of-care criteria can be helpful. For example, analytical variations studies using American data have repeatedly shown that black and uninsured patients have lower coronary angiography rates than those who are insured.65–67 Laouri et al 68 drew on audit data from four teaching hospitals in Los Angeles and assembled a cohort of 352 patients who met explicitly defined criteria for the necessity of coronary angiography as established by an expert panel. The patients were tracked forward for 3 months and, after adjustment for confounding factors, those managed in the public hospital system had a 35% rate of angiography versus 57% for private hospital patients (P 0·005). Two recent studies incorporate appropriateness criteria to provide further evidence for underuse of coronary interventions. The first by Guadagnoli et al 69 examined variations in coronary angiography after AMI in approximately 50 000 elderly Medicare beneficiaries in the USA. Among those patients with ACC–AHA class 1 indications, coronary angiography was used less often among Medicare beneficiaries enrolled in managed-care plans than among those with fee-for-service coverage. Moreover, utilization rates among elderly patients with class I indications for angiography were low in both groups (37% v 46%), suggesting room for improving the care of such patients with acute myocardial infarction. In contrast, the rate of angiography use among those with ACC–AHA class III indications (where angiography was deemed not useful) was similarly low (13%) in both groups. The second prospective study applied appropriateness ratings for coronary revascularization procedures to 2552 patients identified at the time of coronary angiography for various indications. Among 908 patients with indications appropriate for PTCA, 34% were treated medically. Among 1353 patients with indications appropriate for CABG, 78

26% were treated medically. Relating processes to outcomes, the research team also found that medically-treated patients deemed appropriate for revascularization were more likely to experience adverse events downstream.70 The lesson, simply put, is that evidence must be sought for both inappropriate overuse and underuse of cardiovascular services in any and all healthcare systems. Outcomes studies and process–outcome relationships Types of outcome studies Researchers, clinicians, and administrators alike are also drawing on outcomes with increasing frequency as a means of assessing quality of care. To repeat a point made earlier, various biases threaten the validity of inferences drawn from these non-randomized studies; but they have a useful role both in monitoring quality of care and as a source of evidence when randomization is not feasible or appropriate. Just as studies in the 1960s and 1970s showed geographic and institutional variations in broad markers of processes of care, so also did the 1980s and 1990s see the publication of research demonstrating significant mortality differences across physicians,71 hospitals,72 regions,73 and health systems.74 The magnitude of mortality variations has been meaningful, even amongst relatively homogeneous groups of patients. For example, Tu et al demonstrated marked interhospital and interregional variations in 1 year risk-adjusted mortality rates for patients hospitalized between 1994 and 1997 in one Canadian province. Mortality ranged from 20·8% to 27·4% across regions, and from 17·6% to 32·3% across hospitals admitting 100 or more AMI cases per year.75 Regional variations persist even in highly selected subpopulations of patients. Pilote et al demonstrated that 1 year AMI mortality rate across eight US census regions ranged from 8·6% to 10·3% among the population enrolled in GUSTO-1.73 As with descriptive studies of variations in process of care, these high-level outcomes studies function largely as screening tests: they often raise more questions than answers. Researchers use multivariate analyses to adjust for prognostic differences in the patient populations being compared. However, since patients are not randomized to different sites or regions, there is uncertainty about the extent to which unmeasured variation in patient characteristics accounts for the residual outcomes variation. Furthermore, the higher the level of comparison and the longer the follow up, the more uncertain the causal inferences become. Regional differences in long-term AMI outcomes, for example, may reflect genetic differences in populations, environmental factors, regional variation in health behaviors and socioeconomic status, as well as more conventional factors such as variations in processes of care


on the index hospitalization and follow up interventions (for example, revascularization or rehabilitation). For convenience, we suggest that outcomes analyses in health services research can be classified variously as qualityof-care screening studies or process/outcome hypothesis studies. Quality-of-care screening studies focus on outcomes to detect variations in quality of care. They are most powerful when applied to short-term outcomes that are closely tied to a particular episode of illness or procedure, and a provider or institution. In these circumstances, causal inferences are more straightforward. Their applicability is clearest for technically demanding procedures, such as PTCA or CABG, where variations in outcomes are taken as proxies for operator skill. However, even in such instances, other factors in pre- and perioperative care may be important. For relatively homogeneous diagnoses, outcomes studies may also sometimes be a useful screen to determine if detailed process-of-care analyses are required. For example, if inhospital mortality were found to be similarly low across a whole set of institutions, there would be little rationale for undertaking a major audit of processes of care. Ultimately, the goal of such studies is to isolate one or more process-of-care factors that can be modified to lead to consistently better outcomes. Outcomes analyses may also be used to validate process-of-care criteria or their application, for example, the study of underuse of revascularization by Hemingway et al cited above.70 In this sense there is overlap between the two categories of non-randomized outcomes studies. But an important distinction should also be drawn. Quality-of-care studies are concerned with the applicability of existing evidence in a particular context. Other outcomes studies may be initiated with a view to deriving or supporting generalizable hypotheses about the process– outcome relationship. They are poor cousins to randomized trials from the standpoint of strength of evidence. For true efficacy assessments, randomized trials are usually possible and always preferable, given the unavoidable biases of observational studies.76 A poorly conducted non-randomized outcomes comparison for quality management purposes may at worst mislead patients and tarnish the reputation of a number of capable cardiologists or cardiac surgeons. A poorly conducted non-randomized outcome comparison of two treatments may, if taken seriously, misguide clinical practice worldwide. That caveat aside, these process/outcome hypothesis studies can be useful to illustrate unanticipated harm from interventions, test the external validity of randomized trial results, generate hypotheses about interventions that may be worth testing with formal experimental designs, and, in special circumstances, provide an acceptable level of evidence for adopting a particular intervention. There are many methods available for examining the relationship between processes of care and outcomes. The

simplest method is to draw broad causal inferences using ecological comparisons, for example, correlating differences in processes and outcomes across two or more institutions or jurisdictions. However, the greater the difference between service settings being compared, the more difficult it is to be sure that patients were similar, or to isolate which aspects, if any, of the process of care relate to the outcomes observed. This is especially true when comparisons are made on a broad geographic footing between regions or countries in which populations and processes of care differ in many ways. In these latter comparisons, we are obviously veering away from the use of non-randomized outcomes data to benchmark technical quality of care for homogeneous procedures, and entering the more complex realm of process/ outcome hypothesis studies. This genre is typified by several studies77–80 showing that Canadian patients have more symptoms, worse functional status, or higher death/re-admission rates after AMI than do American patients. The reasons for these differences, however, are unclear. For example, Mark et al 78 in a GUSTO-1 substudy found that, while rates of revascularization were much higher in the USA, Canadians drew their post-MI care more often from family physicians and general internists, while Americans relied more on cardiologists and received more cardiac rehabilitation services.78 In other words, revascularization was only one factor among many that might explain differences in outcomes across two health systems. In an effort to limit the effects of competing process factors, analysts have borrowed the concept of instrumental variables from econometrics.81 This approach compares patients’ outcomes according to some characteristic that sharply distinguishes the care of two or more groups of patients. Thus, one might attempt to elucidate the impact of differences in the rate of revascularization across hospitals with and without on-site interventional capacity. Alter et al 72 recently used such a design to show that hospitals with on-site revascularization facilities had a lower rate of nonfatal composite outcomes (recurrent cardiac hospitalization and emergency department visits), and were also 3·5 times more likely to refer patients to myocardial revascularization procedures. Yet, despite the markedly higher rates of invasive procedures, the non-fatal outcome advantages of invasive-procedure hospitals were actually explained by their teaching status! In sum, given the relatively weak inferences possible from most observational studies of outcomes, alternative strategies for ensuring the quality of medical care should always be considered. It will often be feasible and more efficient to use randomized trials or meta-analyses of trials to establish optimal management strategies, and then ensure that quality of care is maintained by monitoring the process of care in that well-proven practices are consistently applied to eligible patients. On the other hand, for high volume and technically demanding procedures where reasonable risk 79


adjustment methods can be brought into play, outcomes measurement has merit for quality control so long as the results are interpreted carefully. Finally, studies aimed at delineating process–outcome relationships will continue to be valuable, but researchers and evidence-oriented practitioners alike will often find that the interpretation of the findings plunges them into a thicket of causes, effects, and epiphenomena. Special challenges in non-randomized outcomes studies In this section, we delve more deeply into some of the analytical challenges of non-randomized outcomes studies. Many types of biases have been described in the literature,82,83 but selection bias is a recurrent concern whether one is comparing the outcomes of two cardiac surgeons, or using non-randomized data to develop hypotheses about the effectiveness of pharmacologic or non-pharmacologic therapies in real-world settings. Indeed, the ubiquity of selection bias in health services research arises from the fact that ordinary good judgment in practice inevitably means that there are systematic differences in the characteristics of patients who are selected for particular interventions as compared to those who are not. Patients selected post-MI to undergo coronary angiography, for example, are often younger and healthier than other MI victims.72,82 The survival benefits observed for those undergoing angiography may therefore be due to prognostic characteristics rather than to revascularization consequent upon angiography. This latter phenomenon is known as confounding and is a common result of selection biases. Confounding occurs when particular factors are associated with both a study (process) variable and the outcome of interest. Researchers therefore routinely employ some form of multivariate analysis to adjust for imbalances in prognostic factors between groups under study. A complementary strategy is to confirm the consistency of the findings after restricting the analysis to a relatively low-risk subgroup of the patients being examined.76 Eliminating patients in higher risk categories associated with more widely varying physiologic states increases the likelihood of a “level playing field” for comparisons. For many common procedures and diagnoses, researchers can draw on validated prognostic indices and risk-adjustment algorithms as signposts in carrying out study-specific multivariate analyses. For frequently studied procedures such as CABG, major studies have tended to show relative consistency in the types of prognostic clinical factors that must be taken into account for risk adjustment purposes.84 Not surprisingly, risk-adjustment models appear to perform somewhat better with clinical as compared to administrative data.85 However, the key to predictive performance appears to be better data, not more variables. Studies have suggested 80

that the accuracy of risk-adjustment models reaches a plateau after use of only a few key variables. Tu et al,86 for example, examined risk-adjusted hospital mortality rates for CABG with multisite registry data. They determined that six core variables in a risk-adjustment model (age, gender, emergency surgery, previous CABG, LV dysfunction, left main disease) permitted modest discrimination between patients who did and did not die postoperatively (area under the receiver operating characteristic [ROC] curve 0·77). Statistical performance improved only trivially with the inclusion of six additional characteristics, and the relative rankings in the risk-adjusted mortality rates between hospitals did not change. Notwithstanding these studies, the ultimate number as well as the type of clinical variables required in a risk-adjustment model will obviously depend upon the disease being assessed, the processes and outcomes of interest, and the unit of analysis (for example, risk-adjusted mortality rates per physician v per hospital). Propensity scores can also be used to contain the impact of confounding.87 This method reduces the entire collection of background characteristics into a single composite characteristic (that is, the propensity to receive treatment v no treatment), which is then used to subclassify patients further into categories of relative equal propensities. Accordingly, the case-mix composition of patients with similar propensities is balanced, and outcome differences can be directly compared between those receiving and not receiving treatment. While not a solution for confounding per se, hierarchical statistical modeling has recently found favor as a useful analytical tool in outcome studies.88,89 Data in health research frequently exist in an ordered hierarchical structure: that is, patients are managed by physicians who practice within hospitals. In contrast, traditional multivariate techniques ignore the natural hierarchy of data and treat each observation as if it were independent (Figure 8.2).

Traditional multivariate models Patient level

Physician level

Hospital level

Hierarchical multivariate models

Hospital level

Physician level

Patient level

Figure 8.2 models

Schematic view of hierarchical v traditional


The use of hierarchical modeling makes intuitive sense since patients may share higher-level characteristics, leading to observations that are not necessarily independent of one another. The existence of standardized inhospital processes of care (for example, treatment protocols and care maps) may result in greater homogeneity in treatments across patients admitted to a particular institution. Accordingly, the use of traditional multivariate analyses may lead to an artificially inflated number of independent observations and an underestimate in the magnitude of standard error and potential alpha error.90 While the embedding of multivariate analyses in a hierarchical structure has obvious advantages, neither this technique nor fastidious risk-adjustment methods can match the effectiveness of randomization when balancing the casemix distribution between two groups, especially because researchers and quality-of-care evaluators are unlikely to know all the prognostic factors that interact with processes of care and may alter outcomes. Moreover, even if key prognostic confounders are known, they may not all have been measured or recorded accurately. Box 8.3 sets out some general principles that may be useful when researchers appraise non-randomized outcome studies.38 Box 8.3 User’s guide to appraising an observational outcomes study ● Are the outcome measures accurate and comprehensive? ● Were there clearly identified, sensible comparison groups? ● Were all important determinants of outcome measured accurately and reliably? ● Were the comparison groups similar with respect to important determinants, other than the one of interest? ● Was multivariate analysis used to adjust for imbalances in patient prognostic factors and other outcome determinants? ● Did additional analyses (particularly in low-risk subgroups) demonstrate the same results as the primary analysis? ● Did any multivariate analysis take into account natural heirarchies in the data, such as clustering of patients within providers’ practices and/or within institutions? Adapted from Naylor and Guyatt38

Changing practice patterns General considerations Practices clearly change over time in response to published evidence. At times, these changes can be rapid and dramatic, particularly when an innovation is associated with overwhelmingly positive risk–benefit ratios and is feasible for large numbers of practitioners to adopt. This model of knowledgebased practice change is termed passive diffusion. Its impact

is heightened by the extent to which the mass media pick up major medical advances, and by the marketing initiatives of drug and device manufacturers. However, as implied by studies showing unexplained and undesirable variations in practice patterns, the model of passive diffusion leads to inconsistent uptake of evidence into practice. How, then, can evidence be incorporated into practice more consistently, and what happens when data are in hand showing either that practice departs sharply from what available evidence suggests should be the norm, or that technical competence is below standard? How can the gap between “is” and “ought” in medical care be closed? These questions relate to changing physician (and system) performance, and follow logically from work done to measure or assess practice processes and outcomes. Although there is limited randomized evidence on this topic for specific aspects of cardiovascular care, a wealth of experience – some unhappy – has shown that direct incentives and disincentives, financial and otherwise, can have a major impact on practice. Bonuses are paid in American managed care organizations if practitioners meet certain financial and clinical performance targets. Within the UK National Health Service, meeting targets for prespecified preventive services leads to extra payments for general practitioners; and the new rating system for hospital trusts offers administrative autonomy and preferential access to capital funding as a reward for strong performance on measures of quality, accessibility, and efficiency. Simply shifting the mode of physician payment may be an effective way of modifying behavior. For example, exponents of fee-for-service remuneration of cardiovascular medicine and surgery argue that salary and capitation schemes impose a risk of underservicing. Critics of fee-for-service argue that it undervalues quality and cognitive services, and creates a conflict of interest that promotes the use of procedures. As to non-financial incentives and disincentives, the range of options includes merit awards, disciplinary proceedings, and litigation. Arguably more relevant to the evidence-oriented practitioner is the available information on non-administrative mechanisms to improve physician performance that rely on voluntary knowledge- or information-based change. Such initiatives have the advantage of calling forward the better instincts of health professionals who, with few exceptions, seek first to serve patients as competently as possible. Exponents of clinical guidelines initially believed that dissemination of guidelines might prove a key component in catalyzing knowledge-based improvements in physician performance.91 Guidelines would usefully compile the totality of relevant evidence on several related aspects of a clinical condition, treatment, or procedure. The evidence-oriented practitioner would no longer have to comb through the clinical literature, critically appraise it, and keep the relevant materials at hand or in her/his memory. The guideline would instead provide a convenient source of definitive 81


evidence. Furthermore, because inference, expert judgment, values, and circumstances could be used in developing guidelines, clinicians would be able to rely on regionallydeveloped guidelines to navigate the many “grey zones” of clinical practice26 where evidence alone was insufficient. Finally, guidelines could be developed, endorsed and disseminated by authorities with clinical credibility, lending weight to evidence that might otherwise appear rather impersonally in clinical journals. Lomas92 termed this latter approach the model of active dissemination, and criticized its prospects for success on the grounds that it ignored other factors in the practice environment, and presupposed that information acquisition alone leads to behavior change. The available evidence does suggest that there is some impact from more active approaches to informing and educating physicians about relevant clinical advances or guideline content.93 However, the more passive the educational process, and the more removed it is from physicians’ own practice context, the less likely it appears to succeed. Researchers and administrators have accordingly developed an array of non-coercive interventions designed to improve physician performance (Box 8.4). In 1995 Davis et al 94 and Oxman et al 95 conducted systematic reviews of all the available controlled studies of the effects of these strategies on physicians’ and other health professionals’ performance. They included any strategy designed to persuade physicians “to modify their practice performance by communicating clinical information”. Purely administrative interventions or financial and similar applied incentives and disincentives were excluded. There were 99 studies involving physicians and a further three on other health professionals’ behavior. Most of the studies on physician performance focus on internists or family physicians, and specific cardiovascular studies are limited in number to date. Single-intervention studies had positive effects on process or outcome parameters in 49/81 (60%) of trials where they were applied. Short educational seminars or conferences and dissemination of educational materials (printed or in audiovisual format) were least effective of all the single-intervention modalities explored. This finding supports proponents of implementation as opposed to dissemination. Simple audit-and-feedback studies had limited impact. However, it is important to distinguish the types of studies that fall into this category. For example, in randomized studies from the early 1980s, investigators showed that a computer-based monitoring system with reminders and feedback led to significantly better follow up and blood pressure control for patients with hypertension.96,97 Two controlled studies by Pozen et al 98,99 showed that a point-of-service strategy to facilitate implementation of a predictive algorithm for chest pain diagnosis reduced inappropriate use of coronary care units. These studies can best 82

Box 8.4 Some methods used to alter physician performance/behavior ● Education materials: Distribution of published or printed recommendations, including practice guidelines and audiovisual materials or electronic publications. ● Conferences: Participation of healthcare providers in conferences, lectures, workshops, or traineeships outside their practice settings. ● Outreach visits: Use of a trained person who meets with providers in their practice settings to provide information. The information given may include feedback on the provider’s performance. ● Local opinion leaders: Use of providers explicitly nominated by their colleagues to be “educationally influential”. ● Patient-mediated interventions: Any intervention aimed at changing the performance of healthcare providers for which information was sought from or given directly to patients by others (for example, direct mailings to patients, patient counseling delivered by others, or clinical information collected directly from patients and given to the provider). ● Audit and feedback: Any summary of clinical performance of healthcare over a specified period, with or without recommendations for clinical action. The information may have been obtained from medical records, computerized databases or patients or by observation. ● Reminders: Any intervention (manual or computerized) that prompts the healthcare provider to perform a clinical action. Examples include concurrent or intervisit reminders to professionals about desired actions such as screening or other preventive services, enhanced laboratory reports or administrative support (for example, follow up appointment systems or stickers on charts). ● Marketing: Use of personal interviewing, group discussion (focus groups) or a survey of targeted providers to identify barriers to change and the subsequent design of an intervention. ● Local consensus processes: Inclusion of participating providers in discussion to ensure agreement that the chosen clinical problem is important and the approach to managing it appropriate. Modified from Oxman et al 95

be regarded as “reminder” studies because there is continuous feedback at point of service. Audit-and-feedback studies that appear to be ineffective are those where data are collected and cumulated about processes or outcomes, and fed back only intermittently to practitioners without mechanisms to ensure local buy-in, to address local barriers to change, or to rectify specific gaps in clinical knowledge that may be associated with aberrant practice patterns. The latter distinction also highlights the fact that feedback can occur concurrently with service provision or retrospectively (that is, after the service has been provided). Concurrent audit and feedback arguably is taken to its


administrative conclusion in utilization management programs that refuse to authorize payment for a cardiovascular procedure unless the patient meets certain criteria, or in mandatory second opinion programs. These types of programs were not included in the reviews by Davis et al 94 and Oxman et al.95 The methods that had the most consistent effects were: outreach visits including formal academic detailing and opinion-leader studies, where an educationally influential physician was nominated by local peers to be the vector for the information; physician reminder systems at point of service; and patient-mediated methods, including reminders or educational materials. If two or more modalities were combined, then the effects were greater – that is, combining two effective methods (for example, academic detailing with support from a local opinion leader) had more impact than combining two less effective methods (for example, audit-andfeedback combined with a one-day seminar). Multifaceted interventions showed the strongest effects, with 31 of 39 (79%) positively affecting processes or outcomes of care. Davis et al 94 noted that most interventions appear to have a greater impact on process-of-care measures and other indices of physician performance, than on patient outcomes. They postulated that this may be because the clinical interventions themselves have limited impact (a rationale for the power argument given earlier), and because patients do not always accept physician recommendations. They also suggest that a recurring weakness in interventions designed to improve processes and outcomes of care is a failure to conduct a needs analysis that addresses barriers to change. These systematic reviews of practice-change interventions do not provide definitive evidence about which behavior change interventions are most effective and efficient in particular contexts or clinical conditions. This is because the studies cover a wide range of clinical condition and provider groups, rendering inferences across studies difficult. As in any meta-analysis, cross-study inferences involve nonrandomized comparison with all their potential pitfalls. Furthermore, factorial designs in behavior changes studies have been more the exception than the rule, and it is therefore usually unclear as to which element(s) in a multifactorial strategy was (were) truly effective. Nonetheless, the evidence from controlled trials does suggest that practice changes are best achieved by combining credible evidence or information with active local strategies of implementation using multifactorial methods. Such multifactorial initiatives are further supported in a recent qualitative study examining factors leading to increasing blocker use after AMI.100 Hospitals with greater improvements in blocker use over time, when compared to those having less or no improvement, were more likely to have shared goals, substantial administrative support, strong physician leadership advocating blocker use, and incorporation of credible data feedback programs.

The case of outcomes report cards The interest in outcomes measurement to assure technical competence has led to statewide initiatives whereby all cardiac surgery centers in New York and Pennsylvania, USA, are mandated to provide clinical data to permit compilation of publicly released mortality “report cards” on their CABG patients. (More recently, cardiovascular report cards have included interregional and hospital-specific AMI mortality rates, process indicators, (for example, evidence-based therapies and cardiac intervention rates post-AMI),75,101 and patient satisfaction with hospital care.102) The CABG report cards provide a final case study that bridges some of the material presented above on outcomes assessment and behavior change. In New York between 1989 and 1992, inhospital postoperative mortality of CABG showed an unadjusted relative decline of 21%.103,104 Patients were apparently becoming sicker in the same period, so that the risk-adjusted mortality decline was computed as 41%. Exponents of outcomes reporting claim that this improvement was catalyzed by a reporting system that provided relevant data to patients, administrators, and referring physicians.103,104 There can be no doubt that the New York and Pennsylvania report cards have pinpointed problems with a few operators who had very poor technical outcomes. The key question is how much of the overall improvement in mortality can be attributed to public outcomes reportage. Some critics contend that the trend is confounded by two factors. More assiduous coding of risk factors would artefactually increase the overall expected mortality, and surgeons could generate better mortality profiles by selectively turning down high-risk patients, even though such patients may have most to gain from CABG. There has indeed been a striking increase in the prevalence of various reported risk factors in the New York database since its inception. For example, prevalence of congestive heart failure rose from 1·7% in 1989 to 7·6% in 1991; renal failure rose from 0·4% to 2·8%, chronic obstructive pulmonary disease (COPD) from 6·9% to 17·4% and unstable angina from 14·9% to 21·8% in the same period.105 As well, a survey106 of randomly selected cardiologists and cardiac surgeons in Pennsylvania found that about 60% of cardiologists reported greater difficulty in finding surgeons who would operate on high-risk patients; a similar number of surgeons reported that they were less willing to operate on such patients. However, this type of survey is weak evidence for harm done by untoward case selection, and internal New York data do not support such a trend in the state.107 A more telling criticism is the fact that ecological correlations between falling mortality and initiation of reportage are tantamount to a case series in medicine. They provide weak and uncontrolled evidence for causation. In fact, the above-noted survey106 of randomly selected cardiologists in Pennsylvania showed that most referring physicians did not

83


view the Pennsylvania guide as an important source of information because of concerns about inadequate risk adjustment, unreliable data, and the absence of indicators of quality other than mortality. Schneider and Epstein108 later surveyed patients undergoing cardiac surgery in Pennsylvania to determine the impact of the statewide consumer guide to the performance of hospitals and individual surgeons. Only 12% of the patients were aware of the guide before undergoing a CABG, and less than 1% knew the correct rating of their hospital or surgeon or reported that such information had any meaningful influence on their selection of a provider for open-heart surgery. It is perhaps not surprising that, more generally, a recent overview by Marshall et al 109 found little evidence for consumer-driven market shifts arising from public report cards about specific diseases or procedures. It appears more plausible that the publication of outcomes “report cards” facilitates change by sensitizing politicians, public servants, and the governing bodies of hospitals to the existence of outcome variations. For example, after the publication of the CABG “report card”, New York State insisted on attainment of center-specific minimum case volumes before certifying any cardiac surgery program. On the other hand, in the absence of any report cards, the drop in post-CABG mortality in neighboring Massachusetts110 has rivaled that seen in New York and Pennsylvania. Technical improvements in surgery, together with closer quality monitoring at the institutional level, appear to be the primary reason for these improved outcomes. Given what has been learned about physician behavior change, the controversy about the New York State and Pennsylvania programs is hardly surprising. These externally mandated experiments in outcomes assessment contrast with initiatives that involve influential professionals and promote local buy-in from the outset. O’Connor discusses elsewhere in this volume the successful regional collaboration for continuous quality improvement that was developed in northern New England by involving cardiac surgeons in a systematic examination and improvement of processes and outcomes of care.111–113 In Canada, a similar cooperative venture exists through the Cardiac Care Network of Ontario, which draws together representatives of all major cardiovascular referral centers in the province.114 Historically, confidential report cards on mortality and length of stay were generated for the chief of cardiac surgery and CEO (cheif executive officer) at each center, using risk adjustment algorithms coauthored by leaders of the Cardiac Care Network itself.84 CABG outcomes in Ontario are comparable to those in New York and Pennsylvania. Moreover, as in Massachusetts, the trend to improved outcomes antedates the report card system.115,116 Most recently, hospital-specific CABG outcomes in Ontario have been made available to the public. In summary, the unresolved issues with public outcomes report cards include validity and reliability of the data and 84

the risk adjustment algorithms, as well as inadvertent adverse effects (for example, avoidance of high-risk patients, and consumers’ or referring physicians’ focus on point estimates rather than statistically reliable ranges). Potential harm to the public from substandard technical competence must be weighed against needless patient anxieties and confusion, along with harm to skilled health workers and fine institutions caused by poorly founded and widely publicized inferences about inferior outcomes. Debate continues, but it is untenable to assume that all hospitals or providers are equally technically competent, and the public has an unequivocal right to receive reliable and current data on physician and hospital performance. Thus, the trend must inexorably be toward greater public reporting of both process and outcome indicators of quality of care. The challenges for evidence-oriented practitioners are to ensure that the right indicators are chosen, that reliable data are analyzed appropriately, and that responsible reporting mechanisms are developed.

Conclusions Assessing cardiovascular practices involves observational methods that can focus on either processes or outcomes of care. Methodologies for process-of-care assessments range from simple descriptive studies revealing variations in practice, to highly sophisticated case-specific audits using explicit criteria. Process-of-care assessments are more efficient than outcomes assessments in many respects, and lend themselves to measuring both over- and underuse of necessary cardiovascular services, thereby shedding light on quality and accessibility of care. Observational outcomes measurement is nonetheless useful in assessing provider or institutional quality of care for high volume and relatively homogeneous procedures where technical skill is a factor. These comparisons must be made with caution, given the inevitable influence of unrecognized confounding through selection biases inherent in routine practice. The use of well-validated risk adjustment algorithms is imperative to improve the chances that differences in outcomes arise from the technical quality of care provided, rather than from differences in prognostic characteristics of patients themselves. Observational outcomes studies can also be undertaken cautiously to illustrate unanticipated harm from interventions, test the external validity of randomized trial results, generate hypotheses about interventions that may be worth testing with formal experimental designs, and, very rarely, provide an acceptable level of evidence for adopting a particular intervention. To reduce general inconsistencies in the uptake of evidence into practice, and to redress instances where process or outcomes of clinical care are measured and found wanting, several proven strategies are available. First, while new


evidence published in journals or distilled into educational materials and practice guidelines does change practice through passive diffusion, evidence is most likely to have an impact if actively disseminated and made relevant and salient locally to practitioners. Strategies to achieve this end include: ● ● ● ● ● ●

reminder systems concurrent audit and feedback local outreach through academic detailing patient-mediated interventions local involvement of an educationally influential practitioner, and a local needs assessment with a consensus among providers on the issues as well as the barriers and facilitators to positive change.

In conclusion, the practitioner of evidence-based cardiovascular medicine and surgery is increasingly challenged to stay abreast of his or her field and to maintain technical competence in performing ever more exacting procedures. Information systems in practice can and will be re-engineered to be more conducive to evidence-based clinical decision making. However, it will also remain important to assess practice patterns on a systematic basis, to share that information with patients and providers, and wherever necessary, take steps to improve physician performance with a view to optimizing the quality, accessibility, and efficiency of cardiovascular care. References 1.Payne N, Saul C. Variations in use of cardiology services in a health authority: comparison of coronary artery revascularisation rates with prevalence of angina and coronary mortality. BMJ 1997;314:257–61. 2.Alter DA, Naylor CD, Austin P, Tu JV. Effects of socioeconomic status on access to invasive cardiac procedures and on mortality after acute myocardial infarction. N Engl J Med 1999;341:1359–67. 3.Anderson GM, Grumbach K, Luft HS, Roos LL, Mustard C, Brook R. Use of coronary artery bypass surgery in the United States and Canada. Influence of age and income. JAMA 1993;269:1661–66. 4.Krieger J, Collier C, Song L, Martin D. Linking communitybased blood pressure measurement to clinical care: a randomized controlled trial of outreach and tracking by community health workers. Am J Public Health 1999;89:856–61. 5.Tu JV, Naylor CD, Austin P. Temporal changes in the outcomes of acute myocardial infarction in Ontario, 1992–1996. Can Med Ass J 1999;161:1257–61. 6.Jha P, Deboer D, Sykora K, Naylor CD. Characteristics and mortality outcomes of thrombolysis trial participants and nonparticipants: a population-based comparison. J Am Coll Cardiol 1996;27:1335–42. 7.Jollis JG, Ancukiewicz M, DeLong ER, Pryor DB, Muhlbaier LH, Mark DB. Discordance of databases designed

for claims payment versus clinical information systems. Implications for outcomes research. Ann Intern Med 1993; 119:844–50. 8.Hannan EL, Kilburn H, Jr, Lindsey ML, Lewis R. Clinical versus administrative data bases for CABG surgery. Does it matter? Med Care 1992;30:892–907. 9.Fisher ES, Whaley FS, Krushat WM et al. The accuracy of Medicare’s hospital claims data: progress has been made, but problems remain. Am J Public Health 1992;82:243–8. 10.Dixon J, Sanderson C, Elliott P, Walls P, Jones J, Petticrew M. Assessment of the reproducibility of clinical coding in routinely collected hospital activity data: a study in two hospitals. J Public Health Med 1998;20:63–9. 11.Green J, Wintfeld N. How accurate are hospital discharge data for evaluating effectiveness of care? Med Care 1993;31: 719–31. 12.Krumholz HM, Chen J, Murillo JE, Cohen DJ, Radford MJ. Admission to hospitals with on-site cardiac catheterization facilities :impact on long-term costs and outcomes. Circulation 1998;98:2010–16. 13.Every NR, Larson EB, Litwin PE et al. The association between on-site cardiac catheterization facilities and the use of coronary angiography after acute myocardial infarction. Myocardial Infarction Triage and Intervention Project Investigators. N Engl J Med 1993;329:546–51. 14.Di Salvo TT, Paul SD, Lloyd-Jones D et al. Care of acute myocardial infarction by noninvasive and invasive cardiologists: procedure use, cost and outcome. J Am Coll Cardiol 1996;27:262–9. 15.Rochon PA, Anderson GM, Tu JV et al. Age- and genderrelated use of low-dose drug therapy: the need to manufacture low-dose therapy and evaluate the minimum effective dose. J Am Geriatr Soc 1999;47:954–9. 16.Krumholz HM, Radford MJ, Wang Y, Chen J, Marciniak TA. Early beta-blocker therapy for acute myocardial infarction in elderly patients. Ann Intern Med 1999;131:648–54. 17.Evans RG. Introduction. In: Evans RG, Barer ML, Marmor TR, eds. Why are some people healthy and others not? The determinants of health of populations. New York: Aldine de Gruyter, 1994. 18.Fiebach NH, Viscoli CM, Horwitz RI. Differences between women and men in survival after myocardial infarction. Biology or methodology? JAMA 1990;263:1092–96. 19.Meehan TP, Radford MJ, Vaccarino LV et al. A collaborative project in Connecticut to improve the care of patients with acute myocardial infarction. Conn Med 1997;61:147–55. 20.White KL. Improved medical care statistics and the health services system. Public Health Rep 1967;82:847–54. 21.Outcomes and the management of health care. Health Services Research Group. Can Med Ass J 1992;147:1775–80. 22.Wennberg J, Gittelsohn. Small area variations in health care delivery. Science 1973;182:1102–8. 23.Wennberg J, Gittelsohn A. Variations in medical care among small areas. Sci Am 1982;246:120–34. 24.Wennberg JE, Barnes BA, Zubkoff M. Professional uncertainty and the problem of supplier-induced demand. Soc Sci Med 1982;16:811–24. 25.Wennberg J. Which rate is right? N Engl J Med 1986;314: 310–11.

85


26.Naylor CD. Grey zones of clinical practice: some limits to evidence-based medicine. Lancet 1995;345:840–2. 27.Diehr P, Cain KC, Kreuter W, Rosenkranz S. Can small-area analysis detect variation in surgery rates? The power of smallarea variation analysis. Med Care 1992;30:484–502. 28.Diehr P, Cain K, Connell F, Volinn E. What is too much variation? The null hypothesis in small-area analysis. Health Serv Res 1990;24:741–71. 29.Petticrew M, McKee M, Jones J. Coronary artery surgery: are women discriminated against? BMJ 1993;306:1164–6. 30.Jaglal SB, Goel V, Naylor CD. Sex differences in the use of invasive coronary procedures in Ontario. Can J Cardiol 1994;10:239–44. 31.Ayanian JZ, Epstein AM. Differences in the use of procedures between women and men hospitalized for coronary heart disease. N Engl J Med 1991;325:221–5. 32.Krumholz HM, Douglas PS, Lauer MS, Pasternak RC. Selection of patients for coronary angiography and coronary revascularization early after myocardial infarction: is there evidence for a gender bias? Ann Intern Med 1992;116:785–90. 33.Gan SC, Beaver SK, Houck PM, MacLehose RF, Lawson HW, Chan L. Treatment of acute myocardial infarction and 30-day mortality among women and men. N Engl J Med 2000 343:8–15. 34.Weintraub WS, Kosinski AS, Wenger NK. Is there a bias against performing coronary revascularization in women? Am J Cardiol 1996;78:1154–60. 35.Chen E, Naylor CD. Variation in hospital length of stay for acute myocardial infarction in Ontario, Canada. Med Care 1994;32:420–35. 36.Quality of care: 1. What is quality and how can it be measured? Health Services Research Group. Can Med Ass J 1992;146:2153–8. 37.Quality of care: 2. Quality of care studies and their consequences. Health Services Research Group. Can Med Ass J 1992;147:163–7. 38.Naylor CD, Guyatt GH. Users’ guides to the medical literature. XI. How to use an article about a clinical utilization review. Evidence-Based Medicine Working Group. JAMA 1996;275:1435–9. 39.Fibrinolytic Therapy Trialists’ (FTT) Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet 1994;343:311–22. 40.The Heart and Stroke Foundation of Canada, the Canadian Cardiovascular Society and the Canadian Association of Emergency Physicians for the Emergency Cardiac Care Coalition. Recommendations for ensuring early thrombolytic therapy for acute myocardial infarction. Can Med Ass J 1996;154:483–7. 41.Rogers WJ, Bowlby LJ, Chandra NC et al. Treatment of myocardial infarction in the United States (1990 to 1993). Observations from the National Registry of Myocardial Infarction. Circulation 1994;90:2103–14. 42.Cox JL, Lee E, Langer A, Armstrong PW, Naylor CD. Time to treatment with thrombolytic therapy: determinants and effect on short-term nonfatal outcomes of acute myocardial infarction. Canadian GUSTO Investigators. Global Utilization

86

of Streptokinase and PA for Occluded Coronary Arteries. Can Med Ass J 1997;156:497–505. 43.Birkhead JS. Time delays in provision of thrombolytic treatment in six district hospitals. Joint Audit Committee of the British Cardiac Society and a Cardiology Committee of Royal College of Physicians of London. BMJ 1992;305:445–8. 44.GISSI–Avoidable Delay Study Group. Epidemiology of avoidable delay in the care of patients with acute myocardial infarction in Italy. A GISSI-generated study. Arch Intern Med 1995;155:1481–8. 45.Porter G, Doughty R, Gamble G, Sharpe N. Thrombolysis in acute myocardial infarction: reducing in hospital treatment delay. N Z Med J 1995;108:253–4. 46.Park RE, Fink A, Brook RH et al. Physician ratings of appropriate indications for six medical and surgical procedures. Am J Public Health 1986;76:766–72. 47.Brook RH, Chassin MR, Fink A, Solomon DH, Kosecoff J, Park RE. A method for the detailed assessment of the appropriateness of medical technologies. Int J Technol Assess Health Care 1986;2:53–63. 48.Park RE, Fink A, Brook RH et al. Physician ratings of appropriate indications for three procedures: theoretical indications vs indications used in practice. Am J Public Health 1989;79: 445–7. 49.Hamm CW, Reimers J, Ischinger T, Rupprecht HJ, Berger J, Bleifeld W. A randomized study of coronary angioplasty compared with bypass surgery in patients with symptomatic multivessel coronary disease. German Angioplasty Bypass Surgery Investigation (GABI). N Engl J Med 1994;331:1037–43. 50.King SB III, Lembo NJ, Weintraub WS et al. A randomized trial comparing coronary angioplasty with coronary bypass surgery. Emory Angioplasty versus Surgery Trial (EAST). N Engl J Med 1994;331:1044–50. 51.Coronary angioplasty versus coronary artery bypass surgery: the Randomized Intervention Treatment of Angina (RITA) trial. Lancet 1993;341:573–80. 52.Rodriguez A, Boullon F, Perez-Balino N, Paviotti C, Liprandi MI, Palacios IF. Argentine randomized trial of percutaneous transluminal coronary angioplasty versus coronary artery bypass surgery in multivessel disease (ERACI): in-hospital results and 1-year follow-up. ERACI Group. J Am Coll Cardiol 1993;22:1060–7. 53.Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. The Bypass Angioplasty Revascularization Investigation (BARI) Investigators. N Engl J Med 1996;335:217–25. 54.Hilborne LH, Leape LL, Bernstein SJ et al. The appropriateness of use of percutaneous transluminal coronary angioplasty in New York State. JAMA 1993;269:761–5. 55.Ayanian JZ, Landrum MB, Normand SL, Guadagnoli E, McNeil BJ. Rating the appropriateness of coronary angiography – do practicing physicians agree with an expert panel and with each other? N Engl J Med 1998;338:1896–1904. 56.Brook RH, Kosecoff JB, Park RE, Chassin MR, Winslow CM, Hampton JR. Diagnosis and treatment of coronary disease: comparison of doctors’ attitudes in the USA and the UK. Lancet 1988;1:750–3. 57.McGlynn EA, Naylor CD, Anderson GM et al. Comparison of the appropriateness of coronary angiography and coronary


artery bypass graft surgery between Canada and New York State. JAMA 1994;272:934–40. 58.Chassin MR, Kosecoff J, Park RE et al. Does inappropriate use explain geographic variations in the use of health care services? A study of three procedures. JAMA 1987;258:2533–7. 59.Leape LL, Park RE, Solomon DH, Chassin MR, Kosecoff J, Brook RH. Does inappropriate use explain small-area variations in the use of health care services? JAMA 1990;263: 669–72. 60.Wennberg JE. The paradox of appropriate care. JAMA 1987;258:2568–9. 61.Tu JV, Naylor CD, Kumar D, DeBuono BA, McNeil BJ, Hannan EL. Coronary artery bypass graft surgery in Ontario and New York State: which rate is right? Steering Committee of the Cardiac Care Network of Ontario. Ann Intern Med 1997;126:13–19. 62.Naylor CD, Sykora K, Jaglal SB, Jefferson S. Waiting for coronary artery bypass surgery: population-based study of 8517 consecutive patients in Ontario, Canada. The Steering Committee of the Adult Cardiac Care Network of Ontario. Lancet 1995;346:1605–9. 63.Hux JE, Naylor CD. Are the marginal returns of coronary artery surgery smaller in high-rate areas? The Steering Committee of the Provincial Adult Cardiac Care Network of Ontario. Lancet 1996;348:1202–7. 64.Yusuf S, Zucker D, Peduzzi P et al. Effect of coronary artery bypass graft surgery on survival: overview of 10-year results from randomised trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration. Lancet 1994;344:563–70. 65.Goldberg KC, Hartz AJ, Jacobsen SJ, Krakauer H, Rimm AA. Racial and community factors influencing coronary artery bypass graft surgery rates for all 1986 Medicare patients. JAMA 1992;267:1473–7. 66.Hadley J, Steinberg EP, Feder J. Comparison of uninsured and privately insured hospital patients. Condition on admission, resource use, and outcome. JAMA 1991;265:374–9. 67.Hannan EL, Kilburn H, Jr, O’Donnell JF, Lukacik G, Shields EP. Interracial access to selected cardiac procedures for patients hospitalized with coronary artery disease in New York State. Med Care 1991;29:430–41. 68.Laouri M, Kravitz RL, French WJ et al. Underuse of coronary revascularization procedures: application of a clinical method. J Am Coll Cardiol 1997;29:891–7. 69.Guadagnoli E, Landrum MB, Peterson EA, Gahart MT, Ryan TJ, McNeil BJ. Appropriateness of coronary angiography after myocardial infarction among Medicare beneficiaries. Managed care versus fee for service. N Engl J Med 2000;343:1460–6. 70.Hemingway H, Crook AM, Feder G et al. Underuse of coronary revascularization procedures in patients considered appropriate candidates for revascularization. N Engl J Med 2001;344:645–54. 71.Tu JV, Austin PC, Chan BT. Relationship between annual volume of patients treated by admitting physician and mortality after acute myocardial infarction. JAMA 2001;285:3116–22. 72.Alter DA, Naylor CD, Austin PC, Tu JV. Long-term MI outcomes at hospitals with or without on-site revascularization. JAMA 2001;285:2101–8. 73.Pilote L, Califf RM, Sapp S et al. Regional variation across the United States in the management of acute myocardial infarction. GUSTO-1 Investigators. Global Utilization of

Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries. N Engl J Med 1995;333:565–72. 74.Yusuf S, Flather M, Pogue J et al. Variations between countries in invasive cardiac procedures and outcomes in patients with suspected unstable angina or myocardial infarction without initial ST elevation. OASIS (Organisation to Assess Strategies for Ischaemic Syndromes) Registry Investigators. Lancet 1998;352:507–14. 75.Tu JV, Austin P, Naylor CD, Iron K, Zhang H. Acute myocardial infarction outcomes in Ontario. In Naylor CD, Slaughter PM, eds. Cardiovascular health and services in Ontario. An ICES Atlas. Toronto: Institute for Clinical Evaluative Sciences, 1999. 76.Wen SW, Hernandez R, Naylor CD. Pitfalls in nonrandomized outcomes studies. The case of incidental appendectomy with open cholecystectomy. JAMA 1995;274:1687–91. 77.Rouleau JL, Moye LA, Pfeffer MA et al. A comparison of management patterns after acute myocardial infarction in Canada and the United States. The SAVE investigators. N Engl J Med 1993;328:779–84. 78.Mark DB, Naylor CD, Hlatky MA et al. Use of medical resources and quality of life after acute myocardial infarction in Canada and the United States. N Engl J Med 1994;331: 1130–5. 79.Pilote L, Racine N, Hlatky MA. Differences in the treatment of myocardial infarction in the United States and Canada. A comparison of two university hospitals. Arch Intern Med 1994;154:1090–6. 80.Fu Y, Chang WC, Mark D et al. Canadian-American differences in the management of acute coronary syndromes in the GUSTO IIb trial: one-year follow-up of patients without ST-segment elevation. Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) II Investigators. Circulation 2000;102:1375–81. 81.McClellan M, McNeil BJ, Newhouse JP. Does more intensive treatment of acute myocardial infarction in the elderly reduce mortality? Analysis using instrumental variables. JAMA 1994; 272:859–66. 82.DeLong ER, Nelson CL, Wong JB et al. Using observational data to estimate prognosis: an example using a coronary artery disease registry. Stat Med 2001;20:2505–32. 83.Sackett DL. Bias in analytic research. J Chronic Dis 1979;32: 51–63. 84.Tu JV, Jaglal SB, Naylor CD. Multicenter validation of a risk index for mortality, intensive care unit stay, and overall hospital length of stay after cardiac surgery. Steering Committee of the Provincial Adult Cardiac Care Network of Ontario. Circulation 1995;91:677–84. 85.Krumholz HM, Chen J, Wang Y, Radford MJ, Chen YT, Marciniak TA. Comparing AMI mortality among hospitals in patients 65 years of age and older: evaluating methods of risk adjustment. Circulation 1999;99:2986–92. 86.Tu JV, Sykora K, Naylor CD. Assessing the outcomes of coronary artery bypass graft surgery: how many risk factors are enough? Steering Committee of the Cardiac Care Network of Ontario. J Am Coll Cardiol 1997;30:1317–23. 87.Rubin DB. Estimating causal effects from large data sets using propensity scores. Ann Intern Med 1997;127:757–63. 88.Rice N, Leyland A. Multilevel models: applications to health data. J Health Serv Res Policy 1996;1:154–64.

87


89.Diez-Roux AV, Link BG, Northridge ME. A multilevel analysis of income inequality and cardiovascular disease risk factors. Soc Sci Med 2000;50:673–87. 90.Duncan C, Jones K, Moon G. Context, composition and heterogeneity: using multilevel models in health research. Soc Sci Med 1998;46:97–117. 91.Standards, guidelines and clinical policies. Health Services Research Group. Can Med Ass J 1992;146:833–7. 92.Lomas J. Retailing research: increasing the role of evidence in clinical services for childbirth. Milbank Q 1993;71:439–75. 93.Grimshaw JM, Russell IT. Effect of clinical guidelines on medical practice: a systematic review of rigorous evaluations. Lancet 1993;342:1317–22. 94.Davis DA, Thomson MA, Oxman AD, Haynes RB. Changing physician performance. A systematic review of the effect of continuing medical education strategies. JAMA 1995;274: 700–5. 95.Oxman AD, Thomson MA, Davis DA, Haynes RB. No magic bullets: a systematic review of 102 trials of interventions to improve professional practice. Can Med Ass J 1995;153: 1423–31. 96.Barnett GO, Winickoff RN, Morgan MM, Zielstorff RD. A computer-based monitoring system for follow-up of elevated blood pressure. Med Care 1983;21:400–9. 97.Dickinson JC, Warshaw GA, Gehlbach SH, Bobula JA, Muhlbaier LH, Parkerson GR, Jr. Improving hypertension control: impact of computer feedback and physician education. Med Care 1981;19:843–54. 98.Pozen MW, D’Agostino RB, Selker HP, Sytkowski PA, Hood WB, Jr. A predictive instrument to improve coronary-careunit admission practices in acute ischemic heart disease. A prospective multicenter clinical trial. N Engl J Med 1984; 310:1273–8. 99.Pozen MW, D’Agostino RB, Mitchell JB et al. The usefulness of a predictive instrument to reduce inappropriate admissions to the coronary care unit. Ann Intern Med 1980;92:238–42. 100.Bradley EH, Holmboe ES, Mattera JA, Roumanis SA, Radford MJ, Krumholz HM. A qualitative study of increasing beta-blocker use after myocardial infarction: Why do some hospitals succeed? JAMA 2001;285:2604–11. 101.Tu JV, Austin P, Rochon PA, Zhang H. Secondary prevention after acute myocardial infarction, congestive heart failure and coronary artery bypass graft surgery in Ontario. In: Naylor CD, Slaughter PM, eds. Cardiovascular health and services in Ontario: an ICES Atlas. Toronto: Institute for Clinical Evaluative Sciences, 1999. 102.Decker B, MacInnes R. Assessing the importance of report cards rating patient satisfaction. Health Syst Lead 1997; 4:16–18. 103.Hannan EL, Kilburn HJ, Racz M, Shields E, Chassin MR. Improving the outcomes of coronary artery bypass surgery in New York State. JAMA 1994;271:761–6.

88

104.Hannan EL, Siu AL, Kumar D, Kilburn H, Jr, Chassin MR. The decline in coronary artery bypass graft surgery mortality in New York State. The role of surgeon volume. JAMA 1995; 273:209–13. 105.Green J, Wintfeld N. Report cards on cardiac surgeons. Assessing New York State’s approach. N Engl J Med 1995; 332:1229–32. 106.Schneider EC, Epstein AM. Influence of cardiac-surgery performance reports on referral practices and access to care. A survey of cardiovascular specialists. N Engl J Med 1996;335: 251–6. 107.Hannan EL, Siu AL, Kumar D, Racz M, Pryor DB, Chassin MR. Assessment of coronary artery bypass graft surgery performance in New York. Is there a bias against taking high-risk patients? Med Care 1997;35:49–56. 108.Schneider EC, Epstein AM. Use of public performance reports: a survey of patients undergoing cardiac surgery. JAMA 1998; 279:1638–42. 109.Marshall MN, Shekelle PG, Leatherman S, Brook RH. The public release of performance data: what do we expect to gain? A review of the evidence. JAMA 2000;283:1866–74. 110.Ghali WA, Ash AS, Hall RE, Moskowitz MA. Statewide quality improvement initiatives and mortality after cardiac surgery. JAMA 1997;277:379–82. 111.O’Connor GT, Plume SK, Olmstead EM et al. A regional intervention to improve the hospital mortality associated with coronary artery bypass graft surgery. The Northern New England Cardiovascular Disease Study Group. JAMA 1996; 275:841–6. 112.O’Connor GT, Plume SK, Olmstead EM et al. A regional prospective study of in-hospital mortality associated with coronary artery bypass grafting. The Northern New England Cardiovascular Disease Study Group. JAMA 1991;266: 803–9. 113.Malenka DJ, O’Connor GT. A regional collaborative effort for CQI in cardiovascular disease. Northern New England Cardiovascular Study Group. Jt Comm J Qual Improv 1995; 21:627–33. 114.Tu JV, Naylor CD. Coronary artery bypass mortality rates in Ontario. A Canadian approach to quality assurance in cardiac surgery. Steering Committee of the Provincial Adult Cardiac Care Network of Ontario. Circulation 1996;94:2429–33. 115.Ivanov J, Weisel RD, David TE, Naylor CD. Fifteen-year trends in risk severity and operative mortality in elderly patients undergoing coronary artery bypass graft surgery. Circulation 1998;97:673–80. 116.Tu JV, Naylor CD. Coronary artery bypass mortality rates in Ontario. A Canadian approach to quality assurance in cardiac surgery. Steering Committee of the Provincial Adult Cardiac Care Network of Ontario. Circulation 1996;94:2429–33.

Part II Prevention of cardiovascular diseases Salim Yusuf, Editor

89


GRADE A

GRADE C


Level 5

GRADE B Level 2

Level 3 Level 4

90




9

Global perspective on cardiovascular disease K Srinath Reddy

In the second half of the 20th century, cardiovascular diseases (CVD) became the dominant cause of global mortality and a major contributor to disease related disability. In the first half of the 21st century this pattern will become even more pervasive as the CVD epidemic accelerates in many developing regions of the world, even as it retains its primacy as the leading public health problem in the developed regions.1–7 In 2000, CVD accounted for 16·7 million deaths globally.1 Coronary heart disease (CHD) and stroke were then leading contributors, with a death toll of 6·9 million and 5·1 million, respectively. According to estimates provided by the World Health Organization, in 1998 30·9% of all global deaths were due to CVD. Both men and women experienced these burdens, with CVD contributing to 28% of the deaths in the former and 34% of the deaths in the latter.2 The low and middle income countries contributed 78% of all CVD deaths, and 86·3% of disability adjusted life year (DALY) loss attributed to CVD that year. Although these large absolute burdens reflect the large population sizes of the developing countries, proportional mortality rates of deaths attributable to CVD have also been rising in these countries, from 24·5% in 1990 to 28·5% in 1998. The relative importance of CHD and stroke vary across regions and from country to country. For example, more than twice as many deaths from stroke occurred in the developing countries as in the developed countries.3 CHD was the dominant form of CVD in the developed countries, Latin America and India, whereas stroke was the leading cause of cardiovascular death in sub-Saharan Africa, China and other parts of Asia. Developing countries such as Argentina, Colombia and China now have CVD mortality rates higher than those of most other countries. Argentina currently exceeds many European and North American countries in its CVD mortality rate.7 The rise and recent decline of the CVD epidemic in the developed countries have been well documented.8–11 The identification of major risk factors through population based studies, and effective control strategies combining community education and targeted management of high-risk individuals, have together contributed to the fall in CVD

mortality rates (inclusive of coronary and stroke deaths) that has been observed in almost all industrialized countries. It has been estimated that, during the period 1965–90, CVD related mortality fell by 50% or so in Australia, Canada, France and the United States, and by 60% in Japan.8 Other parts of western Europe reported more modest declines (20–25%).8 The decline in stroke mortality has been more marked than the decline in coronary mortality. In the USA the decline in stroke mortality commenced nearly two decades earlier than that in coronary mortality and maintained a sharper rate of decline.10 During the period 1979–89, the age-adjusted mortality from stroke in that country declined by about one third, and the corresponding decline in coronary mortality was 22%.10 In Canada, Japan, Switzerland and the United States, stroke mortality has declined by more than 50% in men and women aged 65–74 years since the 1970s.11 In Japan, where stroke mortality outweighs coronary mortality, the impressive overall decline in CVD mortality is contributed principally by the former. However, recent trends in some of the developed countries have been of some concern. A flattening of age adjusted mortality rates for major cardiovascular diseases in the USA has been reported since 1990, with an especially well documented absence of a decline in stroke mortality since that year (Figure 9.1). This has been accompanied by Deaths/100 000 Population 900 Deaths/100 000 Population

Introduction

800 700 600 500 400

CVD Heart disease

300 200

CHD

100

Stroke

0 1900 1915 1930 1945

1960 1975 1990 1997

2000

Figure 9.1 Age adjusted (to 2000 standard) mortality rates for major cardiovascular diseases in the United States from 1900 to 199712

91


an increase in mortality from congestive heart failure. Lack of decline in incidence of CHD and stroke, fall in the rate of decrease in cardiovascular risk factor levels and rising levels of obesity since 1990 have all been incriminated as factors responsible for such a plateau effect on CVD mortality rates in the USA over the past decade.12 The discordant trend of rising CVD mortality rates in eastern and central Europe, however, is in sharp contrast to the decline in western Europe.4 In countries such as Bulgaria and Hungary, CHD mortality rates are now the highest in the world in both men and women, and are still rising.8 The average life expectancy in Russian males has fallen rapidly in recent years to below 60 years, a phenomenon to which rising CVD rates have contributed substantially.13–15 Considerable variations in CVD mortality trends have been described across central and eastern Europe.16 Whereas Poland has demonstrated a recent decline in CVD mortality, many other countries are still manifesting a rise. The 25% decline in CHD mortality observed in Poland during 1991–93 has been attributed to an increase in the consumption of fresh fruit and vegetables.17 Rheumatic heart disease (RHD) is also a major burden in the developing countries: it is the most common CVD in children and young adults. Although it is rare in the developed countries, at least 12 million persons are currently estimated to be affected by RHD globally.2 More than 2 million require repeated hospital admission and 1 million will need heart surgery over the next 20 years.2 Annually, 500 000 deaths occur as a result of RHD, and many poor persons, who are preferentially affected, are disabled because of lack of access to the expensive medical and surgical care demanded by the disease. The prevalence of RHD in the developing countries ranges from 1 to 10 per 1000 and the incidence of rheumatic fever ranges from 10 to 100 per 100 000, with a high rate of recurrence.

Early age of CVD deaths in developing countries Although the present high burden of CVD deaths is in itself an adequate reason for attention, a greater cause for concern is the early age of CVD deaths in the developing countries compared to the developed countries. For example, in 1990 the proportion of CVD deaths occurring below the age of 70 years was 26·5% in the developed countries, compared to 46·7% in the developing countries.4 The contrast between the truly developed “established market economies” (22·8% of CVD deaths 70 years) and a large developing country such as India (52·2%) was even sharper.3 Therefore, the contribution of the developing countries to the global burden of CVD, in terms of disability adjusted years of life lost, was 2·8 times higher than that of the developed countries. 92

Epidemiologic transition and the evolution of the CVD epidemic What is the “transition”? The health status and dominant disease profile of human societies have been historically linked to the level of their economic development and social organization at any given stage. The shift from nutritional deficiencies and infectious diseases as the major causes of death and disability, to degenerative disorders (chronic diseases such as CVD, cancer, diabetes) has marked the economic ascent of nations as they industrialized. This has been called the epidemiologic transition. The economic and social changes that propel this transition are related to a rise in per capita income; greater investments in public sanitation, housing and healthcare; assured availability of adequate nutrition; and technological advances in medical care. Life expectancy rises as causes of childhood and early adult mortality decline. This, in turn, leads to a decline in fertility. The age profile of the population changes from a pyramidal distribution dominated by the young to a columnar structure where adults and the elderly progressively expand their numbers. This has been described as the demographic transition. Because the disease profile is also linked to the age profile of the population, the health transition encompasses the effects of the epidemiologic and demographic transitions.

CVD profile at different stages of the epidemiologic transition The model of epidemiologic transition originally described by Omran,18 with three phases (the age of pestilence and famine; the age of receding pandemics; the age of degenerative and manmade diseases), was later modified to include a fourth phase, the age of delayed degenerative diseases.19 Life expectancy increases progressively from around 30 years in the first phase to over 70 years in the fourth phase. The shift to a dominant chronic disease profile occurs in the third phase. As the average life expectancy exceeds 50–55 years, the proportionate mortality due to CVD begins to exceed that of infectious diseases.20 The transition occurs not only between the broad disease categories but also within them. The disease profile within CVD alters at each phase of the epidemiologic transition. In the first phase (the age of pestilence and famine), CVD accounts for 5–10% of deaths.20 The major causes of CVD are, however, related to infectious and nutritional deficiencies. Thus, RHD and cardiomyopathies (for example, Chagas’ disease) are the main CVD in this phase. Even as countries emerge from this phase, the residual burden of chronic valvular heart disease and congestive heart failure often remains for some time. These effects are still evident in subSaharan Africa and parts of South America and south Asia.20

Global perspective on cardiovascular disease

In the second phase (the age of receding pandemics), the decline in infectious disease that accompanies socioeconomic development ushers in changes in diet. As the subsistence nutrition changes to more complete diets, the salt content of the food increases. Hypertension and its sequelae (hypertensive heart disease and hemorrhagic stroke) now affect the population, whose average age also has risen with increased life expectancy.20 Some residual burden of RHD and cardiomyopathies is also evident. These nonatherosclerotic diseases contribute to 10–35% of deaths. This pattern currently prevails in parts of Africa, north Asia and South America.20 In the third phase (the age of degenerative and manmade disease), accelerated economic development and increased per capita incomes promote lifestyle changes in diet, physical activity, stress and addictions. A diet rich in calories, saturated fat and salt is accompanied by reduced physical activity through the increased use of mechanized transport and sedentary leisuretime pursuits. The metabolic mismatch leads to obesity, increased blood lipids, diabetes and elevated blood pressure. Tobacco consumption, especially cigarette smoking, starts as a pleasurable pastime and turns into a severe addiction. These factors result in the onset of clinically manifest atherosclerotic vascular disease (CHD, atherosclerotic stroke and peripheral vascular disease) at around 55 years of age. Such patterns first occur in the upper socioeconomic classes, who have disposable income to expend on rich diets, tobacco and transport vehicles. Several countries in South America and Asia currently manifest this pattern. As the epidemic advances further and involves all social strata, with homogenization of risk behaviors and risk factors across the population, the death toll of CVD rises to range between 35% and 65% of all deaths. This scenario is currently observed in eastern Europe. In the fourth phase (the phase of delayed degenerative disease) a number of changes occur in the society to modify risk behaviors and reduce risk factor levels in the population. Health research augments the knowledge of CVD risk factors. The desire to reduce the adverse impact of CVD on individuals, as well as on the society, steer the community as well as the policymakers to apply this knowledge to disease prevention and health promotion. Community awareness through education, as well as its ability to exercise healthy choices through supportive regulatory measures, empowers its members to adopt healthier lifestyles. Saturated fat and salt consumption declines and leisuretime physical activity and exercise programs are avidly pursued. With concerns about the effects of active and passive smoking, tobacco consumption falls. Simultaneously, medical research makes available new technologies which are very effective in saving lives, modifying the course of disease and reducing the levels of risk factors. All of these changes, in unison, delay the onset of disease, lower the age standardized mortality rates and reduce the disability. The contribution of

CVD to total mortality falls to 50% or below. These patterns are now established in most of North America, western Europe and New Zealand.20 Recent developments in some countries of eastern Europe, with sharp declines in life expectancy and other health indices, led to a fifth phase of health transition being postulated.6 In this stage of “social upheaval and health regression” the CVD spectrum too may witness a reversal, with CHD and stroke occurring at younger ages, resulting in a fall in life expectancy as in Russia. Variations in the transition There are, however, variations on this theme. Even within Europe, for example, northern Europe and the Mediterranean countries have differences in CVD mortality rates which are better explained by cultural differences in diet than by the level of economic development.21 Japan has so far avoided the CHD epidemic.6 Whether recent changes in diet, with a rise in mean plasma cholesterol levels in the population, combined with high smoking rates, will lead to a major CHD epidemic in the future remains to be seen. The question of “arrested epidemiologic transition” is also raised with respect to some of the developing countries. If poverty continues to be a major problem for them, will they experience the CVD epidemic in its full fury or will the pretransitional diseases of nutrition and infection continue to occupy the center stage? Even now, there is evidence that the social gradient has begun to reverse for risk factor levels and even for morbidity measures in some populations in the developing world.4 Unless economic development is greatly stunted in some countries, it is likely that the model of epidemiologic transition will be applicable to most of the developing countries. The transition to the atherothrombotic phase of the epidemic may be preceded by a sharp fall in the burden of hemorrhagic stroke. The recent decline in CVD mortality reported from South Korea reflects such a fall in the contribution from hemorrhagic stroke, whereas thrombotic stroke and coronary heart disease have just begun to rise.22 Whether adherence to traditional diets will result in a continued decline of CVD in South Korea, or CHD rates will rise further to push up the CVD mortality rates, remains to be studied. Cuba and Chile have also been cited as examples of developing countries with declining CVD mortality rates, despite high life expectancy. The model of “health transition”, although very useful, is not immutable and is likely to vary according to both level of development and the nature of public health responses to social transition. The model of health transition should also not lead to complacency regarding the high absolute burdens and early deaths in the developing countries. For example, even in a country in “early transition”, such as Tanzania, the stroke mortality rate in the age group 15–59 years in rural and 93


urban areas is two to four times higher than that in the UK in a similar age group.23 Early and late adopters The pace of epidemiologic transition will vary both between and within countries. Usually lifestyle changes towards riskprone behaviors occur first in the higher socioeconomic groups and urban communities, for whom the innovations of modernity are more easily accessible and affordable. As these innovations diffuse and become routinely available at prices amenable to mass consumption, the poorer sections and rural communities also join the CVD bandwagon. Soon the awareness of CVD risks, as well as the economic independence to make healthy lifestyle choices in relation to diet and leisuretime exercise (along with the greater ability to access healthcare), moves the “early adopters” in the affluent and urban strata into a reduced risk zone. The burden of CVD is then largely concentrated in the lower socioeconomic groups and rural populations, who continue to practice high-risk behaviors and display elevated risk factor levels.20 These “late adopter” groups also will slowly alter their behaviors, lower their levels of risk and reduce their burden of CVD as healthcare responses to the CVD epidemic become universally effective. This is the evolutionary profile of the CVD epidemic, as evident from the analysis of mature epidemics in industrial nations and the advancing epidemics in the developing countries. Differences within and between countries, suggested by cross-sectional views at any point in this evolution, should not obscure the longitudinal perspective of an evolving epidemic in which most countries will traverse similar paths, albeit at different times determined by their pace of development. Global shifts in CVD risk factors and their reflection in global CVD trends indicate that all countries and communities have far more in common in terms of disease causation than the differences that demarcate them. The challenge of epidemiologic transition is not whether it will happen in the developing countries, but whether we can apply the available knowledge to telescope the transition and abbreviate phase three of the model in these countries.

Projections The Global Burden of Diseases study15 estimates that annual mortality from non-communicable diseases will rise from an estimated 28·1 million deaths in 1990 to 49·7 million in 2020. CVD, which accounts for a large proportion of these, will rise as a result of the accelerating epidemic in the developing countries. CHD will continue to be the leading cause of death in the world and, in terms of disability adjusted life years (DALY) lost, will rise from its fifth position in 1990 to top the DALY table in 2020.15 Men as well as women in the 94

developing countries will experience the largest rise in CHD and stroke mortality rates across the world (Table 9.1). Table 9.1 Global % change in CHD and stroke mortality 1990–2020 (adapted from Murray and Lopez3) CHD Men Developed countries 48 Developing countries 137 World 100

Stroke Women

Men

Women

29 127 80

56 124 106

28 107 78

The profile of DALY loss attributable to CVD in 1990 in various regions of the world and the projected estimates for 2020 (Table 9.2) also indicate a large rise.3 Among the developed countries, the sharp decline in the industrial nations is partly offset by the rise in the former socialist countries. Table 9.2 Contribution of cardiovascular disease to DALY loss (percentage of total) (adapted from Murray and Lopez3) Region

1990 (%)

2020 (%)

World Developed countries Developing countries

10·85 25·7 8·9

14·7 22·0 13·8

Deaths attributable to tobacco, a risk factor for CVD and other chronic diseases, are projected to rise from 3·0 million in 1990 to 8·4 million in 2020. The largest increases will be in India, China and other developing countries in Asia, where tobacco-attributable deaths will rise from 1·1 million to 4·2 million in 2020.24

Mechanisms that propel a CVD epidemic in developing countries Demographic changes due to the epidemiologic transition A major public health challenge, identified by recent analyses of global health trends, is the projected rise in both proportional and absolute CVD mortality rates in the developing countries over the next quarter century. The reasons for this are many.4 In the second half of the 20th century, most developing countries experienced a major surge in life expectancy. This was principally as a result of a decline in deaths occurring in infancy, childhood and adolescence, and was related to more effective public health responses to perinatal, infectious and nutritional deficiency disorders and to improved economic indicators, such as per capita income,


and social indicators such as female literacy in some areas. These demographic shifts have augmented the ranks of middle-aged and older adults. The increasing longevity provides longer periods of exposure to the risk factors for CVD, resulting in a greater probability of clinically manifest CVD events. The concomitant decline in infectious and nutritional disorders (competing causes of death) further enhances the proportional burden due to CVD and other chronic lifestyle related diseases. The ratio between deaths due to pretransitional diseases (related to infections and malnutrition) and those caused by post-transitional diseases (such as CVD and cancer) varies among regions and between countries, depending on factors such as the level of economic development and literacy, as well as availability of and access to healthcare. The direction of change towards a rising relative contribution of posttransitional diseases is, however, common to and consistent among the developing countries.25 The experience of urban China, where the proportion of CVD deaths rose from 12·1% in 1957 to 35·8% in 1990, illustrates this phenomenon.26 Population expansion and aging Despite relative declines in fertility, the continuing growth of populations in the developing countries will also increase the absolute numbers at risk of CVD. The world population is expected to rise from 5·71 billion in 1995 to 8·29 billion in 2025. Combined with changes in the demographic profile, this will result in a large number of adults who are potentially vulnerable to CVD. At present there are an estimated 380 million people aged 65 or more, including around 220 million in the developing countries. By 2020, the figures are projected to reach more than 690 million and 460 million, respectively.2 In India, for example, the population is expected to rise from 683·2 million in 1981 to somewhere between 1253·8 and 1480·5 million in 2021. Simultaneously, the proportion of adults aged 35 years or above will rise from 28·4% of the population to 42·4%.27

study (1984–86, 1988–89), and the substantially higher levels of most CVD risk factors in urban population groups compared to rural population groups in India, provide evidence of such trends.26 The increasing use of tobacco in a number of developing countries will also translate into higher mortality rates from CVD, lung cancer and other tobacco related diseases, and undesirable alterations in diet and physical activity are also having adverse effects on cardiovascular health. The global availability of cheap vegetable oils and fats has resulted in greatly increased fat consumption in low-income countries in recent years.28 The transition now occurs at lower levels of the gross national product than previously, and is further accelerated by rapid urbanization. In China, for example, the proportion of upper income persons who were consuming a relatively high-fat diet (30% of daily energy intake) rose from 22·8% to 66·6% between 1989 and 1993. The lower and middle income groups too showed a rise (from 19% to 36·4% in the former, and from 19·1% to 51·0% in the latter).28 The Asian countries, traditionally high in carbohydrates and low in fat, have shown an overall decline in the proportion of energy from complex carbohydrates along with an increase in the proportion of fat.28 The globalization of food production and marketing is also contributing to the increasing consumption of energydense foods that are poor in dietary fiber and several micronutrients.29 The rising tobacco consumption patterns in most developing countries contrast sharply with the overall decline in the industrial nations.30 Recent projections from the World Health Organization suggest that by the year 2020 tobacco will become the largest single cause of death, accounting for 12·3% of deaths worldwide.24 India, China and countries in the Middle Eastern crescent will by then have tobacco contributing to more than 12% of all deaths. In India alone, the toll attributable to tobacco will rise from 1·4% in 1990 to 13·3% in 2020.24 A large component of this will be in the form of cardiovascular deaths. Thrifty gene

Increased standard of living leading to deleterious health behaviors A third reason to arouse concern is that, if population levels of CVD risk factors rise as a consequence of adverse lifestyle changes accompanying industrialization and urbanization, the rates of CVD mortality and morbidity could rise even higher than the rates predicted solely by demographic changes. Both the degree and the duration of exposure to CVD risk factors would increase as a result of higher risk factor levels, coupled with a longer life expectancy. The increase in body weight (adjusted for height), blood pressure and cholesterol levels in Chinese population samples aged 35–64 years between the two phases of the Sino-MONICA

A “programming” effect of factors promoting selective survival may also determine individual responses to environmental challenges and, thereby, the population differences in CVD. The “thrifty gene” has been postulated to be a factor in promoting the selective survival, over generations, of persons who encountered an adverse environment of limited nutritional resources.31 Although this may have proved advantageous in surviving the rigors of a spartan environment over thousands of years, the relatively recent and rapid changes in environment may have resulted in a metabolic mismatch. Thus a salt-sensitive person whose forefathers thrived despite a limited supply of salt now reacts to a salt-enriched diet with high blood pressure. It has also 95


been hypothesized that populations subjected to food scarcity have undergone selection of a gene which increases the efficiency of fat storage through an oversecretion of insulin in response to a meal. Although this favors survival in a situation of low caloric availability, a current excess of caloric intake may lead to obesity, hyperinsulinemia, diabetes and atherosclerosis. Similarly, an insulin-resistant individual whose ancestors may have survived because a lack of insulin sensitivity in the skeletal muscle ensured adequate blood glucose levels for the brain in daunting conditions of limited calorie intake and demanding physical challenges, may now respond to a high-calorie diet and a sedentary lifestyle with varying degrees of glucose intolerance and hyperinsulinemia. Although such mechanisms seem plausible, their contribution to the acceleration of the CVD epidemic in the developing countries remains speculative. Maternal–fetal exposures as a cause of midlife CVD A recently reported association which, if adequately validated by the tests of causation, may have special relevance to the developing countries is the inverse relationship between birth size and CVD in later life.32–38 The “fetal origins hypothesis” states that adverse intrauterine influences, such as poor maternal nutrition, lead to impaired fetal growth, resulting in low birthweight, short birth length and a small head circumference. These adverse influences are postulated to also “program” the fetus to develop adaptive metabolic and physiologic responses which facilitate survival. These responses, however, may lead to disordered responses to environmental challenges as the child grows, with an increased risk of glucose intolerance, hypertension and dyslipidemia in later life, with adult CVD as a consequence. Although some supportive evidence for the hypothesis has been provided by observational studies, it awaits further evaluation for a causal role. If it does emerge as an important risk factor for CVD, the populations of developing countries will be at an especially enhanced risk because of the vast numbers of poorly nourished infants born in the past several decades. The steady improvement in child survival will lead to a higher proportion of such infants surviving to adult life, when their hypothesized susceptibility to vascular disease may manifest itself. Ethnic diversity Although ethnic diversity in CVD rates, risk factor levels and risk factor interactions are evident from population studies, the extent to which genetic factors contribute is unclear. It is only after demographic profiles, environmental factors and possible programming factors are ascertained and adjusted for that differences in gene frequency or expression can be invoked as a probable explanation for 96

interpopulation differences in CVD.39 The extent to which chronic diseases, including CVD, occur within and among different populations is determined by genetic–environmental interaction, which occurs in a wide and variable array, ranging from the essentially genetic to the predominantly environmental. This is perhaps best illustrated by the knowledge gained from studies in migrant groups, where environmental changes due to altered lifestyles are superimposed on genetic influences. These “natural experiments” have been of great value in enhancing the understanding of why CVD rates differ among ethnic groups. The classic Ni-HonSan study of Japanese migrants revealed how blood cholesterol levels and CHD rates rose from Japan to Honolulu and further still to San Francisco, as Japanese communities in the three areas were compared.40 The experience gleaned from the study of south Asians, Chinese and Pima Indians further elucidates the complexities of ethnic variations in CHD.41–43 The comparison of Afro-Caribbeans, south Asians and Europeans in the UK brought out the sharp differences in central obesity, glucose intolerance, hyperinsulinemia and related dyslipidemia between the three groups, despite similar profiles of blood pressure, body mass index and total plasma cholesterol.44 However, urban–rural comparisons within India,27 as well as migrant Indian comparisons with their non-migrant siblings,45 reveal large differences in these conventional risk factors. Thus, where the environment is common but gene pools differ, the non-conventional risk factors appear to be explanatory of risk variance, whereas when the same gene pool is confronted with different environments, the conventional risk factors stand out as being of major importance. To what extent ethnic diversity in response to CVD risk factors influences the course of the CVD epidemic in different developing countries remains to be studied. However, the experience of some of the migrant groups (for example, south Asians) portends severe epidemics in the home countries as they advance in their transition.

Strategies to deal with the coronary epidemic CVD prevention Evolving concepts of risk factors Risk factor – Decades of research, embracing evidence from observational epidemiology and clinical trials, have demonstrated that CHD is multifactorial in causation. The term “risk factor” was first used in the context of CHD.46 Several such risk factors have been identified, ranging from the established “major” factors such as smoking, elevated blood cholesterol and hypertension, to the recently investigated factors such as homocysteine and lipoprotein a. A risk factor must fulfill the criteria of causality: strength of association (high relative risk or odds ratio), consistency of


association (over many studies), temporal relationship (cause preceding the effect), dose–response relationship (greater the exposure, higher the risk), biologic plausibility, experimental evidence and, very importantly, evidence from human studies. “Clinical” v “prevention” norms – The need to make “clinical” decisions related to the management of these risk factors led to a definition of threshold levels of risk and practice guidelines. These “clinical norms” erroneously came to be identified, by the health professionals as well as the community, as also representing the prevention norms. The former are defined by evidence of benefit exceeding risk when an intervention reduces a risk factor below a particular level (the net benefit being demonstrated in clinical trials specifically designed for that purpose). The latter, however, are usually identified from observational studies (long-term longitudinal prospective studies of large cohorts) and denote the optimal values of the risk factor at which the risk of developing disease is minimal. The targeting of individuals is promoted by the “clinical” approach of healthcare providers, who seek to identify persons at “high risk” of disease or its outcomes for intensive investigation and intervention. Thus thresholds are defined to categorize persons with “high cholesterol” or “high blood pressure” and to implement individualized control strategies. Attention and action above this threshold often contrast with indifference and inertia below it. As trial evidence is gathered, the clinical norms may progress towards the prevention norms, as in the case of cholesterol or hypertension, where the thresholds for intervention have been lowered dramatically in the last decade. They may, however, remain higher than the prevention norms, as clinical trials may be conducted at a stage in the natural history where the risks of prior exposure may not be completely reversible, and also because the intervention may itself be associated with some adverse effects. Thus the benefits of lowering a risk factor may appear less than those that may occur by preventing its rise in the first place. The continuum of risk – It is clear that even though lifestyle disorders afflict some individuals, they arise from causes that are widespread in the population as a whole. Risk factors such as cholesterol and blood pressure operate in a continuum of progressively increasing risk, rather than through an all-or-none relationship suggested by cut-off values. For example, a systolic blood pressure (SBP) in the range 130–139 mmHg carries a higher risk for both heart attacks and strokes than values in the range 120–129 mmHg. Whereas an SBP of 180 mmHg carries a much higher risk for an individual than 140 mmHg, the number of persons in any population who have SBP values in the range 130–139 mmHg is higher than those with values of

180 mmHg or higher. The Multiple Risk Factor Intervention Trial’s cohort study in the United States (MRFIT) revealed that of all heart attacks which are attributable to SBP, 7·2% arise from the 0·9% of the population that represents the 180 mmHg range, whereas 20·7% of all such heart attacks occur in the 22·8% which has pressures in the range 130–139 mmHg47 (Figure 9.2). Similarly, 57% of all excess deaths attributable to diastolic blood pressure occur in the range 80–95 mmHg, compared to only 15% which occur in the high range of 105–130 mmHg.

SBP (mmHg)

Excess CHD deaths (%)

Men (%) at each level of SBP

0·9

180 7·2

1·2

170–179

6·8

160–169

10·1

150–159

19·5

6·2

140–149

23·4

12·8

130–139

20·7

22·8

9·9

28·4

2·4

19·0

120–129 110–119 110

–

2·7

High blood pressure

6·1

Figure 9.2 The risk pyramid for blood pressure and coronary heart disease (CHD): baseline SBP and CHD death rates for men screened in MRFIT37 (adapted with permission from Stamler et al 36)

This dichotomy is also clearly seen in the Framingham Study on coronary risk factors.49 People with a blood cholesterol level of 300 mg/dl run three to five times the risk of CHD as people at a cholesterol level of 200 mg/dl. At cholesterol levels over 300 mg/dl, 90 out of 100 persons developed the disease in the next 16–30 years of follow up in Framingham. At cholesterol levels under 200 mg/dl the rate was 20 out of 100 during the same period. However, more than twice as many people developed CHD with cholesterol levels under 200 mg% all their lives as did those with cholesterol levels over 300 mg%. This is because a 20% fraction of a 45% segment of the population is a much larger number than a 90% fraction of a 3–5% segment of the population.49 Thus, for most causal factors there is a “risk pyramid”. Those at the top of the pyramid are at the highest individual risk of disease, but those at the lower levels account for the largest number of cases in the community because they 97


constitute the largest segment of the population. Any approach that targets only those at the highest risk produces limited gains for the community, despite conferring definite benefits to the individuals in that category. The concept that “sick individuals arise from sick populations” was propounded and proved by Geoffrey Rose.50,51 He demonstrated that risk factor “distributions” throughout the population are predictive of disease burden in that community. The mean (average) levels of a risk factor across different populations correlate with the proportions of high-risk individuals in those populations, whatever the cut-off value. Thus, as the average population blood pressure value among populations rises, the proportion of hypertensive individuals also rises. In each population there are groups who represent the extremes of the risk profile (very low risk v very high risk). However the proportion at “high risk” would be determined by the average value of that risk factor in the population. This in turn is dependent on the dominant behaviors that characterize the society at each stage of its development. Multiplicative risk – The process of identifying and estimating the independent risk associated with any single risk factor led to clinical and preventive strategies to target it in isolation. However, observational studies like Framingham and MRFIT have clearly revealed that the coexistence of multiple risk factors confers a magnified risk which is multiplicative rather than merely additive. A smoker with modest elevations of cholesterol and diastolic or systolic blood pressure is at a greater risk of coronary death than a non-smoker with severe hypertension or marked hypercholesterolemia. In the MRFIT study, a non-smoker with SBP less than 118 mmHg and a total serum cholesterol level less than 182 mg% had a 20-fold lower risk of coronary death than a smoker with a SBP exceeding 142 mmHg and a serum cholesterol exceeding 245 mg% (age adjusted CHD mortality of 3·09 v 62·11, per 10 000 person years). A smoker who has a SBP of 132–141 mmHg and a serum cholesterol of 203–220 mg% has a CHD mortality risk of 28·87 per 10 000 person years, compared to a risk of 12·36 in a non-smoker with an SBP below 118 mmHg but with a serum cholesterol exceeding 245 mg%.52 The demonstration of such multiplicative risk has led to the concept of “comprehensive cardiovascular risk” or “total risk”, quantifying an individual’s overall risk of CVD resulting from the confluence of risk factors.48 Both clinical and preventive strategies are veering away from unifactorial risk reduction to multifactorial risk modification, to reduce this overall risk in individuals as well as in populations. High-risk approach for prevention Having recognized that environmental risk factors do not affect only a few individuals in isolation but are spread 98

across populations, with a continuous rather than a threshold relationship to disease, how should that influence disease control strategies? The health policy debate, until recently, was on whether to focus the control strategies on individuals at the highest risk of disease (in view of their markedly elevated risk factor levels) or on the population as a whole (aiming to achieve modest reductions in the risk of most members of that community). The high-risk approach aims to identify persons with markedly elevated risk factors and therefore at the highest risk of disease.50 These individuals are then targeted by interventions which aim to reduce the risk factor levels. If successful, the benefits to individuals are large, because the individuals risks are large. However, as the number of persons in this high-risk category is proportionately much smaller than that in the moderate-risk group, the overall benefits to society are limited in terms of deaths or disability avoided. The strategy also does not minimize the risk for the individuals concerned. Although a fall of blood cholesterol from 300 mg% to 240 mg% does indeed reduce the risk, even this attained value poses greater risk than 200 mg%. Thus there is still a substantial residual risk, despite the impressive risk reduction owing to the change from the initial cholesterol levels. Further, this strategy is behaviorally inappropriate.50 An individual with high blood cholesterol levels may be advised to eat low-fat food, but can he strictly adhere to it if his family and friends consume a very different diet? The main advantage of the high-risk approach, however, is that physicians as well as patients are highly motivated to act, because the projected risks compel attention and the benefits of reduction appear attractive.50 Population approach for prevention In contrast, the population approach aims at reducing the risk factor levels in the population as a whole, through community action.50 Because there is a continuum of risk associated with most risk factors, this mass change will result in mass benefit across a wide range of risks. Although individual benefits are relatively small, the cumulative societal benefits are large (“the prevention paradox”). The strategy is also behaviorally more appropriate.50 If the eating habits in the community alter towards preferred consumption of foods with lower saturated fat and salt content and a greater daily intake of fresh fruit and vegetables, even the high-risk individual on a prescribed diet will find a supportive ambience which does not mark him out as a deviant from social norms. If a new generation grows up in an environment where healthy behavior is considered common practice, its average blood cholesterol level may remain below 200 mg% rather than around 240 mg%, and thus be at a lower risk than even the beneficiary of the high-risk strategy. However, the risks and benefits of such a strategy are less obvious to those in the moderate-risk range. The motivation for change is therefore not as strong as for those in the high-risk


group. The gratification of achieving readily identifiable success in high-risk individuals, through drugs or other powerful interventions, is also denied to the physicians in the population strategy, where the potential beneficiaries, though many, are faceless and nameless. Because such “anonymity of prevention” denies the pleasure of individual rescue acts, physician motivation for community counseling is neither strong nor sustained.50 Policymakers, however, can ill afford to ignore the imperatives of investing in a population approach which will pay large long-term dividends in the control of lifestyle diseases. Health professionals too must recognize the benefits of this strategy to play a strong advocacy role for health-promoting behaviors in the community. The success of the population strategy has been demonstrated both in developed countries (for example, Finland)53 and in some developing countries (for example, Mauritius).54 The North Karelia Project demonstrated large reductions in CVD mortality (50·1% in males and 63·5% in females), CHD mortality (53·4% and 59·8%) and all-cause mortality (39·5% and 40·4%) during the 20-year intervention period. These accompanied changes in CVD risk factors following community-based intervention programmes. Impressive reductions in cigarette smoking, prevalence of hypertension and mean population cholesterol levels, as well as increases in leisuretime physical activity, were noted during the period 1987–92 consequent upon lifestyle intervention programs in Mauritius. The impact of the population strategy is likely to be large, as suggested by an estimate that if every American had a diastolic blood pressure value a mere 2 mmHg lower than his or her current value, the number of heart attacks that could be prevented would exceed those that could be avoided by effectively treating every person with a diastolic pressure of 95 mmHg or higher. The corresponding benefit for preventing paralytic strokes would be 93% of those avoided by drug therapy.55 Such blood pressure changes can be effectively achieved and sustained through modest reductions in weight and salt intake or through exercise. Combining the strategies These strategies are not mutually exclusive but are synergistic, complementary and necessary. The risks and benefits demonstrated in high-risk individuals serve to educate the community about risk factors, whereas the population approach makes it easier to achieve the desired level of lifestyle change in high-risk individuals. The populationbased lifestyle-linked risk reduction approach is particularly relevant in the context of the developing countries, where it is necessary to ensure that communities currently at low risk are protected from the acquisition or augmentation of risk factors (“primordial prevention”). This is true for adults in the rural regions of most developing countries, as well as for children in all populations. It is also eminently applicable

to moderate-risk groups in urban areas, where lifestylebased risk modification will help avoid drug therapy, with its attendant economic and biologic costs. There will still be some who need such pharmacologic or technologic interventions because of their high-risk status. However, their numbers too will decrease as the risk profile of the whole community gradually shifts. Case management Despite these preventive strategies, several individuals will manifest clinical disease because risks are not totally eliminated in the community or because genetic susceptibility is strongly expressed. The success of preventive efforts will reduce their number as well as delay the age of onset of clinical events. Those who develop disease will require optimal clinical care, which can avert early death, reduce disability and ensure an adequate quality of life. This mandates early detection of disease. The cost effectiveness and safety of these diagnostic and therapeutic techniques would have to be established through appropriately designed clinical research. This scientific evidence has to be translated into practice guidelines, which then need to be widely disseminated. The rapid diffusion of these guidelines across various levels of healthcare and their sustained impact on clinical practice will ensure that the burden of cardiovascular disease in the community is mitigated through appropriate application of available knowledge. Postmyocardial infarction risk reduction through thrombolytic agents, aspirin, blockers, ACE inhibitors and statins is clearly illustrative of the benefits of such evidence-based clinical care.56–61 The decline in CVD mortality rates in industrial countries is the collective result of population-based prevention strategies improving the risk factor profile of communities, a highrisk approach of targeted interventions to protect individuals with markedly elevated risk factor levels, and case management strategies to salvage, support and sustain those presenting with clinical problems. These strategies are not diverse and divisive but are continuous and complementary in the effort to control the incidence and impact of CVD. The enormous need for evidence-based medicine in developed and developing countries CVD related expenditure in developed countries The management of CVD is often technology intensive and expensive. Procedures for diagnosis or therapy, drugs, hospitalization and frequent consultations with healthcare providers all contribute to high costs, both to those affected and to society. In developed countries they already account for 99


about 10% of direct healthcare costs, equal to between 0·5% and 1% of a country’s gross national product.2 As life expectancy increases and the duration of the therapy becomes prolonged, the costs may further escalate until preventive strategies succeed in greatly reducing the incidence of CVD. CVD related expenditure in developing countries The costs of CVD related healthcare have not been clearly estimated in the developing countries.62 However, high expenditure on tertiary care in most of these countries probably has a large contribution from CVD. As the epidemic advances many more will be affected, escalating the costs of CVD related healthcare. This may divert scarce resources intersectorally from developmental activities, and intrasectorally from the “unfinished agenda” of infectious and nutritional disorders. As the epidemic matures, the social gradient will reverse and many of the poor who are then afflicted will be unable to afford or access the expensive healthcare that CVD demands. Need for evidence-based medicine The need for cost effective prevention and case management is, therefore, urgent. These practices need to be based on the best available evidence which is generalized to the context of each developing country. Where such evidence is unavailable or insufficient to guide policy and practice, health research must quickly address those information needs. International cooperation can greatly further these efforts to acquire, appraise, analyze and apply such knowledge. Evidence from health research must do justice to the needs of public health! Evidence-based cardiovascular medicine must pursue this advocacy to secure acquittal from CVD for countries under the trial of epidemiologic transition. However, the recommendations also need to be context specific and resource sensitive, in accordance with the specific needs of different regions. The challenge for cardiovascular research is to provide for such relevant knowledge generation, and the challenge for public health and clinical practice is to provide for effective knowledge translation. The course and consequences of the global cardiovascular epidemic should not merely be predicted, but ought to be favorably altered by responding to these challenges. References 1.The World Health Report 2001. Geneva: World Health Organization, 2001. 2.The World Health Report 1997. Geneva: World Health Organization, 1997. 3.Murray CJL, Lopez AD. The Global burden of disease: A comprehensive assessment of mortality and disability from

100

disease, injuries and risk factors in 1990 and projected to 2020. Boston: Harvard University Press, 1996. 4.Reddy KS, Yusuf S. The emerging epidemic of cardiovascular disease in developing countries. Circulation 1998;97: 596–601. 5.Howson CP, Reddy KS, Ryan TJ, Bale JR (eds) Control of cardiovascular disease in developing countries. Research, development and institutional strengthening. Washington, DC: National Academy Press, 1998. 6.Yusuf S, Reddy S, Ounpuu S, Anand S. Global Burden of Cardiovascular Diseases. Part I: General Considerations, the Epidemiological Transition, Risk Factors, and Impact of Urbanisation. Circulation 2001;104:2746; Part II: Variations in Cardiovascular Disease by Specific Ethnic Groups and Geographic Regions and Prevention Strategies. Circulation 2001;104:2855. 7.American Heart Association. 2000 Heart and Stroke Statistical Update. Dallas, TX: American Heart Association, 1999. 8.Lopez AD. Assessing the burden of mortality from cardiovascular disease. Wld Hlth Stat Q 1993;46:91–6. 9.Feinleib M, Ingster L, Rosenberg H, Maurer J, Singh G, Kochanek K. Time trends, cohort effects and geographic patterns in stroke mortality. United States. Ann Epidemiol 1993;3:458–65. 10.Whelton PK, Brancati FL, Appel LJ, Klag MJ. The challenge of hypertension and atherosclerotic cardiovascular disease in economically developing countries. High Blood Press 1995;4:36–45. 11.Marmot M. Coronary heart disease: rise and fall of a modern epidemic. In: Marmot M, Elliott P, eds. Coronary heart disease epidemiology. From aetiology to public health. Oxford: Oxford University Press, 1992. 12.Cooper, Cutler J, Desvigne-Nickens P et al. Trends and disparities in coronary heart disease in the United States. Findings of the National Conference on Cardiovascular Disease Prevention. Circulation 2000;102:3137–47. 13.AR. The epidemiologic transition: a key of the epidemiology of population change. Millbank Mem Fund Q 1971;49: 509–38. 14.Bobadilla JL, Costello CA, Mitchell F (eds) Premature death in the new independent States. Washington, DC: National Academy Press, 1997. 15.Notzon FC, Komarow YM, Ermakov SP et al. Causes of declining life expectancy in Russia. JAMA 1998;279: 793–800. 16.Chenet L, Mckee M, Fulop N et al. Changing life expectancy in central Europe: Is there a single reason? J Public Health Med 1996;18:329–36. 17.Zatonski WA, McMichael AJ, Powles JW. Ecological study of reasons for sharp decline in mortality for ischaemic heart disease in Poland since 1991. BMJ 1998;317:678. 18.Omran AR. The epidemiologic transition: a key of the epidemiology of population change. Millbank Mem Fund Q 1971;49:509–38. 19.Olshansky SJ, Ault AB. The fourth stage of the epidemiologic transition: the age of delayed degenerative diseases. Millbank Mem Fund Q 1986;64:355–91. 20.Pearson TA, Jamison DT, Trejo-Gutierrez H. Cardiovascular disease. In: Jamison DT, ed. Disease control priorities in developing countries. New York: Oxford University Press, 1993.


21.Verschuren WMM, Jacobs DR, Bloemberg BPM et al. Serum total cholesterol and long-term coronary heart disease mortality in different cultures: twenty-five year follow-up of the Seven Country Study. JAMA 1995;274:131–6. 22.Suh I. Cardiovascular mortality in Korea: a country experiencing epidemiologic transition. Acta Cardiol 2001;56:75–81. 23.Unwin N, Setel P, Rashid S et al. Noncommunicable diseases in sub-Saharan Africa: where do they feature in the health research agenda? Bull WHO 2001;79: 947–53. 24.Tobacco or health: First global status report. Geneva: World Health Organization, 1996. 25.Bulatao RA, Stephens PW. Global estimates and projections of mortality by cause 1970–2015. Pre-working paper 1007. Washington, DC: Population Health and Nutrition Department, World Bank, 1992. 26.Yao C, Wu Z, Wu J. The changing pattern of cardiovascular diseases in China. Wld Hlth Stat Q 1993;46:113–18. 27.Reddy KS. Cardiovascular disease in India. Wld Hlth Stat Q 1993;46:101–7. 28.Drewnowski A, Popkin BM. The nutrition transition: new trends in the global diet. Nutr Rev 1997;55:31–43. 29.Lang T. The public health impact of globalisation of food trade. In: Shetty PS, McPherson K, eds. Diet, nutrition and chronic disease. Lessons from contrasting worlds. Chichester: Wiley, 1997. 30.Peto R. Tobacco – the growing epidemic in China. JAMA 1996;275:1683–4. 31.Thrifty genotype rendered detrimental by progress [editorial]. Lancet 1989;ii:839–40. 32.Barker DJP, Martyn CN, Osmond C, Haleb CN, Fall CHD. Growth in utero and serum cholesterol concentrations in adult life. BMJ 1993;307:1524–7. 33.Martyn CN, Barker DJP, Jespersen S, Greenwald S, Osmond C, Berry C. Growth in utero, adult blood pressure and arterial compliance. Br Heart J 1995;73:116–21. 34.Law CM, Shiell AW. Is blood pressure inversely related to birth weight? The strength of evidence from a systematic review of the literature. J Hypertens 1996;14:935–41. 35.Joseph KS, Kramer MS. Review of evidence on fetal and early childhood antecedents of adult chronic disease. Epidemiol Rev 1996;18:158–74. 36.Eriksson JG, Forsen T, Tuomilehto J, Osmond C, Barker DJP. Early growth, adult income, and risk of stroke. Stroke 2000; 31:869–74. 37.Eriksson JG, Forsen T, Tuomilehto J, Osmond C, Barker D. Fatal and childhood growth and hypertension in adult life. Hypertension 2000;36:790. 38.Eriksson JG, Forsen T, Tuomilehto J, Osmond C, Barker DJP. Early growth coronary heart disease in later life: longitudinal study. BMJ 2001;322:949–53. 39.Reddy KS. Coronary heart disease in different racial groups. In: Yusuf S, Wilhelmsen L, eds. Advanced issues in prevention and treatment of atherosclerosis. Surrey: Euromed Communications, 1996. 40.Robertson TL, Kato H, Rhoads GG et al. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California. Incidence of myocardial infarction and death from coronary heart disease. Am J Cardiol 1977;39:239–49.

41.Enas EA, Mehta JL. Malignant coronary artery disease in young Asian Indians: thoughts on pathogenesis, prevention and treatment. Clin Cardiol 1995;18:131–5. 42.Li N, Tuomilehto J, Dowse G et al. Prevalence of coronary heart disease indicated by electrocardiogram abnormalities and risk factors in developing countries. J Clin Epidemiol 1994;47:599–611. 43.Sievers ML, Nelson RG, Bennet PH. Adverse mortality experience of a southwestern American Indian community: overall death rates and underlying causes of death in Pima Indians. J Clin Epidemiol 1990;43:1231–42. 44.Chaturvedi N, McKeigue PM, Marmot MG. Relationship of glucose intolerance to coronary risk in Afro-Caribbeans compared with Europeans. Diabetologia 1994;37:765–72. 45.Bhatnagar D, Anand IS, Durrington PN et al. Coronary risk factors in people from the Indian Subcontinent living in West London and their siblings in India. Lancet 1995;345:404–9. 46.Kannel WB, Dawber TR, Kagan A, Revotskie N, Strokes J III. Factors of risk in the development of coronary heart disease – sixyear follow-up experience. Ann Intern Med 1961; 55:33–50. 47.Stamler J, Stamler R, Neaton JD. Blood pressure, systolic and diastolic, and cardiovascular risks: US population data. Arch Intern Med 1993;153:598–615. 48.WHO Expert Committee. Hypertension Control in Populations. World Health Organization Technical Report No 862. Geneva: World Health Organization, 1996. 49.Castelli WP, Anderson K, Wilson PW, Levy D. Lipids and risk of coronary heart disease. The Framingham Study. Ann Epidemiol 1992;2:23–8. 50.Rose G. Sick individuals and sick populations. Int J Epidemiol 1985;14:32–8. 51.Rose G, Day S. The population mean predicts the number of deviant individuals. BMJ 1990;301:1031–4. 52.Neaton JD, Kuller LH, Wentworth D, Borhani NO, for the Multiple Risk Factor Intervention Trial Research Group. Total and cardiovascular mortality in relation to cigarette smoking, serum cholesterol concentration, and diastolic blood pressure among black and white males followed for five years. Am Heart J 1984;108:759–69. 53.Puska P, Tuomilehto J, Aulikki N, Enkki V. The North Karelia Project. 20 years results and experiences. Helsinki: National Public Health Institute, 1995. 54.Dowsen GK, Gareeboo H, George K et al. Changes in population cholesterol concentrations and other cardiovascular risk factor levels after five years of non-communicable disease intervention programme in Mauritius. BMJ 1995; 311:1255–9. 55.Cook NR, Cohen J, Hebert PR, Taylor JO, Hennekens CH. Implications of small reductions in diastolic blood pressure for primary prevention. Arch Intern Med 1995;155:701–9. 56.Matching the Intensity of Risk Factors Management with the Hazard for Coronary Disease Events (27th Bethesda Conference). J Am Coll Cardiol 1996;27:957–1047. 57.Antiplatelet Trialists Collaboration. Collaborative overview of randomized trials of antiplatelet therapy. I. Prevention of death, myocardial infarction and stroke by prolonged antiplatelet therapy in various categories of patients. BMJ 1994;308:235–46. 58.Yusuf S, Wittes J, Friedman L. Overview of results of randomized clinical trials in heart disease. I. Treatments following myocardial infarction. JAMA 1988;260:2088–93.

101


59.Walsh JT, Gray D, Keating NA et al. ACE for whom? Implications for clinical practice of post-infarct trials. Br Heart J 1995;73:470–4. 60.Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383–9. 61.Sacks FM, Pfeffer MA, Moye LA et al. The effect of pravastatin on coronary events after myocardial infarction in patients

102

with average cholesterol levels. N Engl J Med 1996;335: 1001–9. 62.Chockalingam A, Balaguer-Vinto I (eds) Impending Global Pandemic of Cardiovascular Diseases. Challenges and opportunities for the prevention and control of cardiovascular diseases in developing countries and economies in transition (World Heart Federation). Barcelona: Prous Science, 1999.

10

Tobacco: global burden and community solutions Terry F Pechacek, Samira Asma, Nicole Blair, Michael P Eriksen

Introduction While smoking is universally known to be deadly, few are aware of precisely how deadly. In the United States, smoking is the leading preventable cause of death, killing over 440 000 people each year, wasting 5 million years of potential life and costing over $75 billion in health expenditures. Since the first Surgeon General’s Report on smoking in 1964, over 10 million Americans have died from smoking. If current trends continue, an additional 25 million Americans alive today, including 6·4 million children, will die a painful and premature death caused by smoking. While the United States statistics are appalling, the global projections are even more dire. Globally, if current trends continue, the number of people killed by tobacco will more than triple to 10 million a year by the year 2025. The only ray of hope is that the precursors for the projected global tobacco epidemic are not yet all in place. While high smoking rates among men are nearly universal, the same cannot be said for women and teens. Thus, despite the unprecedented toll of tobacco and the gloomy projections, there is the potential for prevention. This chapter will explore that potential, particularly for coronary heart disease, by: 1. 2. 3.

examining the current global burden of tobacco and future projections reviewing the mixed evidence for community-based tobacco control interventions proposing a new and dynamic model for communitybased tobacco control, based on state innovations, proven to be effective in the United States, that may be able to be applied throughout the world.

Current global burden of tobacco and future projections Worldwide, the only two major causes of death whose effects are now increasing rapidly are HIV and tobacco. If current smoking patterns persist, there will be about one billion deaths from tobacco during the twenty-first century, compared to “only” about 0·1 billion (100 million) during the

whole of the twentieth century. About half of these deaths will be in middle age (35 to 69) rather than old age, and those killed by tobacco in middle age lose, on average, more than 20 years of non-smoker life expectancy.1 Tobacco use is estimated to have caused about 4 million deaths a year, more or less evenly split between developed and developing countries. These numbers reflect smoking patterns several decades ago, and worldwide cigarette consumption has increased substantially over the past half century.2 Currently, about 30% of young adults become persistent smokers, and relatively few quit. The main diseases by which smoking kills people are substantially different in America, where vascular disease and lung cancer predominate;1 in China, where chronic obstructive pulmonary disease causes even more tobacco deaths than lung cancer;3,4 and in India, where almost half the world’s tuberculosis deaths take place and the ability of smoking to increase the risk of death from TB may well be of particular importance.5,6 There are already a billion smokers, and by 2030 about another billion young adults will have started to smoke. If current smoking patterns persist, worldwide mortality from tobacco is likely to rise from about four million deaths a year currently to about 10 million a year around 2030 and will rise somewhat further in later decades. This means that tobacco use will cause about 150 million deaths in the first quarter of the century and 300 million in the second quarter. Predictions beyond that are inevitably speculative, but if over the next few decades a quarter to a third of the young adults become persistent smokers and about half are eventually killed by their habit, about 15% of adult mortality in the second half of the century will be due to tobacco, implying some 600 million to 900 million tobacco deaths between 2050–2099.7 First, globally in 1995, 29% of the world’s population aged 15 years and over smoked daily (Table 10.1). Low-income and middle-income countries whose populations account for four fifths of the global adult population, accounted for 82% of the world’s smokers. East Asia and the Pacific, which includes China, accounted for 36% (43 million) of all smokers, but only 32% of the population aged 15 years and over. Overall, smoking prevalence was highest in Europe and Central Asia at 40% and lowest in Sub-Saharan Africa at 18%. 103


For both males and females, there was wide variation in smoking prevalence between regions. The prevalence of smoking amongst males was highest in East Asia and the Pacific, and in Europe and Central Asia, at about 60% in each case, and lowest in Sub-Saharan Africa at 29%. Among females, the prevalence of smoking was highest in Europe and Central Asia at 26% and lowest in South Asia at 5% (for cigarettes and bidis combined) and Middle East and North Africa at 6%. Second, the prevalence of smoking amongst males was higher overall for men (47%) than for women (11%). WHO data at country level suggest that the proportion of men who smoke is well above 50% in many low-income and middle-income countries: 82% in Indonesia, 78% in the Philippines, 75% in Cuba, 72% in China.2

● ● ● ●

Globally, males account for four in five of all smokers. The majority of epidemiologic studies suggest that individuals who avoid starting to smoke in adolescence or young adulthood are unlikely ever to become smokers. Nowadays, the overwhelming majority of smokers start before age 25, often in childhood or adolescence (Figure 10.1); in highincome countries, eight out of 10 begin in their teens. In middle- and low-income countries for which data are available, it appears that most smokers start in their early US (both sexes, born 1952–61)

100

US (both sexes, born 1910–14)

80 Cumulative uptake (%)

China (males, 1996)

India (males, 1995)

60

40

20

0 15

20

25

Age

Figure 10.1 Smoking initiation age in China, India and the United States. Source: Gupta 1996; USDHHS 1989 and 1994; Chinese Academy of Preventive Medicine 1997

104

twenties, but the trend is towards younger ages. For example, in China between 1984 and 1996, there was a significant increase in the number of young men aged between 15 and 19 years who took up smoking. A similar decline in the age of starting has been observed in high-income countries. Tobacco: a risk factor for coronary heart disease It is well established that prolonged smoking is an important cause of chronic disease. Prolonged smoking causes many diseases in addition to lung cancer, notably other cancers and chronic respiratory and cardiovascular diseases. However, the toll of death and disability from smoking outside the high income countries has yet to be felt. This is because the diseases caused by smoking can take several decades to develop. Even when smoking is common in a population, the damage to health may not yet be visible. The alarming size of the hazards now observable in populations that have been smoking for many decades. Thus in the first 20 years of follow up of the British doctors, cohort (1951–71), smokers had, on average, about a 1·5 to twofold higher death rate at each age, similar to the excess reported in other studies around that time (see Table 10.1). With longer duration of smoking, death rates of smokers have increased substantially so that during the second period of follow up (1971–91), smokers in middle age had a threefold higher death rate than non-smokers. A similar excess mortality ratio was found in the CPS-II cohort based on follow up in the latter half of the 1980s. These relative risks suggest that, on average, a smoker who begins smoking in young adult life and continues to do so has at least a 50% chance of eventually being killed by tobacco, either in middle age or in old age. The evidence from these two studies on the diseasespecific risks associated with smoking are similar.8 Current smokers have about a 20-fold higher death rate from lung cancer than never smokers, among whom lung cancer death rates have remained low and constant. There is epidemiologic evidence to suggest that this is also the case in other populations. For example, based on the two American Cancer Society studies with follow up to 1959–65 and 1982–86 respectively, lung cancer death rates among lifelong non-smokers were remarkably constant at 15·4 and 14·7 per 100 000 (age-standardized) for men, and 9·6 and 12·0 for women; the rates for current smokers were 187·1 and 341·3 for men, and 26·1 and 154·6 for women.9 Smokers also incur a 10–20-fold excess mortality from chronic obstructive lung disease (primarily chronic bronchitis and emphysema), and a risk of death from major vascular diseases that is about twice that of non-smokers. The excess mortality of smokers from vascular disease is particularly noteworthy. Vascular disease death rates are typically much higher than those for cancer or other causes

Tobacco: global burden and community solutions

Table 10.1 Prevalence of smoking among adults aged 15 and over, by World Bank region, 1995 World Bank Region

East Asia and Pacific Europe and Central Asia Latin America and Caribbean Middle East and North Africa South Asia (cigarettes) South Asia (bidis) Sub-Saharan Africa Low-income & Middle-income High-income World

Smoking prevalence (%)

Total smokers

Males

Females

Overall

(millions)

(% of all smokers)

61 57 40 44 21 21 29 49 38 47

4 26 21 5 1 4 9 9 21 11

33 40 30 25 11 13 18 29 29 29

413 145 95 40 88 99 59 939 205 1143

36 13 8 3 8 9 5 82 18 100

6.3 times as common in smokers as in non-smokers aged 30-39

30–39

4·7

40–49 Age

6·3

3·1

50–59

2·5

60–69

Not caused by smoking Excess with cigarette use

1·9

70–79 0

1

2

3

4 Ratio

5

6

7

8

Figure 10.2 Source: Based on the ISIS study of over 10 000 UK heart attacks, Parish et al. Cigarettes smoking, Lar yields, and non-fatal myocardial infarction: 14,000 cases and 32,000 controls in the United Kingdom. BMJ 1994;311:471–710

associated with smoking. Cardiovascular diseases (especially ischemic heart disease and stroke), therefore, contribute more to smoking-attributable deaths at a population level than do other causes, including lung cancer for which the relative risk is much higher, although this pattern will change as cardiovascular disease mortality declines. Finally, it is worth noting that the all-age excess mortality ratio of about 2 from cardiovascular diseases masks a very significant age gradient in relative risks. This is clearly shown in Figure 10.2 based on a large (46 000 persons) case–control study carried out in the United Kingdom. At younger ages (50 years), smokers have a five to six times higher death rate than non-smokers, with the relative excess declining with age. What these data suggest is that if a smoker dies from vascular disease before about the age of 50 years, there is a 70–80% chance that smoking caused it, and that this is the prinicipal mechanism through which smoking causes a threefold excess mortality rate in middle age.

Cigarette smoking is only one of several causative factors that produce disease. This is especially true for ischemic heart disease where smoking interacts synergistically with other factors such as hypercholesterolemia and hypertension to greatly increase risk of heart disease. Evidence suggests that the independent risk attributable to smoking is comparable to that of other major risk factors.10 This interaction with dietary parameters probably explains the currently lower proportions of ischemic heart disease attributable to smoking in populations such as China where lowfat diets have predominated.3 The extent to which smoking is responsible for deaths from diseases other than lung cancer varies substantially from one population to another. For example, smoking is particularly cardiotoxic for people who already have other risk factors such as high blood cholesterol. The range of other diseases that are caused by smoking is so extensive that the influence of other specific risk factors may effectively average 105


out even in different populations. For example, although in many developing countries, cholesterol levels are low (limiting the cardiotoxic effects of tobacco), a high prevalence of respiratory diseases may greatly increase the pulmonary vulnerability to tobacco.11 Smokers have twice the risk of heart attack compared with non-smokers. Smoking is also a major risk factor for sudden death from heart attack, with smokers having two to four times the risk of non-smokers. The risk increases with the number of cigarettes smoked. Overall, cigarette smokers have coronary heart disease (CHD) rates 70% higher than those of non-smokers, with heavy smokers dying from CHD at a rate two to three times that of non-smokers.1 In addition, recent epidemiologic evidence shows that never-smokers exposed to environmental tobacco smoke (ETS) have an increased risk not only for lung cancer but also for cardiovascular disease. Two recent prospective trials12–14 and meta-analyses15 estimate the relative risk for cardiovascular diseases at 1·2 to 1·3 individuals exposed to ETS.12–15 Of the deaths caused by ETS, the number of deaths from heart disease is about three times the number of non-cardiac deaths16. Cardiovascular deaths account for a significant portion of adult deaths in all countries. Worldwide, slightly more than 50 million people are estimated to have died in 1990, 53% of whom were males. Ischemic heart disease (IHD) was the leading cause of death worldwide, accounting for just under 6·3 million deaths – 2·7 million in established market economies (EME) and formerly socialist economies of Europe (FSE); 3·6 million in the developing regions. Stroke was the next most common cause of death (4·38 million deaths – almost 3 million in developing countries), closely followed by acute respiratory infections.17 Of the various coronary heart disease pathologies, IHD and stroke predominate in the developed regions, accounting for 75–80% of all cardiovascular deaths. Stroke is proportionately more important as a cause of cardiovascular disease death in FSE (31%) than in EME (25%). Rheumatic heart disease is estimated to cause between 1% and 6% of all CHD deaths in the developing regions (and about 2·4% globally). The category labeled as inflammatory heart disease (pericarditis, endocarditis, myocarditis, and cardiomyopathies) accounts for similar proportions of CHD deaths, being highest in SubSaharan Africa (SSA) (7·8%). It is also worth noting the substantial contribution of IHD in all developing regions, ranging from 52% of cardiovascular deaths in India to 26% in SSA. Stroke, on the other hand, is by far the leading cause of cardiovascular deaths in China and SSA, causing roughly half of all coronary heart disease deaths in 1990.18

Future projections Policy makers must be concerned not so much by the current mortality from past smoking patterns, but by the much 106

larger death rates that are projected in coming decades as a result of current smoking, especially for low- and middleincome countries. Smoking-attributable deaths are projected to increase for two reasons: first, increases in the susceptible population size; and second, increase in age-specific disease rates. For example in China, male per capita consumption of cigarettes rose 10-fold between 1952 and 1992. The incidence of lung cancer in China has increased more than sixfold during the period 1970 to 1980,19 and is likely to increase 7·5-fold in the near future. During the same period, the population that will contract lung cancer will increase fourfold. The net result is that 30 000 lung cancer deaths per year in 1975 will increase to 90 000 per year by 2025. Tobacco will cause 0·5 billion deaths among smokers alive today. At some point in the second decade of the twenty-first century, annual deaths from tobacco will average 10 million a year. This total may appear earlier or later. Depending on smoking patterns, there will be about 450 million tobacco deaths between 2000 and 2050.8 Projections beyond 2050 are more uncertain. If the proportion of people taking up smoking continues, as at present, to be between one quarter and one third of young adults then, given population growth, an additional 500 billion tobacco deaths are expected in the second half of the twenty-first century. Thus, in the twenty-first century overall, tobacco would be expected, on current patterns, to kill about one billion people, or ten times as many people in the twentieth century.7 Direct estimates for China based on retrospective and prospective studies3,4 suggest that, on current patterns, smoking may account for one in three of all adult male deaths in China, or about 100 million of the 300 million Chinese males now aged 0–29. Annual tobacco deaths will rise to 1 million before 2010 and 2 million before 2025, when young adults of today reach old age. Similar preliminary estimates for India based on large retrospective and prospective studies suggest that about 30% of all male deaths in middle age are attributable to smoking and about 80 million Indian males currently aged 0–34 will eventually be killed by tobacco. Projections of tobacco mortality based on econometric models by Murray and Lopez suggest that there will be 8·3 million tobacco-attributable deaths per year in 2020. These researchers have predicted elsewhere that global deaths attributed to tobacco would rise from 6% of all deaths in 1990 to about 12% in 2020.18 Worldwide, a very large increase in deaths from noncommunicable diseases (group 2) is expected, with a rise in annual mortality from an estimated 28·1 million deaths in 1990 to 49·7 million in 2020. Conversely, annual mortality from communicable maternal, perinatal, and nutritional disorders (group 1) is predicted to decline from 17·2 million in 1990 to 10·3 million in 2020 (Figure 10.3). It is of interest to examine how DALYs (disability adjusted life years is a measure of life lost due to disability or premature


death) from various leading causes are expected to change over the next three decades (Figure 10.4). Figure 10.5 shows the change in cause of mortality. IHD is projected to be the leading cause of disability and death by 2020. It has been

Deaths (millions)

60

40

20

0

1990 CVD

2000 Cancer

2010

Year Respiratory

2020

Digestive

Other

Figure 10.3 Baseline projections of deaths from group 2 causes, world, 1990–2020 (from Murray and Lopez, 1996, with permission)

plausibly predicted that the current global total of about 3 million deaths per year from tobacco (2 million developed, 1 million developing) would reach approximately 10 million deaths per year (3 million developed, 7 million developing) during the second quarter of next century (Figure 10.6). This would mean that over 200 million of today’s children and teenagers will be killed by tobacco, as well as a comparable number of today’s adults, predicting that a total of about half a billion of the world’s population today will be killed by tobacco. About 250 million will die in middle age (35–69), with each person losing about 20 years of life.20 In terms of DALYs, the contribution of tobacco is projected to increase to account for nearly 9% of worldwide burden (18·2% of burden in developed countries and 7·7% in developing countries) in 2020 (Figure 10.7). Tobacco is also projected to cause about 12% of deaths worldwide (17·7% of deaths) in developed countries and 10·9% in developing countries) by 2020 (Figure 10.8). DALYs from cancers are expected to rise from 5·1% to 9·9% of the worldwide total in 2020. The proportionate share of the global burden of disease due to cardiovascular diseases is projected

1990

2020 (Baseline scenario) Disease or injury

Disease or injury

Figure 10.4 permission)

Lower respiratory infections

1

Diarrheal diseases

2

Conditions arising during the perinatal period Unipolar major depression Ischemic heart disease

3

Cerebrovascular disease

6

Tuberculosis

7

Measles

8

Road traffic accidents

9

Congenital anomalies

10

Malaria

11

Chronic obstructive pulmonary disease Falls

12

Iron-deficiency anemia

14

Anemia

15

4 5

1

Ischemic heart disease

2

Unipolar major depression

3


4


5

Chronic obstructive pulmonary disease

6 7

Lower respiratory infections Tuberculosis

8

War

9

Diarrheal diseases

10 HIV 11 Conditions arising during the perinatal period 12 Violence 13 Congenital anomalies

13

14 Self-inflicted injuries 15 Trachea, bronchus and lung cancers

16

19

17

24

19

25

28

37

33

39

Change in rank order of DALYs for the 15 leading causes, world, 1990–2020 (from Murray and Lopez, 1996, with

107


1990

2020 (Baseline scenario) Disease or injury

Disease or injury

Million deaths/year

Figure 10.5 permission)

12 10 8 6 4 2 0 1965

Ischemic heart disease

1

1 Ischemic heart disease


2

2 Cerebrovascular disease

Lower respiratory infections Diarrheal diseases

3

Conditions arising during the perinatal period Chronic obstructive pulmonary disease

5

3 Chronic obstructive pulmonary disease 4 Lower respiratory infections 5 Trachea, bronchus and lung cancers

Tuberculosis

7

7 Tuberculosis

Measles

8

8 Stomach cancer


9

9 HIV

Trachea, bronchus and lung cancers Malaria

10 11

12 Cirrhosis of the liver

Self-inflicted injuries

12

13 Liver cancer

Cirrhosis of the liver

13

14 Violence

Stomach cancer

14

15 War

Diabetes mellitus

15

4

6

6 Road traffic accidents

10 Self-inflicted injuries 11 Diarrheal diseases

16

16

20

19

21

27

30

29

Change in rank order of deaths for the leading 15 causes, world, 1990–2020 (from Murray and Lopez, 1996, with

Lessdeveloped countries Developed countries 1995

1990 2020

2·5

World

8·9

12·1

Developed

18·2

2025

Figure 10.6 Annual deaths attributed to tobacco (WHO Program on Substance Abuse; WHO, 1995 A48/9)

to rise from 11·1% to 14·75%.17 In conclusion, tobacco is projected to be the leading cause of death and disability globally.

1·4

Developing

7·7

0

5

10 15 Percentage of all DALYs

20

25

Figure 10.7 Tobacco as a cause of DALYs, 1990 and 2020 (WHO Program on Substance Abuse)

Community solutions The relationship between smoking and cardiovascular morbidity and mortality was extensively documented throughout the last half of the twentieth century. Therefore, 108

the reduction of tobacco use within populations has been a recommended strategy in the primary and secondary prevention of cardiovascular diseases for many years.21 However, the development and testing of specific strategies to implement


1990 2020

6

World

12·3

14·5

Developed

17·7

3·7

Developing

10·9

0

5

10 15 Percentage of all deaths

20

25

Figure 10.8 Tobacco as a cause of death, 1990 and 2020 (WHO Program on Substance Abuse)

this recommendation has proceeded more slowly. During this interval, there has been a paradigm shift from an individual or clinical approach to smoking prevention and cessation to a more public health or population-based approach.22–24 Community-based cardiovascular prevention trials started in the early 1980s were conducted during this shift in paradigm.25 Results from all of these trials showed significant declines in the prevalence of smoking overall; however, the declines in the intervention communities did not exceed the declines in the comparison communities by a statistically significant amount in several of the trials 26–28 nor in a joint analysis of the three US community trials.29 The community-based cardiovascular prevention trials initiated in the 1980s recognized that the critical behaviors related to cardiovascular risk (for example, diet, exercise, smoking) all involved individual choices but also involved societal or cultural barriers and enticements, monetary and opportunity costs, local and regional policies, and other communitywide factors.25 The intervention methods which were developed in these trials shared many common elements, but largely were restricted from applying an ecological and policy oriented health promotion approach that combines educational, political, regulatory, and organizational supports for changes in the target health behaviors.30 Green and Richard posit that the early community-based cardiovascular prevention trials could rely on an expansion of the traditional health education models which would initiate change in early adopters in societies.30 However, as the rate of diffusion of the adoption of heart-healthy lifestyle changes (including smoking cessation) accelerated, these more traditional approaches lost efficacy. The smoking cessation results from the community-based cardiovascular prevention trials initiated in the 1980s must be viewed as modest at best. While the Stanford Five-City Project observed a significantly greater decline (13%) in smoking rates in the intervention communities among

the cohort samples,28,31 no effect on smoking rates was observed in the cross-sectional surveys by end of treatment28,31 or at the follow up during which the comparison communities were declining somewhat, but not significantly, more rapidly than the intervention communities.32 In the Minnesota Heart Health Project, the long-term smoking cessation results were mixed, with evidence of an intervention effect only for women in cross-sectional survey data.26,33 Unexpectedly strong secular declines in smoking prevalence, especially among men, were observed in comparison communities. In the Pawtucket Heart Health Program, the prevalence of cigarette smoking declined slightly, but not significantly, more in the comparison community.27 More recently, the German Cardiovascular Prevention Study has reported more encouraging treatment effects for smoking, observing a 6·7% decline in smoking, with the strongest effect in men.34 Among men, the prevalence of smoking among 25–69 year olds declined from 41·8% in 1985 to 39·2% in 1991 in the national reference sample, in comparison with the significantly greater decline from 44·5% to 37·4% in the intervention regions. This result is consistent with the diffusion model posited by Green30 that the largely individually oriented health educational approaches applied in the community-based cardiovascular prevention trials initiated in the 1980s have their largest impact among populations who are at the earlier stages of adoption of the recommended preventive lifestyle. In addition to the community-based cardiovascular prevention trials initiated in the 1980s, the Community Intervention Trial for Smoking Cessation (COMMIT) was started in the late 1980s. COMMIT focused solely on smoking cessation and built upon the initial experience in the ongoing cardiovascular prevention trials. Additionally, COMMIT was planned as a randomized community trial with 11 pairs of communities and had adequate power to detect relatively small intervention effects.35 The modest effects observed in this trial were very sobering for the public health community. No cessation effect was observed for the “heavy” smokers (defined as smoking 25 or more cigarettes per day at baseline) for whom the trial was specifically designed. Among the evaluation cohorts of light-tomoderate smokers, a significantly greater quit rate (30·6% v 27·5%) was observed over the 4 year intervention period, with the effect strongest among the less educated residents of the communities.36,37 Overall, the prevalence of smoking declined slightly, but non-significantly, more in the intervention communities (3·5 percentage points) than in the comparison communities (3·2 percentage points). While the COMMIT intervention protocol sought to apply the best smoking cessation strategies available, investigators were limited in their ability to be involved in many of the ecological and policy oriented health promotion strategies which Green and others25,30 recommend due to the federal sources 109


of funds for the study. While an intervention “receipt index” of the strategies applied significantly correlated with quit rate differences across the 11 community pairs among the light-to-moderate smokers, process data showed that implemented protocol did not change many important intermediate variables (for example, MD/DDS counseling rates, worksite smoking bans, public attitudes toward smoking). Several reviewers have provided some perspectives on the modest smoking cessation effects which have been observed in these community trials.38–40 Common themes are: (1) the difficulty in observing intervention effects relative to the large secular declines in cardiovascular risk factors, including smoking, occurring during the period when the trials were implemented, and (2) the need for a more comprehensive health promotion approach to be applied. Concurrent with the implementation of these intervention trials, a broader national movement to reduce tobacco use emerged with a focus on the principles of health promotion. This concept, which included an organized approach to changing social, economic, and regulatory environments, emerged as a more effective mechanism for population behavior change than traditional health education, and included mobilization at the national, state, and local level.41 In 1991, the National Cancer Institute launched the American Stop Smoking Intervention Study for Cancer Prevention (ASSIST) program as a 7 year demonstration project. ASSIST included 17 states, and was the largest tobacco control project in the United States ASSIST was predicated on a coalition model, and was designed to demonstrate that a comprehensive, coordinated intervention effort could significantly reduce smoking and tobacco use. In 1993, CDC began the Initiatives to Mobilize for the Prevention and Control of Tobacco Use (IMPACT) program, which provided funding for 32 states to build capacity in state health departments to conduct effective tobacco control. Based on the lessons learned from ASSIST, IMPACT, and large state programs such as those of California and Massachusetts, the National Tobacco Control Program (NTCP) was developed to support all 50 states, seven territories, and the District of Columbia plan, establish, and evaluate comprehensive tobacco prevention and control programs.41 Through evidence-based analyses of California and Massachusetts, in-depth involvement with settlement States, and published evidence of effective tobacco control strategies, the federal government has set forth “best practices” recommendations for state-based programs which contain the following elements42: ● ● ●

110

community programs to reduce tobacco use chronic disease programs to reduce the burden of tobacco-related diseases school programs

● ● ● ● ● ●

enforcement statewide programs counter-marketing cessation programs surveillance and evaluation administration and management.

Because tobacco use is ultimately an individual behavior, educational and clinical public health approaches have historically been individually focused. In the area of tobacco use prevention among youth, this has been particularly true. In the past two decades, however, excellent social-psychological approaches have been applied to school-based prevention programs.41 Evidence shows that school-based smoking prevention programs that identify social influences to smoke and teach skills to resist those influences have demonstrated consistent and significant reductions in adolescent smoking prevalence, and that larger-scale implementation of intensive interventions can achieve long-term reductions in cigarette smoking among young people.41 The durability of this effect is enhanced by community wide programs that involve parents, mass media, community organizations, and other elements of an adolescent’s social environment.42 Educational strategies, conducted in conjunction with community and media-based activities, can postpone or prevent smoking onset in 20–40% of adolescents.41 Unfortunately, the full range of recommended community wide efforts to modify the social environments of adolescents,42 including removal of pervasive imagery-based pro-tobacco advertising, significant tobacco tax increases, enhanced enforcement of minors’ access laws, and well financed and sustained youth oriented counter-advertising campaigns, need to be applied in conjunction with experimentally tested school-based tobacco use prevention curricula and tobacco-free school policies. The efficacy of a comprehensive approach to youth tobacco use prevention was originally demonstrated in Massachusetts and California, who funded their programs with dedicated excise tax dollars. During the period of the 1990s when smoking rates among the youth in the United States were consistently increasing, rates in Massachusetts and California appear to have risen more slowly43 and even declined among 7–8 graders in Massachusetts.44 With the influx of revenues resulting from state settlements with the tobacco industry and increases to state tobacco excise taxes, additional states, such as Minnesota, Florida, Arizona, and Oregon have also been able to implement comprehensive tobacco control programs.41 In many ways, efforts to assist adult smokers to quit smoking have made the slowest progress in the paradigm shift from the clinical to the public health model. However, advancements in treating tobacco use and nicotine addiction were summarized in a recent guideline: Treating tobacco use and dependence: a clinical practice guideline, published by the


US Public Health Service. The guideline provides a blueprint to healthcare professionals and health insurance providers for implementing appropriate medical services that will help treat nicotine addiction. Less intensive interventions, as simple as physicians advising their patients to quit smoking, can produce cessation rates of 5–10% per year. More intensive interventions, combining behavioral counseling and pharmacologic treatment, can produce 20% to 25% quit rates in one year.45 The most significant and sustained declines in population levels of cigarette consumption have been observed in states where changes in the social environments rather than enhanced clinical services have been the focus of the programs.46 For example, studies have found that moderate or extensive laws for clean indoor air are associated with a lower smoking prevalence and higher quit rates.41 There is clear and compelling scientific evidence which demonstrates that increasing the price of cigarettes is an effective way to prevent smoking initiation among youth, promote smoking cessation among adults, and reduce cigarette consumption among continuing smokers.41 Therefore, because increased excise taxes increase the price of cigarettes, they provide a cost effective short-term strategy to reduce tobacco use. Research indicates that for every 10% increase in price, overall smoking rates would decrease by 3–5%, and as high as 7% among youth.41 Studies of smokeless tobacco products suggest that increasing their prices would reduce the prevalence of smokeless tobacco use as well.41 Even greater decreases can be achieved when an adequately funded comprehensive tobacco prevention and control programs are combined with a price increase. This has been demonstrated in California, where a tobacco control program has been funded by excise tax revenues since 1989, and tobacco rates have declined at rates two or three times faster than the rest of the country. California also has the distinction of being the first state to demonstrate a reduction in tobacco-related deaths. The incidence of lung cancer in California has declined significantly faster than in other parts of the United States and this state has also seen dramatic declines in cardiovascular disease death rates.47 Tobacco products have been largely unregulated in comparison to other consumer products. While the importance of nicotine addiction is now well recognized as a factor maintaining tobacco use behaviors,48 regulatory efforts to decrease the addictiveness of the product are only now emerging.49 Smokers receive very little information regarding chemical constituents when they purchase a tobacco product. Without information about toxic constituents in tobacco smoke, the use of terms such as “light” and “ultra light” on packaging and in advertising may be misleading to smokers. Also, because cigarettes with low tar and nicotine contents are not substantially less hazardous than higheryield brands, consumers may be misled by the implied promise of reduced toxicity underlying the marketing of such brands.41

Currently all 50 states and the District of Columbia have tobacco control programs in place that have the potential to achieve positive results in reducing tobacco use. CDC has synthesized an evidence-based comprehensive framework for statewide programs to reduce tobacco use. The framework integrates four program goals with four program components; optimally, each of the goals would be fully addressed in the implementation of each of the components, within each of the Best practices guidelines. The program goals for reducing tobacco use statewide include: ● ● ● ●

Prevent initiation among young people. Promote quitting among adults and youth people. Eliminate exposure to environmental tobacco smoke. Identify and eliminate disparities among population groups.

The program components for reducing tobacco use statewide include: ● ● ● ●

community interventions counter-marketing program policy and regulation surveillance and evaluation.

Aggressive and comprehensive tobacco control programs in a number of states have produced substantial declines in cigarette use. The findings from multiple states were reviewed by the US Surgeon General in the report, Reducing tobacco use.41 For example: ●

●

●

In California, home to one of the longest-running tobacco control programs, the overall prevalence of tobacco use has declined at nearly twice the rate of that in the United States. The declines in the rates of lung cancer and heart disease have also been significantly faster than in other parts of the country. California is also the first state to experience a decrease in tobacco-related deaths. In 1992, Massachusetts initiated a comprehensive statewide tobacco control program. From 1992–2000, per capita consumption declined by 36%, when the rate of decline in the remaining 48 states was only 16%. A decline in smoking prevalence among adults was also greater than in the rest of the country (excluding California). From 1995–1999, smoking declined by 70% among 6th graders, and by 38% among 7th and 8th grade students. Florida’s tobacco control program, which combined a counter-marketing media campaign, community-based activities, education and training, and an enforcement program, in concert with a state excise tax increase, has effectively reduced teen tobacco use. Among middle school students, tobacco use declined by 47%, from 18·5% in 1998 to 9·8% in 2001. Among high school students, current cigarette use declined by 30%, from 27·4% in 1998 to 9·0% in 2001. 111


●

●

With the support of a dedicated excise tax, Arizona was able to begin funding a comprehensive tobacco control program in 1996 that includes all nine Best practices components. From 1996 to 1999, the proportion of healthcare providers encouraging patients to quit smoking increased significantly. Also during this time, smoking prevalence has declined significantly in women and men, whites and Hispanics, and people with low income and low education. With the support of a dedicated excise tax, Oregon launched a comprehensive statewide tobacco control program in 1997. From 1997–1999, Oregon experienced a 2·3% decline in consumption from 1996 to 2001, and in prevalence of smoking among adults, pregnant women, and youth. By following the lessons learned by more experienced states, Oregon was able to operate more efficiently, and has seen reductions in prevalence in spite of spending less than the Best practices minimum guidelines. This “Oregon Model” is now being quickly diffused out to other states to guide the development of newly funded programs. The Oregon program included an implementation of CDC’s Guidelines for school health programs to prevent tobacco use and addiction in 30% of their schools. This demonstration found that a comprehensive school-based tobacco prevention program that includes tobacco-free school policies and community involvement as one component of a statewide tobacco program may contribute to reductions in current smoking among 8th graders. Also, the significantly greater declines in smoking prevalence in the schools that rated high and medium on implementation criteria emphasize the importance of monitoring activity in funded programs and the need for on-going assistance to facilitate implementation of evidence-based recommendations.

As results are obtained from these most recent states as well as continuing data from California and Massachusetts, our understanding of the potential effectiveness of the full multicomponent population-based approach to tobacco prevention and control will be expanded. However, the data already sufficient for the US Surgeon General to conclude that if the recommended intervention strategies were fully implemented, rates of tobacco use in the US could be cut in half by the year 2010.41 References 1.Peto R, Lopez AD, Boreham J, Thun M, Heath C Jr. Mortality from smoking in developed countries 1950–2000. Indirect estimates from National Vital Statistics. Oxford: Oxford University Press, 1994. 2.World Health Organization. Tobacco or health: global status report. Geneva: WHO, 1997.

112

3.Liu B-Q, Peto R, Chen Z-M. Emerging tobacco hazards in China: 1 Retrospective proportional mortality study of one million deaths. BMJ 1998;317:1411–22. 4.Niu SR, Yang G-H, Chen Z-M. Emerging tobacco hazards in China: 2. Early mortality results from a prospective study. BMJ 1998;317:1423–4. 5.Gajalakshmi CK, Peto R. Tobacco epidemiology in the state of Tamil Nadu, India. The Proceedings of the XV Asia Pacific Cancer Conference in Chennai, 1999. 6.Gupta PC, Mehta HC. Cohort of all cause mortality among tobacco users in Mumbai, India. Bull World Health Organ 2000;78(Suppl.7):877–83. 7.Peto R, Lopez AD. The future worldwide health effects of current smoking patterns. In Koop EC, Pearson CE and Schwarz RM eds. Global health in the 21st century. New York: Jossy-Bass, 2000. 8.Peto R, Chen ZM, Boreham J. Tobacco: the growing epidemic. Nat Med 1999;5:15–17. 9.Thun MJ, Day-Lally C, Myers DG et al. Trends in tobacco smoking and mortality from cigarette use in cancer prevention, Studies I (1959 through 1965) and II (1982 though 1988). In: National Cancer Institute. Changes in cigarette-related disease risks and their implication for prevention and control. Smoking and tobacco control, Monograph 8. Bethesda, MD: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute (NIH Pub No 97–4213), 1997. 10.US Department of Health and Human Services. Reducing the health and consequences of smoking: 25 years of progress (A report of the Surgeon General). Atlanta, GA: US Department of Health and Human Services, Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention of Health Promotion, Office on Smoking and Health, 1989. DHHS Publication (CDC) 89–8411. 11.US Department of Health and Human Services. The health consequences of smoking: cardiovascular disease. (A report of the Surgeon General.) DHHS publication PHS 84-50204. Rockville, MD: Public Health Service Office on Smoking and Health, 1984. 12.Steenland K, Thun M, Lally C, Heath C. Environmental tobacco smoke and coronary heart disease in the American Cancer Society CPS-II cohort. Circulation 1996;94:622–8. 13.Kawachi I, Colditz GA, Speizer FE et al. A prospective study of passive smoking and coronary heart disease. Circulation 1997;95:2374–9. 14.Howard G, Wagenknecht LE, Burke GL et al. for the ARIC investigators. Cigarette smoking and progression of atherosclerosis: the Atherosclerosis Risk in Communities (ARIC) study. JAMA 1998;279:119–24. 15.Glantz SA, Parmley WW. Passive smoking and heart disease: epidemiology, physiology, and biochemistry. Circulation 1991; 83:1–2. 16.Glantz, SA, Parmley WW. Passive and active smoking: a problem for adults. Circulation 1996;4:596–8. 17.Murray CJL, Lopez AD. Alternative projections of mortality and disability by cause 1990–2020: global burden of disease study. Lancet 1997;349:1502. 18.Murray CJL, Lopez AD, eds. The global burden of disease. A comprehensive assessment of mortality and disability from diseases, injuries, and risk factors in 1990 and projected to 2020. Cambridge, MA: Harvard University Press, 1996.


19.Sidel R, Sidel VW. The health of China. Boston: Beacon Press 1982. 20.Peto R, Lopez AD, Boreham J, Thun M. Health C Jr, Doll R. Mortality from smoking worldwide. Br Med Bull 1996;52: 12–21. 21.World Health Organization. Report of the WHO expert committee on smoking control. Controlling the smoking epidemic. WHO Technical Report Series No. 636; Geneva: WHO, 1979. 22.Jeffery TW. Risk behaviors and health. Am J Psychol 1989; 44:1194–202. 23.Lichtenstein E, Glasgow RE. Smoking cessation: what have we learned over the past decade? J Consult Clin Psychol 1992;60:518–27. 24.National Cancer Institute. Strategies to control tobacco use in the United States: a blueprint for public health action in the 1990s. Smoking and Tobacco Control Monographs No. 1. Rockville, MD: US Department of Health and Human Services, National Cancer Institute, 1991. 25.Luepker RV. Community trials. Prev Med 1994;23:602–5. 26.Luepker RV, Murray DM, Jacobs DR et al. Community education for cardiovascular disease prevention: risk factor changes in the Minnesota Heart Health Program. Am J Public Health 1994;84:1383–93. 27.Carleton RA, Lasater TM, Assaf AR et al. The Pawtucket Heart Health program: community changes in cardiovascular risk factors and projected disease risk. Am J Public Health 1995;85: 777–85. 28.Farquhar JW, Fortmann SP, Flora JA et al. Effects of community wide education on cardiovascular disease risk factors. JAMA 1990;264:359–65. 29.Winkleby MA, Feldman HA, Murray DM. Joint analysis of three US community intervention trials for reduction of cardiovascular disease risk. J Clin Epidemiol 1997;50:645–58. 30.Green LW, Richard L. The need to combine health education and health promotion: the case of cardiovascular disease prevention. Promotion Educ 1993;Dec:11–17. 31.Fortmann SP, Taylor CB, Flora JA, Jatulis DE. Changes in adult cigarette smoking prevalence after five years of community health education: the Stanford Five-City Project. Am J Epidemiol 1993;137:82–96. 32.Winkelby MA, Taylor CB, Jatulis D, Fortmann SP. The longterm effects of a cardiovascular disease prevention trial: the Stanford Five-City Project. Am J Public Health 1996;86: 1773–9. 33.Lando HA, Pechacek TF, Pirie PL et al. Changes in adult cigarette smoking in the Minnesota Heart Health Program. Am J Public Health 1995;85:201–8 34.Hoffmeister H, Mensink GBM, Stolzenberg H et al. Reduction of coronary heart disease risk factors in the German Cardiovascular Prevention Study. Prev Med 1996;25:135–45.

35.Gail MH, Byar DP, Pechacek TF, Corle DK, for the COMMIT Study Group. Aspects of statistical design for the community intervention trial for smoking cessation (COMMIT). Control Clin Trials 1992;13:6–21 and erratum, Control Clin Trials 1993;14:253–4. 36.The COMMIT Research Group. Community intervention trial for smoking cessation: I. Cohort results from a 4-year community intervention. Am J Public Health 1995;85:183–92. 37.The COMMIT Research Group. Community intervention trial for smoking cessation: II. Changes in adult cigarette smoking prevalence. Am J Public Health 1995;85:193–200. 38.Winkleby MA. The future of community-based cardiovascular disease intervention studies (Editorial). Am J Public Health 1994;84:1369–72. 39.Fisher EB: The results of the COMMIT Trial. Am J Public Health 1995;85:159–60. 40.Susser M. Editorial: The tribulation of trials – intervention in communities. Am J Public Health 1995;85:156–8. 41.US Department of Health and Human Service. Reducing Tobacco Use (A Report of the Surgeon General). DHHS Publication (CDC), August 2000. 42.Centers for Disease Control and Prevention. Best practices for comprehensive tobacco control programs – August 1999. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, 1999. 43.Centers for Disease Control and Prevention. Cigarette smoking before and after an excise tax increase and an antismoking campaign. MMWR 1996;45:966–70. 44.Briton NJ, Clark TW, Baker AK, Posner J, Soldz S, Krakow M. Adolescent tobacco use in Massachusetts. Trends among public school students 1984–1996. Boston: Health and Addictions Research, 1997. 45.US Department of Health and Human Services, Public Health Service. Treating tobacco use and dependence. Clinical practice guideline. Rockville, MD: US Department of Health and Human Services, Public Health Service, 2000. 46.Green LW, Johnson JL. Dissemination and utilization of health promotion and disease prevention knowledge: theory, research, and experience. Can J Public Health 1996;87:S11–17. 47.Centers for Disease Control and Prevention. Declines in lung cancer rates – California, 1988–1997. MMWR 2000;47: 1066–9. 48.Fiore MC, Novotny TE, Pierce JP et al. Methods used to quit smoking in the US: do cessation programs help? JAMA 1990;263:2760–5. 49.Consensus Statement. The Agency for Health Care Policy and Research Smoking Cessation Clinical Practice Guideline. JAMA 1996;275;1270–80.

113

11

Tobacco and cardiovascular disease: achieving smoking cessation Godfrey H Fowler

Worldwide, there are about one billion current smokers and about three million die annually from their smoking, half before the age of 70; this includes about 150 000 annually in the UK and half a million in the USA.1 Even in countries where the health hazards of smoking are widely acknowledged, it remains a common behavior: in the USA and Canada, for example, about a quarter of all adults smoke, and, in the UK, the situation is worse with about one third of adults smoking. Cardiovascular disease, in particular ischemic heart disease, is the commonest smoking-related cause of death in developed countries.2 This is because, although the relative risk of death from cardiovascular disease in smokers, compared with non-smokers, is much lower than the relative risk from cancer (in particular lung cancer) and chronic obstructive lung disease, ischemic heart disease is much the commonest cause of death in these countries. Overall, the relative risk of death from cardiovascular disease in smokers compared with non-smokers is roughly doubled, though this varies with the different cardiovascular diseases and is greater at a younger age. Passive smoking also increases the risk of cardiovascular disease but the extent of this increase remains uncertain.3,4

Strategies for tobacco control Strategies for reducing the health consequences of smoking should aim to: ● ● ● ●

reduce the uptake of smoking by young people; increase the numbers of smokers stopping smoking; encourage a shift to less harmful tobacco use; decrease exposure to environmental tobacco smoke.

Reducing the uptake of smoking by young people is a priority in many countries. Laws to ban tobacco sales to those below a certain age and to prohibit tobacco advertising and promotion are common in developed countries but are frequently contravened. Other measures include restrictions on smoking in public places, fiscal policies to increase the cost of smoking, and a variety of educational programs. In spite of these, smoking prevalence in teenagers has remained remarkably resistant to change over the last decade, and in 114

the UK about a quarter of young people are regular smokers by the age of 16 years. Modification of cigarettes, particularly with regard to tar yield, over the last two or three decades has undoubtedly contributed to less harmful tobacco use, but it should be emphasized that this is no substitute for tobacco avoidance. However, although such changes have certainly contributed to a decline in lung cancer, possible benefits from these changes relating to cardiovascular disease have not yet been established with certainty.5 Decreased exposure to environmental tobacco smoke is a desirable objective in itself but, again, the contribution this might make to reducing cardiovascular disease risk is very difficult to estimate. For established smokers, smoking cessation is the most important step for safeguarding future health, and this chapter will consider evidence-based methods of achieving this objective.

Evidence of benefits from smoking cessation Many observational epidemiologic studies have investigated the effect of stopping smoking on smoking-related diseases, and there is a wealth of evidence that, not only is tobacco smoking a major risk factor for cardiovascular disease, but also stopping smoking reduces this risk. Grade B However, there is less agreement about the rate at which the risk attenuates after smoking cessation. In the 20 year follow up of the British Doctors Study, for example, excess risk was halved within 2 or 3 years of smoking cessation, and by 10 years the risk had returned to that of a nonsmoker (Figure 11.1).6 However, follow up of the cohort men in the British Regional Heart Study indicates that attenuation of risk is much slower, and even men who had given up smoking for more than 10 years still had an increased risk, compared with non-smokers (Figure 11.2).7 Following myocardial infarction (MI), smoking cessation confers substantial benefits and is particularly important. In one observational study, stopping smoking halved both the number of non-fatal recurrences and the number of cardiovascular deaths (Figure 11.3).8

Tobacco and cardiovascular disease: achieving smoking cessation

Cigarettes smoked daily 20+ 1–19 Non-smoker

2·0 1·7 1·5 1·2 1·0

Current smokers

1

1–4 5–9 10–19 20+ Years as ex-smokers

Nonsmokers

Figure 11.1 Diminished risk of death from coronary heart disease in former light and heavy smokers. Both light and heavy smokers show a steady decline in risk after stopping until, after 10–20 years, it is little different from the risk of nonsmokers. (Source: Royal College of Physicians. Smoking or health? London: Pitman Medical, 1977.)

4

Relative odds

3

100 Non-smokers Cumulative mortality (%)

Mortality ratios

2·5

Stopped smoking 80

Continued smoking

60

40

20

2

4

6 8 10 Years of follow up

12

14

Figure 11.3 Cumulative mortality for 498 survivors of a coronary attack by smoking habit. Life-table curves start 2 years after attack. Average annual mortality was 6.5% in non-smokers, 3.7% in those who stopped smoking, and 10.2% in those who continued smoking. (Source: Daly et al.8)

processes; free radical damage to vascular endothelium has been demonstrated, as have effects on platelet survival, platelet aggregation, and fibrinogen levels.11–13

2

The nature of tobacco smoking 1

0·5 Never smoked cigarettes

>20

11–20

6–10

0–5 Current cigarette smoker

Years since giving up cigarettes

CHD events

30

22

30

20

42

189

Number of men

1819

603

719

577

762

3185

Figure 11.2 Relative odds of a major CHD event in relation to years since stopping smoking cigarettes. (Source: Cook et al.7)

In another study, follow up over 13 years of post-MI patients showed a 37% mortality in those who had stopped smoking, compared with 82% mortality in those who continued smoking.9 Furthermore, a UK trial of smoking cessation advice in smokers with evidence of ischemic heart disease showed a (non-significant) 13% difference in cumulative CHD deaths over 20 years in those given smoking cessation advice, compared with those who were not.10 The mechanisms through which tobacco smoking mediates its adverse cardiovascular effects are largely unknown and certainly multiple. There is evidence that smoking contributes to both the atherosclerotic and the thrombotic

Before individual smoking interventions and cessation methods are considered, it is helpful to review briefly the nature of tobacco smoking and the consequent implications for interventions. Tobacco smoking is a complex behavior to which psychologic, social, and pharmacologic factors contribute.14 Its acquisition is almost invariably in adolescence, as the result of desire for experimental rebellious behavior, which is perceived as adult and encouraged by peer group pressure. However, pharmacologic addiction usually then becomes a factor determining persistence of the behavior, making it difficult to stop because of the addictive effects of nicotine and the discomforts associated with withdrawal. Although the balance between psychologic factors and pharmacologic addiction varies from smoker to smoker, there is now increasing awareness of the importance of nicotine addiction in maintaining smoking behavior, and the powerful nature of this addiction has been compared with addiction to heroin or cocaine.15 The evidence basis for smoking cessation Emphasis on smoking cessation in individual established smokers is a vital component of any tobacco control strategy and should complement efforts to prevent the uptake of smoking by young people. Individual approaches to both cessation and avoidance can only be supplementary 115


to “whole population” approaches, including legislation (banning sales to “minors” and controlling advertising promotion), restrictions on smoking, public information and campaigns, tax measures, and so on. Grade A For many established smokers, stopping smoking is very difficult, for both behavioral and psychopharmacologic reasons, and only a minority of established smokers ever succeed in stopping for good. Most of those who succeed in stopping find the process of stopping is a dynamic one rather than a single discrete event (Figure 11.4). Thinking about stopping

Deciding to try

Trying to stop Stopping

“Contented” Relapsing smokers Staying stopped

Figure 11.4 Stopping smoking is a process

Relapses are common and eventual success is usually the outcome of many attempts. Motivation to stop and confidence in the ability to succeed are important predictors of success. Relapse in the first few weeks is a common pattern, but the tendency to relapse declines as time progresses and most of those who manage to avoid relapse for a year are then able to achieve sustained abstinence. The great majority of those smokers who do achieve longterm smoking cessation do so on their own without any special aids or assistance, but research has shown that smoking cessation advice and support from a health professional and the use of smoking cessation aids can enhance the chances of success.17 As already indicated, motivation to stop smoking is critical to success and this motivation may be determined by a variety of concerns – personal health, family health, financial anxieties, social pressures, and so on. The so-called Stages of Change model acknowledges different levels of motivation and activity, from precontemplation through contemplation, preparation and action, to maintenance or relapse.18 Identification of the stage at which an individual smoker is can enable motivational or interventional methods to be targeted more appropriately (though the evidence basis for the effectiveness of such targeting is currently lacking). A favorable factor is that the majority of smokers in many countries report that they want to quit smoking and have tried to do so, often many times. They also cite advice from a health professional as being important to them in influencing their motivation to quit. Surprisingly perhaps, only a minority of smokers say they have ever been asked by a doctor about smoking and advised to stop.14 116

Community interventions Community- or population-based smoking cessation interventions have been implemented in a number of settings. Typically, they involve use of mass media to promote public awareness and education, to encourage health professionals to raise smoking as an issue in consultations with patients, and to offer self-help materials. Evaluation of the effectiveness of such programs is difficult and they are discussed in Chapter 10.

Individual advice Nicotine addiction is now acknowledged as a treatable condition16 and there is substantial scientific evidence of the effectiveness of behavioral and pharmacologic interventions. Individual smoking interventions by health professionals have been extensively studied.17 An early and influential trial in the UK was conducted by Russell and colleagues in general practices in London in the 1970s. In this study, over 2000 smokers attending their general practitioners for routine consultations were randomly allocated to a nonintervention control group and three intervention groups: ● ● ●

Grade B completion of a brief smoking questionnaire; brief (1–2 minutes) smoking cessation advice; and brief advice supplemented with a simple self-help smoking cessation leaflet.

Smoking cessation rates achieved at 1 month and sustained for 1 year were 1·6%, 3·3%, and 5·1% respectively in the three intervention groups compared with 0·3% in the control group (Figure 11·5).19 Many similar randomized controlled trials of simple, brief smoking cessation advice in medical settings have subsequently been conducted and this finding of a small percentage of biochemically validated long-term smoking cessation, resulting from such interventions, has been replicated.20

5·1%

28 GPs (5 London practices) Approx. 2000 patients Random allocation to four groups

3·3%

1·6% 0·3% Non-intervention control

Questionnaire only

Questionnaire + advice to stop

Questionnaire + advice to stop + leaflet + warning of follow up

% = proportion who stopped smoking during the first month and were still not smoking 1 year later (P < 0.001) Equivalent to 25 long term successes per GP each year

Figure 11.5 Effect of GP’s advice against smoking. (Based on Russell et al.19)


Nicotine replacement therapy (NRT) The advent of nicotine chewing gum in the 1970s provided the first specific pharmacologic “treatment” for smoking cessation. Grade B Subsequent development of transdermal nicotine patches, nicotine nasal sprays, and nicotine oral inhalers has increased the range of products available. The objective in using these “nicotine replacement” products is to provide a temporary alternative source of nicotine to allay withdrawal symptoms and so enhance the potential for smoking cessation. Large placebo-controlled trials have clearly demonstrated that the use of such products as an adjunct to advice from a health professional can approximately double smoking cessation rates, compared with placebo.21,22 Several systematic reviews of the many trials that have now been conducted with them have confirmed the benefits and shown that all preparations are effective, but the evidence is particularly substantial for nicotine gum and nicotine patches.23,24 Both appear to have similar effectiveness but, because of greater social acceptability, ease of use, and simpler compliance, transdermal patches have been found by many to be preferable, though gum appears more effective for the most dependent smokers. A summary of the effectiveness of NRT and estimates of the number needed to treat (NNT) to achieve one success are provided in Table 11.1. Table 11.1 Nicotine replacement therapy preparations and abstinence NRT preparation % Quitting OR (95% CI) (n of trials) Active Control

NNT

Gum (39) Patches (9) Nasal spray (1) Inhaler (1)

13 10 6 10

18·2 20·5 25·9 15·2

10·6 10·8 9·9 5·0

1·6 (1·5–1·8) 2·1 (1·6–2·6) 2·9 (1·5–5·7) 3·0 (1·4–6·6)

Abbreviations: NRT, nicotine replacement therapy; OR, odds ratio

Concern has been expressed about the use of NRT in patients with cardiovascular disease because of the potential adverse effects of nicotine on the cardiovascular system. In considering this issue, it is important to be aware that use of NRT is advocated only as a temporary substitute source of nicotine in those already self-administering this drug through tobacco smoking. It is also important to bear in mind that blood levels of nicotine achieved with NRT are substantially lower than those achieved by moderate or heavy smoking. Furthermore, there is no evidence that nicotine itself contributes to the atherogenic or thrombotic processes, unlike tobacco smoking. In a placebo-controlled randomized trial specifically investigating the safety of transdermal nicotine patches in

patients with cardiac disease, no increase was found in rates of arrhythmia, MI, or death in high-risk patients (with a history of MI or coronary revascularization procedure, or of angina, heart failure, arrhythmia, peripheral vascular disease, or cerebral vascular disease) in those using nicotine patches compared with those using placebo patches.25 A review of the potential adverse effects of NRT in patients with cardiovascular disease concluded that there is no evidence to justify the withholding of these products in such patients who smoke and are motivated to stop.27

Bupropion (Zyban) Originally developed as an antidepressant, bupropion (which is an inhibitor of neuronal uptake of norepinephrine, dopamine and serotonin) was found to apparently aid smoking cessation. Subsequent evidence from placebo-controlled trials26 has confirmed this and indicate that it is at least as effective as NRT. Although side effects are rare, there is a risk of seizures in about 1 in 1000 users. It should not therefore be used by patients who have a past history of seizures and who are on drugs known to lower the threshold for seizures; this includes antipsychotics, antidepressants, antimalarials, theophylline, quinalones, and sedating antihistamines. There is no evidence of other adverse effects in patients with cardiovascular disease and it may therefore be used by such patients. Whereas NRT should be started only when stopping smoking is actually attempted, bupropion treatment needs to be initiated 2 weeks before the attempt is made. The limited evidence currently available suggests that the combination of NRT and bupropion may be more effective than either alone, but more research is needed to elucidate this.

Review of cessation studies Grade A In a comprehensive systematic review of 188 randomized controlled trials of the efficacy of a wide range of interventions aimed at helping people to stop smoking, it was concluded that simple advice, even on one occasion only, given by a doctor in general or family practice or in a hospital clinic to all smokers who consulted, resulted in sustained cessation of about 2% and that additional encouragement and support (additional visits, exhaled CO measurement, letters, etc.) further enhances this effect.28 Whether similar interventions delivered by nurses are equally effective remains uncertain,29 although there is evidence that nurse support, subsequent to doctor advice, can enhance the effect of this.30 This comprehensive review25 also endorsed the use of nicotine replacement therapy but concluded that a variety of other smoking cessation interventions – hypnosis, acupuncture, aversion therapy, and pharmacologic agents other than nicotine, which are sometimes advocated – have 117


Table 11.2 Summary estimates of randomized controlled trials of interventions to help people to stop smoking Intervention

% Estimate of efficacy (95% CI)

Statistical significance

Subjects (trials) (n)

Comment

Simple physician advice (once)

2 (1–3)

0·001

14 438 (17)

Effective

Physician advice with additional encouragement/support

5 (1–8)

0·01

6466 (10)

Effective

Nurse advice

1 (1–3)

0·10

3369 (2)

Unproven

Advice in infarct survivors

36 (23–48)

0·001

223 (1)

Important

Advice in healthy men at high CHD risk

21 (10–31)

0·001

13 205 (4)

Important

Hypnosis

24 (10–38)

0·001

3 (1–6)

0·10

Acupuncture

646 (10)

2759 (8)

No trials with biochemical validation No trials with biochemical validation

Adapted from Law et al 28

not been shown by rigorous scientific evidence to be effective, although it must be acknowledged that the methodologic problems associated with attempts to evaluate these have yet to be overcome. A summary of estimates of effectiveness is provided in Table 11.2.

●

●

●

Specialist smoking cessation clinics Specialist smoking cessation clinics have been shown to deliver effective interventions and can make a useful contribution to the provision of individual interventions, usually by providing regular group treatments. There is some evidence that they can achieve enhanced attendance and abstinence rates as high as 20% or more, but interpretation of their success should take account of the fact that they recruit widely and participants are generally highly motivated to stop, compared with the majority of those expressing an intention to do so. When available, they offer a self-referral and secondary referral service and can provide valuable opportunities for smoking cessation research.31 However, as they are relatively few in number in relation to the huge need for such interventions, their overall contribution will inevitably be small. Practical aspects of smoking cessation in clinical practice The essential features of individual smoking cessation interventions in medical practice are to: ●

118

assess in any medical consultation the smoking status of the patient, whether a non-smoker, smoker or ex-smoker;

advise all smokers about the desirability and importance of stopping smoking because of health hazards, especially those who already have smoking-related diseases; assist smokers to stop smoking, particularly those with smoking-related diseases and especially if expressing interest to do so; follow up at subsequent consultations to assess the outcome and, if necessary, further assist those trying to stop smoking while encouraging ex-smokers to maintain their non-smoking status.32

Assessment The smoking status of all patients should be recorded in medical records in such a way that the information is easily accessible in future consultations. Assessment of smoking should include a brief history of the patient’s smoking, including attempts to stop and their current tobacco consumption. Assessment of nicotine addiction should also be made by inquiring how soon after waking they smoke their first cigarette and some assessment of their motivation to stop smoking. Advice Grade B All patients with smoking-related diseases should be advised to stop smoking and any reasons that patients put forward for wanting to stop smoking should be strongly reinforced. Assistance Specific help with smoking cessation should be strongly influenced by patients’ preferences and patients should


themselves be active in deciding what to do. There is no set “prescription” for how to go about stopping smoking, but it is possible to provide guidance, which experience has shown to be useful. It is important to adopt an individual approach but guidance might include: ● ● ●

●

Grade A setting a target date for stopping; some preparation for stopping, review of motivation and reasons for stopping; awareness of times when a particular need is felt for a cigarette, and attempts to change routines to avoid association of these times with smoking; eliciting support for cessation from friends and colleagues and, ideally, recruiting a fellow smoker (particularly a spouse) to join in the attempt to give up smoking.

Generally, sudden complete withdrawal is likely to be more successful than attempts to gradually reduce smoking. Strategies need to be planned for coping with withdrawal symptoms and other difficulties likely to be encountered immediately after cessation; avoidance of other smokers and smoking environments is likely to be important, particularly at “danger times”, like teabreaks and after meals or when having a drink. A number of self-help leaflets are available from a variety of sources to supplement and reinforce such simple guidance. These leaflets have particular value and effectiveness when handed out by health professionals as an adjunct to brief advice. Ask about smoking and record in notes Non-smoker

Use of NRT should be encouraged in all those (except perhaps the lightest smokers) for whom advice and self-help are not enough. Assessment of nicotine dependence is most simply done by asking how soon after waking the first cigarette is smoked. If this is within half an hour, this is evidence of at least moderate dependence and suggests likely benefit from using NRT. As already indicated, although there is debate about the safety of using NRT in patients with cardiovascular disease, evidence of harm from doing this is lacking but there is some evidence that suggests that it is safe. This is likely to be so if nicotine gum or patches are used (as they should be) as a temporary substitute for smoking. A combination of smoking and nicotine replacement may well be potentially harmful and should be strongly discouraged. A simple smoking cessation protocol is illustrated in Figure 11.6. Key points ●

●

●

●

●

Smoker ●

Wants to stop

Encourage to remain non-smoker

Contented smoker

Assess motivation Give information to motivate Ready to stop Not ready to stop Give advice, leaflets, offer support & follow up Stops smoking

Fails to stop

Assess timing of first cigarette and nicotine dependence & suggest gum, patches, inhaler or nasal spray if appropriate Stops smoking

Fails to stop Try again later

Figure 11.6 Smoking cessation protocol for doctor/nurse intervention

●

Tobacco smoking is a critically important, modifiable cardiovascular risk factor (especially in those with established cardiovascular disease). Smoking cessation attenuates cardiovascular risk and early benefits accrue (again, especially in those with established cardiovascular disease). There is good evidence for the effectiveness of simple, brief smoking cessation advice and the use of NRT as an adjunct to this. NRT is effective and safe. Smoking and smoking cessation should be routinely addressed by health professionals in any consultations with patients who smoke. Simple cessation advice and support should be routinely offered by healthcare professionals in any consultations with patients who smoke. NRT – chewing gum, transdermal patches, nasal spray, or oral inhaler – should be recommended to all smokers trying to quit. Encouragement and support should accompany this, and compliance for 2 or 3 months should be encouraged in those who achieve short-term abstinence with it. Bupropion (Zyban) is a proven alternative to NRT.

References 1.Peto R, Lopez AD, Boreham J, Thun M, Heath C. Mortality from smoking in developed countries 1950–2000. Oxford: Oxford University Press, 1994. 2.Doll R, Peto R, Wheatley K, Gray R, Sutherland I. Mortality in relation to smoking: 40 years’ observation on male British doctors. BMJ 1994;309:901–11. 3.Fourth Report of Independent Scientific Committee on Smoking and Health. London: HMSO, 1988. 4.Glanz SA, Parmley WW. Passive smoking and heart disease: mechanisms and risk. JAMA 1995;273:1047–53. 5.Darby SC, Doll R, Stratton IM. Trends in mortality from smoking-related diseases in England and Wales. In: Wald N,

119


Frogatt P, eds. Nicotine, smoking and the low tar programme. Oxford: Oxford University Press, 1989. 6.Doll R, Peto R. Mortality in relation to smoking: 20 years’ observation of British male doctors. BMJ 1976;4:1525–36. 7.Cook DG, Pocock SJ, Shaper AG et al. Giving up smoking and the risk of heart attacks. Lancet 1986;2:1376–80. 8.Daly LE, Mulcahy R, Graham IM, Hickey M. Long term effect on mortality of stopping smoking after unstable angina and myocardial infarction. BMJ 1983;287:324–6. 9.Wilhelmssen C, Vedin J, Elmfeld D et al. Smoking and myocardial infarction. Lancet 1975;1:415–17. 10.Rose G, Colwell L. Randomised controlled trial of antismoking advice. J Epidemiol Community Health 1992;46:75–7. 11.Pittilo RM, Woolf N. Cigarette smoking, endothelial injury and atherosclerosis. J Smoking-related Dis 1993;4:17–25. 12.Hawkins RI. Smoking, platelets and thrombosis. Nature 1972;263:450–2. 13.Meade TW, Imeson J, Sterling Y. Effect of changes in smoking on clotting factors and on risk of ischaemic heart disease. Lancet 1987;ii:986–8. 14.Marsh A, Matheson J. Smoking attitudes and behaviour. London: HMSO, 1993. 15.US Department of Health and Human Services. The health consequences of smoking and nicotine addiction. Report of Surgeon General 1988. Washington DC: DHHS, 1989. 16.RCPL. Nicotine addiction in Britain. London: Royal College of Physicians of London, 2000. 17.Kottke T, Battista R, DeFriese G, Brekke M. Attributes of successful smoking cessation interventions in medical practice: a meta-analysis of 39 controlled trials. JAMA 1988;259: 2883–9. 18.Prochaska JO, DiClemete C. Towards a comprehensive model of change. In: Miller WR, Heather N, eds. Treating addictive behaviours: processes of change. New York: Plenum, 1986. 19.Russell MAH, Wilson C, Taylor C, Bales CD. Effect of general practitioner’s advice against smoking. BMJ 1979;ii:231–5.

120

20.Jamrozik K, Vessey M, Fowler G et al. Controlled trial of three different antismoking interventions in general practice. BMJ 1984;288:1499–503. 21.Russell MAH, Merrium L, Stapleton J, Taylor W. Effect of nicotine chewing gum as an adjunct to general practitioners’ advice against smoking. BMJ 1983;287:1782–5. 22.Imperial Cancer Research Fund General Practice Research Group. Randomised trial of nicotine patches in general practice: results at one year. BMJ 1994;308:1476–7. 23.Silagy C, Mant D, Fowler G, Lodge M. Meta-analysis of the efficacy of nicotine replacement in smoking cessation. Lancet 1994;343:139–42. 24.Tang TL, Law M, Wald N. How effective is nicotine replacement in helping people to stop smoking? BMJ 1994;308:21–6. 25.Joseph AM, Norman SM, Ferry LH et al. The safety of transdermal nicotine as an aid to smoking cessation in patients with cardiac disease. N Engl J Med 1996;335:1792–8. 26.Jorenby DE, Leischow SJ, Nides MA et al. A controlled trial of sustained-release bupropion, a nicotine patch, as both for smoking cessation. N Engl J Med 199;341:685–91. 27.Benovitz NL, Goursley SG. Cardiovascular toxicity of nicotine: implications for nicotine replacement therapy. J Am Coll Cardiol 1997;29:422–31. 28.Law M, Tang TL, Wald N. An analysis of the effectiveness of interventions intended to help people stop smoking. Arch Intern Med 1995;155:1933–41. 29.Sanders D, Fowler G, Mant D et al. Randomised controlled trial of anti-smoking advice by nurses in general practice. J Roy Coll Gen Pract 1989;39:273–6. 30.Hollis J, Lichenstein E, Vogt T et al. Nurse-assisted counselling for smokers in primary care. Ann Intern Med 1993;118:521–5. 31.Sutherland G, Stapleton J, Russell MAH et al. Randomised controlled trial of nasal nicotine spray in smoking cessation. Lancet 1992;340:324–9. 32.US Department of Health and Human Services. 1996 Clinical practice guidelines no. 18: Smoking cessation. Washington DC: DHHS, 1996.

12

Lipids and cardiovascular disease Malcolm Law

For many years the issue of lipids and cardiovascular disease was seen as controversial and difficult to resolve, yet the high-fat diet typical of many western countries over the greater part of the 20th century has proved to be the major underlying factor in the epidemic of ischemic heart disease, and modern cholesterol lowering drugs can reduce risk more than any other single intervention.1

Serum total and low density lipoprotein cholesterol Typical values of serum total and low density lipoprotein (LDL) cholesterol in western countries are high in comparison to those in agricultural and hunter–gatherer communities, because of the high saturated fat content of the western diet. Average serum cholesterol concentration (in men aged 45–60) is about 3–3·5 mmol/l in hunter–gatherer societies and rural China (where heart disease is rare), 5·0 mmol/l in Japan, 5·5 mmol/l in Mediterranean populations and a little higher in the USA, and 6 mmol/l in Britain and several other European countries.2 Average levels of LDL cholesterol are about 2 mmol/l lower.2 Use of the term “normal” in reference to usual or average western cholesterol values may therefore be misleading. Of the average total serum cholesterol in western populations, two thirds is low density lipoprotein (LDL) cholesterol and one quarter is high density lipoprotein (HDL) cholesterol. The atherogenic properties lie in the LDL fraction (sometimes measured as its carrier protein, apolipoprotein B, with which it is highly correlated). Many of the large epidemiological studies and randomized trials measured only total serum cholesterol, and results based on total serum cholesterol have been taken to estimate effects of LDL cholesterol. Fortuitously, the approximation is a good one. The absolute reduction in total serum cholesterol produced by diet and by most drugs (including statins1) is similar to the reduction in LDL cholesterol. Observational differences in total cholesterol between individuals are close to the corresponding differences in LDL cholesterol, because HDL cholesterol is independent of total serum cholesterol.3,4 This arises because the tendency for HDL cholesterol to be positively associated with total cholesterol (as HDL cholesterol is part of total) is offset by the small inverse association between HDL and LDL cholesterol. Much epidemiologic

and clinical trial data are therefore available to estimate quantitatively the effect of lowering serum LDL cholesterol on the risk of ischemic heart disease.

Serum cholesterol and ischemic heart disease Evidence from genetics, animal studies, experimental pathology, epidemiologic studies and clinical trials indicates conclusively that increasing serum cholesterol is an important cause of ischemic heart disease and that lowering serum cholesterol reduces the risk,5,6 and the results of six large randomized trials of statins have ensured that this is now widely accepted.1,7–11 Three important practical questions arise: the nature of the dose–response relationship, the size of the effect, and the speed of the reversal of risk. To answer these questions data from both observational epidemiology (cohort studies) and randomized controlled trials are necessary. The two are complementary; examining trial data alone is misleading. Table 12.1 summarizes the advantages of each. In cohort (or prospective) studies serum cholesterol is measured in a large number of individuals and subsequent heart disease mortality (or incidence of myocardial infarction) is recorded. Cohort studies are easier to conduct than trials (as there is no intervention) and can therefore be much larger. Accordingly, their statistical power is greater and they can examine the association across a wider range of serum cholesterol values and a wider range of ages than trials have done. Most of the cohort studies and trials of cholesterol and ischemic heart disease recruited men, for reasons of economy, as ischemic heart disease is more common in men. The more limited data from women indicate a similar effect as in men.1

The nature of the dose–response relationship: is there a threshold? Figure 12.1 shows mortality from ischemic heart disease plotted according to quintile groups (fifths) of the ranked serum cholesterol measurements in a large cohort study of serum cholesterol and ischemic heart disease (MRFIT Screenees).12 With ischemic heart disease plotted on a logarithmic scale, the relationship is described almost perfectly by a straight line linking the proportional change in 121


Table 12.1 Relative advantages of cohort studies and randomized trials in assessing the relation between serum cholesterol and ischemic heart disease Objective

Advantage (comment)

Statistical power

Cohort studies (recorded about three times more ischemic heart disease events than the trials) Cohort studies (observation across wide range of cholesterol values) Cohort studies (ischemic heart disease events at age 35–85, but mostly 55–65 in trials) Cohort studies (on recruitment the serum cholesterol was the same in intervention and control groups) Randomized trials (on recruitment serum cholesterol was the same in intervention and control groups) Randomized trials (not a major advantage – bias in cohort studies can be allowed for)

Dose–response relationship Wide age range Long-term effects of cholesterol differences Short-term effects of cholesterol differences Avoid bias

Ischemic heart disease mortality per 10 000 man years

30

20

10

7·5 4

5 6 Serum cholesterol (mmol/l)

7

Figure 12.1 Mortality from ischemic heart disease (with 95% confidence intervals) according to serum cholesterol in a large cohort study12

ischemic heart disease to the absolute difference in serum cholesterol (r 0·997). Other cohort studies show the same relationship.6 The 95% confidence limits of the risk estimates in each quintile group do not overlap, establishing that there is no threshold below which a further decrease in serum cholesterol is not associated with a further decrease in risk of ischemic heart disease. The exponential relationship indicated by the straight line means that a given absolute difference in serum cholesterol concentration from any point on the cholesterol distribution is associated with a constant proportional difference in the incidence of ischemic heart disease. 122

This absence of a threshold has been contentious; many published guidelines on lowering cholesterol invoke one, commonly advocating cholesterol lowering drugs only in patients whose serum cholesterol exceeds 5 mmol/l, yet the evidence is firmly against any threshold. The data in Figure 12.1 (which alone are conclusive) are supported by data from other large cohort studies,6 including an important study from China which shows that the continuous relationship extends below serum cholesterol values of 4 mmol/l.13 The results of a subgroup analysis in one statin trial8 showing no reduction in coronary events in persons with the lowest serum cholesterol levels have been misinterpreted, because the confidence interval on the result was consistent both with no reduction in coronary events and with the expected reduction from the continuous association shown in cohort studies. Large randomized trials have now confirmed the result from cohort studies that the constant proportional reduction in risk extends below 5 mmol/l.9,11,14 Experimental data on the transfer of cholesterol from the blood into atheromatous lesions exclude a threshold as low as 1 mmol/l.15 Patients at high risk of an ischemic heart disease event (especially those with existing disease) should be offered a statin irrespective of their existing level of total or LDL cholesterol. The size of the effect Cohort studies provide the best estimates because they cover a wide age range and have high statistical power, and because the serum cholesterol differences between individuals recorded on entry to a cohort study will have been present on average for decades beforehand (so cohort studies show long-term associations). Trials, on the other hand, show the effect of short-term differences. Cohort studies are subject to bias, but this can be corrected. The major bias is the so-called “regression dilution bias”.3

Lipids and cardiovascular disease

Age (years)

Estimated percentage decrease in risk for a serum cholesterol reduction (mmol/l) of 0·3 (5%)

0·6 (10%)

1·2 (20%)

1·8 (30%)

32 22 15 11 10

54 39 27 20 19

79 63 47 36 34

90 77 61 49 47

0% Reduction in risk of IHD

Table 12.2 Estimates (from 10 cohort studies) of the percentage decrease in risk of ischemic heart disease according to extent of serum cholesterol reduction and age6

10%

20%

30% Randomized trials 40%

40 50 60 70 80

Table 12.2 shows estimates of the long-term percentage decrease in the risk of an ischemic heart disease event according to the decrease in serum cholesterol concentration and age at event. The estimates are taken from an analysis of the 10 largest cohort studies, corrected for the regression dilution bias and for the minor distinction between differences in total and in LDL cholesterol discussed above.6 A reduction in total or LDL cholesterol of 0·6 mmol/l (about 10%) is associated with a decrease in risk of ischemic heart disease of about 50% at age 40, 40% at 50, 30% at 60, and 20% at 70–80. The proportional decrease in risk decreases with age, but the absolute benefit increases because the disease becomes more common with age. The increasing reduction in risk with greater reduction in serum cholesterol shown in Table 12.2 follows from the exponential dose–response relationship described above. For a 0·6 mmol/l cholesterol reduction at age 60, for example, the reduction in risk is 27% and the relative risk is therefore 0·73; with a serum cholesterol reduction three times as great (1·8 mmol/l), the relative risk is 0·733 (0·73 0·73 0·73) or 0·39, and the reduction in risk is 61%. Speed of reversal and consistency of observational and trial data Data have been analyzed from the “old generation” of 28 randomized trials in which the average serum cholesterol reduction was about 0·6 mmol/l (10%).6 Figure 12.2 shows the reduction in incidence of ischemic heart disease in all trials combined according to time since entry. In the first 2 years there was little reduction in risk. From 2 to 5 years the average reduction in risk was 22%, and after 5 years the reduction was 25%. The ischemic heart disease events in these trials mostly occurred at an average age of about 60, and at this age the estimate of the long-term effect from the cohort studies is 27% (Table 12.2). The similarity of the estimates of effect from the cohort studies and from the trial

0

Cohort studies

2 4 6 8 10 Long term Time after cholesterol reduction (years)

Figure 12.2 Reduction in the incidence of ischemic heart disease (IHD) per 0.6 mmol/l (about 10%) decrease in serum cholesterol, as estimated from randomized trials according to time since entry and from cohort studies (which reflect the long term association)6

data from the third year onwards therefore indicates that the reversal of risk is near maximal after 2 years – a surprisingly rapid effect. The trial data show that the proportional reduction in risk from lowering serum cholesterol is similar in persons with and without previous myocardial infarction or other clinical evidence of coronary artery disease.6 The six large trials of statins1,7–11 have achieved significantly larger reductions in total and LDL cholesterol. These trials too showed a relatively small reduction in ischemic heart disease events in the first 2 years, but a reduction after 2 years that is close to the maximum indicated by cohort studies. Most of these trials achieved an average reduction in total and LDL cholesterol of about 1 mmol/l in treated relative to placebo patients. The reduction in ischemic heart disease events was about 40% from the third year onwards, similar to the long-term estimate corresponding to this cholesterol reduction at age 60 (the average age at the time of the events) from the cohort studies. With relatively little reduction in risk in the first 2 years, the average reduction over the entire duration of the trials (5 years or so) was about a third. Importantly, however, the randomized trials of statins do not show their full potential for preventing ischemic heart disease events because of “contamination” – some patients allocated to the treated group leave the trial and stop taking the tablets, whereas some patients allocated to the placebo group take statins. Atorvastatin and simvastatin can reduce serum total and LDL cholesterol by about 1·8 mmol/l, but no trial has maintained this difference between all patients allocated to the treated and placebo groups over the 5 year duration of a trial. Two trials have maintained a difference of about 1·6 mmol/l – one statin trial1 and a trial in which serum cholesterol was reduced by ileal bypass surgery.6 A long-term reduction in ischemic heart disease events of about 55% at age 60 would be 123


expected from the cohort study data in Table 12.2, and this was approximately the observed reduction after 2 years in these two trials. With a serum cholesterol of 1·8 mmol/l a reduction in heart disease events of about 60% would be expected in the longer term, as Table 12.2 shows. The randomized trials therefore confirm the dose–response relationship shown in the cohort studies (the greater the cholesterol reduction the greater the reduction in heart disease events) and confirm the estimates from the cohort studies in Table 12.2 of the reduction in risk. We can therefore be confident that using atorvastatin or simvastatin in doses of around 20 mg/day to reduce cholesterol by 1·8 mmol/l will reduce risk by about 60% in the longer term.

Dietary fat and serum cholesterol The relationship between dietary saturated fat and serum cholesterol is shown by the data from Japan and Britain in Table 12.3. This comparison is a useful one because dietary saturated fat differs greatly, yet dietary polyunsaturated fat and cholesterol are similar in the two countries. As in other situations (salt and blood pressure, for example) the size of the association varies with age, yet there has been a tendency to generalize to older age groups the results of studies conducted in younger age groups. Many dietary trials, for example, have been conducted in people under 30. The few that have been conducted in people over 50 tend to support the above Japan–Britain comparison.6 In older people a reduction in dietary saturated fat equivalent to 10% of calories will lower serum cholesterol by about 1 mmol/l, which in turn will reduce ischemic heart disease mortality in the long term by about 40%. The chain lengths of saturated fatty acids influence the extent to which they increase blood cholesterol. Palmitic (C16:0) and myristic (C14:0) acids have the major effect, lauric acid (C12:0) some effect, and stearic acid (C18:0) and medium chain fatty acids have little or no effect.

Table 12.3 Serum cholesterol and dietary saturated fat in Japan and Britain. Data compiled from national surveys in each country2 Age

Britain

Difference

Dietary saturated fat (% calories) All ages 6%

16%

10%

Serum cholesterol (mmol/l) 20–9 30–9 40–9 50–9 60–9

5·0 5·6 6·0 6·2 6·2

0·5 0·6 0·9 1·0 1·2

124

Japan

Trans unsaturated fatty acids are also important: randomized trials show that they increase serum total and LDL cholesterol by about as much as these longer chain saturated fatty acids.16,17 They are scanty in naturally occurring fats but are generated by the hydrogenation of vegetable oils for use as hardening agents in manufactured foods. They constitute 6–8% of dietary fat, or 2% of calories, in western diets. Naturally occurring cis unsaturated fatty acids reduce serum cholesterol by approximately half as much as longer chain saturated fatty acids increase it. Reduction in dietary cholesterol has a small effect on blood cholesterol concentration.5 Substitution of cis unsaturated for saturated fats in the western diet is thus the most appropriate change in lowering the high levels of blood cholesterol in western populations. The reduction in serum total or LDL cholesterol that can easily be attained by individuals trying to alter their diet in isolation from family, friends and workmates is relatively small (about 0·3 mmol/l, or 5%). A larger serum cholesterol reduction, about 0·6 mmol/l (10%), is realistic on a community basis, as the availability of palatable low-fat food increases when other family members or the community alter their diet, and the dietary change is perceived more positively. A reduction by about 7% of calories, a realistic target for a high-fat population, would lower serum cholesterol by 0·6 mmol/l, which in turn would reduce the mortality from ischemic heart disease at age 60 by 25–30%. Reductions in serum cholesterol of about 0·6 mmol/l through dietary change have occurred in entire western communities, in the United States and Finland for example.2 Measures that facilitate such a change include wider public education, labeling of foods sold in supermarkets, and the provision of information on the fat content of restaurant meals. Most important is the implementation of national and international policies on food subsidies that are linked to health priorities.

Serum cholesterol and circulatory diseases other than ischemic heart disease Table 12.4 shows the death rates from all circulatory diseases according to total serum cholesterol concentration, observed in the same large cohort study (MRFIT Screenees18) as shown in Figure 12.1. Apart from ischemic heart disease, serum cholesterol is associated with stroke and with other circulatory diseases.

Stroke 4·5 5·0 5·1 5·2 5·0

The data from the large cohort study of the MRFIT Screenees (Table 12.4) are useful because thrombotic and hemorrhagic stroke were distinguished. For deaths from thrombotic stroke the data are consistent with a continuous dose–response relationship with serum cholesterol,


Table 12.4 Death rates per 100 000 man years (number of deaths) from circulatory diseases according to serum cholesterol in a large cohort of men18 Cause of death (IDC-9 code)

Serum cholesterol (mmol/l) (% of all men) 4·1 (6%)

Ischemic heart disease (410–4) Stroke thrombotic (433–8) intracranial hemorrhage (431–2) Other circulatory diseases All circulatory diseases (390–459)

6·2 (24%)

4·1–5·1 (31%)

5·2–6·1 (39%)

65 (160)

98 (1239)

169 (2731)

289 (2804)

0·001

6 (14) 9 (22) 31 (77) 110 (273)

6 (73) 4 (55) 39 (483) 147 (1850)

8 (135) 5 (86) 41 (670) 224 (3622)

13 (126) 6 (57) 57 (556) 365 (3543)

0·001 — 0·001 0·001

analogous to that shown for ischemic heart disease in Figure 12.1. For hemorrhagic stroke (intracranial and subarachnoid), however, there is an excess risk at lower serum cholesterol levels.18,19 Cohort studies that have distinguished thrombotic and hemorrhagic strokes are fairly consistent in showing a positive association of serum cholesterol with thrombotic stroke mortality but an inverse association with hemorrhagic stroke mortality, and cohort studies that do not distinguish the two types of stroke tend to show little or no association between cholesterol and stroke mortality, consistent with the two associations cancelling each other out.20 Whether the inverse association is cause and effect is uncertain. It is more difficult to see how a spurious (non-causal) association with intracranial hemorrhage might arise through the disease (or predisposition to the disease) lowering serum cholesterol than is the case with depression and suicide or cancer. It is also difficult to see any mechanism by which the inverse association might be cause and effect, although experimental data lend some support to an interpretation of a causal effect of low cholesterol in that the endothelium of intracerebral arteries might be weaker at very low serum cholesterol levels.19 The randomized trials of serum cholesterol reduction, especially the statin trials, have shown a lower incidence of stroke (all types combined) in treated than control patients:9–11,21 statins reduced stroke by about 26%.21 However, these were nearly all non-fatal strokes. Thrombotic stroke has a lower case fatality than hemorrhagic stroke, so the majority of nonfatal strokes (recorded in the trials) will be thrombotic but little more than half of fatal strokes (recorded in the cohort studies) will be thrombotic. Also, the randomized trials tended to recruit patients in the upper half of the serum cholesterol distribution, where thrombotic stroke will be more common because of its positive association with serum cholesterol. These observations can probably reconcile the reduction in the incidence of stroke in trials (where most of the strokes will have been thrombotic) with the absence of an association between serum cholesterol and stroke mortality in cohort studies. Grade A Trials that distinguished thrombotic from hemorrhagic stroke showed that the risk of

P (trend)

thrombotic stroke was significantly reduced by a statin,11 but there are too few data on hemorrhagic stroke to confirm or refute the inverse association with cholesterol shown in cohort studies. Even if the association between low cholesterol and hemorrhagic stroke is cause and effect, however, the increased mortality from hemorrhagic stroke due to very low cholesterol concentrations is small compared to the lower mortality from other vascular diseases. For example, in Table 12.4 the mortality from all circulatory diseases at the lowest serum cholesterol (4·1 mmol/l) was 110 per 100 000 man years, lower than the rate of 147 per 100 000 man years in the next highest cholesterol group. Patients who have had a thrombotic stroke are at high risk of a recurrent event and should receive statins, as should patients with carotid artery disease and others at high risk. Patients who have had a hemorrhagic stroke should not receive statins.

Peripheral arterial disease Observational data show the expected association between peripheral arterial disease and serum cholesterol. In a large case–control study the association was equivalent in magnitude to an increase in risk of intermittent claudication of about 24% for a 0·6 mmol/l increase in serum cholesterol22 (uncorrected for regression dilution bias), similar in magnitude to the association of serum cholesterol with ischemic heart disease. In the 4S trial (serum cholesterol reduction 1·8 mmol/l) the incidence of intermittent claudication was reduced by 38% (95% confidence interval 12%, 56%; 52 v 81 cases).23

Abdominal aortic aneurysm The pathology of the condition is complex, but abdominal aortic aneurysms are associated with atheromatous disease and tend to coexist with coronary artery or peripheral arterial disease. Abdominal aortic aneurysms are associated with 125


a higher serum LDL cholesterol and triglyceride and a lower HDL cholesterol. Other circulatory diseases Table 12.4 shows a strong association between serum cholesterol and all circulatory diseases other than ischemic heart disease and stroke. Deaths from peripheral arterial disease and abdominal aortic aneurysm are too infrequent to account fully for this association. It is probably attributable also to poorly certified ischemic heart disease: deaths certified due to atrial fibrillation, heart failure, myocardial degeneration and atherosclerosis, for example, are in many cases due to ischemic heart disease.

Safety of cholesterol reduction

occurs in the first year after lowering cholesterol. Crosssectional studies of dietary saturated fat and serum cholesterol showed little or no relationship, an observation that was wrongly interpreted as indicating that lowering dietary saturated fat did not reduce cholesterol, until randomized trials established that it did. (The weak cross-sectional association arises because the inaccuracy in measuring individual dietary saturated fat is large in comparison to the small degree of variation between individuals in true saturated fat consumption.27) Clinicians were reluctant to accept that there was benefit in lowering average levels of serum cholesterol in high-risk patients: the notion that the average serum cholesterol in entire western populations is high appeared counterintuitive. The issue of safety caused concern, as discussed above. Lastly, it has seemed inconsistent that serum cholesterol is a poor screening test yet an important cause of heart disease, as discussed below. All these issues are now satisfactorily resolved.

The uncertainty concerning the excess mortality from hemorrhagic stroke at low serum cholesterol concentrations is unresolved, as discussed above. This apart, there are no material grounds for concern about hazard. Trials of “statin” drugs, particularly informative on safety because of the large reduction in serum cholesterol that they achieve, have resolved the issue of safety because they show no excess mortality from non-circulatory causes.1,7–11 The excess mortality from cancer and accidents and suicide at very low serum cholesterol in observational studies is attributable to cancer or depression lowering serum cholesterol, not the reverse.19 Further reassurance on safety is provided by the condition of heterozygous familial hypobetalipoproteinemia, in which serum cholesterol levels are as low as 2–3 mmol/l. Life expectancy is prolonged because coronary artery disease is avoided, and no adverse effects from the low cholesterol are recognized24,25 – an important natural experiment. Statins as drugs are safe, with few adverse effects. The rare complication of rhabdomyolysis, with severe muscular pain and myoglobulinuria, has received attention with the withdrawal from the market of cerivastatin, but this is thought to affect only about one in 250 000 patients using the other statins. It is commoner with concomitant therapy with cytochrome P450 metabolized drugs, of which erythromycin and fibrates (especially gemfibrozil) are the most common.

Dietary fat increases blood levels of coagulation factor VII and hence increases the risk of thrombosis, myocardial infarction and cerebral thrombosis.28,29 Saturated and unsaturated fat increase factor VII equally, and the increase appears directly related to the extent of postprandial lipemia. The importance of this effect in increasing the risk of cardiovascular death is difficult to quantify. However, analyses of differences in serum cholesterol and ischemic heart disease mortality between different populations (so-called “ecological” comparisons), such as the Seven Countries Study, yield significantly larger estimates of the relationship than obtained from the cohort studies and trials discussed above, and differences between populations in serum cholesterol are largely attributable to differences in dietary fat, whereas genetic differences account for over half the variation in serum cholesterol between individuals in a cohort. At age 60, for example, the ecological estimate is a 38% difference in risk for a 0·6 mmol/l cholesterol difference, compared to a 27% difference in the cohort studies (Table 12.2).3 The difference may partly reflect the effect of dietary fat on heart disease risk.

Why was cholesterol reduction contentious?

Triglycerides

A few years ago many clinicians regarded serum cholesterol reduction with uncertainty or suspicion. Until the early 1990s unfavorable evidence had been reported at regular intervals over the previous 30 years. The earliest trials used toxic agents to lower serum cholesterol, notably estrogen (in men) and thyroxine. Some early trials were short in duration26 and showed no reduction in risk, because none

Serum triglyceride concentration was associated with the risk of ischemic heart disease in many cohort studies, but the association is subject to confounding by serum LDL and HDL cholesterol, diabetes and other factors.4,30 The effect of dietary fat increasing factor VII will also produce an indirect association between triglycerides and heart disease mortality. Whether an independent association exists is

126

Dietary fat and coagulation


contentious. Very high serum levels of triglyceride caused by genetic defects (familial lipoprotein lipase deficiency, for example) are not associated with atheroma or coronary artery disease, and this observation, together with the potential for confounding in cohort studies, suggests that a material cause and effect relationship between serum triglyceride and heart disease is unlikely.

double the population average.30 The 5% of men with highest LDL cholesterol (or its carrier protein, apolipoprotein B) experience 17% of the heart disease deaths.30 Grade A Including HDL cholesterol improves this poor detection by only about one percentage point. Lipids cannot identify a small minority of the population in whom the majority of future heart disease deaths will cluster.

High density lipoprotein cholesterol

Appropriate policy

There is an inverse association between HDL cholesterol (or apolipoprotein A1) and ischemic heart disease. An absolute increase corresponding to 0·12 mmol/l (about 10% of the average value) is associated with about a 15% decrease in the risk of ischemic heart disease at age 604,30 or a 20% decrease with adjustment for the regression dilution bias.30 The effect of alcohol in increasing HDL cholesterol is the major mechanism for the lower risk of heart disease in drinkers.31 The effect of smoking in decreasing HDL cholesterol contributes to the excess risk of heart disease in smokers. The statin cholesterol lowering drugs increase HDL cholesterol relatively little. Certain other cholesterol lowering drugs (such as fibrates and niacin) increase HDL cholesterol more, but even in persons with relatively low HDL cholesterol the overall protective effect of these drugs is smaller because they reduce LDL cholesterol less, and so they should not be preferred to statins.

In a small proportion of the population, notably persons with familial hypercholesterolemia, the absolute risk of death from ischemic heart disease at a young age is so great that affected persons should be identified and treated, even though the condition accounts for a fraction of all heart disease deaths in a population. Grade A The most appropriate screening strategy has not yet been devised; measuring lipids in relatives of known cases will not identify all cases. Because screening cannot identify a group who would not benefit from a reduction in dietary fat and serum cholesterol, such measures should be directed at the entire population. Serum cholesterol reductions of 0·6 mmol/l (10%), as discussed above, have occurred in entire western communities, facilitated by health education, the wider availability of healthy food in restaurants and supermarkets, and a positive image of healthy eating. A reduction of 0·6 mmol/l is less likely when an individual attempts dietary change in isolation. The most important measures to lower cholesterol in healthy people therefore involve wider public education, encouragement of labeling of the nutrient content of foods, and the widespread availability of palatable low-fat foods. Clinicians need to direct their activities towards high-risk patients, and the most important high-risk group (based on the proportion of all heart disease deaths that can be anticipated) are patients who have had a myocardial infarction. As a group, these patients face a risk of death from ischemic heart disease of about 5% per year (untreated), a risk that varies relatively little with age or sex. As in healthy people, serum cholesterol testing cannot identify a substantial group at either materially higher or materially lower than average risk of death. Also, the evidence strongly indicates that there is no threshold below which serum cholesterol reduction is not effective. It follows that serum cholesterol should be reduced in all survivors of myocardial infarction. Simvastatin and atorvastatin can lower serum cholesterol by 1·8 mmol/l (30%) and this, as discussed above, will reduce mortality from heart disease by about 60% after 2 years, a substantially larger reduction in risk than can be achieved by any other single intervention. Other high-risk groups include patients with angina, patients who have had a thrombotic stroke, patients with carotid artery disease, patients with peripheral arterial disease and claudication,

Lipids as screening tests Serum cholesterol reduction is important in reducing the risk of ischemic heart disease, but cholesterol and other lipids are poor population screening tests for ischemic heart disease. The reason for the apparent discrepancy is that the screening potential of a factor depends not only on the strength of its relationship with disease, but also on its variation in magnitude across individuals in a community. In the case of lipids, the high average values in western societies place everyone at risk, and the variation between individuals is too small for use in population screening. By analogy, if everybody smoked between 15 and 25 cigarettes per day, cases of lung cancer would not cluster in the minority who smoked 25 cigarettes a day to the extent that those who smoked 15 or 20 could be ignored. Moreover, the Gaussian distribution of serum cholesterol means that many people have values around the average and few have relatively high values, so that most ischemic heart disease events will occur in people whose serum cholesterol is about average. Among men aged 35–64, the 5% with the highest serum total cholesterol experience only about 12% of all deaths from ischemic heart disease – their risk is little more than

127


and diabetics. All these patients should receive statins routinely. The “population” and “high-risk” approaches are complementary – the first primarily a public health issue aimed at altering the population diet and hence the incidence of ischemic heart disease, the second primarily a clinical activity, identifying and treating with statins patients with coronary artery disease.

is uncertain. The possible hazard, however, is greatly outweighed by the benefit of low mortality from heart disease at very low cholesterol levels.

Screening ●

●

Conclusions ●

The high levels of serum cholesterol found in western populations are a major cause of the high mortality from ischemic heart disease and, to a lesser extent, stroke and other circulatory diseases. Realistic dietary change in a community can lower serum cholesterol by 0·6 mmol/l (10%) and reduce heart disease mortality by about 25–30%. Simvastatin and atorvastatin can lower cholesterol by 1·8 mmol/l (30%) and reduce the risk of heart disease death by about 60% from the third year onwards, and should be offered to all high-risk survivors. Despite the importance of lowering cholesterol, lipids are poor screening tests of individual risk, because the average risk is high and the range across a population is relatively narrow. Serum cholesterol and ischemic heart disease Grade A ● The effect of serum cholesterol reduction on ischemic heart disease mortality is large and important. ● There is little reduction in risk in the first year, but the expected reduction in risk is largely attained from the third year onwards. ● There is no threshold across the range of cholesterol values in western countries below which reducing serum cholesterol reduction is not worthwhile. In particular there is no justification for withholding statins from high risk patients whose serum cholesterol is below 5 mmol/l. ● The greater the reduction in serum cholesterol, the greater the reduction in risk. ● Simvastatin and atorvastatin can lower cholesterol by 1.8 mmol/l (30%) and reduce the risk of heart disease death at age 60 by about 60% from the third year onwards; their use should be routine in high-risk patients. ● Realistic dietary change in a community can lower serum cholesterol by 0.6 mmol/l (10%) and reduce the risk of heart disease death by 25–30% at age 60. In an individual acting alone, the realistic change is half this. ● The benefits are similar in men and women.

Serum cholesterol and other circulatory diseases ● ●

128

Statins reduce the risk of thrombotic stroke and peripheral arterial disease and should be used in high-risk patients. Cohort studies show excess mortality from hemorrhagic stroke at very low cholesterol levels. The interpretation

Lipids are poor screening tests for predicting heart disease death in an individual: it is not possible to identify a small minority in a community who will experience the majority of heart disease deaths. The 5% of men with highest total serum cholesterol experience only about 12% of heart disease deaths. In the extremity of the distribution, familial hypercholesterolemia is important to detect because the absolute risk of heart disease death at a young age is high, even though the condition accounts for only a small proportion of heart disease deaths.

References 1.Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383–9. 2.Law MR, Wald NJ. An ecological study of serum cholesterol and ischaemic heart disease between 1950 and 1990. Eur J Clin Nutr 1994;48:305–25. 3.Law MR, Wald NJ, Wu T, Hackshaw A, Bailey A. Systematic underestimation of association between serum cholesterol concentration and ischaemic heart disease in observational studies: data from the BUPA study. BMJ 1994;308:363–6. 4.Pocock SJ, Shaper AG, Phillips AN. Concentrations of high density lipoprotein cholesterol, triglycerides, and total cholesterol in ischaemic heart disease. BMJ 1989;298:998–1002. 5.American Heart Association, National Heart, Lung, and Blood Institute. The cholesterol facts: a summary of the evidence relating dietary fats, serum cholesterol, and coronary heart disease. Circulation 1990;81:1721–33. 6.Law MR, Wald NJ, Thompson SG. By how much and how quickly does reduction in serum cholesterol concentration lower risk of ischaemic heart disease? BMJ 1994;308: 367–72. 7.Shepherd J, Cobbe SM, Ford I et al. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. N Engl J Med 1995;333:1301–7. 8.Sacks FM, Pfeffer MA, Moye LA et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med 1996;335:1001–9. 9.The Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) Study Group. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. N Engl J Med 1998;339:1349–57. 10.Downs JR, Clearfield M, Weis S et al. Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels. JAMA 1998;279:1615–22. 11.Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in


20536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002;360:7–22. 12.Neaton JD, Wentworth D. Serum cholesterol, blood pressure, cigarette smoking, and death from coronary heart disease. Arch Intern Med 1992;152:56–64. 13.Chen Z, Peto R, Collins R et al. Serum cholesterol concentration and coronary heart disease in a population with low cholesterol concentrations. BMJ 1991;303:276–82. 14.Scandinavian Simvastatin Survival Study Group. Baseline serum cholesterol and treatment effect in the Scandinavian Simvastatin Survival Study (4S). Lancet 1995;345:1274–5. 15.Smith EB, Slater RS. Relationship between low-density lipoprotein in aortic intima and serum-lipid levels. Lancet 1972;i:463–9. 16.Mensink RP, Katan MB. Effect of dietary trans fatty acids on high-density and low-density lipoprotein cholesterol levels in healthy subjects. N Engl J Med 1990;323:439–45. 17.Nestel P, Noakes M, Belling B et al. Plasma lipoprotein lipid and Lp[a] changes with substitution of elaidic acid for oleic acid in the diet. J Lipid Res 1992;33:1029–36. 18.Neaton JD, Blackburn H, Jacobs D et al. Serum cholesterol level and mortality findings for men screened in the multiple risk factor intervention trial. Arch Intern Med 1992;152: 1490–500. 19.Law MR, Wald NJ, Wu T, Bailey A. Assessing possible hazards of reducing serum cholesterol. BMJ 1994;308:373–9. 20.Prospective Studies Collaboration. Cholesterol, diastolic blood pressure and stroke: 13 000 strokes in 450 000 people in 45 prospective cohorts. Lancet 1995;346:1647–53. 21.Crouse JR, Byington RP, Furberg CD. HMG-CoA reductase inhibitor therapy and stroke risk reduction: an analysis of clinical trials data. Atherosclerosis 1998;138:11–24.

22.Fowkes FGR, Housley E, Riemersma RA et al. Smoking, lipids, glucose intolerance, and blood pressure as risk factors for peripheral atherosclerosis compared with ischaemic heart disease in the Edinburgh Artery Study. Am J Epidemiol 1992; 135:331–40. 23.Kjekshuj J, Pedersen TR, Pyorala K, Olsson AG. Effect of simvastatin on ischaemic signs and symptoms in the Scandinavian Simvastatin Survival Study (4S). J Am Coll Cardiol 1997;29(Suppl A):75A. 24.Linton MF, Farese RV, Young SG. Familial hypobetalipoproteinemia. J Lipid Res 1993;34:521–41. 25.Glueck CJ, Gartside P, Fallat RW, Sielski J, Steiner PM. Longevity syndromes: familial hypobeta and familial hyperalpha lipoproteinemia. J Lab Clin Med 1976;88:941–57. 26.Frantz ID, Dawson EA, Ashman PL et al. Test of effect of lipid lowering by diet on cardiovascular risk. The Minnesota coronary survey. Arteriosclerosis 1989;9:129–35. 27.Jacobs DR, Anderson JT, Blackburn H. Diet and serum cholesterol. Am J Epidemiol 1979;110:77–87. 28.Miller GJ, Cruickshank JK, Ellis LJ et al. Fat consumption and factor VII coagulant activity in middle-aged men. An association between a dietary and thrombotic coronary risk factor. Atherosclerosis 1989;78:19–24. 29.Salomaa V, Rasi V, Pekkanen J et al. The effects of saturated fat and n-6 polyunsaturated fat on postprandial lipemia and hemostatic activity. Atherosclerosis 1993;103:1–11. 30.Wald NJ, Law M, Watt HC et al. Apolipoproteins and ischaemic heart disease: implications for screening. Lancet 1994; 343:75–9. 31.Gaziano JM, Buring JE, Breslow JL et al. Moderate alcohol intake, increased levels of high-density lipoprotein and its subfractions, and decreased risk of myocardial infarction. N Engl J Med 1993;329:1829–34.

129

13

Use of lipid lowering agents in the prevention of cardiovascular disease Jeffrey L Probstfield

Introduction According to the most recent data from the American Heart Association, 12 600 000 American adults have coronary heart disease (CHD).1 CHD is the leading cause of death among US adults, responsible for one of every five deaths in the United States in 1999. Although the age adjusted death rate from CHD decreased by 24% from 1989 to 1999, the actual number of deaths decreased by only 6·8% over this same time.1 The associated morbidity, treatment of related conditions and preventive approaches for CHD are reviewed in other chapters of this book. Discussed here is the practical use of lipid lowering agents to prevent hypercholesterolemia – a well-established risk factor for the development of CHD. Major trials have clearly demonstrated that decreases in low density lipoprotein cholesterol (LDL-C) are associated with reductions in total mortality,2–4 CHD mortality,2–4 fatal and non-fatal CHD as well as strokes.2–5 Other major trials have also shown that lowering LDL-C can retard the progression of coronary artery atherosclerosis6 and carotid atherosclerosis7 and may even cause their regression, as well as slow the progression and occlusion of atherosclerosis in saphenous vein bypass grafts.8 In both primary3 and secondary2,4 CHD prevention settings, decreases in total and causespecific mortality have been demonstrated, and these benefits have been shown in both subjects with elevated,2,3 average,5 and normal LDL-C levels. Evidence of the benefits of statins in reducing the risk of stroke and observations concerning their pleiotropic effects are also reviewed. This chapter’s primary purpose is to review the evidence regarding plasma lipid-altering medications, their mechanisms of action, dosages and dosing schedules, effects on lipid and lipoprotein variables, adverse effects, and clinical uses. The evidence for cholesterol lowering in such subgroups as the elderly, women, diabetic patients, and those with small-dense LDL particles will also be summarized. The 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitor agents (statins), which are highly effective LDL-C lowering agents, are reviewed, as are niacin, bile acid sequestering agents (resins), fibrates, and ezetimibe. A brief review of the costs/1% LDL-C lowering/year and cost effectiveness concludes this chapter. 130

The National Cholesterol Education Program (NCEP) has been instrumental in developing and promulgating guidelines for initiating LDL-C lowering. These guidelines, developed on the basis of the patient’s established baseline LDL-C and presence or absence of CHD or its risk factors, recommend treatment goals to attain desired levels of plasma LDL-C. To date, the NCEP has issued three Adult Treatment Panel (ATP) reports. ATP I emphasized primary prevention of CHD in persons with high (160 mg/dl) or borderlinehigh LDL levels (130–159 mg/dl) and 2 risk factors for development of CHD. In ATP II, persons with established CHD were targeted for intensive lipid lowering therapy. ATP III, disseminated in 2001, continues to identify elevated LDL-C as the primary target of cholesterol lowering therapy and to maintain attention on intensive treatment of patients with CHD.9 It expands the indications for intensive therapy to lower levels of cholesterol in clinical practice. A major new feature is that intensive LDL-C lowering treatment is a primary prevention measure for persons with multiple risk factors for developing CHD, as identified by the estimated 10 year CHD risk score developed from the Framingham data. ATP III sets the optimal LDL-C level as 100 mg/dl and defines low HDL-C as 40 mg/dl (previous cutpoint was 35 mg/dl).9 ATP III also recommends that persons with the metabolic syndrome – a constellation of major lipid and non-lipid risk factors, life-habit risk factors, and emerging risk factors – should be targeted for intensive therapeutic lifestyle changes. Characteristics of the metabolic syndrome include abdominal obesity, elevated blood pressure, insulin resistance, and atherogenic dyslipidemia – elevated triglycerides, small LDL particles, and low HDL-C. Atherogenic dyslipidemia should be treated with lipid-altering agents.9 Boxes 13.1–13.4 summarize the major new recommendations of ATP III and classifications of cholesterol levels.9

Use of individual lipid-altering agents In this short evidence-based overview, we focus on documented activities of known lipid- and lipoprotein-altering drugs on lipid and lipoprotein variables, and their related

Use of lipid lowering agents in the prevention of cardiovascular disease

Box 13.1 New features of adult treatment panel III Focus on multiple risk factors ● Raises persons with diabetes without CHD, most of whom display multiple risk factors, to the risk level of CHD risk equivalent. ● Uses Framingham projections of 10 year absolute CHD risk, (that is, the per cent probability of having a CHD event in 100 years) to identify certain patients with multiple (2) risk factors for more intensive treatment. ● Identifies persons with multiple metabolic risk factors (metabolic syndrome) as candidates for intensified therapeutic lifestyle changes. Modifications of lipid and lipoprotein classification ● Identifies LDL-C 100 mg/dl as optimal ● Raises categorical low HDL-C from 35 to 40 mg/dl ● Lowers the triglyceride classification cutpoints to give more attention to moderate elevations Adapted from Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III) 9

Box 13.2 ATP III classification of LDL, total and HDL cholesterol LDL cholesterol: 100 Optimal Near optimal/above optimal Borderline high High 190 Very high Total cholesterol: 200 Desirable Borderline high 240 High HDL cholesterol: 40 Low 60 High Reproduced from Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III) 9

adverse effects. We identify those issues that remain more speculative as such. The interested reader is referred to the excellent and more complete reviews by Lousberg et al,10 as well as the recent ATP III guidelines.9 HMG-CoA reductase inhibitors (statins) These agents have a powerful LDL-C lowering and those currently approved for use differ only in their dose-response curves and unit cost. Mevastatin was first isolated in 1976 by Endo and colleagues as a natural product from Penicillium species. A related natural product, lovastatin, was approved by the FDA for cholesterol lowering in 1987. Subsequently,

Box 13.3 Major risk factors (exclusive of LDL cholesterol) that modify LDL goals* ● Cigarette smoking ● Hypertension (blood pressure 140/90 mmHg or on antihypertensive medication) ● Low HDL cholesterol (40 mg/dl)† ● Family history of premature CHD (CHD in male firstdegree relative 55 years; CHD in female first-degree relative 65 years) ● Age (men 45 years; women 55 years) * Diabetes is regarded as a coronary heart disease (CHD) risk equivalent. LDL indicates low density lipoprotein; HDL high density lipoprotein. † HDL cholesterol 60 mg/dl counts as a “negative” risk factor; its presence removes 1 risk factor from the total count. Adapted from Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III) 9

Box 13.4 Three categories of risk that modify LDL cholesterol goals Risk category LDL goal (mg/dl) CHD and CHD risk equivalents 100 Multiple (2) risk factors* 130 0–1 risk factor 160 * Risk factors that modify the low density lipoprotein (LDL) goal are listed in Box 13.3. CHD indicates coronary heart disease. Adapted from Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III)9

simvastatin, pravastatin, fluvastatin, atorvastatin, and cerivastatin were developed and approved for use in the US.11 Cerivastatin was withdrawn later because of adverse effects.

Mechanism of action: lipid-altering effects Brown and colleagues demonstrated that lovastatin inhibits HMG-CoA reductase, the rate-limiting enzyme in cholesterol biosynthesis.12 Total body cholesterol synthesis is reduced by at least 20%. Ultimately a critical reduction in cholesterol concentration occurs in the liver cell leading to enhanced production of hepatic LDL receptors,13 and increased cellular uptake of LDL-C. Further, reduced very(V) LDL biosynthesis occurs. Although speculative, it appears that the mechanism by which an increased removal of VLDL from the plasma occurs best fits with the upregulation of LDL receptors and an enhanced removal of VLDLs from the plasma due to an alteration in VLDL structure (specifically apo B-100).14 131


Pleiotropic effects In addition to reducing cholesterol biosynthesis, other potential antiatherogenic mechanisms of action for the statins are under current, intense investigation. Their exact role and importance remains speculative. Inhibition of 3-HMG-CoA reductase may be pleiotropic.15 Pleiotropic effects of statins on the vascular system and the arterial walls – affecting endothelial function, inflammation, coagulation, plaque stabilization, and smooth muscle cell migration – have been identified.15–19 Several statins have been shown to decrease smooth muscle cell migration and inhibit cholesterol accumulation in macrophages.15 The small GTP-binding protein, Rho, has membrane localization and activity affected by posttranslational isoprenylation. Its role in mediating the direct vascular effects of statins is also under intense study.20 The Pravastatin Inflammation/CRP Evaluation (PRINCE) provides clinical evidence of anti-inflammatory properties of a statin.21 In this prospective, randomized, cohort study, pravastatin lowered levels of C-reactive protein (CRP), an inflammatory biomarker that is predictive of cardiovascular risk. Decreased CRP levels were seen as early as 12 weeks in pravastatin-treated participants (P 0·001). Pravastatin lowered the median CRP level by 16·9% versus placebo (P 0·001) at 24 weeks. The decreases occurred in both the primary and secondary prevention groups and occurred regardless of sex, age, smoking, body mass index, baseline lipid levels, diabetes, and use of aspirin or hormone replacement therapy.21 Results of the recently completed Prospective Pravastatin Pooling (PPP) Project – a meta-analysis of three large, placebo-controlled, randomized trials including almost 20 000 patients and 102 559 person-years of follow up – provide further clinical evidence that statins may be antiinflammatory and/or antithrombotic. In particular, statins may be beneficial in reducing strokes.22 Pooled data from two of the trials, CARE and LIPID, involving more than 13 000 patients, showed a 22% reduction in total strokes and a 25% reduction in non-fatal stroke.22 WOSCOPS, the third trial pooled for analysis, had a similar, but smaller, trend for reduction in total stroke. Pravastatin reduced the risk of non-hemorrhagic stroke over a wide range of lipid values in patients with documented CHD.22 These results contrast importantly with those of a 1995 meta-analysis, which found no effect of lipid lowering on stroke in earlier non-statin clinical trials.23 The Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL) trial provides clinical evidence of an anti-ischemic effect with statins. Atorvastatin reduced early recurrent ischemic events in patients with acute coronary syndromes.24 The statin (80 mg/day) was initiated 24 to 96 hours after an acute coronary syndrome to over 3000 adults with unstable angina or non-Q-wave myocardial infarction (MI). In the atorvastatin group, 14·8%

132

of patients had a primary end point (death, non-fatal acute MI, cardiac arrest with resuscitation, or recurrent symptomatic myocardial ischemia requiring emergency rehospitalization) versus 17·4% in the placebo group (P 0·048).23 The MIRACL investigators suggest that patients with acute coronary syndromes begin statin therapy before hospital discharge, regardless of baseline LDL-C levels. Atorvastatin Versus Revascularization Treatment (AVERT) compared the efficacy of aggressive cholesterol lowering therapy versus percutaneous transluminal coronary angioplasty in low-risk, stable patients with CHD. Results favor the use of aggressive lipid lowering over angioplasty in patients with mild to moderate CHD. In addition to significantly reducing LDL-C levels, atorvastatin was associated with a 36% reduction in ischemic events and a significant delay in time to first ischemic event.25 Dosage The recommended dosages of these agents have been described and are shown in Table 13.1.11 Statins are to be Table 13.1 Dose response lipid and lipoprotein changes (% change) Total dose (mg/d)

5

10

20

40

80

Lovastatin BID Simvastatin Pravastatin Fluvastatin Atorvastatin

Total cholesterol reductions 19 24 29 34 19 23 28 31 36 16 24 25 27 17 19 25 29 33 37 45 LDL cholesterol reductions

Lovastatin BID Simvastatin Pravastatin Fluvastatin Atorvastatin

28 26

34 30 22 39

40 38 32 22 43

42 41 34 25 50

47 37 35 60

19 19 11 12 26

27 18 24 14 29

24 19 19 37

9 12 4 6

8 13 7 5

Triglycerides Lovastatin BID Simvastatin Pravastatin Fluvastatin Atorvastatin

7 12

16 15 15 19

HDL cholesterol Lovastatin BID Simvastatin Pravastatin Fluvastatin Atorvastatin

8 10

9 12 7 6

From Physicians’ Desk Reference11

10 8 2 3 9


taken with the evening meal or if they are taken twice daily with the morning and evening meals. Higher dosages of both simvastatin and pravastatin have been approved by the FDA and are now on the market. Impact on lipid levels All of these agents except atorvastatin will lower plasma total cholesterol between 20 and 40% and LDL-C by 25 to 45% at maximum approved doses. Fluvastatin appears to lower cholesterol by up to 20% and LDL-C by 25% at maximum doses. To achieve 35–45% LDL-C lowering, daily doses of 40 mg of simvastatin or 80 mg of lovastatin are required. Triglycerides are reduced between 10 and 30%. HDL-C plasma levels are frequently increased by 5–10%, but the increases may be more modest or absent in those with inherently low levels. Lp(a) levels are not affected.11 Statin therapy alters small dense LDL particles to a larger more buoyant form and also normalizes the responsiveness of coronary vessels to vasoactive stimulus.26 E-selectin, a cell adhesion molecule with increased expression in atherosclerotic states, is reduced with simvastatin or atorvastatin as monotherapy or in combination with colestipol.27 Atorvastatin is a more powerful member of the statin class. Reductions in total cholesterol of 45–50%, LDL-C of up to 60%, and triglycerides of 35–45% are seen at 80 mg/day doses. Reductions in apo B levels of 35–40% have been observed.11 Changes in plasma levels of Lp(a) are small, if they occur.28 Increases in HDL-C are inconsistent but may reach 12%.29 Table 13.1 summarizes the effects of approved statins. Adverse reactions Overall adverse reactions occur in less than 2% of individuals. From 1 to 3% of persons taking a statin will have doserelated, elevated, hepatic enzyme levels.30 Most of these abnormalities are seen within the first 3 months of beginning treatment and require monitoring.30 In patients who abuse alcohol, there is an increased risk of hepatic toxicity. An extremely low incidence of adverse events (not significantly different from placebo) has been documented over 5·5 years in the Heart Protection Study (HPS),4 to be discussed in more detail below. Statins compete with other drugs for specific metabolic pathways of the cytochrome P450 system,31 whose enzymes act as a major catalyst for drug oxidation in the liver.32 Lovastatin and simvastatin undergo extensive firstpass metabolism by CYP3A4, and caution is urged in using them with cyclosporin (a known inhibitor of CYP3A4), particularly when other inhibitors of the cytochrome P450 system, such as azole-derived antifungal drugs, erythromycin and clarithromycin, are in use, as well as nefazodone and many HIV protease inhibitors. Atorvastatin is also at least

partially metabolized by CYP3A4 but inhibitors of this enzyme only mildly increase serum concentrations. Fluvastatin is metabolized mostly by CYP2C9 and few drug–drug interactions have been noted. Pravastatin has less potential for drug interaction with other substrates, inhibitors, or inducers of the CYP3A4 and CYP2C9 systems than the other statins because it is metabolized by sulfation, not the cytochrome system.32 From 5 to 10% of individuals taking statins may develop muscle enzyme elevations. However, one should consider discontinuing statin therapy if CPK increases by more than threefold. Rare (less than 0·1%) and reversible increases of greater than 10-fold in CPK levels have been described. The causes of CPK elevation remain unexplained. Statin monotherapy or combination therapy can cause myopathy, which, although rare, can progress to rhabdomyolysis.33 This effect can be seen with any statin; however, cerivastatin was voluntarily withdrawn from the world market in 2001 because of an increased rate of rhabdomyolysis compared with other statins. Rhabdomyolysis occurred more often in patients taking full-dose cerivastatin (0·8 mg/day) and with concomitant gemfibrozil, and it led to kidney failure and death in 52 cases.34 Clinical use Although the biggest proportional reduction in LDL-C levels occurs at low doses, the clinical response to statins is dosedependent, and it appears to be independent of patient characteristics, such as age, gender, smoking status and initial lipid and lipoprotein levels.35 ATP III calls for LDL-C lowering drug therapy in persons with CHD and CHD risk equivalents when the LDL-C is 130 mg/dl.9 In persons with two or more risk factors for the development of CHD, ATP III suggests that lipid lowering drug therapy also begin at LDL-C levels 130 mg/dl.9 ATP III also recommends that LDL-C be measured, either at admission or within 24 hours, in all patients hospitalized with a major coronary event.9 Lipid lowering drug therapy should be initiated at hospital discharge in a person with a coronary event or procedure if LDL-C is 130 mg/dl.9 Treatment initiation at hospital discharge takes advantage of patients’ likely higher motivation to comply with therapy at that time and may avoid the “treatment gap” that can occur if outpatient follow up is less consistent. ATP III still recommends lifestyle changes, including reduced cholesterol and saturated fat diets, weight loss if overweight, and physical activity in hyperlipidemic patients.9 Novel agents The newest statin, rosuvastatin, has been submitted for approval to both the FDA and regulatory authorities in Europe. Called a superstatin because of its potency, rosuvastatin 133


rapidly lowers total C and LDL-C while increasing levels of HDL-C.36 A long half life (20 hours) and lack of metabolism via the cytochrome P4503A4 isoenzyme have also been demonstrated.37 In a phase II study, rosuvastatin across a dose range of 1–80 mg lowered LDL-C by 34–65%. Phase III trials demonstrated greater reductions in LDL-C for rosuvastatin versus atorvastatin as well as greater increases in HDL-C. A starting dose of 10 mg will reduce LDL-C by approximately 50%. The drug appears to be well tolerated at doses up to 80 mg/day. Nicotinic acid In the early 1950s Attschult noted profound reductions in plasma total cholesterol and triglyceride levels in association with use of nicotinic acid. Nicotinic acid has the most marked clinical effect on triglycerides and HDL,38 and is the only lipid-altering agent to consistently lower Lp(a) plasma levels.39 It also can alter small, dense LDL particles to larger, more buoyant forms.39 Mechanism of action Nicotinic acid’s predominant effect on plasma lipid levels is to reduce production of very low-density lipoprotein (VLDL) particles40 with subsequently reduced production of intermediate density lipoprotein (IDL) and LDL particles. Nicotinic acid’s major effect on VLDL metabolism results from an inhibition of hormone-sensitive, lipase-induced lipolysis in adipose tissue, and decreased triglyceride esterification in the liver. HDL-C increases, to a greater extent with niacin than with other drugs, and appears to be related to reduced apo A-I clearance and increased production of apo A-II. How Lp(a) levels are reduced is unknown, but early nicotinic acid induced hepatotoxicity may play a role. Dosage Crystalline nicotinic acid is available in 0·1 and 0·5 g tablets. There is a sustained-release form in dosages of 0·125, 0·25 and 0·5 g. The maximum daily dose is usually 3 g (Table 13.2).41 A new extended-release form of niacin, available since 2000, has relatively mild hepatic effects and can be taken at bedtime to lessen cutaneous flushing.42 Extended-release niacin is essentially equivalent to immediate-release niacin in increasing HDL-C.43 Results Regardless of the patient’s clinical lipoprotein abnormality, dose-dependent reductions in total and LDL-C and plasma triglycerides have been achieved with use of nicotinic acid. HDL-C levels may increase 15–40%; the average increase is 25%, with increases commonly plateauing at a dosage 134

between 1·5 and 3·0 g/d. Reductions in Lp(a) of 25–30% are achieved.44 As noted above, small-dense LDL particles become larger and more buoyant during nicotinic acid therapy.45 In certain patients, optimal responses may be formulation- and dosing regimen-dependent.46 Adverse reactions Even at very low doses (0·05–0·10 g), nicotinic acid often causes cutaneous flushing (80%) and pruritus (50%). Other frequently noted adverse effects are gastrointestinal symptoms (5–20%), liver enzyme elevations (3–10%), and uric acid increases (5–10%). Liver enzyme elevations occur more commonly with slow-release preparations and rapid dose increases. The clinical picture of mild liver function abnormalities usually resolves with continued therapy or reduced doses. Some 5 to 10% of patients who are taking nicotinic acid will have abnormal glucose tolerance tests or fasting blood sugar levels. A flu-like syndrome that can include hepatitislike findings on liver biopsy, a secretory defect with profound decreases in LDL-C, decreases in HDL-C and a prothrombin time abnormality may occur. This clinical picture is dosedependent and resolves when the agent is stopped.47 Blurred vision with macular edema occurs very rarely. Prednisone is contraindicated for use with nicotinic acid; co-administration can result in patients manifesting clinical diabetes. Clinical use Many prescription and non-prescription forms of nicotinic acid are available in the US. Although non-prescription forms are usually less expensive, bioavailability may be a problem. Niaspan, Nicolar and Rugby brands are highly effective prescription nicotinic acids, and the latter is also available as an over-the-counter formulation. The larger crystalline-form tablets are scored, which allows easy tailoring of the therapeutic regimen starting with a single low dose of either 0·1 or 0·25 g/d. Dosing with the crystalline form requires three or four administrations a day. No preparation or dosing regimen has been shown to be superior to multiples of 0·1 g crystalline tablets administered four times a day. Many patients will have little or no effect from two administrations a day, unless using sustained-release preparations. Increases in the dosage are implemented only every few days.46 Clinicians commonly reduce the number of administrations to three times per day and use 0·5 g tablets starting with 0·25 g qd for the first week. Sustained-release preparations should be used only in those patients with a documented response to immediate-release forms. Nicotinic acid should always be taken with food. Hot drinks and alcoholic beverages should be avoided at time of administration and dosages should be reduced or perhaps restarted if several successive doses are missed. Cutaneous


Table 13.2 Summary of effects of non-statin lipid-altering agents Agent

Lipid/ lipoprotein indication

Dosage and dosing

Nicotinic acid

aTriglyceride (TGs) aLDL-C bHDL-C aLp(a)

1–3 g/d

Bile acid sequestrants

Fibric acid derivatives

aLDL-C

6–8 g/d maximum dose 3–4 admin/d 4–24 g cholestyramine 5–30 g colestipol

bHDL-C

2 admin/d,1 at major meal

aTGs

Clofibrate 1 g bid

Response expected

Common adverse effects

Comments

bTGs 20–80%

Cutaneous flushing, pruritis, GI symptoms, “Flu-like” syndrome

Start low dose Advance slowly Relative contraindications: aFBS, aLiver function test (LFTs)

GI symptoms

Premix, slow admin Alters absorption of other drugs, for example, glycosides, warfarin, etc.

bLDL-C 25–40% aHDL-C 25% bLp(a) 10–30%

bLDL-C 25–35% at maximum dose aTGs 15–20% aHDL-C 4–7%

Contraindicated in hypertriglyceridemia

bLDL-C 10–20% bTGs 40–55% bHDL-C 15–20%

GI symptoms

Gemfibrozil 0·6 g/bid Fenofibrate 0·4 g qd Selective cholesterol absorption inhibitor

bLDL-C aHDL-C

Ezetimibe 5 mg/d 10 mg/d

Will aLDL-C in hypertriglyceridemic patients Contraindicated in those with gall stones Marked dose alteration in those with chronic renal failure

bLDL-C 16–19% aHDL-C 3–3.5%

flushing and pruritus will occur routinely if these precautions are not followed. If symptoms occur, they are the most severe during the first administration. Pretreatment with aspirin or ibuprofen may lessen cutaneous reactions. Although nicotinic acid may profoundly alter glucose metabolism in some, many diabetic patients have had their lipid disorders successfully managed with this agent. A fasting blood sugar 115 mg/dl predicts subjects who will lose the acute insulin response with an intravenous glucose tolerance test.48 A fasting blood sugar level 100 mg/dl should identify those patients who can take nicotinic acid without development of clinical diabetes. Bile acid sequestering agents (resins) (Table 13.2) This class of agents was first developed for the treatment of cholestasis-related pruritus by Carey and Williams in 1960. Hashim and Van Itallie subsequently demonstrated that cholestyramine lowered plasma cholesterol and it has been

No common AEs shown to date

in clinical use for 30 years. Other agents in this class are colestipol and the recently approved colesevelam. Mechanism of action The enterohepatic circulation of bile acids allows for only 6 or 7% of them to be excreted each day. These polymers with a molecular weight of over 106 are not absorbed and function by binding bile acids in the gastrointestinal lumen. Since an increase in bile acid excretion from the body and an increased production in the liver occur, relative depletion of cholesterol from the liver cells occurs inducing an increased level of hepatic LDL-receptor activity.49,50 The net effect is an increase in the catabolism of LDL-C and decreased plasma levels. Dosage Resins are dispensed in individual packets and are also available in a cost effective bulk formulation. Scoops, equivalent 135


in size to the number of grams in one packet, are used to dispense from the parent container. The newest resin, colesevelam, has a hydrogel tablet formulation. Results Resins are associated with significant reductions in plasma total and LDL-C and with small increases in plasma HDL-C levels.51 Plasma triglycerides are inconsistently affected, but substantial increases may occur, if used in those with already elevated plasma triglyceride levels.52 In familial dysbetalipoproteinemia (type III or remnant removal disease) plasma triglyceride levels may increase by more than threefold. Adverse reactions No long-term adverse effects have been demonstrated.51 Drugs that are highly charged, including the cardiac glycosides, the anticoagulant warfarin, diuretic agents, as well as thyroid hormone, will have their absorption affected53 if taken in close proximity to resin administration. Concomitant warfarin and resin therapy may be extremely challenging. If a resin’s effect on the absorption of a specific medication is not known, the resin should be taken at least 4 hours before or 2 hours after other medications. In clinical situations of existing gastrointestinal malabsorption, the absorption of fat-soluble vitamins may also be reduced. Clinical use The biggest proportional reduction in lipid levels occurs at low doses and in those who have moderately elevated levels of cholesterol.54 Careful selection of the vehicle and logistics used in resin administration will promote long-term patient adherence. Premixing with cold water (taking advantage of the resin’s hygroscopic nature) and drinking the preparation slowing is by far the most frequent and successful method of administration. Still, some patients prefer mixing with a heavily textured juice. Pre-existing gastrointestinal symptoms should be addressed before resin therapy is started. Bloating, belching and increased flatus are related to rapid ingestion. Dyspepsia and increased stool consistency or frank constipation can be managed with increases in fluids or dietary fiber intake. The newest agent in the resin class is colesevelam, available for use in the US since 2000. It is a polymeric, highpotency, water-absorbing hydrogel with a non-systemic mechanism of action.55 Based on data from approximately 1400 subjects, colesevelam reduced LDL-C by a median of 20%; the reduction is dose-dependent. When combined with lovastatin, simvastatin, or atorvastatin, colesevelam will reduce LDL-C levels by 8 to 16% over that seen with the statin alone.55 Colesevelam has also been shown to increase HDL-C up to 9%; however, increases in triglycerides, as much 136

as 25%, have also been reported.56 Colesevelam does not cause constipation, which is likely to improve patient adherence,55 and is formulated as a tablet, which should eliminate the palatability problems that some patients have with resin powders.56 In drug-interaction studies, colesevelam was coadministered with digoxin, warfarin, sustained-release metoprolol and verapamil, quinidine and valproic acid, and no clinically significant effects on absorption were reported.57

Fibric acid derivatives (Table 13.2) The fibrates currently marketed in the US are clofibrate, gemfibrozil, and fenofibrate. Fibrates available in other countries include bezafibrate, fenofibrate, ciprofibrate, beclafibrate, etiofibrate and clinofibrate. In a WHO study clofibrate was shown to reduce modestly (P 0·05) all cardiovascular events. However, increases in non-cardiovascular morbidity and mortality and total mortality occurred.58 In the Helsinki Heart Study, gemfibrozil was associated with a 35% reduction in MIs, particularly in those with elevated levels of plasma LDL-C and triglycerides and low levels of plasma HDL-C. Increases in non-cardiovascular deaths and no reduction in total mortality was observed,59 leading to concerns about the use of fibrates. No fibrate trial has yet shown a significant decrease in total mortality. These agents are approved for use primarily in those with hypertriglyceridemia. Clofibrate can be toxic; in some early studies there was a high mortality rate from malignancy and gastrointestinal disease in association with its use. Therefore, its use should be restricted to patients with severe hypertriglyceridemia unresponsive to other fibrates, niacin, or a combination of niacin and fibrate. Gemfibrozil has been shown to lower the risk of CHD and stroke in men with previous CHD, and low HDL-C and low LDL-C levels. In the Veterans Affairs HDL Intervention Trial (VA-HIT),60 2531 men with CHD (mean HDL-C 31·5 mg/dl and mean LDL-C 111 mg/dl) were randomized to receive gemfibrozil 1200 mg/day or placebo. There was a 22% reduction in CHD over 5 years.60 A more recent VA-HIT study describes the effect of therapy on stroke. There were 134 confirmed strokes (90% ischemic), 76 and 58 in the placebo and gemfibrozil groups, respectively (P 0·03). Risk reduction was evident after 6 to 12 months of gemfibrozil use. Adjusted for baseline variables, the relative risk reduction with gemfibrozil was 31%.61 Attributing the reduction in CHD to a change in HDL-C levels has been questioned by some. Clearly the reduction in CHD may more properly be associated with changes in other lipoprotein particles62 than with modest changes in HDL-C levels. Although the number of strokes in the study are modest,61 this is the first suggestion that stroke can be reduced with a form of lipid-altering therapy that has little effect on LDL-C.


Mechanism of action Decreased synthesis of VLDL with more efficient lipolysis and increased VLDL triglyceride catabolism has long been speculated as the mechanism of action of fibrates on lipid metabolism. Schoonjans et al in 1996 offered direct evidence that fibrates and fatty acids work as ligands for a class of compounds called peroxisome proliferator-activated receptors, of the nuclear receptor superfamily.63 Peroxisome proliferator-activated receptor alpha partially mediates the inductive effects of fibrates on HDL-C levels by regulating the transcription of HDL apolipoproteins, apo A-1 and apo A-II. Four specific actions are noted: 1. 2.

3. 4.

increased hydrolysis of plasma triglycerides due to induction of LPL and reduction of apo-CIII expression stimulation of cellular fatty acid uptake and conversion to acyl-CoA derivatives due to increased expression of genes for fatty acid transport protein and acyl-CoA synthetase increased peroxisomal and mitochondrial beta-oxidation decreased synthesis of fatty acids and triglycerides with a concomitantly decreased production of VLDLs.

Gemfibrozil was associated with a greater reduction in clinical events than the amount of cholesterol lowering or increase in HDL would predict.64 This suggests that its effects on CHD are mediated by a different mechanism, possibly related to its effects on triglycerides or other lipoprotein particles and HDL. Fibrates also shift the size of LDL particles from smaller, denser forms to larger, more buoyant forms, which could be less atherogenic. Dosage Table 13.2 lists dosing information for the three fibrates marketed in the US. Bezafibrate, 0·2 g, is given tid. (0·4 g, sustained-release qd), fenofibrate, 0·3–0·4 g, is given qd, and ciprofibrate, 0·1–0·2 g, is given qd. Results In patients with familial combined hyperlipidemia, LDL-C levels may be reduced by fibrates, but, particularly in those with elevated baseline levels of plasma triglycerides there will almost uniformly be an increase in LDL-C levels as VLDL-C levels decrease.65 Gemfibrozil and clofibrate had similar impact on lipids and lipoproteins in a double-blind crossover study.66 In patients with moderate to severe forms of hypertriglyceridemia, reductions in plasma triglycerides of 40–60% may occur with concomitant increases of 12–30% in HDL-C levels, but 100% increases in LDL-C may occur.67 Adverse reactions Fibrates are associated with adverse effects in 5–10% of patients. GI side effects (5%) are the most common, but only

rarely are these sufficient to warrant discontinuation of the medication. The increased incidence of hepatobiliary disease (particularly gallstones) occurs with all agents in this class.68 Minor alterations in several plasma biochemical values may occur, but these are dose-dependent and usually transient. The effective non-toxic dose-range is narrow, and at high doses fibrates cause myositis. They may potentiate the effects of oral anticoagulants and oral hypoglycemics and might also interact with statins to raise the risk of rhabdomyolysis. Clinical use The primary indication for the use of these agents has shifted to treatment of severe hypertriglyceridemia and more specifically for familial dysbetalipoproteinemia, or remnant removal disease. They are preferred by those who are less experienced in the use of nicotinic acid in clinical circumstances with increases in both plasma LDL-C and reductions in HDL-C levels. Because of the long-term adverse effects on hepatobiliary function and the potential for increases in LDL-C levels, liver function tests and LDL-C levels must be monitored closely. Chronic renal failure requires a 50% reduction in gemfibrozil dose.69 Novel agents Cholesterol lowering agents with different mechanisms of action are in development. Ezetimibe is a novel cholesterol absorption inhibitor that selectively and potently inhibits intestinal absorption of dietary and biliary cholesterol.70 In phase II clinical trials, ezetimibe at 10 mg/day reduced LDL-C by 15% in 68% of patients and by 25% in 22% over 12 weeks. HDL-C increased by 3·5% and the drug was well tolerated.70 Ezetimibe may have additive effects if given in combination with a statin. When given in a fixed combination tablet with simvastatin, LDL-C was reduced by 52%.71 Policosanol is a phytochemical that is a mixture of higher primary aliphatic alcohols isolated from sugar cane wax.72 At dosages of 10 to 20 mg/day, it decreased total-C by 17 to 21% and LDL-C by 21 to 29%, while raising HDL-C by 8 to 15%.72 Policosanol appears to have an acceptable safety/tolerability profile. Combination therapy Combination drug therapy should be used when diet and single drug therapy do not reduce LDL-C levels to the desired levels. Verification of adherence to and the efficacy of a prescribed regimen should be made on at least two occasions at monthly intervals before adding to the regimen. Table 13.3 describes a stepped approach to combination therapy depending on the lipid or lipoprotein variable(s) that are the therapeutic objective. Recall that reduction in LDL-C is the only alteration in lipid(s) or lipoprotein(s) that has been unequivocally demonstrated to reduce risks for CHD 137


Table 13.3 Stepped approach to lipid medication altering therapies Elevated LDL-C

Elevated TG/LDL decreased HDL-C

1. Statin

Niacin

2. Statin resin 3. aStatin resin 4. aStatin resin niacin or ezetimibe

Niacinresin

Elevated Markedly Lp(a) elevated TGs Niacin

Fibrate/ niacin

Guidelines for selecting combination therapy Practitioners should review four questions before adding other agents to initial diet and lipid-altering drug therapy regimens.73 1. 2.

3.

aNiacinresin aNiacinstatin or

4. Statinfibrate

Has adherence to and efficacy of the initial regimen been verified? Does the patient have fasting hypertriglyceridemia? (Bile acid sequestering agents should be used as second or third agent only.) What contraindications are present mitigating addition of other lipid-altering agents? (Other diseases or clinical conditions, or other lipid-altering agents.) What are the total costs of additional drug therapy to the patient?

Efficacy of various combinations in clinical trials. Epidemiologic data demonstrate an increased risk associated with reduced levels of HDL-C,64 increased plasma Lp(a) (usually in association with increased LDL-C levels), and, to a lesser extent, increased plasma triglyceride levels (usually in association with other risk factors). Intervention studies demonstrating reduction in CHD risk with changes in other lipid and lipoprotein particles have yet to be done.

Selected examples of maximum lipid and lipoprotein alterations are given in Table 13.4. Prior to the development of atorvastatin the maximum lowering of LDL-C was demonstrated with a combination of lovastatin (40 mg/day), colestipol (30 g/day) and nicotinic acid (5·5 g/day) at 70%. Triglyceride reductions of 80% can be effected with nicotinic acid alone with little to be gained in efficacy by adding another agent. Lp(a) levels are affected substantially only by nicotinic

Table 13.4 Efficacy of selected combination of hyperlipidemic drug therapy in modifying plasma concentrations of total, LDL and HDL cholesterol levels Drug combination

138

% change

Reference

Total

LDL

HDL

Cholestyramine niacin lovastatin pravastatin

26 51 36

32 61 43

23 21 18

Angelin et al, 1986 Leren et al, 1988 Jacob et al, 1993

Colestipol niacin lovastatin niacin simvastatin fenofibrate

41 45 55 41 39

48 54 66 50 54

25 2 32 9 15

Packard et al, 1980 Illingworth et al, 1981 Malloy et al, 1987 Simons et al, 1992 Heller et al, 1981

Lovastatin gemfibrozil

34

40

7

Illingworth et al, 1989

Simvastatin gemfibrozil

54

58

18

Feussner et al, 1992

Atorvastatin colesevelam

31

48

11

Ezetimibe fenofibrate atorvastatin

27 38

36 55

1·9 1·1

Hunninghake et al, 2001

Bays et al, 2001 Bays et al, 2001


acid. HDL-C can be consistently raised by 25% with nicotinic acid alone with little further gain by adding other agents. Adverse effects The important adverse effects of the single agents are described in Table 13.2. As noted previously, the most serious interaction is seen when a statin drug is used in combination with cyclosporin and myopathy develops. While cessation of the statin allows symptomatic myopathy and elevated muscle enzymes to resolve, continued therapy at the same dose may lead to frank rhabdomyolysis necessitating hemodialysis. Statins have also been associated with myopathic syndromes in patients using erythromycin, niacin and gemfibrozil. Reduced levels of any statin should be used in transplant patients in association with niacin and gemfibrozil with careful monitoring of muscle enzyme levels. Erythromycin use should be absolutely contraindicated in transplant patients already on cyclosporin and a statin. If erythromycin is used the statin must be temporarily discontinued. Clinical use Although single-drug therapy offers a simple regimen, combination therapy with low-dose statin and low-dose bile acid sequestrant has been investigated.74,75 Since the largest portion of lipid alteration is effected at low doses of both of these classes of agents and they work by very different mechanisms, an additive or synergistic response may occur. Low-dose combinations provide a good clinical alternative for patients who have symptoms at higher statin dosages and for organ transplant patients. They also appear to be more cost effective than using a statin as a single agent. The newest resin, colesevelam, has been studied in combination with statins. Low-dose colesevelam and low-dose lovastatin were given in a double-blind, placebo-controlled study to 135 hypercholesterolemic patients.76 The combination lowered LDL-C by 34% and 32% (the agents were either taken at the same time or colesevelam was taken at dinner, lovastatin at bedtime). Both combinations were superior to either agent alone, and both decreased total-C by 21%. Neither combination treatment significantly changed HDL-C or triglycerides. All treatments were well tolerated.76 The resin was also studied alone or in combination with low-dose atorvastatin in hypercholesterolemic men and women. Combination therapy reduced LDL-C by 48%, statistically different from either low-dose atorvastatin or colesevelam alone, but did not affect triglycerides.77 All treatment groups had similar frequency of adverse effects and the combination was well tolerated.77 Colesevelam was also given in combination with simvastatin to 251 hypercholesterolemic patients in a randomized, double-blind, placebo-controlled format. All groups, including the placebo-treated patients,

had decreased LDL-C levels versus baseline. Among all combination treatment groups (given different dosages of colesevelam and simvastatin), the mean decrease in LDL-C was 42%; this exceeded the decrease with either simvastatin or colesevelam alone.78 Combination therapy was not significantly different from simvastatin monotherapy in effects on HDL-C and triglycerides. Side effects were similar among all treatment groups.78

Informed decisions about “gray zones” Data have now accumulated on the use of lipid lowering drugs in previously less-well-studied subgroups such as the elderly, women, and diabetic patients. How and when should we use lipid lowering drugs in the following groups? The elderly If one lives until age 80 in the US, the average additional life expectancy is 8 years. Older individuals appear to be at least as responsive to cholesterol lowering agents as those in younger age groups. While some have suggested that risk attenuates for those who have hypercholesterolemia at older age, the absolute risk for developing CHD outcomes in the elderly over a short time interval is much higher than it is in younger individuals. ATP III notes that most new CHD events and most coronary deaths occur in persons older than 65 and that a high LDL-C/low HDL-C level still has predictive power for development of CHD in an older person.9 WOSCOPS,3 (primary intervention, or PI) included patients up to the age of 64 years, 4S2 (secondary intervention, or SI) up to the age of 70 years, Post-CABG8 (SI) up to 74 years and CARE5 (SI) previously provided limited data for those up to 75 years. All except Post-CABG showed benefit on CHD and CVD mortality. WOSCOPS and 4S showed benefit on total mortality, although statistical significance of the data from WOSCOPS was marginal. All of these studies included limited analyses by age group. When data from WOSCOPS were pooled in the PPP, pravastatin significantly reduced relative risk of coronary events in older patients.79 More recently, the Heart Protection Study (HPS) enrolled over 20 000 participants – including 5082 women, 3982 type 2 diabetic patients and 1263 elderly patients between the ages of 75 and 80 years. It also enrolled 3421 subjects with low baseline LDL-C levels, and follow up lasted for 5·5 years.4 Recently reported results show that a dose of 40 mg simvastatin once daily yielded striking results in terms of reduced events: 12% reduction in total mortality, 17% reduction in vascular mortality, 22% reduction in CHD events, 27% reduction in all strokes, and 16% reduction in non-coronary revascularizations.4 Statin therapy appeared to be beneficial at all cholesterol levels – even in participants 139


whose baseline levels were well below the currently recommended target levels of 100 mg/dl.4,80 Other ongoing trials that will provide needed information on lipid therapy in the elderly are the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Disease Trial (ALLHAT) and the Women’s Health Initiative (WHI).81 Complete ALLHAT results, involving 10 000 participants without an upper age limit, are expected in 2002.81 WHI will evaluate the effects of diet and lowering of fats in 48 000 female participants, aged 79 or younger, with completion expected in 2007.81 According to ATP III, “hard-and-fast” age restrictions do not appear to apply to the use of lipid lowering drugs in elderly persons with established CHD.9 For primary prevention, ATP III recommends therapeutic lifestyle changes, including lowfat diet, exercise, and weight loss if overweight, and LDL lowering drugs if older persons are at increased risk because of multiple risk factors or advanced subclinical atherosclerosis.9 Women Women were not included in early cholesterol lowering trials because of concerns about confounding hormonal effects on lipids, specifically in premenopausal women. Yet the relationship between increases in plasma cholesterol and CHD exists for women at all ages. Based on recent secondary and primary prevention trials that did not convincingly show that hormone replacement therapy reduced CHD risk in postmenopausal women and did show benefits with statins, ATP III recommends a cholesterol lowering drug over hormone replacement for CHD risk reduction in women.9 The later onset of CHD in women should be factored into clinical decision making regarding cholesterol lowering drugs.9 Of the “older” major clinical trials, CARE (Cholesterol and Recurrent Events) enrolled a fairly high percentage of women, 14%. It reported a 46% reduction in major coronary events among women participants versus a 20% commensurate risk reduction in men.4 In the PPP, which included analysis of pooled data from CARE, pravastatin significantly reduced relative risk of coronary events in women.22 The first CHD primary prevention trial of statins to include women was the Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS).82 Among 997 postmenopausal women who received either placebo or lovastatin (20 to 40 mg/day), statin therapy showed consistent numerical decreases in first acute coronary major events and in all prespecified secondary end points (for example coronary revascularization, MI). The study was insufficiently powered, however, to detect significant differences between treatment groups.82 In HPS, 34% of participants enrolled were women, and benefits were consistent regardless of sex.4 Male and female participants experienced similar reductions in risk. 140

Diabetic patients Although aggressive control of blood glucose levels in type 2 diabetic patients reduces microvascular clinical outcomes, its effect on macrovascular disease outcomes remains unknown. Other traditional CHD risk factors are believed to increase dramatically the risk for clinical CHD events in these patients. Inherent in the diabetic disease process is an abnormality of lipoprotein lipase activity that is partially but not completely corrected by optimal glucose control. Any additional lipid and lipoprotein disorder(s) present in diabetic patients because of either inherited or secondary causes (obesity, alcohol consumption, etc.), accelerate atherosclerotic progression and increase the risk of clinical CHD events. Treatment of lipid disorders in diabetic patients with commensurate lowering of blood cholesterol levels suggests a similar treatment benefit in diabetic as in non-diabetic patients.2,3,5 The use of niacin in diabetic patients has traditionally not been recommended because of concerns about adverse effects on glycemic control. In the Arterial Disease Multiple Intervention Trial (ADMIT), however, niacin was given to diabetic patients in a prospective, randomized, placebo-controlled study enrolling 468 subjects, 125 of them with diabetes and diagnosed peripheral arterial disease.83 Niacin was given at 3000 mg/day or to maximally tolerated dose, for up to 60 weeks. Niacin significantly increased HDL-C and decreased triglycerides and LDL-C in all participants (P 0·01: niacin v placebo for all). It modestly raised glucose levels in all participants while HbA1c was unchanged from baseline through follow up.83 ADMIT investigators conclude that lipid-modifying doses of niacin can be safely used in patients with diabetes and that niacin may be considered an alternative therapy in such patients who do not tolerate statins or in whom statins do not correct hypertriglyceridemia or low HDL-C levels.83 In CARE, 586 normocholesterolemic diabetic patients with CHD (14% of total sample) were given pravastatin or placebo for 5 years. In the diabetic patients given pravastatin, there were 8% and 25% reductions respectively in absolute and relative risks of coronary events.84 Pravastatin also reduced the risk for revascularization procedures among diabetic patients by 32%. In subjects who were not diabetic but who had impaired glucose tolerance, pravastatin also substantially lowered the risk of recurrent coronary events.84 According to the pooled data in PPP, pravastatin significantly reduced relative risk of coronary events in diabetic patients.79 The HPS trial included 3980 persons with diabetes and 2930 of these had no CVD. As noted above, the event reductions seen with simvastatin occurred in diabetic patients as well. There was a 24% decrease in CVD and a 25% decrease in total CHD.4 Peripheral vascular disease In the Rancho Bernardo studies, patients with peripheral vascular disease had a several-fold increased risk of dying of


CHD. Greater than 80% of these individuals have CHD although some will manifest few symptoms. It is reasonable, but unsubstantiated, to treat these individuals as if they have CHD.

Small dense LDL-C particles (phenotype B) The entire population may be generally divided into two categories on the basis of predominant LDL species present in plasma. People with a predominance of smaller, more dense LDL particles exhibit an increased propensity for oxidative susceptibility of these species.85 These individuals have a higher risk for CHD, which may be associated to interrelated changes in plasma lipids, specifically an increase in triglycerides and reduced plasma levels of HDL. Alternatively, this pattern may be related to the insulin resistance syndrome, or syndrome X, which consists of impaired glucose tolerance, increased insulin levels, hypertension and abnormalities of coagulation factors. No trial of clinical outcomes and intervention of LDL subspecies has been done.

Costs and cost effectiveness of lipid alterations for CHD prevention True benefits for individuals and the public health have only been demonstrated for alteration of plasma LDL-C. One method for comparing the costs of cholesterol lowering is shown in Table 13.5 where the cost of the various statins is given in terms of the number of dollars per per cent of LDL-C lowering per year. Table 13.5 Comparative cost, dose and LDL-C lowering of statins Agent

Dose (mg)

LDL-C b AWP (%)b ($/day)c

Cost/1%/ LDLR ($/yr)

Lovastatin

10a 20 10a 80 20a 80 20a 40 10a 80

21 24 30 47 32 37 22 24 39 60

11·45 23·09 27·66 75·27 32·47 58·61 11·80 12·87 32·87 79·71

Simvastatin Pravastatin Fluvastatin Atorvastatin

Key points Lipid and lipoprotein alteration for prevention of coronary heart disease 1. Plasma LDL cholesterol lowering is effective in both men and women. Grade A1a 2. Plasma LDL cholesterol lowering is cost effective above the age of 35 years. Grade A1a 3. Plasma LDL cholesterol lowering is effective for ages at least to age 80 years. Grade A1a 4. Plasma LDL cholesterol lowering extends the patency survival of coronary artery saphenous vein grafts. Grade A1a 5. The variables (NIDDM, hypertension, increased plasma triglycerides, low plasma HDL-C, small-dense LDL and increased PAI-1 levels) that are components of the so-called “insulin resistance syndrome” appear to be a marker for individuals with small-dense LDLs. Grade B2 6. Alteration of plasma lipids (triglycerides) beyond total plasma cholesterol and LDL-C has not been demonstrated to affect CHD-cause specific morbidity or mortality. Grade A1a 7. Therapeutic increases in HDL-C may be associated with reductions in CHD. Associated changes in VLDLtriglyceride-rich particles appear to have an important role in CHD prevention. Grade A1a 8. Since it is effective even in high risk individuals with low initial LDL levels, consider initiation of appropriate therapy without initial determination of plasma lipid and lipoprotein levels. Grade A1a Best medical practice supports monitoring LDL-C levels during treatment. Grade C5

1·49 2·63 2·52 4·40 2·78 4·34 1·47 1·47 2·30 3·64

a

Common starting dose, qd. From Physicians’ Desk Reference11 c From Red Book Update. Montvale, NJ: Medical Economics Company; 2002 Cost/1%/LDLR was derived as cost/year/1% LDL reduction. b

Until the release of results from the 4S study, reductions in CHD morbidity from plasma cholesterol-related CHD had been modest and reductions in CHD and total mortality had not been demonstrated. Since elevated plasma cholesterol is a major risk factor for CHD and is prevalent in Western countries, evaluation of the cost effectiveness of plasma cholesterol lowering is important because of the size of the potential population for intervention and the associated healthcare costs of what can be lifelong medical therapy. In a sensitivity analysis,86 data from 4S demonstrate cost effectiveness of intervention for both men and women from 35–70 years and at plasma cholesterol levels above 213 mg/dl. The estimates of treatment costs for benefits observed in the 4S study, indicate that treatment is cost effective, among both men and women and at all plasma cholesterol levels between 213 and 309 mg/dl and with evidence of vascular disease. A WOSCOPS economic analysis, comparing pravastatin with dietary changes alone, showed the economic efficiency of therapeutic intervention with a statin.87 Caro et al used a generalized model of cardiovascular disease prevention and followed hypercholesterolemic men over a given time period to quantify the effect on cardiovascular diseases 141


avoided. Over a broad range of inputs and regardless of country, cost effectiveness ratios are below $35 000 per lifeyears gained. Pravastatin is cost-efficient in preventing CHD.87 Based on US medical price levels and the clinical trial evidence up to 1998, Hay et al concluded that statin therapy is cost effective (that is, cost less than $50 000/year of life saved) in any patient with an annual CHD risk 1%, including those with previous CHD or diabetes.88 In a recent cost-effectiveness analysis of the VA-HIT trial, the cost per year of life gained with gemfibrozil was estimated. Using the prices for gemfibrozil negotiated by the VA, gemfibrozil was cost-saving.89 Using prices for gemfibrozil paid outside the VA system, the cost of a quality-adjusted life-year saved by gemfibrozil ranged from $6300 to $17 000. The VAHIT investigators concluded that gemfibrozil reduced cardiovascular events in men with CHD and low levels of HDL-C and LDL-C at annual drug cost-savings of $100 or less in 1998 dollars. Even at higher drug prices, the cost of a life-year saved is well below the threshold considered cost effective.89 Malik et al 90 provide a graphic demonstration of how cost effective statin therapy is versus other widely used therapies for CHD, such as ACE inhibitors, blockers, and non-CV interventions, such as driver’s side air bags (Figure 13.1). The cost effectiveness threshold here is £25 000/year of life saved (US $35 000/year of life saved).

Ramipril high risk Ramipril HOPE population Ramipril low risk ACEI post-MI (AIRE) ACEI post-MI (SAVE) β blockers post-MI Statins (4S) Statins (WOSCOPS) Mild HT treatment Angiography Post-MI Driver’s side air bag

Hemodialysis

10 20 3 40 50 1000 Cost effectiveness (£/life year saved)

Figure 13.1 Comparison* of cost effectiveness of different strategies in prevention * Estimates from previous studies are in 1994–1995 UK pounds. Adapted from: Malik IS, et al Heart, 2001;85:539–43

Future directions The benefits of statins beyond the coronary vascular bed are now being intensively investigated. In addition to their lipid lowering properties, statins demonstrate pleiotropic effects on many aspects of atherosclerosis, such as plaque thrombogenicity, cellular migration, endothelial function, and thrombotic tendency.18 Whether or not these effects can be related to a decrease in progression of atherosclerosis or to reductions in acute coronary syndromes remains to be seen. Another area for more study is the benefit of moderate 142

versus aggressive LDL lowering. For example what are the risks and benefits of lowering LDL-C to 75 mg/dl compared to about 100 mg/dl? The exact role of manipulation of other lipoprotein particles also remains to be demonstrated. Summary The HMG-CoA reductase inhibitors are an important advance in the treatment of CHD and there is compelling evidence that LDL-C lowering with these agents can decrease the risk of CHD events and total mortality in both primary and secondary prevention. In addition to their effects on LDL-C, statins have pleiotropic effects, which may affect the development and the occurrence of clinical events of atherosclerosis. Lipid lowering therapy benefits the elderly, women, and diabetic patients, even if these individuals have normal LDL-C levels. Effective single- and combination-agent regimens for intervention for other plasma lipid and lipoprotein variables are also available. Key points Cardinal issues for using lipid altering agents 1. Is the diagnosis of hyperlipidemia certain? 2. Are there currently medications in the patient’s regimen that cause dyslipidemia or offer the potential for drug interactions with hypolipidemic therapy? 3. ALWAYS start the therapeutic regimen with diet and other lifestyle modifications. Grade A 4. The statins act as a class of agents, but possess different dose response curves. Some may have substantial levels of adverse effects. Grade A 5. The currently approved statins have powerful lipidaltering effects and a very low order of adverse effects. Grade A 6. Nicotinic acid is a powerful agent which can be effective in many people including some diabetic patients when used carefully. Grade A 7. Nicotinic acid is the most effective of any agent on HDL-C levels and the only one with a possible effect on Lp(a). Grade A 8. Resin therapy can be effective for lowering LDL-C plasma levels with careful attention to details of dosing and administration, particularly when added to low-dose statin or low-dose niacin. Grade A 9. Fasting hypertriglyceridemia is a relative contraindication to primary or combination resin therapy. Grade A 10. Fibrates are effective agents for lowering triglyceride particularly when extremely high and moderately raising HDL-C levels, but changes in LDL-C levels need to be monitored. Grade A 11. Are there contraindications to the specific hypolipidemic drug combinations? 12. Consider the direct and indirect costs before initiating primary lipid-altering therapy and with the addition of each agent to the combined regimen.


References 1.American Heart Association. 2002 Heart and Stroke Statistical Update. Dallas, TX: The American Heart Association, 2001. 2.Scandinavian Simvastatin Survival Study Group. Randomized trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1995;344:1383–9. 3.Shepherd J, Cobbe SM, Ford I et al for the West of Scotland Coronary Prevention study Group. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. N Engl J Med 1995;333:1301–7. 4.MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20 536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002;360:7–22. 5.Sacks FM, Pfeffer MA, Moye LA et al for the Cholesterol and Recurrent Events Trial Investigators. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med 1996;335: 1001–9. 6.Brown G, Albers JJ, Fisher LD et al. Regression of coronary artery disease as a result of intensive lipid-lowering therapy in men with high levels of apolipoprotein B. N Engl J Med 1990;323:1289–98. 7.Furberg CD, Adams HP, Applegate WB et al for the Asymptomatic Carotid Plaque Study (ACAPS) Research Group. Effect of lovastatin and warfarin on early carotid atherosclerosis and cardiovascular events. Circulation 1994;90:1670–87. 8.The Post Coronary Artery Bypass Graft Trial investigators. The effect of aggressive lowering of low-density lipoprotein cholesterol levels and low-dose anticoagulation on obstructive changes in saphenous-vein coronary-artery bypass grafts. N Engl J Med 1997;336:153–62. 9.Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). Executive summary. NIH Publication No. 01-3670 May 2001. 10.Lousberg TR, Denham AM, Rasmussen JR. A comparison of clinical outcome studies among cholesterol-lowering agents. Ann Pharmacother 2001;35:1599–607. 11.Physicians’ Desk Reference. Montvale, NJ: Medical Economics Co; 2002. 12.Brown MS, Faust JR, Goldstein JL. Inhibition of 3-hydroxy-3methylglutaryl coenzyme A reductase activity in human fibroblasts incubated with compactin (ML-236B), a competitive inhibitor of the reductase. J Biol Chem 1978;253:1121–8. 13.Brown MS, Goldstein JL. A receptor-mediated pathway for cholesterol homeostasis. Science 1986:232:34–47. 14.Berglund LF, Beltz WF, Elam RL, Witztum JL. Altered apolipoprotein B metabolism in very low density lipoprotein from lovastatin-treated guinea pigs. J Lipid Res 1994;35: 956–65. 15.Corsini A, Bellosta S, Baetta R, Faumgalli R, Paoletti R, Bernini F. New insights into the pharmacodynamic and pharmacokinetic properties of statins. Pharmacol Ther 1999;84: 413–28. 16.Gotto AM Jr, Farmer JA. Pleiotropic effects of statins: do they matter? Curr Opin Lipidol 2001;12:391–4. 17.Farmer JA. Pleiotropic effects of statins. Curr Atheroscler Rep 2002;2:208–17.

18.Vaughan CJ, Gotto AM Jr, Basson CT. The evolving role of statins in the management of atherosclerosis. J Am Coll Cardiol 2000;35:1–10. 19.Faggiotto A, Paoletti R. State-of-the-art lecture. Statins and blockers of the renin-angiotensin system: vascular protection beyond their primary mode of action. Hypertension 1999;34: 987–96. 20.Laufs U, Liao JK. Direct vascular effects of HMG-CoA reductase inhibitors. Trends Cardiovasc Med 2000;10:143–8. 21.Albert MA, Danielson E, Rifai N, Ridker PM, for the PRINCE Investigators. Effects of statin therapy on C-reactive protein levels: the pravastatin inflammation/CRP evaluation (PRINCE): a randomized trial and cohort study. JAMA 2001;286:64–70. 22.Byington RP, Davis BR, Plehn JF et al. Reduction of stroke events with pravastatin: the Prospective Pravastatin Pooling (PPP) Project. Circulation 2001;103:387–92. 23.Hebert PR, Gaziano JN, Hennekens CH. An overview of trials of cholesterol lowering and risk of stroke. Ann Intern Med 1995;155:50–5. 24.Schwartz GG, Olsson AG, Ezekowitz MD et al. Myocardial Ischemia Reductions with Aggressive Cholesterol Lowering (MIRACL) Investigators. Effects of atorvastatin on early recurrent ischemic events in acute coronary syndromes: the MIRACL study: a randomized controlled trial. JAMA 2001;285:1711–18. 25.Waters DD. Medical therapy versus revascularization: the atorvastatin versus revascularization treatment AVERT trial. Can J Cardiol 2000 Jan;16(Suppl. A):11A–13A. 26.Anderson TJ, Meredith IT, Yeung AC, Frei B, Selwyn AP, Ganz P. The effect of cholesterol-lowering and antioxidant therapy on endothelium-dependent coronary vasomotion. N Engl J Med 1995; 332:488–93. 27.Hackman A, Yasunori A, Insull W Jr, Pownall H, Smith L, Dunn K, Gotto AM Jr, Ballantyne CM. Levels of soluble cell adhesion molecules in patients with dyslipidemia. Circulation 1996; 93:1334–8. 28.Nawrocki JW, Weiss SR, Davidson MH et al. Reduction of LDL cholesterol by 25% to 60% in patients with primary hypercholesterolemia by atorvastatin, a new HMG-CoA reductase inhibitor. Arterioscler Thromb Vasc Biol 1995;15:678–82. 29.Bakker-Arkema RG, Davidson MH, Goldstein RJ et al. Efficacy and safety of a new HMG-CoA reductase inhibitor, atorvastatin, in patients with hypertriglyceridemia. JAMA 1996;275: 128–33. 30.Farmer JA, Torre-Amione G. Comparative tolerability of the HMG-CoA reductase inhibitors. Drug Saf 2000;23:197–213. 31.Neuvonen PJ, Jalava K-M. Itraconazole drastically increases plasma concentrations of lovastatin and lovastatin acid. Clin Pharmacol Ther 1996;60:54–61. 32.Beaird SL. HMG-CoA reductase inhibitors: assessing differences in drug interactions and safety profiles. J Am Pharm Assoc (Wash) 2000;40:637–44. 33.Omar MA, Wilson JP, Cox TS. Rhabdomyolysis and HMG-CoA reductase inhibitors. Ann Pharmacother 2001;35:1096–107. 34.Furberg CD, Pitt B. Withdrawal of cerivastatin from the world market. Curr Control Trials Cardiovasc Med 2001;2:205–7. 35.Shear CL, Franklin FA, Stinnett S et al. Expanded Clinical Evaluation of Lovastatin (EXCEL) study results: effect of patient characteristics on lovastatin-induced changes in plasma concentrations of lipids and lipoproteins. Circulation 1992;85:1293–303.

143


36.Hanefeld M. Clinical rationale for rosuvastatin, a potent new HMGCoA reduction inhibitor. Int J Clin Pract 2001;55: 399–405. 37.McTaggart F, Buckett L, Davidson R et al. Preclinical and clinical pharmacology of rosuvastatin, a new 3-hydroxy-3-methylglutaryl-coenzyme A reductase inhibitor. Am J Cardiol 2001;87:28B–32B. 38.Alderman JD, Pasternak RC, Sacks FM, Smith HS, Monrad S, Grossman W. Effect of modified, well tolerated niacin regimen on serum total cholesterol, high density lipoprotein cholesterol and the cholesterol to high density lipoprotein ratio. Am J Cardiol 1989;64:725–9. 39.Carlson LA, Hampsten A, Asplund A. Effects of hyperlipidemic drugs on serum levels of lipoprotein Lp(a) in hyperlipidemic subjects treated with nicotinic acid. J Int Med 1989; 226:271–6. 40.Grundy SM, Mok HYI, Zech L, Berman M. Influence of nicotinic acid on metabolism of cholesterol and triglycerides in man. J Lipid Res 1981;22:24–36. 41.Illingworth DR, Stein EA, Mitchel YB et al. Comparative effects of lovastatin and niacin in primary hypercholesterolemia. Arch Intern Med 1994;154:1586–95. 42.White Robinson A, Sloan HL, Arnold G. The antilipidemic effects of plain and extended-release niacin. Prev Cardiol 2000;3:131–5. 43.Knopp RH. Evaluating niacin in its various forms. Am J Cardiol 2000;86:51L–56L. 44.Superko HR, Krauss RM. Differential effects of nicotinic acid in subjects with different LDL subclass patterns. Atherosclerosis 1992;95:69–76. 45.Zambon A, Brown BG, Hokanson JE, Brunzell JD. Hepatic lipase changes predict coronary artery disease progression/ regression in Familial Atherosclerosis Treatment Study (FATS). Circulation 1997;94:1–539. 46.Probstfield JL, Hunninghake DB. Nicotinic acid as a lipoproteinaltering agent: therapy directed by the primary physician. Arch Intern Med 1994;154:1557–9. 47.Patterson DJ, Dew EW, Gyorkey F, Graham DY. Niacin hepatitis. South Med J 1983;76:239–41. 48.Brunzell JD, Robertson RP, Lerner RL et al. Relationships between fasting plasma glucose levels and insulin secretion during intravenous glucose tolerance tests. J Clin Endocrinol Metab 1976;42:222–9. 49.Shepherd J, Packard CJ, Bicker S, Lawrie TD, Morgan HG. Cholestyramine promotes receptor-mediated low-densitylipoprotein catabolism. N Engl J Med 1980;302:1219–22. 50.Kovanen PT, Bilheimer DW, Goldstein JL, Jaramillo JJ, Brown MS. Regulatory role for hepatic low density lipoprotein receptors in vivo in the dog. Proc Natl Acad Sci 1981; 78: 1194–8. 51.Lipid Research Clinics Program. The Lipid Research Clinics Coronary Primary Prevention Trial results, I: reduction in the incidence of coronary heart disease. JAMA 1984;251:351–64. 52.Levy RI (Moderator). Discussants: Dietary and drug treatment of primary hyperlipoproteinemia, NIH Conference. Ann Intern Med 1972;77:267–94. 53.Hunninghake DB. Bile acid sequestrants. In Rifkind BM, ed. Drug treatment of hyperlipidemia. New York: Marcel Dekker, Inc., 1991.

144

54.Superko HR, Greenland P, Manchester RA. Effectiveness of low dose colestipol therapy in patients with moderate hypercholesterolemia. Am J Cardiol 1992;70:135–40. 55.Davidson MH, Dicklin MR, Maki KC, Kleinpell RM. Colesevelam hydrochloride: a non-absorbed, polymeric cholesterol-lowering agent. Expert Opin Investig Drugs 2000; 9:2663–71. 56.Aldridge MA, Ito MK. Colesevelam hydrochloride: a novel bile acid-binding resin. Ann Pharmacother 2001;35:898–907. 57.Donovan JM, Stypinski D, Stiles MR, Olson TA, Burke SK. Drug interactions with colesevelam hydrochloride, a novel, potent lipid-lowering agent. Cardiovasc Drugs Ther 2000; 14: 681–90. 58.Committee of Principal Investigators. WHO cooperative trial on primary prevention of ischemic heart disease with clofibrate to lower serum cholesterol: mortality follow-up. Lancet 1980;2: 279–385. 59.Frick MJ, Elo O, Haapa K et al. Helsinki Heart Study: primary prevention trial with gemfibrozil in middle-aged men with dyslipidemia: safety of treatment, changes in risk factors, and incidence of coronary heart disease. N Engl J Med 1987; 317:1237–45. 60.Bloomfield Rubin H, Robins S et al. Gemfibrozil for the secondary prevention of coronary heart disease in men with low levels of high density lipoprotein cholesterol. N Engl J Med 1999;341:410–18. 61.Bloomfield Rubins H, Davenport J, Babikiam V et al. for the VAHIT Study Group. Reduction in stroke with gemfibrozil in men with coronary heart disease and low HDL cholesterol: the Veterans Affairs HDL Intervention Trial (VA-HIT). Circulation 2001;103:2828–33. 62.Robins SJ, Collins D, Wittes JT et al. VA-HIT Study Group. Veterans Affairs High-Density Lipoprotein Intervention Trial. Relation of gemfibrozil treatment and lipid levels with major coronary events: VA-HIT: a randomized controlled trial. JAMA 2001;285:1585–91. 63.Schoonjans K, Staels B, Auwrex J. Role of the peroxisome proliferator-activated receptors (PPAR), in mediating the effects of fibrates and fatty acids on gene expression. J Lipid Res 1996;37:907–25. 64.Gordon DJ, Probstfield JL, Garrison RJ, et al. High density lipoprotein cholesterol and cardiovascular disease: four prospective American studies. Circulation 1989; 79:8–15. 65.Hokanson JE, Austin ME, Zambon A, Brunzell JD. Plasma triglyceride and LDL heterogeneity in familial combined hyperlipidemia. Arterioscler Thromb 1993;13:427–34. 66.Illingworth DR. Fibric acid derivatives. In Rifkind BM, ed. Drug treatment of hyperlipidemia. New York: Marcel Dekker, Inc., 1991. 67.Rabkin SW, Hayden M, Frohlich J. Comparison of gemfibrozil and clofibrate on serum lipids in familial combined hyperlipidemia. A randomized placebo-controlled, double-blind, crossover clinical trial. Atherosclerosis 1988;73: 233–40. 68.Palmer RH. Effects of fibric acid derivatives on biliary litogenicity. Am J Med 1987;83(Suppl. 5B):37–43. 69.Goldberg AP, Sherrard DJ, Hass LB, Brunzell JD. Control of clofibrate toxicity in uremic triglyceride. Clin Pharmacol Ther 1977;21:317–25. 70.Bays HE, Moore PB, Drehobl MA et al. Ezetimibe Study Group. Effectiveness and tolerability of ezetimibe in patients with


primary hypercholesterolemia: pooled analysis of two phase II studies. Clin Ther 2001;23:1209–30. 71.Meng CQ. Ezetimibe (Schering-Plough). Curr Opin Invest Drugs. 2001;2:389–92. 72.Gouni-Berthold I, Berthold HK. Policosanol: clinical pharmacology and therapeutic significance of a new lipid-lowering agent. Am Heart J 2002;143:356–65. 73.Hoeg JM. Combination drug therapy. In Rifkind BM, ed. Drug treatment of hyperlipidemia. New York: Marcel Dekker, Inc., 1991. 74.Simons LA, Simons J, Parfitt A. Successful management of primary hypercholesterolaemia with simvastatin and low-dose colestipol. Med J Aust 1992;15:455–59. 75.Jacob BG, Richter WO, Schwandt P. Long-term treatment (2 years) with the HMG CoA reductase inhibitors lovastatin or pravastatin in combination with cholestyramine in patients with severe primary hypercholesterolemia. J Cardiovasc Pharmacol 1993;22:396–400. 76.Davidson MH, Toth P, Weiss S et al. Low-dose combination therapy with colesevelam hydrochloride and lovastatin effectively decrease low-density lipoprotein cholesterol in patients with primary hypercholesterolemia. Clin Cardiol 2001;24: 467–74. 77.Hunninghake D, Insull W Jr, Toth Davidson D, Donovan JM, Burke SK. Coadministration of colesevelam hydrochloride with atorvastatin lowers LDL cholesterol additively. Atherosclerosis 2001;158:407–16. 78.Knapp JJ, Schrott H, Ma P et al. Efficacy and safety of combination simvastatin and colesevelam in patients with primary hypercholesterolemia. Am J Med 2001;110:352–60. 79.Sacks FM, Tonkin AM, Shepherd J, et al. Effects of pravastatin on coronary disease events in subgroups defined by coronary risk factors: the Prospective Pravastatin Pooling Project. Circulation 2000;102:1893–1900. 80.Fox R. American Heart Association 2001 scientific sessions: late-breaking science – statins: the new aspirin? Circulation 2001;104:E9051–2. 81.The Cholesterol Treatment Trialists’ (CTT) Collaboration. Protocol for a prospective collaborative overview of all current and planned randomized trials of cholesterol treatment regimens. Am J Cardiol 1995;75:1130–4.

82.Clearfield M, Downs JR, Weis S et al. Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS): efficacy and tolerability of long-term treatment with lovastatin in women. J Womens Health Gender Based Med 2001; 10: 971–81. 83.Elam MB, Hunninghake DB, Davis KB et al. Effect of niacin on lipid and lipoprotein levels and glycemic control in patients with diabetes and peripheral arterial disease: the ADMIT study: a randomized trial. Arterial Disease Multiple Intervention Trial. JAMA 2000;284:1263–70. 84.Goldberg RB, Mellies MJ, Sacks FM et al. Cardiovascular events and their reduction with pravastatin in diabetic and glucose-intolerant myocardial infarction survivors with average cholesterol levels: subgroup analyses in the cholesterol and recurrent events (CARE) trial. The CARE Investigators. Circulation 1998;98:2513–19. 85.Chait A, Brazg RL, Tribble DL, Krauss RM. Susceptibility of small, dense low-density lipoproteins to oxidative modification in subjects with the atherogenic lipoprotein phenotype, pattern B. Am J Med 1993;94:350–6. 86.Weinstein MC, Stason WB. Foundations of cost effectiveness analysis for health and medical practices. N Engl J Med 1977;296:716–21. 87.Caro J, Klittich W, McGuire A, Ford I, Pettitt D, Norrie J, Shepherd J. International economic analysis of primary prevention of cardiovascular disease with pravastatin in WOSCOPS. West of Scotland Coronary Prevention Study. Eur Heart J 1999;20:263–8. 88.Hay JW, Yu WM, Ashraf T. Pharmacoeconomics of lipidlowering agents for primary and secondary prevention of coronary artery disease. Pharmacoeconomics 1999;15:47–74. 89.Nyman JA, Martinson MS, Nelson D, the VA-HIT Study Group. Cost-effectiveness of gemfibrozil for coronary heart disease patients with low levels of high-density lipoprotein cholesterol: the Department of Veterans Affairs High-Density Lipoprotein Cholesterol Intervention Trial. Arch Intern Med 2002;162:177–82. 90.Malik IS, Bhatia VK, Kooner JS. Cost effectiveness of ramipril treatment for cardiovascular risk reduction. Heart 2001; 85:539–43.

145

14

Blood pressure and cardiovascular disease Curt D Furberg, Bruce M Psaty

Definition A new classification of elevated blood pressure (BP) that places greater emphasis on systolic BP was introduced in the 1993 Fifth Report of the Joint National Committee on Detection, Evaluation and Treatment of High Blood Pressure (JNC V)1 and slightly modified in the 1997 JNC VI.2 Hypertension is defined as systolic blood pressure (SBP) 140 mmHg or greater and/or diastolic blood pressure (DBP) 90 mmHg or greater (Table 14.1). The new classification Table 14.1 Classification of blood pressure for adults aged 18 years and older a Category

Systolic (mmHg)

Optimalb Normal High normal Hypertension Stage 1c Stage 2c Stage 3c

120 130 130–139

and and or

140–159 160–179 180

or or or

a

Diastolic (mmHg) 80 85 85–89 90–99 100–109 110

Not taking antihypertensive drugs and not acutely ill. When systolic and diastolic blood pressures fall into different categories, the higher category should be selected to classify the individual’s blood pressure status. For example, 160/92 mmHg should be classified as stage 2 hypertension and 174/120 mmHg should be classified as stage 3 hypertension. Isolated systolic hypertension is defined as SBP 140 mmHg and DBP 90 mmHg and staged appropriately (for example 170/82 mmHg is defined as stage 2 isolated systolic hypertension). In addition to classifying stages of hypertension on the basis of average blood pressure levels, clinicians should specify presence or absence of target organ disease and additional risk factors. This specificity is important for risk classification and treatment. b Optimal blood pressure with respect to cardiovascular risk is 120/80 mmHg. However, unusually low readings should be evaluated for clinical significance. c Based on the average of two or more readings taken at each of two or more visits after an initial screening. Adapted from JNC VI 2

146

addresses the issue of severity or increased risk by defining three stages of hypertension, ranging from stage 1 (SBP 140–159 mmHg and/or DBP 90–99 mmHg) to stage 3 (SBP 180 mmHg and/or DBP 110 mmHg). In middle-aged populations, the most common type of elevated BP is combined systolic-diastolic hypertension. A second type, isolated systolic hypertension, generally occurs in older persons, probably as a result of age-related stiffening of the arteries. A recent Clinical Advisory Statement3 recommends that SBP should be the principal measure for the detection, evaluation, and treatment of hypertension in both middle-aged and older individuals. Hypertension is also classified as complicated or uncomplicated according to the presence or absence of target organ manifestations. These manifestations, which can be cardiac, cerebrovascular, peripheral vascular, renal, or retinal, represent complications of hypertension, and they also increase the risk of other hypertension-related complications (Box 14.1). Thus, hypertension is classified by its type (combined systolic-diastolic or isolated systolic hypertension), its

Box 14.1 Components for cardiovascular risk stratification in patients with hypertension Major risk factors ● Smoking ● Dyslipidemia ● Diabetes mellitus ● Age older than 60 years ● Gender (men and postmenopausal women) ● Family history of cardiovascular disease: women under age 65 or men under age 55 Target organ damage/clinical cardiovascular disease ● Heart diseases ● Left ventricular hypertrophy ● Angina/prior myocardial infarction ● Prior coronary revascularization ● Heart failure ● Stroke or transient ischemic attack ● Nephropathy ● Peripheral arterial disease ● Retinopathy Adapted from JNC VI 2

Blood pressure and cardiovascular disease

severity (stage 1–3), and by coexisting target organ manifestations (if present, complicated, or uncomplicated). These classifications are clinically important, since they have risk as well as treatment implications.

Prevalence In cohort analysis,4 mean SBP increases gradually with age, regardless of initial BP (Figure 14.1). Mean DBP also increases until the age of 55–60 years, when it levels off.4 Later in life there is a reduction in mean DBP, especially in those with high initial levels (Figure 14.1). The age-related changes in BP explain the increase in overall prevalence of hypertension with age and the increase in the prevalence of isolated systolic hypertension with advanced age. The prevalence of target organ manifestations also increases with age, as a result of the increasing prevalence and the longer duration of hypertension. The prevalence of hypertension is greater for AfricanAmericans than for non-Hispanic Whites and MexicanAmericans,5 and for less educated than more educated people.

Natural history Hypertension is one of the major risk factors for cerebrovascular disease (stroke), coronary heart disease (acute myocardial infarction [MI]), congestive heart failure (both systolic and diastolic dysfunction), and renal dysfunction. The risk is directly associated with the BP level and with the presence

of target organ manifestations and other cardiovascular risk factors. Ferrucci et al.6 calculated the cardiovascular risk score for each participant of the Systolic Hypertension in the Elderly Program (SHEP) using the Multiple Risk Factor Assessment Equation.7 The simple risk score is based on age, sex, total and HDL-cholesterol, SBP, smoking, and diabetes. In the placebo group, the 5 year rates of MI, stroke, and heart failure were progressively higher with higher quartiles of risk score in those who were free of cardiovascular disease at baseline. The relative event protection conferred by chlorthalidone-based treatment was similar across quartiles of risk. Thus, the absolute risk reduction increased by quartile of risk. This was reflected in a 2- to 10-fold lower “number needed to treat” (NNT) to prevent hypertensive complications in the highest-risk quartile (Figure 14.2). The authors concluded that hypertensive patients with additional cardiovascular risk factors should be the prime candidates for antihypertensive treatment. Disease burden Hypertension is one of the most common medical conditions in the developed world. It has been estimated that as many as 43 million adult non-institutionalized Americans have hypertensive BP levels or are taking antihypertensive medications. Another seven million persons may be controlling their hypertension using non-pharmacologic methods.5 Over the last two decades, the National High Blood Pressure Education Program has markedly raised the population’s awareness concerning the high prevalence and complications of hypertension.1 The number of patients taking

Groups determined at index examination Group 1 SBP60 n = 39

Figure 48.1 The frequency of supraventricular tachycardia (SVT), atrial fibrillation (AF), and non-sustained ventricular tachycardia (VT) at different ages in a consecutively referred population at St George’s Hospital, London (unpublished data)

hypertrophy of the interventricular septum (ASH) was considered to be the sine qua non of the disease. However, subsequent two-dimensional echocardiographic studies have shown that any pattern of hypertrophy is compatible with the diagnosis.53 The proportion of patients with concentric versus asymmetric hypertrophy depends on the definition employed. Thus, when a septal to posterior wall thickness ratio of 1·3:1 is used to define asymmetry, only 1–2% of patients have concentric left ventricular hypertrophy.53 However, this proportion rises to approximately 30% when a ratio of 1·5:1 is used.54 Criteria for abnormal wall thickness vary, but values exceeding two standard deviations from the mean corrected for age, sex, and height are generally accepted as diagnostic in the absence of any other cardiac or systemic cause. Doppler echocardiography is used to quantify the gradient across the left ventricular outflow tract using the modified Bernoulli equation: peak gradient 4V max 2 where Vmax is the maximum velocity across the left ventricular outflow tract. When it is not possible to obtain accurate Doppler measurements, the gradient can be estimated using M-mode recordings of the mitral valve and the formula: peak gradient 25(X/Y ) 25 where X is the duration of mitral–septal contact, and Y the period from the onset of systolic anterior motion of the mitral valve to the onset of mitral–septal contact.18 Cardiopulmonary responses to exercise In most patients with HCM, peak oxygen consumption is below the predicted value corrected for age, sex, and 706

Cardiac catheterization In the modern era, cardiac catheterization is performed only in patients with refractory symptoms (particularly those with severe mitral regurgitation), and in order to exclude epicardial coronary artery disease in older patients with chest pain. In addition to an outflow gradient, a variety of hemodynamic abnormalities are described including elevated left ventricular end-diastolic and pulmonary capillary wedge pressures, and a “spike and dome” appearance in the aortic waveform. Right atrial and right ventricular pressures are usually normal unless there is a substantial right ventricular outflow gradient or severe “restrictive” physiology. Resting cardiac output is typically normal or increased, except in patients with “end stage” ventricular dilatation. In patients with hypertrophy confined to the distal left ventricle, ventriculography may show a characteristic “spade-shaped” appearance caused by the encroachment of hypertrophied papillary muscles. Coronary arteriography is usually normal, but systolic obliteration of epicardial vessels is described. Muscle bridges are also described but their relevance to an individual patient’s symptoms is often difficult to assess. Radionuclide studies Several studies have used stress radionuclide imaging to study myocardial perfusion in patients with HCM. Fixed 201 thallium perfusion defects have been associated with increased left ventricular cavity dimensions, impaired systolic function, and reduced exercise capacity, and are thought to represent myocardial scars.29 Reversible regional 201 thallium defects are present in over 25% of patients, but correlate poorly with symptomatic status.27,29 It has been suggested that reversible defects are associated with a poor prognosis, but one large prospective study has failed to demonstrate any relation with medium-term outcome.56 Using positron emission tomography (PET) a reduction in coronary vasodilator reserve has been observed both in hypertrophied and non-hypertrophied regions of myocardium during dipyridamole-induced coronary microvascular vasodilatation.26 The reduction in vasodilator reserve may be more

Hypertrophic cardiomyopathy

pronounced in patients with a history of chest pain and STsegment depression.26 PET has also demonstrated subendocardial hypoperfusion after dipyridamole infusion across the septum of patients with asymmetrical septal hypertrophy. PET has been used to investigate the relationship between myocardial blood flow and metabolism using fluorine-18 labeled deoxyglucose (FDG).57,58 Areas of blood flow/FDG mismatch thought to indicate the presence of ischemic myocardium have been described both at rest and during exercise. Other studies, however, have demonstrated selective abnormalities of glucose metabolism, independent of coronary flow59 and, more recently, studies have suggested that heterogeneous FDG uptake may relate to regional systolic function and age. Radionuclide angiography has been used to investigate global and regional left ventricular function in HCM, and has shown prolonged isovolumic relaxation, delayed peak filling, reduced relative volume during the rapid filling period, and increased atrial contribution to filling and regional heterogeneity in the timing, rate, and degree of left ventricular relaxation and diastolic filling.60,61 A reduced peak filling rate has been shown to be associated with an increased diseaserelated mortality,61 but its predictive value is not high and adds little to conventional risk stratification.

Hypertension Left ventricular hypertrophy occurs in up to 50% of hypertensive patients. The hypertrophic response is determined by a number of factors including the degree of hypertension, sex, and race.65 In general, patients with HCM tend to have more severe hypertrophy than hypertensives, and the presence of a maximal wall thickness of more than 2 cm in a Caucasian patient should always raise the suspicion of HCM (Table 48.2).66,67 Concentric hypertrophy is more frequent in patients with hypertension, and asymmetric septal hypertrophy more so in HCM, but the specificity of each pattern is not high. In contrast, isolated distal ventricular hypertrophy does seem to be highly predictive of HCM. Systolic anterior motion of the mitral valve occurs in both diseases, but the combination of complete SAM with a substantial left ventricular outflow gradient and asymmetric septal hypertrophy is more indicative of HCM. A number of other echoderived parameters such as left ventricular cross-sectional area and direction-dependent contraction have been suggested as discriminants, but these require further study.68

Table 48.2 Relation of the pattern of left ventricular hypertrophy to underlying etiology

Differential diagnosis In adults, unexplained left ventricular hypertrophy exceeding two standard deviations from the normal (typically, 1·5 cm) is usually sufficient to make a diagnosis of HCM. In children and adolescents the diagnosis can be more difficult as young “gene carriers” may not manifest the complete phenotype. A number of rare genetically determined disorders can present with a cardiac phenotype similar to HCM, but most are distinguished by the presence of other clinical features. Rare exceptions include patients with Friedreich’s ataxia that present with cardiac disease before the onset of obvious neurological deficit,62 Noonan syndrome patients with only very mild somatic abnormalities,63 and patients with primary mitochondrial disease that do not have clinical evidence for neuromuscular disease (unpublished data). Recently mutations in the gene encoding the 2 subunit of AMP-activated protein kinase (7q36) have been described in two families with left ventricular hypertrophy with Wolff–Parkinson–White syndrome. When activated, this gene functions to protect the cell from critical depletion of ATP by activating glycolysis and fatty acid uptake during hypoxic stress or extreme metabolic demand.64 In routine clinical practice the two most commonly encountered areas of difficulty are the differentiation of HCM from “secondary” left ventricular hypertrophy as seen in hypertension and the “athlete’s heart”, and the more recently identified problem of incomplete penetrance in adults.

Sensitivity Specificity Predictive value of positive test

ASHa (%)

Distal (%)

56 (83)b 81 (56) 83 (70)

10

81

100

66

100

58

Symmetrical Wall (%) thickness 2.0 cm (%) 40 (40) 93 (93) 81 (83)

a

Defined by an interventricular septum to posterior wall thickness ratio of 1·5:1. b Values in parentheses from Keller et al. Sensitivity, specificity and predictive value of asymmetric hypertrophy (ASH and distal) in diagnosing hypertrophic cardiomyopathy and symmetrical hypertrophy in diagnosing secondary hypertrophy. The same parameters are shown for a maximal wall thickness or septal thickness of 2·0 cm in diagnosing HCM in patients with symmetric hypertrophy. (Taken from Shapiro et al 54 and Keller et al.56)

Athlete’s heart While HCM is the commonest cause of unexpected sudden death in young athletes,69,70 cardiovascular adaptation to regular training can make differentiation of the “athlete’s heart” from HCM problematic. The ability to distinguish 707


these two entities is of crucial importance, as continued competitive activity in a young person with HCM may threaten that individual’s life, whereas an incorrect diagnosis of HCM in a normal athlete may unnecessarily deprive them of their livelihood. The presence of symptoms, a family history of HCM and/or premature sudden death should always raise the level of suspicion for HCM. In general, athletic training is associated with only a modest increase in myocardial mass, with 2% of elite athletes having a wall thickness 13 mm.71 A diagnosis of HCM in an elite athlete is very likely when an individual has a left ventricular wall thickness 16 mm in men or 13 mm in women. Other echocardiographic features favoring a diagnosis of HCM include small left ventricular cavity dimensions (athletes tending to have increased left ventricular end-diastolic dimensions), left atrial enlargement, and the presence of a left ventricular outflow gradient.72 Doppler evidence of diastolic impairment is also highly suggestive of HCM. The “athletic” ECG often displays voltage criteria for left ventricular hypertrophy, sinus bradycardia, and sinus arrhythmia, but Q waves, ST segment depression, and/or deep T wave inversion is highly suggestive of HCM. Incremental exercise testing may also be useful in distinguishing patients with HCM, a maximal oxygen consumption 50 ml/kg/min or 20% above the predicted maximal value being highly suggestive of athletic adaptation.55 The type of training may also be relevant to diagnosis as hypertrophy is greatest in specific sports such as rowing and cycling. Isometric activities do not appear to cause a substantial hypertrophic response. Very occasionally a period of detraining over 3–6 months is required to distinguish HCM from the athlete’s heart.

Incomplete penetrance in adults It is increasingly recognized that some adults with sarcomeric protein mutations do not fulfill conventional echocardiographic criteria for HCM. New clinical diagnostic criteria for HCM based on the assumption that the probability of disease in a first-degree relative of a patient with HCM is 50%, have recently been proposed (Box 48.1).75 It is important to realize that they are intended to apply only to unexplained ECG and echocardiographic abnormalities in first-degree adult relatives of individuals with proven HCM, and not to isolated cases of minor echocardiographic and ECG abnormalities. HCM in the elderly “Inappropriate” or idiopathic left ventricular hypertrophy has long been recognized in patients over the age of 65 years.74–77 The pattern of disease in this age group is said to differ from that observed in younger patients with HCM in that symptoms occur late in life, the prognosis for most patients is relatively good, and many have mild hypertension. The echocardiographic features of HCM in the elderly are often the same as in the young, but some morphological differences are described: in comparison to their younger counterparts, patients with “elderly HCM” tend to have relatively mild hypertrophy localized to the anterior interventricular septum; the left ventricular cavity is commonly ovoid or ellipsoid rather than crescentic. Elderly patients with left ventricular outflow tract obstruction tend to have more severe narrowing of the left ventricular outflow tract, anterior displacement of the mitral valve apparatus,

Box 48.1 Proposed diagnostic criteria for hypertrophic cardiomyopathy in first-degree relatives of patients with definite diagnosis of hypertrophic cardiomyopathy MAJOR MINOR ● Echocardiography Left ventricular wall thickness 13 mm in the anterior Left ventricular wall thickness of 12 mm in the anterior septum or posterior wall or 15 mm in the posterior septum or posterior wall, or of 14 mm in the posterior septum or free wall septum or free wall Severe SAM (septal leaflet contact)

Moderate SAM (no leaflet-septal contact) Redundant MV leaflets

●

Electrocardiography LVH repolarization changes (Romhilt & Estes) T wave inversion in leads I and aVL (3 mm) (with QRS-T wave axis difference 300), V3–V6 (3 mm) or II and III and aVF (5 mm) Abnormal Q waves (40 ms or 25% R wave) in at least two leads from II, III, aVF (in the absence of left anterior hemiblock), V1–V4; or I, aVL, V5–V6

Complete BBB or (minor) interventricular conduction defects (in LV leads) Minor repolarization changes in LV leads

Deep S in V2 (25 mm)

Unexplained syncope, chest pain, dyspnea It is proposed that diagnosis of hypertrophic cardiomyopathy in first-degree relatives of patients with the disease would be fulfilled in the presence of one major criterion, or two minor echocardiographic criteria, or one minor echocardiographic plus two minor ECG criteria. (From McKenna WJ et al.73.)

708


100

(145)

90 Cumulative survival rate (%)

restricted anterior excursion of the anterior mitral valve leaflet in systole, and a larger area of contact between the mitral valve leaflet and the septum. Mitral valve calcification is often seen in elderly patients, but it is not associated with a greater degree of left ventricular outflow tract obstruction. The frequency of moderate to severe symptoms is similar in young and elderly patients, but the limited published evidence indicates that the elderly respond well to pharmacologic and surgical therapy and have a relatively good prognosis.76 In spite of recent evidence demonstrating de novo hypertrophy in middle-aged patients with myosin binding protein-C mutations, it remains uncertain whether the majority of patients with this so-called elderly phenotype have a separate disease entity reflecting a polygenic response to hypertrophic stimuli. Hypertension is more frequent in the elderly population, but the failure to demonstrate any difference in left ventricular morphology in hypertensive and non-hypertensive HCM patients (with the possible exception of posterior wall thickness) has led some to suggest that it is not an important factor.77 Other “hypertrophic” stimuli that may be present in older patients include increased angulation and decreased compliance of the aorta.

(95)

80

(46) (21)

70 (16)

60 50

(9) Diagnosis of hypertrophic cardiomyopathy in adult life (n = 184) Diagnosis of hypertrophic cardiomyopathy in childhood (n = 27)

40 0 0

1

2

3

4 5 6 7 Years from diagnosis

8

9

10

Figure 48.2 Cumulative survival from the year of diagnosis in 211 medically treated patients with hypertrophic cardiomyopathy92

100

Val606Met

Risk stratification in HCM 80

Although sudden death in HCM is a relatively uncommon event, the fact that it frequently occurs in young asymptomatic individuals gives it a particular significance to families affected by HCM and to the wider community. Clinical risk stratification in patients with HCM is based on the premise that, if sudden death can be prevented, the natural history of the disease for most patients is relatively benign. The absence of risk factors also facilitates reassurance of low risk individuals. A number of studies have shown that individual sarcomeric protein gene mutations have different prognostic implications (Figure 48.2). For example, most families with troponin-T mutations described to date have a poor prognosis, whereas -myosin heavy chain mutations may have a benign or malignant course. Early investigations of HCMrelated -tropomyosin disease have suggested a favorable prognosis.79 In spite of these data, genetic testing at present has a limited role in risk stratification because the number of families studied is small, and even within families there is marked heterogeneity of disease expression. Clinically a young age at diagnosis is associated with an increased risk of sudden death (Figure 48.3). Other recognized risk markers in this age group include a family history of multiple premature sudden deaths, and recurrent unexplained syncopal episodes.39 More recently abnormal blood pressure responses during exercise have been shown to be

Cumulative survival (%)

Markers of sudden death risk in HCM

60

40 Arg249Gln Arg453Cys 20

Arg403Gln 0 0

20

40 Years of age

60

80

Figure 48.3 Kaplan–Meier survival curves for individuals with HCM and different gene mutations. Two -myosin points are reported to be associated with near normal survival: Val606 l Met (● ●) and Leu908 l Val. The mutations Arg403 l Gln (●), Arg453Cys ( ), and Arg249Gln (■) are associated with a poorer prognosis10

709


associated with a higher mortality in patients less than 40 years of age.30 Abnormal blood pressure responses are seen more frequently in patients with a family history of sudden death and small left ventricular cavity dimensions.31,80 Two recent studies have shown a correlation between severe left ventricular hypertrophy (defined as a maximal wall thickness 30 mm) and prognosis.81,82 However, taken in isolation, left ventricular hypertrophy has a relatively low positive predictive accuracy; furthermore, the majority of patients who die suddenly have wall thickness values 30 mm. Severe left ventricular hypertrophy may be more prognostically important in the young, but further studies are required. Two studies49,50 have shown that NSVT in adults with HCM is associated with an increased risk of sudden death. Its clinical value is however, limited by a modest positive predictive accuracy of 22%, and a low incidence in children. Recently it has been suggested that NSVT is significant only when episodes are repetitive, prolonged and/or associated with symptoms. There are however no data to support this. A number of other non-invasive and invasive electrophysiologic parameters have been evaluated in an attempt to further refine clinical risk stratification. QT and QTc intervals are often prolonged in patients with HCM, but no study has shown a convincing association with the risk of sudden death.83–85 QT dispersion may be a more sensitive marker of the propensity to ventricular arrhythmia but further studies in large well-characterized populations are necessary. Abnormal signal averaged ECGs (SAECGs) are relatively common in patients with HCM and NSVT, the best predictor of NSVT being a reduced voltage (150 V) in the initial portion of the high gain QRS complex (sensitivity 95%, specificity 74%, positive predictive accuracy 64%).86 Unfortunately, abnormal SAECGs are not associated with other clinical risk factors and do not identify patients who go on to develop sustained ventricular arrhythmia or sudden death. Similarly, while, global and specific vagal components of heart rate variability (HRV) are reduced in patients with HCM and NSVT, abnormal HRV is not predictive of sudden catastrophic cardiac events.87 The role of programmed electrical stimulation in patients with HCM remains controversial. The largest series from a single center48,88 reports that programmed ventricular stimulation using up to three premature stimuli in the right and/or left ventricle produces sustained ventricular arrhythmia (that is, lasting for more than 30 seconds or associated with hypotension) in 43% of patients selected on the basis of a history of previous cardiac arrest, syncope, palpitations, or non-sustained ventricular tachycardia on Holter. Inducible sustained ventricular arrhythmia was associated with a history of cardiac arrest or syncope and, in a subsequent study, was associated with a reduced survival. The sensitivity, specificity, and predictive accuracy for predicting subsequent cardiac events were 82%, 68%, and 17% respectively. However, almost three quarters of the patients with 710

sustained ventricular arrhythmias required three premature stimuli for induction. The experience in other cardiac diseases has shown that, whilst “aggressive” protocols using three or more stimuli are highly sensitive, their specificity is low. In addition 76% of patients had polymorphic ventricular tachycardia or fibrillation rather than sustained monomorphic ventricular tachycardia, which is generally thought to be a more sensitive and specific marker of sudden death risk. The interpretation of these published data in HCM and their translation into clinical practice is further complicated by the selection criteria used to select patients as “low-risk” patients were under-represented in the analysis. The general view at present is that programmed stimulation is of limited use in the assessment of risk in HCM. Recently, the putative arrhythmogenic substrate in HCM has been investigated by analyzing changes in individual paced ECG transitions (“fractionation”) recorded at three sites in the right ventricle.89 Compared with controls, patients with a history of ventricular fibrillation have prolongation of the paced ECG at relatively long extrastimulus coupling intervals. Patients with a family history of premature sudden death or NSVT exhibit responses that span the range from “high risk” (ventricular fibrillation) to “low risk” (no adverse prognostic features). Identification of high-risk patients The identification of individuals at high risk of sudden death has been hampered by the inherent difficulties of studying a disease with a low prevalence and event rate. This is further compounded by the low positive predictive accuracy of most suggested risk markers for sudden death. Recent data have suggested that risk may be assessed using a small number of easily determined risk markers, specifically, nonsustained ventricular tachycardia, left ventricular wall thickness (30 mm), abnormal blood pressure response in those under 80 years of age, family history of multiple sudden deaths, and recurrent unexplained syncope. Patients with none of the above risk factors have 1% estimated annual risk of sudden death compared with patients with two or more risk factors who have a 4–6% annual risk of sudden death.90 Grade B Risk stratification remains problematic in those patients with a single risk factor, some of whom are clearly at increased risk of sudden death. Further work to identify which of these individuals would benefit from prophylactic therapy is required. Management of the “high-risk” patient There is now general agreement that low-risk adult patients – that is, those without symptoms or risk factors – can be readily identified, and in most populations represent the majority of individuals with the disease. For patients who are considered to be at high risk of sudden death,


(21)

(21)

100

Cumulative survival rate (%)

(121)

(120) (118)

90

(22)

(20) 80 (19) 0 6

12 18 24 30 Months from ECG monitoring

36

Figure 48.4 Cumulative survival curves for patients with nonsustained ventricular tachycardia treated with either conventional antiarrhythmic drugs (▲) or amiodarone (●), and for patients without non-sustained ventricular tachycardia ( ).91

implantable cardioverter defibrillators (ICDs) are increasingly seen as the preferred therapy. The ICD has been accepted as a superior treatment to antiarrhythmic medication for the prevention of life-threatening ventricular arrhythmias and sudden death in high-risk patients with other cardiovascular diseases. Although randomized data comparing amiodarone to ICD in high-risk patients with hypertrophic cardiomyopathy are not currently available, ICDs have been shown to be effective in terminating lifethreatening tachyarrhythmias. In addition the pacing capabilities of the most recent defibrillators offer protection from cardiac arrest from bradyarrhythmias. Studies have shown that low-dose amiodarone can reduce the incidence of sudden death in adults with NSVT (Figure 48.4),91 and in children considered to be at risk,92 although the finding of appropriate discharges in patients with ICDs already taking amiodarone suggests that the ICD may be superior in preventing sudden death. Grade B Approximately 30% of patients with HCM and a history of cardiac arrest have a further event within 6 years of their first episode. There is general agreement that, in patients with a history of ventricular fibrillation, arrest ICD is the preferred therapy.93–97 In patients with multiple risk factors ICDs are increasingly seen as the preferred therapy. In patients with a single risk factor, management needs to be individualized; amiodarone or ICD may be appropriate in selected individuals. Symptomatic therapy Obstructive hypertrophic cardiomyopathy Most physicians still use blockade as the first-line drug therapy in patients with left ventricular outflow obstruction.

Grade B While some studies have suggested that up to 70% of patients improve with blockers, high doses are frequently required and side effects may be limiting. The beneficial effects of blockers on symptoms (principally dyspnea and chest pain) and exercise tolerance appear to result largely from a decrease in the heart rate with a consequent prolongation of diastole and increased passive ventricular filling and myocardial blood flow. By reducing the inotropic response, blockers may also lessen myocardial oxygen demand and decrease the outflow tract gradient during exercise, when sympathetic tone is increased. Verapamil has favorable effects on symptoms secondary to improved ventricular filling and probable reduction in myocardial ischemia. In patients with a substantial outflow tract gradient or markedly elevated pulmonary pressure (or both), verapamil should be used with caution, however, as the drug’s vasodilatory effect may lead to serious hemodynamic complications. There is no evidence that the administration of blockers and verapamil together is more advantageous than the use of either drug alone, and there is no evidence that either protects patients from sudden death. Disopyramide reduces left ventricular outflow tract gradients and relieves symptoms by virtue of its negative inotropic properties and has been extensively used in some centers for symptomatic therapy in patients with significant outflow obstruction.99,100 Reduction in SAM (systolic anterior motion), left ventricular ejection time and improved exercise capacity and functional status are all described but, in common with other therapies, the initial hemodynamic and clinical benefits may decrease with time. Because disopyramide may shorten the atrioventricular nodal conduction time and thus increase the ventricular rate during paroxysmal atrial fibrillation, supplementary therapy with blockers or verapamil in low doses is advisable. The anticholinergic effects of disopyramide (dry mouth, urinary retention, glaucoma) may limit the drug’s use, particularly in elderly patients. When drug therapy fails or is only partially effective, surgery remains the “gold standard” treatment.101–107 The most frequently performed operation is septal myotomy– myectomy in which a wedge of myocardium is excised from the upper interventricular septum via a transaortic approach. Operative mortality in experienced centers is now 1–2%, but may be higher in less experienced units. Most studies indicate that operative mortality is higher when myectomy is combined with other cardiac operations. The incidence of non-fatal complications such as conduction system disease and ventricular septal defect has declined with modification of surgical practice and the use of intraoperative transesophageal echocardiography. Some data suggest that surgery reduces or abolishes resting gradients in 95% of cases, and that 70% of patients show useful symptomatic and functional improvement. However, at least 10% continue to experience significant symptoms. Mitral valve replacement has been proposed as an alternative to myectomy, its 711


attraction being that it avoids potential complications of ventricular septal defect and complete heart block. In a series of 58 patients, mitral valve replacement resulted in a substantial reduction in left ventricular outflow gradient, improved symptomatic class, and an actuarial survival at 3 years of 86%.108 However, early operative mortality was 9%, and only 68% of patients were free from thromboembolism, anticoagulant-related problems, congestive cardiac failure, and reoperation. Thus mitral valve replacement, whilst successfully treating outflow tract obstruction, is usually advocated only in selected patients. These include patients with severe mitral regurgitation from intrinsic abnormalities of the valve apparatus; patients with mid-cavity obstruction from anomalous insertion of papillary muscle into the anterior mitral leaflet; and patients with only mild septal hypertrophy, which suggests that muscular resection would be associated with a high risk of septal perforation or an inadequate hemodynamic result. Grade B Mitral valvuloplasty has also been combined with myotomy–myectomy in some patients with particularly elongated mitral leaflets. Dual chamber pacing has been proposed as a less invasive alternative to surgery. Several studies have described significant gradient reduction in patients treated with atrioventricular synchronous pacemakers programmed with a short atrioventricular delay to ensure constant capture of the right ventricle.109–113 It was initially thought that pacing reduced the outflow gradient by causing paradoxical movement of the interventricular septum, but it is now realized that many aspects of ventricular activation are altered by right ventricular pacing, and it is likely that reduced or delayed septal thickening, reduced contractility, and altered papillary muscle movement contribute to gradient reduction. In general, outflow gradients can be reduced by approximately 50%, but the translation of this benefit into useful clinical improvement is very variable and unpredictable. Some workers suggest that suboptimal responses to pacing may be caused by short native PR intervals that make it impossible to achieve maximum pre-excitation simultaneously and maintain normal atrial filling of the left ventricle. This can be overcome in some patients by pharmacologically increasing the PR interval with blockers and/or calcium antagonists, but some groups controversially advocate radiofrequency ablation of the atrioventricular node in order to achieve “optimal” AV pacing. Other unresolved issues include the significance of the appreciable placebo effect of pacing demonstrated in at least two randomized studies, the effect of long-term pacing on left ventricular wall thickness, and the role of pacing in the young. Despite the drawbacks, pacing may be an option in a minority of drug refractory patients in whom surgery poses an unacceptable operative risk. Several centers are now examining a novel approach to gradient reduction that uses injection of alcohol into the first septal perforator branch of the left anterior descending artery to produce a “chemical myectomy”.114–116 Published 712

data suggest that procedure-related mortality is less than 1% in experienced centers, but deaths from conduction system damage and inadvertent injection of alcohol into other myocardial segments are recognized. This later problem can be minimized by the use of intracoronary myocardial contrast echocardiography. Preliminary data indicate that significant gradient reduction and improvement in symptoms can be achieved. However, as with dual chamber pacing, the actual mechanism of gradient reduction and symptomatic improvement is likely to be more complex than the creation of a “localized” scar in the interventricular septum. The most frequent complication reported to date is complete heart block, although the incidence varies between the small numbers of centers currently performing the procedure. There has been some concern regarding the short- and long-term consequences of deliberately producing a myocardial infarct such as a possible predisposition to ventricular dysarrhythmias and progressive left ventricular wall thinning. Long-term follow up data is not yet available in patients who have undergone pacing or alcohol septal ablation. Non-obstructive HCM The treatment options in symptomatic patients without an outflow tract gradient are limited. Blockers and calciumchannel antagonists can be used alone or in combination, and, in patients with symptoms suggestive of pulmonary venous congestion, diuretics may also be helpful. In a small number of patients with severe refractory chest pain, a variety of techniques, such as transcutaneous nerve stimulation and cardiac denervation, have been used with variable success. For the minority of patients with HCM, who arrive at an end stage characterized by wall thinning, cavity enlargement, and systolic impairment treatment, should include standard therapeutic agents for heart failure associated with systolic dysfunction, including diuretics, ACE inhibitors, and digitalis. Ultimately these patients may become candidates for heart transplantation. Grade C Management of supraventricular arrhythmia Paroxysmal episodes of supraventricular tachycardias, such as atrial fibrillation or flutter, can lead to rapid clinical deterioration by reducing diastolic filling and cardiac output, usually as a consequence of the high ventricular rate. Conversely chronic supraventricular arrhythmias are often well tolerated if the heart rate is adequately controlled.47 Established atrial fibrillation/flutter should be cardioverted, but when restoration of sinus rhythm is not possible, blockers and verapamil are usually efficacious in controlling the heart rate. Grade B Occasionally radiofrequency ablation of the atrioventricular node and implantation of a pacemaker may be necessary in selected patients to achieve adequate rate control. In patients with recurrent


supraventricular arrhythmias, treatment is indicated only if they are sustained (30 seconds) or associated with symptoms. Specific medical therapy with low dose amiodarone (1000–1400 mg/week), or blockers with class III action (for example, sotalol) is effective in maintaining sinus rhythm, and in controlling the ventricular rate during breakthrough episodes. The role of other drugs such as class 1 agents is uncertain. Atrial fibrillation/flutter in HCM is associated with a significant risk of thromboembolism, and anticoagulation should be considered in all patients when atrial fibrillation/flutter is sustained or recurs frequently.

Conclusion HCM is a disorder of diverse etiology, pathology, and clinical presentation. While recent advances in molecular genetics and clinical characterization have led to greater understanding of the disease and its management, several clinical issues remain unresolved. Nevertheless, the pace of current research suggests that many of these controversies will be resolved over the next decade. Summary 1 ●

●

●

The majority of cases of hypertrophic cardiomyopathy (HCM) are caused by mutations in genes encoding cardiac sarcomeric proteins. Although symptoms of chest pain, dyspnea, palpitation, and syncope are common, many patients are asymptomatic and may present for the first time with sudden death. Recurrent syncope, a family history of premature sudden death, non-sustained ventricular tachycardia during ambulatory ECG, left ventricular wall thickness 30 mm, and abnormal exercise blood pressure responses are associated with an increased risk of sudden death.

Summary 2 ●

●

●

●

Symptomatic patients with left ventricular outflow gradients should be initially treated with blockers or disopyramide. If drug therapy is ineffective, patients should be considered for surgery. Grade B Pacing and alcohol septal myectomy are a viable option for patients with symptomatic left ventricular outflow gradient who are unsuitable or not keen on surgery. Grade B All patients should undergo non-invasive risk stratification using ambulatory electrocardiography and exercise testing. Grade B Low-risk adults can generally be reassured that their prognosis is good. High-risk patients require further assessment and should be considered for amiodarone or ICD therapy. Grade B

References 1.Kimura A, Harada H, Park J-E et al. Mutations in the cardiac troponin I gene associated with hypertrophic cardiomyopathy. Nature Genet 1997;16:379–82. 2.Jarcho JA, McKenna WJ, Pare JA et al. Mapping a gene for familial hypertrophic cardiomyopathy to chromosome 14q1. N Engl J Med 1989;321:1372–8. 3.Thierfelder L, MacRae C, Watkins H et al. A familial hypertrophic cardiomyopathy locus maps to chromosome 15q2. Proc Nat Acad Sci USA 1993;90:6270–4. 4.Watkins H, MacRae C, Thierfelder L et al. A disease locus for familial hypertrophic cardiomyopathy maps to chromosome 1q3. Nature Genet 1993;3:333–7. 5.Bonne G, Carrier L, Bercovici J et al. Cardiac myosin binding protein C gene splice acceptor site mutation is associated with familial hypertrophic cardiomyopathy. Nature Genet 1995; 11:438–40. 6.Poetter K, Jiang H, Hassanzadeh S et al. Mutations in either the essential or regulatory light chains of myosin are associated with a rare myopathy in human heart and skeletal muscle. Nature Genet 1996;13:63–9. 7.Oslon TM, Doan TP, Kishimoto NY, Whitby FG, Ackerman MJ, Fananapazir L. Inherited and de novo mutations in the cardiac actin gene cause hypertrophic cardiomyopathy. J Mol Cell Cardiol 2000;32:1687–94. 8.Satoh M, Takahashi M, Sakamoto T, Hiroe M, Marumo F, Kimura A. Structural analysis of the titin gene in hypertrophic cardiomyopathy: identification of a novel disease gene. Biochem Biophys Res Commun 1999;262:411–17. 9.Watkins H, Thierfelder L, Hwang D, McKenna WJ, Seidman JG, Seidman CE. Sporadic hypertrophic cardiomyopathy due to de novo myosin mutations. J Clin Invest 1992;90:1666–71. 10.Watkins H, Rozenzweig A, Hwang DS et al. Characteristics and prognostic implications of myosin missense mutations in familial hypertrophic cardiomyopathy. N Engl J Med 1992; 326:1108–14. 11.Lechin M, Quinones MA, Omran A et al. Angiotensin converting enzyme genotypes and left ventricular hypertrophy in patients with hypertrophic cardiomyopathy. Circulation 1995;92:1808–12. 12.Rayment I, Holden HM, Sellers JR et al. Structural interpretation of the mutations in the beta-cardiac myosin that have been implicated in familial hypertrophic cardiomyopathy. Proc Nat Acad Sci USA 1995;92:3864–8. 13.Lankford EB, Epstein ND, Fananapazir L, Sweeney HL. Abnormal contractile properties of muscle fibres expressing beta-myosin heavy chain gene mutations in patients with hypertrophic cardiomyopathy. J Clin Invest 1995;95:1409–14. 14.Geisterfer-Lowrance AA, Christe M, Conner DA, Ingwall JS, Schoen FJ, Seidman CE, Seidman JG. A mouse model of familial hypertrophic cardiomyopathy. Science 1996;272: 731–4. 15.Davies MJ, McKenna WJ. Hypertrophic cardiomyopathy: pathology and pathogenesis. Histopathology 1995;26: 493–500. 16.Maron BJ, Wolfson JK, Epstein SE, Roberts WC. Intramural (“small vessel”) coronary artery disease in hypertrophic cardiomyopathy. J Am Coll Cardiol 1986;8:545–57.

713


17.Varnava AM, Elliott PM, Sharma S, McKenna WJ, Davies MJ. Hypertrophic cardiomyopathy: the interrelation of disarray, fibrosis, and small vessel disease. Heart 2000;84: 476–82. 18.Wigle ED, Sasson Z, Henderson MA et al. Hypertrophic cardiomyopathy: the importance of the site and extent of hypertrophy: a review. Prog Cardiovasc Dis 1985;28:1–83. 19.Levine RA, Vlahakes GJ, Lefebvre XP et al. Papillary muscle displacement causes systolic anterior motion of the mitral valve. Experimental validation and insights into the mechanism of subaortic obstruction. Circulation 1995;91:1189–95. 20.Klues HG, Maron BJ, Dollar AL, Roberts WC. Diversity of structural mitral valve alterations in hypertrophic cardiomyopathy. Circulation 1992;85:1651–60. 21.Sugrue DD, McKenna WJ, Dickie S et al. Relation between left ventricular gradient and relative stroke volume ejected in early and late systole in hypertrophic cardiomyopathy. Assessment with radionuclide cineangiography. Br Heart J 1984;52:602–9. 22.Maron BJ, Epstein SE. Clinical significance and therapeutic implications of the left ventricular outflow tract pressure gradient in hypertrophic cardiomyopathy. Am J Cardiol 1986; 58:1093–6. 23.Hanrath P, Mathey DG, Siegert R, Biefield W. Left ventricular and filling patterns in different forms of left ventricular relaxation and filling patterns in different forms of left ventricular hypertrophy. An echocardiographic study. Am J Cardiol 1980; 45:15–23. 24.Maron BJ, Spirito P, Green KJ, Wesley YE, Bonow RO, Arce J. Noninvasive assessment of left ventricular diastolic function by pulsed Doppler echocardiography in patients with hypertrophic cardiomyopathy. J Am Coll Cardiol 1987;10: 733–42. 25.Cannon RO, Rosing DR, Maron BJ et al. Myocardial ischemia in patients with hypertrophic cardiomyopathy: contribution of inadequate vasodilator reserve and elevated left ventricular filling pressures. Circulation 1985;71:234–43. 26.Camici P, Chiriatti G, Lorenzoni R et al. Coronary vasodilatation is impaired in both hypertrophied and non hypertrophied myocardium of patients with hypertrophic cardiomyopathy: A study with Nitrogen-13 ammonia and positron emission tomography. J Am Coll Cardiol 1991;17:879–86. 27.Cannon RO, Dilsizian V, O’Gara P et al. Myocardial metabolic, hemodynamic, and electrocardiographic significance of reversible thallium-201 abnormalities in hypertrophic cardiomyopathy. Circulation 1991;83:1660. 28.Elliott PM, Kaski JC, Prasad K et al. Chest pain during daily life in patients with hypertrophic cardiomyopathy: an ambulatory electrocardiographic study. Eur Heart J 1996;17: 1056–64. 29.O’Gara PT, Bonow RO, Maron BJ et al. Myocardial perfusion abnormalities in patients with hypertrophic cardiomyopathy: assessment with thallium-201 emission computed tomography. Circulation 1987;76:1214–23. 30.Sadoul N, Prasad K, Slade AKB, Elliott PM, McKenna WJ. Abnormal blood pressure response during exercise is an independent marker of sudden death in young patients with hypertrophic cardiomyopathy. Circulation 1997;96:2987–91. 31.Frenneaux MP, Counihan PJ, Caforio A, Chikamori T, McKenna WJ. Abnormal blood pressure response during exer-

714

cise in hypertrophic cardiomyopathy. Circulation 1990; 82:1995–2002. 32.Hada Y, Sakamoto T, Amano K et al. Prevalence of hypertrophy cardiomyopathy in a population of adult Japanese workers as detected by echocardiographic screening. Am J Cardiol 1987;59:183–4. 33.Savage DD, Castelli WP, Abbott RD et al. Hypertrophic cardiomyopathy and its markers in the general population: the great masquerader revisited: the Framingham Study. J Cardiovasc Ultrason 1983;2:41–7. 34.Maron BJ, Gardin JM, Flack JM, Gidding SS, Kurosaki TT, Bild DE. Prevalence of hypertrophic cardiomyopathy in a population of young adults. Echocardiographic analysis of 4111 subjects in the CARDIA study. Coronary Artery Risk Development in (Young) Adults. Circulation 1995;92:785–9. 35.Codd MB, Sugrue DD, Gersh BJ, Melton LJ. Epidemiology of idiopathic dilated and hypertrophic cardiomyopathy: a population based study in Olmsted County, Minnesota, 1975–1984. Circulation 1989;80:564–72. 36.Maron BJ, Peterson EE, Maron MS, Peterson JE. Prevalence of hypertrophic cardiomyopathy in an outpatient population referred for echocardiographic study. Am J Cardiol 1994;73: 577–80. 37.Maron BJ, Mathenge R, Casey SA, Poliac LC, Longe TF. Clinical profile of hypertrophic cardiomyopathy identified de novo in rural communities. J Am Coll Cardiol 1999 May; 33(6):1590–5. 38.Frank S, Braunwald E. Idiopathic hypertrophic subaortic stenosis: clinical analysis of 126 patients with emphasis on the natural history. Circulation 1968;37:759–88. 39.McKenna WJ, Deanfield J, Faruqui A, England D, Oakley C, Goodwin J. Prognosis in hypertrophic cardiomyopathy. Role of age and clinical, electrocardiographic and haemodynamic features. Am J Cardiol 1981;47:532–8. 40.Spirito P, Chiarella F, Carratino L, Zoni-Berisso M, Bellotti P, Vecchio C. Clinical course and prognosis of hypertrophic cardiomyopathy in an outpatient population. N Engl J Med 1989;320:749–55. 41.Cecchi F, Olivotto I, Montereggi A, Santoro G, Dolara A, Maron BJ. Hypertrophic cardiomyopathy in Tuscany: clinical course and outcome in an unselected regional population. J Am Coll Cardiol 1995;26:1529–36. 42.Cannan CR, Reeder GS, Bailey KR, Melton LJ III, Gersh BJ. Natural history of hypertrophic cardiomyopathy. A population based study, 1976 through 1990. Circulation 1995;92: 2488–95. 43.Maron BJ, Tajik AJ, Ruttenberg HD et al. Hypertrophic cardiomyopathy in infants: clinical features and natural history. Circulation 1982;65:7–17. 44.Savage DD, Seides SF, Clark CE et al. Electrocardiographic findings in patients with obstructive and non-obstructive hypertrophic cardiomyopathy. Circulation 1978;58:402–9. 45.Lemery R, Kleinebenne A, Nihoyannopoulos P, Alfonso F, McKenna WJ. Q-waves in hypertrophic cardiomyopathy in relation to the distribution and severity of right and left ventricular hypertrophy. J Am Coll Cardiol 1990;16:368–74. 46.Cosio FG, Moro C, Alonso M, Saenz de la Calzada C, Llovet A. The Q-waves of hypertrophic cardiomyopathy. N Engl J Med 1980;302:96–9.


47.Yamaguchi H, Ishimura T, Nishiyama S et al. Hypertrophic nonobstructive cardiomyopathy with giant negative T-waves (apical hypertrophy): ventriculographic and echocardiographic features in 30 patients. Am J Cardiol 1979;44: 401–12. 48.Fananapazir L, Tracey CM, Leon MB et al. Electrophysiological abnormalities in patients with hypertrophic cardiomyopathy: a consecutive analysis in 155 patients. Circulation 1989;80:1259. 49.McKenna WJ, England D, Doi Y, Deanfield JE, Oakley CM, Goodwin JF. Arrhythmia in hypertrophic cardiomyopathy. 1. Influence on prognosis. Br Heart J 1981;46:168–72. 50.Maron BJ, Savage DD, Wolfson JK, Epstein SE. Prognostic significance of 24 hour ambulatory electrocardiographic monitoring in patients with hypertrophic cardiomyopathy: a prospective study. Am J Cardiol 1981;48:252–7. 51.Alfonso F, Frenneaux MP, McKenna WJ. Clinical sustained uniform ventricular tachycardia in hypertrophic cardiomyopathy: association with left ventricular apical aneurysm. Br Heart J 1989;61:178–81. 52.Robinson K, Frenneaux MP, Stockins B, Karatasakis G, Poloniecki J, McKenna WJ. Atrial fibrillation in hypertrophic cardiomyopathy: a longitudinal study. J Am Coll Cardiol 1990;15:1279–85. 53.Maron BJ, Gottdiener JS, Epstein SE. Patterns and significance of the distribution of left ventricular hypertrophy in hypertrophic cardiomyopathy: a wide angle, two-dimensional echocardiographic study of 125 patients. Am J Cardiol 1981;48:418–28. 54.Shapiro LM, McKenna WJ. Distribution of left ventricular hypertrophy in hypertrophic cardiomyopathy: a twodimensional echocardiographic study. J Am Coll Cardiol 1983;2:437–44. 55.Sharma S, Elliott PM, Whyte G et al. Utility of metabolic exercise testing in distinguishing hypertrophic cardiomyopathy from physiologic left ventricular hypertrophy in athletes. J Am Coll Cardiol 2000;36:864–70. 56.Dilsizian V, Bonow RO, Epstein SE, Fananapazir L. Myocardial ischemia detected by thallium scintigraphy is frequently related to cardiac arrest and syncope in young patients with hypertrophic cardiomyopathy. J Am Coll Cardiol 1993;22:796–804. 57.Nienaber CA, Gambhir SS, Moddy FV et al. Regional myocardial blood flow and glucose utilization in symptomatic patients with hypertrophic cardiomyopathy. Circulation 1993;87:1580–90. 58.Grover-McKay M, Schwaiger M, Krivokapich J, Perloff JK, Phelps ME, Schelbert HR. Regional myocardial blood flow and metabolism at rest in mildly symptomatic patients with hypertrophic cardiomyopathy. J Am Coll Cardiol 1989;13: 317–24. 59.Gould KL. Myocardial metabolism by positron emission tomography in hypertrophic cardiomyopathy. J Am Coll Cardiol 1989;13:325–6. 60.Betocchi S, Bonow RO, Bacharach SL, Rosing DR, Maron BJ, Green MV. Isovolumic relaxation period in hypertrophic cardiomyopathy: assessment by radionuclide angiography. J Am Coll Cardiol 1986;7:74–81. 61.Chikamori T, Dickie S, Poloniecki JD, Myers MJ, Lavender JP, McKenna WJ. Prognostic significance of radionuclide-assessed

diastolic dysfunction in hypertrophic cardiomyopathy. Am J Cardiol 1990;65:478–82. 62.Child JS, Perloff JK, Bach PM, Wolfe AD, Perlman S, Kark RA. Cardiac involvement in Friedreich’s ataxia. A clinical study of 75 patients. J Am Coll Cardiol 1986;7:1370. 63.Burch M, Sharland M, Shinebourne E, Smith G, Patton M, McKenna WJ. Cardiologic abnormalities in Noonan syndrome: phenotypic diagnosis and echocardiographic assessment of 118 patients. J Am Coll Cardiol 1993;22:1189–92. 64.Blaire E, Redwood C, Ashrafian H, Oliveira M, Broxholme J, Kerr B, Salmon A, Ostram-Smith I, Watkins H. Mutations in the gamma (2) subunit of AMP-activated protein kinase cause familial HCM: evidence for the central role of energy compromise in disease pathogenesis. Hum Mol Genet 2001;10:1215–20. 65.Devereux RB. Cardiac involvement in essential hypertension. Prevalence, pathophysiology and prognostic implications. Med Clin N Am 1987;71:813–26. 66.Shapiro LM, Kleinebenne A, McKenna WJ. The distribution of left ventricular hypertrophy in hypertrophic cardiomyopathy: comparison to athletes and hypertensives. Eur Heart J 1985;6:967–74. 67.Keller H, Wanger K, Goepfrich M, Stegaru B, Buss J, Heene DL. Morphological quantification and differentiation of left ventricular hypertrophy in hypertrophic cardiomyopathy and hypertensive heart disease. Eur Heart J 1990;11:65–74. 68.Hattori M, Aoki T, Sekioka K. Differences in direction-dependent shortening of the left ventricular wall in hypertrophic cardiomyopathy and in systemic hypertension. Am J Cardiol 1992;70:1326–32. 69.Maron BJ, Roberts WC, McAllister HA, Rosing DR, Epstein SE. Sudden death in young athletes. Circulation 1980;62: 218–29. 70.Burke AP, Farb A, Virmani R, Goodin J, Smialek JE. Sportsrelated and non-sports related sudden cardiac death in young adults. Am Heart J 1991;121:568–75. 71.Pelliccia A, Maron BJ, Spataro A, Proschan MA, Spirito P. The upper limit of physiologic cardiac hypertrophy in highly trained elite athletes. N Engl J Med 1991;324:295. 72.Maron BJ, Pellicia A, Spirito P. Cardiac disease in young trained athletes. Insights into methods for distinguishing athlete’s heart from structural heart disease, with particular emphasis on hypertrophic cardiomyopathy. Circulation 1995;91:1569. 73.McKenna WJ, Spirito P, Desnos M, Dubourg O, Komajda M. Experience from clinical genetics in hypertrophic cardiomyopathy: proposal for new diagnostic criteria in adult members of affected families. Heart 1997;77:130–2. 74.Lewis JF, Maron BJ. Clinical and morphology expression of hypertrophic cardiomyopathy in patients 65 years of age. Am J Cardiol 1994;73:1105–11. 75.Topol EJ, Traill TA, Fortuin NJ. Hypertensive hypertrophic cardiomyopathy of the elderly. N Engl J Med 1985:312: 277–83. 76.Faye WP, Taliercio CP, Ilstrup DM, Tajik AJ, Gersh BJ. Natural history of hypertrophic cardiomyopathy in the elderly. J Am Coll Cardiol 1990;16:821–6. 77.Karam R, Lever HM, Healy BP. Hypertensive hypertrophic cardiomyopathy or hypertrophic cardiomyopathy with hypertension? A study of 78 patients. J Am Coll Cardiol 1989;13: 580–4.

715


78.Varnava AM, Elliott PM, Baboonian C, Davison F, Davies MJ, McKenna WJ. Hypertrophic cardiomyopathy: histopathological features of sudden death in cardiac troponin T disease. Circulation 2001;104:1380–4. 79.Coviello DA, Maron BJ, Spirito P et al. Clinical features of hypertrophic cardiomyopathy caused by mutation of a “hotspot” in the alpha-tropomyosin gene. J Am Coll Cardiol 1997;29:635–40. 80.Elliott PM, Gimeno Blanes JR, Mahon NG, Poloniecki JD, McKenna WJ. Relation between severity of LVH and prognosis in patients with hypertrophic cardiomyopathy. Lancet 2001;357:407–8. 81.Spirito P, Bellone P, Harris KM, Bernabo P, Bruzzi P, Maron BJ. Magnitude of left ventricular hypertrophy and risk of sudden death in hypertrophic cardiomyopathy. N Engl J Med 2000;342:1778–85. 82.Counihan PJ, Frenneaux MP, Webb DJ, McKenna WJ. Abnormal vascular responses to supine exercise in hypertrophic cardiomyopathy. Circulation 1991;84:686–96. 83.Dritsas A, Sabarouni E, Gilligan D, Nihoyannopoulos P, Oakley CM. QT-Interval abnormalities in hypertrophic cardiomyopathy. Clin Cardiol 1992;15:739–42. 84.Fei L, Slade AK, Grace AA, Malik M, Camm AJ, McKenna WJ. Ambulatory assessment of the QT interval in patients with hypertrophic cardiomyopathy: risk stratification and effect of low dose amiodarone. Pacing Clin Electrophys 1994;17: 2222–7. 85.Buja G, Miorelli M, Turrini P, Melacini P, Nava A. Comparison of QT dispersion in hypertrophic cardiomyopathy between patients with and without ventricular arrhythmias and sudden death. Am J Cardiol 1993;72:973–6. 86.Kulakowski P, Counihan PJ, Camm AJ, McKenna WJ. The value of time and frequency domain, and spectral temporal mapping analysis of the signal-averaged electrocardiogram in identification of patients with hypertrophic cardiomyopathy at increased risk of sudden death. Eur Heart J 1993;14: 941–50. 87.Counihan PJ, Fei L, Bashir Y, Farrell TG, Haywood GA, McKenna WJ. Assessment of heart rate variability in hypertrophic cardiomyopathy. Association with clinical and prognostic features. Circulation 1993;88:1682–90. 88.Fananapazir L, Chang AC, Epstein SE, McAreavey D. Prognostic determinants in hypertrophic cardiomyopathy: prognostic evaluation of a therapeutic strategy based on clinical, Holter, hemodynamic and electrophysiological findings. Circulation 1992;86:730–40. 89.Saumarez RC, Slade AKB, Grace AA, Sadoul N, Camm AJ, McKenna WJ. The significance of paced electrocardiogram fractionation in hypertrophic cardiomyopathy. A prospective study. Circulation 1995;91:2762–8. 90.Elliott PM, Poloniecki J, Dickie S et al. Sudden death in hypertrophic cardiomyopathy: identification of high-risk patients. J Am Coll Cardiol 2000;36:2212–18. 91.McKenna WJ, Oakley CM, Krikler DM et al. Improved survival with amiodarone in patients with hypertrophic cardiomyopathy and ventricular tachycardia. Br Heart J 1985;53:412–16. 92.McKenna WJ, Deanfield JE. Hypertrophic cardiomyopathy: an important cause of sudden death. Arch Dis Childhood 1984;59:971–5.

716

93.Elliott PM, Sharma S, Varnava A, Poloniecki J, Rowland E, McKenna WJ. Survival after cardiac arrest or sustained ventricular tachycardia in patients with hypertrophic cardiomyopathy. J Am Coll Cardiol 1999;33:1596–601. 94.Maron BJ, Shen WK, Link MS et al. Efficacy of implantable cardioverter-defibrillators for the prevention of sudden death in patients with hypertrophic cardiomyopathy. N Engl J Med 2000;342:365–73. 95.Primo J, Geelen P, Brugada J et al. Hypertrophic cardiomyopathy: role of the implantable cardioverter defibrillator. J Am Coll Cardiol 1998;31:1081–5. 96.Silka MJ, Kron J, Dunnigan A, Dick M. Sudden cardiac death and the use of implantable cardioverter-defibrillator in paediatric patients. Circulation 1993;87:800–7. 97.Kron J, Oliver RP, Norsted S, Silka MJ. The automatic implantable cardioverter defibrillator in young patients. J Am Coll Cardiol 1990;16:896–902. 98.Elliott PM, Sharma S, Poloniecki J, Prasad K, Murd’Ah, McKenna WJ. Amiodarone and sudden death in hypertrophic cardiomyopathy. (Abstract) Circulation 1997;96:I–464. 99.Pollick C. Muscular subaortic stenosis: hemodynamic and clinical improvement after disopyramide. N Engl J Med 1982; 307:997–9. 100.Pollick C. Disopyramide in hypertrophic cardiomyopathy. Hemodynamic assessment after intravenous administration. Am J Cardiol 1988;62:1248–51. 101.Morrow AG, Reitz BA, Epstein SE, Henry WL, Conkle DM, Itscoitz SB, Redwood DR. Operative treatment in hypertrophic subaortic stenosis: techniques, and the results of pre and postoperative assessments in 83 patients. Circulation 1975;52:88–102. 102.Maron BJ, Merrill WH, Freier PA, Kent KM, Epstein SE, Morrow AG. Long-term clinical course and symptomatic status of patients after operation for hypertrophic subaortic stenosis. Circulation 1978;57:1205–13. 103.Williams WG, Wigle ED, Rakowski H, Smallhorn J, LeBlanc J, Trusler GA. Results of surgery for hypertrophic obstructive cardiomyopathy. Circulation 1987;76(Suppl. V): V104–8. 104.Heric B, Lytle BW, Miller DP, Rosenkranz ER, Lever HM, Cosgrove DM. Surgical management of hypertrophic obstructive cardiomyopathy. Early and late results. J Thorac Cardiovas Surg 1995;110:195–208. 105.Robbins RC, Stinson Eb. Long-term results of left ventricular myotomy and myectomy for obstructive hypertrophic cardiomyopathy. J Thorac Cardiovasc Surg 1996;111:586–94. 106.Schulte HD, Bircks WH, Loesse B, Godehardt EAJ, Schwartzkopff B. Prognosis of patients with hypertrophic obstructive cardiomyopathy after transaortic myectomy. Late results up to 25 years. J Thorac Cardiovasc Surg 1993;106: 709–17. 107.McCully RB, Nishimura RA, Tajik J, Schaff HV, Danielson GK. Extent of clinical improvement after surgical treatment of hypertrophic obstructive cardiomyopathy. Circulation 1996; 94:467–71. 108.McIntosh CL, Greenberg GJ, Maron BJ, Leon MB, Cannon RO, Clark RE. Clinical and hemodynamic results after mitral valve replacement in patients with hypertrophic cardiomyopathy. Ann Thorac Surg 1989;47:236–46.


109.Slade AKB, Sadoul N, Shapiro L et al. DDD pacing in hypertrophic cardiomyopathy: a multicenter clinical experience. Heart 1996;75:44–9. 110.Jeanrenaud X, Goy JJ, Kappenberger L. Effects of dualchamber pacing in hypertrophic obstructive cardiomyopathy. Lancet 1992;339:1318–23. 111.Fananapazir L, Epstein ND, Curiel RV, Panza JA, Tripodi D, McAreavey D. Long-term results of dual chamber (DDD) pacing in obstructive hypertrophic cardiomyopathy. Evidence for progressive symptomatic and hemodynamic improvement and reduction of left ventricular hypertrophy. Circulation 1994;90:2731–42. 112.Nishimura RA, Trusty JM, Hayes DL et al. Dual chamber pacing for hypertrophic cardiomyopathy: a randomised

double-blind crossover trial. J Am Coll Cardiol 1997;29: 435–41. 113.Kappenberger L, Linde C, Daubert C et al. (PIC study Group). Pacing in hypertrophic obstructive cardiomyopathy. A randomised crossover study. Eur Heart J 1997;18:1249–56. 114.Sigwart U. Non-surgical myocardial reduction for hypertrophic obstructive cardiomyopathy. Lancet 1995;346:211–14. 115.Gleichman U, Seggewiss H, Faber L, Fassbender D, Schmidt HK, Strick S. [Catheter treatment of hypertrophic cardiomyopathy]. [German]. Deut Medizin Woch 1996;121:679–85. 116.Knight C, Kurbaan AS, Seggewiss H et al. Non-surgical reduction for hypertrophic obstructive cardiomyopathy: outcome in the first series of patients. Circulation 1997;95: 2075–81.

717

49

Other cardiomyopathies José A Marin-Neto, Marcus Vinícius Simões, Benedito Carlos Maciel

Introduction From the vast array of clinical entities that comprise the cardiomyopathies, two conditions were selected for this chapter: Chagas’ heart disease and endomyocardiofibrosis. In neither disease have large randomized controlled trials been conducted to support recommendations for therapeutic options. Knowledge of natural history and pathophysiology is based almost entirely on observational studies, mostly of the case series kind. Most are flawed by heterogeneous criteria for patient selection and investigation. Thus, particularly in Chagas’ heart disease, a large volume of incomplete and biased information has been obtained, so that metaanalysis of available data has been limited to the etiologic treatment in a subgroup of patients. Despite the paucity of evidence-based knowledge, the diseases epitomize quite different unique pathophysiological conditions. In essence, Chagas’ heart disease is a myocarditis of parasitic origin, although the role of the etiologic agent in the chronic phase of the disease is still somewhat controversial. Endomyocardiofibrosis has no defined etiology or pathogenesis and reasonably good animal models exist only for the study of Chagas’ disease. There are many reasons for the lack of solid evidencebased data in both diseases. However, while the apparently low prevalence of endomyocardiofibrosis is an obvious obstacle, the high prevalence of Chagas’ heart disease in many countries has not helped to produce large randomized trials on therapeutic management. Chagas’ heart disease Epidemiology Chagas’ disease is caused by Trypanosoma cruzi infection. Its transmission is mainly vectoral, through the feces of infected bloodsucking insects of the family Reduviidae (subfamily Triatominae). Many case series reports have also documented that the infection can occur by transplacental and oral transmission, blood transfusion, laboratory contamination, and organ transplantation. Although virtually every organic system may be involved and megaesophagus and megacolon can produce florid clinical conditions, it is the cardiac involvement – the object of this chapter – that constitutes the most serious form of the disease. 718

The current prevalence of Chagas’ heart disease is unknown because no recent large scale screening has been carried out even in endemic countries. Besides, the epidemiological information available from the different countries is strikingly varied. This reflects the diversity of public health programs, including the control of vectoral and transfusional transmission. Thus, a survey carried out from 1988 to 1990 in 850 municipalities of Brazil revealed that serological screening for Chagas’ disease was performed in only two thirds of all blood donors.1 Also, a review of serological surveys for Chagas’ infection among blood donors conducted over the 1980s in several countries in the American continent disclosed a highly variable rate of prevalence, from 0% to 63%.2 Neither is case reporting reliable, even in high endemicity areas. Rough estimates by the World Health Organization, based upon limited serological surveys, suggest that 15–18 million people are infected in extensive areas of the South American subcontinent.3 Moreover, some 65 million are at risk. Cross-sectional epidemiological studies have been carried out in scattered areas of Brazil and Venezuela to assess the prevalence of clinical manifestations and mortality due to Chagas’ heart disease. However, probably because of marked variations in the genetic background, parasite strain, climate, socioeconomic and related hygienic alimentary conditions, and healthcare measures, the prevalence of both morbidity and mortality is extremely variable even within each country.1 Nevertheless, Chagas’ heart disease is by far the most common form of cardiomyopathy in Latin American countries. Further, because of modern migratory trends, it is likely to become ubiquitous. This tendency is shown in the United States, where based on a prevalence of 45% of T. cruzi infection detected serologically in 205 Latin American immigrants and on rough estimates of the number of such legal and illegal immigrants, 400 000–500 000 infected people are believed to be living there now.4 Also, rural–urban migration from endemic areas in Brazil is believed to have brought half a million infected people to cities such as São Paulo and Rio de Janeiro in the past three decades. Chagas’ heart disease has a very high social impact. It has been estimated that over 750 000 years of productive life are lost annually due to premature deaths in Latin American countries, at a cost of about US$1200 million/year.5 These

Other cardiomyopathies

figures substantiate the need for the elimination of transmission – a goal achieved in some regions7 and proved to be a highly cost effective public health policy. Natural history and prognostic factors There is sound experimental, pathological, and clinical evidence that Chagas’ heart disease presents two phases, acute (immediately following infection) and chronic. The long period – 10–30 years – between the acute condition and the clinically manifest chronic Chagas’ heart disease is known as the indeterminate form of the disease and constitutes one of its most intriguing features. Its classical definition – now the subject of controversy6 – requires that the patients be asymptomatic, have no physical signs of Chagas’ disease, and normal ECG and radiological exploration of the chest, esophagus and colon. Megaesophagus and megacolon are also frequently diagnosed in chagasic patients in Brazil, Argentina, and Chile, but not in Mexico, Colombia or Venezuela. The hypothesis that different T. cruzi strains or environmental factors may cause this difference in morbidity8 has not been evaluated by appropriately designed studies. The natural history of Chagas’ heart disease is relatively well known from observational studies conducted mainly in endemic areas in Brazil, Argentina, and Venezuela since the early 1940s. There is also a wealth of case series reports dealing with acute Chagas’ disease acquired through nonvector transmission. Most of these investigations consist of cross-sectional observations of infected people in rural areas. Very few studies have been conducted using case–control populations of chagasic and non-chagasic people. There have also been some observational investigations focusing on the description and follow up of hospital based cohorts of chagasic patients. Both the rural and the hospital based studies have limitations. There is usually no adequate identification of cardiac involvement provided in the rural based studies. Furthermore, because of the protracted course of heart involvement, from the acute phase to end stage heart failure, no prospective studies encompassing the whole span of the disease are available. Conversely, in hospital based studies the heart disease is often well characterized, but results could not be extended to the whole chagasic population. Prognosis in the acute phase Case series using serological tests in endemic areas have shown that in no more than 10% of the acute cases were clinical manifestations sufficient to make a correct diagnosis.9 This is a major deterrent for understanding the transition from the acute to the chronic stages of Chagas’ disease. However, the scarce clinical data are in general agreement with findings from experimental models of Chagas’ disease.

For the minority of patients in whom the clinical diagnosis was possible, cardiac involvement occurred in around 90% of 313 successive cases; in 70–80%, cardiac enlargement was seen on x rays, contrasting with only 50% of cases showing ECG abnormalities. The severity of myocarditis was inversely proportional to age; signs of heart failure were twice as intense in children aged up to 2 years than in those aged between 3 and 5 years.9 Mortality in the acute phase, as seen in the 313 cases, was 83%. This was higher than the 3–5% reported in similar studies in other endemic areas in Brazil, Argentina, and Uruguay. The ECG was normal in 633% of the non-fatal cases and in only 143% of those who died in the acute phase. Seventy-five per cent of all deaths were seen in children aged under 3 years. Heart failure was the constant finding in all fatal cases, with or without concomitant encephalitis.9 Survival is characterized by disappearance of symptoms and signs of heart failure within 1–3 months and normalization of the ECG in over 90% of the cases after 1 year of the infection. However, there is no evidence of spontaneous cure of the infection, as demonstrated by serial xenodiagnosis and serological tests in studies of several hundreds of chagasic patients. Of 172 patients who were followed in Bambuí (central Brazil) for up to 40 years after the acute infection, the development of cardiac involvement (based on clinical signs, ECG, and chest x ray changes) occurred in 338%, 393%, and 581% for follow up periods of 10–20 years, 21–30 years, and 31–40 years respectively.9 In another review from the same area, for 268 patients whose acute phase of the disease had been diagnosed an average of 27 years before, the general mortality for the period was 138%.9 Prognosis in the indeterminate phase Factors that affect the varying rates of disease progression are currently unknown. A 1–3% per year rate of appearance of heart involvement has been observed in several studies. Of 400 young adults followed for 10 years, 91 (23%) showed clinical, ECG or chest x ray markers of cardiac disease. Of note, eight deaths were recorded in that period, of which only one could be ascribed to reagudization (that is, a full-blown infective illness with parasitemia) of Chagas’ heart disease.10 Another longitudinal study in Bambui, central Brazil, contrasted the evolution of 885 young chagasic patients in the indeterminate phase for 10 years with that of 911 chagasic patients with initially abnormal ECGs in the same period. Survival after 10 years was 974% and 613% respectively for the indeterminate group and the group with cardiac involvement.11 A third longitudinal study in a rural Venezuelan community, with 47% prevalence of positive serology for Chagas’ disease, followed 364 patients for a mean period of 4 years. 719


It revealed the appearance of heart disease at a rate of 11% per year in seropositive individuals. Mortality was 3% in the 4 years of follow up and Chagas’ heart disease was the cause of death in 69% of all fatal cases.12 In 1973 a longitudinal study was initiated in a rural community in northeast Brazil. In the initial cross-sectional study, of 644 individuals aged 10 years, 537% were seropositive. The population initially described in 1973–1974 was reexamined in 1977, 1980, and 1983. The overall rate of development of abnormal ECG was 257% in seropositive (248) as compared to 125% per year in seronegative (332) individuals, a relative risk of 2 for the same geographical area. The age adjusted mortality rate was higher in seropositive (89/1000/year of 488 patients) than in seronegative individuals (78/1000/year of 509 individuals). Mortality in this study was strongly associated with ventricular conduction defects and arrhythmias.13 These results were obtained in chagasic populations with more than 50% of the patients younger than 20 years. It is relevant that fewer indeterminate cases are found in the older age groups because of the evolutive nature of the disease (that is, more aging patients presenting clinical signs of cardiac or digestive involvement). Key points ● ●

●

As long as the patients remain in the indeterminate phase, their prognosis is good. Grade B2 After 10 years almost 80% of patients remain in the indeterminate phase of the disease and probably 50% of the entire population will have no signs of heart disease throughout their lives. Grade B2 There are no clues as to why some chagasic patients remain in the indeterminate phase, while others sooner or later go through the chronic phase of heart involvement.

Prognosis of chronic Chagas’ heart disease The evidence provided by the studies mentioned above shows that the mere appearance of ECG changes entails a bad prognosis. Also, a retrospective analysis of seropositive individuals followed over 18 years revealed that right bundle branch block was three times more common in fatal cases than in survivors.14 Another important negative prognostic factor once heart disease is manifest is male gender. This is borne out by several studies carried out with long-term follow up of different cohorts of chagasic patients.15 Only two case–control follow up studies have been reported in Brazilian endemic areas. In central Brazil16 two cross-sectional clinical assessments (1974 and 1984) included 12-lead ECG and radiological evaluation of heart size. Seropositive patients and controls were matched by age and gender. In the first cross-sectional study 264 pairs of 720

subjects were evaluated, of which 110 were re-examined after the 10 year follow up period with the same clinical, ECG, and chest x ray assessment. The incidence of heart disease in previously healthy but serologically positive individuals was 383% in the 10 year period. In those patients with previous heart involvement, a rate of 345% of deterioration was observed in the same period. In the chagasic population the overall 10 year mortality was 23%, compared to 103% in the controls. Moreover, cardiac mortality, including sudden death and death in heart failure, was 17% among chagasic patients and only 23% in the control population. Again, the overall mortality was much higher in chagasic males and largely predominated in the group aged 30–59 years.13 The same group of investigators, applying similar methods in northeastern Brazil, showed that mortality rates were 16% and 0% for 125 matched pairs of respectively chagasic and non-chagasic patients followed for 45 years.17 Progression of disease as assessed by ECG changes occurred in 104% of patients, as compared to 48% of controls. The hypothesis that the different morbidity and mortality rates in the two regions were due to differences in the pathogenicity characteristics of T. cruzi strains was not substantiated by direct evidence. There is persuasive evidence to support the concept that the mortality associated with Chagas’ disease strongly correlates with severity of the myocardial dysfunction. For example, survival 2 years after the first episode of heart failure was only 334% in 160 cases.18 Of note, 10% of deaths were sudden. In addition, 98% of the deceased people were autopsied, revealing 20% prevalence of amastigote forms of T. cruzi in the cardiac tissue, but with a clear predominance of this finding in male patients.18 In a study of 107 chagasic patients followed for 10 years, a significant reduction in life expectancy, as compared to that of 22 non-chagasic patients, was detected only in those with ECG or clinical changes. A mortality rate of 82% over the 10 year follow up period was seen in the group of 34 patients with signs of heart failure at the beginning of the study. In contrast, a 65% 10 year survival was associated with ECG abnormalities but in the absence of signs of heart failure.19 Another study of 104 male chagasic patients admitted to hospital with heart failure revealed a mortality rate of 52% after 5 years. The strongest predictors of survival were reduced LV ejection fraction and maximal oxygen uptake during exercise.20 In a series of 42 patients with Chagas’ heart disease in the United States 11 deaths occurred during a mean follow up of nearly 5 years, always in association with global or regional LV dysfunction. Established or developing heart failure was a strong predictor of mortality but aborted sudden death or the presence of sustained ventricular tachycardia were not predictors for mortality in this series.21 These results conflict with the evidence from 44 chagasic patients


followed for a mean period of 2 years that ventricular tachycardia detected during exercise testing is a marker of increased risk of sudden death.22 This discrepancy probably reflects the limitations of small numbers and relatively short follow up in both studies. Key points ●

● ●

There is substantial evidence that the most important prognostic factor in established Chagas’ heart disease is the degree of myocardial dysfunction. However, ECG changes also herald increased risk. Grade B2 Once overt cardiac failure ensues the prognosis is poor and approaches 50% in 4 years. Grade B2 It is possible – but not proven by good evidence – that ventricular dysrhythmia and sudden death play a more prominent role in mortality due to Chagas’ disease than in heart failure due to other etiologies. Grade B4

Clinical features of Chagas’ heart disease Cardiac abnormalities are present in all stages of Chagas’ disease, but their clinical expression is highly variable. The paucity of clinical indicators of the typical myocarditis of acute Chagas’ disease has already been pointed out. There is also solid evidence – from necropsy studies as well as from in vivo investigations – that virtually all patients, even in the indeterminate phase of the disease, have some subtle degree of cardiac structural or functional involvement.23–29 Patients with Chagas’ heart disease are classified following the criteria shown in Table 49.1. The anatomical and functional disturbances detected during life are consistent with the autopsy findings reported on several series of chagasic patients who died in the various stages of the disease.23,24,30 It is not uncommon for patients with ECG and marked LV regional abnormalities to be asymptomatic hard workers and capable of performing as such under laboratory conditions.26 When symptoms occur, they are usually in the form of fatigue and exertional dyspnea, palpitations, dizziness and syncope or chest pain. These are the expression of a reduction of the cardiac reserve (including minor early signs of diastolic dysfunction), the presence of ventricular dysrhythmias, and atrioventricular block. The chest pain is usually atypical for myocardial ischemia but in sporadic cases may mimic an acute coronary syndrome. Systemic and pulmonary embolism, arising from mural thrombi in the cardiac chambers, is a conspicuous complication of Chagas’ heart disease, but post-mortem findings show they are often overlooked. In 1345 necropsies on patients with Chagas’ heart disease, 595 cases (44%) had cardiac thrombi and/or visceral thromboembolism. The presence of cardiac thrombi was related to severity of ventricular enlargement. Embolism most frequently involved

lungs (36%), kidneys (36%), spleen (14%), and brain (10%).31 Congestive heart failure is more commonly expressed by prominent signs of systemic congestion, with less intense pulmonary congestion.32 This peculiar feature of Chagas’ heart disease is linked to early severe damage of the RV, a chamber frequently neglected in investigations of cardiac function.33,34 Sudden unexpected death occurs with undefined but not negligible frequency even in patients previously asymptomatic. It is usually precipitated by physical exercise and associated with ventricular tachycardia and fibrillation or, more rarely, with complete AV block. Limited evidence from autopsy studies in these patients indicates variable degrees of inflammatory and neuronal cardiac alterations.24 Patients with chronic Chagas’ heart disease invariably have one or more positive serological tests. There is also recent and limited experience with polymerase chain reaction-based methods for detecting T. cruzi DNA sequences in the blood of chagasic patients. This method is likely to replace the cumbersome and unreliable method of direct demonstration of parasite infection by xenodiagnosis.35 ECG abnormalities are present in most patients with chronic Chagas’ heart disease, mainly in the form of conduction disturbances and ventricular arrhythmias. In more advanced stages pathological Q waves are found, compatible with extensive areas of myocardial fibrosis. The combination of right bundle branch block and left anterior hemiblock is very typical in chronic Chagas’ heart disease. Nevertheless, no ECG changes can be considered specific to the disease. Many case series reports have documented the typical feature of striking segmental-wall motion abnormalities in several hundreds of chronic chagasic patients. The most characteristic lesion is the apical aneurysm, but it is the posterior basal dysynergy that best correlates with the occurrence of malignant ventricular arrhythmia. A few small retrospective studies have evaluated the correlation between ventricular arrhythmia and symptoms in Chagas’ heart disease. It is apparent that complex ventricular dysrhythmia may occur in asymptomatic patients, but it is usually a conspicuous manifestation associated with poor LV function. The aneurysms are also sources of emboli. In spite of chest pain being a common complaint in many chagasic patients, coronary angiography is usually normal. However, functional abnormalities in the myocardial blood flow regulation have been described and all types of myocardial perfusion defects have been detected in several small groups of selected patients, possibly implying microvascular coronary disturbances.28 Cardiac autonomic dysfunction, mainly parasympathetic, has been shown in various groups of several hundreds of chagasic patients (including those with isolated digestive disease) whose response to various autonomic tests was 721


Table 49.1 A clinical classification of Chagas’ heart disease Clinical phase Acute

Indeterminate IA

IB

Overt heart disease

Heart failure

Symptoms

Fairly common

Absent

Absent

Minimal

Present

Physical

Usually Abnormal

Normal

Normal

May be abnormal

Abnormal

ECG changes

Common

Absent

Absent

RBBB, LAHB, AVB, PVCs

Q waves VT

Heart size (x rays)

Usually abnormal

Normal

Normal

Normal

Enlarged

RV function

Usually abnormal

Normal

May be depressed

Usually abnormal

Abnormal

LV diastolic function

?

?

Mild impairment

Abnormal

Abnormal

LV systolic function

Abnormal

Normal

Mild segmental dysynergy

Segmental dysynergy

Global depression

Perfusion defects

?

Mild abnormalities

Mild abnormalities

Common

Common

Autonomic function

?

May be abnormal

May be abnormal

May be abnormal

Usually abnormal

RV biopsy

Abnormal

May be abnormal

Usually abnormal

Abnormal

Abnormal

Exercise stress test

?

Normal

May be abnormal – Arrhythmia – Chronotropic deficit

May be abnormal – Arrhythmia – Chronotropic deficit

Abnormal Reduced exercise capacity

Arrhythmia/Sudden death

?

Absent

Very uncommon

May be detected

Common

Abbreviations: AVB, atrioventricular block; ECG, electrocardiogram; LAHB, left anterior hemiblock; LV, left ventricle; PVCs, premature ventricular complexes; RBBB, right bundle branch block; RV, right ventricle; VT, ventricular tachycardia; ?, unknown

compared to that of control subjects.28,34,36,37 However, in Chagas’ disease patients, the autonomic abnormalities do not correlate with any symptoms or cause postural hypotension. Recent scintigraphic studies demonstrated early sympathetic denervation, topographically related to the segmental-wall motion and perfusional abnormalities often detected in patients with more advanced stages of disease.38 Pathophysiology and pathogenetic mechanisms The clinical manifestations arising during the acute phase are closely related to parasite presence in target organs such as the gastrointestinal tract, central nervous system and 722

heart, coexisting with high grade parasitemia. Lymphadenopathy, liver and spleen enlargement are markers of widespread immunologic reaction. As the parasitemia abates and the systemic inflammatory reaction subsides, it is believed that a silent, relentless, focal myocarditis ensues during the indeterminate phase. In predisposed hosts, encompassing approximately 30–50% of the infected population, this chronic myocarditis may evolve to cumulative destruction of cardiac fibers and marked reparative fibrosis. During this phase ventricular arrhythmias and sudden death may rarely occur as manifestations of the underlying focal inflammatory process, and possibly, of the early autonomic denervation. The incessant myocarditis


is eventually responsible for myocardial mass loss attaining critical degrees, thereby leading to regional and global ventricular remodeling, chamber dilation and setting the anatomic substrate for malignant dysrhythmia. This hypothesis is based on experimental models for Chagas’ heart disease. Additional evidence has been provided by studies correlating clinical and pathological findings in autopsied humans dying in all phases of the disease. Most studies included case series of dozens of patients for the acute and indeterminate phases and ranging from hundreds to thousands of cases for the chronic phase. The most intriguing challenge for understanding the pathophysiology of Chagas’ heart disease lies in the complex host-parasite relationship.39,40 It is not clear why in many patients (who remain in the indeterminate phase) the myocardial aggression is controlled at low levels, whereas in others the development of full blown chronic Chagas’ cardiomyopathy is triggered. In brief, auto-immune mediated myocardial injury is probably sustained and exacerbated by continuous antigen presentation related to low grade persistent T. cruzi tissue parasitism. Evidence gathered from pathophysiological studies in animal models and in humans is consistent with the hypothesis that four main pathogenetic mechanisms participate in the genesis of chronic Chagas’ heart disease: ● ● ● ●

neurogenic mechanisms parasite-dependent inflammatory aggression microvascular disturbances immune mediated cardiac damage.

Neurogenic mechanisms – As discussed above, abnormal autonomic cardiac regulation, preceding the development of myocardial damage, has been conclusively shown in many functional investigations.28,34,36,37,38,41–46 Accordingly, intense neuronal depopulation has been clearly demonstrated in several pathologic studies.30,47 Antibodies against autonomic receptors may be detected in experimental and human chronic Chagas’ disease. Functional abnormalities in the cardiac electrogenesis can be caused by auto-antibodies against beta1-adrenergic and muscarinic M2-receptors.42–46 It is still speculative whether receptor stimulation or inhibition thus triggered could mediate myocardial damage. However, various kinds of evidence militate against neurogenic derangements being a main pathogenetic mechanism. Even in endemic areas cardiac denervation shows marked individual variability in intensity and frequency.37,48 Also, no correlation has been shown between cardiac parasympathetic denervation and the extent of myocardial dysfunction or the presence of dysrhythmia. Moreover, the typical chagasic cardiomyopathy is found in geographical regions where the disease is apparently caused by parasite strains devoid of neurotropism. Interestingly, in such

regions, the typical chagasic digestive syndromes – considered to be causally related to parasympathetic denervation of the esophagus and colon – are rarely described.37 Furthermore, no follow up studies correlating autonomic regulation, myocardial function, and cardiac rhythm assessment have been reported. In conclusion, the role of dysautonomia remains to be determined. Furthermore, the attractive hypothesis implicating autonomic impairment in triggering sudden death has never been appropriately tested. Parasite-dependent inflammatory aggression – A direct cause-effect relationship between parasitism and inflammatory findings in the chronic phase of Chagas’ heart disease was initially discarded.47 Very low-grade parasitemia was detected by xenodiagnosis. Also, the scanty tissular nests of amastigotic T. cruzi bear no topographic relation with the diffuse focal inflammation, as seen by classical histopathological staining techniques. However, more sensitive molecular biology methods have shown that parasitemia may be persistent in the chronic phase of Chagas’ disease.49,50 In myocardial biopsy specimens from chronic chagasic patients techniques using amplification of DNA sequences of T. cruzi 51 and immunofluorescent monoclonal antibodies52,53 showed parasite antigens in the inflammatory infiltrates. Microvascular disturbances – Several independent studies in animal models54,55 and in humans with Chagas’ disease,56–59 focusing on histopathological changes, platelet activation and endothelial function disturbances support the hypothesis of microvascular abnormalities. These derangements may cause ischemic-like symptoms and transient ECG changes often detected in chagasic patients. They might also constitute the mechanism responsible for the myocardial perfusion abnormalities described in chagasic patients with angiographically normal coronary arteries.4,28 On the basis of the evidence from these investigations, it has been postulated that microvascular derangements could be a relevant mechanism for the amplification and perpetuation of myocardial damage triggered by the inflammatory process;55 however, there is no information on their prognostic implications. Immune mediated cardiac damage – Studies in humans and in animal experiments provide evidence for the role played by immunological mechanisms in chronic Chagas’ heart disease. It is widely accepted that mononuclear inflammatory infiltrates seen in chronic chagasic myocarditis are the expression of cell mediated aggression. Ultrastructural microscopic studies in animal models show that immune effector cells cause lysis of non-parasitized myocardial 723


cells.60 Depletion of the TDC4 lymphocyte subpopulation prevents myocardial injury in the murine model of Chagas’ heart disease.61 Conversely, myocardial damage is induced in non-chagasic animals, by passive transfer of TDC4 lymphocytes from infected mice.61,62 Furthermore, the identification of antigenic epitopes related to cross-reactivity between T. cruzi and myocardial protein has been reported.63,64 These findings support the notion of an organ-specific autoimmune aggression against self-antigens modified since the acute phase. Also plausible is an incessant aggression maintained by persistent presentation of cross-reacting parasite antigens to the macrophage system, as a consequence of lifelong tissue parasitism. Key points ●

●

●

●

●

The evidence gathered from pathophysiological studies in animal models and in humans is consistent with the hypothesis of immune mediated injury being a key pathogenetic mechanism in chronic Chagas’ heart disease. The immune responses are probably related to the persisting low-grade T. cruzi tissue parasitism but the mechanisms triggering exacerbated responses in some cases and deterring significant immune damage in others are still unknown. The presence of the parasite (or its remnants) in direct topographic relation to the inflammatory foci has potential therapeutic implication. Microvascular disturbances probably constitute important amplification mechanisms for the inflammatory myocardial injury. Cardiac dysautonomia is a well characterized feature that may precede other manifestations of myocardial damage but its pathogenetic role is still debated.

Management of Chagas’ heart disease Etiologic treatment Although recent developments in basic biochemistry of the parasite allowed the identification of potential targets for chemotherapy, such as protein prenylation, sterol, proteases and phospholipid metabolism,65 few antitrypanosomal agents are currently available for clinical use. Nifurtimox and benznidazole have been shown to be comparable in efficacy and high incidence of side effects including dermatitis, polyneuritis, leukopenia, gastrointestinal intolerance. This often warrants discontinuation of treatment, and nifurtimox was abandoned in most centers. Acute phase – There is consensus that in the acute phase etiologic treatment is mandatory to control symptoms and life threatening myocarditis and encephalitis. Guidelines have been developed to recommend etiologic treatment also 724

for laboratory or surgical accidents and in organ transplant recipient and donors.66,67 After adequate treatment a negative xenodiagnosis is found in over 90% of cases and serological tests are negative in 80%. A more recent study suggested that molecular methods can be more effective to show parasite persistence; the etiologic treatment in the acute phase led to PCR negative results in 73% of the cases, while xenodiagnosis was negative in 86% after 3 years.68 Despite these current recommendations, in the absence of long-term follow up trials, there is no evidence for the prevention of chronic organ damage even for patients treated in the acute phase. The impact on prognosis has not been established, again due to lack of appropriately designed follow up studies.

Chronic phase – From experimental studies there is scarce evidence for benefit of trypanocide treatment against the development of chronic tissue damage.69 Also, evidence for potential benefits of etiologic chemotherapy in chronic human Chagas’ disease is extremely limited, due mainly to misleading criteria being employed in small trials. Besides several case-series studies, a prospective, nonrandomized, controlled trial involving 131 patients treated with benznidazole (5 mg/kg/day for 30 days) and 70 untreated patients with a mean follow up period of 8 years has been reported.70 Progression of disease was assessed by ECG changes. Treated patients presented fewer ECG changes than the control group (42% v 30%) and less deterioration in the clinical condition (21% v 17%). These results suggest that etiologic treatment may impact favorably in the chronic phase. However the parasitological criterion of persistent negativity of the xenodiagnosis is unreliable as this test is commonly negative in the chronic phase of Chagas’ disease – 60–80% – despite the presence of overt and progressive cardiac involvement. Moreover, large fluctuations of parasitemia occur over time. There may also be bias in the trials caused by selection of patients with persistent parasitemia in the pretreatment period. Furthermore, results of experimental studies have shown that in the chronic phase the parasitemia is low or not detectable at all, while there is a predominant tissular parasitism by amastigotic forms of T. cruzi. Conversely, because persistently positive serological tests may merely reflect mechanisms of immunological memory or be associated with crossreactivity to altered host antigens, results of any of the serological criteria used to assess etiological treatment are clearly inadequate. In fact, the observed rate of negativation of serological tests following treatment in the chronic phase is consistently low (4–8%) in trials suffering the same methodological limitations discussed above. Thus, until an adequate laboratory method is available for assessment of cure, the only acceptable criteria for etiologic


treatment must be based on the prevention of the appearance of the clinical manifestation or the arrest of damage already detected. For this, a long follow up period of large cohorts of patients would be required. Using better diagnostic tests and research designs, more recent clinical trials have reported high rates of parasitologic cure in children with early chronic T. cruzi infection and claimed trypanocidal therapy for the indeterminate phase.71 A recent systematic review of studies testing the efficacy of trypanocide therapy in the chronic asymptomatic T. cruzi infection has been carried out.72 Only five studies met the inclusion criteria requiring that published trials randomly allocated participants with chronic T. cruzi infection without symptomatic Chagas’ heart disease to one or more of the trypanocidal drugs (benznidazole, nifurtimox, allopurinol) given for at least 30 days at any dose, and to control treatment with or without placebo.71–76 Studies had to report on at least one of the following outcomes: all-cause mortality, sudden death, incidence of heart failure, side effects of treatment, ECG changes (collectively named here as “clinical outcomes”) or reduction in parasite load, reduction in antibody titres to T. cruzi or negative seroconversion (collectively named as “parasite-related outcomes”). General characteristics of the five studies included are shown in Table 49.2. Data synthesis for pooled outcomes including all information available are shown in Figure 49.1.

The most important finding in this review was that trypanocide therapy improved parasite-related outcomes. The strongest effect was found for benznidazole that significantly reduced the proportion of positive xenodiagnosis in both children and adults and increased the rate of negative seroconversion in children, when serology was tested using the ELISA with Antigen-total stimulation (AT ELISA) technique. Allopurinol and itraconazole failed to demonstrate a significant effect on these outcomes and had severe side effects. Although these results are in favor of the use of trypanocide therapy in children and asymptomatic adults for reducing antibodies or the parasite load respectively, whether this effect will result in clinical benefit remains to be proven. None of these trials was designed primarily to assess clinical outcomes and failed to report key methodological issues. In addition, because of the variability in their designs and the small size of each trial, the meta-analysis performed for the pooled outcomes could never include all randomized participants. Thus, all observations on the effects of these agents for chronic asymptomatic T. cruzi infection should be interpreted in the light of the small number of participants in studies not intended to evaluate clinical outcomes. Hence, at present, no experimental evidence is available to support any recommendation on the clinical use of trypanocide therapy for improving clinical outcomes in chronic asymptomatic T. cruzi infection. Large randomized controlled studies encompassing

Table 49.2 General characteristics of included studies Author country (year)

Participants (% IP)

Interventions (n randomized) dose

Outcomesa (Primary and secondary)

Andrade Brazil (1996)

School children (90%)

Benznidazole (64) 75 mg/kg/day – 8 weeks v placebo (65)

Seroconversion Antibodies changes

Apt Chile (1998)

Adults (70%)

Allopurinol (187) 85 mg/kg/day – 8 weeks v itraconazol (217)

Seroconversion n positive xenodiagnosis ECG changes Side effects

6 mg/kg/day – 16 weeks v placebob (165) Coura Brazil (1997)

Adults (NA)

Benznidazole (26) 5 mg/kg/day – 4 weeks v nifurtimox (27) 5 mg/kg/day – 4 weeks v placebo (24)

n positive xenodiagnosis

Gianella Bolivia (1997)

Adults (NA)

Allopurinol (20) 300 mg tid – 8 weeks v placebo (20)

n positive xenodiagnosis

Sosa-Estani Argentina (1998)

School children (95%)

Benznidazole (55) 5 mg/kg/day – 8 weeks v placebo (51)

Seroconversion Antibodies changes

a

As stated by the authors in the report. Participants initially in placebo arm were re-allocated to one of the active arms after two months of treatment. Abbreviations: NA, information not available; IP, indeterminate phase. Reproduced with permission72 b

725


I – Incidence of ECG abnormalities/ BZD – children Study

BZD (n/N)

Plac (n/N)

Weight %

Andrade

1/59

4/58

71

0·28 (0 ·05–1·69)

Sosa-Estani

1/40

1/41

29

1·03 (0·06–16·99)

Total

2/99

5/99

100

0·41 (0·09–1·85)

OR (95% CI)

Heterogeneity test 2 = 0·58 P = 0·45/Overall effect test Z = –1·16 P = 0·2

0·01

0·1

1

10

100

0·01

0·1

1

10

100

Heterogeneity test 2 =27·72 P < 0·001/Overall effect test Z = –6·35 P = 0 < 0·001 0·01

0·1

1

10

100

II – Negative seroconversion/BZD – AT ELISA – children Study

BZD (n/N)

Plac (n/N)

Weight %

OR (95% CI)

Andrade

37/58

3/54

57·8

12·35 (5·72–26·68)

Sosa-Estani

24/44

3/44

42·2

9·19 (3·73–22·64)

Total

61/102

6/98

100

10·91 (6·07–19·58)

Heterogeneity test 2 = 0·24 P = 0·63/Overall effect test Z = 8·0 P = 0 < 0·001 III – Positive xenodiagnosis/all tests/AIl available studies Study

All TT (n/N)

Plac (n/N)

Weight %

OR (95% CI)

Apt

22/336

32·6

1·21 (0·56–2·61)

Coura

10/193

23/67

28·0

0·07 (0·03–0·17)

Gianella

12/33

17/23

17·4

0·23 (0·08–0·66)

Sosa-Estani

2/42

22/43

22·0

0·10 (0·04–0·27)

46/604

71/298

100

0·24 (0·15–0·37)

Total

9/165

IV – Antibody titers mean differences/All available studies (IIF) Study (Weight %) Andrade

Gianella

Sosa-Estani

Total

n TT/n Plac (Weight %)

All TT mean (sd)

Plac mean (sd)

SMD (95% CI)

58/54

–1409

–566

0·68

(48·4)

(1052)

(1400)

(–1·06– –0·30)

–30·1

0·04

(234·7)

(–0·69–0·76)

13/17

–19·7

(13·5)

(317·5)

44/44

–1·40

0·17

–0·66

(38·1)

(2·31)

(2·40)

(–1·09– –0·23)

115/115

–0·58

(100)

(–0·84–0·31)

Heterogeneity test 2 = 3·20 P = 0·20/Overall effect test Z = 4·25 P < 0·001

–1·5

–0·5

0

0·5

1·5

Figure 49.1 Overview of the effect, estimates for data on four outcomes pooled. Estimates are expressed as odds ratios, using the method proposed by Yusuf and Peto (Peto OR), or standardized mean differences (SMD) and its 95% confidence intervals using the fixed models statistical approach (95% CI Fixed). Antibodies mean changes are given in the units originally reported by authors. A negative sign means reduction of levels after being treated. Reproduced from Villar et al 72 with permission.

patients in different stages of disease have to be performed to overcome the current dilemma. Prevention of reagudization Trypanosomicide therapy has been shown in several cases to prevent the parasitological reactivation of Chagas’ disease following corticosteroid therapy.77 It is debatable whether 726

primary chemoprophylaxis would be justifiable in all patients undergoing treatment with immunosuppressant drugs for associated diseases.78 Treatment of congestive heart failure Hemodynamic derangements in chronic chagasic patients with heart failure are comparable to those reported in


dilated cardiomyopathies of other etiologies. Similarly, the classic therapeutic interventions – sodium restriction, diuretics, digitalis, and vasodilation with nitrates and hydralazine – have been successful in relieving congestive symptoms. Small studies have documented short-term hemodynamic beneficial effects of these agents and, to a lesser extent, improvement in exercise tolerance in chronic chagasic patients. However, no studies reported improvement in survival or even in long-term outcome. Small prospective studies on ACE inhibitors have shown promising results in heart failure complicating Chagas’ disease. A multicenter, prospective, non-controlled trial assessed the impact of adding an ACE inhibitor to conventional therapy in 115 patients with heart failure (of whom 20 were chagasics). At the end of 12 weeks, irrespective of etiology, the NYHA functional class was significantly improved in most patients (852%).79 A single-blind, crossover trial of ACE inhibitor and placebo for 6 weeks each, with a washout period of 2 weeks, was reported on 18 NYHA class IV chagasic patients.80 Treatment with the ACE inhibitor was associated with significant reduction in neurohumoral activation and ventricular arrhythmias. These results indicate a potentially beneficial role for this class of drugs in reducing active mechanisms related to sudden death. However, no long-term controlled study has assessed the impact on survival of chagasic patients treated with ACE inhibitors. Other neuro-humoral blocking drugs such as beta adrenergics and spironolactone have not been objectively tested in any clinical trial including a large enough number of chagasic patients to permit efficacy assessment comparative to other etiologies of congestive heart failure. Surgical treatment Heart transplantation – Heart transplantation has been performed in small numbers of patients with refractory heart failure due to Chagas’ disease. However, transplantation is limited by socioeconomic factors in the areas where the disease is endemic and by problems related to the obligatory immunosuppression. Acute myocarditis with marked transitory LV systolic depression has previously been reported in small case series as a frequent complication in patients receiving the usual dose cyclosporin therapy.81 Although the reactivation of acute infection was usually responsive to antiparasite therapy, the possibility of definitive damage to the allograft could not be ruled out and early concern was raised that this could constitute a severe limitation or even contraindication for heart transplantation in Chagas’ disease. Nevertheless, recent data have shown more encouraging results to circumvent this limitation, through the use of reduced immunosuppression regimen. The long-term impact of heart transplantation in chagasic patients has recently been

described in a subgroup of a large cohort of 792 patients submitted to orthotopic heart transplantation in 16 centers in Brazil, The mean overall follow up period was 287305 years, and 117 patients with chronic Chagas’ heart disease constituted the subgroup. The entire cohort population also included 407 patients with idiopathic dilated cardiomyopathy and 196 with ischemic heart disease.82 Among chagasic patients the reported criteria and contraindications for transplantation were similar to those used for non-chagasic patients, except for the detection of megacolon or megaesophagus, also considered a contraindication for transplantation. The survival rate of Chagas’ disease patients at 1 and 12 years was respectively 76%, and 46%. These survival rates were statistically better in comparison with the rest of the cohort group in which the respective survival rates were 72% and 27%. It is worthy of note that reactivation of T. cruzi infection with myocarditis and meningoencephalitis was rarely reported, and was the cause of death, in only 03% of the entire chagasic cohort. Even allowing for the poor control of other relevant characteristics of chagasic and non-chagasic patients in this retrospective analysis of a cohort study, the results suggest that heart transplantation is a valid therapeutic option in end stage Chagas’ heart disease with expected survival rate at least comparable to other patients submitted to this procedure. Dynamic cardiomyoplasty – Reported experience with this procedure in chagasic patients is limited. While initial results in very few patients showed encouraging symptom and LV function improvement,83 a survey of surgical centers in South America (112 patients of whom 96 had heart failure due to dilated cardiomyopathy and 13 due to Chagas’ heart disease) was less optimistic.84 Comparative analysis showed survival rates of 861% and 498% for patients with dilated cardiomyopathy and 40% and 95% for chagasic patients at 1 and 5 years follow up respectively. These results were corroborated by another recent observational study, again including a quite reduced number of chagasic patients.85 There are no clues from these data to elucidate why the prognosis for chagasic patients was worse. Clearly large controlled randomized trials are needed to define any value of dynamic cardiomyoplasty as a temporary approach, before refractory heart failure due to Chagas’ heart disease can be treated by more radical interventions such as cardiac transplantation.86 Partial left ventriculectomy and synchronization therapy – The so-called Batista operation has been performed in small numbers of chagasic patients in many scattered surgical centers in Brazil, without any systematic approach specific for this disease. Because no systematic outcome information is available, and also due to the lack of consistent results with the procedure in other etiologies, currently, partial left 727


ventriculotomy can not be recommended for the treatment of chagasic heart failure. Also, recent small case-series studies reported acute symptomatic and hemodynamic improvement after dual-chamber or multisite pacemaker implantation, but on an entirely empirical basis. Prevention of thromboembolic events There is very limited clinical information on the risk of embolism in patients with mural thrombus or apical aneurysm. In 65 selected patients with apical aneurysm, a follow up study ranging from 19 to 176 months documented 17 episodes of thromboembolism in 14 patients (245%)87 – seven to the brain, nine to the lung and one to the iliac artery. These patients also had congestive heart failure and 11 died in the observation period. In eight of those patients, the cause of death was related to heart failure and in three it was a consequence of cerebral embolism. Another small study in an endemic region of South America addressed the contribution of Chagas’ heart disease in 69 patients having embolic strokes.88 Of 13 patients with non-ischemic dilated cardiomyopathy, Chagas’ heart disease was detected in nine (130%). It was the third most frequently identified cause of embolism after atrial fibrillation (29%) and rheumatic valvular heart disease (203%). However, the real risk of thromboembolism in patients with Chagas’ heart disease is unknown, as no specific studies have addressed this problem. Furthermore, despite the preliminary evidence that thromboembolic events are relevant factors in the natural history of Chagas’ disease, no clinical studies have been conducted to date on adequate treatment and prevention. Current recommendations for anticoagulant therapy are based on information derived from other dilated cardiomyopathies. Thus, chagasic patients presenting global LV dysfunction, atrial fibrillation, previous embolic episodes, and dyskinetic areas with detected mural thrombus are candidates for treatment with intravenous and/or oral anticoagulants. Social and economic factors limit the implementation of this treatment, however, even in chagasic patients with otherwise apparently clear indications for prevention of thromboembolism. Management of rhythm disturbances A wide spectrum of rhythm disturbances is one of the main hallmarks of Chagas’ heart disease. Sinus node dysfunction and other atrial dysrhythmias are common findings and usually present at the early appearance of symptoms. Management of rhythm disturbances does not differ from that recommended for other cardiomyopathies, although there is no sound evidence to support any specific treatment. Complex ventricular dysrhythmia is the most important disturbance because of its implication for sudden death. It is 728

believed that this is more common in chagasic patients than in other dilated cardiomyopathies, but no adequate comparative study has been reported to support this hypothesis. As may be expected, there is reasonable evidence that the more complex and frequent the ventricular dysrhythmia, the worse the ventricular function. However, there is convincing evidence that complex ventricular dysrhythmia may also occur in chagasic patients with preserved global LV function. This is more remarkable when dyskinesia in the posterior basal LV region seems to provide the electrophysiologic substrate for refractory ventricular tachycardia. Although no prospective controlled trial has been conducted, the scarce experience reported suggests that this subgroup may benefit from surgical excision of fibrotic tissue following careful electrophysiologic mapping of LV dyskinetic regions. Equally limited is the reported experience with implantable cardioverter defibrillators in chagasic patients surviving episodes of sudden death.4,89 Except for several small case series reports, very scanty information has been published on the efficacy of pharmacological antiarrhythmic therapy in Chagas’ heart disease. A prospective, double-blind, placebo-controlled, randomized crossover study in a reduced number of patients reported similar effects of disopyramide and amiodarone for controlling ventricular dysrhythmia.90 Another prospective open, parallel, randomized study in 81 chagasic patients with ventricular dysrhythmia compared the efficacy of flecainide and amiodarone.91 The final evaluation by Holter monitoring after 60 days showed a significant and comparable reduction in the frequency of ventricular tachycardia achieved with both flecainide (965%) and amiodarone (926%). However, the follow up was insufficient for conclusions to be drawn on the long-term efficacy or the impact of arrhythmia control on the incidence of sudden death. Two moderately large randomized trials included chagasics among patients treated with amiodarone. The GESICA (Grupo de Estudio de la Sobrevida en la Insuficiencia Cardiaca en Argentina) study concluded after 2 years of follow up that low-dose amiodarone was effective in reducing mortality and hospital admission in patients with severe heart failure, independent of the presence of complex ventricular dysrhythmia.92 Unfortunately, the subgroup of chagasic patients was very modest (48 of 516 patients) and subgroup analysis was not provided. Neither would it have been likely to be useful. An ongoing prospective, multicenter, randomized, controlled study designed to evaluate the impact of amiodarone on survival of treatment of asymptomatic ventricular arrhythmia also included chagasic patients.93 In its pilot phase, this trial enrolled 127 patients (24 with Chagas’ heart disease) with LVEF 35%, presenting frequent ventricular premature complexes and/or repetitive forms of asymptomatic ventricular arrhythmia. The preliminary results after 12 months of follow up showed a significant


reduction in the incidence of sudden death in the amiodarone group (70% v 204%). However, owing to a high dropout rate (16%), follow up data were obtained in only 106 patients. Nevertheless, it is to be hoped that the final results, recruiting a larger number of chagasic patients, will provide some helpful evidence on treating ventricular dysrhythmia with amiodarone in Chagas’ disease. Complete atrioventricular block may also contribute to low cardiac output and cause syncope and sudden death in chagasic patients. In this situation pacemakers are used, as in other cardiac conditions. The evidence on the effects of pacemaker implantation comes from limited case series reports with historic control series of patients in whom this treatment was not possible.89 The common association of atrioventricular disturbances and ventricular complex dysrhythmia in the same patient also requires pacemaker implantation associated with pharmacological antiarrhythmic therapy. This management is regarded as prophylactic, although it is not based on unquestionable evidence. Key points ●

●

●

●

Etiologic treatment is clearly warranted for treatment of acute phase or chronic infection reactivation associated with immune-suppressive states. Grade A/B4 Despite recent evidence supporting the participation of persistent tissue parasitism in the chronic phase of disease, and preliminary persuasive evidence that treatment of chronic asymptomatic patients results in benefit from the parasite outcomes point-of-view, there is no evidence that clinical outcomes are influenced. Grade A/C, 1c and 5 Digitalis, diuretics and neurohumoral blocking drugs are empirically used for treating chagasic patients with heart failure. Grade B2 Heart transplantation can be considered a promising treatment for refractory heart failure in Chagas’ patients, even though this position is based in only one prospective multicenter cohort including small number of patients. Grade B2 Pharmacological, surgical, and device-based strategies for the treatment of ventricular dysrhythmia in chagasic patients are empirical and not supported by any large randomized, controlled trials. Grade B4

Endomyocardial fibrosis (EMF) EMF is a restrictive cardiomyopathy with still unknown etiology occurring most frequently in tropical and subtropical countries. Major endocardial fibrotic involvement of the inflow portion of one or both ventricles, including the subvalvular region, leads to cavity obliteration, restriction of diastolic filling, and clinical manifestations of congestive heart failure and valvular regurgitation. A remarkably

similar cardiac involvement occurring in non-tropical countrieshas been described as endomyocardial disease. This is commonly named Löffler endocarditis or hypereosinophilic syndrome. Although a still disputed issue, it has been postulated that the two conditions represent different stages of the same disease.94 Another controversial hypothesis is that eosinophilderived factors have a toxic role in the pathogenesis of endomyocardial damage. A recent report combines circumstancial evidence for the association of vector-borne etiology and helminth hypereosinophilia as an etiologic hypothesis for endemic EMF in tropical rain forest zones.95 Epidemiology and natural history The low prevalence of EMF inhibits the study of the epidemiology and natural history. Even the larger published series have included only around 100 patients. Symptoms and signs Biventricular involvement has been documented in approximately half of the patients with EMF, while isolated right or left ventricular disease is variably described in 10–40% of cases in different published series. Depending upon predominant involvement of either chamber, symptoms and signs related to pulmonary congestion (left-sided) and systemic congestion (right-sided disease), and to mitral or tricuspid reflux, will tend to be more conspicuous. Constrictive pericarditis is an important differential diagnosis in EMF, especially when the right ventricle is markedly involved.96 Demonstration of ventricular obliteration by imaging techniques is essential for the diagnosis, but endomyocardial biopsy can be decisive in selected patients. The magnitude of symptoms, the grade of ventricular obliteration (especially of the right ventricle), and the occurrence of valvar regurgitation are important prognostic determinants of mortality in this disease. These clinical markers are useful in selecting patients for surgery since a good longterm prognosis has been reported for patients who have mild ventricular dysfunction and no valvular regurgitation.97 Surgical management Extensive surgical excision of the fibrotic tissue, preserving or replacing the atrioventricular valves, can ameliorate symptoms and improve hemodynamics and has been suggested to improve the long-term prognosis.98 An operative mortality ranging from 46% to 250% has been reported in small case series studies. Ten year 70% and 17 year 55% survival rates have been respectively reported in European and Latin American series.99,100 It must be emphasized that no reports based on randomized controlled trials of treatment strategies are available. 729


Key points ● ●

●

The etiology and pathogenesis of EMF are still to be determined. The epidemiology, natural history, and pathophysiology are very incompletely understood, with available data based solely on retrospective evidence from small observational investigations. Grade B4 Promising preliminary results obtained with surgical approaches await validation in large randomized, controlled studies before any general recommendation for improving quality of life and survival rates can be made. Grade B4

References 1.Wanderley DMV, Corrêa FMA. Epidemiology of Chagas’ heart disease. São Paulo Med J 1995; 113: 742–9. 2.Schmunis GA. Trypanosoma cruzi, the etiologic agent of Chagas’ disease: status in the blood supply in endemic and nonendemic countries. Transfusion 1991;31:547–57. 3.WHO. Control of Chagas’ disease. WHO Technical Report Series 811. Geneva: World Health Organization, 1991. 4.Hagar JM, Rahimtoola SH. Chagas’ heart disease. Curr Prob Cardiol 1995;10:825–928. 5.Schofield CJ, Dias JCP. A cost benefit analysis of Chagas’ disease control. Mem Inst Oswaldo Cruz 1991;86:285–95. 6.Acquatella H, Catalioti F, Gomez-Mancebo JR, Davalos V, Villalobos L. Long-term control of Chagas’ disease in Venezuela: effects on serologic findings, electrocardiographic abnormalities, and clinical outcome. Circulation 1987;76: 556–62. 7.Prata A. Clinical and epidemiological aspects of Chagas disease. Lancet Infect Dis 2001;1:92–100. 8.Miles MA, Póvoa MM, Prata A, Cedillos RA, De Souza AA, Macedo V. Do radically dissimilar trypanosoma cruzi strains (Zymodemes) cause Venezuelan and Brazilian forms of Chagas’ Disease? Lancet 1981;20:1338–40. 9.Dias JCP. Cardiopatia chagásica: história natural. In: Cançado JR, Chuster M. eds. Cardiopatia chagásica. Belo Horizonte Fundação: Carlos Chagas de Pesquisa Médica, 1985. 10.Dias JCP. The indeterminate form of human chronic Chagas’ disease. A clinical epidemiological review. Rev Soc Bras Med Trop 1989;22:147–56. 11.Forichon E. Contribution aux estimations de morbidité et de mortalité dans la maladie de Chagas. Toulouse: Universite Paul-Sabatier, 1974. 12.Puigbó JJ, Rhode JRN, Barrios HG, Yépez CG. A 4-year follow up study of a rural community with endemic Chagas’ disease. Bull World Health Organ 1968;39:341–8. 13.Mota EA, Guimarães AC, Santana OO et al. A nine year prospective study of Chagas’ disease in a defined rural population in northeast Brazil. Am J Trop Med 1990;42:429–40. 14.Dias JCP, Kloetzel K. The prognostic value of the electrocardiographic features of chronic Chagas’ disease. Rev Inst Med Trop São Paulo 1968;10:158–62. 15.Lima e Costa MFF, Barreto SM, Guerra HL, Firmo JOA, Uchoa E, Vidigal PG. Ageing with Trypanosoma cruzi

730

infection in a community where the transmission has been interrupted: the Bambuí Health and Ageing Study (BHAS). Int J Epidemiol 2001;30:887–93. 16.Coura JR, Abreu LL, Pereira JB, Willcox HP. Morbidade da doença de chagas. IV. Estudo longitudinal de dez anos em Pains e Iguatama, Minas Gerais. Mem Inst Oswaldo Cruz 1985; 80:73–80. 17.Pereira JB, Cunha RV, Willcox HP, Coura JR. Development of chronic human Chagas’ cardiopathy in the hinterland of Paraìba, Brazil, in a 45 year period. Rev Soc Bras Med Trop 1990;23:141–7. 18.Pugliese C, Lessa I, Santos Filho A. Estudo da sobrevida na miocardite crônica de chagas descompensada. Rev Inst Med Trop São Paulo 1976;18:191–201. 19.Espinosa R, Carrasco HA, Belandria F et al. Life expectancy analysis in patients with Chagas’ disease: prognosis after one decade (1973–1983). Int J Cardiol 1985;8:45–56. 20.Mady C, Cardoso RHA, Barreto ACP et al. Survival and predictors of survival in congestive heart failure due to Chagas’ cardiomyopathy. Circulation 1994;90:3098–102. 21.Hagar JM, Rahimtoola SH. Chagas’ heart disease in the United States. N Engl J Med 1991;325:763–8. 22.de Paola AA, Gomes JA, Terzian AB et al. Ventricular tachycardia on exercise testing is significantly associated with sudden cardiac death in patients with chronic chagasic cardiomyopathy and ventricular arrhythmias. Br Heart J 1995;74:293–5. 23.Laranja FS, Dias E, Nobrega G, Miranda A. Chagas’ disease: a clinical, epidemiologic, and pathologic study. Circulation 1956;14:1035–59. 24.Lopes ER, Chapadeiro E, Almeida HO, Rocha A, Rocha A. Contribuição ao estudo da anatomia patológica dos corações de Chagásicos falecidos subitamente. Rev Soc Bras Med Trop 1975;9:269–82. 25.Marin-Neto JA, Simoes MV, Sarabanda AVL. Chagas’ heart disease. Arq Bras Cardiol 1999;72:264–80. 26.Gallo Jr L, Maciel BC, Marin-Neto JA et al. Control of heart rate during exercise in health and disease. Brazilian J Med Biol Res 1995;28:1179–84. 27.Barreto ACP, Arteaga-Fernandez E. RV endomyocardial biopsy in chronic Chagas’ disease. Am Heart J 1986;111:307–12. 28.Marin-Neto JA, Marzullo P, Marcassa C. Myocardial perfusion defects in chronic Chagas’ disease. Assessment with thallium201 scintigraphy. Am J Cardiol 1992;69:780–4. 29.Barreto ACP, Ianni BM. The undetermined form of Chagas’ heart disease: concept and forensic implications. São Paulo Med J 1995;113:797–801. 30.Oliveira JSM. A natural human model of intrinsic heart nervous system denervation: Chagas’ cardiopathy. Am Heart J 1985;110:1092–8. 31.Oliveira JSM, Araújo RRC, Mucillo G. Cardiac thrombosis and thromboembolism in chronic Chagas’ heart disease. Am J Cardiol 1983;52:147–51. 32.Prata A, Andrade Z, Guimarães AC. Chagas’ heart disease. In: Shaper AG, Hutt MSR, Fejfar Z, eds. Cardiovascular disease in the tropics. London: British Medical Association, 1974. 33.Marin-Neto JA, Marzullo P, Sousa ACS et al. Radionuclide angiographic evidence for early predominant right ventricular


involvement in patients with Chagas’ disease. Can J Cardiol 1988;4:231–6. 34.Marin-Neto JA, Bromberg-Marin G, Pazin-Filho A, Simões MV, Maciel BC. Cardiac autonomic impairment and early myocardial damage involving the right ventricle are independent phenomena in Chagas’ disease. Int J Cardiol 1998;65:261–9. 35.Britto C, Cardoso A, Silveira C, Macedo V, Fernandes O. Polymerase chain reaction (PCR) as a laboratory tool for the evaluation of the parasitological cure in Chagas’ disease after specific treatment. Medicina (B Aires) 1999;59:176–8. 36.Marin-Neto JA, Gallo L Jr, Manço JC, Rassi A, Amorin DS. Mechanisms of tachycardia on standing: studies in normal individuals and in chronic Chagas’ heart patients. Cardiovasc Res 1980;14:541–50. 37.Amorim DS, Marin-Neto JA. Functional alterations of the autonomic nervous system in Chagas’ heart disease. São Paulo Med J 1995;113:772–83. 38.Simões MV, Pintya AO, Marin GB et al. Relation of regional sympathetic denervation and myocardial perfusion disturbances to wall motion impairment in Chagas’ cardiomyopathy. Am J Cardiol 2000;86:975–81. 39.Rassi Jr A, Rassi A, Little WC. Chagas’ heart disease. Clin Cardiol 2000;23:883–9. 40.Higuchi ML. Chronic chagasic cardiopathy: the product of a turbulent host-parasite relationship. Rev Inst Med Trop Sao Paulo 1997;39:53–60. 41.Ribeiro ALP, Moraes RS, Ribeiro JP et al. Parasympathetic dysautonomia precedes left ventricular systolic dysfunction in Chagas’ disease. Am Heart J 2001;141:260–5. 42.Sterin-Borda L, Gorelik G, Postan M, Gonzalez CS, Borda E. Alterations in cardiac beta-adrenergic receptors in chagasic mice and their association with circulating beta-adrenoceptorrelated antibodies. Cardiovasc Res 1999;41:116–25. 43.Wallukat G, Nissen E, Morwinski R, Muller J. Autoantibodies against the beta- and muscarinic receptors in cardiomyopathy. Herz 2000;25:261–6. 44.Chiale PA, Ferrari I, Mahler E et al. Differential profile and biochemical effects of antiautonomic membrane receptor antibodies in ventricular arrhythmias and sinus node dysfunction. Circulation 2001;103:1765–71. 45.Mahler E, Sepulveda P, Jeannequin O et al. A monoclonal antibody against the immunodominant epitope of the ribosomal P2beta protein of Trypanosoma cruzi interacts with the human beta 1-adrenergic receptor. Eur J Immunol 2001;31:2210–16. 46.Costa PCS, Fortes FSA, Machado AB et al. Sera from chronic chagasic patients depress cardiac electrogenesis and conduction. Braz J Med Biol Res 2000;33:439–46. 47.Köberle F. Chagas’ heart disease and Chagas’ syndromes: the pathology of American trypanosomiasis. Adv Parasitol 1968; 6:63–116. 48.Amorim DS, Manço JC, Gallo L Jr, Marin-Neto JA. Chagas’ heart disease as an experimental model for studies of cardiac autonomic function in man. Mayo Clin Proc 1982;57:48–60. 49.Avila HA, Sigman DS, Cohen LM, Millikan RC, Simpson L. Polymerase chain reaction amplification of Trypanosoma cruzi kinetoplast minicircle DNA isolated from whole blood lysates: diagnosis of chronic Chagas’ disease. Mol Biochem Parasitol 1991;48:211–21.

50.Monteon-Padilha V, Hernandez-Becerril N, Ballinas-Verdugo MA, Aranda-Faustro A, Reyes PA. Persistence of Trypanosoma cruzi in chronic chagasic cardiopathy patients. Arch Med Res 2001; 32:39–43. 51.Jones EM, Colley DG, Tostes S et al. A Trypanosoma cruzi DNA sequence amplified from inflammatory lesions in human chagasic cardiomyopathy. Trans Assoc Am Phys 1992; 105:182–9. 52.Bellotti G, Bocchi EA, Moraes AV et al. In vivo detection of Trypanosoma cruzi antigens in hearts of patients with chronic Chagas’ disease. Am Heart J 1996;131:301–7. 53.Anez N, Carrasco H, Parada H et al. Myocardial parasite persistence in chronic chagasic patients. Am J Trop Med Hyg 1999;60:726–32. 54.Morris SA, Tanowitz HB, Wittner M, Bilezikian JP. Pathophysiological insights into the cardiomyopathy of Chagas’ disease. Circulation 1990;83:1900–9. 55.Rossi MA. Microvascular changes as a cause of chronic cardiomyopathy in Chagas’ disease. Am Heart J 1990;120:233–6. 56.Reis DD, Jones EM, Tostes S. Expression of major histocompatibility complex antigens and adhesion molecules in hearts of patients with chronic Chagas’ disease. Am J Trop Med Hyg 1993;49:192–200. 57.Torres FW, Acquatella H, Condado J, Dinsmore R, Palacios I. Coronary vascular reactivity is abnormal in patients with Chagas’ heart disease. Am Heart J 1995;129:995–1001. 58.Simões MV, Ayres-Neto EM, Attab-Santos JL, Maciel BC, MarinNeto JA. Chagas’ heart patients without cardiac enlargement have impaired epicardial coronary vasodilation but no vasotonic angina. J Am Coll Cardiol 1996;27:394–5A. 59.Higuchi ML. Human chronic chagasic cardiopathy: participation of parasite antigens, subsets of lymphocytes, cytokines and microvascular abnormalities. Mem Inst Oswaldo Cruz 1999;94:263–7. 60.Andrade ZA, Andrade SG, Correa R, Sadigursky M, Ferrans VJ. Myocardial changes in acute Trypanosoma cruzi infection. Am J Pathol 1994;144:1403–11. 61.Santos RR, Rossi MA, Laus JL et al. Anti-CD4 abrogates rejection and re-establishes long-term tolerance to syngeneic newborn hearts grafted in mice chronically infected with Trypanosoma cruzi. J Exp Med 1992;175:29–39. 62.Ribeiro-dos-Santos R, Mengel JO, Postol E et al. A heartspecific CD4 T-cell line obtained from chronic chagasic mouse induces carditis in heart-immunized mice and rejection of normal heart transplants in the absence of Trypanosoma cruzi. Parasite Immunol 2001;23:93–101. 63.Cunha-Neto E, Duranti M, Gruber A et al. Autoimmunity in Chagas’ disease cardiopathy: biological relevance of a cardiac myosin-specific epitope crossreactive to an immunodominant Trypanosoma cruzi antigen. Proc Natl Acad Sci USA 1995;92: 3541–5. 64.Giordanengo L, Maldonado C, Rivarola HW et al. Induction of antibodies reactive to cardiac myosin and development of cardiac alteration in cruzipain-immunized mice and their offspring. Eur J Immunol 2000;30:3181–9. 65.Docampo R. Recent developments in the chemotherapy of Chagas’ disease. Curr Pharm Des 2001;7:1157–64. 66.Cançado JR. Etiological treatment of chronic Chagas’ disease. Rev Inst Med Trop Sao Paulo 2001;43:173–81.

731


67.Sosa-Estani S, Segura EL. Treatment of Trypanosoma cruzi infection in the undetermined phase. Experience and current guidelines of treatment in Argentina. Mem Inst Oswaldo Cruz 1999;94:363–5. 68.Solari A, Ortiz S, Soto A et al. Treatment of Trypanosoma cruzi-infected children with nifurtimox: a 3 year follow up by PCR. J Antimicrob Chemother 2001;48:515–19. 69.Andrade SG, Stocker-Guerret S, Pimentel AS, Grimaud JA. Reversibility of cardiac fibrosis in mice chronically infected with Trypanosoma cruzi, under specific chemotherapy. Mem Inst Oswaldo Cruz 1991;86:187–200. 70.Viotti R, Vigliano C, Armenti H, Segura E. Treatment of chronic Chagas’ disease with benznidazole: clinical and serologic evolution of patients with long-term follow up. Am Heart J 1994;127:151–62. 71.Andrade AL, Zicker F, deOliveira RM et al. Randomized trial of efficacy of benznidazole in treatment of early Trypanosoma cruzi infection. Lancet 1996;348:1407–13. 72.Villar JC, Marin-Neto JA, Ebrahim S, Yusuf S. Trypanocidal drugs for chronic asymptomatic Trypanosoma cruzi infection (Cochrane Review). In: The Cochrane Library 2002; (Issue 2). Oxford: Update Software CD003463. 73.Coura JR, de Abreu LL, Faraco Willcox HP, Petana W. Estudo comparativo controlado com emprego de Benznidazole, Nifurtimox e placebo, na forma crônica da doença de Chagas, em uma área de campo com transmissão interrompida. Avaliação preliminar. Rev Soc Bras Med Trop 1997;30: 139–44. 74.Gianella A, Holzman A, Lihoshi N, Barja Z, and Peredo Z. Eficacia del Alopurinol en la enfermedad de Chagas crónica. Resultados del estudio realizado en Santa Cruz, Bolivia. Bol Cientif CENETROP 1997;16:25–30. 75.Sosa-Estani S, Segura EL, Velazquez E, Ruiz AM, Porcel BM, Yampotis C. Efficacy of chemotherapy with benznidazole in children in the indeterminate phase of Chagas’ disease. Am J Trop Med Hyg 1998;59:526–9. 76.Apt W, Aguilera X, Arribada A et al. Treatment of chronic Chagas’ disease with itraconazole and allopurinol. Am J Trop Med Hyg 1998;59:133–8. 77.Rassi A, Amato Neto V et al. [Protective effect of benznidazole against parasite reactivation in patients chronically infected with Trypanosoma cruzi and treated with corticoids for associated diseases]. Rev Soc Bras Med Trop 1999;32:475–82. 78.Nishioka Sde A. [Benznidazole in the primary chemoprophylaxis of the reactivation of Chagas’ disease in chronic chagasic patients using corticosteroids at immunosuppressive doses: is there sufficient evidence for recommending its use?]. Rev Soc Bras Med Trop 2000;33:83–5. 79.Batlouni M, Barretto AC, Armaganijan D et al. Treatment of mild and moderate cardiac failure with captopril. A multicenter trial. Arq Bras Cardiol 1992;58:417–21. 80.Roberti RR, Martinez EE, Andrade JL et al. Chagas’ cardiomyopathy and captopril. Eur Heart J 1992;13:966–70. 81.Bocchi EA, Bellotti G, Mocelin AO et al. Heart transplantation for chronic Chagas’ heart disease. Ann Thorac Surg 1996;61:1727–33. 82.Bochi EA, Fiorelli A. On behalf of the First Guidelines Group for Heart Transplantation of the Brazilian Society of Cardiology. The Paradox of Survival Results after Heart

732

Transplantation for Cardiomyopathy caused by T. cruzi. Ann Thor Surg 2001;71:1833–8. 83.Jatene AD, Moreira LF, Stolf NA et al. Left ventricular function changes after cardiomyoplasty in patients with dilated cardiomyopathy. J Thorac Cardiovasc Surg 1991;102:132–8. 84.Moreira LF, Stolf NA, Braile DM, Jatene AD. Dynamic cardiomyoplasty in South America. Ann Thorac Surg 1996;61: 408–12. 85.Braile DM, Godoy MF, Thevenard GH et al. Dynamic cardiomyoplasty: long-term clinical results in patients with dilated cardiomyopathy. Ann Thorac Surg 2000;69:1445–7. 86.Bocchi EA. Cardiomyoplasty for treatment of heart failure. Eur J Heart Fail 2001;3:403–6. 87.Albanesi-Filho FM, Gomes JB. O tromboembolismo em pacientes com lesão apical da cardiopatia chagásica crônica. Rev Port Cardiol 1991;10:35–42. 88.Rey RC, Lepera SM, Kohler G, Monteverde DA, Sica RE. Cerebral embolism of cardiac origin. Medicina (Buenos Aires) 1992;52:206–16. 89.Rassi Jr A, Rassi SG, Rassi A. Sudden death in Chagas’ disease. Arq Bras Cardiol 2001;76:75–96. 90.Carrasco HA, Vicuna AV, Molina C et al. Effect of low oral doses of disopyramide and amiodarone on ventricular and atrial arrhythmias of chagasic patients with advanced myocardial damage. Int J Cardiol 1985;9:425–38. 91.Rosembaum M, Posse R, Sgammini H et al. Comparative multicenter clinical study of flecainide and amiodarone in the treatment of ventricular arrhythmias associated to chronic Chagas cardiopathy. Arch Inst Cardiol Mex 1987;57: 325–30. 92.Doval HC, Nul DR, Grancelli HD et al. Randomized trial of low-dose amiodarone in severe congestive heart failure. Lancet 1994;344:493–8. 93.Garguichevich JJ, Ramos JL, Gambarte A et al. Effect of amiodarone therapy on mortality in patients with left ventricular dysfunction and asymptomatic complex ventricular arrhythmias: Argentine pilot study of sudden death and amiodarone (EPAMSA). Am Heart J 1995;130:494–500. 94.Olsen EGJ, Spry CJF. Relationship between eosinophilia and endomyocardial disease. Prog Cardiovasc Dis 1985;27: 241–54. 95.Ogunowo PO, Akpan NA, Odigwe CO, Ekanem IA, Esin RA. Helminth associated hypereosoniphilia and tropical endomyocardiofibrosis (EMF) in Nigeria. Acta Trop 1998;69:127–40. 96.Somers K, Brenton DP, Sood NK. Clinical features of endomyocardial fibrosis of the right ventricle. Br Heart J 1968;30: 309–21. 97.Barreto ACP, Luz PL, Mady C, Bellotti G, Pilleggi F. Determinants of survival in patients with endomyocardial fibrosis. Circulation 1988;78:526–30. 98.Oliveira SA, Barreto ACP, Mady C, Bellotti G, Pilleggi F. Surgical treatment of endomyocardial fibrosis: a new surgical approach. J Am Coll Cardiol 1990;5416:1246–51. 99.Schneider U, Jenni R, Turina J, Turina M, Hess OM. Longterm follow up of patients with endomyocardial fibrosis: effects of surgery. Heart 1998;79:362–7. 100.Moraes F, Lapa C, Hazin S, Tenorio E, Gomes C, Moraes CR. Surgery for endomyocardiofibrosis revisited. Eur J Cardiothorac Surg 1999;15:309–12.

Part IIIf Specific cardiovascular disorders: Pericardial disease Bernard J Gersh, Editor

733


GRADE A

GRADE C


Level 5

GRADE B Level 2

Level 3 Level 4

734




50

Pericardial disease: an evidence-based approach to diagnosis and treatment Bongani M Mayosi, James A Volmink, Patrick J Commerford

Pericardial disease is a potentially curable cause of heart disease that accounts for about 7% of all patients who are hospitalized for cardiac failure in Africa.1 Although there are no good epidemiologic data on the incidence or prevalence of pericarditis in different populations,2 hospital-based series indicate that the spectrum of pericardial disease is determined by the epidemiologic setting of the patient. In Western countries, most cases of primary pericarditis are of unknown cause, whereas tuberculosis accounts for the majority of patients in the developing world.3,4 Thus, evidence-based guidelines should be adapted according to the prevalence of certain diseases in particular geographic areas and patient populations. A discussion of the large number of diseases that may affect the pericardium5 (Box 50.1) cannot be covered in this short chapter. Consequently, this overview will focus on the diagnosis and treatment of idiopathic and tuberculous pericarditides. It will, in particular, aim to examine the extent to which existing treatments are supported by evidence from well-designed prospective studies. The findings reported

Box 50.1 Causes of acute pericarditis5 ● Malignant tumor ● Idiopathic pericarditis ● Uremia ● Bacterial infection ● Anticoagulant therapy ● Dissecting aortic aneurysm ● Diagnostic procedures ● Connective tissue disease ● Postpericardiotomy syndrome ● Trauma ● Tuberculosis ● Other ● radiation ● drugs inducing lupus-like syndrome ● myxedema ● postmyocardial infarction syndrome ● fungal infections ● AIDS-related pericarditis

here are based on a comprehensive search of electronic databases and bibliographies of articles on pericarditis.

Primary acute pericardial disease Acute pericarditis may be caused by a variety of disorders (Box 50.1). Among the secondary forms of pericarditis, the underlying disorder is usually evident before pericardial involvement. The most challenging dilemma for the physician is the patient with acute pericardial disease without apparent cause at the initial evaluation (primary acute pericardial disease). In Western series a specific etiology has been found in only 14–22% of these patients when they are subjected to a prospective diagnostic protocol (Table 50.1).3,6 Diagnosis Acute pericarditis is the occurrence of two or more of the following: characteristic chest pain, pericardial friction rub (pathognomonic of acute pericarditis), and an electrocardiogram (ECG) showing characteristic ST segment elevation or typical serial changes.7 The chest radiograph, echocardiogram, and radionuclide scans are of little diagnostic value in uncomplicated acute pericarditis. The first step in the etiologic diagnosis of acute pericarditis consists of a search for an underlying disease that may require specific therapy. In most cases of suspected viral pericarditis, special studies for etiologic agents are not necessary because of the low diagnostic yield of viral studies and lack of specific therapy for viral disease.7 However, a treatable condition such as Mycoplasma-associated pericarditis must be considered and treated with antibiotics if the serologic test is consistent with the diagnosis.8 The Permanyer-Miralda et al protocol3 for the evaluation of acute pericardial disease is discussed under “Pericardial effusion” below. Treatment Although there are no controlled trials, it is generally accepted that bed rest and oral non-steroidal anti-inflammatory drugs 735


Table 50.1 Etiology of primary acute pericarditis in the West

Acute idiopathic pericarditis Neoplastic pericarditis Tuberculous pericarditis Other infections Collagen vascular disease Other

Permanyer-Miralda et al 19853 (n 231)

Zayas et al 19955 (n 100)

199 13 9 6 2 2

78 7 4 3 3 5

(86%) (6%) (4%) (3%) (0.5%) (0.5%)

(NSAIDs) are effective in most patients with acute pericarditis.7 The use of corticosteroids for acute idiopathic pericarditis when the disease does not subside rapidly is also untested in randomized trials, but it may be unnecessary and even dangerous in acute non-relapsing pericarditis in view of the availability of other agents, such as the parenteral NSAID ketorolac tromethamine.9 Ketorolac is an extremely potent analgesic agent that appears to cause rapid resolution of symptomatic acute pericarditis. However, the limitation of this study of 20 patients with acute pericarditis was that there was no control group for comparison.9 Idiopathic relapsing pericarditis is the most troublesome complication of acute pericarditis, affecting about 20% of cases. There are no established therapeutic guidelines for patients who do not respond to NSAIDs.7 Corticosteroids provide symptomatic relief in most of these patients, but symptoms recur in many when the prednisone dosage is reduced and severe complications are associated with prolonged steroid use.10 Claims of effectiveness have been made in small uncontrolled studies for pericardiectomy, azathioprine, high-dose oral and intravenous corticosteroids,

(78%) (7%) (4%) (3%) (3%) (5%)

and colchicine (Table 50.2).10 The results of these studies are inconsistent and the effectiveness of these potentially harmful therapeutic modalities remains to be established in well-designed randomized studies. Nevertheless, colchicine, used on the basis of its efficacy in the recurrent polyserositis seen in familial Mediterranean fever,16 has aroused much interest following the dramatic effects which were initially reported with its use in recurrent pericarditis.15 The accumulating experience with colchicine indicates that, whilst its long-term use is well tolerated, it is associated with a variable remission rate of 33–100% (Table 50.2), and there is a tendency for a small proportion of patients to relapse after cessation of therapy.16 In a multicenter cohort study involving 51 patients with recurrent pericarditis who did not respond to conventional treatments, colchicine induced remission in 86%, and 60% remained recurrence-free after discontinuation of the drug.20 These data support the use of colchicine to prevent recurrent attacks of pericarditis as an adjunct to conventional treatment, although the effectiveness of the agent remains to be evaluated in randomized controlled trials. Grade B

Table 50.2 Therapeutic strategies previously evaluated in recurrent pericarditis (after failure of non-steroidal anti-inflammatory drugs) Study

Patients (n)

Therapeutic strategy evaluated

Remission rate

Fowler10 Hatcher11 Asplen12 Melchior13

9 24 2 2

2/9 (22%) 20/24 (83%) 2/2 (100%) 2/2 (100%)

Marcolongo14

12

Pericardiectomy Pericardiectomy Azathioprine IV Methylprednisolone as pulse therapy High-dose prednisone with aspirin

Guindo15 and de la Serna16 Spodick17 Adler18 Millaire19 Guindo 20

9 8 8 19 51

736

Colchicine Colchicine Colchicine Colchicine Colchicine

11/12 (92%) Major side effects in 3 9/9 (100%) 3/9 (33%) 4/8 (50%) 14/19 (74%) 44/51 (86%)

Pericardial disease: an evidence-based approach to diagnosis and treatment

Pericardial effusion The spectrum of causes of pericardial effusion is similar to acute pericarditis (Box 50.1). However, prospective studies indicate that large pericardial effusions are more likely to be a result of serious underlying illnesses such as tuberculosis and cancer, where rapid diagnosis may lead to earlier therapy and improved survival.3 The clinical features vary depending on the rate of accumulation of the fluid, the amount of fluid that accumulates, and the stage at which the patient is first seen.

Table 50.3 Protocol for evaluation of primary acute pericardial disease3 Stage Stage I General studies and echocardiogram

Stage II Pericardiocentesis

Diagnosis The radiographic signs of pericardial effusion include an enlarged cardiac silhouette, a pericardial fat stripe, a predominant left-sided pleural effusion, and an increase in transverse cardiac diameter compared with previous chest radiographs. However, these signs cannot reliably confirm or exclude the presence of pericardial effusion, thus making radiography poorly diagnostic of pericardial effusion.21 Similarly, ECG is useful only in that it may suggest a cardiac abnormality. The QRS complexes are usually small, with generalized T wave inversion. Electrical alternans, which suggests the presence of massive pericardial effusion, is uncommon. Even more uncommon is total electrical alternans (P-QRS-T alternation), which is pathognomonic of tamponade.22 Echocardiography, computed tomography (CT), and magnetic resonance imaging (MRI) can accurately detect and quantify pericardial effusion. Echocardiography, which is relatively inexpensive, sensitive (capable of detecting as little as 17 ml pericardial fluid), harmless, and widely available, is the diagnostic method of choice for pericardial effusion.5 Furthermore, it may also provide prognostic information. A large effusion with a circumferential echo-free space of 1 cm in width at any point is reported to be a powerful predictor of tamponade23 and intrapericardial echo images are associated with an increased likelihood of subsequent constriction.24 Permanyer-Miralda et al 3 have evolved a systematic approach for the evaluation of acute primary pericardial disease in developed countries with a low prevalence of tuberculosis (Table 50.3). It is based on a prospective study of 231 consecutive patients who were evaluated to determine the safest and most sensitive approach to the etiologic diagnosis of acute pericardial disease. The findings were confirmed in a subsequent prospective study of a similar diagnostic protocol involving 100 patients with primary pericardial disease.5 First, these prospective studies indicate that a specific etiology is found in only 14–22% of patients with acute primary pericardial disease (Table 50.1). Secondly, while therapeutic pericardiocentesis is absolutely indicated for cardiac tamponade, it is not warranted as a routine investigation because of low diagnostic yield. The

Stage III Surgical biopsy of the pericardium

Stage IV Empirical antituberculous treatment

Evaluation Electrocardiogram Chest radiograph Tuberculin skin test Serologic tests Therapeutic pericardiocentesis: absolutely indicated for cardiac tamponade Diagnostic pericardiocentesis: clinical suspicion of purulent or tuberculous pericarditis Illness lasting for more than 1 week “Therapeutic” biopsy: as part of surgical drainage in patients with severe tamponade relapsing after pericardiocentesis Diagnostic biopsy: in patients with more than 3 weeks illness and without an etiologic diagnosis having been reached by previous procedures Fever and pericardial effusion of unknown origin persisting for more than 5–6 weeks

indications for diagnostic pericardiocentesis are the clinical suspicion of purulent or tuberculous pericarditis and those with an illness lasting longer than 1 week. Thirdly, the diagnostic yield of pericardiocentesis and pericardial biopsy is similar. Whereas biopsy is more invasive and may entail the need for general anesthesia, it is a safe procedure and direct histologic examination of the pericardium may allow immediate diagnosis in the case of tuberculosis. Furthermore, pericardial biopsy may allow a more direct visualization of the pericardium. However, even when detailed investigations are performed, including pericardioscopy and surgery, the etiology of pericardial effusion remains obscure in a significant number of patients.25 Cardiac tamponade A pericardial effusion may result in the life-threatening complication of cardiac tamponade, a condition caused by 737


compression of the heart and impaired diastolic filling of the ventricles. It is an indication for pericardiocentesis. Cardiac tamponade is a clinical diagnosis, which is confirmed by echocardiography. The clinical examination shows elevated systemic venous pressure, tachycardia, dyspnea, and pulsus paradoxus.26 Pulsus paradoxus may be absent in some instances such as left ventricular dysfunction, atrial septal defect, regional tamponade, and positive pressure breathing. Systemic blood pressure may be normal, decreased, or even elevated. The diagnosis is usually confirmed by the echocardiographic demonstration of a large circumferential pericardial effusion and some of the features listed in Table 50.4. However, as a diagnostic test for tamponade, echocardiography may lack both sensitivity and specificity in certain clinical situations. For example, echocardiographic features of right heart collapse may be absent in the presence of loculated effusions causing regional left ventricular compression or in patients with pulmonary hypertension. This is particularly important after cardiac surgery when the absence of a circumferential effusion and right atrial collapse and right ventricular diastolic collapse may not exclude the presence of life-threatening tamponade.27,28 Furthermore, a dilated non-collapsing inferior vena cava and an abnormal respiratory pattern of diastolic flow are not specific signs of tamponade (Table 50.4).

Constrictive pericarditis The etiology of constrictive pericarditis has changed over the past four decades.29 Tuberculous constrictive pericarditis, which was a common cause of constriction worldwide before the 1960s, has since declined in incidence and is now rare in Western countries. In these countries the

diminished importance of tuberculous pericarditis has been associated with a large contribution made by idiopathic cases. Postradiotherapy constriction, which was first recognized as an important disease in the 1960s, continues to feature prominently among the causes of constrictive pericarditis, while postsurgical constriction has emerged as an important cause. Constrictive pericarditis is characterized anatomically by an abnormally thickened and non-compliant pericardium that limits ventricular filling in mid to late diastole. Consequently, nearly all ventricular filling occurs very rapidly in early diastole. This results in elevated cardiac filling pressures and the characteristic hemodynamic waveforms during which the diastolic pressures of the cardiac chambers equalize. The clinical manifestations of constrictive pericarditis, which are secondary to systemic venous congestion, mimic a variety of cardiopulmonary disorders, making the diagnosis of this condition difficult in some cases. Diagnosis The chest radiograph and ECG are usually abnormal, drawing attention to the heart, but the abnormalities are largely non-specific. Chest radiography may reveal a small, normal, or enlarged cardiac silhouette, pleural effusions in 60%, and pericardial calcification in 5–50% of cases.29–31 Calcification is not specific for constrictive pericarditis, as a calcified pericardium does not necessarily imply constriction. Nonspecific but generalized T wave changes are seen in most cases, while low voltage complexes occur in about 30%. The ideal imaging technique for the accurate preoperative diagnosis of pericardial constriction would simultaneously provide both anatomic data describing the thickness of the pericardium and physiologic/hemodynamic data describing

Table 50.4 Echocardiographic features of cardiac tamponade26 Echocardiographic/Doppler criteria

Comments

1. Right heart collapse: right atrial compression, right ventricular diastolic collapse

Changes in blood volume may affect the sensitivity and specificity of right heart collapse as a sign of tamponade. False positives and false negatives may occur

2. Abnormal respiratory changes in ventricular dimensions

Inconstant finding

3. Abnormal respiratory changes in tricuspid and mitral flow velocities

May also be seen in obstructive airways disease, pulmonary embolism, and right ventricular infarction

4. Dilated inferior vena cava with lack of inspiratory collapse

Often seen with congestive heart failure and constrictive pericarditis

5. Left ventricular diastolic collapse

Frequent sign of regional cardiac tamponade and useful marker of tamponade in postoperative patients in a retrospective investigation27

6. Swinging heart

Not sensitive, specificity unknown

738


the characteristic differential diastolic filling to the left ventricle and the right ventricle with respiration. In these regards, echocardiographic findings of abnormal ventricular septal motion (septal bounce or shudder), dilated inferior vena cava, and hepatic veins in patients with right heart failure are suggestive of constrictive pericarditis. Respiratory variation in mitral inflow velocities and hepatic veins is quite characteristic for constrictive pericarditis, although the lack of respiratory variation does not exclude constrictive pericarditis. Specificity of these Doppler findings of constrictive pericarditis is enhanced by demonstrating no significant respiratory variation in the superior vena cava velocity. In patients with increased respiratory effort such as chronic obstructive pulmonary disease, which simulates the interventricular dependence resulting in similar two-dimensional and Doppler echocardiographic findings, superior vena caval velocities are markedly increased with inspiration.32 New tissue Doppler recording of mitral annulus velocity adds more confidence in distinguishing constrictive pericarditis from restrictive process because of myocardial disease (Table 50.5).33

CT and MRI can demonstrate the extent and distribution of pericardial thickening. While this does not make the diagnosis of constriction, it is often very useful to know that the pericardium is abnormal in a patient in whom this diagnosis is suspected. In addition, CT or MRI features of myocardial atrophy or fibrosis predict a poor outcome following pericardiectomy.34 A promising new imaging technique is cine-CT, which simultaneously provides both anatomic and physiologic data that may allow accurate preoperative diagnosis of pericardial constriction.35 Unless the diagnosis is very obvious, cardiac catheterization is usually performed. The characteristic finding is equal end-diastolic pressures in the two ventricles, persisting with respiration and fluid challenge. However, the diagnosis of constrictive pericarditis remains a challenge because it is often mimicked by restrictive cardiomyopathy. A number of studies, using different techniques, have attempted to distinguish the two conditions, including studies of left ventricular filling rate, mitral and tricuspid diastolic flow patterns, pulmonary venous flow velocity, hepatic flow velocity patterns, hemodynamic

Table 50.5 The differentiation of constrictive pericarditis from restrictive cardiomyopathy Type of evaluation

Constrictive pericarditis

Restrictive cardiomyopathy

Physical examination

Regurgitant murmurs uncommon

Regurgitant murmurs may be present

Chest radiography

Pericardial calcification may be present

Pericardial calcification absent

Echocardiography

Normal wall thickness

Increased wall thickness, thickened cardiac valves and granular sparkling texture (amyloid). Enlarged atria Pericardium usually normal

Pericardium may be thickeneda Prominent early diastolic filling with abrupt displacement of interventricular septum due to increased ventricular interaction Doppler studies

Early mitral flow is reduced with onset of inspiration and reciprocal effect on tricuspid flow Expiratory increase of hepatic vein diastolic flow reversal Mitral annulus velocity 7 cm/s

Cardiac catheterization

RVEDP and LVEDP usually equal RV systolic pressure 50 mmHg RVEDP one third of RV systolic pressure

Endomyocardial biopsy CT/MRI

May be normal or show non-specific myocyte hypertrophy or myocardial fibrosis Pericardium may be thickened 3 mma

No respiratory variation in diastolic flow with short deceleration time Inspiratory increase of hepatic vein diastolic flow reversal Mitral and tricuspid regurgitation may be present Mitral annulus velocity 7 cm/s LVEDP often 5 mmHg greater than RVEDP, but may be identical RV systolic pressure 50 mmHg May reveal specific cause of restrictive cardiomyopathy Pericardium usually normal

Abbreviations: CT, computed tomography; LV, left ventricular; LVEDP, left ventricular end-diastolic pressure; MRI, magnetic resonance imaging; RV, right ventricular; RVEDP, right ventricular end-diastolic pressure a Normal thickness of the pericardium does not rule out pericardial constriction.

739


investigations, endomyocardial biopsy, and CT and MRI studies.36 Table 50.5 summarizes the important differences between the two conditions. No technique is totally reliable and in some patients, the only way of making the diagnosis is to perform a pericardiectomy.37 Treatment The treatment for chronic constrictive pericarditis is complete resection of the pericardium. The average hospital mortality following pericardiectomy in several series ranges from about 5% to 16%.29,38–40 Poor outcome is related mainly to preoperative disability, the degree of constriction, and myocardial involvement. The majority of early deaths are associated with low cardiac output, which has been attributed to myocardial atrophy. Thus, early pericardiectomy is recommended in patients with non-tuberculous constrictive pericarditis before severe constriction and myocardial atrophy occur. Among patients who survive the operation, symptomatic improvement can be expected in about 90% and complete relief of symptoms in about 50%. The 5 year survival rate ranges from 74% to 87%. Long-term survival and symptomatic relief do not appear to be influenced by age, choice of median sternotomy or left thoracotomy, or transient low output syndrome postoperatively. However, long-term survival is unfavorably influenced by the presence of severe preoperative functional disability (NYHA class III or IV, diuretic use), renal insufficiency in the preoperative state, the presence of extensive non-resectable calcifications, incomplete pericardial resection, and the presence of radiation pericarditis, which is commonly complicated by myocardial fibrosis and restrictive myocardial disease.

Tuberculous pericarditis The prevalence of tuberculous pericarditis follows the same pattern as that of tuberculosis in general. It is the most common cause of pericarditis in developing countries where tuberculosis remains a major public health problem, but accounts for only about 5% of cases in the West.3,4 In Africa the incidence of tuberculous pericarditis is rising as a direct result of the human immunodeficiency virus epidemic41 and this trend is likely to occur in other parts of the world where the spread of AIDS is leading to the resurgence of tuberculosis. Tuberculosis caused by drug resistant Mycobacterium tuberculosis has emerged in the past few years as a serious threat to global public health, but its impact on pericardial tuberculosis has not been studied. Tuberculous pericarditis appears to be more common in blacks than whites and males than females,42,43 although the sex difference was reversed in the large prospective studies of Strang et al 30,44 The disease can occur at any age. 740

Tuberculous pericardial effusion Tuberculous pericarditis is usually detected clinically either in the effusive stage or after the development of constriction. Tuberculous pericarditis has a variable clinical presentation and it should be considered in the evaluation of all instances of pericarditis without a rapidly self-limited course.43 While tuberculous pericarditis may cause effusions that do not produce cardiac compression, more commonly there is at least some degree of cardiac compression, which may be severe, causing tamponade. Tuberculous pericardial effusion usually develops insidiously, presenting with non-specific systemic symptoms such as fever, night sweats, fatigue, and loss of weight.4,42,45 Chest pain, cough, and breathlessness are common.45–47 Severe pericardial pain of acute onset characteristic of idiopathic pericarditis is unusual in tuberculous pericarditis.42,45,48 Right upper abdominal aching due to liver congestion is also common in these patients.4,42,45 In South African patients with tuberculous pericardial effusion, evidence of chronic cardiac compression that mimics heart failure is by far the most common presentation (Table 50.6).4,47,49 While there is marked overlap between the physical signs of pericardial effusion and constrictive pericarditis, the presence of increased cardiac dullness extending to the right of the sternum favors a clinical diagnosis of pericardial effusion.

Diagnosis A definite diagnosis of tuberculous pericarditis is based on the demonstration of tubercle bacilli in pericardial fluid or on histologic section of the pericardium and a probable diagnosis is made when there is proof of tuberculosis elsewhere in a patient with unexplained pericarditis. A definite or probable diagnosis is made in up to 73% of patients treated for tuberculous pericarditis.44,50 The chest radiograph shows features of active pulmonary tuberculosis in only 30% and pleural effusion is present in 40–60% of cases.43 The ECG is usually abnormal, drawing attention to the heart.44,51 The ST segment elevations characteristic of acute pericarditis are usually absent. ECG findings are not specific for a tuberculous etiology.50 Pericardiocentesis is recommended in all patients in whom tuberculosis is suspected. The pericardial fluid is bloodstained in 80% of cases4 and, since malignant disease and the late effects of penetrating trauma cause bloody pericardial effusion, confirmation of tuberculosis as the cause is important. The difficulty in finding tubercle bacilli in the direct smear examination of pericardial fluid is well known. Culture of tubercle bacilli from pericardial effusions can be improved considerably by inoculation of the fluid into double strength liquid Kirchner culture medium at the bedside. A prospective study of the value of the double strength liquid Kirchner culture medium in patients considered to have


Table 50.6 Physical signs documented by a single observer in 88 patients with pericardial effusion and 67 patients with constrictive pericarditis in South Africa4 Signs

Pericardial effusion (n 88)

Constrictive pericarditis (n 67)

Hepatomegaly

84 (95%)

67 (100%)

Increased cardiac dullness

83 (94%)

17 (25%)

Raised jugular venous pulse

74 (84%)

67 (100%)

Soft heart sounds

69 (78%)

51 (76%)

Sinus tachycardia

68 (77%) (Transient AF in 3)

47 (70%) (Persistent AF in 2)

Ascites

64 (73%)

60 (89%)

Apex palpable

53 (60%)

39 (58%)

Significant pulsus paradoxus

32 (36%)

32 (48%)

Edema

22 (25%)

63 (94%)

Pericardial friction rub

16 (18%)

–

Pericardial knock

–

14 (21%)

Third heart sound

–

30 (45%)

Sudden inspiratory splitting of the second heart sound

–

24 (36%)

Abbreviation: AF, atrial fibrillation

tuberculosis reported a 75% yield compared to a 53% yield with conventional culture.52 For Mycobacterium tuberculosis, the radiometric method (BACTEC) permits an average recovery and drug sensitivity testing time of 18 days, compared to 38·5 days for conventional methods, but the low yield of 54% is the major disadvantage of the former method.52 Sputum with acid-fast bacilli will be found in only about 10% of patients.4 Gastric washings from such patients may be studied and urine culture and lymph node biopsy may also demonstrate tubercle bacilli. In developing countries tuberculin skin testing is of little value owing to the high prevalence of primary tuberculosis, mass BCG immunizations and the likelihood of cross-sensitization from mycobacteria present in the environment.53 Furthermore, the limited utility of the tuberculin skin test has also been documented in a prospective study performed in a non-endemic area.43 There is considerable urgency to establish the correct diagnosis of tuberculosis since early initiation of therapy is associated with a favorable outcome.45 Since tubercle bacilli are often not found on stained smears of pericardial fluid46,54 and their growth on culture requires 3–6 weeks, there is a need for other means of making an early diagnosis. Unfortunately, a rapid, simple, inexpensive, sensitive, and specific diagnostic test for pericardial tuberculosis is not available.53 Pericardial biopsy is an important option, but a normal result does not exclude the diagnosis. Recently, the usefulness of measuring adenosine deaminase activity for the rapid diagnosis of pericardial

tuberculosis has been reported in different study populations with consistent results showing a pericardial fluid activity of 40 U/l to have a sensitivity and specificity of more than 90%.55–57 An analysis of the largest prospective study of the usefullness of the adenosine deaminase test for the diagnosis of pericardial tuberculosis,58 which included a wide spectrum of patients with pericardial effusion, yielded a likelihood ratio of 3·8 and 0·05 for positive and negative tests respectively (Table 50.7). Fagan’s nomogram (Figure 50.1) for interpreting diagnostic test results can be used to determine the usefulness of a positive (adenosine deaminase 40) or negative (adenosine deaminase 40) test result.59,60 Although the likelihood ratio for a positive test is 3·8, a high pretest probability of 80% is associated with a post-test probability of 95% if the adenosine deaminase result is positive. The likelihood ratio of a negative adenosine deaminase test is 0·05, which should confer conclusive changes on pretest to post-test probabilities. In addition to the adenosine deaminase test, the measurement of interferon levels in pericardial fluid may offer another means of early diagnosis. A study involving 12 patients, with definite tuberculous pericardial effusion, and 19 controls indicated that elevated interferon measured by radioimmunoassay in pericardial aspirate is a sensitive (92%) and highly specific (100%) marker of a tuberculous etiology in patients with a pericardial effusion.61 This promising report needs confirmation in larger studies. The polymerase chain reaction is useful in detecting Mycobacterium tuberculosis DNA in pericardial fluid,62–64 741


Table 50.7 Test properties of the adenosine deaminase (ADA) assay derived from Latouf et al 58 ADA level

Definite diagnosis of pericardial tuberculosis Present

ADA 40 U/l ADA 40 U/l Total

Absent

n

Proportion

n

Proportion

77 3 80

77/80 0·963 3/80 0·038

26 77 103

26/103 0·253 77/103 0·748

.1 .2

95

1

1000 500

90

2

200 100 50

80

20 10 5 2 1

60

5 10 %

20

70 50 40 30

30

.5

40

.2 .1 .05

50 60 70 80 90

.02 .01 .005 .002 .001

95

%

20 10 5

2 1 .5 .2

99 Pretest probability

Likelihood ratio

.1 Post-test probability

Figure 50.1 Nomogram for interpreting diagnostic test results. (Adapted from Fagan59,60)

but the technique is less sensitive than established methods and is prone to contamination and false positive results and thus not yet suitable for routine clinical use.53,64 At present, serum antibody tests against specific tuberculoprotein epitopes have not offered a significant diagnostic advance over other methods.53 742

3·80 0·05

Treatment

99

.5

Likelihood ratio

In areas and communities with a high prevalence of tuberculosis, a pericardial effusion is often considered to be tuberculous, unless an alternative cause is obvious, and treatment often has to be commenced before a bacteriologic diagnosis is established.52 A definite diagnosis is not made in about a third of patients treated for tuberculous pericarditis and an adequate response to antituberculous chemotherapy serves as confirmation. When systematic investigation fails to yield a diagnosis in patients living in non-endemic areas, good prospective data indicate that there is no justification for starting antituberculous treatment empirically.7 Grade A Antituberculous chemotherapy dramatically increases survival in tuberculous pericarditis. In the preantibiotic era, mortality was 80–90% and currently it ranges from 8% to 17%.47,65,66 A regimen consisting of rifampicin, isoniazid, and pyrazinamide in the initial phase of at least 2 months, followed by isoniazid and rifampicin for a total of 6 months of therapy has been shown to be highly effective in treating patients with extrapulmonary tuberculosis.67,68 Treatment for 9 months or longer gives no better results and has the added disadvantages of increased costs and poor compliance.68 Short-course chemotherapy is also highly effective in curing tuberculosis in HIV-infected patients,69 although it has not been evaluated specifically in tuberculous pericarditis. In 1988 Strang et al 44 reported a prospective doubleblind evaluation of patients with tuberculous pericardial effusion treated with antituberculous drugs who were randomly allocated to prednisolone or placebo during the first 11 weeks of therapy (Figure 50.2): 240 patients entered the study and 198 were evaluated at 24 months; 42 patients (18%) were excluded from analysis mainly owing to loss to follow up and non-compliance with medication. In this trial, five of 97 patients given prednisolone compared with 11 of 101 given placebo died of pericarditis; seven and 17 needed repeat pericardiocentesis; three and seven open surgical drainage, and 91 and 88 had a favorable functional status at 24 months, respectively. Table 50.8 shows the outcomes for patients in the prednisolone and control groups together with the associated odds ratios (95% CI) and P values for the 198 patients who were analyzed in the trial. Patients treated with prednisolone were significantly less likely to


240 eligible patients entered into the trial

118 refused to enter the drainage/no drainage comparison

122 agreed to enter the drainage/no drainage comparison

29 randomized to drainage/ prednisolone

35 randomized to drainage/ placebo

31 randomized to no drainage/ prednisolone

27 randomized to no drainage/ placebo

57 randomized to prednisolone

61 randomized to placebo

21 included in analysis






Figure 50.2

Tuberculous pericardial effusion trial profile.44 A total of 198 patients were included in the analysis.

Table 50.8 Pericardial effusion: prednisolone versus placebo44 Outcome

Group prednisolone (n 97)

Group placebo (n 101)

Peto’s odds ratio (95% CI)

P value

1. Favorable status at 24 monthsa 2. Repeat pericardiocentesis 3. Subsequent open drainage 4. Pericardiectomy 5. Total with one or more adverse eventsb 6. Death from pericarditis

91/97 7/97 3/97 7/97 21/97 5/97

88/101 17/101 7/101 10/101 35/101 11/101

2·15 (0·84–5·53) 0·41 (0·17–0·95) 0·45 (0·13–1·60) 0·71 (0·26–1·92) 0·53 (0·29–0·98) 0·46 (0·17–1·29)

0·11 0·04 0·22 0·50 0·04 0·14

a

Patients were classified as having a favorable status if the following criteria were fulfilled or if only one was still abnormal: pulse rate 100, jugular venous pulse 5 cm, arterial pulsus paradoxus 10 mmHg, ascites and edema absent/just detectable, physical activity unrestricted, cardiothoracic ratio 55%, and electrocardiogram voltage 6 mm in V6 or 4 mm along the frontal axis. b Includes outcomes numbered 2, 3, 4, and 6.

require repeat pericardiocentesis and had fewer combined adverse events than the placebo group. Although there is a suggestion that prednisolone may have a beneficial effect with regard to death from pericarditis, the 95% confidence intervals are consistent with a null effect. It appears from these data that the adjuvant use of prednisolone in tuberculous pericarditis is associated with a reduced risk of reaccumulation of pericardial fluid and less morbidity during the treatment period, which may be clinically significant. It should, however, be noted that the

exclusion of a high proportion of randomized patients from the analysis may be a source of substantial bias in the findings reported in this study. In support of the possibility of bias, a re-analysis of this trial that includes all the participants in the groups to which they were randomized showed that the tendency for prednisolone to reduce the incidence of cardiac tamponade requiring pericardiocentesis was not statistically significant (RR 0·43, 95% CI 0·19–1·01).70 Similarly, the effect of prednisolone on all-cause mortality showed a promising but non-significant effect (RR 0·53, 743


95% CI 0·23–1·18). Therefore, on the basis of the currently available data, prednisolone cannot be recommended for routine use in patients with tuberculous pericardial effusion. We concur with the recommendation that corticosteroids should be reserved for critically ill patients with recurrent large effusions who do not respond to pericardial drainage and antituberculous drugs alone.31 Grade A In the study by Strang et al,44 which compared prednisolone and placebo, those who were willing were also randomized to open complete drainage by substernal pericardiotomy and biopsy under general anesthesia followed by suction drainage on admission or percutaneous pericardiocentesis as required to control symptoms and signs (Figure 50.2); 101 patients participated in this comparison. Complete open drainage abolished the need for pericardiocentesis (odds ratio 0·12, 95% CI 0·04–0·39) but did not influence the need for pericardiectomy for subsequent constriction (odds ratio 0·45, 95% CI 0·10–2·06) or the risk of death from pericarditis (odds ratio 1·51, 95% CI 0·33–6·96). The impact of antituberculous treatment on the development of constrictive pericarditis in patients with chronic pericardial effusion of unknown cause has been investigated in a randomized trial in India:71 Twenty-five adults were randomized in a prospective 2:1 fashion to receive either three-drug antituberculous treatment (group A) or placebo (group B) for 6 months; 21 patients (14 in group A and seven in group B) completed the study protocol, and were included in the analysis. The primary end points were the development of pericardial thickening diagnosed by CT scan and constrictive pericarditis diagnosed by cardiac catheterization. There was no significant difference between the groups in the development of the combined end point of pericardial thickening and constrictive pericarditis (group A: n 3, 21·4% v group B: n 2, 29·6%; P NS); and pericardial fluid had disappeared in 10 patients (six in group A and four in group B). Thus, antituberculous treatment did not prevent the development of constrictive pericarditis and did not alter the clinical course in patients with large chronic pericardial effusions of undetermined etiology in an endemic area. However, the results of this trial should be considered with caution because of the small sample size involved. Nevertheless, the study makes a very important observation that requires further evaluation. In endemic areas antituberculous chemotherapy, which is not without hazard, is often administered to patients with large pericardial effusions in the absence of proof of tuberculosis.4 Recently, Hakim et al 72 reported the first double-blind randomized placebo controlled trial of adjunctive steroids in the treatment of effusive tuberculous pericarditis in HIV seropositive patients. This Zimbabwean study randomized 58 HIV positive patients aged 18–55 years with suspected tuberculous pericarditis to receive prednisolone (n 29) or placebo (n 29) for 6 weeks, in addition to standard 744

short-course antituberculous chemotherapy. The primary end points were resolution of pericardial fluid and death over an 18-month period of observation. There was no difference in the rate of radiologic and echocardiographic resolution in pericardial effusion. By contrast, there were fewer deaths in the intervention group (5/29) compared with the placebo group (10/29), but the numbers were small and the result could have occurred by chance (RR 0·50, 95% CI 0·19–1·28). Thus the trials of steroids for the treatment of tuberculous pericarditis suggest that prednisolone has a potentially large beneficial effect on survival in immunocompetent and HIV seropositive patients, but the individual trials were too small to be sure that this is a true effect.70 We believe that well-designed and adequately powered trials of steroids in tuberculous pericarditis are warranted. Tuberculous pericardial constriction Constrictive pericarditis is one of the most serious sequelae of tuberculous pericarditis and it occurs in 30–60% of patients despite prompt antituberculous treatment and the use of corticosteroids.42,43 Tuberculosis is the most frequent cause of constrictive pericarditis in developing countries.3,4 The presentation is highly variable, ranging from asymptomatic to severe constriction. The diagnosis is often missed on cursory clinical examination (Table 50.6). The diastolic lift (pericardial knock) with a high-pitched early diastolic sound and sudden inspiratory splitting of the second heart sound are subtle but specific physical signs, and found in 21–45% of patients with constrictive pericarditis. These signs are often missed by the inexperienced observer unless specifically sought. Furthermore, if the investigation is not clinically guided, echocardiography has the potential to miss the signs that are suggestive of this diagnosis. Diagnosis Most patients with constrictive pericarditis in South Africa have the subacute variety, in which a thick fibrinous exudate fills the pericardial sac, compressing the heart and causing a circulatory disturbance. As a result, calcification of the pericardium will be absent in the majority.30 The chest radiograph findings are non-specific. In a study reported by Strang et al, 70% of 143 patients had a cardiothoracic ratio greater than 55% and only 6% had a ratio greater than 75%.30 It is uncommon to find concomitant pulmonary tuberculosis. Nonspecific but generalized T wave changes are seen in most cases, while low voltage complexes occur in about 30% of cases. Atrial fibrillation occurs in less than 5% of cases, is persistent, and usually occurs with a calcified pericardium. As with tuberculous pericardial effusion, the ECG is useful only in drawing attention to the presence of a cardiac abnormality. Echocardiography is particularly valuable in confirming the diagnosis of subacute constrictive pericarditis. Typically,


a thick fibrinous exudate is seen in the pericardial sac and is associated with diminished movements of the surface of the heart, normal sized chambers, absence of valvular heart disease, and absence of myocardial hypertrophy.30 In time, the pericardial exudate condenses into a thick skin surrounding the heart, which usually, but not always, can be distinguished from myocardium. Treatment The treatment of tuberculous pericardial constriction involves the use of antituberculous drugs and pericardiectomy for persistent constriction in the face of drug therapy. In a double-blind, randomized, controlled trial in South Africa, 143 patients with tuberculous pericarditis and clinical signs of a constrictive physiology were allocated to receive prednisolone or placebo in addition to antituberculous drugs

during the first 11 weeks of treatment (Figure 50.3)30: 114 patients were available for evaluation at 24 months; 20% of patients were excluded from analysis mainly due to loss to follow up and non-compliance with medication. Although clinical improvement occurred more rapidly in the prednisolone group and there was a lower mortality from pericarditis at 24 months (4% v 11%) and a lower requirement for pericardiectomy (21% v 30%), these findings were not statistically significant (Table 50.9). The remarkable finding of this study is that constriction resolved on antituberculous chemotherapy within 6 months in most patients, and only 29 (25%) of the 114 patients with constrictive pericarditis required pericardiectomy for persistent or worsening constriction. No controlled studies have compared early pericardiectomy with late pericardiectomy in this condition. We recommend pericardiectomy if the patient’s condition is static

143 eligible patients entered into the trial

Figure 50.3

70 randomized to prednisolone

73 randomized to placebo

53 included in the analysis

61 included in the analysis

Favorable status at 24 months = 50 Pericardiectomy = 11 Death due to pericarditis = 2

Favorable status at 24 months = 52 Pericardiectomy = 18 Death due to pericarditis = 7

Tuberculous constrictive pericarditis trial profile30

Table 50.9 Constrictive pericarditis: prednisolone versus placebo30 Outcome

Group prednisolone (n 53)

Group placebo (n 61)

Peto’s odds ratio (95% CI)

P value

1. Favorable status at 24 monthsa 2. Pericardiectomy 3. Death from pericarditis

50/53 11/53 2/53

52/61 18/61 7/61

2·60 (0·79–8·59) 0·63 (0·27–1·47) 0·35 (0·09–1·36)

0·116 0·29 0·13

a

See note a in Table 50.8.

745


hemodynamically or deteriorating after 4–6 weeks of antituberculous therapy. However, if the disease is associated with pericardial calcification, which is a marker of chronic disease, surgery should be undertaken earlier under the antituberculous drug cover. The reported risks of death with pericardiectomy in patients with tuberculous constrictive pericarditis are variable, ranging from 3% to 16%.40,73 Grade C Effusive constrictive tuberculous pericarditis This mixed form is a common presentation in Southern Africa. There is increased pericardial pressure owing to effusion in the presence of visceral constriction and the venous pressure remains elevated after pericardial aspiration. In addition to physical signs of pericardial effusion, a diastolic knock may be detected on palpation and an early third heart sound on auscultation. In patients with the effusive constrictive syndrome echocardiography may show a pericardial effusion between thickened pericardial membranes, with fibrinous pericardial bands apparently causing loculation of the effusion. The treatment of effusive constrictive pericarditis is a problem because pericardiocentesis does not relieve the impaired filling of the heart and surgical removal of the fibrinous exudate coating the visceral pericardium is not possible. Antituberculous drugs should be given in the standard fashion and serial echocardiography performed to detect the development of a pericardial skin, which is amenable to surgical stripping. The place of corticosteroids in such patients is unknown.

Acknowledgments The authors wish to acknowledge valuable comments made by Dr JK Oh on the earlier version of this chapter. References 1.Maharaj B. Causes of congestive heart failure in black patients at King Edward VIII Hospital, Durban. Cardiovasc J S Afr 1991;2:31–2. 2.Maisch B. Pericardial diseases, with a focus on etiology, pathogenesis, pathophysiology, new diagnostic imaging methods and treatment. Curr Opin Cardiol 1994;9:379–88. 3.Permanyer-Miralda G, Sagrista-Sauleda J, Soler-Soler J. Primary acute pericardial disease: a prospective series of 231 consecutive patients. Am J Cardiol 1985;56:623–9. 4.Strang JIG. Tuberculous pericarditis. Clin Cardiol 1984;7: 667–70. 5.Fowler NO. Pericardial disease. Heart Dis Stroke 1992;1: 85–94. 6.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;75:378–82.

746

7.Permanyer-Miralda G, Sagrista-Sauleda J, Shebatai R et al. Acute pericardial disease: an approach to etiologic diagnosis and treatment. In: Soler-Soler J et al., eds. Pericardial disease: new insights and old dilemmas. Dordrecht: Kluwer Academic Publishers, 1990. 8.Farraj RS, McCully RB, Oh JK, Smith TF. Mycoplasma-associated pericarditis. Mayo Clin Proc 1997;72:33–6. 9.Arunsalam S, Siegel RJ. Rapid resolution of symptomatic acute pericarditis with ketorolac tromethamine: a parenteral nonsteroidal antiinflammatory agent. Am Heart J 1993;125: 1455–8. 10.Fowler NO, Harbin AD. Recurrent pericarditis: follow-up of 31 patients. J Am Coll Cardiol 1986;7:300–5. 11.Hatcher CR, Logue RB, Logan WD et al. Pericardiectomy for recurrent pericarditis. J Thorac Cardiovasc Surg 1971;62: 371–8. 12.Asplen CH, Levine HD. Azathioprine therapy of steroid responsive pericarditis. Am Heart J 1970;80:109–11. 13.Melchior TM, Ringsdal V, Hildebrandt P, Torp-Pedersen C. Recurrent acute idiopathic pericarditis treated with intravenous methylprednisolone given as pulse therapy. Am Heart J 1992;123:1086–8. 14.Marcolongo R, Russo R, Lavender F, Noventa F, Agostini C. Immunosuppressive therapy prevents recurrent pericarditis. J Am Coll Cardiol 1995;26:1276–9. 15.Guindo J, Rodriguez de la Serna A, Ramio J et al. Recurrent pericarditis. Relief with colchicine. Circulation 1990;82: 1117–20. 16.Rodriguez de la Serna A, Guindo Soldevila J, Marti Claramunt V, Bayes de Luna A. Colchicine for recurrent pericarditis. Lancet 1987;ii:1517. 17.Spodick DH. Colchicine therapy for recurrent pericarditis. Circulation 1991;83:1830. 18.Adler Y, Zandman-Goddard G, Ravid M et al. Usefulness of colchicine in preventing recurrences of pericarditis. Am J Cardiol 1994;73:916–17. 19.Millaire A, deGroote P, Decoulx E et al. Treatment of recurrent pericarditis with colchicine. Eur Heart J 1994;15:120–4. 20.Guindo J, Adler Y, Spodick H et al. Colchicine for recurrent pericarditis: 51 patients followed up for 10 years. Circulation 1997;96(Suppl. I):I-29 (Abstract). 21.Eisenberg MJ, Dunn MM, Kanth N, Gamsu G, Schiller NB. Diagnostic value of chest radiography for pericardial effusion. J Am Coll Cardiol 1993;22:588–92. 22.Spodick DH. Electrical alternation of the heart. Its relation to the kinetics and physiology of the heart during cardiac tamponade. Am J Cardiol 1962;10:155–65. 23.Eisenberg MJ, Oken K, Guerrero S, Saniei MA, Schiller NB. Prognostic value of echocardiography in hospitalized patients with pericardial effusion. Am J Cardiol 1992;70:934–9. 24.Sinha PR, Singh BP, Jaipuria N et al. Intrapericardial echogenic images and development of constrictive pericarditis in patients with pericardial effusion. Am Heart J 1996; 132: 1268–72. 25.Nugue O, Millaire A, Porte H et al. Pericardioscopy in the etiologic diagnosis of pericardial effusion in 141 consecutive patients. Circulation 1996;94:1635–41. 26.Fowler NO. Cardiac tamponade: a clinical or an echocardiographic diagnosis? Circulation 1993;87:1738–41.


27.Chuttani K, Pandian NG, Mohanty PK et al. Left ventricular diastolic collapse: an echocardiographic sign of regional cardiac tamponade. Circulation 1991;83:1999–2006. 28.Chuttani K, Tischler MD, Pandian MG, Lee RT, Mohanty PK. Diagnosis of cardiac tamponade after cardiac surgery: relative value of clinical, echocardiographic and hemodynamic signs. Am Heart J 1994;127:913–18. 29.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: 1380–6. 30.Strang JIG, Kakaza HHS, Gibson DG et al. Controlled trial of prednisolone as adjuvant in treatment of tuberculous constrictive pericarditis in Transkei. Lancet 1987;ii:1418–22. 31.Lorell BH. Pericardial diseases. In: Braunwald E, ed. Heart disease: a textbook of cardiovascular medicine. Philadelphia: WB Saunders, 1997. 32.Boonyaratavej S, Oh JK, Tajik AJ, Appleton CP, Seward JB. Comparison of mitral inflow and superior vena cava Doppler velocities in chronic obstructive pulmonary disease and constrictive pericarditis. J Am Coll Cardiol 1998;32:2043–8. 33.Ha J-W, Oh JK, Ling LH, Nishimura RA, Seward JB. Annulus paradoxus: transmitral flow velocity to mitral annular velocity ratio is inversely proportional to pulmonary capillary wedge pressure in patients with constrictive pericarditis. Circulation 2001;104:976–8. 34.Reinmuller R, Gurgan M, Erdmann E et al. CT and MR evaluation of pericardial constriction: a new diagnostic and therapeutic concept. J Thorac Imaging 1993;8:108–21. 35.Oren RM, Grover-McKay M, Stanford W, Weiss RM. Accurate preoperative diagnosis of pericardial constriction using cine computed tomography. J Am Coll Cardiol 1993; 22:832–8. 36.Fowler NO. Constrictive pericarditis: its history and current status. Clin Cardiol 1995;18: 341–50. 37.Kushwaha SS, Fallon JT, Fuster V. Restrictive cardiomyopathy. N Engl J Med 1997;336:267–76. 38.Tirilomis T, Unverdorben S, von der Emde J. Pericardiectomy for chronic constrictive pericarditis: risks and outcome. Eur J Cardiothor Surg 1994;8:487–92. 39.McCaughan BC, Schaff HV, Piehler JM et al. Early and late results of pericardiectomy for constrictive pericarditis. J Thorac Cardiovasc Surg 1985;89:340–50. 40.Bashi VV, John S, Ravikumar E et al. Early and late results of pericardiectomy in 118 cases of constrictive pericarditis. Thorax 1988;43:637–41. 41.Cegielski JP, Ramaiya K, Lallinger GJ, Mtulia IA, Mbaga IM. Pericardial disease and human immunodeficiency virus in Dar es Salaam, Tanzania. Lancet 1990;335:209–12. 42.Schrire V. Experience with pericarditis of Groote Schuur Hospital, Cape Town: an analysis of one hundred and sixty cases over a six-year period. S Afr Med J 1959;33:810–17. 43.Sagrista-Sauleda J, Permanyer-Miralda G, Soler-Soler J. Tuberculous pericarditis: ten-year experience with a prospective protocol for diagnosis and treatment. J Am Coll Cardiol 1988;11:724–8. 44.Strang JIG, Kakaza HHS, Gibson DG et al. Controlled clinical trial of complete open surgical drainage and of prednisolone in treatment of tuberculous pericardial effusion in Transkei. Lancet 1988;ii:759–64.

45.Hageman JH, d’Esopo ND, Glenn WWL. Tuberculosis of the pericardium: a long-term analysis of forty-four proved cases. N Engl J Med 1964;270:327–32. 46.Fowler NO, Manitsas GT. Infectious pericarditis. Prog Cardiovasc Dis 1973;16:323–36. 47.Desai HN. Tuberculous pericarditis: a review of 100 cases. S Afr Med J 1979;55:877–80. 48.Quale JM, Lipschik GY, Heurich AE. Management of tuberculous pericarditis. Ann Thorac Surg 1987;43:653–5. 49.Heimann HL, Binder S. Tuberculous pericarditis. Br Heart J 1940;2:165–76. 50.Fowler NO. Tuberculous pericarditis. JAMA 1991;266: 99–103. 51.Schrire V. Pericarditis (with particular reference to tuberculous pericarditis). Aust Ann Med 1967;16:41–51. 52.Strang G, Latouf S, Commerford P et al. Bedside culture to confirm tuberculous pericarditis. Lancet 1991;338:1600–1. 53.Ng TTC, Strang JIG, Wilkins EGL. Serodiagnosis of pericardial tuberculosis. Quart J Med 1995;88:317–20. 54.Schepers GWH. Tuberculous pericarditis. Am J Cardiol 1962; 9:248–76. 55.Koh KK, Kim EJ, Cho CH et al. Adenosine deaminase and carcinoembryonic antigen in pericardial effusion diagnosis, especially in suspected tuberculous pericarditis. Circulation 1994; 89:2728–35. 56.Martinez-Vasquez JM, Ribera E, Ocana I et al. Adenosine deaminase activity in tuberculous pericarditis. Thorax 1986; 41:888–9. 57.Komsouglu B, Goldeli O, Kulan K, Komsouglu SS. The diagnostic and prognostic value of adenosine deaminase in tuberculous pericarditis. Eur Heart J 1995;16:1126–30. 58.Latouf SE, Levetan BN, Commerford PJ. Tuberculous pericardial effusion: analysis of commonly used diagnostic methods. S Afr Med J 1996;86(Suppl.):15 (Abstract). 59.Fagan TJ. Nomogram for Bayes’ theorem (C). N Engl J Med 1975;293:257. 60.Jaeschke R, Guyatt GH, Sackett DL III. How to use an article about a diagnostic test: B. What are the results and will they help me in caring for my patients? JAMA 1994;271:703–7. 61.Latouf SE, Ress SR, Lukey PT, Commerford PJ. Interferongamma in pericardial aspirates: a new, sensitive and specific test for the diagnosis of tuberculous pericarditis. Circulation 1991;84(Suppl.):II-149. 62.Brisson-Noel A, Gicquel B, Lecossier D et al. Rapid diagnosis of tuberculosis by amplification of mycobacterial DNA in clinical samples. Lancet 1989;ii:1069–71. 63.Godfrey-Faussett P, Wilkins EGL, Khoo S, Stoker N. Tuberculous pericarditis confirmed by DNA amplification. Lancet 1991;337:176–7. 64.Cegielski JP, Blythe BH, Morris AJ et al. Comparison of PCR, culture, and histopathology for diagnosis of tuberculous pericarditis. J Clin Microbiol 1997;35:3254–7. 65.Harvey AM, Whitehill MR. Tuberculous pericarditis. Medicine 1937;16:45–94. 66.Bhan GL. Tuberculous pericarditis. J Infect 1980;2:360–4. 67.Cohn DL, Catlin BJ, Peterson KL, Judson FN, Sbarbaro JA. A 62-dose, 6-month therapy for pulmonary and extrapulmonary tuberculosis. A twice-weekly directly-observed, cost-effective regimen. Ann Intern Med 1990;112:407–15.

747


68.Combs DL, O’Brien RJ, Geiter LJ. USPHS Tuberculosis ShortCourse Chemotherapy Trial 21: effectiveness, toxicity and acceptability. The report of final results. Ann Intern Med 1990;112:397–406. 69.Perriens JH, St Louis M, Mukadi YB et al. Pulmonary tuberculosis in HIV-infected patients in Zaire: a controlled trial of treatment for either 6 months or 12 months. N Engl J Med 1995;332:779–84. 70.Mayosi BM, Volmink JA, Commerford PJ. Interventions for treating tuberculous pericarditis (Cochrane Review). In: Cochrane Collaboration. Cochrane Library, Issue 2. Oxford: Update Software, 2001.

748

71.Dwivendi SK, Rastogi P, Saran RK, Rarain VS, Puri VK, Hasan M. Antituberculous treatment does not prevent constriction in chronic pericardial effusion of undetermined aetiology. Indian Heart J 1997;49:411–14. 72.Hakim JG, Ternouth I, Mushangi E, Siziya S, Robertson V, Malin A. Double blind randomised placebo controlled trial of adjunctive prednisolone in the treatment of effusive tuberculous pericarditis in HIV seropositive patients. Heart 2000; 84:183–8. 73.Pitt Fennell WM. Surgical treatment of constrictive tuberculous pericarditis. S Afr Med J 1982;62:353–5.

Part IIIg Specific cardiovascular disorders: Valvular heart disease Bernard J Gersh, Editor

749


GRADE A

GRADE C


Level 5

GRADE B Level 2

Level 3 Level 4

750




51

Rheumatic heart disease: prevention and acute treatment Edmund AW Brice, Patrick J Commerford

Rheumatic fever is the most important cause of acquired heart disease in children and young adults worldwide. Initiated by an oropharyngeal infection with group A hemolytic streptococci (GAS) and following a latent period of approximately 3 weeks, the illness is characterized by an inflammatory process primarily involving the heart, joints, and central nervous system. Pathologically, the inflammatory process causes damage to collagen fibrils and connective tissue ground substance (fibrinoid degeneration) and thus rheumatic fever is classified as a connective tissue or collagen vascular disease. It is the destructive effect on the heart valves that leads to the important effects of the disease, with serious hemodynamic disturbances causing cardiac failure or embolic phenomena resulting in significant morbidity and mortality at a young age. There have been many publications concerning the primary and secondary prevention of rheumatic fever and the treatment of the acute attack. The evidence from randomized controlled clinical trials is strongest in the field of primary prevention or the treatment of pharyngitis caused by GAS. There are few randomized trials concerning secondary prevention. In the treatment of the acute attack, most publications have been observational studies with only a small minority of randomized trials.

Epidemiology In the developed countries of the world, the incidence of rheumatic fever fell markedly during the 20th century. For example, in the USA the incidence per 100 000 was 100 at the start of this century, 45–65 between 1935 and 1960, and is currently estimated to be approximately 2 per 100 000. This decrease in rheumatic fever incidence preceded the introduction of antibiotics and is a reflection of improved socioeconomic standards, less overcrowded housing, and improved access to medical care. The current prevalence of rheumatic fever in the USA and Japan, 0·6–0·7 per 1000 population, contrasts sharply with that in the developing countries of Africa, Asia, and South America, where rates as high as 15–21 per 1000 have been reported. For example, in a study of 12 050 schoolchildren in Soweto, South Africa, a peak prevalence of rheumatic heart disease of 19·2/1000 children was reported.1

As GAS pharyngitis and rheumatic fever are causally related, both diseases share similar epidemiologic features. The age of first infection is commonly between 6 and 15 years. Also, the risk for developing rheumatic fever is highest in situations where GAS is more common, for example where people live in crowded conditions.

Pathogenesis Clinical, epidemiologic, and immunologic observations tend to support strongly the causative role of untreated GAS pharyngitis in rheumatic fever. Beyond this, however, the pathogenesis of acute rheumatic fever and clinical heart disease remains unclear and several important and unexplained observations render the management of this important disease extremely difficult. These are: ● ● ● ●

●

individual variability of susceptibility to GAS pharyngitis; individual variability of development of symptomatic GAS pharyngitis; individual variability of development of acute rheumatic fever after an episode of GAS pharyngitis; individual variation in the development of carditis and chronic rheumatic heart disease after an attack of acute rheumatic fever; the development of chronic rheumatic heart disease in patients who have no definite history of acute rheumatic fever.

Streptococcal skin infection (impetigo) has not been shown to cause rheumatic fever. While effective antibiotic treatment virtually abolishes the risk of rheumatic fever, in situations of untreated epidemic GAS pharyngitis up to 3% of patients develop it.2 Worryingly, as many as a third of patients who develop rheumatic fever do so after virtually asymptomatic GAS and in more recent outbreaks, 58% denied preceding symptoms.3 This does not augur well for the primary prevention of rheumatic fever where prompt diagnosis of GAS pharyngitis and treatment are essential. The virulence of the streptococcal infection is dependent on the organisms’ M protein serotype, which determines the antigenic epitopes shared with human heart tissue, especially sarcolemmal membrane proteins and cardiac myosin.4 It is 751


these variations in virulence, as a result of M protein variation, that are thought to explain the occasional outbreaks of rheumatic fever in areas of previously low incidence.5 Other factors influencing the risk for rheumatic fever are the magnitude of the immune response and the persistence of the organism during the convalescent phase of the illness.2 Evidence suggests that host factors play a role in the risk for rheumatic fever. In patients who have suffered an attack of rheumatic fever, the incidence of a repeat attack is approximately 50%. A specific B cell alloantigen has been found to be present in 99% of patients with rheumatic fever versus 14% of controls.6 Certain HLA antigens appear to be associated with increased risk for rheumatic fever. Approximately 60–70% of patients worldwide are positive for HLA-DR3, DR4, DR7, DRW53, or DQW2.7 Such genetic markers for rheumatic fever risk may be useful to identify those in need of GAS prophylaxis. However, in view of the frequency with which these markers occur, they are unlikely to be of practical benefit in the short term.

Clinical features While there is no specific clinical, laboratory or other test to confirm conclusively a diagnosis of rheumatic fever, the diagnosis is usually made using the clinical criteria first formulated in 1944 by T Duckett Jones8 and subsequently modified by the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease of the Council on Cardiovascular Disease in the Young (American Heart Association).9 The revised criteria emphasize the importance of diagnosing initial attacks of rheumatic fever. The criteria are often incorrectly applied to the diagnosis of recurrent attacks, for which they were not originally intended. The diagnosis is suggested if, in the presence of preceding GAS infection, two major criteria (carditis, chorea, polyarthritis, erythema marginatum, and subcutaneous nodules) or one major and two minor criteria (fever, arthralgia, elevated erythrocyte sedimentation rate, elevated C-reactive protein, or a prolonged PR interval on ECG) are present. Evidence of preceding GAS infection, essential for the diagnosis, may be obtained from throat swab culture (only positive in approximately 11% of patients at the time of diagnosis of acute rheumatic fever)3 or by demonstrating a rising titer of antistreptococcal antibodies, either antistreptolysin O (ASO) or anti-deoxyribonuclease B (anti-DNase B).

Prevention The most recent recommendations on the prevention of rheumatic fever have been published by the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease of the Council on Cardiovascular Disease in the Young (American Heart Association).10 752

Prevention of rheumatic fever may be considered to be either prevention of the initial attack (primary prevention) or prevention of recurrent attacks (secondary prevention). True primary prevention of rheumatic fever depends more on socioeconomic than medical factors. Upgrading housing and other aspects of urban renewal will do more toward eradicating the disease than antibiotic prophylaxis. Primary prevention Prevention of the initial attack of rheumatic fever depends on the prompt recognition of GAS pharyngitis and its effective treatment. Whilst it has been demonstrated that therapy initiated as long as 9 days after the onset of GAS pharyngitis can prevent an attack of rheumatic fever,11 early treatment reduces both the morbidity and the period of infectivity. The first report of the use of penicillin for the treatment of GAS pharyngitis and prevention of most attacks of rheumatic fever was published in 1950.11 Over the following 40 years, attention focused on accurate diagnosis and treatment of GAS pharyngitis. A single dose of intramuscular benzathine penicillin G became the most common mode of treatment and avoided problems of non-compliance. Subsequently, as a result of the pain and possibility of allergic reaction associated with benzathine penicillin G, oral penicillin became the treatment of choice12 and remains so today.13 In situations where compliance with a 10 day course of oral penicillin would be unreliable, a single dose of IM benzathine penicillin G would be preferred (dosage 1·2 million U if 27 kg, otherwise 600 000 U). Early studies established a 10 day course of oral penicillin as optimal14,15 and this has been supported in more recent studies.16,17 Shorter treatment periods are associated with significant decreases in bacteriological cure while longer courses of treatment do not increase cure rate. Current recommendations10 for penicillin therapy in children cite a dose of 250 mg 2–3/day. These recommendations are based on trials (Table 51.1) of 250 mg given 2–4/day resulting in equivalent cure rates.18–21 A dose of 750 mg/day penicillin yielded significantly worse results than 250 mg 3/day when compared in a randomized study.22 There is no evidence available for optimal doses of penicillin in adults but 500 mg 2–3/day is currently recommended.10 Grade A Over the past decade, many trials have been published comparing penicillin VK to a variety of other antimicrobial agents, most commonly cephalosporins and macrolides. This has been prompted by the reported increase in treatment failures with penicillin. It has been suggested that treatment failure rates of up to 38% are possible. This contention, however, has been thoroughly investigated in a study by Markowitz et al 23 in which treatment failure rates of penicillin were compared between two time periods,

Rheumatic heart disease: prevention and acute treatment

Table 51.1

Cure rates for various penicillin dosage schedules used in treatment of streptococcal pharyngitis

Reference

Agent/dose

Bacteriologic cure rate (%)

21

Pen V 250 mg 2/day 10 days Pen V 250 mg 3/day 10 days

82·0 71·5

Gerber et al. (1989)22


78·0 92·0

Potassium Pen G 80 000 U 2/day 10 days

88·0


83·0 84·0

Gerber et al. (1985)

Vann and Harris (1972)19 Spitzer and Harris (1977)

20

Abbreviation: Pen, penicillin

1953–1979 and 1980–1993. Of the almost 2800 patients with GAS serotyping, treatment failures ranged between 10·5% and 17%, with no significant difference between each time period. It was thus concluded that the overreporting of treatment failures was due to problems with the design of the individual studies. An increased bateriologic cure rate for streptococcal pharyngitis by cephalosporins was demonstrated in a meta-analysis24 of 19 randomized comparisons of a variety of cephalosporins with 10 days of oral penicillin therapy. Throat swab cultures were used to determine the presence of GAS and clearance after treatment. The results showed a statistically significant advantage of cephalosporins for which a bacteriologic cure rate of 92% was reported versus 84% for penicillin. The corresponding clinical cure rates were 95% and 89% respectively. It is suggested that the resistance of cephalosporins to penicillinase-producing anaerobes and staphylococci present in the pharyngeal flora may explain these findings. This difference in efficacy would mean that 12–13 patients would require cephalosporin treatment to potentially prevent one penicillin bacteriological treatment failure. More recently, a multicenter comparison of 10 day therapy with cefibuten oral suspension (9 mg/kg/day in one dose) and penicillin V (25 mg/kg/day in three divided doses)25 revealed a bacteriological cure rate of 91% versus 80% respectively (corresponding clinical cure rates were 97% v 89%). Grade A Shorter courses of selected cephalosporins26 (4 or 5 days) have been shown to be effective, but current recommendations10 suggest that further study of these regimens is required before their adoption. The cephalosporins offer statistically significant advantages over penicillin in controlled clinical trials. It remains to be demonstrated, however, whether this statistical benefit can be translated into clinical or epidemiological benefit in regions where the disease is endemic. Given the financial constraints on healthcare resources of developing nations and the considerable cost difference, it would seem that this is unlikely in the foreseeable future. Greater benefit is likely to be achieved by concerted efforts to identify, treat, and

ensure compliance in large numbers of patients with the established, albeit inferior, penicillin schedules. In patients allergic to penicillin, erythromycin has been shown to have an equivalent cure rate.27 The recommended dosage for erythromycin estolate is 20–40 mg/kg/ day in 2–4 divided doses, and for erythromycin ethylsuccinate, it is 40 mg/kg/day in 2–4 divided doses, both for 10 days.28 Grade A The efficacy of erythromycin estolate is superior to that of erythromycin ethylsuccinate and is associated with fewer gastrointestinal tract side effects.29 GAS strains resistant to erythromycin have been reported in some parts of the world.30 Thus, penicillin V remains the treatment of choice in nonpenicillin allergic patients as it has a long record of efficacy and is probably the most cost effective option. Appropriate antibiotic therapy in children with streptococcal pharyngitis should result in a clinical response within 24 hours – most children will become culture negative within the first or second day of treatment.31 After completion of therapy, only patients who have persistent or recurring symptoms, or those at an increased risk for recurrence, require repeat throat swab culture. If symptomatic patients are still harboring GAS in the oropharynx, a second course of antibiotics, preferably with another agent (amoxicillin clavulanate, cephalosporins, clindamycin or penicillin and rifampicin), is recommended.10 Failure to eradicate GAS occurs more frequently following the administration of oral penicillin than IM benzathine penicillin G.32 Further treatment of asymptomatic patients, who are frequently chronic GAS carriers, is only indicated for those with previous rheumatic fever or their family members. Secondary prevention Following an initial attack of rheumatic fever, there is a high risk of recurrent attacks, which increase the likelihood of cardiac damage, and continuous antibiotic therapy is required. This is especially important as GAS infections need not be symptomatic to trigger a recurrence of rheumatic fever, 753


nor does optimal GAS treatment preclude a recurrence. It is recommended that patients who have suffered either proven attacks of rheumatic fever or Sydenham’s chorea be given long-term prophylaxis following the initial treatment to eradicate the pharyngeal GAS organisms. Recommendations regarding the duration of such prophylaxis are largely empiric and based on observational studies. The duration of prophylaxis should be individualized and take into account the socioeconomic conditions and risk of exposure to GAS for that patient. Individuals who have suffered carditis, with or without valvular involvement, are at higher risk for recurrent attacks33,34 and should receive prophylaxis well into adulthood and perhaps for life. If valvular heart disease persists then prophylaxis is indicated for at least 10 years after the last attack of rheumatic fever and at least until 40 years of age. Those patients who have not suffered rheumatic carditis can receive prophylaxis until 21 years of age or 5 years after the last attack.35 The choice of prophylactic agent has to be made with due regard for the likelihood of compliance with a regimen over a period of many years. Grade A Therefore, despite associated pain (which can be ameliorated by using lidocaine as a diluent 36), intramuscular injection of benzathine penicillin G is the method of choice in most situations. The recommended dose is 1·2 million U every 3–4 weeks. A comparison of 3 weekly (n 90) versus 4 weekly (n 63) benzathine penicillin prophylaxis37 demonstrated the superiority of the 3 weekly dosage. The only prophylaxis failure in the 3 weekly dosage group was due to partial compliance versus five true failures in the 4 weekly dosage group. A long-term follow up study38 for a mean period of 6·4 years (range 1–12 years) in 249 consecutively randomized patients to 3 or 4 weekly regimens further supported the former schedule (0·25% v 1·29% prophylaxis failures respectively). Assays for penicillin levels in blood have also shown that 4 weekly dosage did not provide adequate drug levels throughout the intervening period between doses.39 Therefore, only those considered at low risk should receive a 4 weekly dose. Oral prophylaxis has been shown to be less effective than intramuscular penicillin G prophylaxis, even when compliance is optimal.32 Penicillin V 250 mg 2/day for adults and children is the recommended dose. No published data exist on other penicillins, macrolides, or cephalosporins for secondary prophylaxis of rheumatic fever. However erythromycin, at a dose of 250 mg 2/day is usually recommended for those allergic to penicillin. Patients who have either had prosthetic valves inserted and/or who are in atrial fibrillation require warfarin anticoagulation. This is a situation that may necessitate the use of an oral prophylaxis regimen. In such patients intramuscular injections of penicillin may carry the risk of hematoma formation, especially in patients rendered asthenic as a consequence of their underlying illness. This important 754

circumstance is, as far as we are aware, not addressed in the literature.

Acute management The aim of the acute treatment of a proven attack of rheumatic fever is to suppress the inflammatory response and so minimize the effects on the heart and joints, to eradicate the GAS from the pharynx, and provide symptomatic relief. The longstanding recommendation of bed rest would appear to be appropriate, mainly in order to lessen joint pain. Grade C The duration of bed rest should be individually determined but ambulation can usually be started once the fever has subsided and acute phase reactants are returning towards normal. Strenuous exertion should be avoided, especially for those with carditis. Even though throat swabs taken during the acute attack of rheumatic fever are rarely positive for GAS, it is advisable for patients to receive a 10 day course of penicillin V (or erythromycin if penicillin allergic). Although conventional, this strategy is untested. Thereafter, secondary prophylaxis should commence as described in the previous section. The choice of anti-inflammatory agent is between salicylates and corticosteroids. Grade A Recently, a metaanalysis of trials comparing these two agents has been published.40 In this review, a total of 130 publications from 1949 were assessed. While 11 studies had been randomized, only five (Table 51.2)41–45 fulfilled the meta-analysis criteria of: ● ● ● ●

adequate case definition by the Jones criteria; proper randomization to either salicylates or some form of corticosteroid; non-overlap of subjects between studies; and follow up for at least 1 year for assessment of the presence of an apical systolic murmur suggesting structural cardiac damage as a result of carditis.

The trials varied in the use of steroid agent used, either cortisone, ACTH, or prednisone. The largest study of the five selected for the meta-analysis was that of the Rheumatic Fever Working Party where ACTH, cortisone, and aspirin were compared in a trial involving 505 children in the USA and UK.44 This study found no long-term advantage to be associated with either therapy. While apical systolic murmurs disappeared more rapidly in the steroid-treated groups, the prevalence of a cardiac murmur at 1 year follow up was the same as for the salicylate-treated group. The erythrocyte sedimentation rate was found to normalize and nodules resolved faster in the steroid group. When the five studies were examined in the meta-analysis, it was found that the advantage of corticosteroids over salicylates, in preventing the development of a pathologic apical


Table 51.2 Randomized trials of acute rheumatic fever treatment Reference

Number of patients analyzed

Agent/dose

Apical murmur present at 1 year (%)

Combined Rheumatic Fever Study Group (1960)41

57

Prednisone 60 mg/day 21 days then taper v ASA 50 mg/lb/day 9 weeks, then taper

Steroids 57·1% v ASA 37%

Combined Rheumatic Fever Study Group (1965)42

73

Prednisone 3 mg/lb/day 7 days then taper v ASA 50 mg/lb/day 6 weeks

Steroids 25·3% v ASA 32·1%

Dorfman et al (1961)43

129

Hydrocortisone 250 mg then taper and/or ASA to 20–30 mg%


Rheumatic Fever Working Party (1955)44

497

ACTH 80–120 U and taper v cortisone 300 mg and taper v ASA 60 mg/lb/day and taper

Steroids 48·6% v ASA 44%

Stolzer et al (1965)45

128

ASA 30–60 mg/lb/day 6 weeks v cortisone 50–300 mg/day v ACTH 20–120 v mg/day


systolic murmur after 1 year of treatment, was not statistically significant (estimated odds ratio 0·88, 95% CI 0·53–1·46). All these trials may be criticized on two important points. Firstly, the method used to assess cardiac involvement was clinical with the development or persistence of an apical systolic murmur the usual criterion. It could be argued that observer error and interobserver variability of clinical methodology could invalidate the results and that the question should be re-examined using modern noninvasive techniques. It has, however, been shown that, at least during the acute phase of the illness, transthoracic twodimensional echocardiography with color flow imaging does not add significantly to the clinical evaluation of the degree of cardiac involvement.46 The second point relates to the duration of follow up. Lack of clinical evidence of cardiac involvement at 1 or 2 years following the initial attack of acute rheumatic fever is no guarantee that the important sequelae of valvular incompetence or stenosis will not develop in the ensuing decades. Appropriate dosages of anti-inflammatory agents are aspirin 100 mg/kg/day in four or five divided doses or prednisone 1–2 mg/kg/day. Patients with severe cardiac involvement appear to respond more promptly to corticosteroids.47 The duration of therapy must be gauged from the severity of the attack, the presence of carditis, and the rate of response to treatment. Milder attacks with little or no carditis may be treated with salicylates for approximately a month or until inflammation has subsided, as assessed by clinical and laboratory evidence. More severe cases may require 2–3 months of steroid therapy before this can be gradually weaned. Up to 5% of patients may still have rheumatic activity despite treatment at 6 months. Occasionally a “rebound”

of inflammatory activity can occur when anti-inflammatory therapy is reduced, and may require salicylate treatment. Alternative non-steroidal anti-inflammatory agents have not been adequately assessed in trials and would be of benefit only in individuals allergic to or intolerant of aspirin. A recent prospective randomized controlled trial demonstrated no benefit for intravenous immunoglobulin over placebo when administered during the first episode of rheumatic fever.48 In patients whose initial attack of rheumatic fever is inadequately treated, there is a high risk that the rheumatic activity will continue and result in valvular incompetence, most commonly of the mitral valve. The end result of an ongoing rheumatic process with deteriorating valvular function is heart failure. Experience has shown that in such cases prompt surgical management49 is the sole option and can result in the survival of up to 90% of patients.50 It has been suggested that the reduction in cardiac workload following valve surgery results in a settling of the rheumatic process – akin to the beneficial effect observed for bed rest.

Conclusion While questions regarding the pathogenesis of rheumatic fever remain, sufficient evidence is available to offer guidance on the appropriate prevention and acute treatment of this common illness (Table 51.3). It must be remembered that as most sufferers of this disease are in poor socioeconomic environments and in countries where resources are scarce, the regimens used must be cost effective and chosen with a view to maximizing patient compliance. 755


Table 51.3 Recommendations for prophylaxis and therapy Agent Primary prevention Benzathine penicillin G Penicillin V Erythromycin estolate

Dose

Route

Duration

600 000 U if 27 kg, 1 200 000 U if 27 kg Children 250 mg, 2–3/day Adults 500 mg 2–3/day 20–40 mg/kg/day 2–4/day (max 1g/day)

Intramuscular injection Oral

Once 10 days

Oral

10 days

Secondary prevention (prevention of recurrent attacks)a Benzathine penicillin G 1 200 000 U every 3 weeks (low risk, every 4 weeks) Penicillin V 250 mg 2/day Erythromycin 250 mg 2/day

Intramuscular injection Oral Oral

Treatment of the acute attack of rheumatic fever: Bed rest ● Salicylates 100 mg/kg/day in 4–5 doses (in severe attacks with cardiac involvement, prednisone 1–2 mg/kg/day) ● Valve repair/replacement surgery for severe valve dysfunction. a Duration of secondary prophylaxis depends on history of carditis and if valvular involvement persists. For details see text. ●

A recent study of the effect of a 10 year education program on the reduction of rheumatic fever incidence51 demonstrated what can be achieved by a structured approach to patient identification, community education, and effective diagnosis and treatment. This intervention resulted in a 78% reduction in the incidence of rheumatic fever within 10 years. Much could be achieved through the establishment of similar programs where rheumatic fever is rife.

References 1.McLaren MJ, Hawkins DM, Koornhof HJ et al. Epidemiology of rheumatic heart disease in black schoolchildren of Soweto, Johannesburg. BMJ 1975;3:474–8. 2.Siegel AC, Johnson EE, Stollerman GH. Controlled studies of streptococcal pharyngitis in a pediatric population. 1. Factors related to the attack rate of rheumatic fever. N Engl J Med 1961;265:559–65. 3.Dajani AS. Current status of nonsuppurative complications of group A streptococci. Pediatr Infect Dis J 1991;10:S25–7. 4.Dale JB, Beachey EH. Sequence of myosin cross-reactive epitopes of streptococcal M protein. J Exp Med 1986;164: 1785–90. 5.Schwartz B, Facklam RR, Breiman RF. Changing epidemiology of group A streptococcal infection in the U.S.A. Lancet 1990;336:1167–71. 6.Khanna AK, Buskirk DR, Williams RC et al. Presence of non-HLA B cell antigen in rheumatic fever patients and their families as defined by a monoclonal antibody. J Clin Invest 1989;83:1710–16. 7.Haffejee I. Rheumatic fever and rheumatic heart disease: the current state of its immunology, diagnostic criteria and prophylaxis. Quart J Med 1992;84:641–58.

756

8.Jones TD. Diagnosis of rheumatic fever. JAMA 1944;126: 481–4. 9.Dajani AS, Ayoub EM, Bierman FZ et al. Guidelines for the diagnosis of rheumatic fever: Jones criteria, updated 1992. JAMA 1992;268:2069–73. 10.Dajani A, Taubert K, Ferrieri P et al. Treatment of acute streptococcal pharyngitis and prevention of rheumatic fever: a statement for health professionals. Paediatrics 1995;96:758–64. 11.Denny FW, Wannamaker LW, Brink WR, Rammelkamp CH Jr, Custer EA. Prevention of rheumatic fever: treatment of the preceding streptococci infection. JAMA 1950;143: 151–3. 12.Gerber MA, Markowitz M. Management of streptococcal pharyngitis reconsidered. Pediatr Infect Dis 1984;4:518–26. 13.Nelson JD, McCracken GH Jr, Streptococcal infections (editorial). Pediatr Infect Dis J Newsletter 1993;12:12. 14.Wannamaker LW, Rammelkemp CR Jr, Denny FW et al. Prophylaxis of acute rheumatic fever by the treatment of the preceding streptococcal infection with varying amounts of depot penicillin. Am J Med 1951;10:673–95. 15.Breese BB. Treatment of beta haemolytic streptococcal infections in the home: relative value of available methods. JAMA 1953;152:10–14. 16.Schwartz RH, Wientzen RL, Pedreira F et al. Penicillin V for group A streptococcal pharyngotonsillitis: a randomised trial of seven vs. ten day therapy. JAMA 1981;246:1790–5. 17.Gerber MA, Randolf MF, Chanatry J et al. Five vs. ten days of penicillin V therapy for streptococcal pharyngitis. Am J Dis Child 1987;141:224–7. 18.Breese BB, Disney FA, Talpey WB. Penicillin in streptococcal infections: total dose and frequency of administration. Am J Dis Child 1965;110:125–30. 19.Vann RL, Harris BA. Twice a day penicillin therapy for streptococcal upper respiratory infections. South Med J 1972;65: 203–5.


20.Spitzer TG, Harris BA. Penicillin V therapy for streptococcal pharyngitis: comparison of dosage schedules. South Med J 1977;70:41–2. 21.Gerber MA, Spadaccini LJ, Wright LL, Deutsch L, Kaplan EL. Twice daily penicillin in the treatment of streptococcal pharyngitis. Am J Dis Child 1985;139:1145–8. 22.Gerber MA, Randolf MF, DeMeo K, Feder HM, Kaplan EL. Failure of once-daily penicillin therapy for streptococcal pharyngitis. Am J Dis Child 1989;143:153–5. 23.Markowitz M, Gerber MA, Kaplan EL. Treatment of streptococcal pharyngotonsillitis: reports of penicillin’s demise are premature. J Pediatr 1993;123:679–85. 24.Pichichero ME, Margolis PA. A comparison of cephalosporins and penicillins in the treatment of group A beta-haemolytic streptococcal pharyngitis: a meta analysis supporting the concept of microbial copathogenicity. Pediatr Infect Dis J 1991;10:275–81. 25.Pichichero ME, McLinn SE, Gooch WM IIIrd et al. Cefibuten vs. penicillin V in group A beta-haemolytic streptococcal pharyngitis. Members of the Cefibuten Pharyngitis International Study Group. Pediatr Infect Dis J 1995; 14: S102–7. 26.Aujard Y, Boucot I, Brahimi N, Chiche D, Bingen E. Comparative efficacy and safety of four-day cefuroxime axetil and ten day penicillin treatment of group A beta-haemolytic streptococcal pharyngitis in children. Pediatr Infect Dis J 1995;14:295–300. 27.Shapera RM, Hable KA, Matsen JM. Erythromycin therapy twice daily for streptococcal pharyngitis: a controlled comparison with erythromycin or penicillin phenoxymethyl four times daily or penicillin G benzathine. JAMA 1973;226: 531–5. 28.Derrick CW, Dillon HC. Erythromycin therapy for streptococcal pharyngitis. Am J Dis Child 1976;130:175–8. 29.Ginsberg CM, McCracken GH Jr, Crow SD et al. Erythromycin therapy for group A streptococcal pharyngitis. Results of a comparative study of the estolate and ethylsuccinate formulations. Am J Dis Child 1984;138:536–9. 30.Seppala H, Missinen A, Jarvinen H et al. Resistance to erythromycin in group A streptococci. N Engl J Med 1992;326: 292–7. 31.Krober MS, Bass JW, Michels GN. Streptococcal pharyngitis placebo controlled double-blind evaluation of clinical response to penicillin therapy. JAMA 1985;253:1271–4. 32.Feinstein AR, Wood HF, Epstein JA et al. A controlled study of three methods of prophylaxis against streptococcal infection in a population of rheumatic children. N Engl J Med 1959;260:697–702. 33.Majeed HA, Yousof AM, Khuffash FA et al. The natural history of acute rheumatic fever in Kuwait: a prospective six year follow up report. J Chronic Dis 1986;39:361–9. 34.Kuttner AG, Mayer FE. Carditis during second attacks of rheumatic fever – its incidence in patients without clinical evidence of cardiac involvement in their initial rheumatic episode. N Engl J Med 1963;268:1259–61. 35.Berrios X, delCampo E, Guzman B, Bisno AL. Discontinuing rheumatic fever prophylaxis in selected adolescents and young adults. Ann Intern Med 1993;118:401–6.

36.Amir J, Ginat S, Cohen YH et al. Lidocaine as a diluent for administration of benzathine penicillin G. Pediatr Infect Dis J 1998;17:890–3. 37.Lue HC, Wu MH, Hseih KH et al. Rheumatic fever recurrences: controlled study of 3-week versus 4-week benzathine penicillin prevention programs. J Pediatr 1986;108: 299–304. 38.Lue HC, Wu MH, Wang JK et al. Long-term outcome of patients with rheumatic fever receiving benzathine penicillin G prophylaxis every three weeks versus every four weeks. J Pediatr 1994;125:812–6. 39.Kaplan EL, Berrios X, Speth J et al. Pharmacokinetics of benzathine penicillin G: serum levels during the 28 days after intramuscular injection of 1 200 000 units. J Pediatr 1989; 115:146–50. 40.Albert DA, Harel L, Karrison T. The treatment of rheumatic carditis: a review and meta-analysis. Medicine (Baltimore) 1995;74:1–12. 41.Combined Rheumatic Fever Study Group (RFSG). A comparison of the effect of prednisone and acetylsalicylic acid on the incidence of residual rheumatic carditis. N Engl J Med 1960;262:895–902. 42.Combined Rheumatic Fever Study Group (RFSG). A comparison of short-term intensive prednisone and acetyl salicylic acid therapy in the treatment of acute rheumatic fever. N Engl J Med 1965;272:63–70. 43.Dorfman A, Gross JI, Lorincz AE. The treatment of acute rheumatic fever. Pediatrics 1961;27:692–706. 44.Rheumatic Fever Working Party (RFWP) of the MRC, Great Britain, and the Subcommittee of Principal Investigators of the American Council on Rheumatic Fever and Congenital Heart Disease, American Heart Association. The treatment of acute rheumatic fever in children: a cooperative clinical trial of ACTH, cortisone and aspirin. Circulation 1955;11: 343–71. 45.Stolzer BL, Houser HB, Clark EJ. Therapeutic agents in rheumatic carditis. Arch Intern Med 1955;95:677–88. 46.Vasan RS, Shrivastava S, Vijayakumar M et al. Echocardiographic evaluation of patients with acute rheumatic fever and rheumatic carditis. Circulation 1996;94:73–82. 47.Czoniczer G, Amezcua F, Pelargonio S, Massel BF. Therapy of severe rheumatic carditis: comparison of adrenocortical steroids and aspirin. Circulation 1964;29:813–19. 48.Voss LM, Wilson NJ, Neutze JM et al. Intravenous immunoglobulin in acute rheumatic fever: a randomized controlled trial. Circulation 2001;103:401–6. 49.Lewis BS, Geft IL, Milo S, Gotsman MS. Echocardiography and valve replacement in the critically ill patient with acute rheumatic carditis. Ann Thorac Surg 1979;27:529–35. 50.Barlow JB, Kinsley RH, Pocock WA. Rheumatic fever and rheumatic heart disease. In: Barlow JB, ed. Perspectives on the mitral valve. Philadelphia: FA Davis, 1987. 51.Bach JF, Chalons S, Forier E et al. 10-year educational programme aimed at rheumatic fever in two French Caribbean islands. Lancet 1996;347:644–8.

757

52

Mitral valve disease: indications for surgery Blasé A Carabello

Introduction In mitral valve disease symptomatic status, ventricular functional status and the kind of operation that will ultimately be performed all affect the indication for valve surgery. This chapter will integrate these aspects into a strategy for surgical correction. It should be noted that in surgery for valve disease there are few large controlled trials of therapy. Most knowledge of the response of valve disease to surgery accrues from reports of surgical outcome in both selected and unselected patients.

Mitral regurgitation Surgical objectives Like all valvular lesions, mitral regurgitation imposes a hemodynamic overload on the heart. Ultimately, this overload can only be corrected by surgically restoring valve competence. For valve surgery in general, the timing of surgery has two opposing tenets. First, as surgery has an operative risk and, if a prosthesis is inserted, imposes the risks inherent in valve prosthesis, surgery should be delayed for as long as possible. Second, surgery which is delayed until the hemodynamic overload has caused irreversible left ventricular dysfunction will result in a suboptimal outcome. In some patients, far advanced left ventricular dysfunction may militate against operating at all. The timing of valve surgery is made even more complex in mitral regurgitation, as frequently valve repair rather than valve replacement can be effected. Because valve repair does not involve the use of a valvular prosthesis, and because it also helps to preserve left ventricular function, it is applicable at the two ends of the spectrum of mitral regurgitation. Repair might be considered in asymptomatic patients with normal left ventricular function because the disease could be cured then without the need for intense follow up and without the use of a valve prosthesis.1 At the other end of the spectrum, patients with severe impairment of left ventricular function who might not be candidates for mitral valve replacement with chordal disruption might have a good result from valve repair.2 However, for most 758

patients mitral valve surgery is performed for the relief of symptoms or to prevent worsening of asymptomatic left ventricular dysfunction. Etiology The mitral valve apparatus consists of the mitral valve annulus, the valve leaflets, the chordae tendineae and the papillary muscles. Abnormalities of any of these structures may cause mitral regurgitation. The common causes of mitral regurgitation include infective endocarditis, the mitral valve prolapse syndrome with myxomatous degeneration of the valve, spontaneous chordal rupture, rheumatic heart disease, collagen disease such as Marfan’s syndrome, and coronary artery disease leading to papillary muscle ischemia or necrosis. These etiologies are especially important with regard to surgical correction. For instance, the spontaneous rupture of a posterior chorda tendinea leads to mitral valve repair in almost 100% of cases. On the other hand, severe rheumatic deformity of the valve which has led to mitral regurgitation may be irreparable, necessitating replacement. Pathophysiology Hemodynamic phases of mitral regurgitation Figure 52.1 depicts the pathophysiologic phases of mitral regurgitation.3 In the acute phase, such as might occur with spontaneous chordal rupture, there is sudden volume overload on both the left ventricle and the left atrium. The regurgitant volume, together with the venous return from the pulmonary veins, distends both chambers. Distention of the left ventricle increases use of the Frank–Starling mechanism, by which increased sarcomere stretch increases enddiastolic volume modestly and also increases left ventricular stroke work. The new orifice for left ventricular ejection (the regurgitant pathway) facilitates left ventricular emptying and end-systolic volume decreases. Acting in concert, these two effects increase ejection fraction and total stroke volume. However, as shown in Figure 52.1, panel A, if only 50% of the total stroke volume is ejected into the aorta there is a net loss of 30% of the initial forward stroke volume. At the same time, volume overload on the left atrium increases

Mitral valve disease: indications for surgery

Chronic compensated MR Normal

A

Acute MR

B

C

70cc

100cc

95cc EDV 240

EDV 170

EDV 150 LA 10 mmHg

LA 25 mmHg

15 mmHg

70cc

ESV 50

Chronic compensated MR

Acute MR

Chronic decompensated MR 65cc

95cc

70cc EDV 170 LA 25 mmHg70cc

EDV 260

EDV 240 LA

LA 25 mmHg 85cc

95cc 15 mmHg

EF

FSV RF CC’s

N

2·07

90

N

·67

·0

100

AMR

2·25

60

N

·82

·50

70

ESV 110

ESV 50

ESV 30

Preload Afterload SL ESS µ Kdyne/cm2 CF

95cc

EDV 30

ESV 50


EF


FSV RF CC’s

AMR

2·25

60

N

·62

·5

70

CCMR

2·19

90

CCMR

2·19

90

N

·79

·5

95

CDMR

2·19

120

N

EF

FSV RF CC’s

·79

·50

95

·58

·57

65

Figure 52.1 (Panel A) Normal hemodynamic state compared to acute mitral regurgitation (AMR). In AMR, total stroke volume and ejection performance increase as preload is increased and afterload is reduced. However, forward stroke volume is reduced and left atrial pressure increased. (Panel B) AMR compared to chronic compensated mitral regurgitation (CCMR). In CCMR, increased enddiastolic volume permitted by eccentric hypertrophy increases both total and forward stroke volume. Enlargement of atrium and ventricle allows increased volume to be accommodated at lower filling pressure. Increase in afterload toward normal in this state of compensation reduces ejection performance slightly. (Panel C) Chronic decompensated mitral regurgitation (CDMR) compared with CCMR: contractile function is reduced and afterload is increased in CDMR. Both reduce ejection performance and forward cardiac output. There is further cardiac dilation in CDMR, leading to worsening mitral regurgitation, further compromising pump function by reducing forward stroke volume and increasing filling pressure. CF, contractile function; EDV, end-diastolic volume; EF, ejection fraction; ESS, end-systolic stress; ESV, end-systolic volume; FSV, forward stroke volume; LA, left atrial pressure; N, normal hemodynamic state; RF, regurgitant fraction; SL, sarcomere length. (Reproduced with permission from Carabello.3)

left atrial pressure. At this point in time the patient suffers from low output and pulmonary congestion and appears to be in left ventricular failure, although left ventricular muscle function is either normal or even augmented by sympathetic reflexes. Acute severe mitral regurgitation may lead to shock and pulmonary edema, requiring intra-aortic balloon counterpulsation and urgent mitral valve repair or replacement. However, if the patient can be maintained in a relatively stable condition, he or she may then enter the chronic compensated phase (Figure 52.1, panel B) within 3–6 months. In the chronic compensated phase of mitral regurgitation, eccentric cardiac hypertrophy, in which sarcomeres are laid down in series, allows enlargement of the left ventricle, enhancing its total volume pumping capacity. Total stroke volume is increased, allowing normalization of forward stroke volume. Enlargement of the left atrium accommodates the volume overload at a lower pressure, eliminating pulmonary congestion. In this phase the patient may be remarkably asymptomatic, able to perform normal daily activities, and can even engage in sporting events of modest physical demands.4

The patient may remain in the compensated phase for months or years. However, eventually the persistent volume overload leads to a decline in left ventricular function (Figure 52.1, panel C). A loss of myofibrils or an insensitivity to cyclic AMP may be responsible, at least in part, for loss of left ventricular contractility.5,6 In this phase, left ventricular end-systolic volume increases because the reduced force of contraction results in poor left ventricular emptying, forward stroke volume falls, and left ventricular dilation may worsen the mitral regurgitation. At this time there is re-elevation of the left atrial pressure, resulting again in pulmonary congestion. Notably, the still favorable loading conditions of mitral regurgitation (increased preload and normal afterload) permit a “normal” ejection fraction even though left ventricular dysfunction has developed. Importance of the mitral valve apparatus Although the contribution of the mitral valve apparatus to left ventricular function was noted by Rushmer and Lillehei 759


decades ago,7,8 its physiologic significance and impact on patient care have only recently received widespread appreciation. It is quite clear that the mitral valve apparatus has a wider role than simply to prevent mitral regurgitation. Rather, the apparatus is an integral part of the left ventricular internal skeleton. In early systole, tugging on the apparatus by the chordae tendineae may shorten the major axis while lengthening the minor axis, in turn augmenting preload there during the pre-ejection phase of systole. In addition, the apparatus helps to maintain the normal and efficient ellipsoid shape of the left ventricle. Transection of the chordae causes an immediate fall in left ventricular function.9 Until the importance of chordal preservation during mitral valve surgery was recognized, ejection fraction almost always fell following surgery. This was attributed to increased afterload from surgical closure of the low-impedance pathway which, preoperatively, had facilitated ejection into the left atrium. However, it is now clear that closure of the same low-impedance pathway in which chordal integrity is maintained results in no fall in ejection fraction, or only a modest decline, suggesting that the increased postoperative load theory is not the sole mechanism for ejection fraction falls.2,11–13 In fact, chordal preservation can actually effect a lowering of systolic wall stress (afterload) instead of an increase as left ventricular radius decreases following surgery [stress pressureradius/2 thickness].10 Thus, chordal integrity should be maintained whenever possible. A recent randomized study demonstrated that maintenance of just the posterior apparatus lowers mortality and leads to superior postoperative function compared to posterior and anterior chordal transection.14 Apart from the benefits on left ventricular function, if mitral valve repair can be performed instead of replacement, operative mortality is lower, postoperative survival is better and the need for anticoagulation is removed while thromboembolism remains low.14–17 Even if the mitral valve is so badly damaged that a prosthesis must be inserted, chordal preservation, especially of the posterior chords, can usually be performed, resulting in better ventricular function than if all the chords were removed.10 Unfortunately, despite recognition of the importance of the mitral valve apparatus, repair is only performed in about 30% of all operations for mitral regurgitation, varying from zero in some institutions to 90% in others. Indications for surgery Severity of mitral regurgitation Grade B Under most circumstances only severe mitral regurgitation is corrected surgically. Mild to moderate regurgitation (regurgitant fraction 40%) under most circumstances neither causes symptoms nor leads to left ventricular dysfunction, even over a protracted period of time. Severity is 760

difficult to ascertain by physical examination alone, especially in acute mitral regurgitation. As noted above, in acute mitral regurgitation there has been no time for cardiac dilation to occur. Thus, palpation of the precordium does not reveal a hyperdynamic left ventricular impulse. Although the murmur of mitral regurgitation is present, severity cannot be gauged from its intensity. In most cases of severe mitral regurgitation an S3 should be present. This finding does not necessarily indicate heart failure, but may simply be the result of a large regurgitant volume filling the left ventricle under a higher than usual left atrial pressure. In chronic mitral regurgitation there should be evidence on physical examination of an enlarged hyperdynamic left ventricle, unless the patient’s size or habitus makes physical examination difficult. Failure to find evidence of an enlarged heart suggests that the mitral regurgitation is not either severe enough or chronic enough to cause left ventricular enlargement. In chronic severe mitral regurgitation the chest radiograph should also show cardiac enlargement, and the electrocardiogram is likely to demonstrate left atrial abnormality and left ventricular hypertrophy. In most cases, quantification of regurgitant severity is estimated during echocardiography, with Doppler interrogation of the mitral valve. In acute mitral regurgitation, transthoracic echocardiography may underestimate regurgitant severity.18 In such cases, transesophageal echocardiography is helpful. It should be noted that Doppler flow studies visually demonstrate blood flow velocity across the mitral valve, and not true flow. Because of this, both under- and overestimation of regurgitant severity is possible. Flow mapping, which expresses the regurgitant jet in terms relative to left atrial size, has been used extensively. However, the limitations of this method are well known and the technique is semiquantitative at best.19,20 Other methods, such as the proximal isovelocity surface area, have been employed experimentally and in clinical investigations.21–23 In using proximal isovelocity surface area to estimate regurgitation flow, the area of convergence of the regurgitant jet on the ventricular side of the mitral valve is measured at the point of aliasing. By multiplying proximal isovelocity surface area by the known aliasing velocity, actual flow is obtained, which should be a better indication of regurgitant severity. Unfortunately, the convergence pattern is often difficult to pinpoint clinically and is not applicable in many cases. As with mitral valve repair, practice varies from center to center, with some centers routinely accurately quantifying the severity of disease24 whereas others rely on a visual estimation. When regurgitation severity is in doubt because of discordance between left ventricular size and the regurgitant signal, that is a small left ventricle and left atrium suggesting mild disease and a Doppler signal suggesting severe disease, the issue should be resolved at cardiac catheterization. During cardiac catheterization, hemodynamics and a left ventriculogram give


additional (although also imperfect) information about the degree of mitral regurgitation. The left ventriculogram, unlike the Doppler study, visualizes the actual flow of contrast medium from the left ventricle into the left atrium. Care must be taken to inject enough contrast agent (at least 60 ml) to opacify both the enlarged left ventricle and the left atrium in mitral regurgitation. Coronary arteriography is also performed at cardiac catheterization if there is any suspicion of an ischemic etiology for mitral regurgitation, or when risk factors for coronary disease coexist. Acute mitral regurgitation Grade C In almost all cases of severe acute mitral regurgitation the patient is symptomatic. The acute hemodynamic changes noted above cause decreased forward output and sudden left atrial hypertension, resulting in pulmonary congestion, reduced forward flow, and the symptoms of dyspnea, orthopnea, exercise intolerance and fatigue. Vasodilator therapy may be successful in alleviating symptoms by preferentially increasing forward flow while simultaneously decreasing left ventricular size, thereby partially restoring mitral valve competence.25 If vasodilators fail, or if the patient is so severely decompensated that hypotension contraindicates their use, intra-aortic balloon counterpulsation is necessary. In such cases surgery should follow soon after. This is especially true for the patient with ischemic mitral regurgitation. Such patients may have a volatile course, with initially mild heart failure which progresses unpredictably in severity. These patients require close follow up.26,27 In milder cases where symptoms can be relieved by medical therapy, patients should be given a trial of medical therapy, during which they may enter the compensated chronic phase. In such cases patients may then become asymptomatic for months or years. However, one study28 suggests that such patients are at risk of sudden death. If confirmed, this would indicate that relief of symptoms with medical therapy might be masking hemodynamic or electrical instability, and thus be dangerous. Chronic mitral regurgitation Symptomatic disease Grade B – The onset of symptoms of congestive heart failure, or a more subtle decrease in exercise tolerance, is usually indicative of a change in physiologic status which usually has important clinical significance. The onset of new atrial fibrillation is also probably indicative of a significant change in disease status. Further, atrial fibrillation by itself leads to increased morbidity and decreased cardiac output. In most cases, the onset of symptoms or persistent atrial fibrillation is an indication for mitral valve surgery even when objective indicators of left ventricular function do not show advancement to dysfunction. Early surgery in the mildly

symptomatic patient is especially indicated when there is a high probability that mitral valve repair can be effected. In this circumstance there is no need to delay longer, waiting for more severe symptoms or the onset of more apparent left ventricular dysfunction. A valve repair will allow improvement in lifestyle while at the same time avoiding the risks of a prosthesis. Early surgery may be especially important when mitral regurgitation is due to a flail leaflet, because this condition may be associated with a modest increase in the risk of sudden death.29 If preoperative evaluation indicates that repair is unlikely, close follow up of the patient is indicated. If symptoms continue to worsen, or if left ventricular dysfunction develops, mitral competence should be restored. In the patient with mild symptoms and normal left ventricular function, transesophageal echocardiography to determine valve anatomy is crucial. This procedure is the best preoperative test to define whether or not repair can be performed or if replacement will be necessary.

Assessment of left ventricular function – A major goal in the management of the patient with mitral regurgitation is to correct the lesion prior to the development of irreversible left ventricular contractile dysfunction. Unfortunately, contractility is difficult to measure clinically. Standard ejection phase indices, such as ejection fraction, which are used to gauge left ventricular function in most cardiac diseases, are confounded by the abnormal loading conditions present in mitral regurgitation, necessitating alterations in the way these indices are used.30 Because ejection fraction is augmented by increased preload in mitral regurgitation31 the value for ejection fraction should be supernormal in the face of normal contractility. A “normal” ejection fraction for the patient with mitral regurgitation is probably 0·65–0·75. Indeed, Enriquez-Sarano and colleagues32 have demonstrated that once the ejection fraction falls to less than 0·60 in patients with mitral regurgitation, long-term mortality is increased, suggesting that left ventricular dysfunction has already developed at that threshold for ejection fraction. End-systolic dimension, which is less dependent upon preload, has also been developed as an important indicator of left ventricular dysfunction in this disease. As demonstrated in Figure 52.2, when the end-systolic dimension exceeds 45 mm, the postoperative outcome is worsened.33 This figure, or its angiographic equivalent, has been found to be predictive in other studies.34,35 Careful evaluation of the patient with mitral regurgitation with history and physical examination, augmented by serial echocardiograms, should avoid the situation in which unrecognized left ventricular dysfunction develops. Yearly follow up is probably adequate as long as the ejection fraction exceeds 0·65 and the endsystolic dimension is less than 40 mm. If the ejection fraction is lower or the end-systolic dimension is higher, more 761


frequent follow up is indicated. When the ejection fraction approaches 0·60 or when the end-systolic dimension approaches 45 mm, surgery should be contemplated. 1·0

0·8

Probability

0·6

0·4

0·2

0·0 20

30 40 50 60 Preoperative end-systolic diameter (mm)

70

Figure 52.2 Plot of an S-shaped curve-related computed probability of postoperative death or severe heart failure to measured preoperative end-systolic diameter. Individual data coordinates are indicated by solid squares and bars represent upper 95% confidence intervals that were computed from the standard error (some points are overlapping; total n 61). (Reproduced with permission from Wisenbaugh et al.31)

Indications for surgery in the asymptomatic patient with mitral regurgitation Patients with normal left ventricular function Grade B – At first glance the asymptomatic patient with mitral regurgitation who has normal left ventricular function would not seem to require surgery. In this patient, surgery will neither improve lifestyle nor prevent reversible left ventricular dysfunction from developing imminently. However, patients with flail leaflet may become symptomatic within the next year29 and may be at some increased risk for sudden death.

In other cases where it is apparent that the severity of mitral regurgitation will eventually necessitate surgery, it could be argued that if mitral valve repair can be performed, little is to be gained by waiting. This circumstance is much like atrial septal defect, where at low operative mortality (less than 1%) the defect can be repaired without the use of a prosthesis before unwanted sequelae develop (in the case of atrial septal defect, persistent atrial arrhythmias and pulmonary hypertension; in the case of mitral regurgitation, left ventricular dysfunction). If this approach is taken it must be clear that repair rather than replacement can be effected. If the asymptomatic patient with normal left ventricular function is ultimately treated with a prosthesis when a repair had been anticipated, it should be considered a complication of surgery, as the unwanted risks of a prosthesis could have been at least temporarily avoided. Asymptomatic patients with left ventricular dysfunction Grade B – It is the asymptomatic patient with left ventricular dysfunction at whom serial follow up is aimed. If left ventricular dysfunction has developed (ejection fraction 0·6, endsystolic dimension 45 mm), surgery should be performed to prevent further irreversible left ventricular dysfunction even if it entails a prosthetic valve. As left ventricular dysfunction has already been indicated by noninvasive testing in such patients, every effort should be made to spare at least part of the mitral valve apparatus to prevent a further decline in left ventricular function postoperatively. Asymptomatic elderly patients Grade C – Patients over the age of 75 with mitral regurgitation are at increased risk for operative death and a poor outcome. This is especially true if replacement instead of repair is performed, or if concomitant coronary disease – a consequence of aging – is present.12,36 Thus, elderly asymptomatic patients with mild left ventricular dysfunction should probably be managed medically. Only patients with severe symptoms in whom medical therapy is ineffective should undergo this relatively high-risk procedure. A summary of indications for surgery is given in Table 52.1.

Table 52.1 Indications for mitral surgery in asymptomatic patients with severe non-ischemic mitral regurgitation Repair likely

Repair unlikely

Patient aged 75 with flail leaflet Patient aged 75 with persistent atrial fibrillation Patient aged 75 with EF 0·60 or ESD 45 mm

— — Patient aged 75 with EF 0·60 or ESD 45 mm

Abbreviations: EF, ejection fraction; ESD, end-systolic minor axis dimension

762


Establishment of symptom status – Because of the insidious nature of mitral valve disease, patients may subtly alter their lifestyle to maintain their asymptomatic status. Thus, history alone may fail to identify this gradual decline in exercise tolerance. Therefore, in patients with mitral valve disease, formal exercise testing is useful to objectively quantify changes in exercise tolerance over the time of follow up and to separate truly asymptomatic patients from those who avoid situations that produce symptoms. Far advanced disease Grade B Occasionally patients reach the first attention of the physician when in severe congestive heart failure, with far advanced left ventricular dysfunction. Many patients in this category may benefit from surgery because correction of mitral regurgitation will lower left atrial pressure and perhaps increase forward output. However, in such cases postoperative left ventricular function will remain depressed and lifespan is likely to be shortened. It is often difficult to decide whether left ventricular dysfunction is so far advanced that surgery should not be performed. The answer to this question is predicated upon the kind of operation that is contemplated. If repair with sparing of most chordal structures can be performed, patients with an ejection fraction as low as 30% can survive surgery with postoperative ejection performance maintained at this relatively low level.2 However, for patients with an ejection fraction 40% in whom only mitral valve replacement can be performed, operative mortality might be prohibitive. Wisenbaugh33 has further suggested that if the end-systolic dimension exceeds 50 mm in patients with rheumatic mitral regurgitation, postoperative risk is extremely high whether repair can be effected or not. Ischemic mitral regurgitation Grade C The prognosis for ischemic mitral regurgitation remains substantially worse than for non-ischemic disease.37,38 A worsened prognosis probably accrues from the automatic presence of a second potentially fatal and independently progressive cardiac disease, and from the presence of ischemic myocardial dysfunction. Guidelines for surgery are not well developed. Common sense indicates that surgery should be performed when ischemic mitral regurgitation has caused shock or intractable pulmonary congestion. Medical therapy Grade C Apart from the use of prophylactic antibiotics against infective endocarditis, there is no proven medical therapy for chronic mitral regurgitation. Although vasodilators are effective in treating the acute disease, no large long-term trials have been performed to examine their effect in chronic disease.

The trials that have been performed differ regarding benefit from this therapy.39,40 Further, because afterload is not typically elevated in chronic mitral regurgitation, the physiologic underpinnings for vasodilators used for afterload reduction are less clear. In fact, vasodilators in this case might lead to cardiac atrophy, potentially putting the patient at a disadvantage when mitral valve replacement is finally performed. Summary Patients with acute mitral regurgitation and severe hemodynamic instability require surgical correction. In less severe situations medical therapy may allow the patient to enter the chronic compensated phase, in which surgery can be delayed. When symptoms develop in chronic mitral regurgitation, they are usually an indication for valve surgery. This is especially true if left ventricular dysfunction is developing, or if it is certain that a mitral valve repair can be performed. In asymptomatic patients with normal ventricular function surgery should only be contemplated when there is a certainty of repair. On the other hand, if left ventricular dysfunction is developing surgery should be performed to prevent further deterioration, whether or not a repair can be effected.

Mitral stenosis Etiology and pathophysiology Most mitral stenosis in adults is acquired through rheumatic heart disease. In developed countries it typically appears in women in their fourth or fifth decade. In developing nations, where the rheumatic process appears to be more aggressive, stenosis may develop in adolescence or early adulthood. As mitral stenosis worsens, a gradient develops between the left atrium and left ventricle during diastole. At the same time the stenotic valve impairs left ventricular filling, limiting cardiac output. The combination of pulmonary congestion caused by left atrial hypertension and diminished forward cardiac output caused by inflow obstruction mimics the hemodynamics of left ventricular failure, even though the left ventricle itself is usually spared from the rheumatic process, especially in developed countries.41 However, in approximately one third of patients left ventricular ejection performance is reduced despite no impairment in contractility.42 Reduced ejection fraction is caused by reduced preload from the impairment of left ventricular filling and from increased left ventricular afterload secondary to reflex systemic vasoconstriction in the face of decreased cardiac output. Ejection performance may return to normal shortly after mitral stenosis is relieved.43 763


Although the left ventricle is usually spared from direct involvement in this disease, the right ventricle experiences pressure overload because it supplies the hemodynamic force propelling blood across the stenotic mitral valve. Thus, as left atrial pressure rises, pulmonary pressure and right ventricular pressure also must increase, placing a pressure overload on the right ventricle. For reasons that are unclear, as the disease progresses reversible pulmonary vasoconstriction develops, leading to a worsening of pulmonary hypertension and eventually to right ventricular failure. Indications for surgery Grade B In most cases mitral stenosis can be relieved by balloon valvotomy, which offers results comparable to those of open commissurotomy, as shown in a randomized trial.44 Surgery is reserved for those cases in which valve anatomy is unfavorable for balloon valvotomy, or in which balloon valvotomy has been attempted and failed. Although in some instances open surgical commissurotomy can be successful even though balloon valvotomy was predicted to be unsuccessful, the unfavorable anatomy for balloon valvotomy will also be unfavorable for commissurotomy, necessitating valve replacement. Thus when surgery is anticipated, the risks and complications of a prosthesis should also be anticipated. The timing of surgery for mitral stenosis can largely be predicated on symptomatic status, as shown in Figure 52.3.45

100 G ro u

G ro

80

General population

ps II– I

ups

II (su

IV ( sur gica Me l) les di c en al gro ly tr up eat s B ed an dC )

% Survival

40

References

M (O edic les all e n y tr gr ea ou ted p D)

20

0 0

1

2

3

4

5

6

7

8

9

10 11 12

Figure 52.3 Comparison between surgical and medical treatment in patients with mitral stenosis. Groups II, III and IV, equivalent to NYHA classifications II, III and IV, are approximately similar to the groups represented by letters B, C and D, respectively. Class IV patients had better improved survival when treated surgically than did class D patients who were treated medically. Class II and III patients also had better survival when treated surgically than did the patients in groups B and C, although the difference is not as dramatic. (Reproduced with permission from Roy and Gopinath.45)

764

Summary Grade B In most cases mitral stenosis can be treated successfully with balloon valvotomy. However, if this procedure is unfeasible, open commissurotomy or valve replacement is indicated for those with NYHA symptoms greater than class II, or for the development of pulmonary hypertension.

rgica l)

(O

60

Once more than New York Heart Association (NYHA) class II symptoms develop, mortality increases abruptly and surgery should be performed before class III symptoms appear. In addition, some studies indicate that the presence of pulmonary hypertension substantially increases operative risk,35,46 and so surgery should be contemplated in patients who develop asymptomatic pulmonary hypertension (pulmonary artery systolic pressure 50 mmHg). When surgery precedes severe pulmonary hypertension, operative mortality, even with the insertion of a prosthesis, is 1–3%. The most difficult situation for timing surgery arises in the young woman who wishes to bear children. In such patients in whom balloon valvotomy has already been ruled out because of unfavorable valve anatomy, the choice of prosthetic valve becomes quite difficult. If a mechanical valve is placed it will require anticoagulation, which is problematic during pregnancy. Administration of warfarin causes a particularly high incidence of fetal malformation, especially when used during the first trimester. It can be substituted by daily injections of heparin, but serious thrombotic complications have occurred in such circumstances, suggesting that this therapy is inadequate in at least some cases.47 On the other hand, if a bioprosthesis is placed in a young woman it is likely to degenerate within a decade or sooner, forcing the patient to have a reoperation with its attendant increased surgical risk. There is no correct solution to this dilemma, and the prosthesis that is eventually inserted is chosen after lengthy consultation between patient and surgeon.

1.Carabello BA. Timing surgery for mitral regurgitation in asymptomatic patients. Choices Cardiol 1991;5:137–8. 2.Goldman ME, Mora F, Guarino T, Fuster V, Mindich BP. Mitral valvuloplasty is superior to valve replacement for preservation of left ventricular function. An intraoperative two-dimensional echocardiographic study. J Am Coll Cardiol 1987;10:568–75. 3.Carabello BA. Mitral regurgitation, Part 1: basic pathophysiologic principles. Mod Concepts Cardiovasc Dis 1988;57:53–8. 4.Cheitlin MD, Douglas PS, Parmley WW. 26th Bethesda Conference: recommendations for determining eligibility for competition in athletes with cardiovascular abnormalities. Task Force 2: acquired valvular heart disease. J Am Coll Cardiol 1994;24:874–80. 5.Urabe Y, Mann DL, Kent RL et al. Cellular and ventricular contractile dysfunction in experimental canine mitral regurgitation. Circ Res 1992;70:131–47.


6.Mulieri LA, Leavitt BJ, Martin BJ, Haeberle JR, Alpert NR. Myocardial force–frequency defect in mitral regurgitation heart failure is reversed by forskolin. Circulation 1993;88:2700–4. 7.Rushmer RF. Initial phase of ventricular systole: asynchronous contraction. Am J Physiol 1956;184:188–94. 8.Lillehei CW, Levy MJ, Bonnabeau RC. Mitral valve replacement with preservation of papillary muscles and chordae tendineae. J Thorac Cardiovasc Surg 1964;47:532–43. 9.Hansen DE, Cahill PD, DeCampli WM et al. Valvularventricular interaction: importance of the mitral apparatus in canine left ventricular systolic performance. Circulation 1986; 73:1310–20. 10.Rozich JD, Carabello BA, Usher BW et al. Mitral valve replacement with and without chordal preservation in patients with chronic mitral regurgitation. Mechanisms for differences in postoperative ejection performance. Circulation 1992; 86:1718–26. 11.David TE, Burns RJ, Bacchus CM, Druck MN. Mitral valve replacement for mitral regurgitation with and without preservation of chordae tendineae. J Thorac Cardiovasc Surg 1984;88:718–25. 12.Enriquez-Sarano M, Schaff HV, Orszulak TA et al. Valve repair improves the outcome of surgery for mitral regurgitation. A multivariate analysis. Circulation 1995;91:1022–8. 13.Duran CG, Pomar JL, Revuelta JM et al. Conservative operation for mitral insufficiency: critical analysis supported by postoperative hemodynamic studies in 72 patients. J Thorac Cardiovasc Surg 1980;79:326–37. 14.Horskotte D, Schulte HD, Bircks W, Strauer BE. The effect of chordal preservation on late outcome after mitral valve replacement: a randomized study. J Heart Valve Dis 1993;2:150–8. 15.Cohn LH, Couper GS, Aranki SF et al. Long-term results of mitral valve reconstruction for regurgitation of the myxomatous mitral valve. Cardiovasc Surg 1994;107:143–51. 16.Cosgrove DM, Chavez AM, Lytle BW et al. Results of mitral valve reconstruction. Circulation 1986;74(Suppl. I):I-82–I-87. 17.Wells FC. Conservation and surgical repair of the mitral valve. In: Wells FC, Shapiro LM, eds. Mitral valve disease, 2nd edn. Oxford: Butterworth–Heinemann, 1996. 18.Castello R, Fagan L Jr, Lenzen P, Pearson AC, Labovitz AJ. Comparison of transthoracic and transesophageal echocardiography for assessment of left-sided valve regurgitation. Am J Cardiol 1991;68:1677–80. 19.Smith MD, Kwan OL, Spain MG, DeMaria AN. Temporal variability of color Doppler jet areas in patients with aortic and mitral regurgitation. Am Heart J 1992;123:953–60. 20.Slater J, Gindea AJ, Freedberg RS et al. Comparison of cardiac catheterization and Doppler echocardiography in the decision to operate in aortic and mitral valve disease. J Am Coll Cardiol 1991;17:1026–36. 21.Recusani F, Bargiggia GS, Yoganathan AP et al. A new method for quantification of regurgitant flow rate using color Doppler flow imaging of the flow convergence region proximal to a discrete orifice: an in vitro study. Circulation 1991;83: 594–604. 22.Utsunomiya T, Ogawa T, Doshi R et al. Doppler color flow “proximal isovelocity surface area” method for estimating volume flow rate: effects of orifice shape and machine factors. J Am Coll Cardiol 1991;17:1103–11.

23.Vandervoort PM, Rivera JM, Mele D et al. Application of color Doppler flow mapping to calculate effective regurgitant orifice area: an in vitro study and initial clinical observations. Circulation 1993;88:1150–6. 24.Enriquez-Sarano M, Miller FA Jr, Hayes SN et al. Effective mitral regurgitant orifice area: clinical use and pitfalls of the proximal isovelocity surface area method. J Am Coll Cardiol 1995;25:703–9. 25.Yoran C, Yellin EL, Becker RM et al. Mechanisms of reduction of mitral regurgitation with vasodilator therapy. Am J Cardiol 1979;43:773–7. 26.Nishimura RA, Schaff HV, Shub C et al. Papillary muscle rupture complicating acute myocardial infarction: analysis of 17 patients. Am J Cardiol 1983;51:373–7. 27.Nishimura RA, Schaff HV, Gersh BJ, Holmes DR Jr, Tajik AJ. Early repair of mechanical complications after acute myocardial infarction. JAMA 1986;256:47–50. 28.Grigioni F, Enriquez-Sarano M, Ling LH et al. Sudden death in mitral regurgitation due to flail leaflet. J Am Coll Cardiol 1999;34:2078–85. 29.Ling LH, Enriquez-Sarano M, Seward JB et al. Clinical outcome of mitral regurgitation due to flail leaflet. N Engl J Med 1996;335:1417–23. 30.Eckberg DL, Gault JH, Bouchard RL, Karliner JS, Ross J Jr. Mechanics of left ventricular contraction in chronic severe mitral regurgitation. Circulation 1973;47:1252–9. 31.Wisenbaugh T, Spann JF, Carabello BA. Differences in myocardial performance and load between patients with similar amounts of chronic aortic versus chronic mitral regurgitation. J Am Coll Cardiol 1984;3:916–23. 32.Enriquez-Sarano M, Tajik AJ, Schaff HV et al. Echocardiographic prediction of survival after surgical correction of organic mitral regurgitation. Circulation 1994;90:830–7. 33.Wisenbaugh T, Skudicky D, Sareli P. Prediction of outcome after valve replacement for rheumatic mitral regurgitation in the era of chordal preservation. Circulation 1994;89:191–7. 34.Zile MR, Gaasch WH, Carroll JD, Levine HF. Chronic mitral regurgitation: predictive value of preoperative echocardiographic indexes of left ventricular function and wall stress. J Am Coll Cardiol 1984;3:235–42. 35.Crawford MH, Souchek J, Oprian CA et al. Determinants of survival and left ventricular performance after mitral valve replacement. Department of Veterans Affairs Cooperative Study on Valvular Heart Disease. Circulation 1990;81:1173–81. 36.Nair CK, Biddle WP, Kaneshige A et al. Ten-year experience with mitral valve replacement in the elderly. Am Heart J 1992;124:154–9. 37.Connolly MW, Gelbfish JS, Jacobowitz IJ et al. Surgical results for mitral regurgitation from coronary artery disease. J Thorac Cardiovasc Surg 1986;91:379–88. 38.Akins CW, Hilgenberg AD, Buckley MJ et al. Mitral valve reconstruction versus replacement for degenerative or ischemic mitral regurgitation. Ann Thorac Surg 1994;58: 668–75. 39.Schon HR, Schroter G, Barthel P, Schomig A. Quinapril therapy in patients with chronic mitral regurgitation. J Heart Valve Dis 1994;3:303–12. 40.Wisenbaugh T, Sinovich V, Dullabh A, Sareli P. Six month pilot study of captopril for mildly symptomatic, severe isolated mitral

765


and isolated aortic regurgitation. J Heart Valve Dis 1994;3:197–204. 41.Hildner FJ, Javier RP, Cohen LS et al. Myocardial dysfunction associated with valvular heart disease. Am J Cardiol 1972;30: 319–26. 42.Gash AK, Carabello BA, Cepin D, Spann JF. Left ventricular ejection performance and systolic muscle function in patients with mitral stenosis. Circulation 1983;67:148–54. 43.Liu C-P, Ting C-T, Yang T-M et al. Reduced left ventricular compliance in human mitral stenosis. Role of reversible internal constraint. Circulation 1992;85:1447–56.

766

44.Reyes VP, Raju BS, Wynne J et al. Percutaneous balloon valvuloplasty compared with open surgical commissurotomy for mitral stenosis. N Engl J Med 1994;331:961–7. 45.Roy SB, Gopinath N. Mitral stenosis. Circulation 1968; 38(Suppl. V):V68–76. 46.Ward C, Hancock BW. Extreme pulmonary hypertension caused by mitral valve disease. Natural history and results of surgery. Br Heart J 1975;37:74–8. 47.Sbarouni E, Oakley CM. Outcome of pregnancy in women with valve prostheses. Br Heart J 1994;71:196–201.

53

Indications for surgery in aortic valve disease Heidi M Connolly, Shahbudin H Rahimtoola

Evidence-based management of patients with aortic valve disease is limited by the absence of prospective randomized trials of surgery versus medical therapy. There is one prospective randomized trial evaluating patient outcome with use of a pharmacologic agent in patients with aortic valve regurgitation. However, evidence can also be obtained from retrospective studies. This evidence is extremely useful and important in the management of patients. Sir Thomas Lewis pointed out 80 years ago the inadequacy of knowledge of prognosis in patients with heart disease. He proposed a system for prospective follow up of patients, which we now call “databases” or “registries”. The latter are, of course, the major evidence used in this chapter to delineate the indications for surgery. The American College of Cardiology and American Heart Association published Guidelines for the management of patients with valvular heart disease.1 This document has provided an important framework upon which clinical decisions can be based.

Aortic valve stenosis Etiology A wide variety of disorders may produce aortic valve obstruction;2 however, those that result in severe stenosis in adults are: ● ●

●

congenital acquired ● calcific (degenerative) ● autoimmune rheumatic.

The most common cause of aortic stenosis in younger adults is a congenital bicuspid valve, which is found in 1–2% of the general population. Rheumatic heart disease is still common in developing countries. In most patients aged 40 years, the severely stenotic valve is calcified. In patients aged 65 years, 90% of severely stenotic valves are tricuspid. Non-rheumatic calcified valves are thought to be “degenerative” but recent data suggest that it may be the result of an autoimmune reaction to antigens present in the

valve;3 and that the initial process may be an atherosclerotic lesion.4,5 Grading the degree of stenosis The natural history of aortic stenosis is variable depending on the degree of stenosis and the rate at which it progresses. Cardiac catheterization and echocardiographic–Doppler ultrasound studies indicate the systolic pressure gradient increases on an average by 10–15 mmHg per year. The 10–15 mmHg increase is a linearized value whereas the increase is not linear but a stepwise function with periods of steady state interspersed by an increase in gradient. The range of progression is also wide. Recent data suggest that the progression of aortic stenosis may be related to cardiovascular risk factors.6 The systolic gradient across the stenotic aortic valve is dependent on the following: ●

● ●

the stroke volume (not the cardiac output because the gradient and valve area are a per beat, and not a per minute, function) the systolic ejection period systolic pressure in the ascending aorta.

The stenotic valve area is inversely related to the square root of the mean systolic gradient. Thus, measurement of valve area is an important part of the assessment of the severity of aortic valve stenosis. The valve area may decrease by as much as 0·12 0·19 cm2 per year.7 Valve area is related to the body surface area and increases in larger individuals, probably because of the need for a larger stroke volume and cardiac output. The normal aortic valve area ranges from 3 to 4 cm2. It is reduced to half its size before a systolic gradient occurs.8 The orifice area has to be reduced to one third of its size before significant hemodynamic changes are seen;9 gradients increase precipitously after that. The obvious clinical problem is that in an individual patient with aortic stenosis one usually does not know the valve area prior to the onset of disease. Echocardiography is usually the initial procedure used to confirm the presence and determine the severity of aortic valve stenosis.10 In an experienced center the severity of aortic stenosis determined by Doppler echocardiography correlates reasonably well with the severity determined by 767


cardiac catheterization.11 A comprehensive echocardiographic examination in aortic valve stenosis should include assessment of the aortic valve peak and mean gradient as well as aortic valve area.12 When the clinical picture does not correlate with the hemodynamic data obtained by Doppler echocardiography, re-evaluation by cardiac catheterization is indicated. The outcome of patients with severe aortic valve stenosis was described by Ross and Braunwald13 after review of seven autopsy studies published before 1955, and Horstkotte and Loogen14 reported on 35 patients (10 of whom were asymptomatic) with aortic valve area of 0·8 cm2 by cardiac catheterization who refused surgery. The findings are shown in Table 53.1. The mortality of symptomatic patients with “severe” aortic stenosis from eight studies15 is given in Table 53.2.

Valvular Heart Disease.1 In this document aortic valve stenosis is defined as mild when the aortic valve area was 1·5 cm2. In two studies, patients with aortic valve area 1·5 cm2 by catheterization had no mortality on follow up. At the end of 10 years, in one study 8% had severe stenosis, and in the other 15% had a cardiac event. At the end of 20 years, aortic stenosis had become severe in only 20% and continued to be mild in 63%.14,15

Mild aortic stenosis

Severe aortic stenosis

The classification of the severity of aortic valve stenosis was defined in the guidelines provided by the Committee on

Several criteria have been used to define severe aortic stenosis. The guidelines provided by the Committee on Valvular Heart Disease1 describes severe aortic valve stenosis as an aortic valve area 1·0 cm2 and a mean aortic pressure gradient, in the setting of normal cardiac output, of 50 mmHg. This definition was supported by data from a large multicenter database (492 patients) which suggested that the 1 year mortality of those with aortic valve areas after catheter balloon valvuloplasty for calcific aortic stenosis of 0·7 cm2 versus that of those with valve areas 0·7 cm2 was 37% versus 42%, respectively.16 Kennedy and coworkers17 reported on 66 patients with aortic valve areas of 0·7–1·2 cm2 (0·920·15 cm2), normal left ventricular volumes and ejection fraction, whose average age was 67 years. In an average follow up of 35 months, 21% died and 32% had valve replacement; at 4 years, the actuarial incidence of death or valve replacement was 41%.17 Thus, these studies show that patients with aortic valve areas of 0·7–1·0 cm2 have an outcome without valve replacement that is not benign, and is not consonant with moderate

Table 53.1 Survival, according to symptoms, of patients with “severe” aortic stenosis Symptoms

Overall Anginac Syncope Heart failure

Average survival Autopsy dataa (years)

Post cardiac catheterizationb (months)

3 5 3 2

23 45 27 11

a

Data of Ross and Braunwald.13 Data of Horstkotte and Loogen.14 c Angina in patients with aortic stenosis occurs even in those without associated obstructive CAD. b

Moderate aortic stenosis Moderate aortic valve stenosis is defined as a valve area of 1·0–1·5 cm2. In one study in which patients were followed after cardiac catheterization, the 1 year and 10 year mortality was 3% and 15%, respectively; and at 10 years 65% of patients had had a cardiac event.15

Table 53.2 Mortality of symptomatic patients with “severe” aortic stenosis15 Authors

Year of publication

23

Frank et al Rapaport84 Chizner et al 85 Schwarz et al 30 O’Keefe et al 86 Turina et al 87 Kelly et al 88 Horstkotte et al 14

768

1973 1975 1980 1982 1987 1987 1988 1988

Patients (n)

Mortality follow up time (years) 1

2

15 23 19 50 50 39 35

26% 43% 40% 38%

3

5

10

36%

52% 62% 64%

90% 80%

48% 63%

79% 75%

82%

11

94%

Indications for surgery in aortic valve disease

stenosis; these patients should be considered as having severe stenosis. Since gradients are frequently measured initially by Doppler ultrasound, a suggested conservative guideline for relating Doppler ultrasound gradient to severity of aortic stenosis (AS) in adults with normal cardiac output and normal average heart rate is shown in Table 53.3. A suggested grading of the degree of aortic stenosis is given in Table 53.4. Table 53.3 Doppler ultrasound gradient as an indicator of severe aortic stenosis (AS) Peak gradient

Mean gradient

AS severe

80 mmHg 60–79 mmHg 60 mmHg

70 mmHg 50–69 mmHg 50 mmHg

High likely Probable Uncertain

From Rahimtoola,15 with permission

Table 53.4 Grading of stenosis by aortic valve area (AVA) Aortic stenosis

AVA (cm2)

AVA index (cm2/m2)

Mild Moderate Severea

1·5 1·1–1·5 1·0

0·9 0·6–0·9 0·6

a

Patients with AVAs that are at borderline values between the moderate and severe grades (0·9–1·1 cm2; 0·55– 0·65 cm2/m2) should be individually considered. From Rahimtoola15 with permission

Natural history The duration of the asymptomatic period after the development of severe aortic stenosis is uncertain. In a study of asymptomatic patients with varying degrees of severity of aortic stenosis, 21% of 143 patients18 with a mean age of 72 years required valve replacement within 3 months of evaluation at a referral center. At 2 years the mortality was 10% and the event rate (death/valve replacement) in the remaining patients was 26%. Moreover, it is important to recognize that most patients in this study had only moderate aortic stenosis. In another study of 123 asymptomatic adults,7 also with varying grades of severity of aortic stenosis aged 6316 years, only the actuarial probability of death or aortic valve surgery is provided. It was 75% at 1 year, 388% at 3 years and 7410% at 5 years. The event rate at 2 years for aortic jet velocity by Doppler ultrasound of 4·0 m/s (peak gradient by Doppler ultrasound 64 mmHg) was 79 18%, for a velocity of 3·0–4·0 m/s (peak gradient 36–64 mmHg) was 6613%, and for a velocity of 3·0 m/s

(peak gradient of 36 mmHg) was 1616%.7 Aortic jet velocity is influenced by the same parameters as aortic valve gradient (see above). Thus, the duration of the asymptomatic period, particularly in those aged 60 years, is probably very short.19,20 Paul Dudley White in 195121 credited the first recorded occurrence of sudden death to T Bonet in 1679.22 In the past 70 years the reported incidence of sudden death in eight series has ranged from 1 to 21%. Ross and Braunwald,13 after reviewing seven autopsy series published before 1955, concluded the incidence was 3–5%. The incidence in asymptomatic adult patients has been 33% (one in three)23 and 30% (three of ten).14 This information is difficult to use in clinical decision making because important data are not available – that is, the incidence by actuarial analysis of sudden death in a significant number of asymptomatic patients with severe stenosis. It is reasonable to conclude that the true incidence of sudden death in adults with severe aortic valve stenosis is unknown and that sudden death usually occurs after the onset of symptoms, however minor or minimal the symptoms may be. The incidence of sudden death is believed to be higher in children. The development of symptoms of angina, syncope, or heart failure, changes the prognosis of the patient with aortic valve stenosis. Average survival after the onset of symptoms is 2–3 years. Nearly 80% of asymptomatic patients with peak aortic valve velocity measured by Doppler echocardiography 4 m/s develop symptoms within 3 years, and therefore careful clinical monitoring for the development of symptoms and progressive disease is indicated. Management Patients with valvular heart disease need antibiotic prophylaxis against infective endocarditis; those with rheumatic valves need additional antibiotic prophylaxis against recurrences of rheumatic fever.24 Grade A Surgery is recommended in those with severe valve stenosis and is the only specific and direct therapy for most adults with severe aortic stenosis. Rarely, in young patients, the aortic valve is suitable for balloon or surgical valvotomy. In most adults, surgery for aortic stenosis means valve replacement.24,25 Grade B The operative mortality of valve replacement is 5%.25–27 In those without associated coronary artery disease, heart failure or other comorbid conditions, it is 2% in experienced and skilled centers.28 Aortic valve replacement in conjunction with coronary artery bypass carries a surgical mortality of about 7%.27 The operative mortality in those 70 years and in octogenarians is much higher, averaging 8% for valve replacement and 13% for those undergoing valve replacement and associated coronary bypass surgery;25 however, operative mortality in these patients is also dependent on the associated factors listed above.29 769


Patients with associated coronary artery disease (CAD) should have coronary bypass surgery at the same time as valve replacement, because it results in a lower operative mortality (4·0% v 9·4%) and better 10 year survival (49% v 36%).28 This was in spite of the fact that those who underwent coronary bypass surgery had more CAD (34% had three vessel disease, 11% had left main artery disease, and 38% had single vessel disease) than those who did not undergo coronary bypass surgery (13% had three vessel disease, 1% had left main disease, and 65% had single vessel disease).28 Although this approach to CAD is generally approved, there are no randomized trials to support these recommendations. The presence of CAD, its site and severity can be estimated only by selective coronary angiography, which should be performed in all patients 35 years of age or older who are being considered for aortic valve surgery, and in those aged (A)

35 years if they have left ventricular dysfunction, symptoms or signs suggesting CAD, or they have two or more risk factors for premature CAD (excluding gender).25 The incidence of associated CAD will vary considerably depending on the prevalence of CAD in the population;15,24 in general, in persons 50 years of age or older it is about 50%.25 In severe aortic stenosis, valve replacement results in an improvement of survival (Figure 53.1) even if they have normal left ventricular function preoperatively.14,30 Normal preoperative left ventricular function remains normal postoperatively if perioperative myocardial damage has not occurred.31 Left ventricular hypertrophy regresses toward normal;31,32 after 2 years, the regression continues at a slower rate up to 10 years after valve replacement.32 In patients with excessive preoperative left ventricular hypertrophy,33 the hypertrophy may regress slowly or not (B)

Aortic stenosis 100

BSA 80

80

Cumulative survival (%)

% Survival

60

40

60

P < 0·000001 40

χ2 = 23·5

20

P < 0·001 20 < 0·05

< 0·001

0

NH 1

2

3

4

5

Years 125

87

51

35

9

0

19

8

2

1

0

5

2

4

6

8

10

Time (years)

Figure 53.1 There are no prospective randomized trials of aortic valve replacement in severe aortic stenosis (AS), and there are unlikely to be any in the near future. Two studies have compared the results of aortic valve replacement with medical treatment in their own center during the same time period in symptomatic patients with normal left ventricular systolic pump function. (A) Patients who had valve replacement (closed circles) had a much better survival than those treated medically (open circles). (From Schwarz et al 30 with permission.) (B) Patients who were treated with valve replacement (BSA) had a better survival than those treated medically (NH). (From Horstkotte and Loogen14 with permission.) These differences in survival between those treated medically and surgically are so large that there is a great deal of confidence that aortic valve replacement significantly improves the survival of those with severe aortic stenosis. Grade A

770


100 AS 65

94%

90 AS < 65 81%

80 Relative survival (%)

at all. Preoperatively, these patients have a small left ventricular cavity, severe increase in wall thickness, and “supernormal” ejection fraction; this occurs in 42% of women and 14% of men in those aged 60 years.33 After valve replacement their clinical picture often resembles that of hypertrophic cardiomyopathy without outflow obstruction, which is a difficult clinical condition to treat, both in the early postoperative period and after hospital discharge;33 therefore, surgery should be performed prior to development of excessive hypertrophy. Surviving patients are functionally improved.25 After valve replacement, the 10 year survival is 60% and 15 year survival is about 45%.25,34 One half or more of the late deaths are not related to the prosthesis but to associated cardiac abnormalities and other comorbid conditions.34 Thus, the late survival will vary in different subgroups of patients. The older patients (60 years) have a 12 year actuarial survival of 60%.35 Relative survival refers to survival of patients compared to age- and gender-matched people in the population. The relative 10 year survival after surgery is significantly better in those aged 65 than in those aged 65 years (94% v 81% respectively, Figure 53.2);36 the 94% relative survival is not significantly different from the 100% relative survival. Thus, surgery should not be denied to those 60–65 years old and should be performed early.25,35–37 Patients who present with heart failure related to aortic valve stenosis should undergo surgery as soon as possible. Medical treatment in hospital prior to surgery is reasonable but ACE inhibitors should be used with great caution in such patients, and in such a dosage that hypotension and significant fall of blood pressure is avoided. They should not be used if the patient is hypotensive. If heart failure does not respond satisfactorily and rapidly to medical therapy, surgery becomes a matter of considerable urgency.25 Catheter balloon valvuloplasty has a very limited role in adults with calcific aortic stenosis and carries a risk of 10%. In addition, restenosis and clinical deterioration occur within 6 to 12 months. In adults with aortic stenosis, balloon valvuloplasty is not a substitute for valve replacement but can be a bridge procedure in selected patients.38 It usually improves patients’ hemodynamics and may make them better candidates for valve replacement. The operative mortality for patients with heart failure has declined: 25 years ago the operative mortality was 20%,39 but in the current era it is 10%.40 Although this is higher than in patients without heart failure, the risk is justified, because late survival in those who survive the operation is excellent and is far superior to that which can be expected with medical therapy. The 7 year survival of patients who survive operation is 84%.41 The 5 year survival in those without associated CAD is greater than in those with CAD (69% v 39%, P 0·02).40 Left ventricular function improves in most patients provided there has been no perioperative myocardial damage and becomes normal in two thirds of the patients, unless there was irreversible preoperative

70

60

50

40

0

5

10 Years Post-op

Figure 53.2 Data from the Karolinska Institute in Sweden has provided an interesting perspective on the long-term survival after valve replacement in patients with aortic stenosis (AS) aged 65 years. They have examined the relative survival – compared the survival of the patient who has undergone aortic valve replacement with another age and sex matched person in the same population. Actuarial survival 95% confidence interval is shown. Patients under the age of 65 had a relative survival of 81% which is significantly lower than 100%, and is also lower than that of those aged 65 years. On the other hand, patients who underwent valve replacement at age 65 had a relative survival of 94% at the end of 10 years and this was not significantly different from 100%. These data indicate that survival following valve replacement for aortic stenosis in patients aged 65 is not significantly different from age- and sex-matched individuals in the population without aortic stenosis; and the late relative survival of patients aged 65 years is much better than that of patients aged 65. (From Lindblom et al 36 with permission.)

myocardial damage (Figure 53.3).39,40 In addition, the operative survivors are functionally much improved.39,40 Left ventricular hypertrophy and left ventricular dilation, if present preoperatively, regress toward normal.39 Despite the excellent results of valve replacement in patients with severe aortic stenosis who are in heart failure, these results are not as good as for those who are not in heart failure; therefore, it is important to recognize that surgery should not be delayed until heart failure develops. Grade B Six per cent of older patients with aortic stenosis present in cardiogenic shock.38 The hospital mortality in such patients is near 50%. The subsequent mortality is also very high if the patients have not had their aortic stenosis relieved.38 Thus, these patients need to be managed aggressively by emergency surgery with or without catheter balloon valvuloplasty as a “bridge” procedure.38 771


1·0

Ejection fraction P < 0·001

0·9 Mean ± SE

Box 53.1 Results of valve replacement in patients with severe aortic valve stenosis ● Improved symptoms and survival in symptomatic patients, especially in those with left ventricular systolic dysfunction, clinical heart failure, and in those aged 65 years ● Improvement in left ventricular systolic dysfunction, which normalizes in two thirds of patients ● Regression of left ventricular hypertrophy ● Improvement in functional class, more marked in those with severe symptoms preoperatively

0·8

0·7

0·6

Box 53.2 Factors predictive of a less favorable outcome ● Extent and severity of associated comorbid conditions ● Presence and severity of clinical heart failure preoperatively ● Severe associated coronary artery disease ● Severity of depression of preoperative left ventricular ejection fraction ● Duration of preoperative left ventricular systolic dysfunction ● Extent of preoperative irreversible myocardial damage ● Skill and experience of operating and other associated professional teams ● Extent of perioperative myocardial damage ● Complications of a prosthetic heart valve

0·5

0·4

0·3

0·2

0·1

Pre-op

Post-op

Peri-op MI and late CHB Post-op: Perivalvular aortic incompetence

Figure 53.3 Examination of changes in LVEF in each individual patient among those who had left ventricular systolic dysfunction and clinical heart failure. After valve replacement the LVEF improved from 0·34 to 0·63. All but one patient showed an improvement in LVEF; the only patient who showed deterioration in ejection fraction suffered a perioperative myocardial infarction and had a complete heart block; and the only patient who showed only a small increase in ejection fraction had had a myocardial infarct prior to valve replacement. Note that

772

Boxes 53.1 and 53.2 summarize the results of valve replacement in those with severe aortic stenosis and the factors predictive of a worse postoperative survival, less recovery of left ventricular function, and less improvement of symptoms in those with severe aortic stenosis and preoperative left ventricular systolic dysfunction.15,25,29–32,34–36,39–41

Patients with severe left ventricular dysfunction, low aortic valve gradient, and small calculated aortic valve area represent a difficult patient population. There is controversy regarding the best management of these patients, in part related to the difficulty differentiating patients with true severe aortic valve stenosis from patients with moderate aortic valve stenosis and severe left ventricular dysfunction. Differentiating these two patient groups may have an important impact on the management decision and the operative outcome. Thus, patients with low gradient aortic valve stenosis should not be denied aortic valve replacement. A recent series confirms that ejection fraction normalized in two thirds of the patients and, in the two patients with the lowest ejection fraction (0·18 and 0·19), ejection fraction normalized in both. These data indicate that there is probably no lower limit of ejection fraction at which time these patients become inoperable. This also indicates that the lower the ejection fraction, the more urgent the need for valve replacement. (From Smith et al 39 with permission.)


the surgical mortality is high and late survival lower than expected. Importantly however, most survivors experienced improvement in functional class and ejection fraction.42 A small gradient across the valve may be associated with a small calculated aortic valve area that would be in a range indicating severe aortic stenosis. There are at least two possible causes for this clinical circumstance. First, there is a small or reduced stroke volume and a normal or near normal systolic ejection time; thus, the gradient is small and the calculated aortic valve area correctly indicates severe aortic stenosis. The second consideration is that the stroke volume is reduced, and thus the valve needs to open only to a small extent to allow the left ventricle to eject the small stroke volume. The calculated aortic valve area accurately reflects the extent to which the valve has opened but overestimates the severity of aortic stenosis. Use of a provocative test using an inotropic agent, such as dobutamine,43,44,45 may allow one to make the correct differentiation between the two. Dobutamine increases systolic flow per second owing to increases in stroke volume or shortening of ejection time or both. In the first circumstance described above, dobutamine will result in an increase in gradient but the calculated valve area remains more or less unchanged. On the other hand, in the second circumstance described above, the gradient may or may not increase with dobutamine but the calculated valve area increases significantly, indicating that the stenosis is not severe. When the dobutamine test is used, it is important to measure cardiac output and simultaneous left ventricular and aortic pressures both before and during dobutamine infusion. Alternatively, the gradient and valve area may be assessed by echocardiography/Doppler during dobutamine infusion; however, one needs to be certain that cardiac output has increased significantly with dobutamine. Grade B Surgery should be advised for the symptomatic patient who has severe aortic stenosis. In young patients, if the valve is pliable and mobile, simple balloon valvuloplasty or surgical commissurotomy may be feasible. Older patients and even young patients with calcified, rigid valves will require valve replacement. In view of the dismal natural history of symptomatic patients with severe aortic stenosis, the excellent outcome after surgery, and the uncertain natural history of the asymptomatic patient, it is reasonable to recommend aortic valve replacement in select asymptomatic patients in centers with the appropriate skill and experience. The combined risk of surgery and late complications of a valve prosthesis must be weighed against the risk of sudden death. There is no consensus about valve replacement in the truly asymptomatic patient. Clearly, if the patient has left ventricular dysfunction, obstructive CAD or other valve disease that needs surgery, and has severe aortic stenosis, then aortic valve replacement should be performed. Some would recommend valve replacement in all asymptomatic patients with severe aortic stenosis, while others would recommend it in all those with

aortic valve area of 0·70 cm2 and in selected patients only with aortic valve area of 0·71–1·0 cm2. Exercise testing should be avoided in symptomatic patients with aortic stenosis but has been used by some cardiologists to help determine which patients with asymptomatic aortic stenosis should be referred for aortic valve replacement.19 In a small series, Amato and colleagues reported no serious exercise-related complications. During follow up, 6% of the asymptomatic patients (4/66) experienced sudden death; all had a positive exercise test and an aortic valve area of 0·6 cm2. The exercise test was considered positive if there was a horizontal or down sloping ST segment depression of 1 mm in men or 2 mm in women, or an up sloping ST segment depression of 3 mm in men, measured 0·08 seconds after the J point. The exercise test was also considered positive if precordial chest pain or near syncope occurred, if the ECG showed a complex ventricular arrhythmia, or if systolic blood pressure failed to rise by 20 mmHg during exercise compared with baseline. Grade B It must be emphasized that this is a controversial issue. Some cardiologists advise against exercise testing in any patient with severe aortic valve stenosis, especially when the extent of coronary artery disease is not known. Recommendations: aortic valve replacement/repair in severe aortic stenosis1 Indication Class ● Symptomatic patients I ● Asymptomatic patients with: ● associated significantly obstructed I CAD needing surgery ● other valve or aortic disease needing I surgery ● left ventricular systolic dysfunction IIa ● aged 60–65 years IIa ● abnormal response to exercise IIa ● severe left ventricular hypertrophy IIb (15 mm) ● significant arrhythmias IIb ● left ventricular dysfunction on exercise IIb ● Prevention of sudden death III in asymptomatic patients CAD, coronary artery disease Class I: Conditions for which there is evidence and/or general agreement that a given procedure or treatment is useful and effective. Class II: Conditions for which there is conflicting evidence and/or a divergence of opinion about the usefullness/efficacy of a procedure or treatment. IIa: Weight of evidence or opinion is in favor of usefullness/efficacy. IIb: Usefullness/efficacy is less well established by evidence/opinion. Class III: Conditions for which there is evidence and/or general agreement that the procedure/treatment is not useful, and in some cases, may be harmful.

773


Chronic aortic valve regurgitation

not common, and the outcome in syphilitic aortic regurgitation may be more benign in the current era.

Etiology The causes of chronic aortic regurgitation are:46

Era of rheumatic fever/carditis

aortic root/annular dilation congenital bicuspid valve previous infective endocarditis rheumatic in association with other diseases. In developed countries, aortic root/annular dilation and congenital bicuspid valve are the commonest causes of severe chronic aortic regurgitation.

Although the incidence of rheumatic valve disease is low in developed countries, rheumatic heart disease remains the most common form of valve disease in many parts of the world. Moreover, some people now domiciled in the developed world have had their initial attack(s) of acute rheumatic fever whilst living in less developed countries. The detection of a murmur after the episode of acute rheumatic fever averages 10 years.49 The average interval from detection of murmur to development of symptoms is 10 years and the percentage of patients remaining symptomfree 10 years after detection of the murmur is 50%.49 In 1971, Spagnuolo and coworkers50 reported the 15 year actuarial follow up of 174 young people who had a median follow up of 10 years. Patients were considered to be in a cumulative high-risk group if they had systolic blood pressure 140 mmHg and/or diastolic blood pressure 40 mmHg, moderate or marked left ventricular enlargement on chest radiography, and two of three ECG abnormalities (S in V2R in V5 51 mm, ST segment depression or T wave inversion in left ventricular leads). The group’s findings are summarized in Table 53.5.

● ● ● ● ●

Natural history During the first world war, Sir Thomas Lewis and his colleagues47 at Hampstead and Colchester Military Hospitals reported to the Medical Research Council highlighting the inadequacy of the knowledge of heart disease, especially from the standpoint of prognosis. Sir Thomas Lewis proposed a system,48 subsequently called “after histories”,48 which was a prospective follow up of patients. All patients in RT Grant’s “after histories”48 had valvular heart disease – most had aortic regurgitation – in which the patient characteristics were defined and described in detail, particularly by the degree of cardiac enlargement and the grade of cardiac failure. This probably was the start of databases or registries in cardiovascular medicine. Chronic aortic valve regurgitation is a condition of combined volume and pressure overload. With progression of the disease, compensatory hypertrophy and recruitment of preload reserve permit the left ventricle to maintain a normal ejection performance despite the elevated afterload. The majority of patients remain asymptomatic throughout the compensated phase, which may last decades. The natural history of chronic aortic valve regurgitation can be considered by three different eras: the era of syphilis, the era of rheumatic fever/carditis, and the current era of noninvasive quantification of left ventricular function.

Table 53.5 Reported outcome in 174 young people followed for a mean of 10 years after an episode of rheumatic fever Symptoms/outcome ●

●

Era of syphilis The data are from the 1930s and 1940s, and thus largely prior to availability of antibiotics.49 The duration from syphilis infection to death was 20 years. The duration of the asymptomatic period after aortic regurgitation was 5 years in 60% of patients; and the 5 year survival was 95%. Once symptoms had developed, the 10 year survival ranged from 40 to 60%. Heart failure was associated with a 1 year survival of 30–50%, and 10 year survival of 6%. In a study of 161 patients reported in 1935, the 10 year survival after heart failure had developed was 34% but was 66% in those treated with arsenic.49 Syphilis still occurs, but current therapy of syphilis is cheap and efficacious if diagnosed early. Syphilitic aortic regurgitation is 774

Cumulative high-risk group ● mortality ● angina ● heart failure ● mortality or angina or heart failure Cumulative low-risk group ● Mortality ● ●

●

Angina Heart failure Mortality or angina or heart failure

Outcome (years) 6 7 6 6

6 15 5 6 15 15

%

30 60 60 87

0 5a 2 2 5 8

a

The one patient (of the 72 patients) in this subgroup who died had developed two of the three risk factors.

In 1973, Goldschlager and coworkers51 reported on the duration of the asymptomatic period in 126 patients with varied etiology (Table 53.6).


Table 53.6 Asymptomatic period observed in 126 patients following an episode of rheumatic fever

Table 53.7 Outcomes of patients with severe aortic regurgitation

Age group (years)

Patients symptomatic at 10 yearsa (%)

Outcome

11–20 21–30 31–40 41–50 51–60 61–70

0 24 35 71 77 89

a

Symptoms were those of dyspnea, fatigue and, less frequently, chest pain and palpitations. Patients deteriorated from NYHA functional Class I to Classes II, III, or IV. From Goldschlager et al.50

Current era In the current era, patients have been followed after noninvasive tests (echocardiography/Doppler ultrasound, radionuclide LVEF) or after invasive studies (cardiac catheterization or angiography). Reported outcomes are shown in Table 53.7. As outlined in Table 53.7,52–58,64 the natural history of patients with chronic aortic valve regurgitation depends on the presence or absence of symptoms and on the status of the left ventricle. In asymptomatic patients with normal left ventricular function, data would suggest the progression to symptoms and or left ventricular systolic dysfunction in approximately 4% per year. Sudden death occurs very rarely, 0·1% per year, and asymptomatic left ventricular dysfunction occurs at a rate of 1–3% per year, depending on the frequency of follow up. There are limited data on asymptomatic patients with reduced left ventricular systolic function. However, available data would suggest that most of these patients will develop symptoms warranting surgery within two to three years, at an average rate of 25% per year. Limited data are available on the natural history of symptomatic patients with severe aortic valve regurgitation. These patients have a poor prognosis despite medical therapy, with reported mortality rates of 10 and 20% per year in patients with angina and heart failure, respectively. Important limitations of some of the studies in the literature must be kept in mind. For example, the “natural history” group in one study was composed of several subsets of patients53 and 36% of this group were on medications for symptoms. Another concern is the true rate of the development of asymptomatic left ventricular dysfunction.54 At least 25% of patients who develop left ventricular systolic dysfunction do so before they have symptoms, thus emphasizing

Asymptomatic patients with normal left venticular systolic function52–59 progression to symptoms and/or left ventricular systolic dysfunction progression to asymptomatic left ventricular dysfunction: follow up at 12 month intervalsa54 follow up at 6 month intervalsa58 Sudden death Asymptomatic patients with left ventricular systolic dysfunction60–61 progression to cardiac symptoms Symptomatic patients50,62–64 mortality rate angina heart failure a

Incidence

2·4–5·7% per year (average 3·8% per year)

0·9% per year 3·4% per year 0·1% per year

25% per year average 10% per year 10% per year 20% per year

See text for details.

the need for quantitative assessment of left ventricular systolic function at follow up in asymptomatic patients with severe aortic regurgitation and normal left ventricular systolic function. More recent studies indicate a poor outcome of symptomatic patients with medical therapy, even among those with preserved systolic function (Table 53.8).57,65 Sir William Broadbent66 stated 100 years ago that “The age of the patient at the time when the lesion is acquired is

Table 53.8 Likelihood of symptoms or left ventricular dysfunction or death ●

●

Left ventricular end-diastolic dimension 70 mm 70 mm Left ventricular end-systolic dimension 50 mm End-systolic dimension 25 mm/m2 40–49 mm 40 mm

10% per year 2% per year 19% per year 8% per year 6% per year 0% per year

775


the most important consideration in prognosis …”. In asymptomatic patients with normal left ventricular systolic function, the independent predictors of symptoms, left ventricular systolic dysfunction, and death by multivariate analysis were: older age, decreasing resting LVEF, and left ventricular dimension on M-mode echocardiography.54 However, in many of these patients, M-mode images were not obtained from two dimensionally directed echocardiograms. Very importantly, most of these dimensions were obtained in the United States, and US women have smaller left ventricular dimensions than men, even when they become symptomatic.67 Thus, it is unlikely that the above criteria apply to women and almost certainly will not be applicable to populations of smaller body size, for example, Asians, Latin Americans, sub-Saharan Africans, and Orientals. The left ventricular dimension should be corrected to body surface area.68 Patients also develop symptoms and/or left ventricular systolic dysfunction at a faster rate if their initial left ventricular end-diastolic volume is 150 ml/m2 when compared to those with volumes 150 ml/m2.53 Older age also appears to increase the annual mortality.68 Patients with severe ventricular dilation when exercised have shown mean pulmonary artery wedge pressure 20 mmHg and/or exercise ejection fraction 0·50, and such patients have demonstrated reduced exercise capacity, with reduced maximum V O2.69,70 Patients who present with ventricular tachycardia, ventricular fibrillation or syncope and have inducible ventricular tachycardia on electrophysiologic studies have an 80% probability of a serious arrhythmic event up to 4 years of follow up, versus 47% in those in whom ventricular tachycardia could not be induced (P 0·005).71 Acute severe aortic valve regurgitation usually causes sudden severe symptoms of heart failure or cardiogenic shock. The sudden large regurgitant volume load is imposed on a normal size left ventricle causing marked elevation in left ventricular end-diastolic pressure and left atrial pressure. Echocardiography is invaluable in determining the severity and etiology of aortic valve regurgitation.10 The etiology of acute aortic valve regurgitation may have an important impact on the treatment, which is usually emergency surgery. Management options Angina is a result of a relative reduction of myocardial blood flow because of an increased need or associated obstructive CAD or both.25 It does not respond to nitrates as well as in aortic stenosis. The options are to reduce the amount of aortic regurgitation and/or to revascularize the myocardium by coronary bypass surgery or by percutaneous catheter techniques. Clinical heart failure is treated with the traditional first-line triple therapy, that is, digitalis, diuretics, and ACE inhibitors. Parenteral inotropic and vasodilator therapy may be needed for those in severe heart failure.72 The only 776

direct method(s) to reduce the amount of regurgitation is by arterial dilators73 and valve surgery – that is, valve replacement or valve repair. Arterial dilators In chronic aortic valve regurgitation, therapy with vasodilating agents is designed to improve forward stroke volume and reduce regurgitant volume. These effects should translate into reductions in left ventricular end-diastolic volume, wall stress, and afterload, resulting in preservation of left ventricular systolic function and reduction in left ventricular mass. These effects have been observed in small numbers of patients receiving hydralazine.73 In a trial of 80 patients over 2 years74 in which 36% of patients were symptomatic (NYHA class II) and were being treated with digitalis and diuretics, hydralazine produced very minor improvements of left ventricular size and function.74 Side effects associated with long-term use of hydralazine seriously impaired compliance and only 46% of the patients completed the trial. Hydralazine is rarely used currently. Occasionally it is used for a short period of time, to tide the patient over an acute reversible complication or in preparation for elective surgery in selected patients with left ventricular dysfunction. Less consistent results have been reported with ACE inhibitors, depending on the degree of reduction in arterial pressure and end-diastolic volume. In an acute study in the catheterization laboratory, 20 patients were randomized to either oral nifedipine or oral captopril.75 Nifedipine produced a reduction of regurgitant fraction but captopril did not. Nifedipine produced a greater increase of forward stroke volume and cardiac output and a greater fall of systemic vascular resistance. This study showed that, acutely, nifedipine was superior to an ACE inhibitor. A shortterm 6 month randomized trial of a small number of patients showed that the results with captopril were similar to placebo – that is, there were no significant changes in M-mode echocardiographic left ventricular dimensions.76 A randomized trial of 72 patients for 12 months of longacting nifedipine showed statistically significant reductions of left ventricular end-diastolic volume index and left ventricular mass, and increase of LVEF.58 The role of long-acting nifedipine on patient outcome has been evaluated in a prospective, randomized trial of 143 asymptomatic patients with chronic, severe aortic valve regurgitation, and normal left ventricular systolic function; 69 patients were randomized to long-acting nifedipine and 74 patients to digoxin. The patients were evaluated at 6 month intervals for medication complication and had a history, physical examination, ECG, chest radiograph, and echocardiographic/ Doppler study. Two independent blinded observers read each echocardiographic/Doppler study. Criteria for valve replacement were established prior to the start of the study. If left ventricular dysfunction developed, this had to be


confirmed by a repeat echocardiographic/Doppler study at 1 month and by preoperative left ventricular angiographic study. At 6 years, the need for valve replacement was 346% in the digoxin-treated group and 153% in the nifedipine-group, P 0·001 (Figure 53.4).58 Thus, for every 100 patients treated with nifedipine, 19 fewer valve replacements were needed at the end of 6 years; note that even after 6 years, the curves are not parallel and do not converge (see Figure 53.4). Compared to the digoxin group, the nifedipine-treated group demonstrated a reduction in left ventricular volume and mass. Ejection fraction increased in the digoxin arm of the trial, and left ventricular volumes and mass increased. After aortic valve replacement, 12 of 16 patients (75%) in the digoxin group and all six patients in the nifedipine group who had an abnormal LVEF before surgery had a normal ejection fraction. Eighty-five per cent of patients in the digoxin arm of the trial, who underwent valve replacement, developed an abnormal ejection fraction and only three patients had valve replacement for symptoms. Moreover, patients in the digoxin arm of the trial had an outcome similar to that reported in the natural history studies. Long-acting nifedipine is the drug of choice for asymptomatic patients with severe chronic aortic valve regurgitation and normal left ventricular systolic function unless there is a contraindication to its use.25 The goal of vasodilator therapy is to reduce systolic blood pressure. The dose should be increased until there is a measurable decrease in blood pressure or side effects. Vasodilator therapy is not indicated

Incidence of aortic-valve replacement (%)

40

30 P < 0·001 20

10

Digoxin Nifedipine

0 0

1

2 3 4 5 Years after randomization

6

Figure 53.4 The role of long term, long acting nifedipine therapy in asymptomatic patients with severe aortic regurgitation and normal left ventricular systolic pump function was evaluated in 143 asymptomatic patients in a prospective randomized trial. By actuarial analysis, at 6 years, 34 6% of patients in the digoxin group underwent valve replacement versus 15 3% of those in the nifedipine group (P 0·001). This randomized trial demonstrates that long term arteriolar dilator therapy with long acting nifedipine reduces and/or delays the need for aortic valve replacement in asymptomatic patients with severe aortic regurgitation and normal left ventricular systolic pump function. (From Scognamiglio et al 52 with permission.)

in patients with normal left ventricular dimension and/or normal blood pressure. ACE inhibitors are not of proven benefit in asymptomatic patients with severe chronic aortic valve regurgitation and normal left ventricular systolic function. Grade A Valve surgery (replacement/repair) Surgery for aortic valve regurgitation should only be considered when the degree of regurgitation is severe. However, the presence of severe aortic valve regurgitation does not mandate surgery. The critical issue is to choose the best time for surgical intervention. Aortic valve repair or replacement should be performed in most symptomatic patients irrespective of the degree of left ventricular dysfunction. Postoperative survival is better after valve replacement in symptomatic patients with normal or mildly impaired left ventricular systolic function (ejection fraction [EF] 0·45) than in those with greater impairment of left ventricular systolic function (EF 0·45).77 In one study, patients with preoperative left ventricular EF of 0·60 had a better survival than those with left ventricular EF of 0·60.78 Extreme left ventricular dilation (end-diastolic dimension 80) associated with aortic valve regurgitation occurs primarily in men and is often associated with left ventricular dysfunction. Extreme left ventricular dilation, however, is not a marker of irreversible left ventricular dysfunction. Operative risk and late postoperative survival are acceptable in these patients.79 In the setting of severe left ventricular dysfunction (EF 0·25), the risk of aortic valve surgery increases and potential benefits decline, since left ventricular dysfunction may be on the basis of irreversible myocardial damage. However, even in the highest risk patients, the risk of surgery and postoperative medical therapy for heart failure are usually less than the risk of longterm medical management alone. Grade B Aortic valve surgery for asymptomatic patients is more controversial but is indicated in the setting of left ventricular dysfunction with an EF 0·50 and in the setting of severe left ventricular dilation (end-diastolic dimension 75 mm or end systolic dimension 55 mm), even if the ejection fraction is normal. The threshold values of end-diastolic and endsystolic dimension recommended for aortic valve replacement in asymptomatic patients may need to be adjusted to body surface area. In one series, it was noted that a left ventricular end-systolic dimension corrected for body surface area (LVS/BSA) of 25 mm/m2 was associated with increased mortality when followed conservatively.1,68,79 After valve replacement, patients with normal preoperative left ventricular systolic function have reductions of left ventricular volumes and hypertrophy.80 In the majority of patients with normal preoperative left ventricular function, there is an increase in EF after valve replacement, presumably because of a reduction of myocardial stress.31,81 Left 777


ventricular hypertrophy continues to decline for up to 5–8 years in those with normal preoperative left ventricular systolic function, but at a slower rate after 18–24 months.31,81 Most patients are symptomatically improved and are in NYHA class I.25 After valve replacement in those with abnormal preoperative left ventricular systolic function (EF 0·25–0·49), there is a reduction of heart size and left ventricular end-diastolic pressure, end-diastolic and end-systolic volumes and hypertrophy.77 Left ventricular EF improves or normalizes only if the EF was abnormal for 12 months prior to surgery.81 Very early after valve replacement, there may be a reduction in EF. The left ventricular end-diastolic volume has not yet decreased but the regurgitant volume has been eliminated; this causes a decline in EF. An early decrease in left ventricular end-diastolic dimension is a good indicator of functional success of aortic valve replacement as the magnitude of reduction in end-diastolic dimension after operation correlates with the magnitude of late increase in EF.1 Moreover, unless there is a perioperative complication, most patients are symptomatically improved and are in NYHA class I or II.25 Box 53.3 Results of valve replacement in patients with severe chronic aortic valve regurgitation ● Improved survival in those with mild to moderate impairment of left ventricular systolic function and in those with severe left ventricular enlargement irrespective of their symptomatic status ● Improvement in left ventricular systolic dysfunction; function normalizes if the dysfunction is of 12 months’ duration preoperatively ● Regression of left ventricular hypertrophy ● Improvement in functional class, particularly in those with preoperative mild to moderate impairment and in those with preoperative left ventricular dysfunction

Box 53.4 Factors predictive of a less favorable outcome ● Extent and severity of associated comorbid conditions ● Severe obstructive coronary artery disease ● Presence and severity of clinical heart failure preoperatively ● Severity of depression of preoperative LVEF ● Duration of preoperative left ventricular systolic dysfunction ● Extent of preoperative irreversible myocardial damage ● Severity of increase in left ventricular end-diastolic and end-systolic size (left ventricular end-diastolic and endsystolic volumes of 210 and 110 ml/m2, respectively, or end-diastolic and end-systolic dimensions of 80 mm and 60 mm, respectively) ● Skill and experience of operating and associated professional teams, for example, anesthetists ● Extent of perioperative myocardial damage ● Complications of a prosthetic heart valve

778

In those with severe symptoms and severe reduction of EF or severe left ventricular dilation preoperatively, survival as well as the beneficial effects on left ventricular function and functional class are less marked.80,82 Boxes 54.3 and 54.4 summarize the results of valve replacement in those with severe chronic aortic valve regurgitation and the factors predictive of a worse postoperative survival, less recovery of left ventricular function, and less improvement in symptomatic state in those with severe regurgitation and preoperative left ventricular systolic dysfunction. There are two controversial questions regarding patients with severe aortic valve regurgitation. First, when does the symptomatic patient become inoperable? Second, when should one operate on asymptomatic patients with severe aortic valve regurgitation (assuming that associated comorbid conditions do not make the patient inoperable or at high risk for surgery)? Severe left ventricular systolic dysfunction is the major factor that makes the patient with severe aortic valve regurgitation inoperable. In the published study of left ventricular dysfunction in which the patient and left ventricular function improved after valve replacement, the patients had an EF of 0·25–0·49.77,80 Personal experience indicates that with skilled and experienced surgery, patients with an EF of 0·18–0·24 are improved with operation. There are limited data on the results of valve replacement in patients with severe aortic valve regurgitation and severe left ventricular systolic dysfunction with a left ventricular EF of 0·18, these patients are very high risk for conventional valve surgery and many would consider such patients inoperable. The asymptomatic patient with severe aortic valve regurgitation poses a challenging clinical dilemma. If patients have developed left ventricular systolic dysfunction, then their outcome is poor without surgery, and left ventricular dysfunction, if present for 12 months or longer, does not normalize after surgery;81 therefore, surgery is advisable. Patients who need surgery for associated conditions, for example, obstructive CAD, thoracic aortic disease, such as an aortic aneurysm, or another valve lesion, should have surgery for the severe aortic regurgitation. Patients who have developed severe left ventricular dilation are on the edge of developing symptoms at a high rate. One could wait for symptoms to develop and follow these patients very carefully at frequent intervals. Asymptomatic patients who do not have severe left ventricular dilation and those who do not have left ventricular dysfunction at rest or exercise should not have surgery for chronic aortic valve regurgitation. The current status of aortic valve repair prevents recommending this as an early prophylactic procedure. It is difficult to determine which aortic valves will be amenable to repair. In addition, the current rate of reoperation is at a level that prevents regular use of this procedure in asymptomatic patients with minimal left ventricular enlargement.83


Recommendations: aortic valve replacement/repair in severe chronic aortic regurgitation ● Indication Class ● Symptomatic patients with: ● NYHA class III or IV symptoms and normal LV systolic function (LVEF 0·5) I ● NYHA class II symptoms, preserved systolic function (LVEF 0·5) but with progressive LV dilation or declining EF at rest, or declining exercise capacity on exercise testing I ● Canadian Heart Association class II or greater angina with or without CAD I ● NYHA class II symptoms with preserved LV systolic function (LVEF 0·5) with stable LV size and systolic function on serial studies and stable exercise tolerance IIa ● LV systolic dysfunction –LVEF 0·25–0·49 I –LVEF 0·18–0·24 IIb ● Asymptomatic patients with: ● LV systolic dysfunction

–LVEF 0·25–0·49 –LVEF 0·18–0·24

I IIb

normal LV function and: associated severe obstructive CAD needing surgery I ● other valve or thoracic aortic disease needing surgery I ● severe LV dilation with EDD 70 mm or ESD 55 mm and normal LV systolic function (LVEF 0·50) IIb ● normal systolic function at rest (LVEF 0·5) but decline in EF (0·50) on exercise radionuclide angiography IIb ● normal systolic function at rest (LVEF 0·5) but decline in EF (0·50) on stress echocardiography III ● LV dilation is not severe (EDD 70 mm, ESD 50 mm) III Abbreviations: NYHA, New York Heart Association; EDD, end-diastolic dimension; ESD, end-systolic dimension; LVEF, left ventricular ejection fraction; EF, ejection fraction; LV, left ventricular For definition of classes, see p. 773 ● ●

References 1.Bonow R, Carabello B, de Leon A et al. Guidelines for the management of patients with valvular heart disease: executive summary. A report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients with Valvular Heart Disease). Circulation 1998;98:1949–84.

2.Rahimtoola S. Aortic valve stenosis. St. Louis: CV Mosby; 1997. 3.Olsson N, Dalsgaaro C, Haegerstrand A, Rosenqvist M, Ryden L, Nilson J. Accumulation of T lymphocytes and expression of interluken-2 receptors in nonrheumatic stenotic aortic valves. J Am Coll Cardiol 1994;23:1162–70. 4.Otto C, Kuusisto J, Reichenbach D, Gown A, O’Brien K. Characterization of the early lesion of “degenerative” valvular aortic stenosis. Histological and immunohistochemical studies. Circulation 1994;90:844–53. 5.Otto C. Aortic stenosis – listen to the patient, look at the valve. N Engl J Med 2000;343:652–4. 6.Pohle K, Maffert R, Ropers D et al. Progression of aortic valve calcification: association with coronary atherosclerosis and cardiovascular risk factors. Circulation 2001;104:1927–32. 7.Otto C, Burwask I, Legget M et al. Prospective study of asymptomatic valvular aortic stenosis: clinical, echocardiographic, and exercise predictors of outcome. Circulation 1997; 95:2262–70. 8.Rahimtoola S. The problem of valve prosthesis–patient mismatch. Circulation 1978;58:20–4. 9.Tobin J Jr, Rahimtoola S, Blundell P, Swan H. Percentage of left ventricular stroke workloss: a simple hemodynamic concept for estimation of severity in valvular aortic stenosis. Circulation 1967;35:868–79. 10.Cheitlin M, Alpert J, Armstrong W et al. ACC/AHA Guidelines for the Clinical Application of Echocardiography. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Clinical Application of Echocardiography). Developed in collaboration with the American Society of Echocardiography. Circulation 1997;95:1686–744. 11.Currie P, Seward J, Reeder G et al. Continuous-wave Doppler echocardiographic assessment of severity of calcific aortic stenosis: a simultaneous Doppler-catheter correlative study in 100 adult patients. Circulation 1985;71:1162–9. 12.Rahimtoola S. “Prophylactic” valve replacement for mild aortic valve disease at time of surgery for other cardiovascular disease? No. J Am Coll Cardiol 1999;33:2009–15. 13.Ross JJr, Braunwald E. Aortic stenosis. Circulation 1968;36 (Suppl. IV):61–7. 14.Horstkotte D, Loogen F. The natural history of aortic valve stenosis. Eur Heart J 1988;(Suppl. E):57–64. 15.Rahimtoola S. Perspective on valvular heart disease: Update II. New York: Elsevier, 1991. 16.O’Neill W. Predictors of long-term survival after percutaneous aortic valvuloplasty: report of the Mansfield Scientific Balloon Aortic Valvuloplasty registry. J Am Coll Cardiol 1991;17:193–8. 17.Kennedy K, Nishimura R, Holmes DJ, Bailey K. Natural history of moderate aortic stenosis. J Am Coll Cardiol 1991;17: 313–19. 18.Pellikka P, Nishimura R, Bailey K, Tajik A. The natural history of adults with asymptomatic, hemodynamically significant aortic stenosis. J Am Coll Cardiol 1990;15:1012–27. 19.Amato M, Moffa P, Werner K, Ramires J. Treatment decision in asymptomatic aortic valve stenosis: role of exercise testing. Heart 2001;86:361–2. 20.Rosenhek R, Binder T, Porenta G et al. Predictors of outcome in severe, asymptomatic aortic stenosis. N Engl J Med 2000; 343:611–17.

779


21.White P. Heart Disease, 4th ed. New York: Macmillan, 1951. 22.Bonet T. Sepulchretum, sire Anatomia Practica, 4th ed. Geneva: Leonard Chouet; New York: Macmillan, 1951. 23.Frank S, Johnson A, Ross JJ. Natural history of valvular aortic stenosis. B Heart J 1973;35:41–6. 24.Rahimtoola S. Aortic valve stenosis, 10th ed. New York: McGraw-Hill, 2001. 25.Rahimtoola S. Aortic valve regurgitation, 9th ed. New York: McGraw-Hill, 1998. 26.Sethi GK, Miller DC, Souchek J et al. Clinical, hemodynamic and angiographic predictors of operative mortality in patients undergoing single valve replacement. J Thorac Cardiovasc Surg 1987;93:884–7. 27.Edwards F, Peterson E, Coombs L et al. Predication of operative mortality after valve replacement surgery. J Am Coll Cardiol 2001;37:885–92. 28.Mullany C, Elveback E, Frye R et al. Coronary artery disease and its management: influence on survival in patients undergoing aortic valve replacement. J Am Coll Cardiol 1987;10: 66–72. 29.Rahimtoola S. Lessons learned about the determinants of the results of valve surgery. Circulation 1988;78:1503–7. 30.Schwarz F, Banmann P, Manthey J et al. The effect of aortic valve replacement on survival. Circulation 1982;66:1105–10. 31.Pantely G, Morton M, Rahimtoola S. Effects of successful uncomplicated valve replacement on ventricular hypertrophy, volume, and performance in aortic stenosis and aortic incompetence. J Thorac Cardiovasc Surg 1978;75:383–91. 32.Monrad E, Hess O, Murakami T, Nonogi H, Corin W, Krayenbuehl H. Time course of regression of left ventricular hypertrophy after aortic valve replacement. Circulation 1988; 77:1345–55. 33.Carroll J, Carroll EP, Feldman T et al. Sex-associated differences in left ventricular function in aortic stenosis of the elderly. Circulation 1992;86:1099–107. 34.Hammermeister K, Sethi G, Henderson W, Oprian C, Kim T, Rahimtoola S. A comparison of outcomes in men 11 years after heart-valve replacement with a mechanical valve or bioprosthesis. N Engl J Med l993;328:1289–96. 35.Murphy E, Lawson R, Starr A, Rahimtoola S. Severe aortic stenosis in patients 60 years of age and older: left ventricular function and ten-year survival after valve replacement. Circulation 1981;64(Suppl. II):184–8. 36.Lindblom D, Lindblom U, Qvist J LundströmH Long-term relative survival rates after heart valve replacement. J Am Coll Cardiol 1990;15:566–73. 37.Sprigings D, Forfar J. How should we manage symptomatic aortic stenosis in the patient who is 80 or older? Br Heart J 1995;74:481–4. 38.Rahimtoola S. Catheter balloon valvuloplasty for severe calcific aortic stenosis: a limited role. J Am Coll Cardiol 1994; 23:1076–8. 39.Smith N, McAnulty J, Rahimtoola S. Severe aortic stenosis with impaired left ventricular function and clinical heart failure: results of valve replacement. Circulation 1978;58:255–64. 40.Connolly H, Oh J, Orszulak T et al. Aortic valve replacement for aortic stenosis with severe left ventricular dysfunction. Prognostic indicators. Circulation 1997;95:2395–2400. 41.Rahimtoola S, Starr A. Valvular surgery. New York: Grune and Stratton, 1982.

780

42.Connolly H, Oh J, Schaff H et al. Severe aortic stenosis with low transvalvular gradient and severe left ventricular dysfunction: result of aortic valve replacement in 52 patients. Circulation 2000;101:1940–6. 43.deFilippi C, Willett D, Brickner M et al. Usefulness of dobutamine echocardiography in distinguishing severe from nonsevere valvular aortic stenosis in patients with depressed left ventricular function and low transvalvular gradients. Am J Cardiol 1995;75:191–4. 44.Monin J, Monchi M, Gest V, Duval-Moulin A, Dubois-Rande J, Gueret P. Aortic stenosis with severe left ventricular dysfunction and low transvalvular pressure gradients: risk stratification by low-dose dobutamine echocardiography. J Am Coll Cardiol 2001;37:2101–7. 45.Nishimura RA, Grantham JA, Connolly HM, Schaff HV, Higano ST, Holmes DR Jr. Low-output, low-gradient aortic stenosis in patients with depressed left ventricular systolic function: the clinical utility of the dobutamine challenge in the catheterization laboratory. Circulation 2002;106(Suppl. 7):809–13. 46.Rahimtoola S. Aortic valve regurgitation. St. Louis: CV Mosby, 1997. 47.Lewis TS. Special Report Series of the National Health Insurance Joint Committee, Medical Research Committee, Report No. 8. London: MRC, 1917. 48.Grant R. After histories for 10 years of a thousand men suffering from heart disease. Heart 1933;16:275–334. 49.McKay C, Rahimtoola S. Natural history of aortic regurgitation. New York: Kluwer, 1980. 50.Spagnuolo M, Kloth H, Taranta A, Doyle E, Pasternack B. Natural history of rheumatic aortic regurgitation. Criteria predictive of death, congestive heart failure, and angina in young patients. Circulation 1971;34:368–80. 51.Goldschlager N, Pfeifer J, Cohn K, Popper R, Selze R. Natural history of aortic regurgitation: a clinical and hemodynamic study. Am J Med 1973;54:577–88. 52.Scognamiglio R, Fasoli G, Dalla Volta S Progression of myocardial dysfunction in asymptomatic patients with severe aortic insufficiency. Clin Cardiol 1986;9:151–6. 53.Siemienczuk D, Greenberg B, Morris C et al. Chronic aortic insufficiency: factors associated with progression to aortic valve replacement. Ann Intern Med 1989;110:587–92. 54.Bonow R, Lakatos E, Maron B, Epstein S. Serial long-term assessment of the natural history of asymptomatic patients with chronic aortic regurgitation and normal left ventricular systolic function. Circulation 1991;84:1625–35. 55.Scognamiglio R, Fasoli G, Ponchia A, Dalla Volta S Long-term nifedipine unloading therapy in asymptomatic patients with chronic severe aortic regurgitation. J Am Coll Cardiol 1990; 16:424–9. 56.Tornos M, Olona M, Permanyer-Miralda G et al. Clinical outcome of severe asymptomatic chronic aortic regurgitation. A long-term prospective follow-up study. Am Heart J 1995; 130:333–9. 57.Ishii K, Hirota Y, Suwa M, Kita Y, Onaka H, Kawamura K. Natural history and left ventricular response in chronic regurgitation. Am J Cardiol 1996;78:357–61. 58.Scognamiglio R, Rahimtoola S, Fasoli G, Nistri S, Dalla Volta S. Nifedipine in asymptomatic patients with severe aortic regurgitation and normal left ventricular function. N Engl J Med 1994;331:689–95.


59.Henry W, Bonow R, Rosing D, Epstein S. Observations on the optimum time for operative intervention for aortic regurgitation. II. Serial echocardiographic evaluation of asymptomatic patients. Circulation 1980;61:484–92. 60.McDonald I, Jelinek V. Serial M-mode echocardiography in severe chronic aortic regurgitation. Circulation 1980;62: 1291–6. 61.Bonow R. Radionuclide angiography in the management of asymptomatic aortic regurgitation. Circulation 1991;84: I.296–I.302. 62.Hegglin R, Scheu H, Rothlin M. Aortic insufficiency. Circulation 1968;38(Suppl. 15):V77–V92. 63.Rapaport E. Natural history of aortic and mitral valve disease. Am J Cardiol 1975;35:221–7. 64.Rahimoola S. Aortic Valve regurgitation, 10th ed. New York: McGraw-Hill, 2001. 65.Aronow W, Ahn C, Kronzon I, Nanna M. Prognosis of patients with heart failure and unoperated severe aortic valvular regurgitation and relation to ejection fraction. Am J Cardiol 1994;74:286–8. 66.Broadbent W. Aortic incompetence. In: Heart disease: with special reference to prognosis and treatment. London: Baillière, Tindall & Cox, 1897. 67.Klodas E, Enriquez-Sarano M, Tajik A, Mullany C, Bailey K, Seward J. Surgery for aortic regurgitation in women: contrasting indications and outcomes compared with men. Circulation 1996;94:2472–8. 68.Dujardin K, Enriquez-Sarano M, Schaff H, Bailey K, Seward J, Tajik A. Mortality and morbidity of aortic regurgitation in clinical practice. A long-term follow-up study. Circulation 1999; 99:1851–7. 69.Boucher C, Wilson R, Kanarek D et al. Exercise testing in asymptomatic or minimally symptomatic aortic regurgitation: relationship of left ventricular ejection fraction to left ventricular filling pressure during exercise. Circulation 1983;67: 1091–1100. 70.Kawanishi D, McKay C, Chandraratna P et al. Cardiovascular response to dynamic exercise in patients with chronic symptomatic mild-moderate and severe aortic regurgitation. Circulation 1986;73:62–72. 71.Martinez-Rubio A, Schwammenthal Y, Schwammenthal E et al. Patients with valvular heart disease presenting with sustained ventricular tachyarrhythmias or syncope: results of programmed ventricular stimulation and long-term follow-up. Circulation 1997;96:500–8. 72.Rahimtoola S. Management of heart failure in valve regurgitation. Clin Cardiol 1992;15(Suppl. I):22–7. 73.Rahimtoola S. Vasodilator therapy in chronic, severe aortic regurgitation. J Am Coll Cardiol 1990;16:430–2.

74.Greenberg B, Massie B, Bristow J et al. Long-term vasodilator therapy of chronic aortic insufficiency: a randomized doubleblind, placebo-controlled clinical trial. Circulation 1988;78: 92–103. 75.RöthlisbergerC, Sareli P, Wisenbaugh T. Comparison of singledose nifedipine and captopril for chronic severe aortic regurgitation. Am J Cardiol 1993;72:799–804. 76.Wisenbaugh T, Sinovich V, Dullabh A, Sareli P. Six month pilot study of captopril for mildly symptomatic, severe isolated mitral and isolated aortic regurgitation. J Heart Valve Dis 1994;3:197. 77.Greves J, Rahimtoola S, Clinic M et al. Preoperative criteria predictive of late survival following valve replacement for severe aortic regurgitation. Am Heart J 1981;101:300–8. 78.Cunha C, Giuliani E, Fuster V, Seward J, Brandenburg R, McGoon D. Preoperative M-mode echocardiography as a predictor of surgical results in chronic aortic insufficiency. J Thorac Cardiovasc Surg 1980;79:256–65. 79.Klodas E, Enriquez-Sarano M, Tajik A, Mullany C, Bailey K, Seward J. Optimizing timing of surgical correction in patients with severe aortic regurgitation: role of symptoms. J Am Coll Cardiol 1997;30:746–52. 80.Clark D, McAnulty J, Rahimtoola S. Valve replacement in aortic insufficiency with left ventricular dysfunction. Circulation 1980;61:411–21. 81.Bonow R, Dodd J, Maron B et al. Long-term serial changes in left ventricular function and reversal of ventricular dilatation after valve replacement for chronic aortic regurgitation. Circulation 1988;78:1108–20. 82.Klodas E, Enriquez-Sarano M, Tajik A, Mullany C, Bailey K, Seward J. Aortic regurgitation complicated by extreme left ventricular dilation: long-term outcome after surgical correction. J Am Coll Cardiol 1996;27:670–7. 83.Izumoto H, Kawazoe K, Ishibashi K et al. Aortic valve repair in dominant aortic regurgitation. Japan J Thorac Cardiovasc Surg 2001;49:355–9. 84.Rapaport E. Natural history of aortic and mitral valve disease Am J Cardiol 1975;35(Suppl. 2):221–7. 85.Chizner MA, Pearle DL, deLeon AC Jr. The natural history of aortic stenosis in adults. Am Heart J 1980;99(Suppl. 4):419–24. 86.O’Keefe JH Jr, Vlietstra RE, Bailey KR, Holmes DR Jr. Natural history of candidates for balloon aortic valvuloplasty. Mayo Clin Proc 1987;62(Suppl. 11):986–91. 87.Turina J, Hess O, Sepulcri F, Krayenbuehl HP. Spontaneous course of aortic valve disease. Eur Heart J 1987;8(Suppl. 5): 471–83. 88.Kelly TA, Rothbart RM, Cooper CM, Kaiser DL, Smucker ML, Gibson RS. Comparison of outcome of asymptomatic to symptomatic patients older than 20 years of age with valvular aortic stenosis. Am J Cardiol 1988:61(Suppl. 1):123–30.

781

54

Balloon valvuloplasty: aortic valve Daniel J Diver, Jeffrey A Breall

Aortic stenosis: natural history and prognosis

782

100

Latent period (increasing obstruction, myocardial overload

80 % Survival

Grade A There are three major etiologies for valvular aortic stenosis in the adult patient: rheumatic aortic stenosis; congenital bicuspid aortic stenosis with secondary calcification; and senile calcific or degenerative aortic stenosis. In rheumatic aortic stenosis the major pathologic feature is commissural fusion, with associated thickening and fibrosis of the valve leaflets. Symptoms may not occur until the age of 50 or 60 and are often accompanied by evidence of other valvular disease, usually mitral. Patients with congenital bicuspid aortic stenosis develop progressive narrowing and calcification of the aortic valve over time, with symptoms often present by age 40–50. Degenerative calcific (senile) aortic stenosis appears to result from years of normal mechanical stress on the aortic valve, with progressive immobilization of cusps secondary to calcium accumulation in the pockets of the aortic cusps, and eventual fibrosis. Degenerative calcific aortic stenosis is now the most common cause of aortic stenosis in patients presenting for aortic valve replacement.1 Most data regarding the natural history of aortic stenosis are derived from clinical experience during the presurgical era. The natural history of aortic stenosis is characterized by a long latent period marked by slowly increasing obstruction and adaptive myocardial hypertrophy. The majority of patients are free of cardiovascular symptoms until relatively late in the course of the disease. However, once patients with aortic stenosis develop symptoms of angina, syncope or heart failure, survival with medical therapy is dismal, with death occurring within 2–5 years in most patients following the development of symptoms (Figure 54.1). Average survival in patients with aortic stenosis and angina or syncope is 2–3 years, and may be as short as 1·5 years in patients with aortic stenosis who develop heart failure.2 Concomitant atrial fibrillation decreases survival in all symptom groups. Asymptomatic patients with aortic stenosis have an excellent prognosis and rarely die without premonitory symptoms. A study by Pellikka et al 4 showed that mortality was slightly higher in asymptomatic patients treated with “prophylactic” valve replacement than in patients not operated on until symptoms develop. A recent study by Otto and colleagues reported follow up of 123 patients with asymptomatic aortic stenosis. During the 2·5 year, follow up period

Onset severe symptoms

60

Angina Syncope Failure 0 2 4 6 Av. Survival (yr)

40

Average death Age ( )

20

0

40

50

60 Age (yr)

70

80

Figure 54.1 Natural history of aortic stenosis without operative treatment. (Reproduced with permission from Ross and Braunwald.3)

there were no sudden cardiac deaths. This study suggested that the rate of hemodynamic progression and clinical outcome in adults with asymptomatic aortic stenosis may be predicted by echocardiographic aortic jet velocity. Of those patients who entered the study with a peak aortic jet velocity 4 m/s only 21% were alive and free of valve replacement 2 years later.5 The timing of aortic valve replacement in patients with aortic stenosis is predicated on the development of symptoms or deterioration in left ventricular performance, rather than severity of valve gradient or reduction in valve orifice area. Carabello6 has proposed a definition of “critical” aortic stenosis as that valve area small enough to cause the symptoms of aortic stenosis that often presage sudden death: a “critical” situation indicating the need for aortic valve replacement. The aortic valve area associated with such symptom development varies significantly from patient to patient. Aortic stenosis: natural history and prognosis ● ● ●

Long latent period without symptoms Poor prognosis following symptom development with death in 2–5 years Prognosis significantly improved by valve replacement surgery.

Balloon valvuloplasty: aortic valve

Surgery for aortic stenosis Grade B The initial surgical approach to treatment of aortic stenosis involved surgical valvuloplasty. In contrast to the situation with pulmonary and mitral stenosis, the stenotic aortic valve did not respond favorably to surgical valvuloplasty techniques. Closed aortic commissurotomy was associated with a high incidence of acute aortic regurgitation and operative mortality, and was abandoned after the development of open aortic valve surgical techniques. Surgical valvuloplasty under direct vision for aortic stenosis was first described in 1956, but was limited by a high rate of restenosis leading to subsequent aortic valve replacement, as well as a significant incidence of complications, including aortic regurgitation, infective endocarditis and systemic embolization.7 Although ultrasonic decalcification and careful surgical sculpting procedures carried out under direct vision are initially effective in some patients, restenosis remains a serious problem.8 However, open surgical valvulotomy remains an important treatment for infants and children with critical aortic stenosis, a situation where initial prosthetic valve replacement is undesirable. The development and refinement of surgical aortic valve replacement significantly improved morbidity and mortality in patients with symptomatic aortic stenosis. Although there is no prospective randomized controlled study comparing aortic valve replacement with medical therapy in such patients, long-term follow up in high-quality case series has convincingly demonstrated the long-term benefits of aortic valve replacement, including hemodynamic improvement, regression of left ventricular hypertrophy, improvement of left ventricular function and improved survival.9–11 Operative mortality for aortic valve replacement ranges from 2 to 8%, but may be as low as 1% in patients less than 70 years of age without significant comorbidity. Aortic valve replacement, however, is associated with increased morbidity and mortality in certain subgroups.10,12–15 Aortic valve replacement in the presence of left ventricular failure may be associated with perioperative mortality as high as 10–25%, and the need for emergency aortic valve replacement with operative mortality as high as 40%. Surgical risk is increased in the elderly patient, and may be increased severalfold with the need for concomitant bypass or multiple valve surgery. Although advanced age remains a strong predictor of operative death for aortic valve replacement even in recent studies, age alone is not a contraindication to aortic valve replacement in patients with aortic stenosis.16 The Society of Thoracic Surgeons National Cardiac Surgery Database identified risk factors in nearly 50 000 patients who had valve surgery between 1994 and 1997: for patients with isolated aortic valve replacement, age was not a strong predictor of risk.17 Fremes and colleagues18 at the University of Toronto described the result of valve surgery in 469 consecutive patients over 70 years of

age, and found that the predicted probability of operative mortality ranged from 0·9 to 76%, depending on the presence of other risk factors, including urgent operation, double valve surgery, coronary artery disease, female gender and left ventricular dysfunction. The authors suggested that elderly patients in good risk categories should be offered surgical intervention for the correction of valvular lesions, whereas alternative therapy might be indicated in patients with multiple risk factors in whom surgical mortality was prohibitively high. Levinson and colleagues at the Massachusetts General Hospital reported on aortic valve replacement for aortic stenosis in octogenarians.19 In their cohort of 64 patients, serious comorbid non-cardiac conditions were infrequent. In-hospital mortality was 9·4%. An additional 10% of patients had permanent severe neurologic deficits and an additional 38% had a “complicated” course, marked by temporary encephalopathy, discharge to a rehabilitation facility or some combination thereof, albeit with ultimately good results. Although most survivors were ultimately free of cardiac symptoms, there was a high price to pay in terms of perioperative mortality and morbidity to achieve these results. However, recent series suggest that surgical results may be improving in very elderly patients. Rosengart and colleagues20 compared results in 100 consecutive patients age 85 years or older who underwent open heart surgery between 1994 and 1997 with results obtained in the prior decade, and noted improvement in 30 day mortality and risk of major complications. Therefore, while surgical aortic valve replacement has clearly improved the outcome in most patients with critical aortic stenosis, the higher risk in some patient subgroups, including the elderly, often leads to physician or patient deferral of aortic valve replacement. In an attempt to define the natural history of such patients, O’Keefe and colleagues21 at the Mayo Clinic performed a case comparison study of 50 patients with severe, symptomatic aortic stenosis in whom surgery was declined by the patient (n 28) or the physician (n 22). The actuarial survival at 1, 2 and 3 years was 57, 37 and 25%, respectively. The survival of age- and sexmatched control subjects was 93, 85 and 77%, respectively (P 0 · 0001 at each 1 year interval) (Figure 54.2). This study suggested that the natural history of untreated aortic stenosis remains dismal and has not improved in the modern era, and confirmed the necessity of evaluating alternative non-surgical therapy, such as balloon aortic valvuloplasty, in patients likely to decline aortic valve replacement, or for whom surgery is not an option.

Development of balloon aortic valvuloplasty Grade C Children and adolescents with congenital aortic stenosis generally have non-calcified valves with commissural fusion. Because aortic valve replacement is not desirable in 783


100

Survival (%)

80

60

* *

40

* 20

Control Aortic stenosis

0 1

2

3

Years

Figure 54.2 Survival among 50 patients with severe aortic stenosis who did not undergo surgical treatment, in comparison with an age- and sex-matched control group from the US population. Asterisks denote significant differences (P 0.0001) between the two groups. Standard errors are shown as vertical lines. (Reproduced with permission from O’Keefe et al.21)

this age group, commissural incision under direct vision is the preferred surgical procedure, and has been shown to confer significant hemodynamic improvement at low risk.22 The contribution of commissural fusion to the etiology of valvular stenosis and mechanism of surgical improvement in this patient group led to the consideration of balloon aortic valvuloplasty as an alternative, non-surgical therapy. In 1984 Lababidi and colleagues23 reported the first series of 23 children and young adults with congenital aortic stenosis treated with percutaneous balloon aortic valvuloplasty. The patients ranged in age from 2 to 17 years. Balloon valvuloplasty was performed by the retrograde approach from the femoral artery, utilizing balloons of 10–20 mm in diameter. Percutaneous balloon dilation resulted in a decrease in the peak aortic valve gradient from 113 to 32 mmHg, with no change in cardiac output. The excellent initial results of percutaneous balloon valvuloplasty for aortic valve stenosis were confirmed by Rosenfeld and colleagues in young adults with congenital aortic stenosis. Long-term follow-up appeared to be excellent, with a 58% event-free rate at mean follow-up of 38 months,24 although a recent multicenter study from Japan reported that progressive aortic insuffiency and recurrence of pressure gradient was not uncommon by 4 years after balloon valvuloplasty.25 The excellent results of balloon valvuloplasty in pediatric patients with congenital aortic stenosis led to consideration of this technique in adult patients with acquired calcific aortic stenosis. Two reports in 1986 described the first successful balloon valvuloplasty procedures in adult patients. Cribier and colleagues in France performed percutaneous balloon dilation in three elderly patients with calcific aortic 784

stenosis.26 The peak aortic gradient decreased from 75 to 33 mmHg, with an increase in calculated aortic valve area from 0·5 to 0·8 cm2. All patients had symptomatic improvement. McKay and colleagues27 at the Beth Israel Hospital in Boston described two elderly patients (aged 93 and 85 years) with calcific aortic stenosis treated with balloon valvuloplasty with 12–18 mm balloons. This report likewise described a substantial reduction in the transaortic pressure gradient and a significant increase in aortic valve area, with symptomatic improvement in both patients and significant improvement in left ventricular function in one. Despite initial concern regarding the possibility of valve disruption or embolization in the calcific valves present in adult patients, no patient in either report developed emboli or a significant increase in aortic regurgitation.

Mechanism of balloon aortic valvuloplasty Grade B To assess the safety, efficacy and mechanism of balloon aortic valvuloplasty, Safian and colleagues28 performed balloon dilation of stenotic aortic valves in 33 postmortem specimens and in six patients undergoing aortic valve replacement, prior to removal of the stenotic valve. The cause of aortic stenosis was degenerative nodular calcification in 28 cases, calcific bicuspid aortic stenosis in eight cases, and rheumatic heart disease in three. The distribution of the etiology of aortic stenosis in this report is in concordance with the observation that degenerative calcific aortic stenosis is now the most common cause of aortic stenosis in adults presenting for aortic valve replacement.1 Safian and colleagues performed balloon dilation with 15–25 mm balloons in the postmortem specimens, and with 18–20 mm balloons in the surgical patients. Balloon dilation resulted in increased leaflet mobility and increased valve orifice dimensions in all patients. The mechanism of successful dilation included fracture of calcified nodules within the leaflets in 16 valves, separation of fused commissures in five valves, both in six valves, and “grossly inapparent microfractures” (or stretching) in 12 valves. Liberation of calcific debris, valve ring disruption and midleaflet tears did not occur in any valve, although valve leaflet avulsion was produced in one postmortem specimen after inflation with a clearly oversized balloon. The authors concluded that there were several mechanisms for successful balloon aortic valvuloplasty, with the predominant mechanism in a given patient depending on the etiology of the stenosis. Furthermore, it appeared that embolic phenomena and acute regurgitation were not likely to be frequent complications following valvuloplasty. The study by Safian and colleagues suggested that the most common etiology of aortic stenosis in the balloon valvuloplasty population is degenerative nodular calcification, and that the predominant mechanism of valve dilation is fracture of calcified nodules within leaflets and leaflet


stretching. Considered in conjunction with the disappointing surgical experience when stenotic aortic valves were dilated or cracked, the results of this mechanistic study predicted that there might be only mild improvement in aortic valve orifice area in patients treated with balloon aortic valvuloplasty, and that any such improvement might be short-lived. As will be seen, these implications were subsequently borne out in clinical trials. Technical aspects In the original reports by Cribier26 and McKay,27 balloon valvuloplasty was performed via the retrograde femoral approach. The most common balloon size used with the single-balloon retrograde approach is 20 mm, although smaller balloons can be used initially in small or frail patients. If no waist is produced in the inflated balloon, or if the aortic valve gradient is not sufficiently decreased by a given balloon size, a larger balloon may produce a better result. Several modifications of the percutaneous retrograde femoral approach have been described. Block and Palacios29 described an antegrade transseptal technique which they advocated for patients with severe iliac occlusive disease, tortuous iliac vessels or abdominal aortic aneurysm. This approach has recently been reported using the Inoue balloon, which may provide a greater increase in aortic valve area than conventional balloons30 and which allows combined mitral and aortic valvuloplasty using a single catheter and access site.31 A retrograde brachial approach may also be useful in such situations, although care must be taken to avoid injury to the brachial artery by the large valvuloplasty balloon. Dorros and colleagues32 described a double-balloon technique, using both femoral arteries or a combined brachial and femoral approach. The combined diameter of the balloons used in this approach usually exceeds the diameter of the largest balloon used with single-balloon techniques. While initial results with double-balloon aortic valvuloplasty showed a greater enlargement of aortic valve area, follow-up studies showed no reduction in subsequent restenosis compared to single-balloon valvuloplasty.33 An important recent technical advance is management of the femoral access site with preloaded suture closure devices,

which may significantly reduce the incidence of vascular complications following balloon valvuloplasty.34,35 Initial results of balloon aortic valvuloplasty Single center studies Grade B Within several years of the initial reports of balloon valvuloplasty in adult patients with aortic stenosis, several centers reported large single-center experiences.36–39 Cribier et al 36 reported their initial experience with 92 adult patients with symptomatic aortic stenosis and a mean age of 75 years. The aortic valve gradient was reduced from 75 to 30 mmHg, with an increase in calculated aortic valve area from 0·5 to 0·9 cm2. The left ventricular ejection fraction rose from 48% at baseline to 51% immediately following the procedure. The majority of patients had marked symptomatic improvement. There were three in-hospital deaths and eight late deaths. Safian et al 37 reported their initial experience with balloon aortic valvuloplasty in 170 consecutive patients treated at the Beth Israel Hospital in Boston. The procedure was completed successfully in 168 patients and resulted in significant increases in mean aortic valve area (0·6–0·9 cm2) and cardiac output (4·6–4·8 l/min) and a significant decrease in aortic valve pressure gradient (71–36 mmHg) (P 0·01 for all comparisons). There were six in-hospital deaths and five patients required early aortic valve replacement. The majority of patients had marked symptomatic improvement following the procedure. The most common complication was vascular, involving the femoral access site: 40 patients required transfusion and 17 required surgical repair. Transient dysrhythmias, most commonly left bundle branch block, occurred in 28 patients. Left ventricular perforation and tamponade occurred in three patients, a marked increase in aortic regurgitation in two patients, and a non-Q wave myocardial infarction in one patient. No patient suffered a stroke. The hemodynamic results and complications of balloon aortic valvuloplasty in several large single-center studies are summarized in Tables 54.1 and 54.2, respectively. The results are remarkable for their similarity across study sites.

Table 54.1 Acute hemodynamic results of balloon aortic valvuloplasty Author

26

Cribier Safian28 Block29 Lewin39

Patients (n)

92 170 162 125

Aortic valve gradient (mmHg)

Aortic valve area (cm2)

Pre

Post

Pre

Post

75 71 61 87

30 36 27 32

0.5 0.6 0.5 0.6

0.9 0.9 0.9 1.0

785


Table 54.2 Complications of balloon aortic valvuloplasty Author

Safian28 Block29 Lewin39 Total

Patients (n)

170 162 125 457

Complications (%) Death

CVA

Perforation

MI

AI

Vascular

3·5 7·0 10·4 6·6

0 2·0 3·2 1·5

1·8 0 0 0.7

0·6 0 1·6 0·7

1·2 0 1·6 0·9

10·0 7·0 9·6 8·8

Abbreviations: AI, aortic insufficiency; CVA, cerebrovascular accident; MI, myocardial infarction

In general, balloon aortic valvuloplasty resulted in a 50–70% decrease in aortic valve gradient and a 50–70% increase in aortic valve area, resulting in early symptomatic improvement in most patients. The most common complication was vascular at the access site; there was a low incidence of lifethreatening procedural complications. Death during the periprocedural period occurred in about 6% of patients. Multicenter studies Grade B Two large multicenter studies reported the initial results of balloon valvuloplasty in adult patients with symptomatic aortic stenosis.40,41 The Mansfield Balloon Aortic Valvuloplasty Registry40 evaluated data from 27 clinical centers in the United States and included 492 patients treated with balloon aortic valvuloplasty between December 1986 and October 1987. The mean age of patients was 79 years. All had severe symptoms, with 92% reporting congestive heart failure. Balloon aortic valvuloplasty was performed via the femoral approach in 92% of patients, by the brachial approach in 6%, and by the transseptal approach in 2%. A single-balloon technique was used in 72% of patients. The largest balloon size was 20 mm in over half of patients. In the Mansfield Registry, balloon aortic valvuloplasty resulted in a decrease in mean aortic valve gradient from 60 to 30 mmHg, an increase in cardiac output from 3·9 to 4·0 l/min and an increase in aortic valve area from 0·5 to 0·8 cm2. Most patients had significant symptomatic improvement. Death occurred during the procedure in 4·9% of patients, and within 7 days of the procedure in an additional 2·6%. The most common complication (11%) was local vascular injury, requiring surgical repair in 5·7% of patients. Embolic complications, ventricular perforation resulting in tamponade, and significant increase in aortic insufficiency each occurred in 1–2% of patients, and significant arrhythmia or myocardial infarction in less than 1%. Emergency aortic valve replacement was required in 1% of patients following balloon valvuloplasty. The National Heart Lung and Blood Institute (NHLBI) Balloon Valvuloplasty Registry enrolled 674 elderly (average 786

age 78 years) patients at 24 centers between November 1987 and November 1989.41 Heart failure was the most common presenting symptom, occurring in 92% of patients; 45% of patients had angina and 35% had syncope. A singleballoon retrograde valvuloplasty technique was used in 94% of patients; the largest balloon used was 20 mm in over half. The mean gradient decreased from 55 to 29 mmHg and the aortic valve area increased from 0·5 to 0·8 cm2, associated with symptomatic improvement in most patients. Procedural mortality was 3%; other major complications associated with the valvuloplasty procedure included cardiac arrest (5%), emergency aortic valve replacement (1%), left ventricular tamponade (2%), cerebral vascular accident (1%), systemic embolus (1%), emergency temporary pacing (5%), and ventricular arrhythmia requiring countershock (3%). In summary, the initial results of the multicenter studies were similar to each other, and to the results of the previously described single-center studies, and suggested that balloon aortic valvuloplasty resulted in modest hemodynamic improvement and significant symptomatic improvement in many patients considered to be at high risk for aortic valve surgery. Left ventricular function Grade B Aortic valve replacement has been shown to improve left ventricular function in many patients with aortic stenosis and left ventricular dysfunction.9–11 Safian and colleagues42 at Beth Israel Hospital examined the effect of balloon aortic valvuloplasty on left ventricular performance in 28 patients with a low left ventricular ejection fraction (mean 37%), severe aortic stenosis and a mean age of 79 years. Balloon valvuloplasty resulted in significant increases in aortic valve area (0·5–0·9 cm2), systolic pressure (120–135 mmHg), and cardiac output (4·2–4·8 l/min) (P 0·01 for all comparisons), and significant decreases in transaortic pressure gradient (69–35 mmHg) and pulmonary capillary wedge pressure (24–20 mmHg) (P 0·01 for both comparisons). All patients were symptomatically improved at the time of discharge. Serial radionuclide ventriculography showed an increase in left ventricular ejection fraction from 37% prior to


valvuloplasty to 44% 48 hours post procedure and 49% at 3 month follow up. However, there was substantial heterogeneity of response, with 13 patients showing progressive increases in left ventricular ejection fraction (34% to 49% to 58%, P 0 · 001), whereas 15 patients showed no significant change in ejection fraction (41% to 40% to 41%, P NS) over 3 months. There was no difference between the groups with respect to age, extent of coronary disease, history of myocardial infarction, or baseline or postprocedure aortic valve area. However, peak systolic wall stress and left ventricular dimensions were higher in those patients who showed no improvement in ejection fraction. It may be that the failure to increase ejection fraction in this group is due to irreversible impairment in myocardial contractile function, secondary to previous infarction or longstanding aortic stenosis. Davidson and colleagues at Duke University also found that fewer than half of patients with a baseline left ventricular ejection fraction less than 45% showed sustained improvement following percutaneous balloon aortic valvuloplasty, even at short-term follow up.43

Follow up Grade B Despite the moderate hemodynamic improvement and significant symptomatic improvement initially achieved in most patients with aortic stenosis following percutaneous balloon valvuloplasty, this technique is severely limited by the high incidence of restenosis. The Beth Israel group reported follow-up results in 170 patients (mean age 77 years) with symptomatic aortic stenosis who underwent balloon aortic valvuloplasty between October 1985 and April 1988.37 The procedure was completed successfully in 168 patients, with significant improvement in aortic valve area and gradient. There were six in-hospital deaths and five patients required early aortic valve replacement. Follow up averaging 9·1 months was available for all 157 patients discharged from the hospital after successful valvuloplasty. In 44 patients (28%), recurrent symptoms developed a mean of 7·5 months after the procedure: 16 were treated by repeat valvuloplasty, 17 by aortic valve replacement and 11 with medical therapy. Two patients had a second restenosis, treated by aortic valve replacement in one case and by a third valvuloplasty procedure in the other. At latest follow up 103 patients (66%) were symptomatically improved, including 15 with restenosis who successfully underwent redilation. Twenty-five patients died after discharge, a mean of 6 months after balloon valvuloplasty. The most common cause of death was progressive congestive heart failure. Repeat cardiac catheterization was performed in 35 patients in the Beth Israel follow-up cohort, including 21 with recurrent symptoms, a mean of 6 months after valvuloplasty. Significant aortic valve restenosis was found in all 21 patients with recurrent symptoms, and in eight of the 14 patients

without symptoms. If restenosis was assumed to have occurred in all 25 patients who died, and in all 44 patients with recurrent symptoms, then the “clinical” rate of restenosis following valvuloplasty was 44% at only 9 months. The probability of survival at 1 year was 74% for the entire study population. However, if both recurrent symptoms and death were considered as events, the probability of event-free survival at 1 year was only 50%. Similarly poor long-term results with high rates of early restenosis were reported by both of the multicenter studies of balloon aortic valvuloplasty. Among the 492 patients treated with balloon valvuloplasty in the Mansfield Registry the 1 year survival rate was 64%, with an event-free survival rate of only 43%.44 Among the 674 patients reported in the National Heart, Lung and Blood Institute Balloon Valvuloplasty Registry, survival at 1, 2 and 3 years was 55, 35 and 23%, respectively.45 Lieberman and colleagues46 reported long-term follow up in 165 patients undergoing balloon aortic valvuloplasty. The median duration of follow up was 3·9 years, with follow up achieved in 99% of patients. Ninety-three per cent of patients died or underwent aortic valve replacement, and 60% died of cardiacrelated causes. The probability of event-free survival, defined as freedom from death, aortic valve replacement or repeat balloon aortic valvuloplasty at 1, 2 and 3 years after balloon valvuloplasty, was 40%, 19% and 6%, respectively. By contrast, the probability of survival 3 years after balloon aortic valvuloplasty in a subset of 42 patients who underwent subsequent aortic valve replacement was 84%. Mechanism of restenosis Because the mechanism by which balloon aortic valvuloplasty increases aortic valve area appears to consist chiefly of fracture of calcified nodules within leaflets and leaflet stretching, and only rarely involves separation of commissural fusion,28 it is not surprising that the initial improvement in aortic valve area is modest at best, and that significant and early restenosis occurs in most patients. Any element of improvement in the aortic valve area due to leaflet stretching is likely to be rapidly compromised by elastic recoil, and in fact postprocedure echocardiographic follow up suggests early loss of initial valve area in many patients.47 Histologic examination in patients who underwent balloon aortic valvuloplasty and had subsequent valve tissue examined at the time of aortic valve replacement or necropsy, showed evidence of closing of fractures in calcified nodules by granulation tissue that may lead to valvular scarring.48,49 The more rapid time course of this type of inflammatory response, compared to the slowly developing valvular calcification that initially led to the aortic stenosis, may explain the relatively rapid progression to symptomatic restenosis following initially successful balloon valvuloplasty. 787


Results of balloon aortic valvuloplasty Initial hemodynamic and symptomatic improvement Early restenosis, with recurrent symptoms No improvement in long-term survival or event-free survival.

● ● ●

Predictors of outcome following balloon aortic valvuloplasty Grade B Following recognition of the high incidence of restenosis after balloon aortic valvuloplasty, attempts were made to identify patient subsets more likely to derive longterm benefit. Kuntz and colleagues50 analyzed event-free survival in 205 patients who underwent balloon valvuloplasty for symptomatic critical aortic stenosis. They evaluated 40 demographic and hemodynamic variables as univariate predictors of event-free survival by Cox regression analysis, and attempted to identify independent predictors of event-free survival by stepwise multivariate analysis. The rate of event-free survival, defined as survival without recurrent symptoms, repeat balloon valvuloplasty or subsequent aortic valve replacement, was 18% over a mean follow-up period of 24 months (Figure 54.3). Direct predictors of long-term event-free survival in the univariate analysis included female gender, left ventricular ejection fraction, and left ventricular and aortic systolic pressure. Inverse Actuarial survival of unselected octogenarians

1·0

Actuarial survival after AVR

0·9 Probability of survival

0·8 0·7 0·6 0·5

Actuarial survival after BAV

0·4 0·3

Event-free survival after BAV

0·2 0·1 0 0 Event-free survival: Total survival:

6

12

18

205 205

137 172

24 30 Months 103 70 154 135

36

42

48

51 32 16 14 123 115 109 107

Figure 54.3 Actuarial total and event-free survival among 205 patients treated by balloon aortic valvuloplasty (BAV). Shown for comparison are the actuarial survival rates among unselected octogenarians in the United States and among octogenarians who undergo aortic-valve replacement (AVR). The numbers below the figure show how many patients were alive or alive without an event at each follow-up. (Reproduced with permission from Kuntz et al.50)

788

predictors of event-free survival included pulmonary capillary wedge and pulmonary artery pressures. Although the pre- and postvalvuloplasty aortic valve area and aortic valve gradient were not associated with event-free survival, the per cent reduction in the peak aortic valve gradient was a strong predictor of long-term event-free survival. For patients with a left ventricular ejection fraction of less than 40% at baseline, improvement in the ejection fraction was also directly associated with event-free survival. Notably, when patients aged 80 or older were analyzed as a subgroup, univariate analysis indicated that the predictors of long-term event-free survival were the same in elderly patients as in the entire patient cohort. In the stepwise multivariate analysis the only independent predictors of event-free survival following balloon aortic valvuloplasty were the baseline aortic systolic pressure, the baseline pulmonary capillary wedge pressure (inversely related), and the per cent reduction in the peak aortic valve gradient. A baseline aortic systolic pressure less than 110 mmHg was associated with a relative risk of late events of 2·03, and a baseline pulmonary capillary wedge pressure greater than 25 mmHg was associated with a relative risk of 1·73, compared to the risk in patients with a baseline aortic systolic pressure greater than or equal to 140 mmHg and a pulmonary capillary wedge pressure less than 18 mmHg, respectively. Furthermore, a reduction of less than 40% in the peak aortic valve gradient was associated with a relative risk of late events of 1·75, compared to the risk in patients in whom valvuloplasty produced a reduction of 55% or more in the peak aortic valve gradient. To facilitate prediction of outcome following aortic valvuloplasty, using only information available prior to the procedure, Kuntz and colleagues utilized the two independent baseline hemodynamic predictors in the Cox model, and estimated the probability of event-free survival at 6, 12, 18 and 24 months for all patients (Table 54.3). According to this two-variable predictive model, patients with baseline pulmonary capillary wedge pressure less than 18 mmHg and aortic systolic pressure greater than or equal to 140 mmHg (the most favorable patient subgroup) had event-free survival rates of 65% at 1 year and 41% at 2 years. On the other hand, patients with baseline pulmonary capillary wedge pressure greater than 25 mmHg and aortic systolic pressure less than 110 mmHg had event-free survival rates of only 23% at 1 year and 4% at 2 years. In summary, Kuntz and colleagues found that the most important predictors of event-free survival following balloon aortic valvuloplasty were factors related to baseline left ventricular performance, a finding confirmed by analysis of long-term outcome in both large multicenter balloon aortic valvuloplasty registries.44,45 The best long-term results following valvuloplasty were observed in patients who would also have been expected to have excellent long-term results after aortic valve replacement. In fact, comparison with the


Table 54.3 Estimated event-free survival according to baseline hemodynamic variables Pre PCWP (mmHg) 18 18 18 18–25 18–25 18–25 25 25 25

Pre AOSP (mmHg)

Event-free survival (%) 6 mth

12 mth

18 mth

24 mth

140 110–139 110

79 73 64

65 58 46

51 42 30

41 31 19

140 110–139 110

74 68 58

59 50 38

43 34 22

32 23 13

140 110–139 110

63 55 43

44 35 23

28 19 10

18 10 4

Abbreviations: AOSP, aortic systolic pressure; PCWP, pulmonary capillary wedge pressure Reproduced with permission from Kuntz et al.50

surgical data on aortic valve replacement in octogenarians suggests that patients with good hemodynamic performance have better survival after aortic valve replacement than after balloon aortic valvuloplasty.19 Among patients with poor left ventricular performance or advanced heart failure, eventfree survival following balloon aortic valvuloplasty was dismal and did not appear to improve the natural history of untreated aortic stenosis.21 Therefore, even in elderly patients with advanced heart failure and higher perioperative risk,13 aortic valve replacement may increase the likelihood of long-term survival compared to balloon aortic valvuloplasty. In such high-risk patients, however, balloon aortic valvuloplasty may have a role in providing transient hemodynamic improvement, perhaps decreasing the risk of subsequent aortic valve replacement. Repeat balloon aortic valvuloplasty Grade B/C In patients who are not candidates for surgery the development of restenosis following balloon aortic valvuloplasty can be managed with a repeat procedure. Studies of repeat valvuloplasty have shown that the absolute aortic valve area tends to be slightly smaller both before and after the repeat valvuloplasty, even when larger balloons or balloon combinations are used.51 The incidence of repeat restenosis remains high: follow up of the 47 patients in the Mansfield Registry who underwent repeat valvuloplasty showed that 66% of patients had died, undergone subsequent valve replacement or required a third valvuloplasty at a mean follow up of 5 months.52 Histologic study of valves treated with balloon valvuloplasty, and excised prior to subsequent surgery or examined at autopsy, has shown active cellular proliferation within the splits in calcified nodules, as well as foci of ossification.48 These findings suggest

an active scarring process in response to balloon valvuloplasty, which may explain the failure to achieve better results with the use of larger balloons, and raises the possibility that balloon-induced injury to the aortic valve may accelerate the natural history of aortic stenosis. Aortic valve surgery after balloon aortic valvuloplasty Grade B Most surviving patients who have undergone balloon aortic valvuloplasty develop clinically significant restenosis within 1–2 years of the procedure. Many of these patients are subsequently treated with aortic valve replacement. Johnson and colleagues at the Beth Israel Hospital reported 45 patients (25% of the initial balloon valvuloplasty cohort) subsequently treated with aortic valve replacement.53 Three patients required emergency operation immediately after unsuccessful valvuloplasty, and the remaining 42 had an elective operation at a mean of 8 months following valvuloplasty, primarily for the development of symptomatic restenosis. Despite the fact that the majority of these patients had initially undergone balloon valvuloplasty because they were considered to be at high risk for surgery, there were only four hospital deaths among the 45 patients. Three additional patients died a mean of 11 months following surgery. All surviving patients had persistent symptomatic improvement following surgery. Lieberman and colleagues at Duke reported 40 patients (24% of the initial balloon valvuloplasty treatment group) who subsequently underwent aortic valve replacement.54 Only one patient (2·5%) suffered a perioperative death. The probability of survival 3 years from the date of the last mechanical intervention was 75% for patients treated with balloon valvuloplasty and subsequent aortic valve 789


replacement, compared to only 20% for patients whose restenosis was treated with repeat balloon valvuloplasty, and 13% for patients who had no further mechanical intervention after developing restenosis. The majority of surgically treated patients remained asymptomatic at last follow up. It is important to note that this study is not a randomized comparison of treatment strategies for restenosis, and the results must be interpreted in light of the probable selection bias with regard to choice of management strategy for aortic valve restenosis. Nevertheless, it appears that in this group of patients initially felt to be at high risk for aortic valve replacement, surgery could be performed with an acceptable operative risk. Furthermore, as opposed to balloon valvuloplasty, aortic valve replacement appears to offer a reasonable chance of long-term freedom from symptoms. Although these reports do not specifically address potential reduction in the risk of subsequent surgery by prior performance of balloon valvuloplasty, a beneficial effect cannot be excluded.

Balloon aortic valvuloplasty v aortic valve surgery Grade B There are no randomized trials comparing balloon aortic valvuloplasty with aortic valve replacement in adult patients with critical aortic stenosis. However, Bernard and colleagues in France compared two non-randomized matched series of patients with aortic stenosis treated with either balloon aortic valvuloplasty or aortic valve replacement at the same institution between January 1986 and March 1989.55 Forty-six patients were treated with balloon aortic valvuloplasty and 23 with aortic valve replacement with a bioprosthesis. Baseline clinical and hemodynamic parameters were similar in both groups; all patients were at least 75 years old. Follow-up was 22 months for the aortic valvuloplasty patients and 28 months for those having surgery. Among patients treated with balloon aortic valvuloplasty, three patients (6·5%) died within 5 days of the procedure, and an additional 24 (42%) died during subsequent follow up, with 16 deaths being due to recurrent heart failure. Sixteen patients (35%) underwent subsequent aortic valve replacement at a mean of 16 months following balloon valvuloplasty. At last follow up, only three valvuloplasty patients (6·5%) remained alive without subsequent aortic valve replacement. Of the patients treated with initial aortic valve replacement, two (8·7%) died in the perioperative period and an additional three (13%) died during the follow up period. All remaining patients (78%) were alive and in New York Heart Association functional class I or II at last follow up. The overall survival rate following balloon valvuloplasty was 75% at 1 year, 47% at 2 years and 33% at 5 years. By contrast, survival following surgery was 83% at 1 and 2 years and 75% at 3 and 4 years. Although selection 790

bias cannot be excluded in this non-randomized case comparison study, nevertheless the results strongly suggest that percutaneous balloon aortic valvuloplasty does not compare favorably with aortic valve surgery in elderly patients with aortic stenosis. Specific indications for balloon valvuloplasty Aortic valvuloplasty prior to non-cardiac surgery Grade B/C Patients with severe aortic stenosis are at increased risk for significant cardiac complications during non-cardiac surgery.56 Three studies described the role of balloon aortic valvuloplasty in the management of patients with critical aortic stenosis requiring major non-cardiac surgery.57–59 In these studies, 29 patients with critical aortic stenosis underwent balloon aortic valvuloplasty which was complicated by procedural death due to ventricular perforation and tamponade in one patient. Valvuloplasty resulted in a significant improvement in aortic valve gradient and aortic valve area. Twenty-eight of the 29 patients underwent the planned surgical procedure under general or epidural anesthesia. All but one patient had uncomplicated non-cardiac surgery, with no significant congestive heart failure, hypotension, myocardial infarction, arrhythmia or conduction abnormality either during or immediately after surgery. One patient developed marked hypotension requiring transient intravenous pressor support during surgery for bowel carcinoma, resulting in interruption of surgery. This patient subsequently underwent aortic valve replacement and coronary artery bypass graft surgery, followed by repeat bowel resection. Procedures performed successfully following palliative balloon aortic valvuloplasty included aortic aneurysm repair, repair of hip fracture, exploratory laparotomy and thoracotomy. However, the cited reports are not randomized or case–control comparisons of preoperative balloon aortic valvuloplasty versus aortic valve replacement or medical therapy, and do not test the hypothesis that routine balloon valvuloplasty reduces the risk of non-cardiac surgery in patients with critical aortic stenosis. O’Keefe and colleagues60 at the Mayo Clinic described 48 patients with severe aortic stenosis who underwent non-cardiac surgery (including vascular, orthopedic and abdominal procedures) without preoperative balloon valvuloplasty. There were no major perioperative complications in this group, who were managed with careful monitoring of systemic and pulmonary artery pressure during anesthesia. Therefore, the available evidence suggests that balloon valvuloplasty prior to urgent non-cardiac surgery may have greatest benefit in those patients with critical aortic stenosis and poor ventricular function, heart failure or hypotension, in whom transient hemodynamic improvement may decrease the risk of perioperative complications.


Aortic valvuloplasty as a bridge to aortic valve replacement Grade B/C As noted earlier, many patients treated with balloon aortic valvuloplasty subsequently undergo aortic valve replacement. Early series of such patients demonstrated an acceptable operative risk and excellent surgical outcome, with long-term freedom from symptoms in most survivors.53,54 In contrast, recent reports of cardiac surgery in octogenarians identified previous percutaneous aortic valvuloplasty as an independent predictor of hospital death following valve replacement.61,62 However, in most patients undergoing surgery in these studies, valve replacement was performed because of failure of the initial balloon aortic valvuloplasty, which was not specifically used to stabilize the patient for subsequent surgery. Smedira and colleagues63 studied critically ill patients with aortic stenosis in whom balloon aortic valvuloplasty was specifically used as a bridge to aortic valve replacement. They reported five patients with severe aortic stenosis, multiple organ failure and severe hemodynamic compromise who were judged to be at excessive risk for aortic valve surgery. Balloon aortic valvuloplasty was used in these patients to provide transient hemodynamic improvement, to improve organ function, and to decrease the risk of subsequent definitive surgical correction. Following successful balloon aortic valvuloplasty and clinical stabilization, subsequent elective valve replacement was performed in all patients without complications. This report suggests that balloon aortic valvuloplasty may have a role as a bridge to subsequent aortic valve replacement for patients in whom heart failure or hypotension is so severe that the risk of primary aortic valve surgery is unacceptable. Aortic valvuloplasty in cardiogenic shock Grade C Of the 674 patients in the multicenter NHLBI Balloon Valvuloplasty Registry, 39 (6%) had cardiogenic shock. The largest reported series specifically describing the role of balloon aortic valvuloplasty in cardiogenic shock is that of Moreno and colleagues from the Massachusetts General Hospital. Moreno64 studied 21 patients with critical aortic stenosis and cardiogenic shock treated with balloon aortic valvuloplasty. All patients had major associated comorbid conditions precluding the use of emergency aortic valve replacement. The hemodynamic results were excellent, with an increase in systolic aortic pressure from 77 to 116 mmHg and an increase in aortic valve area from 0·5 to 0·8 cm2 (P 0·0001 for both comparisons). Cardiac index increased from 1·84 to 2·24 l/min/m2 (P 0·06). Nine treated patients died in hospital, two during the procedure and seven following successful valvuloplasty. Procedural complications were frequent, with five patients suffering vascular complications and one patient each developing stroke,

cholesterol embolus and aortic regurgitation requiring aortic valve replacement. Twelve patients (57%) survived and were discharged from the hospital. During follow up of 15 months, five additional patients died. Actuarial survival was 38% at 27 months. The only predictor of improved survival was the postprocedure cardiac index. In summary, the limited published data suggest that emergency percutaneous balloon aortic valvuloplasty can be successfully performed in patients with critical aortic stenosis and cardiogenic shock. Morbidity and mortality remain high even after hemodynamically successful procedures. Given the poor long-term outcome in patients treated with balloon aortic valvuloplasty, its use in patients with cardiogenic shock should be considered a bridge to subsequent aortic valve replacement in those patients who improve sufficiently to undergo surgery at reasonable risk. Aortic valvuloplasty in patients with low output, low gradient Grade B Patients with left ventricular dysfunction and aortic stenosis in the presence of low cardiac output and low aortic valve gradient present a complex diagnostic and therapeutic challenge. Aortic valve surgery is associated with increased morbidity and mortality in such patients, a subset of whom have irreversible myocardial dysfunction.10–12 Balloon aortic valvuloplasty has been proposed as a diagnostic tool in patients with aortic stenosis and low-output lowgradient hemodynamics, to distinguish those with reversible myocardial dysfunction due to abnormal loading conditions from those with irreversible myocardial dysfunction. It has been suggested that patients with low-output low-gradient hemodynamics who have a significant improvement in either ventricular function or symptoms following successful balloon aortic valvuloplasty are more likely to improve following aortic valve replacement than those patients in whom the former produces no significant benefit. Safian and colleagues studied 28 patients with a low left ventricular ejection fraction (mean 37%) and severe aortic stenosis who underwent balloon aortic valvuloplasty.42 On the basis of response to balloon valvuloplasty they were able to separate patients into a subset with progressive improvement in left ventricular ejection fraction, and a subset which showed no significant change in left ventricular function. Nishimura and colleagues, utilizing data from the multicenter Mansfield Aortic Valvuloplasty Registry, compared 67 patients with low-output low-gradient hemodynamics against 200 patients with a low cardiac index but not a low aortic valve gradient.65 Patients with low-output low-gradient hemodynamics had less of a decrease in aortic valve gradient after valvuloplasty, but a similar improvement in estimated aortic valve area. However, actuarial survival at 12 months was 46% for these patients, as against 64% in the comparison cohort (P 0·05). Furthermore, patients with 791


low-gradient hemodynamics were less likely to show sustained symptomatic improvement. Therefore, as longterm outcome after balloon valvuloplasty is poor in these patients aortic valve replacement may be indicated in those in whom balloon aortic valvuloplasty produces an initial favorable response. Although these reports suggest that it may be possible to identify a subset of patients with aortic stenosis and low-output low-gradient hemodynamics likely to benefit from subsequent aortic valve replacement, the hypothesis that response to aortic valvuloplasty predicts subsequent outcome following surgery has not been tested. Other indications Grade C Case reports have described the use of balloon aortic valvuloplasty for the management of critical aortic stenosis in pregnancy, documenting its safe performance during pregnancy with subsequent normal births.66 Given their age range, pregnant patients are more likely to have congenital or rheumatic aortic stenosis and therefore to have valve stenosis due to commissural fusion, which responds more favorably to balloon dilation than does the more frequently encountered degenerative calcific valvular disease. Use of balloon aortic valvuloplasty as a bridge to subsequent cardiac transplant in a patient with aortic stenosis and end-stage heart failure has also been described.67 Indications for balloon aortic valvuloplasty ● ● ● ●

Symptomatic critical aortic stenosis in patients who are not candidates for aortic valve replacement Bridge to aortic valve replacement in patients with severe hemodynamic compromise Prior to urgent non-cardiac surgery Aortic stenosis with low-output low-gradient hemodynamics.

Conclusions The development and analysis of balloon aortic valvuloplasty as a treatment strategy for adult patients with critical aortic stenosis offers a paradigm for the investigation of new therapeutic techniques. The initial enthusiasm for new treatment modalities, often based on arguments of physiology, first principles or small case series, is often replaced by a sobering realization of limitations and complications, revealed by careful prospective multicenter clinical trials, ultimately resulting in the development of appropriate clinical indications for the new treatment strategy. The development and investigation of balloon aortic valvuloplasty for aortic stenosis followed just such a course and illustrates the impact of careful, early prospective clinical trial data on the evolution and rapid development of appropriate indications for new therapeutic techniques. 792

Although valve replacement clearly improves morbidity and mortality in patients with symptomatic aortic stenosis, concern regarding the higher morbidity in high-risk subgroups led to the investigation of balloon aortic valvuloplasty as an alternative. Early evidence from both single- and multicenter series showing hemodynamic and symptomatic improvement in most patients treated with balloon valvuloplasty, led to initial widespread enthusiasm for this new technique. However, this enthusiasm was quickly tempered as subsequent follow up in these high-quality case series demonstrated a high rate of hemodynamic and clinical restenosis, and failure of balloon valvuloplasty to improve long-term or event-free survival. Critical evaluation of the data from these large case series provided further understanding of the appropriate role of balloon valvuloplasty in the management of patients with aortic stenosis. When patients were stratified by the independent predictors of event-free survival, it became clear that those who did best with balloon aortic valvuloplasty were acceptable candidates for valve surgery and had an even better event-free survival following surgery. On the other hand, patients with baseline profiles that indicated a high risk for surgery also did extremely poorly with balloon valvuloplasty, with event-free survival that did not appear to differ from the natural history of untreated aortic stenosis. The rapid accumulation and careful analysis of clinical trial data on patients treated with balloon valvuloplasty quickly established that the treatment of choice for adult patients with symptomatic aortic stenosis is valve replacement, with balloon valvuloplasty being reserved for those in whom surgery is not possible or practical. Further refinement of the appropriate therapeutic niche for balloon aortic valvuloplasty has been aided by small case series targeted at specific indications for non-surgical therapy of aortic stenosis. The following guidelines on appropriate utilization of balloon aortic valvuloplasty in adult patients with symptomatic critical aortic stenosis are based on case series and case–control studies, and therefore should be considered as Grade B recommendations. Based on the available evidence, balloon aortic valvuloplasty should be considered: 1.

For patients with symptomatic aortic stenosis who are not operable, or who are poor candidates for aortic valve replacement owing to severe comorbid illness or advanced age in the presence of other significant predictors of surgical risk. It should be emphasized that advanced age alone in a patient without other significant surgical risk factors is not a contraindication to aortic valve replacement. It must be further stressed that the goal of balloon aortic valvuloplasty in this patient group is transient symptomatic relief, as there is no evidence that valvuloplasty improves survival or provides long-term freedom from symptoms. Grade B


2.

3.

4.

As a bridge to subsequent aortic valve replacement in patients with advanced heart failure, hypotension or cardiogenic shock, when clinical presentation suggests excessive risk for an initial surgical strategy. The goal of balloon aortic valvuloplasty in this cohort is transient hemodynamic improvement, leading to stabilization of the patient for subsequent aortic valve replacement, the only treatment shown to ultimately improve longterm survival. Grade B For patients with critical aortic stenosis and poor ventricular function, heart failure or hypotension who require urgent or emergency non-cardiac surgery. The goal of balloon aortic valvuloplasty in this patient subset is successful completion of the required noncardiac surgical procedure, with subsequent aortic valve replacement for the underlying aortic stenosis. Grade B For patients with aortic stenosis, diminished left ventricular function and low-output low-gradient hemodynamics, in whom the response to initial “diagnostic” balloon valvuloplasty may help identify those likely to benefit from subsequent aortic valve replacement. Grade B

Given the disparity in outcome between aortic valve replacement and balloon aortic valvuloplasty in large highquality case series and non-randomized case–control studies, it is unreasonable to pursue randomized clinical trials comparing these treatment strategies. However, the highquality case series rapidly performed and reported in patients treated with balloon aortic valvuloplasty not only established the appropriate role for balloon valvuloplasty in the treatment of aortic stenosis, but also confirmed the value of prompt clinical investigation in the rapid development of appropriate indications for new therapeutic techniques. When the goal of therapy is long-term or symptom-free survival, the available clinical trial data clearly support valve replacement as the treatment of choice for aortic stenosis. However, in patients who are not candidates for or who refuse surgery, the trial data have demonstrated a role for balloon aortic valvuloplasty, albeit with the more limited goal of transient, palliative symptomatic relief, without improvement in survival or long-term symptomatic benefit. References 1.Passik CS, Ackermann DM, Pluth JR, Edwards WD. Temporal changes in the causes of aortic stenosis: a surgical pathologic study of 646 cases. Mayo Clin Proc 1987;62:119–23. 2.Frank S, Johnson A, Ross J. Natural history of valvular aortic stenosis. Br Heart J 1973;35:41–6. 3.Ross J, Braunwald E. Aortic stenosis. Circulation 1968;38 (Suppl. V):61–7.

4.Pellikka PA, Nishimura RA, Bailey KR, Tajik AJ. The natural history of adults with asymptomatic, hemodynamically significant aortic stenosis. J Am Coll Cardiol 1990;15:1012–17. 5.Otto CM, Burwash IG, Legget ME et al. Prospective study of asymptomatic valvular aortic stenosis. Clinical, echocardiographic, and exercise predictors of outcome. Circulation 1997;95:2262–70. 6.Carabello BA. Timing of valve replacement in aortic stenosis. Moving closer to perfection. Circulation 1997;95:2241–3. 7.Hsieh K, Keane JF, Nadas AS, Bernhard WF, Castaneda AR. Long-term follow-up of valvotomy before 1968 for congenital aortic stenosis. Am J Cardiol 1986;58:338–41. 8.McBride LR, Nannheim KS, Fiore AC et al. Aortic valve decalcification. J Thorac Cardiovasc Surg 1990;100:36–42. 9.Kennedy JW, Doces J, Stewart DK. Left ventricular function before and following aortic valve replacement. Circulation 1977;56:944–50. 10.Smith N, McAnulty JH, Rahimtoola SH. Severe aortic stenosis with impaired left ventricular function and clinical heart failure: results of valve replacement. Circulation 1978;58: 255–64. 11.Lund O. Preoperative risk evaluation and stratification of longterm survival after valve replacement for aortic stenosis. Circulation 1990;82:124–39. 12.Magovern JA, Pennock JL, Campbell DB et al. Aortic valve replacement and combined aortic valve replacement and coronary artery bypass grafting: predicting high risk groups. J Am Coll Cardiol 1987;9:38–43. 13.Edmunds LH, Stephenson LW, Edie RN, Ratcliffe MB. Openheart surgery in octogenarians. N Engl J Med 1988;319: 131–6. 14.Verheul HA, Van Den Brink RBA, Bouma BJ et al. Analysis of risk factors for excess mortality after aortic valve replacement. J Am Coll Cardiol 1995;26:1280–6. 15.Gehlot A, Mullany CJ, Ilstrup D et al. Aortic valve replacement in patients aged eighty years and older: early and long-term results. J Thorac Cardiovasc Surg 1996;111:1026–36. 16.Asimakopoulos G, Edwards M, Taylor K. Aortic valve replacement in patients 80 years of age and older. Survival and cause of death based on 1100 cases: collective results from the UK Heart Valve Registry. Circulation 1997;96:3403–8. 17.Edwards FH, Peterson ED, Coombs LP et al. Prediction of operative mortality after valve replacement surgery. J Am Coll Cardiol 2001;37:885–92. 18.Fremes SE, Goldman BS, Ivanou J, Weisel RD, David TE, Salerno T. Valvular surgery in the elderly. Circulation 1989; 80(Suppl. I):177–90. 19.Levinson JR, Akins CW, Buckley MJ et al. Octogenarians with aortic stenosis. Outcome after aortic valve replacement. Circulation 1989;80(Suppl. I):149–56. 20.Rosengart TK, Finnin EB, Kim DY et al. Open heart surgery in the elderly: results from a consecutive series of 100 patients aged 85 years or older. Am J Med 2002;112:143–77. 21.O’Keefe JH, Vlietstra RE, Bailey KR, Holmes DR. Natural history of candidates for balloon aortic valvuloplasty. Mayo Clin Proc 1987;62:986–91. 22.Kirklin JW, Barratt-Boyes BG. Congenital aortic stenosis. In: Cardiac Surgery, 2nd edn. New York: Churchill Livingstone, 1993.

793


23.Lababidi Z, Wu JR, Walls JT. Percutaneous balloon aortic valvuloplasty: results in 23 patients. Am J Cardiol 1984; 53:194–7. 24.Rosenfeld HM, Landzberg MJ, Perry SB, Colan SD, Keane JF, Lock JE. Balloon aortic valvuloplasty in young adults with congenital aortic stenosis. Am J Cardiol 1994;73:1112–17. 25.Tomita H, Echigo S, Kimura K et al. Balloon aortic valvuloplasty in children: a multicenter study in Japan. Jpn Circ J 2001;65:599–602. 26.Cribier A, Savin T, Saondi N, Rocha P, Berland J, Letac B. Percutaneous transluminal valvuloplasty of acquired aortic stenosis in elderly patients: an alternative to valve replacement? Lancet 1986;i:63–7. 27.McKay RG, Safian RD, Lock JE et al. Balloon dilatation of calcific aortic stenosis in elderly patients: postmortem, intraoperative, and percutaneous valvuloplasty studies. Circulation 1986;74:119–25. 28.Safian RD, Mandell VS, Thurer RE et al. Postmortem and intraoperative balloon valvuloplasty of calcific aortic stenosis in elderly patients: mechanisms of successful dilation. J Am Coll Cardiol 1987;9:655–60. 29.Block PC, Palacios IF. Comparison of hemodynamic results of anterograde versus retrograde percutaneous balloon aortic valvuloplasty. Am J Cardiol 1987;60:659–62. 30.Eisenhauer AC, Hadjipetrou P, Piemonte TC. Balloon aortic valvuloplasty revisited: the role of the Inoue balloon and transseptal antegrade approach. Cathet Cardiovasc Interv 2000;50:484–91. 31.Bahl VK, Chandra S, Goswami KC. Combined mitral and aortic valvuloplasty by the antegrade transseptal approach using the Inoue balloon catheter. Int J Cardiol 1998;63:313–15. 32.Dorros G, Lewin RF, King JF, Janke LM. Percutaneous transluminal valvuloplasty in calcific aortic stenosis: the double balloon technique. Cathet Cardiovasc Diagn 1987;13:151–6. 33.Fields CD, Lucas A, Desnoyers M et al. Dual balloon aortic valvuloplasty, despite augmenting acute hemodynamic improvement, fails to prevent post-valvuloplasty restenosis. J Am Coll Cardiol 1989;13:148A. 34.Solomon LW, Fusman B, Jolly N, Kim A, Feldman T. Percutaneous suture closure for management of large French size arterial puncture in aortic valvuloplasty. J Invas Cardiol 2001;13:592–6. 35.Michaels AD, Ports TA. Use of a percutaneous arterial suture device (Perclose) in patients undergoing percutaneous balloon aortic valvuloplasty. Cathet Cardiovasc Interv 2001;53:445–7. 36.Cribier A, Savin T, Berland J et al. Percutaneous transluminal balloon valvuloplasty of adult aortic stenosis: report of 92 cases. J Am Coll Cardiol 1987;9:381–6. 37.Safian RD, Berman AD, Diver DJ et al. Balloon aortic valvuloplasty in 170 consecutive patients. N Engl J Med 1988;319: 125–30. 38.Block PC, Palacios IF. Clinical and hemodynamic follow-up after percutaneous aortic valvuloplasty in the elderly. Am J Cardiol 1988;62:760–3. 39.Lewin RF, Dorros G, King JF, Mathiak L. Percutaneous transluminal aortic valvuloplasty: acute outcome and follow-up of 125 patients. J Am Coll Cardiol 1989;14:1210–17. 40.McKay RG, for the Mansfield Scientific Aortic Valvuloplasty Registry. Balloon aortic valvuloplasty in 285 patients: initial

794

results and complications. Circulation 1988;78(Suppl. II): II–594. 41.McKay RG, for the NHLBI Aortic Valvuloplasty Registry. Clinical outcome following balloon aortic valvuloplasty for severe aortic stenosis. J Am Coll Cardiol 1989;13:1218. 42.Safian RD, Warren SE, Berman AD et al. Improvement in symptoms and left ventricular performance after balloon aortic valvuloplasty in patients with aortic stenosis and depressed left ventricular ejection fraction. Circulation 1988;78:1181–91. 43.Davidson CJ, Harrison JK, Leithe ME, Kisslo KB, Bashore TM. Failure of balloon aortic valvuloplasty to result in sustained clinical improvement in patients with depressed left ventricular function. Am J Cardiol 1990;65:72–7. 44.O’Neill WW, for the Mansfield Scientific Aortic Valvuloplasty Registry Investigators. Predictors of long-term survival after percutaneous aortic valvuloplasty: report of the Mansfield Scientific Aortic Valvuloplasty Registry. J Am Coll Cardiol 1991;17:193–8. 45.Otto CM, Mickel MC, Kenedy JW et al. Three year outcome after balloon aortic valvuloplasty. Insights into prognosis of valvular aortic stenosis. Circulation 1994;89:642–50. 46.Lieberman EB, Bashore TM, Hermiller JB et al. Balloon aortic valvuloplasty in adults: failure of procedure to improve longterm survival. J Am Coll Cardiol 1995;26:1522–8. 47.Nishimura RA, Holmes DR, Reeder GS et al. Doppler evaluation of results of percutaneous aortic balloon valvuloplasty in calcific aortic stenosis. Circulation 1988;78:791–9. 48.Feldman T, Glagov S, Carroll JD. Restenosis following successful balloon valvuloplasty: bone formation in aortic valve leaflets. Cathet Cardiovasc Diagn 1993;29:1–7. 49.Isner JM. Aortic valvuloplasty: are balloon-dilated valves all they are “cracked” up to be? Mayo Clin Proc 1988;63: 830–4. 50.Kuntz RE, Tosteson AN, Berman AD et al. Predictors of eventfree survival after balloon aortic valvuloplasty. N Engl J Med 1991;325:17–23. 51.Kuntz RE, Tosteson AN, Maitland LA et al. Immediate results and long-term follow-up after repeat balloon aortic valvuloplasty. Cathet Cardiovasc Diagn 1992;25:4–9. 52.Ferguson JJ, Garza RA, and the Mansfield Scientific Aortic Valvuloplasty Registry Investigators. Efficacy of multiple balloon aortic valvuloplasty procedures. J Am Coll Cardiol 1991;17:1430–5. 53.Johnson RG, Dhillon JS, Thurer RL, Safian RD, Wientraub RM. Aortic valve operation after percutaneous aortic balloon valvuloplasty. Ann Thorac Surg 1990;49:740–5. 54.Lieberman EB, Wilson JS, Harrison JK et al. Aortic valve replacement in adults after balloon aortic valvuloplasty. Circulation 1994;90(Suppl. II):II205–8. 55.Bernard Y, Etievent J, Mourand JL et al. Long-term results of percutaneous aortic valvuloplasty compared with aortic valve replacement in patients more than 75 years old. J Am Coll Cardiol 1992;20:796–801. 56.Goldman L, Caldera DL, Nussbaum SR. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297:845–56. 57.Levine MJ, Berman AD, Safian RD, Diver DJ, McKay RG. Palliation of valvular aortic stenosis by balloon valvuloplasty as preoperative preparation for noncardiac surgery. J Am Coll Cardiol 1988;62:1309–10.


58.Roth RB, Palacios IF, Block PC. Percutaneous aortic balloon valvuloplasty: its role in the management of patients with aortic stenosis requiring major noncardiac surgery. J Am Coll Cardiol 1989;13:1039–41. 59.Hayes SN, Holmes DR, Nishimura RA, Reeder GS. Palliative percutaneous aortic balloon valvuloplasty before noncardiac operations and invasive diagnostic procedures. Mayo Clin Proc 1989;64:753–7. 60.O’Keefe JH, Shub C, Pettke SR. Risk of noncardiac surgical procedures in patients with aortic stenosis. Mayo Clin Proc 1989;64:400–5. 61.Kohl P, Lahaye L, Gerard P, Limet R. Aortic valve replacement in octogenarians: perioperative outcome and clinical follow-up. Eur J Cardiovasc Surg 1999;16:68–73. 62.Kolh P, Kerzmann A, Lahaye L, Gerard P, Limet R. Cardiac surgery in octogenarians: peri-operative outcome and long-term results. Eur Heart J 2001;22:1235–43. 63.Smedira NG, Ports TA, Merrick SH, Rankin JS. Balloon aortic valvuloplasty as a bridge to aortic valve replacement

in critically ill patients. Ann Thorac Surg 1993;55: 914–16. 64.Moreno PR, Ik-Kyung J, Block PC, Palacios IF. The role of percutaneous aortic balloon valvuloplasty in patients with cardiogenic shock and critical aortic stenosis. J Am Coll Cardiol 1994;23:1071–5. 65.Nishimura RA, Holmes DR, Michela ME et al. Follow-up of patients with low output, low gradient hemodynamics after percutaneous balloon aortic valvuloplasty: the Mansfield Scientific Aortic Valvuloplasty Registry. J Am Coll Cardiol 1991;17:828–33. 66.Banning AP, Pearson JF, Hall RJ. Role of balloon dilatation of the aortic valve in pregnant patients with severe aortic stenosis. Br Heart J 1993;70:544–5. 67.Vaitkus PT, Mancini D, Herrman HC. Percutaneous balloon aortic valvuloplasty as a bridge to heart transplantation. J Heart Lung Transplant 1993;12:1062–4.

795

55

Balloon valvuloplasty: mitral valve Zoltan G Turi

Introduction

Mechanisms of valvuloplasty

Percutaneous balloon mitral valvuloplasty is the latest technique in an evolution that began with Elliot Cutler advancing a knife retrograde through the apex of the left ventricle of a beating heart in 1923.1 Neither he nor Henry Suttar, who performed a similar procedure in England two years later received the expected accol ades,2 and there has been continuing dispute about the relative role of mitral obstruction in defining the spectrum of mitral stenosis. Sir Thomas Lewis’ statement that valvotomy was based on an erroneous idea, namely that the valve is the chief source of the trouble3 has few proponents in the modern era and relieving mitral obstruction is the de facto standard of care. After a 20 year hiatus, the battlefield experience with closed heart procedures in the second world war led to the application of these techniques outside the trauma arena. Although early results were confounded by significant morbidity and mortality, closed mitral valvotomy became a routine procedure for severe mitral stenosis, and is still the treatment of choice in many parts of the world where the disease is endemic and medical facilities limited. Large series4,5 have claimed good long-term results, but lack of systematic follow up or comprehensive objective data obscure the actual restenosis rate and survival. In a Mayo Clinic retrospective analysis6 there was 79% 10 year and 55% 20 year survival rate with reoperation in 34% by 10 years; however nearly a quarter of patients were lost to follow up and severity of disease at baseline could only be estimated. Open commissurotomy with the potential advantages of direct vision has supplanted closed procedures in industrialized nations. Controversy remains as to its superiority7–9 with the advantages of direct vision favoring cases where thrombus is present.

The mechanisms responsible for the benefits of balloon mitral valvuloplasty13 arise from the substantial radial force exerted by the enlarging balloon.14 This stretches the mitral annulus, has the capacity to split fused commissures, and occasionally results in the cracking of calcifications. The stretching mechanism has been observed intraoperatively,15 whereas the splitting of commissures16 and cracking of

The percutaneous approach A pediatric cardiac surgeon, Kanji Inoue, developed a double lumen atrial septostomy balloon catheter made of latex, with a mesh weave used to constrain the balloon during inflation into the classic wishbone shape depicted in Figure 55.1.10 He then adapted the device for percutaneous balloon mitral valvuloplasty, demonstrated under direct vision in the operating room its ability to split fused mitral commissures11 and performed the first procedure in 1982.12 796

Figure 55.1 The Inoue balloon during staged deployment. From top to bottom: distal inflation with pullback against the valve; proximal inflation; full deployment. (Reprinted with permission of the American Heart Association, Inc.38)

Balloon valvuloplasty: mitral valve

calcifications have been demonstrated by direct observation in excised valves.17 The largely successful nature of balloon mitral valvuloplasty is derived from commissural splitting; balloon dilatation procedures where the other two mechanisms predominate, such as balloon valvuloplasty for calcific aortic stenosis, have less impressive short- and long-term results.

Preprocedure evaluation The most common reason for exclusion of patients is unsuitable valve anatomy. Specific relevant physical examination findings are diminution of the first heart sound (often indicative of extensive subvalvular disease) and a hyperdynamic ventricle, suggestive of volume loading secondary to mitral or aortic regurgitation, both of which are relative contraindications to the procedure. Non-invasive methods The echocardiographic findings of greatest predictive value have been debated at length. The standard,18 the Wilkins-Weyman score, incorporates a scoring system for mitral valve leaflet thickening, mobility and calcification, and severity of subvalvular disease (Table 55.1), with a score of 8 described as an “ideal” patient population, and echo scores over 12 potentially predicting poorer results. The correlation between this echo score and initial as well as long-term results is only fair, perhaps because it is a semiquantitative system based on partly subjective assessments

and because other factors not included in the system have predictive value. Thus studies have alternately confirmed19–21 or refuted the predictive value of the Wilkins-Weyman score.22–25 One element of the score, leaflet mobility, correlates more strongly with outcome (r value 0·67) than the complete score,26 while another element, severe calcification of the valve,27 alone predicts a fourfold increase in cardiac complications and a 26% increase in 6 year mortality. In addition important anatomic features that predict outcome, such as eccentricity of commissural fusion and a funnel shaped subvalvular apparatus28 (both negative predictors) are not included. Neither are presence of moderate or severe mitral regurgitation or left atrial thrombus, both relative contraindications. In univariate analysis, the scoring system does predict long-term results,20 but so do age, presence of atrial fibrillation,27 and severity of stenosis before and after the procedure.29 Further, multivariate analyses that included the echo score but not its individual components, failed to demonstrate a single preprocedure predictor of event free survival.30 Multivariate analysis that includes commissural calcification did reveal this to be a strong predictor of death, restenosis, and mitral valve replacement.31 Perhaps the most compelling reason for routinely deriving the echo score is to allow for comparison with known data; most mitral valvuloplasty trials incorporate this or similar scoring systems. However, no absolute predictors of short- and long-term outcome have been developed. Routine, preprocedure, transesophageal echocardiography has been recommended because of its superiority for detection of left atrial thrombus,32 as well as other structural

Table 55.1 Grading of mitral valve characteristics from the echocardiographic examination Grade

Mobility

Subvalvar thickening

Thickening

Calcification

1

Highly mobile valve with only leaflet tips restricted Leaflet mid and base portions have normal mobility

Minimal thickening just below the mitral leaflets Thickening of chordal structures extending up to one third of the chordal length Thickening extending to the distal third of the chords Extensive thickening and shortening of all chordal structures extending down to the papillary muscles

Leaflets near normal in thickness (4–5 mm) Midleaflets normal, considerable thickening of margins (5–8 mm)

A single area of increased echo brightness Scattered areas of brightness confined to leaflet margins

Thickening extending through the entire leaflet (5–8 mm) Considerable thickening of all leaflet tissue (8–10 mm)

Brightness extending into the midportion of the leaflets Extensive brightness throughout much of the leaflet tissue

2

3

4

Valve continues to move forward in diastole, mainly from the base No or minimal forward movement of the leaflets in diastole

Note. The total echocardiographic score was derived from an analysis of mitral leaflet mobility, valvar and subvalvar thickening, and calcification which were graded from 0 to 4 according to the above criteria. The total possible score ranges from 0 to 16. Reprinted with permission from Wilkins GT, Weyman AE, Abascal VM et al. Percutaneous balloon dilation of the mitral valve: an analysis of echocardiographic variables related to outcome and the mechanism of dilation. Br Heart J 60:299–309. © 1988 by the BMJ Publishing Group.18

797


abnormalities including vegetations or ruptured chordae. The case is most compelling in patients predisposed to clot formation such as those with spontaneous echo contrast (“smoke”) on surface echocardiography and those with atrial fibrillation. The former was an independent predictor of left atrial thrombus in a prospective study of 100 patients.33

free survival but a similar success rate.37 Thus, the evidence suggests that balloon commissurotomy can still be considered for these patients if they are poor risks for heart surgery. Nevertheless, a theoretical disadvantage is additional volume loading of the left ventricle when antegrade flow is improved after balloon commissurotomy, a concern in the presence of aortic regurgitation as well.

Cardiac catheterization Cardiac catheterization prior to balloon commissurotomy is rarely necessary in young patients, but can be beneficial to exclude coronary artery disease in older subjects. The gradient alone is a poor proxy for assessment of severity of disease pre-valvuloplasty since it can lead to overestimation of disease with poor heart rate control or underestimation in patients who have not had fluids for many hours prior to catheterization.

Contraindications While the usually cited contraindications are left atrial thrombus, greater than mild mitral regurgitation and severe calcification or subvalvular disease, these were largely empirically derived and can be challenged.

Severe calcification Patients with symmetrical severe calcification may not respond at all to balloon commissurotomy;22,38 those with asymmetric calcification are prone to leaflet tearing or rupture.28 While high echo score alone does not predict the occurrence of severe mitral regurgitation,39 one component, severe calcification, does.40 Nevertheless, when the risk of surgery is prohibitive, growing experience with predominantly elderly patients with high echo scores and poor overall morphology has shown moderate improvement in hemodynamics and palliation of symptoms at the cost of high morbidity and mortality.41

Procedure Antegrade v retrograde approaches

Thrombus 34

Hung and others have described at least three series exceeding 90 patients total with apparent organized left atrial appendage clot who underwent uncomplicated balloon commissurotomy. However, valvuloplasty is not attempted when there is left atrial thrombus along the septum, free in the cavity, or on the surface of the valve. Using the conservative approach preferred by most interventionalists, Kang reports successful resolution of left atrial thrombi with warfarin therapy followed by balloon commissurotomy.35 Mitral regurgitation The general presumption that valvuloplasty in patients with moderate or greater mitral regurgitation carried a high risk has not been prospectively tested; however, there have been two retrospective evaluations. A comparison of 25 patients with moderate mitral regurgitation and 25 age and gender matched patients with mild or no regurgitation did indeed demonstrate an increase in severe insufficiency post procedure; however, these patients had much higher echo scores and twice as frequently had severe calcification.36 Further, while 20% of those with initially moderate mitral regurgitation developed severe regurgitation, hemodynamic improvement overall was similar, as was the incidence of post procedure mitral valve replacement. Similarly, patients with mild mitral regurgitation also had less favorable anatomy at baseline and had lower event 798

The predominant approach to percutaneous balloon mitral valvuloplasty is the antegrade transseptal approach. The techniques include single cylindrical balloon, Inoue, double and trefoil balloons, as well as monorail and metal valvulotomes. Inoue and the double cylindrical balloon methods account for virtually all mitral valvuloplasties performed. The procedure has also been performed retrograde.42–44 The advantages include avoidance of transseptal puncture; however large devices are introduced into the femoral artery and balloons are passed across the submitral apparatus without balloon flotation (increasing the risk of catheter entrapment). There are no direct comparison studies between antegrade and retrograde techniques. Inoue technique The Inoue balloon’s principal features are: a modifiable distal tip with reduced profile for transseptal passage, a nylon mesh covering that allows the balloon to straddle the mitral valve, and a compliance curve that allows the balloon to dilate over at least a 4 mm range of sizes (Figure 55.1). A stepwise approach involves evaluating the patient, typically by echocardiography, between each balloon inflation to assess for improvement and detect presence of increasing mitral regurgitation. If improvement is suboptimal and regurgitation has not occurred/increased, the size is typically increased by 1 mm increments. In reviewing 19 series reporting results of Inoue


valvuloplasty, we noted a reported early success rate of 93% in a total of 7091 patients.45,46 Success was variably defined and in some reports overlapped with severe mitral regurgitation, atrial septal defect or embolic events, but included a doubling of the valve area in most studies. Cylindrical balloon techniques The cylindrical balloon technique, introduced in 1985,47 did not uniformly result in adequate gradient reduction and gave way to a double balloon method.48 A stepwise dilation technique is also used with progressively larger balloons placed side by side until adequate gradient reduction is obtained or an increase in mitral regurgitation is noted. The results of 12 studies incorporating 1864 patients reported a 90% overall success rate. Long-term follow up In an extraordinary series of 4832 patients across 120 centers in China, Chen and colleagues have claimed that 98·8% of patients were in NYHA functional class I or II at a mean 32 months follow up, 99·3% success rate, and virtually no complications.49 Restenosis was reported as 5·2% over a mean 32 months follow up. While there were likely problems with data gathering, the evidence from multiple studies of high success and low complication rates in patients with favorable

(A)

anatomy is consistent.20,50 Less favorable long-term results were reported by Cohen et al 51 for 145 patients followed for a mean of 3 years. Their 5 year event free survival was only 51% (freedom from mitral valve replacement, redilation, or death); however, a high percentage of their patients had unfavorable anatomic features. In general, these descriptive series have suffered from incomplete follow up, non-overlapping end points, and lack of serial hemodynamic measurements for assessing hemodynamics and restenosis.

Single v double cylindrical balloons The disadvantages of single balloons are related to the conundrum of a round balloon in an elliptical orifice – resulting in lower gradient reduction. Although no randomized comparisons were made, and much of the data are from sequential individual operator series, or sequential inflations with single followed by double balloons, the latter appears to be superior in retrospective comparisons (Figure 55.2)52–54 as well as in an in vitro study.55 The increased lateral force exerted by two balloons is one presumed mechanism for the superior splitting of the laterally directed commissures. However, a comparison of effective balloon dilating area to body surface area showed that a large single balloon could have similar hemodynamic benefits as two smaller balloons. Thus, geometry is not the sole determinant.

(B)

40

(C)

LA LA

mmHg

LA LV

LV

0

25 mm

LV

25 mm and 15 mm

Figure 55.2 Single v double balloon mitral valvotomy. Note the initial modest reduction of gradient from baseline (A) after single balloon commissurotomy (B), with near complete resolution of gradient after double balloon inflation (C). (Reprinted with permission of the American Heart Association, Inc.115)

799


complication rates reported61,65 may also reflect operator experience with this more complex procedure. Grade B

Inoue v double balloon (Table 55.2) The Inoue technique’s principal advantages are simplicity and short procedure times. The Inoue balloon differs from cylindrical single balloons because of the unique balloon design. The slenderizing feature that facilitates septal passage and the dumb-bell shape of the inflated balloon have been reported by some to result in a lower incidence of atrial septal defect ( 2·5% v up to 10% for the double balloon technique)56 and a much lower likelihood of catastrophic apical perforation. In a prospective randomized comparison between Inoue and double balloon valvotomy, no significant differences were noted in immediate results, including complications.57 A trend toward fewer atrial septal defects with the Inoue balloon was not significant. Because of a lack of other prospective randomized comparisons by physicians equally experienced at both techniques, questions remain unanswered. It is likely that an easier procedure with lower complication rates (the Inoue technique) is a trade off for slightly greater mitral regurgitation,25,58 possibly because the distal portion of the balloon is oversized and may traumatize the subvalvular apparatus. There are also suggestive data that the double balloon technique, by virtue of the lateralization of forces, is advantageous in less favorable anatomy. One example is the result of dilation of asymmetrically fused commissures – where the Inoue technique has been used this led to significant risk of severe mitral regurgitation,59 whereas with double balloon technique use this appeared to be less of a problem.60 The disadvantages of the two balloon technique include longer procedure times, and higher risk of left ventricular apical perforation61–64 although the higher

Other techniques Percutaneous metal mitral commissurotomy is a promising new technique being adopted primarily in a number of developing countries; a series of 153 patients was described by its inventor, Alain Cribier.66 The device, essentially a Tubbs dilator mounted on a cable, is introduced via the right femoral venous approach and can be opened to a maximum of 40 mm. Initial results are encouraging; in particular, what appear to be relatively high postprocedure valve areas (2·2 0·4 cm2) and low rates of mitral regurgitation (severe mitral regurgitation in 1%). Randomized trials comparing this technique to balloon dilatation have not yet been published although several smaller studies have been completed. The metallic head of the device, the most expensive component, is theoretically resterilizable by autoclaving: a potential advantage in parts of the world where mitral stenosis is endemic and the cost of disposables prohibitive. Grade B Additional data on the retrograde non-transseptal technique previously described by Stefanadis and colleagues have been reported67 for the first time from multiple investigational sites. Long-term (up to 9 years) results are relatively comparable to antegrade techniques. However, significant rates of severe mitral regurgitation (3·4%) and of femoral artery injury (1·1%), as well as a relatively modest success rate (88%) in the setting of favorable echocardiography scores (7·7 2·0), suggest that this procedure might best be reserved for patients where transseptal puncture has

Table 55.2 Comparative results of valvuloplasty techniques Inoue

Abdullah117 Arora57 Bassand64 Kasper118 Ortiz119 Park56a Rothlisberger120 Ruiz121 Sharma25 Trevino60 Zhang61 NHLBI122 a

MVA (mean SD)

n

MVA (mean SD)

n

1·9 0·4 2·1 0·4 2·0 0·5 1·7 0·7 1·8 0·4 1·9 0·5 1·6 0·6 1·9 0·3 2·2 0·4 2·0 0·4 1·8 0·3

60 310 71 23 100 59 145 85 120 157 43

2·1 0·5 2·2 0·4 2·0 0·5 2·2 0·8 2·0 0·5 2·0 0·5 1·8 0·7 2·0 0·6 2·1 0·5 2·1 0·5 1·8 0·4 2·0 0·8

60 290 161 22 36 61 90 322 230 56 43 591

Study by Park et al was randomized. Abbreviation: MVA, mitral valve area in cm2

800

Double balloon

Single balloon MVA (mean SD)

n

1·7 0·7

114


unique contraindications. Because of the learning curve associated with this procedure, and the fact that most patients are amenable to the antegrade approach, the longterm role of this technique is uncertain. Similarly, a series of antegrade Inoue balloon valvuloplasties via a jugular venous route had a significant associated complication rate, but represents another alternative approach.68 Finally, Bonhoeffer and colleagues have described a monorail double balloon technique that has potential cost advantages and simplifies the standard double balloon technique; no formal comparison to other techniques has been performed.69

Intraprocedural transesophageal echocardiography Use of transesophageal echocardiography during balloon mitral valvuloplasty has been recommended for early detection of major complications (severe mitral regurgitation, tamponade, and large atrial septal defect).70 In addition, transesophageal echo can confirm needle location during transseptal puncture.71 Finally, decreased procedure time, mitral regurgitation, and residual atrial septal defects have been described in a randomized study of fluoroscopy plus transesophageal echo versus fluoroscopy without echo during balloon commissurotomy.72 The evidence provided by these three studies is not compelling. The latter included a 60% rate of major complications in the non-echo group, suggesting limited experience. Surface two-dimensional echocardiography is sensitive enough to detect increasing mitral regurgitation in most patients, and is an excellent tool for early appreciation of tamponade. Atrial septal defects are becoming substantially less common and are largely limited to 5 mm or smaller and resolve post procedure. Finally, transseptal puncture in experienced hands has limited risk; arguably the procedure should not be performed by those who need transesophageal echo guidance. Intracardiac echo using a transducer placed via the femoral vein may be an alternative but has not yet been tested systematically in this setting.

Complications The learning curve is steep, which has had a major effect both on success and complication rates,73 as well as skewing data in the literature.56 The National Heart Lung Blood Institute (NHLBI) registry reported substantially lower rates of all major complications except acute mitral regurgitation at centers performing more than 25 cases and in the second year that sites enrolled patients; a willingness to attempt balloon commissurotomy in higher-risk subsets in the second year may explain the mitral regurgitation. A recent report compares the first 100 cases of Inoue balloon dilatation versus a subsequent 133 cases, all by the same high volume

operator with extensive prior double balloon experience. The postprocedure valve area, overall success rate and complication rates were significantly improved beyond 100 cases.74 It is likely that the best interests of patients undergoing the procedure would be served by having relatively few centers perform higher volumes. Overall mortality has been approximately 1%, most commonly related to tamponade not only from transseptal catheterization75 but also from fenestration of the left ventricular apex, in particular by the cylindrical balloon technique. The incidence of tamponade has ranged from 2% to 4%, severe mitral regurgitation from 1% to 6%, and cerebral vascular accident and/or thromboembolism in up to 4%. Disturbingly, magnetic resonance imaging detected new hyperintensitivity foci suggestive of cerebral infarcts in 11 of 27 patients.76 All had been evaluated before their procedure by transesophageal echocardiography without detection of clot. Thus, embolization may be common even if not clinically apparent. The probable sources are intracavitary clot, catheter induced thrombus formation and showers of calcium. Atrial septal defects were a significant source of early complications,76 arising from transseptal tearing secondary to inadvertent proximal deployment of cylindrical balloons, withdrawal of winged balloons retrograde, or trauma to the septum from 5 or 8 mm balloons used to dilate the septum. Theoretically these problems should be avoidable by use of a dilator and a shorter balloon system, both features of the Inoue technique, and indeed this has been the finding.77 It should be noted that decompression of the left atrium by a significant sized post procedure atrial septal defect may have influenced the results of some balloon valvuloplasty series and may lead operators to overestimate the mitral valve area post procedure.78

Predictors of outcome Predictors of outcome were addressed in a number of nonrandomized prospective and retrospective analyses. Factors predicting poorer functional class, hemodynamics, overall and event free survival were found to include age, presence of atrial fibrillation, valvular calcification, and postprocedure results, with event free survival at 6–7 years ranging from 15% (unfavorable baseline anatomy) to 83%.79–81 Although these studies were not randomized, they incorporate a broader spectrum of patients with mitral stenosis than the randomized trials, and may represent a more “real world” assessment of results to be expected in the overall population. Additional attention was focused on predictors of adverse outcome, in particular mitral regurgitation. Age and severity of mitral stenosis,82 and degree of anterior leaflet retraction83 correlated with postprocedure insufficiency. The nature of pre- and postprocedure mitral regurgitation was carefully studied in 50 patients.84 As previously noted, 801


severe mitral regurgitation is typically due to leaflet tearing, while most new mitral regurgitation is typically pericommissural in origin. In addition to anatomic predictors, the steep compliance curve of the Inoue balloon was reported as a likely culprit for severe mitral regurgitation.85 Use of balloon sizes in the upper portion of the pressure-volume curve was associated with increased mitral regurgitation; whether this finding, based on retrospective observation, is truly causal is unproven, but has been the subject of numerous anecdotal reports and several abstracts. Previous observations that patients with prior surgical commissurotomy have satisfactory but inferior results were again confirmed.86,87 Perhaps the most comprehensive analysis of outcome was a recently published follow up of up to 15 years in 879 patients. Severe postprocedure mitral regurgitation, echo score 8, age, prior surgical commissurotomy, NYHA functional class IV, moderate preprocedure mitral regurgitation, and elevated pulmonary artery pressures postprocedure were identified as independent predictors of adverse events at long-term follow up.88

Valvuloplasty for mild mitral stenosis Several studies have looked retrospectively at the results of balloon valvuloplasty for patients with valve areas of 1·3–1·5 cm2.89,90 While historical comparisons suggest greater valve area increase than in patients with severe mitral stenosis, there is no evidence that the risk of occasional mortality, need for mitral valve replacement or other major morbidity warrants this approach. The possibility that early commissurotomy may adversely affect the course of the disease, including progression to pulmonary hypertension, atrial fibrillation and stroke remains a hypothesis in need of prospective investigation.91 Grade C Pregnancy There have been multiple reports of successful balloon commissurotomy during pregnancy.92–94 The procedure has been performed with echo guidance and without fluoroscopy95 to avoid radiation exposure to the fetus. Grade B Dilation for restenosis Reoperation for mitral valve stenosis has long been associated with increased morbidity and mortality.96 Several large balloon commissurotomy series have reported inferior overall results compared to de novo dilatation. Davidson reported less symptomatic improvement97 while Jang described a 20% lower success rate (only 51% having valve area 1·5 cm2) and nearly 20% requiring mitral valve replacement by 4 years.98 Cohen described twice the frequency99 and Medina 802

described a 10-fold increase in restenosis rates at 5 years for patients with prior commissurotomy100 (both to approximately 20%). Most significant is the finding by Jang and colleagues that stratification by echo score resulted in nearly superimposable results for de novo and repeat commissurotomy procedures, suggesting that results are defined by valve morphology rather than history of prior commissurotomy.98 Grade B

Bioprosthesis Several case reports have described successful balloon dilatation of bioprosthetic mitral valves, although both the hemodynamic and longer term benefits were obscure in all but one.101–103 However, bioprosthetic valves are typically similar histologically to those seen in calcific aortic stenosis: severe leaflet thickening, immobility and calcification, without commissural fusion.104,105 Grade B Thus, a formal intraoperative study, examining the morphology of severely stenosed bioprosthetic valves before and after balloon dilation, revealed “completely ineffectual” dilation106 with substantial leaflet tearing and cuspal perforation. Although the need for a percutaneous approach to the problem is great, the data do not support bioprosthetic mitral valve dilation.

Balloon v surgical commissurotomy Randomized trials comparing balloon and surgical commissurotomy were begun early in the development phase of the percutaneous technique. Because both use blind dilation of the valve with blunt instruments, and because closed commissurotomy was the predominant procedure in countries where mitral stenosis was prevalent, the early randomized trials compared balloon and closed commissurotomy. In these studies, surgeons were typically more experienced than the operators performing balloon valvuloplasty. In 1988 we randomized 40 patients with relatively ideal anatomy and severe mitral stenosis;107 these patients have been followed with serial catheterization and echocardiography over a 7 year period; there were similar hemodynamic improvements in both groups, sustained through 7 years (Figure 55.3), with one late death in each group and need for repeat commissurotomy in 20%.108 The actual restenosis rate (26% in the balloon group and 35% in the surgical group) as defined by a 50% loss of the gain and a valve area 1·5 cm2 is significantly higher than the repeat commissurotomy rate because restenosis and functional class do not correlate strongly. Thus it is likely that restenosis rates in trials that have not done formal follow up hemodynamics underestimated the true severity of disease during follow up. Two other studies have compared balloon and closed commissurotomy with shorter, non-invasive follow


100

75 Percentile

up only; these have demonstrated balloon results superior to73 or similar to closed commissurotomy.109 However closed commissurotomy in the former study resulted in only a 1·3 cm2 mean valve area, suggesting relatively unaggressive dilation. Finally, a randomized comparison by Ben Farhat and colleagues described superior acute results (2·2 0·4 cm2 v 1·6 0·4 cm2) for balloon valvuloplasty and 4 year restenosis rate of 7% v 37%.110 Thus balloon commissurotomy is at least equal and probably superior to closed surgical commissurotomy. Grade A

50

25

0

Mitral valve area

1 cm2 2·0 1·8 1·6 1·4 1·2 1·0 0·8 0·6 0·4 0·2 0·0

Balloon Surgery

* *

*

*

Balloon *

Base

1 wk

8 mth

*

3·5 yr

Figure 55.3 Mitral valves areas at baseline and each follow up interval over 313 years in patients randomized to percutaneous balloon or surgical closed mitral commissurotomy.108 Asterisk denotes P 0·001 compared with baseline.

Open commissurotomy v balloon The hypothesis that open commissurotomy would be superior to balloon valvuloplasty was based on the potential benefits of direct vision, including surgical splitting and remodeling of the subvalvular apparatus, neither of which are features of closed or balloon commissurotomy. A prospective series of 100 open commissurotomy patients gathered data specifically for historical comparison to the then reported results of balloon valvuloplasty and concluded that open commissurotomy was distinctly superior.111 The results of surgery, mean valve area 2·9 cm2, exceeds expectations and may be related to technique of measurement112 or patient selection, while mitral regurgitation was absent in all but eight cases (where it was reported to be mild), results also testimony to great operator skill but in excess of prior reports.8 Grade A On the contrary, the more compelling evidence from prospective randomized studies is for similar or superior results with balloon commissurotomy. In 1989 we randomized 60 patients to a prospective comparison of balloon versus open commissurotomy.113 Patients had near identical baseline hemodynamics but those undergoing balloon commissurotomy had superior mitral valve areas at 3 years (Figure 55.4). A possible explanation for superior results in balloon commissurotomy patients is the direct and

3 2 Mitral valve area Baseline 1 wk 6 mth 3 yr

4

Surgery

Figure 55.4 Mitral valve areas at baseline and at each follow up interval in patients randomized to balloon or open surgical commissurotomy. The values represent the percentile of patients whose valve areas are the valve areas on the abscissa. The baseline values overlap, but a shift to the right (representing higher valve areas) is seen for the balloon group at each time point. (Reprinted by permission of the New England Journal of Medicine. Copyright © 1994, Massachusetts Medical Society.113)

continuous feedback to the operator of hemodynamics during catheterization laboratory procedures, which even with the advent of transesophageal monitoring in the operating room is not available to the same degree to the surgeon. In the trial referred to earlier, Ben Farhat and colleagues report a three-way randomized comparison of balloon, closed and open surgical commissurotomy in 90 patients.110 Most of the objective information is through 6 month follow up, although clinical status/events and valve areas are described through 7 years. Their results, which include an absence of mortality, NYHA class I function in 90% of the balloon and open mitral commissurotomy (OMC) patients, and residual valve area of 1·8 cm2 in these two groups at 7 years with only 7% restenosis, are exceptionally optimistic. The results of closed commissurotomy were distinctly inferior. Because functional class correlates poorly with hemodynamics in mitral stenosis and because planimetry, the technique used here for mitral valve area assessment beyond 6 months, is subjective when the commissures are open (and was not performed by blinded investigators), the findings of this study need to be confirmed. Less optimistic data, utilizing hemodynamics and blinded interpretation, suggest that restenosis rates may be 25% by 7 years even in patients with relatively ideal valve anatomy preprocedure.114 Nevertheless, this paper confirms that balloon valvuloplasty is at least as effective as open commissurotomy for patients with severe mitral stenosis and ideal valve anatomy. 803


The study’s optimistic findings may perhaps in part be due to a distinguishing feature of all of the randomized comparisons of balloon versus surgical commissurotomy: single site studies that depend to a significant degree on individual physician practices and small patient populations. Grade A

Cost Although formal cost comparison studies have not been reported, charges and costs at hospitals in India and in the United States have been estimated. Lau and Ruiz described cost to a United States hospital of $3000 for balloon valvuloplasty and $6000 for closed commissurotomy (assuming a hospital could be found that still performs this procedure). We published 1991 charges for balloon and closed commissurotomy in the United States and India (Figure 55.5) and demonstrated a sixfold greater expense for balloon valvuloplasty in India. However, our calculations did not include the extensive reuse of disposables in developing countries, where balloons can account for a much higher portion of the charges than physicians’ fees or operating room billings. Percutaneous metallic commissurotomy, as referred to earlier, may also have a significant impact on cost considerations. The results of the randomized trials offer compelling evidence that balloon valvuloplasty is an effective alternative to surgery for patients with good valve anatomy. Even with a number of anatomic features predicting less favorable United States

outcome, balloon commissurotomy, at the cost of higher risk in patients with unfavorable anatomy, still has the potential for palliation. The safety and efficacy of Inoue and double balloon valvuloplasty are not compellingly different in experienced hands and the selection of techniques should be based on operator preference, experience, and equipment availability. Low cost, avoidance of thoracotomy scar and discomfort, shorter hospitalization and excellent follow up results to date mandate consideration of balloon valvuloplasty in most patients with rheumatic mitral valve stenosis without significant contraindications. Since balloon as well as surgical commissurotomy are largely palliative procedures, percutaneous balloon valvuloplasty has the added benefit of delaying the time until eventual thoracotomy. Grade A In summary, percutaneous balloon mitral valvuloplasty is a superior alternative to surgical commissurotomy for a significant subset of patients with rheumatic mitral stenosis. Careful case selection and performance of the procedure by experienced teams will have a significant impact on outcome. Both clinical and financial considerations suggest that balloon valvuloplasty is the procedure of choice for rheumatic mitral stenosis in patients with suitable anatomy. Grade A Key points ●

●

Disposables

14

Anesthesia

Charges ($’000)

12 Physician’s fee

10 8

●

Cath lab/ operating room

6 4

India ICU Room charges

2 0 PBMV

CMC

PBMV

CMC

Figure 55.5 Charges for percutaneous balloon mitral valvuloplasty (PBMV) and closed surgical commissurotomy (CMC) at the Nizam’s Institute of Medical Sciences in Hyderabad, India and at Harper Hospital in Detroit, MI in 1991. With the extensive reuse of disposables in developing countries, the cost of balloon valvuloplasty more closely approximates that for closed commissurotomy. (© 1993, F.A. Davis Co. Reprinted with permission.116)

804

Ideal patients have severe mitral stenosis without: mild mitral regurgitation, severe subvalvular disease, or severe calcification eccentric commissural fusion, clot in left atrium, volume loaded left ventricle Procedure may be of benefit in: critical mitral stenosis, but evidence for favorable long-term risk–benefit ratio is lacking patients with unfavorable anatomy, including moderate mitral regurgitation, but with less favorable results and higher morbidity/mortality patients with mitral restenosis, dependent on anatomic features pregnant patients Balloon valvuloplasty is superior to closed commissurotomy and is equivalent or superior to open commissurotomy in ideal patients

References 1.Cutler EC, Levine SA. Cardiotomy and valvulotomy for mitral stenosis. Experimental observations and clinical notes concerning an operated case with recovery. Boston Med Surg J 1923;188:1023–7. 2.Suttar HS. The surgical treatment of mitral stenosis. BMJ 1925;2:603–6. 3.Lewis T. Diseases of the heart. 3rd edn. London: Macmillan, 1943. 4.John S, Bashi VV, Jairaj PS et al. Closed mitral valvotomy: early results and long-term follow-up of 3724 consecutive patients. Circulation 1983;68:891–6.


5.Toumbouras M, Panagopoulos F, Papakonstantinou C et al. Long-term surgical outcome of closed mitral commissurotomy. J Heart Valve Dis 1995;4:247–50. 6.Rihal CS, Schaff HV, Frye RL, Bailey KR, Hammes LN, Holmes DR Jr. Long-term follow-up of patients undergoing closed transventricular mitral commissurotomy: a useful surrogate for percutaneous balloon mitral valvuloplasty? J Am Coll Cardiol 1992;20:781–6. 7.Scalia D, Rizzoli G, Campanile F et al. Long-term results of mitral commissurotomy. J Thorac Cardiovasc Surg 1993; 105:633–42. 8.Villanova C, Melacini P, Scognamiglio R et al. Long-term echocardiographic evaluation of closed and open mitral valvulotomy. Int J Cardiol 1993;38:315–21. 9.Hickey MS, Blackstone EH, Kirklin JW, Dean LS. Outcome probabilities and life history after surgical mitral commissurotomy: implications for balloon commissurotomy. J Am Coll Cardiol 1991;17:29–42. 10.Inoue K, Kitamura F, Chikusa H, Miyamoto N. Atrial septostomy by a new balloon catheter. Jpn Circ J 1981;45:730–8. 11.Inoue K, Nakamura T, Kitamura F. Nonoperative mitral commissurotomy by a new balloon catheter. [Abstract] Jpn Circ J 1982;46:877. 12.Inoue K, Owaki T, Nakamura T, Kitamura F, Miyamoto N. Clinical application of transvenous mitral commissurotomy by a new balloon catheter. J Thorac Cardiovasc Surg 1984; 87:394–402. 13.Block PC, Palacios IF, Jacobs ML, Fallon JT. Mechanism of percutaneous mitral valvotomy. Am J Cardiol 1987;59: 178–9. 14.Matsuura Y, Fukunaga S, Ishihara H et al. Mechanics of percutaneous balloon valvotomy for mitral valvular stenosis. Heart Vessels 1988;4:179–83. 15.Nabel E, Bergin PJ, Kirsh MM. Morphological analysis of balloon mitral valvuloplasty; intra-operative results. [Abstract] J Am Coll Cardiol 1990;15:97A. 16.Kaplan JD, Isner JM, Karas RH et al. In vitro analysis of mechanisms of balloon valvuloplasty of stenotic mitral valves. Am J Cardiol 1987;59:318–23. 17.McKay RG, Lock JE, Safian RD et al. Balloon dilation of mitral stenosis in adult patients: postmortem and percutaneous mitral valvuloplasty studies. J Am Coll Cardiol 1987;9: 723–31. 18.Wilkins GT, Weyman AE, Abascal VM, Block PC, Palacios IF. Percutaneous balloon dilatation of the mitral valve: an analysis of echocardiographic variables related to outcome and the mechanism of dilatation. Br Heart J 1988;60:299–308. 19.Desideri A, Vanderperren O, Serra A et al. Long-term (9 to 33 months) echocardiographic follow-up after successful percutaneous mitral commissurotomy. Am J Cardiol 1992;69:1602–6. 20.Palacios IF, Tuzcu ME, Weyman AE, Newell JB, Block PC. Clinical follow-up of patients undergoing percutaneous mitral balloon valvotomy. Circulation 1995;91:671–6. 21.Abascal VM, Wilkins GT, O’Shea JP et al. Prediction of successful outcome in 130 patients undergoing percutaneous balloon mitral valvotomy. Circulation 1990;82:448–56. 22.Fatkin D, Roy P, Morgan JJ, Feneley MP. Percutaneous balloon mitral valvotomy with the Inoue single-balloon catheter: commissural morphology as a determinant of outcome. J Am Coll Cardiol 1993;21:390–7.

23.Levin TN, Feldman T, Bednarz J, Carroll JD, Lang RM. Transesophageal echocardiographic evaluation of mitral valve morphology to predict outcome after balloon mitral valvotomy. Am J Cardiol 1994;73:707–10. 24.Herrmann HC, Ramaswamy K, Isner JM et al. Factors influencing immediate results, complications, and short-term follow-up status after Inoue balloon mitral valvotomy: a North American multicenter study. Am Heart J 1992;124:160–6. 25.Sharma S, Loya YS, Desai DM, Pinto RJ. Percutaneous mitral valvotomy using Inoue and double balloon technique: comparison of clinical and hemodynamic short term results in 350 cases. Cathet Cardiovasc Diagn 1993;29:18–23. 26.Reid CL, Chandraratna PA, Kawanishi DT, Kotlewski A, Rahimtoola SH. Influence of mitral valve morphology on double-balloon catheter balloon valvuloplasty in patients with mitral stenosis. Analysis of factors predicting immediate and 3-month results. Circulation 1989;80:515–24. 27.Zhang HP, Allen JW, Lau FY, Ruiz CE. Immediate and late outcome of percutaneous balloon mitral valvotomy in patients with significantly calcified valves. Am Heart J 1995;129: 501–6. 28.Miche E, Fassbender D, Minami K et al. Pathomorphological characteristics of resected mitral valves after unsuccessful valvuloplasty. J Cardiovasc Surg 1996;37:475–81. 29.Ruiz CE, Zhang HP, Gamra H, Allen JW, Lau FY. Late clinical and echocardiographic follow up after percutaneous balloon dilatation of the mitral valve. Br Heart J 1994;71:454–8. 30.Orrange SE, Kawanishi DT, Lopez BM, Curry SM, Rahimtoola SH. Actuarial outcome after catheter balloon commissurotomy in patients with mitral stenosis. Circulation 1997;95:382–9. 31.Cannan CR, Nishimura RA, Reeder GS et al. Echocardiographic assessment of commissural calcium: a simple predictor of outcome after percutaneous mitral balloon valvotomy. J Am Coll Cardiol 1997;29:175–80. 32.Kronzon I, Tunick PA, Glassman E, Slater J, Schwinger M, Freedberg RS. Transesophageal echocardiography to detect atrial clots in candidates for percutaneous transseptal mitral balloon valvuloplasty. J Am Coll Cardiol 1990;16:1320–2. 33.Rittoo D, Sutherland GR, Currie P, Starkey IR, Shaw TR. A prospective study of left atrial spontaneous echo contrast and thrombus in 100 consecutive patients referred for balloon dilation of the mitral valve. J Am Soc Echocardiogr 1994;7:516–27. 34.Hung JS. Cheng TO, eds. Percutaneous balloon valvuloplasty. Mitral stenosis with left atrial thrombi: Inoue balloon catheter technique. New York: Igaku-Shoin, 1992. 35.Kang DH, Song JK, Chae K et al. Comparison of outcomes of percutaneous mitral valvuloplasty versus mitral valve replacement after resolution of left atrial appendage thrombi. Am J Cardiol 1998;81:97–100. 36.Zhang HP, Gamra H, Allen JW, Lau FY, Ruiz CE. Balloon valvotomy for mitral stenosis associated with moderate mitral regurgitation. Am J Cardiol 1995;75:960–3. 37.Alfonso F, Macaya C et al. Early and late results of percutaneous mitral valvuloplasty for mitral stenosis associated with mild mitral regurgitation. Am J Cardiol 1993;71:1304–10. 38.Tuzcu EM, Block PC, Griffin B, Dinsmore R, Newell JB, Palacios IF. Percutaneous mitral balloon valvotomy in patients with calcific mitral stenosis: immediate and long-term outcome. J Am Coll Cardiol 1994;23:1604–9.

805


39.Feldman T, Carroll JD, Isner JM et al. Effect of valve deformity on results and mitral regurgitation after Inoue balloon commissurotomy. Circulation 1992;85:180–7. 40.Herrmann HC, Lima JA, Feldman T et al. Mechanisms and outcome of severe mitral regurgitation after Inoue balloon valvuloplasty. North American Inoue Balloon Investigators. J Am Coll Cardiol 1993;22:783–9. 41.Tuzcu EM, Block PC, Griffin BP, Newell JB, Palacios IF. Immediate and long-term outcome of percutaneous mitral valvotomy in patients 65 years and older. Circulation 1992; 85:963–71. 42.Bahl VK, Juneja R, Thatai D, Kaul U, Sharma S, Wasir HS. Retrograde nontransseptal balloon mitral valvuloplasty for rheumatic mitral stenosis. Cathet Cardiovasc Diagn 1994; 33:331–4. 43.Stefanadis C, Stratos C, Kallikazaros I et al. Retrograde nontransseptal balloon mitral valvuloplasty using a modified Inoue balloon catheter. Cathet Cardiovasc Diagn 1994;33: 224–33. 44.Stefanadis C, Stratos C, Pitsavos C et al. Retrograde nontransseptal balloon mitral valvuloplasty. Immediate results and long-term follow-up. Circulation 1992;85:1760–7. 45.Lau KW, Hung JS, Ding ZP, Johan A. Controversies in balloon mitral valvuloplasty: the when (timing for intervention), what (choice of valve), and how (selection of technique). Cathet Cardiovasc Diagn 1995;35:91–100. 46.Glazier JJ, Turi ZG. Percutaneous balloon mitral valvuloplasty. Prog Cardiovasc Dis 1997;40:5–26. 47.Lock JE, Khalilullah M, Shrivastava S, Bahl V, Keane JF. Percutaneous catheter commissurotomy in rheumatic mitral stenosis. N Engl J Med 1985;313:1515–18. 48.al Zaibag M, Ribeiro PA, Al Kasab S, al Fagih MR. Percutaneous double-balloon mitral valvotomy for rheumatic mitral-valve stenosis. Lancet 1986;1:757–61. 49.Chen CR, Cheng TO. Percutaneous balloon mitral valvuloplasty by the Inoue technique: a multicenter study of 4832 patients in China. Am Heart J 1995;129:1197–203. 50.Iung B, Cormier B, Ducimetiere P et al. Functional results 5 years after successful percutaneous mitral commissurotomy in a series of 528 patients and analysis of predictive factors. J Am Coll Cardiol 1996;27:407–14. 51.Cohen DJ, Kuntz RE, Gordon SP et al. Predictors of long-term outcome after percutaneous balloon mitral valvuloplasty. N Engl J Med 1992;327:1329–33. 52.Shrivastava S, Mathur A, Dev V, Saxena A, Venugopal P, SampathKumar A. Comparison of immediate hemodynamic response to closed mitral commissurotomy, single-balloon, and double-balloon mitral valvuloplasty in rheumatic mitral stenosis. J Thorac Cardiovasc Surg 1992;104:1264–7. 53.Al Kasab S, Ribeiro PA, Sawyer W. Comparison of results of percutaneous balloon mitral valvotomy using consecutive single (25 mm) and double (25 mm and 12 mm) balloon techniques. Am J Cardiol 1989;64:1385–7. 54.Chen CG, Wang YP, Qing D, Lin YS, Lan YF. Percutaneous mitral balloon dilatation by a new sequential single- and double-balloon technique. Am Heart J 1988;116:1161–7. 55.Ribeiro PA, al Zaibag M, Rajendran V et al. Mechanism of mitral valve area increase by in vitro single and double balloon mitral valvotomy. Am J Cardiol 1988;62:264–9.

806

56.Complications and mortality of percutaneous balloon mitral commissurotomy. A report from the National Heart, Lung, and Blood Institute Balloon Valvuloplasty Registry. Circulation 1992;85:2014–24. 57.Park SJ, Kim JJ, Park SW, Song JK, Doo YC, Lee SJ. Immediate and one-year results of percutaneous mitral balloon valvuloplasty using Inoue and double-balloon techniques. Am J Cardiol 1993;71:938–43. 58.Arora R, Kalra GS, Murty GS et al. Percutaneous transatrial mitral commissurotomy: immediate and intermediate results. J Am Coll Cardiol 1994;23:1327–32. 59.Miche E, Bogunovic N, Fassbender D et al. Predictors of unsuccessful outcome after percutaneous mitral valvulotomy including a new echocardiographic scoring system. J Heart Valve Dis 1996;5:430–5. 60.Rodriguez L, Monterroso VH, Abascal VM et al. Does asymmetrical mitral valve disease predict an adverse outcome after percutaneous balloon mitral valvotomy? An echocardiographic study. Am Heart J 1992;123:1678–82. 61.Trevino AJ, Ibarra M, Garcia A et al. Immediate and long-term results of balloon mitral commissurotomy for rheumatic mitral stenosis: comparison between Inoue and doubleballoon techniques. Am Heart J 1996;131:530–6. 62.Zhang HP, Gamra H, Allen JW, Lau FY, Ruiz CE. Comparison of late outcome between Inoue balloon and double-balloon techniques for percutaneous mitral valvotomy in a matched study. Am Heart J 1995;130:340–4. 63.Fu XY, Zhang DD, Schiele F, Anguenot T, Bernard Y, Bassand JP. Complications of percutaneous mitral valvuloplasty; comparison of the double balloon and the Inoue techniques. Arch Mal Coeur Vaiss 1994;87:1403–11. 64.Rihal CS, Nishimura RA, Reeder GS, Holmes DR Jr. Percutaneous balloon mitral valvuloplasty: comparison of double and single (Inoue) balloon techniques. Cathet Cardiovasc Diagn 1993;29:183–90. 65.Bassand JP, Schiele F, Bernard Y et al. The double-balloon and Inoue techniques in percutaneous mitral valvuloplasty: comparative results in a series of 232 cases. J Am Coll Cardiol 1991;18:982–9. 66.Cribier A, Eltchaninoff H, Koning R et al. Percutaneous mechanical mitral commissurotomy with a newly designed metallic valvulotome. Circulation 1999;99:793–9. 67.Stefanadis CI, Stratos CG, Lambrou SG et al. Retrograde nontransseptal balloon mitral valvuloplasty: immediate results and intermediate long-term outcome in 441 cases – a multicenter experience. J Am Coll Cardiol 1998;32:1009–16. 68.Joseph G, Baruah DK, Kuruttukulam SV, Chandy ST, Krishnaswami S. Transjugular approach to transseptal balloon mitral valvuloplasty. Cathet Cardiovasc Diagn 1997;42: 219–26. 69.Bonhoeffer P, Esteves C, Casal U et al. Percutaneous mitral valve dilatation with the Multi-Track System. Catheter Cardiovasc Interv 1999;48:178–83. 70.Goldstein SA, Campbell A, Mintz GS, Pichard A, Leon M, Lindsay J, Jr. Feasibility of on-line transesophageal echocardiography during balloon mitral valvulotomy: experience with 93 patients. J Heart Valve Dis 1994;3:136–48. 71.Ballal RS, Mahan EF, Nanda NC, Dean LS. Utility of transesophgeal echocardiography in interatrial septal puncture


during percutaneous mitral balloon commissurotomy. Am J Cardiol 1990;66:230–2. 72.Ramondo A, Chirillo F, Dan M et al. Value and limitations of transesophageal echocardiographic monitoring during percutaneous balloon mitral valvotomy. Int J Cardiol 1991; 31:223–33. 73.Rihal CS, Nishimura RA, Holmes DR, Jr. Percutaneous balloon mitral valvuloplasty: the learning curve. Am Heart J 1991;122:1750–6. 74.Sanchez PL, Harrell LC, Salas RE, Palacios IF. Learning curve of the Inoue technique of percutaneous mitral balloon valvuloplasty. Am J Cardiol 2001;88:662–7. 75.Schoonmaker FW, Vijay NK, Jantz RD. Left atrial and ventricular transseptal catheterization review: losing skills. Cathet Cardiovasc Diagn 1987;13:233–8. 76.Rocha P, Mulot R, Lacombe P, Pilliere R, Belarbi A, Raffestin B. Brain magnetic resonance imaging before and after percutaneous mitral balloon commissurotomy. Am J Cardiol 1994; 74:955–7. 77.Yoshida K, Yoshikawa J, Akasaka T et al. Assessment of left-toright atrial shunting after percutaneous mitral valvuloplasty by transesophageal color Doppler flow-mapping. Circulation 1989;80:1521–6. 78.Thomas MR, Monaghan MJ, Metcalfe JM, Jewitt DE. Residual atrial septal defects following balloon mitral valvuloplasty using different techniques. A transthoracic and transoesophageal echocardiography study demonstrating an advantage of the Inoue balloon. Eur Heart J 1992;13: 496–502. 79.Petrossian GA, Tuzcu EM, Ziskind AA, Block PC, Palacios I. Atrial septal occlusion improves the accuracy of mitral valve area determination following percutaneous mitral balloon valvotomy. Cathet Cardiovasc Diagn 1991;22:21–4. 80.Lau KW, Ding ZP, Quek S, Kwok V, Hung JS. Long-term (36–63 month) clinical and echocardiographic follow-up after Inoue balloon mitral commissurotomy. Cathet Cardiovasc Diagn 1998;43:33–8. 81.Meneveau N, Schiele F, Seronde MF et al. Predictors of eventfree survival after percutaneous mitral commissurotomy. Heart 1998;80:359–64. 82.Zhang HP, Yen GS, Allen JW, Lau FY, Ruiz CE. Comparison of late results of balloon valvotomy in mitral stenosis with versus without mitral regurgitation. Am J Cardiol 1998;81:51–5. 83.Matsubara T, Yamazoe M, Tamura Y et al. Progression to moderate or severe mitral regurgitation after percutaneous transvenous mitral commissurotomy using stepwise inflation technique. J Cardiol 1998;31:289–95. 84.Mueller UK, Sareli P, Essop MR. Anterior mitral leaflet retraction – a new echocardiographic predictor of severe mitral regurgitation following balloon valvuloplasty by the Inoue technique. Am J Cardiol 1998;81:656–9. 85.Rittoo D, Sutherland GR, Shaw TR. A prospective echocardiographic study of the effects of balloon mitral commissurotomy on pre-existing mitral regurgitation in patients with mitral stenosis. Cardiology 1998;89:202–9. 86.Goel PK, Garg N, Sinha N. Pressure zone used and the occurrence of mitral regurgitation in Inoue balloon mitral commissurotomy. Cathet Cardiovasc Diagn 1998;43:141–6. 87.Ito T, Suwa M, Hirota Y et al. Comparison of immediate and long-term outcome of percutaneous transvenous mitral

commissurotomy in patients who have and have not undergone previous surgical commissurotomy. Jpn Circ J 1997;61: 218–22. 88.Palacios IF, Sanchez PL, Harrell LC, Weyman AE, Block PC. Which patients benefit from percutaneous mitral balloon valvuloplasty? Prevalvuloplasty and postvalvuloplasty variables that predict long-term outcome. Circulation 2002;105: 1465–71. 89.Pan M, Medina A, Suarez De Lezo J et al. Balloon valvuloplasty for mild mitral stenosis. Cathet Cardiovasc Diagn 1991; 24:1–5. 90.Herrmann HC, Feldman T, Isner JM et al. Comparison of results of percutaneous balloon valvuloplasty in patients with mild and moderate mitral stenosis to those with severe mitral stenosis. The North American Inoue Balloon Investigators. Am J Cardiol 1993;71:1300–3. 91.Turi ZG. Mitral balloon valvuloplasty [letter; comment]. Cathet Cardiovasc Diagn 1992;25:343–4. 92.Glantz JC, Pomerantz RM, Cunningham MJ, Woods JR Jr. Percutaneous balloon valvuloplasty for severe mitral stenosis during pregnancy: a review of therapeutic options. Obstet Gynecol Surg 1993;48:503–8. 93.Patel JJ, Mitha AS, Hassen F et al. Percutaneous balloon mitral valvotomy in pregnant patients with tight pliable mitral stenosis. Am Heart J 1993;125:1106–9. 94.Ribeiro PA, Fawzy ME, Awad M, Dunn B, Duran CG. Balloon valvotomy for pregnant patients with severe pliable mitral stenosis using the Inoue technique with total abdominal and pelvic shielding. Am Heart J 1992;124: 1558–62. 95.Kultursay H, Turkoglu C, Akin M, Payzin S, Soydas C, Akilli A. Mitral balloon valvuloplasty with transesophageal echocardiography without using fluoroscopy. Cathet Cardiovasc Diagn 1992;27:317–21. 96.Harken DE, Black H, Taylor WJ, Thrower WB, Ellis LB. Reoperation for mitral stenosis. A discussion of postoperative deterioration and methods of improving initial and secondary operation. Circulation 1961;23:7–12. 97.Davidson CJ, Bashore TM, Mickel M, Davis K. Balloon mitral commissurotomy after previous surgical commissurotomy. The National Heart, Lung, and Blood Institute Balloon Valvuloplasty Registry participants. Circulation 1992;86: 91–9. 98.Jang IK, Block PC, Newell JB, Tuzcu EM, Palacios IF. Percutaneous mitral balloon valvotomy for recurrent mitral stenosis after surgical commissurotomy. Am J Cardiol 1995; 75:601–5. 99.Cohen JM, Glower DD, Harrison JK et al. Comparison of balloon valvuloplasty with operative treatment for mitral stenosis. Ann Thorac Surg 1993;56:1254–62. 100.Medina A, de Lezo JS, Hernandez E et al. eds. Percutaneous balloon valvuloplasty. Mitral restenosis: the Cordoba-Las Palmas experience. New York: Igaku-Shoin, 1992. 101.Calvo OL, Sobrino N, Gamallo C, Oliver J, Dominguez F, Iglesias A. Balloon percutaneous valvuloplasty for stenotic bioprosthetic valves in the mitral position. Am J Cardiol 1987;60:736–7. 102.Cox DA, Friedman PL, Selwyn AP, Lee RT, Bittl JA. Improved quality of life after successful balloon valvuloplasty of a stenosed mitral bioprosthesis. Am Heart J 1989;118:839–41.

807


103.Babic UU, Grujicic S, Vucinic M. Balloon valvoplasty of mitral bioprosthesis. Int J Cardiol 1991;30:230–2. 104.Waller BF, McKay C, VanTassel J, Allen M. Catheter balloon valvuloplasty of stenotic porcine bioprosthetic valves: Part II: Mechanisms, complications, and recommendations for clinical use. Clin Cardiol 1991;14:764–72. 105.Waller BF, McKay C, Van Tassel J, Allen M. Catheter balloon valvuloplasty of stenotic porcine bioprosthetic valves: Part I: Anatomic considerations. Clin Cardiol 1991;14:686–91. 106.Lin PJ, Chang JP, Chu JJ, Chang CH, Hung JS. Balloon valvuloplasty is contraindicated in stenotic mitral bioprostheses. Am Heart J 1994;127:724–6. 107.Turi ZG, Reyes VP, Raju BS et al. Percutaneous balloon versus surgical closed commissurotomy for mitral stenosis. A prospective, randomized trial. Circulation 1991;83:1179–85. 108.Raju BS, Turi ZG, Raju R et al. Three and one-half year followup of a randomized trial comparing percutaneous balloon and surgical closed mitral commissurotomy. [Abstract] J Am Coll Cardiol 1993;21:429A. 109.Arora R, Nair M, Kalra GS, Nigam M, Khalilullah M. Immediate and long-term results of balloon and surgical closed mitral valvotomy: a randomized comparative study. Am Heart J 1993;125:1091–4. 110.Ben Farhat M, Ayari M, Maatouk F et al. Percutaneous balloon versus surgical closed and open mitral commissurotomy: seven-year follow-up results of a randomized trial. Circulation 1998;97:245–50. 111.Antunes MJ, Nascimento J, Andrade CM, Fernandes LE. Open mitral commissurotomy: a better procedure? J Heart Valve Dis 1994;3:88–92. 112.Acar J. Open mitral commissurotomy or percutaneous mitral commissurotomy? [editorial]. J Heart Valve Dis 1994;3:133–5. 113.Reyes VP, Raju BS, Wynne J et al. Percutaneous balloon valvuloplasty compared with open surgical commissurotomy for mitral stenosis. N Engl J Med 1994;331:961–7.

808

114.Turi ZG. Percutaneous balloon valvuloplasty versus surgery: randomized comparisons. J Interv Cardiol 2000;13: 395–401. 115.Palacios I, Block PC, Brandi S et al. Percutaneous balloon valvotomy for patients with severe mitral stenosis. Circulation 1987;75:778–84. 116.Turi ZG. Frankel W, Brest A, eds. Valvular heart disease: comprehensive evaluation and treatment. 2nd edn. Valvuloplasty. Philadelphia: F.A. Davis, 1993. 117.Abdullah M, Halim M, Rajendran V, Sawyer W, al Zaibag M. Comparison between single (Inoue) and double balloon mitral valvuloplasty: immediate and short-term results. Am Heart J 1992;123:1581–8. 118.Kasper W, Wollschlager H, Geibel A, Meinertz T, Just H. Percutaneous mitral balloon valvuloplasty – a comparative evaluation of two transatrial techniques. Am Heart J 1992;124: 1562–6. 119.Ortiz FA, Macaya C, Alfonso F. Mono- versus double-balloon technique for commissural splitting after percutaneous mitral valvotomy. Am J Cardiol 1992;69:1100–1. 120.Rothlisberger C, Essop MR, Skudicky D, Skoularigis J, Wisenbaugh T, Sareli P. Results of percutaneous balloon mitral valvotomy in young adults. Am J Cardiol 1993;72:73–7. 121.Ruiz CE, Zhang HP, Macaya C, Aleman EH, Allen JW, Lau FYK. Comparison of Inoue single-balloon versus doubleballoon technique for percutaneous mitral valvotomy. Am Heart J 1992;123:942–7. 122.Multicenter experience with balloon mitral commissurotomy. NHLBI Balloon Valvuloplasty Registry Report on immediate and 30-day follow- up results. The National Heart, Lung, and Blood Institute Balloon Valvuloplasty Registry Participants. Circulation 1992;85:448–61.

56

Valve repair and choice of valves Paul J Pearson, Hartzell V Schaff

operation for valve repair should be considered for all patients with severe mitral regurgitation.4,5

Introduction Changes in treatment of mitral regurgitation provide a classic example of how advances in surgical technique influence the overall strategy of medical management of valvular heart disease, including both the indications for and timing of operation. In North America, degenerative diseases such as floppy valves and ruptured chordae tendineae are the most common causes of non-ischemic mitral valve regurgitation.1–3 Previously, clinicians observed patients with mitral regurgitation until symptoms developed or until there was evidence of left ventricular failure. Usually, operation resulted in replacement of the valve with a prosthesis. This left the patient with ventricular dysfunction, irreversible in some cases, and also the attendant prosthesis-related risks such as thromboembolism, hemorrhage caused by systemic anticoagulation, infection, and risks of mechanical failure. The advent of mitral valve repair, with its predictability and safety, lead to new criteria for intervention. Indeed, early

Timing of operation for mitral valve regurgitation Grade B Mitral valve regurgitation often progresses slowly and because of favorable loading conditions, left ventricular dysfunction can develop even though systolic indices of left ventricular performance are maintained. Indeed, with severe mitral valve regurgitation, normal ventricular function should result in a hyperdynamic left ventricle with a supranormal ejection fraction. When the ejection fraction falls below 60% in the presence of severe mitral regurgitation, the prognosis of patients after surgical correction worsens (Figure 56.1).4 However, the relative insensitivity of ejection fraction in gauging ventricular performance in patients with mitral regurgitation has led to the development of indices of

Valve repair

100

Valve replacement

Survival (%)

80

60

40 EF ≥ 60% EF < 60%

20

P = 0·005

P = 0·01

0 0 At risk (n) EF ≥ 60% 138

1

2

3

4

5

6

7

8

134

124

90

67

56

41

31

20

49

46

42

32

19

13

7

6

2

EF < 60%

9

0

1

2

3

4

5

6

7

8

9

8 111

103

95

89

81

69

59

48

34

26

72

69

59

46

39

26

22

16

11

2

77

Years

Figure 56.1 Graphs of late survival according to preoperative ejection fraction (EF) after valve repair (left) and valve replacement (right). (From Enriques-Sarano et al.4)

809


left ventricular function that are less dependent on preload, such as end systolic dimension. Again, prognosis after valve repair or replacement is poor when preoperative left ventricular end systolic dimension exceeds 45 mm.4 Thus, even in an asymptomatic patient with an ejection fraction greater than 60%, if left ventricular end systolic diameter approaches 45 mm, valve repair should be seriously considered.4

Table 56.1 Operative mortality for mitral valve replacement v repair

Overall Age 75 years Age 75 years

Valve repair v replacement Grade A There are no prospective, randomized studies comparing outcomes after mitral valve repair with replacement for mitral regurgitation. In addition, it is often difficult to compare these two modes of surgical treatment by review of the literature because of heterogeneous patient populations.6 Patients with anatomy favorable for valve repair may have less advanced disease when compared to those patients in whom valve replacement is necessary.7 However, even with these confounding factors, some generalizations can be made. First, analysis based upon adjustment for baseline differences in patient populations indicates that patients undergoing mitral valve repair have improved survival and better postoperative left ventricular function than patients undergoing mitral valve replacement (Figure 56.2).7 In addition, patients undergoing valve repair have a lower operative mortality than their counterparts having prosthetic replacement (Table 56.1).6 These good

Replacement

Repair

P

n 214 (10·3%) n 39 (30·8%) n 175 (5·7%)

n 195 (2·6%) n 44 (6·8%) n 151 (1·3%)

0·002 0·0005 0·036

From Enriquez-Sarano et al 7

results following valvuloplasty are, at least in part, due to maintenance of normal left ventricular geometry and function through preservation of the valve-chordal-papillary muscle complex.8–11 Importantly, valve repair and replacement have similar low rates of reoperation. A study from our clinic comparing the outcomes of 195 patients undergoing valve repair with 214 who underwent valve replacement for organic mitral regurgitation demonstrated that freedom from reoperation was 90% and 93% (repair and replacement) at 5 years and 75% and 80% at 10 years respectively (P 0·47)7 (Figure 56.3). Valve repair can even be undertaken in some patients with calcification of the leaflets and annulus. Although this

Valve replacement

Valve repair

100

Late survival (%)

80

69 ± 6%

60 58 ± 5%

40 Expected Observed 20 P = 0·0001

P = 0·77

0 0 188

1

2

3

4

5

6

7

8

9

175 164 148 127 108

85

70

50

37

10

0

25 187 Years

1

2

3

4

5

6

7

8

9

10

180 166 122

86

69

48

37

22

10

7

Figure 56.2 Plots of late survival (in operative survivors) of patients with valve replacement (left) and valve repair (right) compared with their expected survival. Note that in patients with valve repair, there is no difference in the expected survival, whereas in patients with valve replacement, the survival is significantly lower than expected. (From Enriquez-Sarano et al.7)

810

Valve repair and choice of valves

3.

Reoperation 100

80 ± 6%

80 75 ± 10%

4.

60

allow coaptation of the leaflets along several millimeters from their free margins, thus decreasing the probability of tears in areas where segments of leaflets or chordae were repaired; restoration of normal annulus shape.

% 40

Repair Replacement

20 0 0 Repair 195 Replacement 214

P = 0.47

1 173 173

2 160 160

3 4 5 117 81 67 146 120 102

6 46 80

7 35 68

8 22 49

9 11 38

10 Years 7 24

Figure 56.3 Plot of freedom from reoperation in valve repair and replacement groups. No significant difference is observed. (From Enriquez-Sarano et al.7)

presents a challenge to the surgeon, repair utilizing standard techniques after tissue decalcification and debridement does not adversely affect surgical outcome.12,13 Mitral valve repair rather than replacement is also possible in the setting of native valve endocarditis, as this results in lower hospital mortality and improved long-term outcome when compared to valve replacement.14 Thus, valve repair for mitral regurgitation, whatever the etiology, should be the first choice of surgical correction. Freedom from reoperation for structural valve-related degeneration has been reported as high as 90% at 10 years and 85% at 15 years following valve repair.15 In patients who exhibit valve failure following repair, successful rerepair can be undertaken in 16–21% of patients.16,17 Thus, the ultimate likelihood of requiring a mitral prosthesis following surgical repair of mitral regurgitation is very low. Basic concepts of repair Prolapse of a segment of the posterior leaflet is treated by triangular or quadrangular resection of the unsupported portion or by plication of the redundant leaflet tissue.18,19 In patients with anterior leaflet prolapse, with or without chordal rupture, we favor chordal replacement with GoreTex suture.20 Dilation of the valve annulus almost always accompanies mitral regurgitation.21 Progressive annular enlargement worsens regurgitation by further decreasing the area of leaflet coaptation. The dilation tends to be asymmetrical, in that it affects the mural leaflet up to the commissures.22 Dilation changes annular shape so that the anteroposterior diameter of the valve becomes greater than the transverse diameter. Because of this, an annuloplasty procedure is an integral part of mitral valve repair. The goals of an annuloplasty are fourfold: 1. 2.

decrease annulus diameter, thereby decreasing the area that the leaflets must seal; prevent further dilation of the annulus;

Annuloplasty is typically performed with a prosthetic ring;23,24 we favor a partial posterior ring25 to reorient the anterior or posterior leaflets for adequate coaptation (Figure 56.4). Postoperative valve function as assessed by degree of regurgitation, transvalvular gradient, and valve area is comparable, irrespective of which technique is utilized.25 It should be noted that the normal mitral annulus changes size and shape during the cardiac cycle.26–29 This “sphincterlike” function results in a reduction in valve area by 26% during systole, which is associated with a change in shape from circular to elliptical.30 If a flexible annuloplasty is utilized for repair rather than a rigid ring, superior left ventricular systolic function can be demonstrated early and late following valve repair.31,32

Valve replacement Grade B Choice of a valve prosthesis requires consideration of the qualities of the valve weighed against the patient’s needs. Durability of the prosthesis is often the primary concern of the patient. Indeed, when discussing valve replacement with a patient, the most commonly asked question is “How long will it [the prosthesis] last?” For currently available mechanical valves in the United States, the answer is a qualified “forever”, qualified in the sense that intrinsic material failure of mechanical valves is now extremely rare.33 However, this does not mean that a valve might not need to be replaced because of extrinsic mechanical failure (for example, pannus ingrowth inhibiting proper function of the closure mechanism), and the patient should understand these differences. Durability of biologic valves is not so well

Figure 56.4

An annuloplasty ring

811


defined for the individual patient. Indeed, as outlined below, by their very nature, tissue valves have a limited lifespan and their use must be matched to a patient’s needs. Anticoagulation and thrombogenicity are the other major issues with prosthetic valves. Mechanical valves offer excellent durability that clearly surpasses that of currently available tissue valves, but thrombus formation and thromboembolism are recognized hazards. Anticoagulation to prevent thromboembolism introduces an incremental risk to a patient. Indeed, taken together, anticoagulant-related hemorrhage and thrombosis account for up to 75% of complications following mechanical valve replacement.34 Finally, when evaluating different types of cardiac valve prostheses, one must understand the concepts of valvular hemodynamics. These are directly related to valve design and determine the work the heart must expend to pump blood through the prosthesis. All currently approved prosthetic valves have a sewing ring which, with the housing of the valve, takes up a certain cross-sectional area in the path of blood flow. This “sewing ring area” is larger for the tissue valves than for the mechanical valves. The effective orifice area (EOA) of a valve is the actual area of the valve available for blood flow. If one divides the EOA by the sewing ring area, one can calculate the performance index of a given valve. The performance index of currently available porcine valves ranges from 0·35 to 0·4, pericardial valves 0·65, and tilting disc valves from 0·67 to 0·70, so that all stent mounted prosthetic valves are, by definition, stenotic compared to normal native valves. The potential for residual outflow obstruction when a small prosthetic valve is used in a large patient gives rise to the condition termed valveprosthesis patient mismatch.35,36 For most patients, the transvalvular gradient is small and of little clinical significance. It should be noted, however, that as the valve size decreases and the EOA concomitantly decreases, there can be a precipitous rise in the transvalvular gradient, which could cancel out the clinical improvement anticipated from valve replacement. Two other aspects of valve function are often overlooked: dynamic and static prosthetic valve regurgitation. Dynamic regurgitant fraction is the amount of regurgitation that occurs through a valve before the occluder has a chance to close fully. This is lowest for the tilting disc valves, followed by the bileaflet prostheses; the greatest dynamic regurgitation is associated with the ball and cage prostheses.37 Static regurgitation occurs through a valve after the valve has closed. Some static regurgitation is engineered into most valves to wash the valve components and eliminate microemboli. Bileaflet valves and Medtronic-Hall tilting disc valves have greater static regurgitation than ball and cage valves.37 Although regurgitant volume through a normally functioning prosthesis is not important in a patient with adequate ventricular function, in the face of decreased ejection fraction, large regurgitant volumes may attenuate the hemodynamic improvement produced by valve replacement. 812

Mechanical valves Grade B Currently in the United States, there are five categories of mechanical valves approved for implication by the Food and Drug Administration. These include the St Jude (St Jude Medical, Minneapolis, MN) bileaflet prosthesis, the Medtronic-Hall (Medtronic Inc, Minneapolis, MN) tilting disc valve, the CarboMedics (CarboMedics, Austin, TX) bileaflet prosthesis, the Starr-Edwards ball valve (Baxter Healthcare, Santa Ana, CA), and the Omnicarbon tilting disc valve, which evolved from the Omniscience valve. There are few prospective, randomized studies comparing outcomes between these categories of valves in the same patient populations. Non-randomized studies and informal comparisons of published series show little difference in late patient outcome, either in morbidity or mortality, following implantation of currently approved mechanical prostheses (for review, see reference37). Several prospective, randomized studies comparing specific valves bear out this assertion. Schulte and associates randomized 150 consecutive patients to receive a tilting disc prosthesis or Starr-Edwards valve or mitral valve replacement; there was no significant difference in late patient survival (mean follow up 14·8 years) between the two tilting disc valves (Bjork-Shiley, Lillehei-Kaster) and a ball and cage prosthesis (Starr-Edwards).38 In another randomized study of 102 patients, Fiore and colleagues found no significant difference in linearized rates of valve-related events and 3 year actuarial survival between a tilting disc (Medtronic-Hall) and bileaflet (St Jude) prosthesis (Figure 56.5).39 Even when comparing an early model tilting disc prosthesis (Bjork-Shiley) with a bileaflet prosthesis (St Jude), no significant difference in early and late survival or major bleeding complications could be demonstrated in 178 patients in a prospectively randomized, European study (mean follow up of 52 months or 778 patient-years).40 Thus, with regard to clinical performance and hemodynamic data, there are no large randomized studies that definitively demonstrate the superiority and thus preferential selection of one mechanical prosthesis over another.

Bioprosthetic valves Grade B The three most commonly used bioprostheses are the Hancock porcine valve (Medtronic Inc) and the Carpentier-Edwards porcine and bovine pericardial valves (Baxter Healthcare). The main drawback of the bioprosthetic valves is structural deterioration which is not a simple linear function of time as the rate of structural dysfunction steadily accelerates after 5–6 years of implantation.41–43 Regurgitation through cusp tears associated with calcific nodules is the most frequent form of bioprosthesis failure; pure stenosis due to calcified leaflets occurs infrequently. Structural dysfunction of bioprostheses is markedly accelerated in children, adolescents, and young adults, but is attenuated in very elderly


80

80

60

60

%

100

%

100

40 20 0 Patients at risk

40

Medtronic St Jude P = NS 0 47 55

1 28 37

2 20 32

3 16 22

4 9 8

Medtronic St Jude P = NS

20 5 Years 0 Medtronic 0 St Jude

0 Patients at risk

0 47 55

1 31 36

2 22 30

3 17 20

4 8 7

5 Years 0 Medtronic 0 St Jude

Figure 56.5 Actuarial freedom from thromboembolism (left) and hemorrhage (right) after mitral valve replacement with the St Jude and Medtronic-Hall valves (NSnot significant). (From Fiore et al.39)

patients. For aortic bioprostheses, patients younger than 39–44 years of age have structurally related, event free estimates ranging from 58% to 70% at 10 years;43–45 this drops to 33% at 15 years.46 This is in contrast to patients over 70 years of age, who have event free estimates of 95% and 93% at 10 and 15 years following implantation.46 Event free estimates for patients between 60 and 69 years of age, 10 years following implantation, range from 92% to 95%.45–47 Many investigators have compared the performance of the Hancock and the Carpentier-Edwards porcine bioprostheses. In general, no significant differences in the shortand long-term performance of these valves have been demonstrated.48–51 Indeed, at 10 year follow up of 174 patients undergoing mitral or aortic valve replacement who were prospectively randomized to receive either a Hancock or Carpentier-Edwards porcine bioprosthesis, there were no significant differences in patient survival, durability of the prosthesis or valve-related complications.48 These findings were confirmed in another study of 147 patients randomized to receive either the Carpentier-Edwards or Hancock porcine bioprosthesis in the mitral position. At 10 years, no significant differences in survival or valve-related complications were apparent.52 Previously, all commercially available bioprostheses were mounted on a stent or frame to give the relatively flaccid tissue valve a fixed base to facilitate implantation. The stent and sewing ring, however, significantly decrease the EOA and make tissue valves relatively obstructive when compared with mechanical prostheses. There has been considerable interest recently in stentless bioprosthetic valves that are inserted in much the same fashion as homografts. Hemodynamic performance of stentless bioprostheses is good and like other tissue valves, no anticoagulation is required.53–56 In one report, 254 patients with the Toronto SPV stentless valve (St Jude Medical) were followed for 3 years and the initially favorable EOAs and transvalvular gradients were said to improve with time.53 In addition, left ventricular mass decreased by 14·3% in the study period.

The primary mode of failure of stentless valves, like all bioprostheses, is valvular regurgitation. Indeed, 27% of 200 patients were found to have aortic insufficiency 1 year following implantation of a stentless aortic valve (PRIMA Edwards; Baxter Healthcare); however, only one patient exhibited grade 3 insufficiency.55 In addition, in a nonrandomized study of 150 patients receiving either a stentless bioprosthesis (PRIMA Edwards), a traditional bioprosthesis or a homograft, no difference in morbidity or mortality was noted between the groups after 1 year.57 While the initial data on stentless bioprostheses in the aortic position are encouraging, further long-term studies will be needed to establish their ultimate role in the management of aortic valve disease. Development of stentless valves for the mitral position has been difficult. Because the mitral valve annulus changes shape during the cardiac cycle, a stentless prosthesis in this location requires additional external support to maintain competence. This engineering challenge has been met by using artificial chordae to anchor the stentless valve to native papillary muscle.58 Short-term success has been reported, but a stentless prosthesis for the mitral position should be considered as experimental. An additional category of tissue valves currently available for implantation are homografts. These are human tissue valves (either aortic or pulmonic) that have been harvested from cadavers, sterilized antibiotically, and cryopreserved.59,60 Homografts have many attractive features including minimal gradients, low thrombogenicity without need for anticoagulation, and low risk of infection, even when used in patients with active endocarditis.61,62 Implantation of a homograft is considered more difficult than implantation of a stent mounted bioprosthesis and both experience of the surgeon and surgical technique appear to influence late results.63 Freedom from reoperation due to structural deterioration of homografts has been reported to range from 83% at 8 years64 to 86% at 14 years.60 When valve failure occurs, it is due to the gradual development of insufficiency. 813


The other homograft available for aortic valve replacement is the pulmonary valve autograft (termed the Ross procedure). The Ross procedure involves excision of a patient’s normal pulmonic valve (autograft) and utilizing it to replace the diseased aortic valve.65,66 A cryopreserved human pulmonary artery homograft (allograft) is then implanted to replace the native pulmonic valve. There are many positive aspects of the operation. First, as both of the valves are tissue valves, no anticoagulation is required. Second, since the pulmonic valve autograft is not exposed to the antibiotic sterilization or cryopreservation process, it is viable and has potential for growth and long-term durability.67,68 In one series of 195 patients, the freedom from reoperation (autograft or allograft) was reported to be 89% at 5 years.69 Compared to allograft replacement of the aortic valve, patients receiving the pulmonary autograft have comparable hemodynamics and earlyto medium-term postoperative recovery.70 However, there are three potential drawbacks to the pulmonary autograft. First, the operation converts single valve disease to a double valve replacement. And even if the pulmonary autograft functions perfectly, there is potential for tissue degeneration and obstruction of the pulmonary allograft; in fact, the need for right-sided valve re-replacement may be underestimated. In the best of hands and in carefully selected patients, cumulative risk of reoperation for pulmonary valve substitute approaches 20% at 20 years postoperatively.65,71 In addition, there is a 10–20% incidence of autograft aortic insufficiency, grade 2, following operation.72 Long-term follow up from multiple institutions is necessary to define the safety and durability of the pulmonary autograft for aortic valve replacement. Comparative studies of mechanical v bioprosthetic valves Grade A In a prospective, randomized study in which 262 patients received either a mechanical (Bjork-Shiley) or porcine bioprosthesis (initially Hancock and subsequently Carpentier-Edwards) in the mitral position, actuarial survival and incidence of thromboembolism was comparable at 7 years follow up.73 Another prospective, randomized study also demonstrated comparable survival following valve replacement with either a mechanical valve or bioprosthesis.74 Five hundred and seventy-five men, scheduled to undergo either aortic or mitral valve replacement, were randomized either to receive a mechanical valve (Bjork-Shiley) or porcine bioprosthesis (Hancock). After 11 years, survival rates and freedom from all valve-related complications were similar for both patient groups. However, the profile of valve-related complications was different in that structural failure was only observed with the bioprosthetic valves, whereas bleeding complications were more frequent in patients with mechanical valves. Thus, while the types of complications might differ between patients with either a 814

bioprosthesis or mechanical valve, the actual incidence of the complications is comparable and survival is similar. As such, the choice between a bioprosthesis and a mechanical valve should be based on other factors. Matching the patient to the prosthesis: factors in selecting a valve for implantation Grade B Because patient survival following valve replacement is independent of the type of prosthesis used and dependent on other factors, one needs to focus on patient variables when selecting a valve. First, one must assess a patient’s life expectancy after valve replacement. For patients aged 65–69 years undergoing aortic and/or mitral valve replacement, survival is approximately 53% at 10 years and 25% at 15 years; for patients 70 years of age or older, survival is 30–38% at 10 years and 25% at 15 years.47 Coronary artery disease requiring bypass grafting at the time of valve replacement further decreases long-term survival.75 All other factors being equal, it has been our practice to suggest a mechanical valve to patients 70 years or younger and a bioprosthesis to those 75 years and older. In the “gray area” between 70 and 75 years, recommendations are made based upon a patient’s general health and personal preference. The other major issue related to the choice of a valve prosthesis is anticoagulation. Obviously, a mechanical valve, with its obligatory need for lifelong oral anticoagulation, would be contraindicated in a patient who: ● ● ● ●

has bleeding tendencies because of geography or psychosocial issues, would be unable to monitor the level of anticoagulation has an occupation with a high risk of trauma is a female of childbearing age who desires a future pregnancy.

In these situations, one of the tissue valves would be indicated. However, if a patient is likely to require anticoagulation for some other condition such as atrial fibrillation, a large left atrium, chronic deep venous thrombosis or a mechanical prosthesis in another location, then a mechanical valve is chosen for its durability. In addition, if for some reason a patient would be at great risk for reoperation and valve re-replacement, a mechanical valve is favored.

References 1.Dare AJ, Harrity PJ, Tazelaar HD, Edwards WD, Mullany CJ. Evaluation of surgically excised mitral valves: revised recommendations based on changing operative procedures in the 1990s. Hum Pathol 1993;24:1286–93. 2.Olson LJ, Subramanian R, Ackermann DM, Orszulak TA, Edwards WD. Surgical pathology of the mitral valve: a study of 712 cases spanning 21 years. Mayo Clin Proc 1987;62:22–34.


3.Waller BF, Morrow AG, Maron BJ et al. Etiology of clinically isolated, severe, chronic, pure mitral regurgitation: an analysis of 97 patients over 30 years of age having mitral valve replacement. Am Heart J 1982;104:276–88. 4.Enriquez-Sarano M, Tajik AJ, Schaff HV et al. Echocardiographic prediction of survival after surgical correction of organic mitral regurgitation. Circulation 1994;90:830–7. 5.Ling LH, Enriquez-Sarano M, Sewrad JB et al. Clinical outcome of mitral regurgitation due to flail leaflet. N Engl J Med 1996;355:1417–23. 6.Perier P, Deloche A, Chauvaud S et al. Comparative evaluation of mitral valve repair and replacement with Starr, Bjork, and porcine valve prostheses. Circulation 1984;70:187–92. 7.Enriquez-Sarano M, Schaff HV, Orszulak TA et al. Valve repair improves the outcome of surgery for mitral regurgitation: a multivariate analysis. Circulation 1995;91:1022–8. 8.Goldman ME, Mora F, Guarino T, Fuster V, Mindich BP. Mitral valvuloplasty is superior to mitral valve replacement for preservation of left ventricular function: an intraoperative twodimensional echocardiographic study. J Am Coll Cardiol 1987; 10:568–75. 9.Rozich JD, Carabello BA, Usher BW et al. Mitral valve replacement with and without chordal preservation in patients with chronic mitral regurgitation. Mechanisms for differences in postoperative ejection performance. Circulation 1992; 86:1718–26. 10.David TE, Uden DE, Strauss HD. The importance of the mitral apparatus in left ventricular function after correction of mitral regurgitation. Circulation 1983;68:1176–83. 11.David TE, Burns RJ, Bacchus CM, Druck MN. Mitral regurgitation with and without preservation of chordae tendineae. J Thorac Cardiovasc Surg 1984;88:718–25. 12.Grossi EA, Galloway AC, Steinberg BM et al. Severe calcification does not affect long-term outcome of mitral valve repair. Ann Thorac Surg 1994;58:685–8. 13.Carpentier AF, Pellerin M, Fuzellier JF, Relland JYM. Extensive calcification of the mitral valve annulus: pathology and surgical management. J Thorac Cardiovasc Surg 1996;111:718–30. 14.Muehrcke DD, Cosgrove DM, Lytle BW et al. Is there an advantage to repairing infected mitral valves? Ann Thorac Surg 1997;63:1718–24. 15.Alvarez JM, Deal CW, Loveridge K et al. Repairing the degenerative mitral valve: ten to fifteen year follow-up. J Thorac Cardiovasc Surg 1996;112:238–47. 16.Cerfolio RJ, Orszulak TA, Pluth JR, Harmsen WS, Schaff HV. Reoperation after valve repair for mitral regurgitation: early and immediate results. J Thorac Cardiovasc Surg 1996;111: 1177–84. 17.Gillinov AM, Cosgrove DM, Lytle BW et al. Reoperation for mitral valve repair. J Thorac Cardiovasc Surg 1997;113: 467–75. 18.Carpentier A. Cardiac valve surgery: the French connection. J Thorac Cardiovasc Surg 1983;86:323–37. 19.McGoon DC. Repair of mitral insufficiency due to ruptured chordae tendinae. J Thorac Cardiovasc Surg 1960;39: 357–62. 20.David TE, Armstrong S, Sun Z. Replacement of chordae tendineae with Gore-Tex sutures: a ten-year experience. J Heart Valve Dis 1996;5:352–5.

21.Ormiston JA, Shah PM, Tei C, Wong M. Size and motion of the mitral valve annulus in man. II. Abnormalities in mitral valve prolapse. Circulation 1982;65:713–19. 22.Carpentier A. Plastic and reconstructive mitral valve surgery. In Kalmanson D, ed. The mitral valve, a pluridisciplinary approach. London: Publishing Science Group, 1976. 23.Carpentier A, Deloche A, Dauptain J et al. A new reconstructive operation for correction of mitral and tricuspid insufficiency. J Thorac Cardiovasc Surg 1971;61:1–13. 24.Duran CMG, Umbago JL. Clinical and hemodynamic performance of a totally flexible prosthetic ring for atrioventricular valve reconstruction. Ann Thorac Surg 1976;22:458–63. 25.Odell JA, Schaff HV, Orszulak TA. Early results of a simplified method of mitral valve anuloplasty. Circulation 1995;92 (Suppl. II):II-150–4. 26.David TE, Strauss HD, Mesher E et al. Is it important to preserve the chordae tendineae and papillary muscles during mitral valve replacement? Can J Surg 1981;24:236–9. 27.Hansen DE, Cahill PD, DeCampli WM et al. Valvular ventricular interactions: importance of the mitral apparatus in canine left ventricular systolic performance. Circulation 1986;73: 1310–20. 28.Hansen DE, Cahill PD, Derby GC, Miller DC. Relative contributions of the anterior and posterior mitral chordae tendineae to canine global left ventricular systolic performance. J Thorac Cardiovasc Surg 1987;93:45–55. 29.Sarris GE, Cahill PD, Hansen DE et al. Restoration of left ventricular systolic performance after reattachment of the mitral chordae tendineae. The importance of the valvular-ventricular interaction. J Thorac Cardiovasc Surg 1988;95:969–79. 30.Ormiston JA, Shah PM, Tei C, Wong M. Size and motion of the mitral annulus in man: a two-dimensional echocardiographic method and findings in normal subjects. Circulation 1981;64:113–20. 31.David TE, Komeda M, Pollick C, Burns RJ. Mitral valve annuloplasty: the effect of the type on left ventricular function. Ann Thorac Surg 1989;47:524–8. 32.Duran CG, Revuelta JM, Gaite L, Alonso C, Fleitas MG. Stability of mitral reconstructive surgery at 10–12 years for predominantly rheumatic valvular disease. Circulation 1988; 78:191–6. 33.Grunkemeier GL, Rahimtoola SH. Artificial heart valves. Annu Rev Med 1990;41:251–63. 34.Edmonds LH. Thrombotic and bleeding complications of prosthetic heart valves. Ann Thorac Surg 1987;44:430–45. 35.Rahimtoola SH. The problem of valve prosthesis-patient mismatch. Circulation 1978;58:20–4. 36.Rahimtoola SH, Murphy E. Valve prosthesis-patient mismatch. A long-term sequela. Br Heart J 1981;45:331–5. 37.Akins CW. Results with mechanical cardiac valvular prostheses. Ann Thorac Surg 1995;60:1836–44. 38.Schulte HD, Horstkotte D, Bircks W, Strauer BE. Results of a randomized mitral valve replacement with mechanical prostheses after 15 years. Int J Artif Organs 1992;15:611–16. 39.Fiore AC, Naunheim KS, d’Orazio S et al. Mitral valve replacement: randomized trial of St. Jude and Medtronic-Hall prostheses. Ann Thorac Surg 1992;54:68–73. 40.Vogt S, Hoffmann A, Roth J et al. Heart valve replacement with the Bjork-Shiley and St. Jude Medical

815


prostheses: a randomized comparison in 178 patients. Eur Heart J 1990; 11:583–91. 41.Jamieson WRE, Murno AI, Miyagishima RT et al. CarpentierEdwards standard porcine bioprosthesis: clinical performance to seventeen years. Ann Thorac Surg 1995;60:999–1007. 42.Glower DD, White WD, Hatton AC et al. Determinants of reoperation after 960 valve replacements with CarpentierEdwards prostheses. J Thorac Cardiovasc Surg 1994;107: 381–93. 43.Pelletier LC, Carrier M, Leclerc Y et al. Influence of age on late results of valve replacement with porcine bioprostheses. J Cardiovasc Surg 1992;33:526–33. 44.Cohn LH, Collins JJ Jr, DiSesa V et al. Fifteen-year experience with 1,678 Hancock porcine bioprosthetic heart valve replacements. Ann Surg 1989;210:435–43. 45.Jones EL, Weintraub WS, Craver JM et al. Ten-year experience with the porcine bioprosthetic valves; interrelationship of valve survival and patient survival in 1,050 valve replacements. Ann Thorac Surg 1990;49:370–84. 46.Burdon TA, Miller DC, Oyer PE et al. Durability of porcine valves at fifteen years in a representative North American population. J Thorac Cardiovasc Surg 1992;103:238–52. 47.Burr LH, Jamieson WRE, Munro AI et al. Porcine bioprostheses in the elderly: clinical performance by age groups and valve positions. Ann Thorac Surg 1995;60:S264–9. 48.Sarris GE, Robbins RC, Miller DC et al. Randomized, prospective assessment of bioprosthetic valve durability: Hancock verses Carpentier-Edwards valves. Circulation 1993;88 (pt 2):55–64. 49.Bolooki H, Kaiser GA, Mallon SM, Palatianos GM. Comparison of long-term results of Carpentier-Edwards and Hancock bioprosthetic valves. Ann Thorac Surg 1986;42:494–9. 50.Hartz RS, Fisher EB, Finkelmeier B et al. An eight-year experience with porcine bioprosthetic cardiac valves. J Thorac Cardiovasc Surg 1986;91:910–17. 51.McDonald ML, Daley RC, Schaff HV et al. Hemodynamic performance of a small aortic valve bioprostheses: is there a difference? Ann Thorac Surg 1997;63:362–6. 52.Perier P, Deloche A, Chauvaud S et al. A ten-year comparison of mitral valve replacement with Carpentier-Edwards and Hancock porcine bioprostheses. Ann Thorac Surg 1989; 48:54–9. 53.Del Rizzo DF, Goldman BS, Christakis GT, David TE. Hemodynamic benefits of the Toronto Stentless Valve. J Thorac Cardiovasc Surg 1996;112:1431–45. 54.Sintek CF, Fletcher AD, Khonsari S. Small aortic root in the elderly: use of a stentless bioprosthesis. J Heart Valve Dis 1996;5(Suppl. 3):S308–13. 55.Dossche K, Vanermen H, Daenen W, Pillai R, Konertz W. Hemodynamic performance of the PRIMA Edwards stentless aortic xenograft: early results of a multicenter clinical trial. Thorac Cardiovasc Surg 1996;44:11–14. 56.Wong K, Shad S, Waterworth PD et al. Early experience with the Toronto stentless porcine valve. Ann Thorac Surg 1995;60(Suppl. 2):S402–5. 57.Dossche K, Vanermen H, Wellens F et al. Free-hand sewn allografts, stentless (Prima Edwards) and stented (CESA) porcine

816

bioprostheses. A comparative hemodynamic study. Eur J Cardiothorac Surg 1995;9:562–6. 58.Deac RF, Simionescu D, Deac D. New evolution in mitral physiology and surgery: mitral stentless pericardial valve. Ann Thorac Surg 1995;60(Suppl. 2):S433–8. 59.McGriffin DC, O”Brien MF, Stafford EG et al. Long-term results of the viable cryopreserved allograft valve: continuing evidence for superior valve durability. J Cardiac Surg 1988; 3(Suppl.):289. 60.O’Brien MF, McGriffin DC, Stafford EG et al. Allograft aortic valve replacement: long-term comparative clinical analysis of the viable cryopreserved and antibiotic 4 C stored valves. J Cardiac Surg 1991;6(Suppl. 4):534. 61.Tuna IC, Orszulak TA, Schaff HV, Danielson GK. Results of homograft aortic valve replacement for active endocarditis. Ann Thorac Surg 1990;49:619–24. 62.Dearani JA, Orszulak TA, Schaff HV et al. Results of allograft aortic valve replacement for complex endocarditis. J Thorac Cardiovasc Surg 1997;113:285–91. 63.Dearani JA, Orszulak TA, Daly RC et al. Comparison of techniques for implantation of aortic valve allografts. Ann Thorac Surg 1996;62:1069–75. 64.Kirklin JK, Naftel DC, Novick W et al. Long-term function of cryopreserved aortic valve homografts: a ten year study. J Thorac Cardiovasc Surg 1993;106:154–66. 65.Ross D, Jackson M, Davies J. Pulmonary autograft aortic valve replacement: long-term results. J Cardiac Surg 1991;6: 529–53. 66.Elkins RC, Santangelo K, Stelzer P, Randolph JD, Knott-Craig CJ. Pulmonary autograft replacement of the aortic valve: an evolution of technique. J Cardiac Surg 1992;7:108–16. 67.Gerosa G, McKay R, Ross DN. Replacement of the aortic valve or root with an autograft in children. Ann Thorac Surg 1991;51:424. 68.Walls JT, McDaniel WC, Pope ER et al. Documented growth of autogenous pulmonary valve translocated to the aortic valve position (letter). J Thorac Cardiovasc Surg 1994;107:1530. 69.Elkins RC, Lane MM, McCue C. Pulmonary autograft reoperation: incidence and management. Ann Thorac Surg 1996; 62:450–5. 70.Santini F, Dyke C, Edwards S et al. Pulmonary autograft versus homograft replacement of the aortic valve: a prospective randomized trial. J Thorac Cardiovasc Surg 1997;113:894–900. 71.Ross D. Replacement of the aortic valve with a pulmonary autograft: the “switch” operation. Ann Thorac Surg 1991;52:1346. 72.Elkins RC. Editorial: pulmonary autograft – the optimal substitute for the aortic valve? N Engl J Med 1994;330:59. 73.Bloomfield P, Kitchin AH, Wheatley DJ et al. A prospective evaluation of the Bjork-Shiley, Hancock, and Carpentier-Edwards heart valve prostheses. Circulation 1993;88:1155–64. 74.Hammermeister KE, Sethi GK, Henderson WG et al. A comparison of outcomes in men 11 years after heart-valve replacement with a mechanical valve or bioprosthesis. N Engl J Med 1993;328:1289. 75.Jones EL, Weintraub WS, Craver JM et al. Interaction of age and coronary disease after valve replacement: implications for valve selection. Ann Thorac Surg 1994;58:378–85.

57

Diagnosis and management of infective endocarditis David T Durack, Michael L Towns

The diagnosis and management of infective endocarditis (IE) raises many questions and clinical decisions which invite application of the principles of evidence-based medicine. Most of these have not been formally asked or answered by means of controlled clinical studies. Current practice is based upon an extensive accumulation of uncontrolled clinical experience, rather than upon validated clinical trials. Here we will discuss common issues that arise during diagnosis and management of IE. Recommendations will be offered, along with an evidence-based grading (on an A, B, C scale) of the basis for each recommendation.

Background Pathophysiology Endocarditis refers to inflammation of the endocardial lining of the heart. The heart valves are most often involved, or less commonly the lining of the heart chambers (mural endocarditis). When the lesions of endocarditis (vegetations) contain micro-organisms, the associated disease is termed infective endocarditis. This general term covers the various clinical subcategories of the disease (for example, acute, subacute, prosthetic valve infection) and also the various etiologic agents (bacteria, yeasts or fungi). The pathophysiology of this disease often begins with the formation of non-bacterial thrombotic endocarditis (NBTE). This lesion, which is sometimes called a “fibrin-platelet plug”, is a receptive precursor site which may become infected by circulating organisms during the course of a bacteremia or fungemia.1–3 NBTE is not normally found in healthy hearts, but it may develop on an endocardial lining which has been damaged by one of several mechanisms. One of the most common pathogenic mechanisms is that of a cardiac valvular lesion, such as scarring or stenosis, leading to high velocity turbulent flow across the valve, with resultant damage to the endothelial lining.4,5 The damaged area may become a locus for deposition of fibrin and platelets, resulting in NBTE. The type of underlying cardiac valvular lesion determines where a vegetation is most likely to form on the endocardial surface. A bacteremia caused by organisms that have the capacity to adhere to this lesion, mediated

by surface factors such as adhesins, may seed the NBTE and lead to development of an infected vegetation.1–3,6 Vegetations are the pathologic hallmark of IE.1,2,6 They are composed of masses of organisms enmeshed with fibrin, platelets, and a variable (often scanty) inflammatory infiltrate. The vegetations may be of various sizes, and may or may not progress to cause further valvular, perivalvular, or extracardiac complications. Valvular complications may include valvular dysfunction, destruction, or obstruction. Perivalvular complications include extension of infection into adjacent structures, which may result in formation of a perivalvular abscess. Extracardiac complications most commonly result from embolic phenomena such as embolization into the coronary arteries or the systemic arterial tree, resulting in ischemia, infarcts, and sometimes secondary bleeding. Less commonly, abscesses or mycotic aneurysms may develop in various organs. Other extracardiac complications may include immune complex mediated disease such as glomerulonephritis. Epidemiology IE has been variously categorized in the past as acute, subacute, chronic, native valve, prosthetic valve, culture-negative, and intravenous drug abuse associated endocarditis. These terms have some value, but they may overlap. It is useful to specify the infecting organism because this allows prediction of the likely natural history, treatment requirements, and prognosis for an individual patient. Here we will briefly discuss the epidemiology of IE in the context of three main categories: native valve, prosthetic valve, and culturenegative endocarditis.7–9 Table 57.1 shows the etiologic agents that are most commonly isolated in native valve IE. Cases caused by virulent pathogens such as Streptococcus pneumoniae or Staphylococcus aureus may develop on previously normal valves. More often, native valve endocarditis develops in association with predisposing congenital or acquired valvular lesions, especially when caused by less virulent organisms such as the viridans streptococci. Prosthetic valve endocarditis can be subcategorized into early (onset up to 60 days after valve replacement), 817


Table 57.1 Frequency of various organisms isolated in native value infective endocarditis Organism Streptococci Viridans, alpha-hemolytic Strep. bovis (group D) Strep. faecalis (group D) Other streptococci Staphylococci Coagulase-positive Coagulase-negative Gram-negative aerobic bacilli Fungi Miscellaneous bacteria Diphtheroids, propionibacteria Other anaerobes Rickettsia Chlamydia Polymicrobial infection Culture-negative endocarditis

NVE (%) 65 35 15 10 5 25 23 5 5 5 5 1 1 1 1 1 5–10

IV drug abusers (%)

Early PVE (%)

Late PVE (%)

15 5 5 8 5 50 50 5 5 5 5 5 1 1 1 5 5

5 5 5 5 5 50 20 30 15 10 5 5 1 1 1 5 5

35 25 5 5 5 30 10 20 10 5 5 5 1 1 1 5 5

These are representative figures collated from the literature; wide local variations in frequency are to be expected. Abbreviations: NVE, native valve endocarditis; PVE, prosthetic valve endocarditis Reproduced with permission, from Durack7

intermediate (onset from 2 to 12 months) or late cases (onset after one year). The observed spectrum of etiologic agents is different for the two categories, with the organisms causing late onset prosthetic valve endocarditis more closely resembling native valve subacute endocarditis, except that coagulasenegative staphylococci remain important (Table 57.1). Culture-negative IE remains fairly common (3–30% of cases in recent series), despite improvements in blood culture techniques and culture media. Organisms that previously were difficult to recover, such as nutritionally variant streptococci and the fastidious Gram-negatives (HACEK group: Haemophilus spp, Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella spp, and Kingella kingae) are now routinely isolated from modern, optimized blood culture media, usually within 3–5 days. An exception to this is Bartonella spp which have recently been found in association with endocarditis among homeless individuals, and as a rare opportunistic infection in patients with AIDS.10,11 Diagnosis Clinical manifestations Patients with acute IE typically present with an accelerated course typified by high fever, chills, and prostration, whereas those with subacute endocarditis present more insidiously. 818

These patients often have a “flu-like illness” consisting of fever, chills, myalgias/arthralgias, and weakness, but there is great variability in the clinical presentation.7 Cardiac manifestations may dominate the clinical presentation in either acute or subacute disease, with the presence of new or worsened murmurs, or development of cardiac failure due to valvular damage. The patient may present with chest pain due to pleuritis, pericarditis or myocardial infarction resulting from coronary arterial embolism. Extracardiac clinical manifestations consist of embolic as well as vascular phenomena. The patient may present with a headache without any definable neurologic abnormalities, or may have focal abnormalities such as areas of cerebritis, infarcts, hemorrhages or mycotic aneurysms. A cellular reaction in the cerebrospinal fluid (with or without meningismus) may also be present, although only a minority of patients have positive cerebrospinal fluid cultures. The patient may present with focal pain such as flank or left-sided upper quadrant abdominal pain due to embolic infarcts, which may at times be complicated by the formation of abscesses, especially in the spleen. There are many other potential sites for embolization with associated clinical findings, although autopsy findings show that many emboli go undetected during life. Various other vascular phenomena may occur, including petechiae, splinter hemorrhages, Osler’s nodes, Janeway lesions, or clubbing of the fingernails.

Infective endocarditis

Laboratory tests Anemia is commonly present, usually of mild to moderate severity with a normochromic, normocytic film typical of the anemia of chronic disease. Although many patients with acute or subacute endocarditis have some degree of leukocytosis, this is not a reliable laboratory finding. In approximately 90% of patients with infective endocarditis the erythrocyte sedimentation rate (ESR) is elevated; the median value is about 65 mm/h, but the range is wide and about 10% are within the normal range. Urinalysis may show microscopic hematuria and/or mild proteinuria in approximately 50% of cases, with occasional red blood cell casts and heavy proteinuria in those patients who develop immune-complex glomerulonephritis. Non-specific serologic abnormalities are common, especially positive rheumatoid factor which is seen in 30–40% of cases of the subacute form of the disease. A polyclonal increase in gammaglobulins is characteristic of active endocarditis. Microbiology Blood cultures remain the definitive microbiologic procedure for diagnosis of infective endocarditis.12–15 The microorganisms isolated from blood cultures may provide the clinician with clues to the diagnosis, given the clinical setting. For example, patients who present from the community with a fever of unknown origin who have multiple positive blood cultures for viridans group streptococci, enterococci, or the HACEK organisms, should be considered to have IE until proven otherwise.12,16,17 In addition, the temporal pattern of positive cultures may assist in the diagnosis. If three or more blood culture sets drawn at least one hour apart all are positive for the same micro-organism, this is termed “persistent bacteremia”, which indicates that an endovascular infection may be present. Table 57.1 shows the leading organisms isolated from patients with acute, subacute, and prosthetic valve infective endocarditis. What are the optimal blood culture techniques required to diagnose infective endocarditis? Background A positive blood culture is one of the two major diagnostic criteria for IE.13 Therefore, blood cultures should be obtained from every patient in whom this diagnosis is suspected. Optimal techniques are required in order to minimize the number of patients with infective endocarditis that fall into the “culture-negative” category, without resorting to an excessive number of costly blood cultures.14,15

of blood.18 If this were true in every case, it would only be necessary to draw one single sample of about one milliliter of venous blood in order to make the diagnosis. In practice, however, some patients with IE have intermittent or fluctuating bacteremia, and some have less than one organism per milliliter of blood. Therefore, the number of positive culture results is directly correlated with the number of blood samples drawn and the volume of blood in each individual sample. Single samples should not be drawn because the most common contaminants of blood cultures, coagulasenegative staphylococci from the skin, can cause IE.12,19–21 Therefore, a single sample drawn from a patient who might have IE, which is positive for a coagulase-negative staphylococcus, is uninterpretable. Overall, about two thirds of all samples drawn from patients with IE are positive. This figure represents the combined results from two patient populations. The first group includes the “classical” untreated IE patient with continuous bacteremia in whom all or nearly all cultures will be positive.22 In such patients, more than 90% will be diagnosed by the first sample drawn, rising to more than 95% from three cultures.19–22 The second population is a mixed group in whom the proportion of positive cultures is much lower. Many of these patients have received some antibiotic treatment, such as empirical oral ampicillin or cephalosporin, which has temporarily or permanently suppressed the bacteremia and turned the blood cultures negative without curing the underlying endocarditis. Others may have difficult-to-culture organisms, fungal infections or culture-negative IE.14 In order to decrease the number of “culture-negative” endocarditis episodes, investigators have tried to improve the yield by drawing blood during fever spikes, or by culturing arterial instead of venous blood.23 These practices are of marginal or no value. The majority of clinical microbiology laboratories routinely hold their blood culture bottles for 5–7 days before issuing a negative report. Because some of the etiologic agents, for example, HACEK group organisms,24,25 have been traditionally regarded as slow-growers, some laboratories have adopted the policy of prolonging incubation times for blood cultures to 14–21 days in cases of suspected infective endocarditis. Recent data, however, suggest that with modern, improved blood culture media this practice may be unnecessary for all but a very few organisms, such as Bartonella spp.10,12,26,27 Should transesophageal echocardiography be performed in all patients with suspected infective endocarditis?

Evidence

Background

Typically, the bacteremia associated with endocarditis is continuous, with 10–200 colony-forming units per milliliter

Transthoracic M-mode echocardiography (TTE) was first used for the detection of vegetations associated with endocarditis 819


Conclusions

Grading

Comments/references

Draw at least two sets (two separate venepunctures, with each sample divided equally between two bottles) for each blood culture ordered

Grade A

This helps to identify contaminants and increases yield of positives12,14,15,19

Inoculate 8–12 ml blood into each bottle

Grade B

This maximizes yield of positives12,14,15,19

Hold the culture bottles for 14–21 days before issuing the final negative report in order to minimize “culture-negative” episodes (not recommended)

Grade C

The yield is very low after 5 days12,27

Draw an arterial blood sample for culture if venous blood samples are negative but the diagnosis of IE still seems likely (not recommended)

Grade C

The benefit of culturing arterial blood is none or very small23,28

in 1973. Several years later a report describing two dimensional transthoracic echocardiographic findings of vegetations was published. Since then there have been many reports on the use of this technology to assist in the diagnosis of endocarditis.29–45 The sensitivity of the procedure for detection of vegetations is 60–75%.29–32 Transesophageal echocardiography (TEE) was initially described in the late 1980s, and has proved especially valuable in evaluating patients with suspected endocarditis. TEE is more sensitive than TTE for detection of vegetations, abscesses, valve perforations, and other complications of IE.30,35 Because a TEE examination is more costly than a TTE examination, many comparative studies have been undertaken to determine which technology should be used in the initial diagnostic evaluation of a patient with suspected IE. Evidence Multiple studies have demonstrated the superior sensitivity of TEE when compared to TTE. However, this fact does not resolve the question of which is the most appropriate and cost effective test for IE in patients with different pretest probabilities of having that disease. Transthoracic echocardiography (TTE) has an overall sensitivity for detection of intracardiac vegetations of 60– 75%.32 Transesophageal echocardiography (TEE) has greater sensitivity – 95% or better overall, although the sensitivity in an individual case varies depending upon factors such as the location and size of the vegetations.30,35–40 TEE is far superior to TTE in detecting abscesses in patients with both native and prosthetic valve endocarditis (PVE), with a sensitivity of detection of 87%, as compared to 28% with TTE in one study.41 Because patients with PVE are more likely than those with native valve endocarditis (NVE) to have perivalvular abscesses, it is now accepted that TEE is the technique of choice in evaluating a patient with suspected PVE. TEE should also be applied in cases of NVE

820

where there is a prolonged clinical course of infection, as well as those patients who do not respond to adequate medical therapy. The need for TEE in the initial evaluation of patients with NVE, however, is not so clear. In a retrospective analysis of 180 patients referred for echocardiography for suspected infective endocarditis, in whom both TTE and TEE were done, the TTE was reported as technically inadequate in 46 patients (25%). In the remaining 134 patients, there was an almost equal distribution of patients who had a positive TTE (41 patients), a negative TTE (46 patients), and an abnormal but non-diagnostic TTE (47 patients). All patients who had a positive TTE were subsequently found to have a positive TEE, while only two patients with a negative TTE were found to have a positive TEE, yielding a sensitivity of 100% and a specificity of 96%. The principal value of the TEE was in the non-diagnostic group, as well as those with a technically inadequate TTE. In the non-diagnostic group, 9 patients (41%) were found to have positive TEE results for vegetations or abscesses. The study concluded that for initial evaluation of suspected native valve endocarditis, a TTE should be the first echocardiographic study. If the TTE is technically inadequate, then a TEE should be performed. If the TTE is clearly positive, or clearly negative, no additional echocardiographic study should be performed, as there was no incremental diagnostic value with TEE. A TEE, however, should be routinely performed if the TTE is abnormal but non-diagnostic. Another study analyzed the diagnostic value of echocardiography in suspected infective endocarditis, based on the pretest probability of disease.42 In this study, both TTE and TEE were performed on 105 consecutive patients with suspected endocarditis. On the basis of clinical criteria and (separately) echocardiography, patients were classified as having either low, intermediate, or high probability of endocarditis. Echocardiography had low diagnostic value in patients with a low clinical probability of endocarditis, using either TTE or TEE. The authors concluded that echocardiography


Conclusions

Grading

Comments/references

Echocardiography should not be used routinely as a screening test to “rule out endocarditis” in patients with fever and murmur

Grade B

Not cost effective unless there is other evidence of IE, raising the pretest probability30,32,42

For suspected native valve infective endocarditis, TTE should be the initial echocardiographic study

Grade A

This is the most cost effective approach32,42

If the TTE is technically inadequate in a patient with intermediate or high clinical probability of IE, then TEE should be performed

Grade A

Otherwise the diagnosis may be missed32,42

If the TTE is abnormal but non-diagnostic in a patient with intermediate or high pretest clinical probability of IE, then TEE should be performed

Grade B

TEE is more sensitive32,42

If the TTE is negative or abnormal but non-diagnostic in a patient with high pretest clinical probability of IE, then TEE should be performed

Grade B

TEE is more sensitive32,42

If the TTE is technically adequate and positive, no additional echocardiographic studies are warranted initially – that is, it is not necessary to “confirm” a positive TTE with a TEE study

Grade B

Note however that TEE may be performed for other reasons, such as to detect abscesses32,42

In patients with suspected prosthetic valve endocarditis, TEE should be performed

Grade A

TEE is best for detection of abscesses30,32,41,42

should not be used to make a diagnosis of IE in patients with a low clinical probability of disease. In addition, for those patients with an intermediate or high clinical probability of IE, TTE should be the initial echocardiographic procedure, reserving TEE for those patients with prosthetic valves and those with either a technically inadequate TTE, or a TTE which indicates an intermediate probability of endocarditis.42

How can the diagnosis of suspected IE be confirmed? Background The vegetations of IE are located in an inaccessible site, and can be visualized directly only at surgery or autopsy. Therefore, for purposes of initial diagnosis of IE they must be visualized indirectly, usually by means of echocardiography. Positive findings on echocardiography are a major criterion for diagnosis of IE, but they are not definitive because of possible false positive or false negative results.43–45 Likewise, blood cultures, which constitute the second major criterion for diagnosis of IE, also can yield false positive or false negative results.

Evidence In 1981, von Reyn and colleagues46 published a paper on infective endocarditis in which they proposed a set of diagnostic criteria which designated cases as definite, probable, possible, or rejected. These criteria, however, contained some confusingly worded definitions and did not utilize findings from echocardiography, which had only recently come into general use. In 1994, Durack and colleagues from the Duke Endocarditis Service published improved criteria which introduced the concept of major and minor diagnostic criteria and included echocardiographic findings.13 (Tables 57.2 and 57.3). Subsequently, multiple studies have analyzed cases diagnosed by the gold standard of pathologic confirmation at surgery or autopsy, comparing both sets of criteria. In each of these studies the Duke criteria were found to be notably more sensitive than the von Reyn criteria.47–51 In most of these studies, it was felt that the inclusion of echocardiographic data was the primary factor resulting in the increased sensitivity, although even when compared with a modified von Reyn classification with addition of echocardiographic data, there still was an increase in sensitivity. Often increased sensitivity is associated with a concomitant decrease in specificity, but two studies indicate that the Duke criteria have good specificity.52,53 These criteria

821


Table 57.2 Criteria for diagnosis of infective endocarditis Definite infective endocarditis Pathologic criteria Micro-organisms: demonstrated by culture or histology in a vegetation, or in a vegetation that has embolized, or in an intracardiac abscess, or Pathologic lesions: vegetation or intracardiac abscess present, confirmed by histology showing active endocarditis Clinical criteria (use specific definitions listed in Table 57.3) 2 major criteria, or 1 major and 3 minor criteria, or 5 minor criteria Possible infective endocarditis Findings consistent with infective endocarditis that fall short of “Definite,” but not “Rejected” Rejected Firm alternative diagnosis for manifestations of endocarditis, or Resolution of manifestations of endocarditis, with antibiotic therapy for 4 days or less, or No pathologic evidence of infective endocarditis at surgery or autopsy, after antibiotic therapy for 4 days or less Adapted from Durack et al 13 with permission

Conclusions

Grading

The diagnosis of IE is certain only if confirmed by suitable pathologic specimens and/or cultures obtained at surgery or autopsy

Grade A Echocardiography can yield false positives13

A “definite” diagnosis of IE (more than 95% confidence) can be made without surgical or autopsy specimens if defined major and minor criteria (the Duke criteria) are properly applied

Grade A

13,47–50

The diagnosis of IE can be rejected with high specificity if defined clinical criteria are properly applied, but this usually requires some delay to allow a period of observation

Grade A

52,53

The decision as to whether or not to begin antibiotics should be made on the overall clinical assessment as to the likelihood of IE, not based solely upon the Duke criteria

Grade B Treatment decisions often need to be made before all diagnostic information is available13,53

should be useful to specify patient entry criteria for epidemiologic studies and clinical trials involving IE. Can IE be cured with bacteriostatic antimicrobials? Background Antimicrobial agents are traditionally classified as bactericidal or bacteriostatic, according to whether they kill or 822

Comments/references

inhibit growth, respectively. In fact, this classification is an oversimplification because an antimicrobial drug may be partially bactericidal, or may be bacteriostatic for one species of bacteria and bactericidal for another. There is a widely quoted “general rule” that IE should be treated only with bactericidal drugs. The rationale often given to support this “rule” is that colonies of bacteria within a vegetation are protected from host defenses, especially neutrophils, which in other sites would usually eliminate organisms that had been inhibited by bacteriostatic antibiotics.


Table 57.3 Definitions of terminology used in the diagnostic criteria for endocarditis Major criteria Positive blood culture for infective endocarditis Typical micro-organism for infective endocarditis from two separate blood cultures: Viridans streptococci,a Streptococcus bovis, HACEK group, or community acquired Staphylococcus aureus or enterococci, in the absence of a primary focus, or Persistently positive blood culture, defined as recovery of a micro-organism consistent with infective endocarditis from: (i) Blood cultures drawn more than 12 hours apart, or (ii) All of three or a majority of four or more separate blood cultures, with first and last drawn at least 1 hour apart Evidence of endocardial involvement Positive echocardiogram for infective endocarditis (i) Oscillating intracardiac mass, on valve or supporting structures, or in the path of regurgitant jet, or on implanted material, in the absence of an alternative anatomic explanation, or (ii) Abscess, or (iii) New partial dehiscence of prosthetic valve, or New valvular regurgitation (increase or change in pre-existing murmur not sufficient) Minor criteria ● Predisposition: predisposing heart condition or injection drug use ● Fever: 38·0 C (100·4 F) ● Vascular phenomena: major arterial emboli, septic pulmonary infarcts, mycotic aneurysm, intracranial hemorrhage, conjunctival hemorrhages, Janeway lesions ● Immunologic phenomena: glomerulonephritis, Osler’s nodes, Roth spots, rheumatoid factor ● Microbiologic evidence: positive blood culture, but not meeting major criterion as previously definedb or serologic evidence of active infection with organism consistent with infective endocarditisc ● Echocardiogram: consistent with infective endocarditis but not meeting major criterion as previously defined a

Including nutritional variant strains. Excluding single positive cultures for coagulase-negative staphylococci or organisms that do not cause endocarditis. c Positive serology for Coxiella burnetii or Bartonella spp may be used as a major criteria. HACEK, Haemophilus spp, Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella spp, and Kingella kingae Source: adapted from Durack et al 13 b

Evidence In the early days of antimicrobial therapy before penicillin was available, patients with IE were often treated with prolonged courses of sulfonamides. This nearly always failed. For example, in one study none of 42 patients with streptococcal IE treated with sulfonamides survived.54 On the other hand, sulfonamide therapy occasionally cured a fortunate patient.55 Sulfonamides were most likely to succeed in the small subgroup of cases of IE caused by Haemophilus spp, which are especially susceptible to sulfonamides. In the special case of IE caused by Coxiella burnetii (the organism causing Q fever) bacteriostatic antibiotics such as tetracyclines are generally used for lack of better alternatives. In most cases they suppress but do not cure the endocardial infection; valve replacement surgery is required to increase

the likelihood of cure. Similarly, few antibiotics are available to treat IE caused by resistant Gram-negative bacilli such as Pseudomonas cepacia or Stenotrophomonas maltophilia, in these the combination of a (bacteriostatic) sulfonamide plus trimethoprim has been used with some success.56 Should combinations of antimicrobials be used to treat IE? Background IE is generally regarded as being difficult and/or slow to cure. Therefore, many attempts have been made to improve cure rates by using optimal antimicrobial regimens, even more so than in most other infections. In the course of this effort, many combinations of antibiotics have been tried, 823


Conclusions

Grading

Bacteriostatic antibiotics often fail when used to treat IE

Grade A There are animal data and case reports to show this54,55

Bacteriostatic antibiotics may be used to treat IE in a few special cases, for example, Q fever, resistant organisms like Pseudomonas cepacia or Stenotrophomonas maltophilia, or suppressive therapy for organisms not likely to be curable, such as Pseudomonas on a prosthetic valve

Grade B Uncontrolled case reports show that cure or useful suppression can be achieved in some patients56,57

and a general impression exists that combination therapy is optimal for treatment of IE. This is only partly true. Evidence There is excellent documentation that enterococcal endocarditis usually is best treated with a combination of two antibiotics. The primary reason for this is that most strains of Enterococcus faecalis are relatively resistant to antibiotics, but are killed synergistically by a combination of a penicillin and an aminoglycoside.58 This does not hold true, however, if the strain shows high level resistance to aminoglycosides (defined as resistance to 2000 micrograms/ml of streptomycin or 500 micrograms/ml of gentamicin). In the latter case, vancomycin should be substituted,59 except for strains that are vancomycin resistant.60 Ample documentation for the value of combination therapy has been published, based upon in vitro studies, and in vivo treatment of both animals and humans. Unfortunately, the frequency of high level resistance among enterococci has greatly increased over the past 15 years, making the choice of optimal therapy more difficult.65 Even when streptococci are fully sensitive to penicillin, combinations of a penicillin and an aminoglycoside or vancomycin act synergistically against them, so long as the strain is not vancomycin resistant (VRE) or has high level resistance to aminoglycosides.61 This has been convincingly demonstrated both in vitro and in experimental animals.62,63 The human correlate is found in the fact that combination therapy cures more than 97% of cases caused by penicillin-sensitive viridans streptococci within 2 weeks, whereas penicillin alone cures only 80–85% in the same interval, and requires 4 weeks to reach 97% cure or better. This fact has been well proven.54,64 What is the optimal duration of treatment for IE? Background Early experience established that endocarditis could not be cured by short courses of antibiotics (7–10 days) that would

824

Comments/references

have been adequate to cure other common infections such as pneumonia or gonorrhea. Trials of longer duration were more successful, eventually leading to the widely followed practice of treating IE for 6 weeks. This remains common practice today, despite the fact that more than half of all cases of IE could be reliably cured by 2–4 weeks of treatment.

Evidence Before 1950 it was reported that IE could not be cured with 10 days of treatment, even when the organisms were highly susceptible to penicillin and/or high doses were given.56,74,75 Subsequently, high cure rates were achieved by extending treatment to 4–6 weeks. For many years, 6 weeks of therapy was regarded as the standard duration for treatment of IE. In fact, this “rule” often led to overtreatment because 4 weeks would have been adequate for the majority of these cases.55 Because of number preferencing for even numbers, 3 and 5 week regimens have not been studied, even though intuitively it seems likely that these durations would work as well as 4 and 6 week regimens, respectively. Some cases of endocarditis can be cured with treatment for only 2 weeks. This is well supported by clinical experience for two important groups of patients: uncomplicated penicillin sensitive streptococcal native valve endocarditis,55,76 and intravenous drug addicts with right-sided S. aureus endocarditis.70 It should be noted that outpatient parenteral antibiotic therapy (OPAT) is appropriate for selected patients with IE.76–79 In most cases, these will be patients without serious complications who have responded promptly to standard therapy begun in hospital. When OPAT is employed for treatment of IE, the total duration of therapy should normally be the same as if the patient had been hospitalized throughout the course of treatment. The B ratings listed in the conclusions below could be improved by publication of larger numbers of cases or by randomized controlled studies.


Conclusions

Grading

Comments/references

Combination therapy should be used for IE caused by enterococci

Grade A

60, 66

A combination of at least two antibiotics (a lactam plus an aminoglycoside) should be used for IE caused by coagulase-negative staphylococci

Grade A

67, 68

A combination of three antibiotics (a lactam plus an aminoglycoside plus rifampin) should be used for IE caused by coagulase-negative staphylococci

Grade C Limited number of patients reported; no comparative trials67,68

An aminoglycoside should be added to a lactam for the first few days of therapy for IE caused by S. aureus

Grade C A definitive outcome study has not been done68,69

Combination therapy should be used for right-sided IE caused by S. aureus if a short course (2 week) regimen is used

Grade B Only one major study available70,71,72

Addition of an aminoglycoside for IE caused by penicillin sensitive streptococci is beneficial and cost effective if a short course (2 week) regimen is used

Grade A

Addition of an aminoglycoside for IE caused by penicillin sensitive streptococci is beneficial and cost effective if a standard (4 week) regimen is used

Grade C No modern cost effectiveness study has been done73

73

Conclusions

Grading

Comments/references

Penicillin sensitive IE can be cured in 2 weeks by combined penicillin plus aminoglycoside, or in 4 weeks by penicillin alone

Grade A

73, 80

Enterococcal endocarditis should be treated for at least 4 weeks

Grade B

59, 73, 81, 82

Most cases of HACEK endocarditis can be cured in 3–4 weeks (Haemophilus, Actinobacilllus, Cardiobacterium, Eikenella, Kingella spp)

Grade B Total number of reported cases is small73,83,84

Most cases of tricuspid valve S. aureus IE in intravenous drug users can be cured in 2 weeks

Grade B Only one major study done70

Results of treatment for IE in HIV infected patients with 200 CD4 lymphocytes/mm3 are similar to results in non-HIV patients

Grade B

HIV infected patients with 200 CD4 lymphocytes and IE have a worse prognosis, and therefore should not receive short-course (2 week) antibiotic therapy for IE

Grade C This has not been formally studied72

Most cases of left-sided native valve S. aureus IE can be cured in 4 weeks

Grade B Controlled study not done73

What are the main indications for surgical intervention during management of IE? Background The introduction of valve replacement, valve repair, and other surgical procedures has revolutionized the management of IE, being second in importance only to the advent of antibiotic therapy for IE. Many studies have indicated that surgical

72

intervention improves the prognosis of IE over medical therapy alone.85–87 The benefits of early rather than late surgical intervention have been appropriately emphasized.86–88 However, surgical placement of artificial cardiac valves is associated with high costs, significant morbidity, especially in the form of late complications, and some mortality. Furthermore, about two thirds of all patients with IE can be cured without any surgical intervention. Therefore, correct selection of the

825


subgroup of about one third of patients who will benefit from surgery becomes of critical importance.

What is the correct timing for valve replacement during management of IE? Background

Evidence Many hundreds of publications have reported on experience with surgery for IE, beginning in 196589 and continuing unabated today. The cumulative experience is based upon thousands of patients. However, this extensive experience does not include randomized studies of medical versus surgical therapy, primarily because selection bias (that is, choosing more seriously ill patients for surgery) is virtually impossible to overcome. Therefore, the conclusions which have emerged, although often well supported by case studies, can only be rated B in terms of evidence-based analyses.

826

In the past, it was often stated on empirical grounds that valve replacement surgery should be postponed until the patient had been cured by antibiotics. If the patient could not survive until cure, it was believed that surgery should be delayed as long as possible to allow suppression of the number of remaining bacteria to the lowest possible level to reduce the risk of relapse or infection of the new prosthetic valve. The available evidence does not support these widely held views. Evidence Actual experience showed that the frequency of relapse and/or infection of the prosthesis after surgery for IE is low,

Conclusions I: Strong indications for surgical intervention during IE

Grading

Comments/references

Heart failure unresponsive to medical therapy

Grade B

7, 8, 87, 90

Presence of a valve ring abscess

Grade B

7, 8, 90, 91, 92

Early prosthetic valve infection (onset within 60 days of surgery)

Grade B

7, 8, 90, 93

Prosthetic valve infection caused by S. aureus

Grade B

7, 8, 68, 71, 92, 93

Prosthetic valve infection caused by Gram-negative bacilli (not HACEK group)

Grade B

7, 8, 93

Endocarditis caused by filamentous fungi (not yeasts)

Grade B

94, 95

Prosthetic valve infection caused by yeasts

Grade B

93, 96

Development of a sinus of Valsalva aneurysm

Grade B

97

Occlusion of valves by very large vegetations

Grade B

7, 8, 98

Conclusions II: Relative (less strong) indications for surgical intervention during IE

Grading

Comments/references

Recurrent arterial emboli

Grade C

7, 8, 99, 100

Native left-sided valve infection caused by S. aureus

Grade C

7, 8, 68, 71

Apparent failure of medical therapy (persistent bacteremia, persistent fever, increase in size of vegetation during treatment)

Grade C

7, 8, 90

Native valve infection caused by Gram-negative bacilli (not HACEK)

Grade C

7, 8

Large-sized left-sided vegetations by echo (greater than 15–20 mm)

Grade C

7, 8

Native valve infection caused by yeasts

Grade C

95, 96, 101–103

Late onset prosthetic valve infection

Grade C

7, 8, 93

Development of cardiac conduction abnormality during IE, but no abscess identified by TEE

Grade C

7, 8, 91


whether or not antibiotics had been given for long periods before operation. For patients with a good indication for valve replacement early in the course of active endocarditis, many authors have strongly advocated early surgery, before antibiotics have cured the patient, to avoid deaths and complications that might occur during antibiotic treatment.85–88

Can IE be prevented? Background IE sometimes develops as a complication of bacteremias associated with medical and dental procedures, such as urinary catheterizations or tooth extractions. Although these cases represent only a small proportion – about 5% – of all cases of IE, much effort has been made to prevent them because IE carries high associated morbidity and mortality. Soon after antibiotics became available, various attempts were made to prevent bacteremias and/or IE by giving antibiotics before dental extractions or other procedures.104–108 Subsequently, the American Heart Association109 and many other groups110–112 have issued recommendations for prevention of IE by this means.

Evidence The evidence that bacteremias induced by medical and dental procedures can cause IE in patients with predisposing heart lesions consists of many uncontrolled case reports. There are sufficient numbers of these to support the conclusion that there is a real risk after tooth extractions and procedures involving an infected genitourinary tract.113–115 The evidence that lower-risk procedures such as gastrointestinal endoscopy, procedures on the uninfected urinary tract, and biopsies and other minor surgical procedures cause a significant number of cases of IE is sketchy.108,116–114 Prophylaxis of IE has been proven unequivocally to be effective in experimental animal models of endocarditis by giving antibiotics before injecting bacteria intravenously.125–132 However, there has been no definitive study to demonstrate efficacy of antibiotic prophylaxis for infective endocarditis in humans.116 One review of patients with prosthetic heart valves indicated that antibiotic prophylaxis before dental and urogenital procedures was effective,133 but this study was retrospective, unrandomized, and unblinded. Analysis of prospectively collected cases in the Netherlands indicated that prophylaxis was either ineffective or, at best, only

Conclusions

Grading

Comments/ references

If there is no indication for early surgery, complete a standard course of antibiotic treatment before valve replacement.

Grade B

No randomized studies available90

If there is an adequate indication for early surgery, proceed to valve replacement without regard to the duration of antibiotic treatment already given. Delay can result in avoidable complications or death.

Grade A

Many uncontrolled reports support early surgery85–88,90

Conclusions

Grading

Comments/ references

Bacteremias following tooth extraction or surgical procedures involving an infected genitourinary tract can cause endocarditis

Grade B

137

Bacteremias following gastrointestinal endoscopy or surgical procedures involving an uninfected genitourinary tract can cause endocarditis

Grade C

124, 137, 138

Prevention of IE by giving antibiotics before medical and dental procedures that cause bacteremia is an empiric practice which has been proven effective in animal models but not in humans

Grade A

116, 129

Attempted prevention of IE in selected high-risk groups undergoing high-risk procedures such as tooth extraction is probably effective

Grade C

133

Attempted prevention of IE in selected high-risk groups undergoing high-risk procedures such as tooth extraction is recommended

Grade B

110–112

Extensive practice of attempted prophylaxis for IE is probably not cost effective

Grade B

139, 140

827


marginally effective.134 Other analyses have indicated that even if prophylaxis is effective, it would probably not be cost effective as a general strategy.135,136 Despite all this uncertainty, most authorities continue to recommend selective use of prophylaxis for patients with higher risk cardiac lesions undergoing higher risk procedures.109,116

References 1.Durack DT, Beeson PB. Experimental bacterial endocarditis. I. Colonization of a sterile vegetation. Br J Exp Pathol 1972;53:44–9. 2.Angrist A, Oka M, Nakao K. Vegetative endocarditis. Pathol Ann 1967;2:155–212. 3.Blanchard DG, Ross RS, Dittrich HC. Nonbacterial thrombotic endocarditis. Chest 1993;102:954–6. 4.Rodbard S, Yamamoto C. Effect of stream velocity on bacterial deposition and growth. Cardiovasc Res 1969;3:68–74. 5.Grant RT, Wood JE, Jr, Jones TD. Heart valve irregularities in relation to subacute bacterial endocarditis. Am Heart J 1928; 14:247–61. 6.Moreillon P, Que YA, Bayer AS. Pathogenesis of streptococcal and staphylococcal endocarditis. Infect Dis Clin North Am 2002;16:297–318. 7.Durack DT. Infective and noninfective endocarditis. In: Hurst JW, ed. The heart, arteries and veins, 7th edn. New York: McGraw-Hill, 1990. 8.Scheld WM, Sande MA. Endocarditis and intravascular infections. In: Mandell GL, Douglas RG, Jr, Dolin R, eds. Principles and practice of infectious diseases, 4th edn. New York: Churchill Livingstone, 1995. 9.Cabell CH, Abruytn E. Progress toward a global understanding of infective endocarditis: early lessons from the International Collaboration on Endocarditis investigation. Infect Dis Clin North Am 2002;16:255–72. 10.Houpikian P, Raoult D. Diagnostic methods: current best practices and guidelines for identification of difficult-toculture pathogens in endocarditis. Infect Dis Clin North Am 2002;16:377–92. 11.Schwartzman WA, Marchevsky A, Meyer RD. Epithelioid angiomatosis or cat scratch disease with splenic and hepatic abnormalities in AIDS: case report and review of the literature. Scand J Infect Dis 1990;22:121–33. 12.Towns ML, Reller LB. Diagnostic methods: current best practices and guidelines for isolation of bacteria and fungi. Infect Dis Clin North Am 2002;16:363–76. 13.Durack DT, Bright DK, Lukes AS, Duke Endocarditis Service. New criteria for diagnosis of infective endocarditis: utilization of specific echocardiographic findings. Am J Med 1994;96:200–9. 14.Washington JA, II. The role of the microbiology laboratory in the diagnosis and antimicrobial treatment of infective endocarditis. Mayo Clin Proc 1982;57:22–32. 15.Washington JA. The microbiological diagnosis of infective endocarditis. J Antimicrob Chemother 1987;20:29–39. 16.Maki DG, Agger WA. Enterococcal bacteremia: clinical features, the risk of endocarditis, and management. Medicine 1988;67:248–69.

828

17.Gullberg RM, Homann SR, Phair JP. Enterococcal bacteremia: analysis of 75 episodes. Rev Infect Dis 1989;11:74–85. 18.Beeson PB, Brannon ES, Warren JV. Observations of the sites of removal of bacteria from the blood in patients with bacterial endocarditis. J Exp Med 1945;81:9–23. 19.Weinstein M, Reller L, Murphy J, Lichtenstein K. Clinical significance of positive blood cultures: a comprehensive analysis of 500 episodes of bacteremia and fungemia in adults. I. Laboratory and epidemiologic observations. Rev Infect Dis 1983;5:35–53. 20.Towns M, Quartey S, Weinstein M et al. Clinical significance of positive blood cultures: a prospective, multicenter investigation. ASM 1993; Abstract No. C232: Abstract. 21.Weinstein M, Murphy J, Reller L, Lichtenstein K. Clinical significance of positive blood cultures: A comprehensive analysis of 500 episodes of bacteremia and fungemia in adults. II. Clinical observations, with special reference to factors influencing prognosis. Rev Infect Dis 1983;5:54–70. 22.Belli J, Waisbren BA. The number of blood cultures necessary to diagnose most cases of bacterial endocarditis. Am J Med Sci 1956;232:284–8. 23.Mallen MS, Hube EL, Brenes M. Comparative study of blood cultures made from artery, vein, and bone marrow in patients with subacute bacterial endocarditis. Am Heart J 1946;XX: 692–5. 24.Geraci JE, Wilson WR. Endocarditis due to gram-negative bacteria: report of 56 cases. Mayo Clin Proc 1982;57:145–8. 25.Chen YC, Chang SC, Luh KT, Hsieh WC. Actinobacillus actinomycetemcomitans endocarditis: a report of four cases and review of the literature. QJM 1992;81:871–8. 26.Drancourt M, Birtles R, Chaumentin G, Vandenesch F, Etienne J, Raoult D. New serotype of Bartonella henselae in endocarditis and cat-scratch disease. Lancet 1996;347: 441–3. 27.Doern GV, Davaro R, George M, Campognone P. Lack of requirement for prolonged incubation of Septi-Chek blood culture bottles in patients with bacteremia due to fastidious bacteria. Diagn Microbiol Infect Dis 1996;24:141–3. 28.Murray M, Moosnick F. Arterial vs venous blood cultures. J Lab Clin Med 1940;26:382–7. 29.Gilbert BW, Haney RS, Crawford F et al. Two-dimensional echocardiographic assessment of vegetative endocarditis. Circulation 1977;55:346–53. 30.Sachdev M, Peterson G, Jollis JG. Diagnostic methods: imaging techniques for diagnosis of endocarditis. Infect Dis Clin North Am 2002;16:319–38. 31.Stewart JA, Silimperi D, Harris P, Wise NK, Fraker TD, Jr, Kisslo JA. Echocardiographic documentation of vegetative lesions in infective endocarditis: clinical implications. Circulation 1980;61:374–80. 32.Irani WN, Grayburn PA, Afridi I. A negative transthoracic echocardiogram obviates the need for transesophageal echocardiography in patients with suspected native valve active infective endocarditis. Am J Cardiol 1996;78: 101–3. 33.Jaffe WM, Morgan DE, Pearlman AS, Otto CM. Infective endocarditis, 1983–1988: echocardiographic findings and factors influencing morbidity and mortality. J Am Coll Cardiol 1990;15:1227–33.


34.Martin RP. The diagnostic and prognostic role of cardiovascular ultrasound in endocarditis: bigger is not better. J Am Coll Cardiol 1990;15:1234–7. 35.Taams MA, Gussenhoven EJ, Bos E et al. Enhanced morphological diagnosis in infective endocarditis by transesophageal echocardiography. Br Heart J 1990;63:109–13. 36.Rohmann S, Erbel R, Gorge G et al. Clinical relevance of vegetation localization by transoesophageal echocardiography in infective endocarditis. Eur Heart J 1992;12:446–52. 37.Shapiro SM, Bayer AS. Transesophageal and Doppler echocardiography in the diagnosis and management of infective endocarditis. Chest 1991;100:1125–30. 38.Pedersen WR, Walker M, Olson JD et al. Value of transesophageal echocardiography as an adjunct to transthoracic echocardiography in evaluation of native and prosthetic valve endocarditis. Chest 1991;100:351–6. 39.Morguet AJ, Werner GS, Andreas S, Kreuzer H. Diagnostic value of transesophageal compared with transthoracic echocardiography in suspected prosthetic valve endocarditis. Herz 1995;20:390–8. 40.Anders K, Foley K, Stern WE, Brown WJ. Intracranial sparganosis: an uncommon infection. Case report. J Neurosurg 1984;60:1282–6. 41.Daniel WG, Mugge A, Martin RP et al. Improvement in the diagnosis of abscesses associated with endocarditis by transesophageal echocardiography. N Engl J Med 1991;324: 795–800. 42.Lindner JR, Case RA, Dent JM et al. Diagnostic value of echocardiography in suspected endocarditis. An evaluation based on the pretest probability of disease. Circulation 1996;93:730–6. 43.Hickey AJ, Wolfers J. False positive diagnosis of vegetations on a myxomatous mitral valve using two-dimensional echocardiography. Aust NZ J Med 1982;12:540–2. 44.Mintz GS, Kotler MN. Clinical value and limitations of echocardiography. Its use in the study of patients with infectious endocarditis. Arch Intern Med 1980;140:1022–7. 45.Sokil AB. Cardiac imaging in infective endocarditis. In: Kaye D, ed. Infective endocarditis. 2nd edn. New York: Raven Press, 1992. 46.von Reyn CF, Levy BS, Arbeit RD, Friedland G, Crumpacker CS. Infective endocarditis: an analysis based on strict case definitions. Ann Intern Med 1981;94:505–17. 47.Bayer AS, Ward JI, Ginzton LE, Shapiro SM. Evaluation of new clinical criteria for the diagnosis of infective endocarditis. Am J Med 1994;96:211–19. 48.Del Pont JM, De Cicco LT, Vartalitis C et al. Infective endocarditis in children: clinical analyses and evaluation of two diagnostic criteria. Pediatr Infect Dis 1995;14:1079–86. 49.Fournier PE, Casalta JP, Habib G, Messana T, Raoult D. Modification of the diagnostic criteria proposed by the Duke Endocarditis Service to permit improved diagnosis of Q fever endocarditis. Am J Med 1996;100:629–33. 50.Cecchi E, Parrini I, Chinaglia A et al. New diagnostic criteria for infective endocarditis. A study of sensitivity and specificity. Eur Heart J 1997;18:1149–56. 51.Olaison L, Hogevik H. Comparison of the von Reyn and Duke criteria for the diagnosis of infective endocarditis: a critical analysis of 161 episodes. Scand J Infect Dis 1996;28: 399–406.

52.Hoen B, Beguinot I, Rabaud C et al. The Duke criteria for diagnosing infective endocarditis are specific: analysis of 100 patients with acute fever or fever of unknown origin. Clin Infect Dis 1996;23:298–302. 53.Dodds GA, Sexton DJ, Durack DT, Bashore TM, Corey GR, Kisslo J. Negative predictive value of the Duke criteria for infective endocarditis. Am J Cardiol 1996;77:403–7. 54.Galbreath WR, Hull E. Sulfonamide therapy of bacterial endocarditis: results in 42 cases. Ann Intern Med 1943;18:201–3. 55.Durack DT. Review of early experience in treatment of bacterial endocarditis, 1940–1955. In: Bisno AL, ed. Treatment of infective endocarditis. New York: Grune & Stratton, 1981. 56.Speller DCE. Pseudomonas cepacia endocarditis treated with co-trimazole and kanamycin. Br Heart J 1972;35:47–8. 57.Street AC, Durack DT. Experience with trimethoprimsulfamethoxazole in treatment of infective endocarditis. Rev Infect Dis 1988;10:915–21. 58.Wilson WR, Wilkowske CJ, Wright AJ, Sande MA, Geraci JE. Treatment of streptomycin-susceptible and streptomycin-resistant enterococcal endocarditis. Ann Intern Med 1984; 100:816–23. 59.Watanakunakorn C, Bakie C. Synergism of vancomycin– gentamicin and vancomycin–streptomycin against enterococci. Antimicrob Agents Chemother 1973;4:120–4. 60.Caron F, Lemeland JF, Humbert G, Klare I, Gutmann L. Triple combination penicillin–vancomycin–gentamicin for experimental endocarditis caused by a highly penicillin- and glycopeptide-resistant isolate of Enterococcus faecium. J Infect Dis 1993;168:681–6. 61.Watanakunakorn C, Glotzbecker C. Synergism with aminoglycosides of penicillin, ampicillin and vancomycin against nonenterococcal group-D streptococci and viridans streptococci. J Med Microbiol 1976;10:133–8. 62.Sande MA, Irvin RG. Penicillin-aminoglycoside synergy in experimental Streptococcus viridans endocarditis. J Infect Dis 1974;129:572–6. 63.Fantin B, Carbon C. In vivo antibiotic synergism: contribution of animal models. Antimicrob Agents Chemother 1992;36: 907–12. 64.Geraci JE. The antibiotic therapy of infective endocarditis: therapeutic data on 172 patient seen from 1951 through 1957: additional observations on short-term therapy (two weeks) for penicillin-sensitive streptococcal endocarditis. Med Clin North Am 1958;42:1101–48. 65.Hoen B. Special issues in the management of infective endocarditis caused by gram-positive cocci. Infect Dis Clin N Am 2002;16:437–52. 66.Rice LB, Calderwood SB, Eliopoulos GM, Farber BF, Karchmer AW. Enterococcal endocarditis: a comparison of prosthetic and native valve disease. Rev Infect Dis 1991; 13:1–7. 67.Karchmer AW, Archer GL, Dismukes WE. Staphylococcus epidermidis causing prosthetic valve endocarditis: microbiologic and clinical observations as guides to therapy. Ann Intern Med 1983;98:447–55. 68.Karchmer AW. Staphylococcal endocarditis. In: Kaye D, ed. Infective endocarditis, 2nd edn. New York: Raven Press, 1992. 69.Sande MA, Courtney KB. Nafcillin-gentamicin synergism in experimental staphylococcal endocarditis. J Lab Clin Med 1976;88:118–24.

829


70.Chambers HF, Miller RT, Newman MD. Right-sided Staphylococcus aureus endocarditis in intravenous drug abusers: two-week combination therapy. Ann Intern Med 1988;109:619–24. 71.Petti CA, Fowler VG. Staphylococcus aureus bacteremia and endocarditis. Infect Dis Clin North Am 2002;16: 419–36. 72.Miro JM, del Rio A, Mestres CA. Infective endocarditis in intravenous drug abusers and HIV-1 infected patients. Infect Dis Clin North Am 2002;16:273–96. 73.Wilson WR, Karchmer A, Dajani A et al. Antibiotic treatment of adults with infective endocarditis due to viridans streptococci, enterococci, staphylococci and HACEK microorganisms. JAMA 1995;274:1706–13. 74.King FH, Schneierson SS, Sussman ML, Janowitz HD, Stollerman GH. Prolonged moderate dose therapy versus intensive short term therapy with penicillin and caronamide in subacute bacterial endocarditis. J Mt Sinai Hosp 1949; 16:35–46. 75.Bloomfield AL, Armstrong CD, Kirby WMM. The treatment of subacute bacterial endocarditis with penicillin. J Clin Invest 1945;24:251–67. 76.Kwon-Chung KJ, Hill WB. Studies on the pink, adeninedeficient strains of Candida albicans. I. Cultural and morphological characteristics. Sabouraudia 1970;8:48–59. 77.Francioli P, Etienne J, Hoigne R, Thys J, Gerber A. Treatment of streptococcal endocarditis with a single daily dose of ceftriaxone sodium for 4 weeks. Efficacy and outpatient treatment feasibility. JAMA 1992;267:264–7. 78.Francioli P, Ruch W, Stamboulian D, International Infective Endocarditis Study Group. Treatment of streptococcal endocarditis with a single daily dose of ceftriaxone and netilmicin for 14 days: a prospective multicenter study. Clin Infect Dis 1995;21:1406–10. 79.Stamboulian D, Bonvehi P, Arevalo C et al. Antibiotic management of outpatients with endocarditis due to penicillin-susceptible streptococci. Rev Infect Dis 1991;13:S160–S3. 80.Wilson WR, Geraci JE, Wilkowske CJ, Washington JA. Shortterm intramuscular therapy with procaine penicillin plus streptomycin for infective endocarditis due to viridans streptococci. Circulation 1978;57:1158–61. 81.Geraci JE, Martin WJ. Antibiotic therapy of bacterial endocarditis. VI. Subacute enterococcal endocarditis: clinical, pathologic and therapeutic consideration of 33 cases. Circulation 1954;10:173–94. 82.Moellering RC, Jr, Wennersten C, Weinstein AJ. Penicillin–tobramycin synergism against enterococci: a comparison with penicillin and gentamicin. Antimicrob Agents Chemother 1973;3:526–9. 83.Shorrock PJ, Lambert PA, Aitchison EJ, Smith EG, Farrell ID, Gutschik E. Serological response in Enterococcus faecalis endocarditis determined by enzyme-linked immunosorbent assay. J Clin Microbiol 1990;28:195–200. 84.Bieger RC, Brewer NS, Washington JA, II. Haemophilus aphrophilus: a microbiologic and clinical review and report of 42 cases. Medicine 1978;57:345–55. 85.Bogers AJJC, van Vreeswijk H, Verbaan CJ et al. Early surgery for active infective endocarditis improves early and late results. Thorac Cardiovasc Surg 1991;39:284–7.

830

86.Aranki SF, Adams DH, Rizzo RJ et al. Determinants of early mortality and late survival in mitral valve endocarditis. Circulation 1995;92:143–9. 87.Middlemost S, Wisenbaugh T, Meyerowitz C et al. A case for early surgery in native left-sided endocarditis complicated by heart failure: results in 203 patients. J Am Coll Cardiol 1991;18:663–7. 88.Jubair KA, Al Fagih MR, Ashmeg A, Belhaj M, Sawyer W. Cardiac operations during active endocarditis. J Thorac Cardiovasc Surg 1992;104:487–90. 89.Wallace AG, Young G, Jr, Osterhout S. Treatment of acute bacterial endocarditis by valve excision and replacement. Circulation 1965;31:450–3. 90.Olaison L, Pettersson G. Current best practices and guidelines: indications for surgical intervention in infective endocarditis. Infect Dis Clin North Am 2002;16:453–76. 91.DiNubile MJ, Calderwood SB, Steinhaus DM, Karchmer AW. Cardiac conduction abnormalities complicating native valve active endocarditis. Am J Cardiol 1986;58:1213–17. 92.Tucker KJ, Johnson JA, Ong T, Mullen WL, Mailhot J. Medical management of prosthetic aortic valve endocarditis and aortic root abscess. Am Heart J 1993;125:1195–7. 93.Karchmer AW, Longworth DL. Infections of intracardiac devices. Infect Dis Clin North Am 2002;16:477–506. 94.Woods GL, Wood P, Shaw BW, Jr. Aspergillus endocarditis in patients without prior cardiovascular surgery: report of a case in a liver transplant recipient and review. Rev Infect Dis 1989;II:263–72. 95.Fowler VG, Durack DT. Infective endocarditis. Curr Opin Cardiol 1994;9:389–400. 96.Guzman F, Cartmill I, Holden MP, Freeman R. Candida endocarditis: report of four cases. Int J Cardiol 1987;16: 131–6. 97.Scully RE, Mark EJ, McNeely WF, McNeely BU. Case records of the Massachussetts General Hospital. N Engl J Med 1996;334:105–11. 98.Khan SS, Gray RJ. Valvular emergencies. Cardiol Clin 1991; 9:689–709. 99.Steckelberg JM, Murphy JG, Ballard D et al. Emboli in infective endocarditis: the prognostic value of echocardiography. Ann Intern Med 1991;114:635–40. 100.Sexton DJ, Spelman D. Current best practices and guidelines: assessment and management of complications of infective endocarditis. Infect Dis Clin North Am 2002;16:507–22. 101.Kawamoto T, Nakano S, Matsuda H, Hirose H, Kawashima Y. Candida endocarditis with saddle embolism: a successful surgical intervention (abstract). Ann Thorac Surg 1989; 48:723–4. 102.Tanka M, Toshio A, Hosokawa S, Suenaga Y, Hikosaka H. Tricuspid valve Candida endocarditis cured by valve-sparing debridement. Ann Thorac Surg 1989;48:857–8. 103.Isalska BJ, Stanbridge TN. Fluconazole in the treatment of candidal prosthetic valve endocarditis. BMJ 1988;297:178–9. 104.Rhoads PS, Schram WR, Adair D. Bacteremia following tooth extraction: prevention with penicillin and N U 445. J Am Dent Assoc 1950;41:55–61. 105.Pressman RS, Bender IB. Antibiotic treatment of the gingival sulcus in the prevention of bacteremia. Antibiotics Annual 1954;92–104.


106.Northrop PM, Crowley MC. Further studies on the effect of the prophylactic use of sulfathiazole and sulfamerazine on bacteremia following extraction of teeth. J Oral Surg 1944;2:134. 107.Budnitz E, Nizel A, Berg L. Prophylactic use of sulfapyridine in patients susceptible to subacute bacterial endocarditis following dental surgical procedures. J Am Dent Assoc 1942;29:346. 108.Camara DS, Gruber M, Barde CJ, Montes M, Caruana JA, Jr, Chung RS. Transient bacteremia following endoscopic injection sclerotherapy of esophageal varices. Arch Intern Med 1983;143:1350–2. 109.Dajani AS, Taubert KA, Wilson WR et al. Prevention of bacterial endocarditis. Recommendations by the American Heart Association. Circulation 1997;96:358–66. 110.Delaye J, Etienne J, Feruglio A et al. Prophylaxis of infective endocarditis for dental procedures. Report of a working party of the European Society of Cardiology. Eur Heart J 1985;6:826–8. 111.Michel MF, Thompson J, Boering G, Hess J, Van Putten PL. Revision of the guidelines of the Netherlands Heart Foundation for the prevention of endocarditis. Geneesmiddelen-bull 1986;20:53–6. 112.Shanson DC. Antibiotic prophylaxis of infective endocarditis in the United Kingdom and Europe. J Antimicrob Chemother 1987;20:119–31. 113.Meneely JK. Bacterial endocarditis following urethral manipulation. N Engl J Med 1948;239:708–9. 114.Slade N. Bacteriaemia and septicaemia after urological operations. Proc R Soc Med 1958;51:331–4. 115.Sullivan NM, Sutter VL, Mims MM, Marsh VH, Finegold SM. Clinical aspects of bacteremia after manipulation of the genitourinary tract. J Infect Dis 1973;127:49–55. 116.Durack D. Prevention of infective endocarditis. N Engl J Med 1995;332:38–44. 117.Shorvon PJ, Eykyn SJ, Cotton PB. Gastrointestinal instrumentation, bacteraemia, and endocarditis. Gut 1983;24: 1078–93. 118.Edson RS, Van Scoy RE, Leary FJ. Gram-negative bacteremia after transrectal needle biopsy of the prostate. Mayo Clin Proc 1980;55:489–91. 119.Livengood CH, III, Land MR, Addison WA. Endometrial biopsy, bacteremia, and endocarditis risk. Obstet Gynecol 1985;65:678–81. 120.Mellow MH, Lewis RJ. Endoscopy-related bacteremia. Incidence of positive blood cultures after endoscopy of upper gastrointestinal tract. Arch Intern Med 1976;136:667–9. 121.Yin TP, Dellipiani AW. Bacterial endocarditis after Hurst bougienage in a patient with a benign oesophageal stricture. Endoscopy 1983;15:27–8. 122.Giglio JA, Rowland RW, Dalton HP, Laskin DM. Suture removal-induced bacteremia: a possible endocarditis risk. J Am Dent Assoc 1992;123:65–70. 123.Ho H, Zuckerman MJ, Wassem C. A prospective controlled study of the risk of bacteremia in emergency sclerotherapy of esophageal varices. Gastroenterology 1991;101:1642–8.

124.Low DE, Shoenut JP, Kennedy JK et al. Prospective assessment of risk of bacteremia with colonoscopy and polypectomy. Dig Dis Sci 1987;32:1239–43. 125.Glauser MP, Bernard JP, Morceillon P, Francioli P. Successful single-dose amoxicillin prophylaxis against experimental streptococcal endocarditis: evidence for two mechanisms of protection. J Infect Dis 1983;147:568–75. 126.Bernard J, Francioli P, Glauser MP. Vancomycin prophylaxis of experimental Streptococcus sanguis; inhibition of bacterial adherence rather than bacterial killing. J Clin Invest 1981;68:1113–16. 127.Moreillon P, Francioli P, Overholser P, Meylan P, Glauser MP. Mechanisms of successful amoxicillin prophylaxis of experimental endocarditis due to Streptococcus intermedius. J Infect Dis 1986;154:801–7. 128.Malinverni R, Overholser CD, Bille J, Glauser MP. Antibiotic prophylaxis of experimental endocarditis after dental extractions. Lab Invest 1988;77:182–7. 129.Glauser MP, Francioli P. Relevance of animal models to the prophylaxis of infective endocarditis. J Antimicrob Chemother 1987;20(Suppl. A):87–93. 130.Durack DT, Petersdorf RG, Beeson PB. Penicillin prophylaxis of experimental S. viridans endocarditis. Trans Assoc Am Phys 1972;85:222–30. 131.Durack DT, Petersdorf RG. Chemotherapy of experimental streptococcal endocarditis. I. Comparison of commonly prophylactic regimens. J Clin Invest 1973;52:592–8. 132.Pelletier LL, Durack DT, Petersdorf RG. Chemotherapy of experimental streptococcal endocarditis. IV. Further observations on antimicrobial prophylaxis. J Clin Invest 1975; 56:319–30. 133.Horstkotte D, Friedrichs W, Pippert H, Bircks W, Loogen F. Benefits of endocarditis prevention in patients with prosthetic heart valves. [German]. Z Kardiol 1986;75:8–11. 134.Van Der Meer JTM, Van Wijk W, Thompson J, Vandenbroucke JP, Valkenburg HA, Michel MF. Efficacy of antibiotic prophylaxis for prevention of native-valve endocarditis. Lancet 1992;339:135–40. 135.Patton JP. Infective endocarditis: economic considerations. In: Kaye D, ed. Infective endocarditis, 2nd edn. New York: Raven Press, 1992. 136.Imperiale TF, Horwitz RI. Does prophylaxis prevent postdental infective endocarditis? A controlled evaluation of protective efficacy. Am J Med 1990;88:131–6. 137.Everett ED, Hirschmann JV. Transient bacteremia and endocarditis prophylaxis. A review. Medicine 1977;56: 61–77. 138.Biorn CL, Browning WH, Thompson L. Transient bacteremia immediately following transurethral prostatic resection. J Urol 1950;63:155–61. 139.Clemens JD, Ransohoff DF. A quantitative assessment of predental antibiotic prophylaxis for patients with mitral-valve prolapse. J Chron Dis 1984;37:531–44. 140.Bor DH, Himmelstein DU. Endocarditis prophylaxis for patients with mitral valve prolapse: a quantitative analysis. Am J Med 1984;76:711–17.

831

58

Antithrombotic therapy after heart valve replacement Alexander GG Turpie, Walter Ageno

Introduction Despite improvements in prosthetic materials and valve design, thromboembolism remains a serious complication in patients following heart valve replacement. It is generally agreed that lifelong oral anticoagulants are indicated in patients with mechanical prosthetic valves and in patients with tissue valves, if they have atrial fibrillation or a history of thromboembolism.1–3 In the absence of antithrombotic therapy, systemic embolism and stroke have been reported in between 5% and 50% of patients, depending upon the valve site, the type of valve replacement and the presence of comorbid conditions.2,3 With the use of anticoagulants, the rate of systemic embolism has been reduced to 1–3% per year.4 Antithrombotic therapy, however, carries an important risk of bleeding, which is related to the level of anticoagulation used.5 Studies of long-term oral anticoagulant therapy for deep vein thrombosis have shown that a less intense regimen (INR 2·0–3·0) is as effective but safer than the more intense regimen (INR 3·0–4·5) that was standard until recently.6–8 Subsequently, studies in patients following either bioprosthetic or mechanical heart valve replacement have shown reduced bleeding with a lower intensity of anticoagulants without loss of efficacy,9–11 and based on these studies there has been marked improvement in the safety of the long-term anticoagulant regimens used following heart valve replacement.

Bioprosthetic heart valves The risk of thromboembolism is less with uncomplicated bioprosthetic valves than with mechanical valves.12–15 Oral anticoagulants, including warfarin, have been shown to be effective and safe when used at a targeted INR of 2·0–3·0 in such patients based on the results of one prospective clinical trial.9 This study compared two intensities of anticoagulation to determine the safety and efficacy of a less intense anticoagulant regimen in patients following tissue valve replacement. One hundred and eight patients were randomized to standard anticoagulant control (INR 3·0–4·5), and 102 patients to a less intensive regimen (INR 2·0–2·5). Treatment was continued for 3 months. In this study there 832

was no difference in the frequency of major systemic emboli (1·8% v 1·8%) between the two treatment groups, but there were significantly fewer major hemorrhagic complications (0·0% v 4·5%; P 0·034) and total hemorrhagic complications (5·4% v 14·6%; P 0·042) in the low intensity group (INR 2·0–3·0) compared to the high intensity group, respectively. This level of anticoagulation (INR 2·0–3·0) is now recommended by the American College of Chest Physicians (ACCP) for patients with tissue valve replacement.4 Grade A1a The risk of thromboembolism is limited mainly to the first 3 months postoperatively in uncomplicated patients with tissue valves, but is present indefinitely in patients with atrial fibrillation.16 A low ejection fraction, an enlarged left atrium, previous history of venous thromboembolism, and the presence of a pacemaker also increase the risk of thromboembolic complications.17,18 Consequently, in uncomplicated patients with mitral bioprosthetic valves, anticoagulant therapy is recommended for 3 months while long-term therapy is indicated in patients with atrial fibrillation, those with an atrial thrombus detected at echocardiography, and those who develop a systemic embolus.4 Grade B3 Patients with uncomplicated bioprosthetic valves in the aortic position are at very low risk of systemic embolism and some authorities therefore suggest they do not require anticoagulant therapy, although this recommendation remains controversial.4,19 Long-term treatment with aspirin 80 mg/day following 3 months of oral anticoagulant therapy is likely to be beneficial to prevent subsequent thromboembolic events in patients with uncomplicated bioprosthetic valves.4 Grade B4 The current recommendations by the ACCP for patients with tissue valves are shown below.

Mechanical prosthetic heart valves Patients with mechanical heart valve prostheses require lifelong anticoagulation therapy. The optimal level of anticoagulation in patients with mechanical heart valve replacements has been placed on a scientific footing based on the results of recent studies. Randomized trials have shown that oral anticoagulants are effective in reducing the risk of systemic embolism in patients with mechanical prosthetic valves

Antithrombotic therapy after heart valve replacement

Antithrombotic therapy in bioprosthetic heart valve replacement: recommendations INR

Duration

Grade of recommendation

Mitral

2·0–3·0

3 mth

Grade B3

Aortic

2·0–3·0

3 mth

Grade B3

Atrial fibrillation

2·0–3·0

Long-term

Grade B3

Long-term (duration uncertain)

Grade B3

Permanent pacemaker 2·0–3·0

Optional

Grade B4

Systemic embolism

2·0–3·0

3–12 mth

Grade B4

Normal sinus rhythm

Long-term aspirin (80 mg/day)

Left atrial 2·0–3·0 thrombosis

Grade B4

From Stein et al.4

when given at lower intensity than has been used in the past. The 2001 guidelines of the ACCP4 recommend two intensity regimens of long-term oral anticoagulant treatment according to the site of the mechanical prosthesis and the presence of concomitant risk factors. A lower INR range between 2·0 and 3·0 is now recommended for patients with a bileaflet valve (St Jude Medical or Carbomedics) or a tilting disc valve (Medtronic–Hall) in the aortic position who are in normal sinus rhythm and have a left atrium of normal size. These newer recommendations, which are levels of evidence Grade A2 for the St Jude Medical valves and Grade B2 for Carbomedics and Medtronic–Hall, are based on the results of long-term follow up studies.20–23 In particular, a study from France20 has confirmed the efficacy of a less intense level of anticoagulation following mechanical heart valve replacement. In this study 433 patients with mechanical prostheses were randomized to anticoagulant therapy monitored to achieve an INR of 2·0–3·0 or 3·0–4·5 and followed for 2·2 years. Thromboembolic outcome events, either clinical events or asymptomatic CNS abnormalities proven on CT scan, occurred in 10 of 185 (5·3%) patients in the low intensity group and 9 of 192 (4·7%) patients in the high intensity group (P 0·78). Importantly, there was a statistically significant difference in the rate of bleeding complications between the two groups. Bleeding events occurred in 34 patients (18·1%) in the low intensity group compared with 56 patients (29·2%) in the high intensity group (P 0·01). The majority of patients in this trial, however, had aortic valve replacements and were in sinus rhythm, which limits the generalizability of the results.

An INR range between 2·5 and 3·5 is still recommended for mechanical valves in the mitral position.4 This recommendation was based on two prospective studies that demonstrated that anticoagulant therapy maintained within this target INR range was as effective as a more intense level of anticoagulation, but with less bleeding. In the first study10 there was no difference in the frequency of major embolic events in the patients treated with a high intensity regimen (INR 9·0) compared with patients treated with a less intense (INR 2·65) anticoagulant regimen (4·0 v 3·7 embolic episodes per 100 patient years, respectively). However, there was significantly less bleeding in the less intense group (6·2 v 12·1 hemorrhagic episodes per 100 patient years; P 0·02). The second study11 compared low intensity (INR 2·0–2·99) with high intensity (INR 3·0–4·5) oral anticoagulants in patients with mechanical valves, all of whom were treated with aspirin (330 mg twice per day) in combination with dipyridamole (75 mg per day). In this study one transient ischemic attack occurred in the low intensity group and two in the high intensity group. There were significantly fewer bleeding events in the low intensity group, in which three episodes occurred compared with 12 in the high intensity group (P 0·02). Although these studies form the basis for the recommendations for a less intense level of anticoagulation, they have a number of limitations. In the first study a very high intensity anticoagulant regimen was compared with a moderately high intensity anticoagulant regimen. The mean daily dose of warfarin in the high intensity group was 8·5 mg and in the low intensity group the mean daily dose of 5·9 mg was similar to the average daily dose of 5·4 mg used in the high intensity group in three venous thrombosis studies,6–8 and in the randomized study in patients with tissue valves.9 This suggests that the low intensity group in the first mechanical valve study10 may have been equivalent to the standard intensity group in the venous thromboembolism studies.8 The second study11 was small, and therefore the claim that the two regimens were identical in efficacy is questionable. This latter study does, however, confirm the marked difference in bleeding between high intensity and low intensity anticoagulant regimens. The ACCP recommendations for mechanical valves are shown below. The ACCP recommendation of an INR target of 2·5–3·5 is lower than that reported in a study conducted in Europe which has recommended a target range of 3·0–4·0.24 However, the European recommendation is based on retrospective data and largely on events that occurred in patients with older caged ball valve prostheses. Thus recommendations based on these data are unlikely to be applicable for use in patients with the modern bileaflet and tilting disc valves that are currently in use. Of interest, two recent studies have reported a high sensitivity to warfarin in the immediate postoperative phase during oral anticoagulation induction and suggested the 833


Antithrombotic therapy in mechanical heart valve replacement: recommendations INR

Grade of recommendation

Uncomplicated bileaflet aortic

2·0–3·0

Grade A1a

Uncomplicated tilting disc aortic

2·0–3·0

Grade B3

Bileaflet aortic and atrial fibrillation

2·5–3·5 or 2·0–3·0 aspirin 80 mg to 100 mg/day

Grade B3/4

Uncomplicated bileaflet mitral

2·5–3·5

Grade B3

Uncomplicated tilting disc mitral

2·5–3·5

Grade B2

Additional risk factors

2·5–3·5 aspirin 80 mg to 100 mg/day

Grade B2

Systemic embolism


Grade B2

Caged ball or caged disc valve


Grade A/C

From Stein et al.4

need for lower starting doses of warfarin to regularly achieve the therapeutic range (that is, 2·5–3·0 mg instead of 5·0 mg) in most patients.25,26

Combination antithrombotic therapy A major limitation to the current approach used to treat highrisk patients with prosthetic heart valves is that systemic

embolism, which may result in disabling stroke, still occurs at a rate of approximately 2–3% per year, despite the use of anticoagulants.4 The addition of antiplatelet agents to oral anticoagulants has been advocated as an improved approach to the treatment of patients with mechanical valves, or highrisk patients with tissue valves, to reduce further the risk of major systemic embolism. In an early study27 the combination of dipyridamole and oral anticoagulants significantly reduced mortality in patients with early models of the Starr–Edwards prosthesis compared with anticoagulants alone. A subsequent study from the Mayo Clinic28 reported that the addition of dipyridamole to oral anticoagulants significantly reduced the risk of thromboembolic events in patients with mechanical heart valve prostheses. A recent meta-analysis of the dipyridamole studies (Table 58.1) has confirmed improved outcomes with combined therapy compared with anticoagulants alone.29 The routine use of dipyridamole in combination with oral anticoagulants is, however, not widely accepted for the treatment of patients with mechanical valves or high-risk patients with tissue valves, because of the frequency of adverse effects, including intractable headache, dizziness, nausea, flushing, and syncope. The combination of aspirin and oral anticoagulants has also been used in the treatment of patients with heart valve replacement with a significant reduction in embolic complications, but with an increased risk of bleeding complications.30 In the early studies reported to date, aspirin was used in high doses (approximately 1 g/day), and in most cases the bleeding with the combination of high-dose aspirin and high-dose oral anticoagulants was gastrointestinal.31 There is good evidence that gastrointestinal irritation and hemorrhage is dose-dependent over a range of 100–1000 mg of aspirin per day and that the antithrombotic effects of aspirin are independent of the dose over this range. One completed study32 compared low-dose aspirin combined with warfarin in the treatment of patients with mechanical heart valve replacements to determine whether low-dose aspirin would result in an improved antithrombotic effect, without the same high risk of bleeding that has

Table 58.1 Dipyridamole plus anticoagulants following heart valve replacement

Thromboembolism Non-fatal T/E Fatal T/E Death Hemorrhage

Oral A/C alone

Oral A/C dipyridamole

% RR

2P

69/582 (11·9%) 48/582 (8·2%) 21/582 (3·6%) 67/582 (11·5%) 87/539 (16·1%)

31/559 (5·5%) 24/559 (4·3%) 7/559 (1·3%) 40/559 (7·2%) 80/501 (16·0%)

56 50 64 40 1

0·007 0·005 0·008 0·013 0·94

Abbreviations: A/C, anticoagulants; T/E, thromboembolism From Pouleur and Buyse29

834

Antithrombotic therapy after heart valve replacement

Table 58.2 Effect of aspirin combined with antithrombotic therapy following heart valve replacement

Systemic embolism Vascular death Death

Aspirin warfarin (n 186)

Placebo warfarin (n 184)

RR% (95% CI)

2P

1·6 0·6 2·8

4·6 4·4 7·4

65·0 (1·8–37·5) 85·5 (36·0–96·7) 61·7 (16·8–82·3)

0·037 0·003 0·009

Figures represent % annualized rates. From Turpie et al.32

been reported for the combination of oral anticoagulants with high-dose aspirin. This was a double-blind, randomized trial to compare the relative efficacy and safety of aspirin (100 mg per day) with placebo in the prevention of systemic embolism or vascular death in patients with mechanical heart valve replacement or high-risk patients with tissue valves who had atrial fibrillation or a history of thromboembolism. Three hundred and seventy patients were treated with oral anticoagulant therapy (warfarin: INR 3·0–4·5) and randomized to receive aspirin (186 patients) or placebo (184 patients) and followed for up to 4 years (average 2·5 years). The outcomes of the study were systemic embolism, valve thrombosis, vascular death, and hemorrhage. Systemic embolism or vascular death occurred in 6 (3·2%) of the aspirin-treated patients and 24 (13·0%) of the placebo-treated patients (RR 77·2%; 90% CI 51·7–89·2; P 0·0002). The corresponding rates for systemic embolism or death from any cause were 13 (7·0%) and 33 (17·9%), respectively (RR 64·7%; 90% CI 39·6–79·5; P 0·0005); for vascular death 2 (1·1%) and 13 (7·1%), respectively (RR 85·4%; 90% CI 49·8–95·9; P 0·0015); and for death from any cause 9 (4·8%) and 22 (12·0%), respectively (RR 62·7%; 90% CI 44·5–80·5; P 0·0048). Major bleeding events occurred in 24 (12%) of the aspirin-treated patients compared with 19 (10·3%) in the placebo-treated patients (absolute difference 2·6%; 90% CI8·3–3.4; P 0·2710). The results of this study, the annualized rates of which are summarized in Table 58.2, demonstrated that in patients with mechanical valve replacement or high-risk patients with tissue valve replacement, the addition of aspirin (100 mg per day) to oral anticoagulation therapy (warfarin: INR 3·0–4·5) reduced mortality, vascular mortality, and systemic embolism, but with some increase in minor bleeding. In a recent study in patients with mechanical prostheses,33 it was shown that low-dose aspirin (100 mg/day) was as effective as high-dose aspirin (650 mg/day) in combination with oral anticoagulants at a target INR of 2·0–3·0, but with a reduced risk of bleeding. Therefore, the addition of low-dose aspirin (80–100 mg/day) is now recommended for patients with concomitant atrial fibrillation or other additional risk factors and patients

who had thromboembolic events despite adequate oral anticoagulant therapy. Ticlopidine may also be useful as an adjunct to oral anticoagulants, but the data are less solid since the one study in which it has been evaluated was not randomized.34 There are currently no available data on the association of aspirin and clopidogrel in this setting.

Summary The demonstration that, for most indications for oral anticoagulant therapy, less intense anticoagulation (INR 2·0–3·0) is as efficacious as standard intensity anticoagulation (INR 3·0–4·5) but with significantly less bleeding is an important advance in antithrombotic therapy. It has greatly improved safety of long-term oral anticoagulant therapy and has resulted in its more widespread use in the prevention and treatment of thromboembolism. It is the level of choice in uncomplicated patients with tissue prostheses. Further evidence is required, however, before this less intense regimen is routinely adopted for patients with mechanical valves and for high-risk patients with tissue valves. The addition of lowdose aspirin to anticoagulants may be more efficacious in the prevention of systemic embolism and vascular death in heart valve replacement patients than anticoagulants alone and may permit a lower intensity of anticoagulation to be used.

References 1.Starr A. Late complications of aortic valve replacement with cloth-covered composite-seat prostheses. Ann Thorac Surg 1975;19:289. 2.Larsen GL, Alexander JA, Stanford W. Thromboembolic phenomena in patients with prosthetic aortic valves who did not receive anticoagulants. Ann Thorac Surg 1977;12:323. 3.Limet R, Lepage G, Grondin CM. Thromboembolic complications with the cloth-covered Starr–Edwards aortic prosthesis in patients not receiving anticogulants. Ann Thorac Surg 1977;23:529.

835


4.Stein PD, Alpert JS, Bussey HI et al. Antithrombotic therapy in patients with mechanical or biological prosthetic heart valves. Chest 2001;119:220S–7S. 5.Levine MN, Hirsh J, Landefeld S et al. Hemorrhagic complications of anticoagulant treatment. Chest 1992;102 (Suppl. 4):352–63. 6.Hull R, Delmore T, Genton E et al. Warfarin sodium versus lowdose heparin in the long-term treatment of venous thrombosis. N Engl J Med 1979;301:855–8. 7.Hull R, Hirsh J, Jay R et al. Different intensities of oral anticoagulant therapy in the treatment of proximal vein thrombosis. N Engl J Med 1982;307:1676–81. 8.Hull R, Delmore T, Carter C et al. Adjusted subcutaneous heparin versus warfarin sodium in the long-term treatment of venous thrombosis. N Engl J Med 1982;306:189–94. 9.Turpie AGG, Gunstensen J, Hirsh J et al. A randomized trial comparing two intensities of oral anticoagulant therapy following tissue heart valve replacement. Lancet 1988;i: 1242–5. 10.Saour JN, Sieck JO, Gallus AS. Trial of different intensities of anticoagulation in patients with prosthetic heart valves. N Engl J Med 1990;332:428–32. 11.Altman P, Rouvier J, Gurfinkel E et al. Comparison of two levels of anticoagulant therapy in patients with substitute heart valves. J Thorac Cardiovasc Surg 1991;101:427–31. 12.Cevese PG. Long-term results of 212 xenograft valve replacements. J Cardiovasc Surg 1975;16:639–42. 13.Pipkin RD, Buch WS, Fogarty TS. Evaluation of aortic valve replacement with a porcine xenograft without long-term anticoagulation. J Thorac Cardiovasc Surg 1976;71:179–86. 14.Stinson EB, Griepp RB, Oyer PE et al. Long-term experience with porcine aortic valve xenografts. J Thorac Cardiovasc Surg 1977;73:54–63. 15.Ionescu MI, Pakrashi BC, Mary DAS et al. Long-term evaluation of tissue valves. J Thorac Cardiovasc Surg 1974;68: 361–79. 16.Cohn LH, Collins JJ Jr, Rizzo RJ et al. Twenty-year follow-up of the Hancock modified orifice porcine aortic valve. Ann Thorac Surg 1998;66(Suppl):S30–S41. 17.Horstkotte D, Scharf RE, Schulteiss HP. Intracardiac thrombosis: patient-related and device-related risk factors. J Heart Valve Dis 1995;4:114–20. 18.Lonuagie YA, Jamart J, Enucher P et al. Mitral valve Carpentier–Edwards bioprosthetic replacement, thromboembolism, and anticoagulants. Ann Thorac Surg 1993;56: 931–37. 19.Moinuddeen K, Quin J, Shaw R et al. Anticoagulation is unnecessary after biological aortic valve replacement. Circulation 1998;98:II-95–II-99. 20.Acar J, Iung B, Boissel JP et al. AREVA: multicenter randomized comparison of low-dose versus standard-dose anticoagulation in

836

patients with mechanical prosthetic heart valves. Circulation 1996;94:2107–12. 21.Horstkotte D, Schulte HD, Birks W et al. Lower intensity anticoagulation therapy results in lower complication rates with the St Jude Medical Prosthesis. J Thorac Cardiovasc Surg 1994;107:1136–45. 22.David TE, Gott VL, Harker LA et al. Mechanical valves. Ann Thorac Surg 1996;62:1567–9. 23.Akins CW. Long term results with the Medtronic–Hall valvular prosthesis. Ann Thorac Surg 1996;61:806–13. 24.Cannegieter SC, Rosendaal FR, Wintzen AR et al. Optimal oral anticoagulant therapy in patients with mechanical heart valves. N Engl J Med 1995;333:11–17. 25.Ageno W, Turpie AGG. Exaggerated initial response to warfarin following heart valve replacement. Am J Cardiol 1999;84:905–8. 26.Ageno W, Turpie AGG, Steidl L et al. Comparison of a daily fixed 2·5 mg warfarin dose with a 5 mg, INR adjusted, warfarin dose initially following heart valve replacement. Am J Cardiol 2001;88:40–4. 27.Sullivan JM, Harken DE, Gorlin R. Pharmacologic control of thromboembolic complications of cardiac valve replacement. N Engl J Med 1971;284:1391–4. 28.Chesebro JG, Fuster V, Elveback LR et al. Trial of combined warfarin plus dipyridamole or aspirin therapy in prosthetic heart valve replacement: danger of aspirin compared with dipyridamole. Am J Cardiol 1983;51:1537–41. 29.Pouleur H, Buyse M. Effects of dipyridamole in combination with anticoagulant therapy on survival and thromboembolic events in patients with prosthetic heart valves. A meta-analysis of the randomized trials. J Thorac Cardiovasc Surg 1995; 110:463–6. 30.Dale J, Myhre E, Storstein O et al. Prevention of arterial thromboembolism with acetylsalicylic acid: a controlled clinical study in patients with aortic ball valves. Am Heart J 1977;94:101–11. 31.Patrono C, Coller B, Dalen JE et al. Platelet-active drugs. The relationship among dose, effectiveness and side effects. Chest 2001;119:39S–63S. 32.Turpie AGG, Gent M, Laupacis A et al. A double-blind randomized trial of acetylsalicylic acid (100 mg) versus placebo in patients treated with oral anticoagulants following heart valve replacement. N Engl J Med 1991;329:1365–9. 33.Altman R, Rouvier J, Gurfinkel E, Scazziota A, Turpie AGG. Comparison of high-dose with low-dose aspirin in patients with mechanical heart valve replacement treated with oral anticoagulant. Circulation 1996;94:2113–16. 34.Hayashi JI, Nakazawa S, Oguma F, Miyamura H, Eguchi S. Combined warfarin and antiplatelet therapy after St Jude mechanical valve replacement for mitral valve disease. J Am Coll Cardiol 1994;23:672–7.

Part IIIh Specific cardiovascular disorders: Other conditions Bernard J Gersh and Salim Yusuf, Editors

837


GRADE A

GRADE C


Level 5

GRADE B Level 2

Level 3 Level 4

838




59

Treatment of patients with stroke Craig S Anderson

The increasing burden of stroke Stroke is a major global healthcare problem.1 In most Western countries, stroke is the third leading cause of death after heart disease and cancer, and is a major cause of longterm disability and a significant cost to health and social services.2 The majority (about 75%) of new cases of stroke occur in people over the age of 65 years,2–4 and about one third of them are dead within one year.5,6 In addition to concerns about dependency and being a burden to others, survivors hope to remain free of recurrent stroke (and other serious vascular events), estimated at about 30–40% over the first five years after the onset of stroke.7,8 After a long period of neglect and an attitude to stroke that has been one of therapeutic nihilism, the past few decades has seen growing interest in the area of stroke medicine and considerable advances in the epidemiology and therapeutics of stroke. This can be explained by a number of factors, such as the ready availability of non-invasive diagnostic technology and an increase in evidence from randomized controlled trials. Computerized tomography (CT) and magnetic resonance imaging (MRI) provide the clinician with the ability to confirm the bedside diagnosis of stroke and transient ischemic attack (TIA), differentiate accurately intracerebral hemorrhage (and subarachnoid hemorrhage) from infarction, and distinguish the cerebral lesions underlying several distinct stroke syndromes in life. In comparison to ischemic heart disease, stroke is a heterogeneous clinical syndrome that encompasses a number of pathological entities that are not necessarily related to atherosclerosis, have different patterns of occurrence and outcome, and may require different management. However, it is often difficult to assign a specific cause for the different types of cerebral infarction (that is, large artery atherothrombosis, cardioembolism, and “small vessel” lacunes) in a particular individual due to the non-specific and overlapping nature of risk factors and other features. Modern neuro-imaging has also allowed a greater awareness of the importance of “silent strokes” and of the broader effects of strokes on the mind and emotion. Depression is an important sequela of stroke,9 while cerebrovascular disease and Alzheimer’s disease often coincide in older people. Indeed, evidence is accumulating that cerebrovascular disease may play a role in the etiology of Alzheimer’s disease as well as vascular dementia.10

Although the continuous decline in mortality and incidence from stroke in some populations over recent decades is an encouraging trend,11–17 there is still no general consensus about the factors responsible, or their relative contributions, to these trends. It is unclear, for example, whether there has been a change in the natural history with fewer and less severe strokes, either of which could be related to the better control of blood pressure and other risk factors; or whether there has been an improvement in survival following stroke related to improvements in acute and longterm medical care. It is also uncertain why there has been recent trend of a slowdown in the decline,15,18,19 or even an increase in mortality rates from stroke in some Eastern European countries.20 One possible explanation is that this is related to the decline in mortality from coronary artery disease and consequent rising prevalence of chronic ischemic heart disease and heart failure, which increase the pool of persons at high risk of stroke. Other potential candidates include changes in the prevalence of risk factors, in particular diabetes, obesity and smoking, and the recognized failure of hypertension detection and control programs. Untangling the puzzle of trends in stroke is a matter of pressing importance because the elderly, the most strokeprone age group, constitute the fastest growing segment of the population. If the incidence of stroke were to stabilize rather than fall, there will soon be an absolute increase in the numbers of disabled survivors of stroke, with major consequences for the health system and informal caregivers. In developed countries, at least, it has been suggested that early death from stroke in patients who were already “handicapped” from other causes may contribute to these communities avoiding an overall increase in the burden of care related to long-term survivors of stroke.21 Even so, countryspecific data on trends in the cause-specific incidence of stroke provide important local feedback on the success (or failure) of preventative strategies, while patterns of case fatality and outcome should bear a closer relationship to the management of acute stroke. Both are required for the planning of services that will inevitably come under increasing pressure from the aforementioned demographic changes. The burden of stroke, in particular, is projected to increase dramatically in developing countries, due to rapid changing population demographics and a shift from traditional, rural 839

840

62·0% 47·0% 62·7%

Control

56·4% 45·8% 56·4%

Intervention

Death or dependency

9% (4–14) 3% (1–5) 10% (5–15)

RRR (95% CI)

5·6% 1·2% 6·3%

ARR

56 12 63

Deaths/ dependents avoided per 1000 treated (n) 18 83 16

NNT

1920 (80%) 1900 (80%) 240 (10%)

Target population (% of all 2400 strokes) 107 (8·3%) 23 (1·8%) 15 (1·2%)

Deaths/ dependents avoided in 1 million population with 2400 strokes (n, (%))

? Nil additional $83 $36 000 (t-PA) $3200 (streptokinase)

Approximate cost per death or dependency avoided (Aus $)

7·0%

Cholesterol lowering drugs

b

a

8·8%

12·0%

5·0%

4·0%

6·0% 5·4% 5·1%

5·3%

4·7%

4·8%*

44% (21–60)

67% (43–80)

13% (4–21) 10% (2–17)b 15% (5–26)a, b

24% (8–38)

33% (29–38)

28% (15–38)a

RRR (95% CI)

Size of effect remains to be confirmed in ongoing trials. Compared with aspirin (ie, over and above effect of aspirin).

Carotid endarterectomy Symptomatic

Anticoagulants

7·0% 7·0% 7·0%

7·0%

Smoking cessation

Antiplatelet drugs Aspirin Clopidogrel Aspirin dipyridamole

7·0%

Control Intervention

Stroke risk per year

Blood pressure lowering drugs

Strategy/ intervention

3·8%

8·0%

1·0% 1·6% 1·9%

1·7%

2·3%

2·2%

ARR

38

80

10 16 19

17

23

22

Strokes avoided per 1000 treated per year (n)

26

12

100 62 (166b) 53 (111b)

59

43

45

TIA/ stroke patients needed to treat to avoid one stroke per year

81

84

132

Strokes avoided per year among target population (n)

3·4%

3·5%

5·5%

% of all 2400 strokes avoided each year in population of 1 million

2000 74 400 18 500

41 000

0 (voluntary) 19 600 (patches for all)

1 350 (diuretic) 18 000 (ACE inhibitor)

Approximate cost per stroke avoided (Aus $)

32

1·3%

850 (8% of 10 650 TIA/ischemic strokes) 182 000

2130 (20% of 10 650 TIA/ischemic strokes) but only up to 1065 (50%) realistically 3·5% 1200 85

8000 (75% of 10 650 TIA/ischemic strokes) 3·3%b 80b 128 (48b) 5·3% (2·0b) 152 (72b) 6·3% (3·0%b)

4800 (40%)

3600 (30%)

6000 (50%)

Target population (% of all prevalent TIA and stroke survivors)

Table 59.2 Summary of secondary stroke prevention strategies and their cost effectiveness for patients with stroke or TIA each year in a large population. (Reproduced with permission from Hankey GJ, Warlow CP. Treatment and secondary prevention of stroke: evidence, costs, and effects on individuals and populations. Lancet 1999;354:1457–63.)

Stroke unit Aspirin Thrombolysis

Intervention

Table 59.1 Summary of effectiveness and costs of acute treatment for stroke each year in a large population. (Modified with permission from Hankey GJ, Warlow CP. Treatment and secondary prevention of stroke: evidence, costs, and effects on individuals and populations. Lancet 1999;354:1457–63.)

Treatment of patients with stroke

to urban lifestyles.1,22 It is imperative, then, that researchers, healthcare providers and policy makers develop appropriate and cost effective interventions for the prevention, treatment and management of stroke. Hankey and Warlow23 provide an excellent overview of the effectiveness and costs of such strategies, and this is summarized in Tables 59.1 and 59.2.

Stroke units and stroke services Arguably the single most important therapeutic advance during the past decade in the treatment of patients with acute stroke has been the development of stroke services and stroke units. Strong evidence exists from pooled clinical trial data that well coordinated multidisciplinary, inpatient, stroke unit care can significantly improve the likelihood of returning home and retaining independence after stroke.24 Grade A1c Compared with conventional care in a general medical ward, stroke units are associated with a relative risk reduction (RRR) of 9% and an absolute risk reduction (ARR) of 5·6%. Thus, the number of patients with stroke that need to be treated (NNT) on a stroke unit to prevent one from dying or becoming dependent is an impressive 18.23 The benefit of stroke units applies across all subgroups of patients including those who are old, severely disabled, or are admitted late to hospital.24 While it is uncertain which part of this expert care is important, there is broad consensus that stroke services, both in the acute and rehabilitation phases, need to be well coordinated and include a multidisciplinary team approach, active participation of family, and special

education and training of staff (see Table 59.3). In common with coronary care units, stroke units also facilitate the conduct of randomized trials and allow the development of protocols and practices to facilitate rapid, early and thorough evaluation, treatment and rehabilitation. There is much interest in extending the “black box” package of stroke unit care into the community. In the past decade, there have been an increasing number of trials of specialist domiciliary (home-based) stroke care and rehabilitation schemes, with the aim of either avoiding the need for admission to hospital, or enabling earlier and more effective discharge and follow up. Although there may be scope to prevent some admissions to hospital after stroke, it is not an easy task, not least because stroke is a frightening illness, with most patients disabled at onset. In most countries, hospitals offer a safe and secure environment for intensive nursing care and rehabilitation. Patients with stroke, therefore, conventionally receive a substantial part of their acute care and rehabilitation in hospitals or in other institutions that offer a 24 hour stay. This emphasis on hospital care, together with greater public and professional education on acute symptoms and the need to regard stroke as an emergency,25,26 mean that it is probably unrealistic to anticipate major service development to the area of “hospital avoidance” for patients with stroke in the future. Conversely, there is much interest in the development of services that allow patients with stroke to be sent home from hospital earlier than usual and receive domiciliary rehabilitation. Advocates of early discharge schemes suggest several advantages: satisfying patient choice, reducing risks associated with prolonged inpatient care, the home setting

Table 59.3 Organization of stroke services – evidence grades Recommendation 1. Every health care organization involved in the care of patients with acute stroke should ensure that the there are specialty service(s) responsible for the management of these patients which comprise the following factors: ● A geographically defined unit as the inpatient service base ● A well coordinated multidisciplinary team ● Staff with special training and expertise in stroke care and rehabilitation ● Educational programs for staff, patients and caregivers ● Agreed-on protocols for common problems 2. Specialist stroke services can be delivered to patients, following the acute phase, equally effectively in hospital or the community 3. Rehabilitation can be provided to patients within a specialist outpatient or domiciliary setting with equal effect 4. Patients with acute stroke who are not admitted to hospital can benefit from a domicilary rehabilitation team that includes an occupational therapist.

Evidence grade Grade A1c

Grade A1c Grade A1c Grade A1c

841


being more focused toward rehabilitation outcomes, and savings in costs.27 Since 1997, several randomized trials of early hospital discharge and domiciliary stroke rehabilitation have been published.28 These data are consistent with regard to no adverse impact on patient outcomes and a reduction in hospital length of stay, and the limited economic analyses available indicate potential cost-savings with such schemes.29,30 There is high quality evidence that allows some broad recommendations to be made regarding the organization of stroke care31 (see Table 59.1).

Treatment of ischemic stroke Another major therapeutic landmark in the management of stroke is the use of thrombolytic therapy for acute ischemic stroke. Most acute strokes are due to cerebral infarction following occlusion of arterial blood vessels. The pathogenesis of resulting brain damage can be separated into two sequential processes: 1. 2.

the vascular and hematological events that produce occlusion and reduce blood flow in the affected area ischemic necrosis of brain cells.

Surrounding blood vessels in the brain may partly maintain blood flow into the damaged area and therefore, the outer regions of the damaged area are less severely affected than within the core. This process produces an area of irreversible severe ischemia surrounded by an area of moderate ischemia, known as the “penumbra”. The recognition that further death of neurons in the penumbra may be prevented has focused attention on treatments to minimize, or even reverse, the damaging effects of ischemia provided they can be initiated within a short period of time after the onset of stroke. This “therapeutic window” may be divided into two partly overlapping components: the “reperfusion window” related to the restoration of blood flow and the “neuroprotective window” related to damaging effects within brain cells. Studies in various animal models and clinical trials suggest that the reperfusion window is very short, perhaps only a few hours, while the neuroprotective window may be much longer, maybe up to 48 hours. While an effective neuroprotective agent for acute stroke has yet to be identified, considerable progress has been made in therapies aimed at restoring blood flow to prevent or lessen the spreading ischemia within the penumbra. In 1996, the United States Food and Drug Administration (FDA) approved intravenous recombinant tissue plasminogen activator (rt-PA) in selected patients with acute ischemic stroke, provided treatment can begin within 3 hours of the onset of symptoms. The approval was based largely on the results of two combined, National Institute of Neurologic Disease and Stroke (NINDS) Acute Stroke Studies, where all patients were treated within 3 hours of the onset of symptoms 842

and half of the patients were treated within 90 minutes.32 Subsequent individual trials of rt-PA (and streptokinase) with time windows extending up to 6 hours after the onset of stroke have failed to show a definite positive benefit on their own, but several meta-analyses of the trials indicate a large beneficial effect of treatment, albeit with significant risk, mainly intracerebral hemorrhage.33,34 Risk of intracerebral hemorrhage is estimated to be 5–10%, and appears more likely to occur in patients with evidence of a large visible infarct on CT, and concurrent use of aspirin among other factors. Despite this risk, use of intravenous rt-PA appears to result in at least a 30% RRR in disability from stroke. The benefits (about 65 patients per 1000 treated avoid “death or long-term dependence”) appear to be several times greater than for aspirin, the only other proven effective medical treatment used early after the onset of stroke.35,36 Grade A1a Although licenses have been granted for the use of rt-PA in several countries and subsequent consensus statements from lead professional organizations such as the American Heart Association37 endorsing the use of intravenous rt-PA. The uptake of this therapy in clinical practice is extremely limited, even despite intensive community and professional awareness campaigns.38 In addition, there has been criticism of editorial format of certain consensus statements39 and concerns raised about the randomization process in the NINDS trial and failure to adjust for imbalance in baseline variables (J Wardlow, personal communication). Thus, there is a long way to go before we can use thrombolysis widely in patients with acute ischemic stroke. Despite the published data, consensus statements, and guidelines, only a very small minority of patients with acute ischemic stroke currently receive intravenous rt-PA, mainly due to various educational barriers and delays in getting people to hospital quickly after the onset of symptoms. Much more data is required to establish reliably the balance of benefits and risks of thrombolysis across different groups of patients, even those patients treated after the 3 hour time window.33,34 A large multicenter trial (International Stroke Trial (IST) – 3) has commenced to address these issues and influence clinical practice. Apart from rt-PA, two very large trials have established that aspirin 300 mg, administered within the first 48 hours after the onset of ischemic stroke, reduce the risk of death or dependency at 6 months by 1·3%, mainly by reducing early recurrent ischemic stroke.35,36 Grade A1a Despite a long history of use as standard therapy for established or threatened acute ischemic stroke, a large number of trials individually, and when combined in a metaanalysis,35,40 have established beyond doubt that heparin in any form, dose or route of administration has no benefit. Even among patients with presumed cardio-embolic stroke, the modest potential benefit is counterbalanced by a significant


excess risk of symptomatic hemorrhage, of which the most lethal is intracerebral hemorrhage.35,40,41 Grade A1a

Prevention of stroke The most realistic approach to lessen the burden of stroke is prevention, with both population-based strategies and the targeting of high-risk individuals being advocated. Epidemiologic studies suggest that significant reductions in the incidence of stroke, as with coronary heart disease, can be expected by reducing the prevalence or shifting the distribution of risk factors across the entire population.42 Thus, identifying risk factors and intervening to control or modify them remains the most important means of reducing the burden of stroke. Favorable lifestyle behaviors, including weight reduction, diets that are high in fish, fruit and vegetables, and increased physical activity, are based on sound epidemiologic data. Although direct evidence is lacking, observational studies suggest that stopping smoking decreases the risk of stroke by at least 30%.43 A substantial reduction in the incidence of stroke has been noted following cessation of smoking, even in older people and in those who have been heavy smokers for many years.44 Grade B2 The “high risk” strategy involves the identification and management of people at high risk of developing stroke. Therapies of proven benefit in the prevention of stroke among certain individuals are blood pressure lowering therapy, antiplatelet therapy, cholesterol lowering therapy, anticoagulation, and carotid endarterectomy. Evidence is mounting that aggressive treatment of hyperglycemia among patients with diabetes mellitus is also effective in reducing the risk of stroke.45 The absolute benefits of these interventions appear to be greater in subjects in whom the absolute risk is particularly high, notably older people. Effective prevention of stroke in individuals depends on the efficient identification and management of these subjects, particularly in primary care. Blood pressure reduction strategies Blood pressure is the single most important reversible risk factor for stroke. Pivotal data about the relationship between blood pressure and stroke comes from both prospective observational studies and clinical trials. Observational studies provide information from which the effects of prolonged blood pressure differences can be estimated,46 while trials provide data about the effects of short-term blood pressure reduction.47 Four major overviews of observational studies on blood pressure and stroke have been conducted to date. Such analysis overcomes many of the limitations of individual studies, which have frequently failed to adjust the size of the association for measurement error, in particular regression dilution bias.46,48–50 A consistent finding of these overviews is a continuous, approximately log linear relationship

between usual levels of blood pressure and the primary incidence of stroke, with no evidence of an upper or lower threshold level of blood pressure and stroke risk. On the basis of this relationship, it is estimated that a 5 mmHg lower diastolic blood pressure (or 10 mmHg lower systolic blood pressure) is associated with a 30–40% lower risk of stroke, and there is no evidence that these associations differ between men and women. Most of the trial data is on the primary prevention of stroke, confirming beyond doubt the benefits of blood pressure lowering in preventing first-ever stroke in middleaged men and women. Several meta-analyses of trials51–53 demonstrate that lowering blood pressure in this age group is effective in preventing stroke, with risk reductions commensurate with predictions based on non-experimental epidemiologic studies of 30–40%. Grade A1a Moreover, the Heart Outcomes Prevention Evaluation (HOPE) trial54 results suggest that activation of the renin– angiotensin system is an independent risk factor for stroke. In this study of 9297 patients with a history of symptomatic vascular disease (mainly coronary artery disease), who were randomized to either ramipril 10 mg daily or placebo on top of best medical therapy, there was a significant reduction in the rate of the combined end point cluster of stroke, myocardial infarction, or death from vascular causes in the patients allocated ramipril (14·0%) compared with those given placebo (17·8%). This RRR of 22% (95% CI 14–30) and an ARR of 3·8% over about 5 years of follow up was much greater than could be expected from the size of the reductions in blood pressure (3 mmHg systolic, 1 mmHg diastolic). The effects of treatment on the end point of any stroke, a RRR of 32% (95% CI 16–44), were consistent across baseline blood pressures, concurrent medication use, and important patient subgroups including those with and without a history of hypertension.55 Moreover, these data support the hypothesis that angiotensin converting enzyme (ACE) inhibition has beneficial effects independent of blood pressure reduction. Existing data from randomized controlled trials and a systematic review including patients with cerebrovascular disease suggested that blood pressure lowering therapy may reduce the risk of recurrent stroke by an equivalent amount to that of primary stroke prevention.56 However, this evidence has not been compelling enough to influence clinical practice. In fact, the approach to blood pressure control among clinicians involved in the care of patients with stroke, and particularly those who are old and normotensive, has been conservative, due in part to concerns about the adverse effects of aggressive blood pressure lowering in this setting. The Perindopril Protection Against Recurrent Stroke Study (PROGRESS) undertaken in over 6000 patients was designed to address this issue by determining the effects of a flexible ACE-inhibitor based blood pressure lowering regimen (perindopril with or without the addition of the diuretic, 843


indapamide) on the risks of stroke and other major vascular events in patients with a history of stroke or TIA.57 Overall, blood pressure was reduced by an average of 9·0/4·0 mmHg (SE 0·3/0·2) among patients assigned active treatment compared with those assigned placebo during the trial. Compared with those assigned to placebo, the blood pressure reductions among those treated with combination therapy (12·3/5·0 mmHg, SE 0·5/0·3) were about twice as high as those treated with single-drug therapy (4·9/2·8 mmHg, SE 0·6/0·3). The study showed that treatment reduced the incidence of stroke, coronary events and major vascular events by 28%, 26% and 26%, respectively. Active treatment reduced the risks of ischemic stroke by around one quarter and hemorrhagic stroke by one half, and was equally effective in patients with and without a history of hypertension. Combination perindopril and indapamide provided even greater benefits. Thus, blood pressure lowering therapy should now be regarded as the most important measure for both the primary and secondary prevention of stroke. Grade A1a Although the choice of therapy will depend on the degree of acceptance of direct evidence, as well as on regulations and prescribing patterns, the evidence is strong for therapy that includes an ACE inhibitor and maximizes the degree of blood pressure reduction, such as a combination of perindopril and indapamide. Moreover, effective implementation of such therapy in high-risk groups in combination with population-wide blood pressure lowering strategies, provides one of the most meaningful, practical, and effective ways of controlling the looming epidemic of cardiovascular disease and stroke, worldwide. Antiplatelet therapy In the late 1950s and mid-1960s two research paths converged. The first involved the identification of platelet fibrin thrombogenesis as a cause of retinal and hemispheric TIAs. The other involved investigators in Toronto, New York, and Oxford who coincidently determined that several drugs – sulfinpyrazone, aspirin, and dipyridamole – altered platelet responsiveness both in the test tube and in experimentally injured arteries.58–60 Clinical trials of platelet inhibitors were subsequently undertaken for the prevention of stroke, interestingly before being tested in other vascular diseases. The first was a small trial of dipyridamole involving only 169 patients, which showed no benefit.61 Next, the Canadian Cooperative Study was undertaken using a factorial study design in which patients received aspirin 1300 mg daily, sulfinpyrazone 800 mg, both or neither in patients with a recognizable arterial origin for their TIA or nondisabling stroke.62 After 585 patients, two thirds male and one third female, had been followed for an average of 28 months, the investigators reported that patients in the two arms of the trial containing aspirin compared with those 844

not on aspirin had a RRR of 31% for the combined end point of stroke and death. No benefit was detected in the sulfinpyrazone arm. A subgroup analysis reported no benefit for the 200 women in the trial and when the results for men alone were analyzed in a data generated subgroup, a 48% benefit in stroke and death was observed in the aspirin containing groups. Since the publication of that important study there has been a flurry of activity. Most importantly, the mode of action and effectiveness of aspirin has been well elucidated, and several new antiplatelet agents identified and tested. Aspirin inhibits thromboxane A2 formation by irreversibly acetylating the platelet enzyme cyclo-oxygenase. Thromboxane A2 is an important stimulus for platelet aggregation and release. Platelet aggregation is thus inhibited for up to 10 days after exposure to aspirin. Absorption of aspirin occurs rapidly and peak plasma concentrations are reached within 1–3 hours. Even though the plasma half life of aspirin is short, antiplatelet activity is prolonged, but bleeding times return to normal within two days of cessation of aspirin. In a meta-analysis of 145 randomized trials of antiplatelet therapy, aspirin was shown to be associated with a RRR of all vascular events (including stroke) of about 22%.63 Grade A1a Aspirin is, therefore, appropriate for all patients with ischemic stroke unless there is specific contraindication, such as aspirin sensitivity. Aspirin given to patients with a history of stroke or TIA reduces the relative risk of stroke and other important vascular events by about 14% (95% CI 4–21),64 from about 7% to 6% per year. This equates to an ARR of 1·0% and an NNT of 100. Aspirin is also beneficial in such patients who also have atrial fibrillation (AF). The annual risk of stroke has been shown to be reduced from 12% to 10%, a RRR of 14% (95% CI 15–36) and an ARR of 2·0%.65 In the mid-1990s, there was much controversy over the most effective dose of aspirin.66–68 The optimal dose for efficacy and tolerability is low dose aspirin (100–300 mg daily). Gastrointestinal side effects such as gastritis and hemorrhage are more common in older people and are dependent on the dose and duration of treatment. The small risk of intracerebral hemorrhage with prolonged use of aspirin is outweighed by the benefits in high-risk patients. Among the other antiplatelet agents, dipyridamole, ticlopidine and clopidogrel have all been shown to be beneficial in the secondary prevention of stroke. Compared with aspirin, clopidogrel reduces the relative risk of stroke and other major vascular events by about 10% (95% CI 3–17)69,70 from about 6·0% (aspirin) to 5·4% (clopridogrel) per year, which is an ARR of 0·6% compared with aspirin, and equates to about a NNT of 166 compared with aspirin, and probably 62 compared with aspirin. Grade A1a Although expensive, clopidogrel is as safe as aspirin and, unlike the closely related agent ticlopidine, is not known to


cause an excess of neutropenia and thrombocytopenia.70,71 In the European Stroke Prevention Study 2 (ESPS 2),72 6602 patients were given aspirin (25 mg twice daily), modified-release dipyridamole (200 mg twice daily), the combination or placebo. The RRR for the combined end points of stroke and death were 13·2% for aspirin, 15·4% for dipyridamole, and 24·4% for the combination of aspirin and dipyridamole. An update of the data indicates that compared with aspirin, the combination of aspirin and modifiedrelease dipyridamole reduces the risk of stroke by about 23% (95% CI 7–37), and that this effect is greater for stroke than other serious vascular events.73,74 Grade A1c Anticoagulants AF is a major risk factor for stroke. It predisposes to the formation of intracardiac thrombi, mainly within the atria, that may embolize to the brain and other organs. The prevalence of AF increases with age, from 0·5% in patients aged 50–59 years to about 9% in patients over the age of 70 years.75 With an aging population, it is likely that AF will become an increasingly important public health problem. Overall, the annual risk of stroke is about 5%, but rates vary from less than 2% to more than 10% according to the presence of one or more clinical characteristics such as congestive cardiac failure, hypertension, older age (75 years), diabetes and previous history of stroke or TIA.76 Grade A1a There is good evidence that warfarin and aspirin are both highly efficacious in preventing stroke (and other vascular and embolic events) in patients with AF. Six randomized trials have evaluated warfarin with placebo,77–82 two other trials have compared warfarin with aspirin,83,84 and a metaanalysis85 indicates a 64% RRR of stroke favoring warfarin over placebo and a 48% RRR favoring warfarin over aspirin. Aspirin is associated with a RRR of 22%.86,87 Given the consistency of the treatment effects across subgroups of patients, the absolute benefits of antithrombotic therapy is high in those patients who are at highest risk of stroke. Thus, the decision to commence warfarin anticoagulation therapy is usually based on an evaluation of the risks and benefits in an individual patient. Bleeding is the main risk associated with warfarin. Results of the pooled analyses of the major studies show a slightly higher frequency of major bleeding events in warfarin treated groups compared with placebo (1·3% v 1·0% per year). On the basis of these data, a target international normalized ratio (INR) of between 2·0 and 3·0 is generally recommended as a safe and effective level of anticoagulation.88 These recommendations are further supported by the results of the Stroke Prevention in Reversible Ischemic Trial (SPIRIT),89 which involved 1316 patients with TIA or minor stroke in which aspirin (30 mg daily) was compared with warfarin and a target INR of 3·0 to 4·5. The trial was stopped after the first scheduled interim analysis due to an excess of major bleeding in the

warfarin group, and no significant difference between groups in the incidence of the non-hemorrhagic end points (hazard ratio 1·03, 95% CI 0·6–1·75). Based on small studies only, warfarin (or heparin) appears to reduce the incidence of stroke in patients with recent anterior myocardial infarction.90 Patients who have had a recent stroke or TIA in conjunction with a recent myocardial infarction should be given heparin therapy followed by 3–6 months of warfarin therapy.91 In the setting of acute ischemic stroke attributed to cardio-embolism, though, it is often advisable to wait at least a week before commencing anticoagulation, in order to prevent hemorrhagic transformation of the infarct. Grade B2 In the absence of a potential cardiac source for the stroke, such as AF, recent transmural, anterior, myocardial infarction, dilated cardiomyopathy, valvular heart disease, there are presently no other clear indications for warfarin therapy for the prevention of stroke. In the recently completed Warfarin-Aspirin Recurrent Stroke Study (WARSS), there was no significant difference in the prevention of recurrent ischemic stroke, death or intracranial hemorrhage between the groups, although there was a slight trend for aspirin superiority.92 Grade A1a There may be benefit of warfarin in certain other subgroups of patients with vascular disease. On the basis of retrospective observational data, the Warfarin-Aspirin Symptomatic Intracranial Disease study93 a clinical trial has commenced comparing warfarin with aspirin in patients with severe intracranial arterial stenosis. Cholesterol lowering therapy The relation between diet, serum cholesterol levels and ischemic heart disease is relatively well understood but the evidence is less reliable and more complex for stroke. This reflects, in part, the diversity of stroke pathogenesis, with qualitatively different associations of serum cholesterol with the risks of ischemic and hemorrhagic stroke.94–98 High serum cholesterol is considered an important risk factor for ischemic stroke, more so in Western societies than in Asian populations,99 whereas a weak inversion association is apparent between serum cholesterol and the risk of intracerebral hemorrhage. Although there have been no completed randomized trials of cholesterol lowering therapy in patients with stroke or TIA alone, several large scale, randomized trials have established that long-term use of certain 3hydrozy-3-methylglutaryl co-enzyme A reductase inhibitors (statins) results in significant reductions in the risk of major cardiovascular events in patients with a wide range of lipid levels, both100 with101–103 and without,104,105 a history of coronary artery disease. The evidence suggests that lowering serum cholesterol with statins can reduce the risk of stroke by about 25% (95% CI 14–41) over just a few years of therapy.106 Grade A1a 845


Carotid endarterectomy Symptomatic disease Atherosclerotic disease of the carotid artery is an important cause of stroke. The risk of recurrent stroke in recently symptomatic patients with severe carotid stenosis is as high as 28% over 2 years. The introduction of carotid artery surgery in 1954 by Eastcott, Pickering, and Rob heralded a new era in stroke prevention. In 1967 a randomized trial was launched to evaluate this exciting prospect of carotid endarterectomy, with results published in 1970.107 The trial, despite its merits, was not supportive of the procedure due to a number of problems including the small sample size of only 316 randomized patients and relatively high perioperative complication rate of 11%. In addition, almost half of the patients had symptoms only in the vertebral basilar vascular territory and too many patients were lost to follow up (12% in the surgical and 1·3% in the medical group). A second small trial conducted at the same time but reported much later, was overwhelmingly negative due to a high perioperative complication rate.108 These negative trials did not reduce the enthusiasm for the procedure. By 1985 a total of 107 000 endarterectomies were being carried out annually in the United States and it was estimated that a cumulative total of one million endarterectomies had been conducted for both symptomatic and asymptomatic disease. The appropriateness of patient selection and the awareness that in many centers, a forbiddingly high level of operative morbidity and mortality existed, led to a requirement for high quality clinical trials. The two major trials for symptomatic disease are the North American Symptomatic Carotid Endarterectomy Trial (NASCET) and the European Carotid Surgery Trial (ECST), which between them included nearly 6000 patients, and they have demonstrated convincingly the benefits of carotid endarterectomy.109,110 A third smaller trial, when stopped,

had a trend towards the same result but involved only 189 patients.111 NASCET and ECST required angiography for entry and demanded focal hemisphere or retinal minor stroke or TIAs within 180 days. Both stopped the randomization of patients with severe stenosis because of compelling evidence on interim analyses demonstrating a clear reduction in ipsilateral strokes with endarterectomy. Both trials also showed no net benefit of surgery for all patients with mild-moderate grades of stenosis.112,113 NASCET and ECST used measurements of the narrowest diameter of the stenosed segment as the numerator based on angiographic criteria. The results from NASCET demonstrated a 2 year 65% relative and a 17% absolute risk reduction favoring endarterectomy (Table 59·2). When the ECST angiograms were remeasured by the NASCET method and the results calculated for the reduced number of patients who would be “severe” by NASCET criteria, the favorable results in the survival curves for surgery were very similar.114 Grade A1a The compelling results in favor of endarterectomy for severe stenosis in NASCET and ECST are dependent on two important caveats. First, the surgical complication rate undertaken by experienced surgeons was low. In NASCET the occurrence of any stroke, disabling or mild, lasting more than 24 hours or death in the 30 day period was 5·8%. For disabling stroke and perioperative death, the complication rate was 2·1%. The long-term benefits of endarterectomy are abolished as the complication rate exceeds 10%. Surgeons must therefore be highly skilled. Second, the results relate to a measurement of the degree of stenosis from conventional carotid angiograms. Conventional angiography carries a 1% risk of stroke; one stroke in five is disabling.115 In NASCET, from 2929 angiograms performed in 100 centers, the minor and non-disabling stroke rate was 0·6%, the disabling stroke rate 0·1%. While this rate was high, it was only one tenth the risk of stroke from

Table 59.4 Risk of ipsilateral stroke or any perioperative stroke or perioperative death – NASCET. (Reproduced with permission from Neurology 1996;46:605) Study time

Risk (%) medical

Risk (%) surgical

Risk (%) difference

RRR (%)

NNT

30 days 1 year 2 years

3·3 17·3 26·0

5·8 7·5 9·0

2·5 9·8 17·0

— 57 65

— 10 6

From the North American Symptomatic Carotid Endarterectomy Trial (NASCET), the risk of stroke for 331 medically treated and 328 surgically treated patients with symptoms appropriate to severe stenosis are given at 30 days, 1 and 2 years. To prevent one stroke in 2 years, six patients need to have endarterectomy. Abbreviations: NNT, number needed to treat by endarterectomy to prevent one stroke within the specified study time; RRR, relative risk reduction

846


endarterectomy: 5·8% for any stroke or death and 2·1% for disabling stroke or death. Modern non-invasive carotid imaging such as MRI, spiral CT and duplex and Doppler ultrasound avoids the risk associated with conventional intraarterial catheter carotid angiography. However, it is important that the reliability of images obtained are well established in order to avoid operating on patients with false-positive scans, or denying benefits of the intervention in those with falsenegative scans.116 Moreover, carotid ultrasound alone will not identify important lesions of the intracranial arteries. Aneurysms and stenosis of intracranial vessels exists in about 2% and 5% of patients symptomatic with extracranial stenosis, respectively. If such lesions are identified, the benefit:risk ratio of endarterectomy will be reduced. Although endarterectomy remains the standard treatment for a well defined group of high-risk patients, there is growing interest in the use of percutaneous, transluminal angioplasty (PTA) and stenting for carotid stenosis to avoid the significant risk of stroke or death of between 6% and 8% associated with surgery. In the Carotid and Vertebral Artery Transluminal Angioplasty Study (CAVATAS) involving over 500 patients, almost all with symptomatic carotid stenosis, randomized to PTA and surgery, the 30 day outcomes were almost identical for the groups.117 The rate of death or any major stroke was 9·9% after surgery and 10·0% after angioplasty. Analysis of the other risks of treatment confirmed that PTA was safer than surgery in terms of minor morbidity, reduced hospital length of stay, but was associated with a higher rate of (asymptomatic) restenosis during follow up. Stents suitable for carotid use have only been available in recent years and few were used in CAVATAS. Given advances in technology in this area, further clinical trials are underway to evaluate whether primary carotid stenting should be the surgical procedure of choice for carotid stenosis based on effectiveness, lower risks and reduced costs.

Asymptomatic disease Design flaws in early trials prevented conclusive evidence being gained about the benefit of endarterectomy in patients with asymptomatic carotid stenosis. In the Carotid Artery Stenosis with Asymptomatic Narrowing: Operation Versus Aspirin (CASANOVA) trial, patients with greater than 90% stenosis were excluded and crossovers were common between the medical and surgical groups, while in the Mayo Asymptomatic Carotid Endarterectomy (MACE) trial, standard medical care including use of aspirin, was not extended to the surgical arm. The United States Veterans Administration (VA) trial was small, with only 444 patients, and reported a perioperative complication rate of 4·4%, but the stroke-free survival curves were similar in patients receiving endarterectomy to those given best medical care alone.118 A fourth trial, the Asymptomatic Carotid Atherosclerosis Study (ACAS), used a more robust design and randomized 1662 patients. A significant benefit was demonstrated in favor of endarterectomy with a RRR of 53% after 2·7 years of average follow up.119 However, there are reservations about the clinical significance of this result (see Table 59.5). The absolute benefits of the procedure in this setting are small, RRR of 1% per year, so that 67 patients are required to undergo endarterectomy to prevent one stroke in 2 years. This is in contrast to the six symptomatic patients required to derive benefit with similarly severe disease. When the perioperative complication rate exceeds 3%, the benefits are negated. A higher figure is known to be common. Thus, endarterectomy for all patients with asymptomatic carotid stenosis is not a cost effective procedure. Grade A1c All of the symptomatic trials and major observational case-series studies have observed a worsening of prognosis with increasing degrees of carotid stenosis.120,121 However, there is also a strong association between cardiovascular risk

Table 59.5 Risk of ipsilateral stroke or any perioperative stroke or perioperative death (based upon 825 patients randomized to CEA) — ACAS. (Reproduced with permission from Neurology 1996;46:605) Study time

Risk (%) medical

Risk (%) surgical

Risk (%) difference

RRR (%)

NNT

30 days 1 year 2 years 5 years

0·4 2·4 5·0 11·0

2·3 3·0 3·5 5·1

1·9 0·6 1·5 5·9

— — 30 53

— — 67 17

From the Asymptomatic Carotid Atherosclerosis Study (ACAS), the risk of stroke is compared between the medically and surgically treated patients. To prevent one stroke in 2 years, 67 patients need to have endarterectomy. Adjusting the surgical arm to include only the patients who had endarterectomy, the 30 day surgical risk becomes 2·6% and the 1, 2 and 5 year risks rise to 3·3%, 3·8% and 5·4% respectively. The number needed to treat to prevent one stroke in 2 years becomes 83. Abbreviations: NNT, number needed to treat by endarterectomy to prevent one stroke within the specified study time; RRR, relative risk reduction

847


factors and carotid stenosis in all age groups.122 Thus, although the reduction in strokes after endarterectomy is found to be greatest in those with the most severe stenosis, there remains uncertainty about the benefits of endarterectomy in the elderly, including those with asymptomatic disease.123 In view of this uncertainty, a fifth and largest trial is being conducted in Europe, the Asymptomatic Carotid Surgery Trial (ACST).124 It is possible that a high-risk patient subgroup, for example those with high grade stenosis (85–99%) and a high vascular risk profile, will be identified in which endarterectomy is definitely cost effective for asymptomatic carotid disease. In the meantime, only carefully selected high risk patients should be recommended for endarterectomy, conducted by the most expert of surgeons.

●

●

Conclusions Four decades of clinical observation and randomized controlled trials in stroke prevention have provided very positive and promising results. Several key points emerge: ●

●

●

●

848

Modifiable risk factors for stroke have been identified. From a public health viewpoint, risk factors of greatest importance in the prevention of stroke are those that carry a high population-attributable risk, such as high blood pressure, obesity, high cholesterol levels, physical inactivity, and cigarette smoking. Stroke prevention in the community requires manipulation of these risk factors in individuals at high risk and in the whole population. Given the continuous relationship between levels of risk factors and stroke risk, effective prevention of stroke involves the management of the patient as a whole person defined by their absolute risk of future major vascular events rather than by a single variable such as the level of blood pressure or serum cholesterol. On the basis of a large body of direct and indirect evidence, clinicians should now consider blood pressure lowering therapy as pivotal to the prevention of recurrent stroke in all patients with cerebrovascular disease, irrespective of blood pressure levels, age and other characteristics. Although there is no evidence at present to guide the timing of blood pressure lowering therapy after the onset of stroke, pragmatically it is probably wise to wait until patients are clinically stable. The evidence is strong for therapy that maximizes the degree of blood pressure reduction using an ACE inhibitorbased regimen. Aspirin is cheap, safe, familiar and acceptable as the antiplatelet agent of first choice for patients with vascular disease. The optimal dose of aspirin is 100–300 mg. Aspirin is also the first choice for patients with acute ischemic stroke, commencing within 48 hours of onset. Clopidogrel is more expensive and combination aspirin

●

●

●

and modified-release dypridamole is less well tolerated but both agents offer modest benefits over aspirin. Warfarin is indicated only for patients with a proven or potential, cardiac (embolic) source of stroke. Warfarin is favored when the risk of stroke is high and aspirin is favored when the risk of stroke is low. Various clinical parameters have been well identified that allow clinical stratification of risk. Carotid endarterectomy is indicated for patients with severe carotid artery stenosis who have had symptoms of retinal or brain ischemia appropriate to the stenosis, and are willing to undergo a small but definite risk of death or disability related to the procedure undertaken by an experienced surgeon. PTA may be undertaken as an alternative to carotid endarterectomy in experienced hands. There remains uncertainty about the cost effectiveness of endarterectomy in patients with high grade asymptomatic stenosis and that of carotid stenting. Cholesterol lowering therapy with statins is safe, well tolerated and cost effective in preventing major vascular events including stroke in all high-risk individuals irrespective of baseline cholesterol levels. Well organized care and rehabilitation within stroke units or services has been shown to save lives and reduce long-term dependency. The key components of such care should include multidisciplinary team approach, active participation of family, special education and training of staff, and early commencement of rehabilitation. The primary focus of thrombolytic therapy is to restore, preserve and improve circulation to acute focal brain ischemia. Although rt-PA remains the only approved hyperacute stroke therapy, all other data suggest that other thrombolytic or neuroprotective approaches are likely to have very short time-windows for efficacy and safety. While the risk of bleeding is significant, it is comparable with the risks associated with other procedures such as carotid endarterectomy. Effective community and professional education, motivation and training are likely to be important for more widespread application as acute stroke treatment.

References 1.Murray CJL, Lopez AD. Global Health Statistics. Geneva: World Health Organization, 1996. 2.Warlow CP. Epidemiology of stroke. Lancet 1998;352 (Suppl. 3):SIII 1–4. 3.Bonita R, Anderson C, Broad J, Jamrozik K, Stewart-Wynne E, Anderson N. Stroke incidence and case-fatality in Australasia: a comparison of the Auckland and Perth population-based stroke registers. Stroke 1994;25:552–7. 4.Bonita R, Broad J, Beaglehole R, Anderson N. Changes in incidence and case-fatality in Auckland, New Zealand, 1981–1991. Lancet 1993;342:1470–3.


5.Anderson CS, Jamrozik KD, Broadhurst RJ, Stewart-Wynne EG. Predicting survival for 1 year among different subtypes of stroke. Results from the Perth Community Stroke Study. Stroke 1994;25:1935–44. 6.Bonita R, Stewart AWS, Ford M. Predicting survival after stroke: a three-year follow up. Stroke 1988;19:669–73. 7.Hankey G, Jamrozik K, Broadhurst R et al. Long-term risk of recurrent stroke in the Perth Community Stroke Study. Stroke 1998;29:2491–500. 8.Burn J, Dennis M, Bamford J, Sandercock P, Wade D, Warlow C. Long-term risk of recurrent stroke after a first-ever stroke: The Oxfordshire Community Stroke Project. Stroke 1994; 25:333–7. 9.House A. Mood disorders after stroke: a review of the evidence. Int J Geriatric Psychiatry 1987;2:211–21. 10.De la Torre JC. Alzheimer Disease as a vascular disorder: nosological evidence. Stroke 2002;33:1152–62. 11.Thom JT. Stroke mortality trends: an international perspective. Ann Epidemiol 1993;3:509–18. 12.Bonita R, Beaglehole R. Monitoring stroke: an international challenge. Stroke 1995;26:541–2. 13.Bonita R, Broad JB, Beaglehole R. Changes in stroke incidence and case-fatality Auckland, New Zealand in 1981–1991. Lancet 1993;342:1470–2. 14.Stegmayr B, Asplund K, Wester PO. Trends in incidence, case fatality rate, and severity of stroke in Northern Sweden, 1985–1991. Stroke 1994;25:1738–45. 15.Brown RD, Whisnant JP, Sicks JD, O’Fallon WM, Wiebers DO. Stroke incidence, prevalence, and survival: secular trends in Rochester, Minnesota, through 1989. Stroke 1996;27: 373–80. 16.Jamrozik K, Broadhurst RJ, Lai N, Hankey GJ, Burvill PW, Anderson CS. Trends in the incidence, severity, and short-term outcome of stroke in Perth, Western Australia. Stroke 1999;30:2105–11. 17.Thorvaldsen P, Davidsen M, Brønnum-Hansen H, Schroll M, for the Danish MONICA Study Group. Stable stroke occurrence despite incidence reduction in an aging population: stroke trends in the Danish Monitoring trends and determinants in cardiovascular disease (MONICA) population. Stroke 1999;30:2529–34. 18.Gillum RF, Sempos CT. The end of the long-term decline in stroke mortality in the United States. Stroke 1997; 28: 1527–9. 19.Sarti C, Tuomilehto J, Sivenious J et al. Declining trends in incidence, case-fatality and mortality of stroke in three geographical areas of Finland during 1983–1989: results from the FINMONICA stroke register. J Clin Epidemiol 1994; 47: 1259–69. 20.Ryglewicz D, Polakowska M, Lechowicz W et al. Stroke mortality rates in Poland did not decline between 1984 and 1992. Stroke 1997;28:752–7. 21.Malmgren R, Bamford J, Warlow C, Sandercock P, Slattery J. Projecting the number of patients with first ever strokes and patients newly-handicapped by stroke in England and Wales. BMJ 1989;298:656–60. 22.Reddy KS, Yusuf S. Emerging epidemic of cardiovascular disease in developing countries. Circulation 1998;97: 596–601.

23.Hankey GJ, Warlow CP. Treatment and secondary prevention of stroke: evidence, costs, and effects on individuals and populations. Lancet 1999;354:1457–63. 24.Stroke Unit Trialists’ Collaboration. Collaborative systematic review of the randomised trials of organised in-patient (stroke unit) care after stroke. BMJ 1997;314:1151–9. 25.The European Ad Hoc Consensus Group. European strategies for early intervention in stroke. A report of an ad hoc consensus group meeting. Cerebrovasc Dis 1996;6:315–24. 26.Organising Committees. Asia Pacific Consensus Forum on Stroke Management. Stroke 1998;29:1730–6. 27.Young J. Is stroke better managed in the community? BMJ 1994;309:1356–8. 28.Langhorne P, Dennis MS and collaborators. Services for reducing duration of hospital care for acute stroke patients (Cochrane Review). In: The Cochrane Library, 1, 1999. Oxford: Update Software. 29.Beech R, Rudd AG, Tilling K, Wolfe CDA. Economic consequences of early inpatient discharge to community-based rehabilitation for stroke in an inner-London teaching hospital. Stroke 1999;30:729–35. 30.Anderson C, Ni Mhurchu C, Rubenach S, Clark M, Spencer C, Winsor A. Home or hospital for stroke rehabilitation? Results of a randomised controlled trial. II: Cost minimisation analysis at 6 months. Stroke 2000;31:1032–7. 31.Intercollegiate Working Party for Stroke. National Clinical Guidelines for Stroke. London, England: Royal College of Physicians; 2000. 32.National Institute of Neurological Disorders and Stroke r-tPA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med 1995;333:1581–7. 33.Wardlaw J, del Zoppo G, Yamaguchi T. Thrombolysis for acute ischemic stroke (Cochrane review). Cochrane Database Syst Rev. 2000;(2):CD000213. 34.Ringleb PA, Schellinger PD, Schranz C, Hacke W. Thrombolytic therapy within 3 to 6 hours after onset of ischemic stroke: useful or harmful? Stroke 2002;33:1437–41. 35.International Stroke Trial Collaborative Group. The International Stroke Trial (IST): a randomized trial of aspirin, subcutaneous heparin, both, or neither among 19 435 patients with acute ischemic stroke. Lancet 1997;349: 1569–14. 36.CAST (Chinese Acute Stroke Trial) Collaborative Group. CAST: randomized placebo-controlled trial of early aspirin use in 20 000 patients with acute ischemic stroke. Lancet 1997;349:1641–9. 37.Adams HP, Brott TG, Furlan AJ et al. Guidelines for thrombolytic therapy for acute stroke: a supplement for the guidelines for the management of patients with acute ischemic stroke: a statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association. Circulation 1996;94:1167–74. 38.Morgenstern LB, Staub L, Chan E et al. Improving delivery of acute stroke therapy: the TLL Temple Foundation Stroke Project. Stroke 2002;33:160–6. 39.Caplan LR, Moht JP, Kistler JP, Koroshetz W. Should thrombolytic therapy be the first-line treatment of acute ischemic stroke? Thrombolysis: not a panacea for ischemic stroke. N Engl J Med 1997;337:1309–10.

849


40.Gubitz G, Counsell C, Sandercock P. Anticoagulants for acute ischemic stroke (Cochrane Review). In The Cochrane Library. Issue 4. Oxford, UK: Update Software, 2001. 41.Adams HP. Emergent use of anticoagulation for treatment of patients with ischemic stroke. Stroke 2002;33:856–61. 42.Rose G. Sick individuals and sick populations. Int J Epidemiol 1985;14:32–8. 43.Hankey GJ. Smoking and risk of stroke. J Cardiovasc Risk 1999;5:207–11. 44.Wolf PA. Epidemiology and risk factor management. In: Welch KMA, Caplan LR, Reis DJ, Siesjö BK, Weir B, eds. Primer on cerebrovascular diseases. San Diego: Academic Press, 1997. 45.The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993; 329:977–86. 46.MacMahon S, Peto R, Cutler J, Collins R, Sorlie Pea. Blood pressure, stroke, and coronary heart disease. Part 1, prolonged differences in blood pressure: prospective observational studies corrected for regression dilution bias. Lancet 1990;335:765–74. 47.Collins R, Peto R, MacMahon S et al. Blood pressure, stroke, coronary heart disease. Part 2, short-term reductions in blood pressure: overview of randomised drug trials in their epidemiological context. Lancet 1990;336:827–38. 48.Prospective Studies Collaboration. Cholesterol, diastolic blood pressure and stroke; 13 000 strokes in 45 000 people in 45 prospective cohorts. Lancet 1995;346:1647–53. 49.Eastern Stroke and Coronary Heart Disease Collaborative Research Group. Blood pressure, cholesterol and stroke in Eastern Asia. Lancet 1998;352:1801–7. 50.Asia Pacific Cohort Studies Collaboration. Determinants of Cardiovascular Disease in the Asia Pacific region: Protocol for a Collaborative Overview of Cohort Studies. CVD Prevention 1999;2:281–9. 51.MacMahon S, Rodgers A. The effects of antihypertensive treatment on vascular disease: reappraisal of the evidence in 1994. J Vasc Med Biology 1993;4:265–71. 52.Insua JT, Sacks HS, Lau TS et al. Drug treatment of hypertension in the elderly: a meta-analysis. Ann Int Med 1994;121: 355–62. 53.Staessen JA, Gasowski J, Wang JG et al. Risks of untreated and treated isolated systolic hypertension in the elderly: metaanalysis of outcome trials. Lancet 2000;355:865–72. 54.The Heart Outcomes Prevention Evaluation (HOPE) Study Investigators. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. N Engl J Med 2000;342:145–53. 55.Bosch J, Yusuf S, Progue J et al. on behalf of the HOPE Investigators. Use of ramipril in preventing stroke: double blind randomized trial. BMJ 2002;324:1–5. 56.Rodgers A, Neal B, MacMahon S. The effects of blood pressure lowering in cerebrovascular disease: an overview of randomized controlled trials. Neurol Rev Int 1997; 12:12–15. 57.PROGRESS Collaborative Group. Randomized trial of a perindopril-based blood-pressure-lowering regimen among

850

6105 individuals with previous stroke or transient ischemic attack. Lancet 2001;358:1033–41. 58.Mustard JF, Rowsell HC, Smythe HA, Senyi A, Murphy EA. The effect of sulfinpyrazone on platelet economy and thrombus formation in rabbits. Blood 1967;29:859–66. 59.Weiss HJ, Aledort LM. Impaired platelet/connective-tissue reaction in man after aspirin ingestion. Lancet 1967;ii:495–7. 60.Emmons PR, Harrison MJ, Honour AJ, Mitchell JR. Effect of a pyrimido pyrimidine derivative on thrombus formation in the rabbit. Nature 1965;208:255. 61.Acheson J, Danta G, Hutchinson EC. Controlled trial of dipyridamole in cerebral vascular disease. BMJ 1969;1:614–5. 62.The Canadian Cooperative Study Group. A randomized trial of aspirin and sulfinpyrazone in threatened stroke. N Engl J Med 1978;299:53–9. 63.Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy in various categories of patients. BMJ 1994;308:81–106. [Published erratum appears BMJ 1994;308:1540.] 64.Alga A, van Gijn J. Cumulative meta-analysis of aspirin efficacy after cerebral ischaemia of arterial origin. J Neurol Neurosurg Psychiatry 1999;66:255. 65.European Atrial Fibrillation Trial Study Group. Secondary prevention in nonrheumatic atrial fibrillation after transient ischemic attack or minor stroke. Lancet 1993; 342: 1255–62. 66.Barnett HJM, Kaste M, Meldrum HE, Eliasziw M. Aspirin dose in stroke prevention: beautiful hypotheses slain by ugly facts. Stroke 1996;27:588–92. 67.Hart RG, Harrison MJG. Aspirin wars: the optimal dose of aspirin to prevent stroke. Stroke 1996;27:585–7. 68.Patrono C, Roth GJ. Aspirin in ischemic cerebrovascular disease: how strong is the case for a different dosing regimen? Stroke 1996;27:756–60. 69.CAPRIE Steering Committee. A randomized, blinded, trial of clopidogrel versus aspirin in patients at risk of ischemic events (CAPRIE). Lancet 1996;348:1329–39. 70.Hankey GJ, Sudlow CLM, Dunbabin DW. Thienopyridines or aspirin to prevent stroke and other serious vascular events in patients at high risk of vascular disease. Stroke 2000; 31:1779–84. 71.Hankey GJ. Clopidogrel and thrombotic thrombocytopenic purpura. Lancet 2000;356:269–270. 72.Diener HC, Cunha L, Forbes C et al. European Stroke Prevention Study 2. Dipyridamole and acetylsalicylic acid in the secondary prevention of stroke. J Neurol Sci 1996;143:1–13. 73.Wilterdink JL, Easton JD. Dipyridamole plus aspirin in cerebrovascular disease. Arch Neurol 1999;56:1087–92. 74.De Schryver ELLM on behalf of the European. Australian Stroke Prevention in Reversible Ischaemia Trial (ESPRIT) Group. Design of ESPRIT: an international randomized trial for secondary prevention after non-disabling cerebral ischaemia of arterial origin. Cerebrovasc Dis 2000;10: 147–50. 75.Kannel WB, Abbott RD, Savage DD, McNamara PM. Coronary heart disease and atrial fibrillation: the Framingham Study. Am Heart J 1983;106:389–96. 76.Cage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for


predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA 2001;285:2864–70. 77.Petersen P, Godtfredsen J, Boysen G. Placebo-controlled, randomized trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation. The Copenhagen AFASAK study. Lancet 1989;i:175. 78.The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators. The effect of low-dose warfarin on the risk of stroke in patients with nonrheumatic atrial fibrillation. N Engl J Med 1990;323:1505–11. 79.Connolly SJ, Laupacis A, Gent M et al. Canadian atrial fibrillation anticoagulation (CAFA) study. J Am Coll Cardiol 1991; 18:349–55. 80.Stroke Prevention in Atrial Fibrillation Investigators. Stroke Prevention in Atrial Fibrillation Study: final results. Circulation 1991;84:527–39. 81.Ezekowitz MD, Bridgers SL, James KE et al. Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. N Engl J Med 1992;327:1406–12. (Erratum, N Engl J Med 1993;328:148.) 82.EAFT (European Atrial Fibrillation Trial) Study Group. Secondary prevention in non-rheumatic atrial fibrillation after transient ischemic attack or minor stroke. Lancet 1993; 342:1255–62. 83.Stroke Prevention in Atrial Fibrillation Investigators. Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation: Stroke Prevention in Atrial Fibrillation II Study. Lancet 1994;343:687–91. 84.Stroke Prevention in Atrial Fibrillation Investigators. Adjusteddose warfarin versus low-intensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation: Stroke Prevention in Atrial Fibrillation III randomized clinical trial. Lancet 1996;348:633–8. 85.The Atrial Fibrillation Investigators. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomised controlled trials. Arch Intern Med 1994;154:1449–57. (Published erratum appears in Arch Intern Med 1994;154:2254.) 86.Hart RG, Benavente O, McBride R, Pearce LA. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis. Ann Intern Med 1999;131: 492–501. 87.Laupacis A, Boysen G, Connolly S et al. The efficacy of aspirin; in patients with atrial fibrillation: analysis of pooled data from 3 randomised trials. Arch Intern Med 1997;157: 1237–40. 88.Ezekowitz MD, Levine JA. Preventing stroke in patients with atrial fibrillation. JAMA 1999;281:1830–5. 89.The Stroke Prevention in Reversible Ischemia Trial (SPIRIT) Study Group. A randomised trial of anticoagulants versus aspirin after cerebral ischemia of presumed arterial origin. Ann Neurol 1997;42:857–65. 90.Cerebral Embolism Task Force. Cardiogenic brain embolism: the second report of the Cerebral Embolism Task Force. Arch Neurol 1989;86:727. 91.Turpie AGG, Robinson JG, Doyle DJ et al. Comparison of highdose with low-dose subcutaneous heparin to prevent left ventricular mural thrombosis in patients with acute transmural anterior myocardial infarction. N Engl J Med 1989;320:352–7.

92.Mohr JP, Thompson JL, Lazar RM et al. for the WarfarinAspirin Recurrent Stroke Study Group. A comparison of warfarin and aspirin for the prevention of recurrent ischemic stroke. N Engl J Med 2001;345:1444–51. 93.Chimowitz MI, Kokkinos J, Strong J et al. The WarfarinAspirin Symptomatic Intracranial Disease Study. Neurology 1995;45:1488–51. 94.Neaton JD, Blackburn H, Jacobs D. Serum cholesterol level and mortality findings for men screened in the Multiple Risk Factor Intervention Trial. Arch Int Med 1992;152: 1490–500. 95.Yano K, Reed DM, MacLean CJ. Serum cholesterol and hemorrhagic stroke in the Honolulu Heart Program. Stroke 1989;20:1460–5. 96.Iribarren C, Jacobs DR, Sadler M, Claxton AJ, Sidney S. Low total serum cholesterol and intracerebral hemorrhagic stroke: is the association confined to elderly men? The Kaiser Permanente Medical Care Program. Stroke 1996;27:1993–8. 97.How Lin C, Shimzu Y, Kato H et al. Cerebrovascular diseases in a fixed population of Hiroshima and Nagasaki, with special reference to relationship between type and risk factors. Stroke 1984;15:653–60. 98.Iso H, Jacobs DRJ, Wentworth D, Neaton JD, Cohen JD. Serum cholesterol levels and six-year mortality from stroke in 350 977 men screened for the Multiple Risk Factor Intervention Trial. N Engl J Med 1989;320:904–10. 99.Eastern Stroke and Coronary Heart Disease Collaborative Research Group. Blood pressure, cholesterol, and stroke in eastern Asia. Lancet 1998;352:1801–7. 100.The Heart Protection Study. Presented at the Scientific Sessions of the American Heart Association; 2001. Available at: http://www.ctsu.ox.ac.uk/~hps/. 101.Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383–9. 102.Sacks FM, Pfeffer MA, Moyé LA et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med 1996;335:1001–9. 103.The Long-term Intervention with Pravastatin in Ischemic Disease (LIPID) Study Group. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. N Engl J Med 1998;339:1349–57. 104.Shepherd J, Cobbe SM, Ford I et al. Prevention of coronary heart disease with pravastatin in men with hypercholesterolaemia. N Engl J Med 1995;333:1301–7. 105.Downs JR, Clearfield M, Wies S et al. Priimary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels: results of AFCAPS/Tex CAPS. JAMA 1998;279:1615–22. 106.Sandercock P. Statins for stroke prevention? Lancet 2001; 357:1548–9. 107.Fields WS, Maslenikov V, Meyer JS et al. Joint study of extracranial arterial occlusion. V. Progress report of prognosis following surgery or nonsurgical treatment for transient cerebral ischemic attacks and cervical carotid artery lesions. JAMA 1970;211:1993–2003.

851


108.Shaw DA, Venables GS, Cartlidge NEF, Bates D, Dickinson PH. Carotid endarterectomy in patients with transient cerebral ischaemia. J Neurol Sci 1984;64:45–53. 109.North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N Engl J Med 1991;325:445–3. 110.European Carotid Surgery Trialists’ Collaborative Group. MRC European Carotid Surgery Trial: interim results for symptomatic patients with severe (70–99%) or with mild (0–29%) carotid stenosis. Lancet 1991;337:1235–43. 111.Mayberg MR, Wilson SE, Yatsu F et al. Carotid endarterectomy and prevention of cerebral ischemia in symptomatic carotid stenosis. JAMA 1991;266:3289–94. 112.European Carotid Surgery Trialists’ Collaborative Group. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC European Carotid Surgery Trial (ECST). Lancet 2000;351: 1379–87. 113.North American Symptomatic Carotid Endarterectomy Trial Collaborators. Benefit of carotid endarterectomy in patients with symptomatic, moderate or severe stenosis. New Engl J Med 1998;339:1415–25. 114.Barnett HJM, Warlow CP. Carotid endarterectomy and the measurement of stenosis. Stroke 1993;24:1281–4. 115.Hankey GJ, Warlow CP, Molyneuz AJ. Complications of cerebral angiography for patients with mild carotid territory ischaemia being considered for carotid endarterectomy. J Neurol Neurosurg Psychiatr 1990;53:542–8. 116.Eliasziw M, Rankin RN, Fox AJ, Haynes RB, Barnett HJM, for the North American Symptomatic Carotid Endarterectomy Trial

852

(NASCET) Group. Accuracy and prognostic consequences of ultrasonography in identifying severe carotid artery stenosis. Stroke 1995;26:1747–52. 117.CAVATAS Investigators. Endovascular versus surgical treatment in patients with carotid stenosis in the Carotid and Vertebral Artery Transluminal Angioplasty Study: a randomised trial. Lancet 2001;357:1729–37. 118.Hobson RW II, Weiss DG, Fields WS et al. Efficacy of carotid endarterectomy for asymptomatic carotid stenosis. N Engl J Med 1993;328:221–7. 119.Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Endarterectomy for asymptomatic carotid artery stenosis. JAMA 1995;273:1421–8. 120.Hennerici M, Hulsbomer HB, Hefter H, Lemmerts D, Rautenberg W. Natural history of asymptomatic extracranial arterial disease: results of a long-term prospective study. Brain 1987;110:777–91. 121.Norris JW, Zhu CZ, Bornstein NM, Chambers BR. Vascular risks of asymptomatic carotid stenosis. Stroke 1991;22: 1485–90. 122.Lernfelt B, Forsberg M, Blomstrand C, Mellström D, Volkmann R. Cerebral atherosclerosis as predictor of stroke and mortality in representative elderly population. Stroke 2002;33:224–9. 123.Rothwell P. Carotid endarterectomy and prevention of stroke in the very elderly. Lancet 2001;357:1142–3. 124.Halliday AW, Thomas D, Mansfield A. The Asymptomatic Carotid Surgery Trial (ACST) rationale and design. Eur J Vasc Surg 1994;8:703–10.

60

Heart disease and pregnancy Samuel C Siu, Jack M Colman

Introduction Women with heart disease comprise approximately 1% of the population in obstetric referral centers,1 though they are less frequently seen in general obstetric practice. Current data on pregnancy outcomes of women with heart disease are primarily from studies that were retrospective, focused on a particular cardiac lesion, or examined populations managed at a single institution or from an earlier era. Treatment recommendations are usually based on institutional experience or extrapolation from observational studies. Pregnancy in most women with heart disease has a favorable outcome for both mother and fetus. With the exception of patients with Eisenmenger syndrome, pulmonary vascular obstructive disease and Marfan syndrome with aortopathy, maternal death during pregnancy in women with heart disease is rare.1–5 However, pregnant women with heart disease do remain at risk for other complications, including heart failure, arrhythmia and stroke. Women with congenital heart disease comprise the majority of pregnant women with heart disease seen at referral centers.1,5 The next largest group includes women with rheumatic heart disease. Other important conditions less frequently encountered include peripartum dilated cardiomyopathy, hypertrophic cardiomyopathy and coronary artery disease. Gestational hypertension, arising de novo or superimposed on pre-existing hypertension, is responsible for around 15% of all maternal mortality and considerable morbidity.6

Cardiovascular physiology and pregnancy Pregnancy is characterized by hormonally mediated changes in blood volume, red cell mass and heart rate, resulting in a 50% increase in cardiac output during the antepartum period.7 Increases in LV end-diastolic dimension and volume are present by 14 weeks’ gestation and reach maximum levels early in the third trimester.8–12 Preload- and afterloadadjusted indices of contractility remain in the normal range during the antepartum period.13 LV mass increases during pregnancy as a consequence of increased LV wall thickness. Gestational hormones, circulating prostaglandins and the low-resistance vascular bed in the placenta result in decreases in peripheral vascular resistance and blood pressure (BP). These physiological changes are exacerbated in

multifetal gestations. During labor and delivery there are additional increases in cardiac output and oxygen consumption.7,14 Immediately following delivery, relief of caval compression and autotransfusion from the emptied uterus result in a transient increase in cardiac output. Most of the hemodynamic changes of pregnancy have resolved by the second postpartum week, but complete return to normal does not occur until 6 months after delivery.15,16 LV diastolic dimension, volume and mass also return to preconception levels by the sixth postpartum month.

Outcomes associated with specific cardiac lesions Congenital heart lesions Left to right shunts The effect of increase in cardiac output on the volumeloaded right ventricle in atrial septal defect (ASD), or the left ventricle in ventricular septal defect (VSD) and patent ductus arteriosus (PDA), is counterbalanced by a decrease in peripheral vascular resistance. Consequently, the increase in volume overload is attenuated. In the absence of pulmonary hypertension, pregnancy, labor and delivery are well tolerated.1,4,5,17 Grade B2 However, arrhythmias, ventricular dysfunction and progression of pulmonary hypertension may occur, especially when the shunt is large or when there is pre-existing elevation of pulmonary artery pressure. Infrequently, particularly in ASD, paradoxical embolization may be encountered if systemic vasodilatation and/or elevation of pulmonary resistance promote transient right to left shunting. Left ventricular outflow tract obstruction When aortic stenosis (AS) complicates pregnancy it is usually due to congenital bicuspid aortic valve, which may also be associated with aortic coarctation and/or ascending aortopathy. Other causes of left ventricular (LV) outflow tract obstruction at, below and above the valve have similar hemodynamic consequences. Women with symptomatic aortic stenosis should delay pregnancy until after surgical correction.18 Grade B4 However, the absence of symptoms is not sufficient assurance that pregnancy will be well 853


tolerated. In a pregnant woman with severe AS the limited ability to augment cardiac output may result in abnormal elevation of LV systolic and filling pressures, which in turn precipitate or exacerbate heart failure or ischemia. In addition the non-compliant, hypertrophied ventricle is sensitive to falls in preload. The consequent exaggerated drop in cardiac output may lead to hypotension. In a compilation of many small retrospective series, 65 patients were followed through 106 pregnancies with a maternal mortality of 11% and a perinatal mortality of 4%.19 In 25 of the same 65 pregnancies managed more recently there was no maternal mortality, although maternal functional deterioration occurred in 20% of pregnancies.19 Women with moderate or severe aortic stenosis continue to be at increased risk for pulmonary edema or arrhythmia during pregnancy.1,5,19 Grade B2 Intrapartum palliation by balloon valvuloplasty may be helpful in selected cases. Grade C5 In the absence of prosthetic dysfunction or residual aortic stenosis, patients with bioprosthetic aortic valves usually tolerate pregnancy well. Although it had been stated that pregnancy might accelerate the rate of degeneration of bioprosthetic or homograft valves, recent studies have shown that this is not the case.20 Grade B2 A study of 14 pregnancies in women who underwent pulmonary autograft aortic valve replacement (Ross procedure) reported favorable maternal and fetal outcomes except in one woman who developed postpartum LV dysfunction.21 Grade B4 Pregnancy in a woman with a mechanical valve prosthesis carries an increased risk of valve thrombosis as a result of the hypercoagulable state. The magnitude of this increased risk (3–14%) is greater if subcutaneous unfractionated heparin rather than warfarin is used as the anticoagulant agent; this may be the result of inadequate dosing, insufficient monitoring or reduced efficacy of heparin18,22 Grade B4 (see anticoagulation section under antepartum management). Coarctation of the aorta Maternal mortality with uncorrected coarctation was 3% in an early series; the risk was higher in the presence of associated cardiac defects, aortopathy or long-standing hypertension; aortic rupture accounted for 8 of the 14 reported deaths and occurred in the third trimester as well as in the postpartum period.23 Grade B4 The results of recent studies have been more encouraging. In 182 pregnancies reported in three recent studies, the only maternal death occurred in a woman with Turner syndrome who had previously undergone coarctation repair.1,5,24 Grade B2 The management of hypertension in uncorrected coarctation is particularly problematic in pregnancy, because satisfactory control of upper body hypertension may lead to excessive hypotension below the coarctation site, thereby compromising the fetus. Intrauterine growth restriction and premature 854

labor and delivery are more common. Following coarctation repair, the risk of dissection and rupture is reduced but not eliminated. Pregnant women with repaired coarctation are at increased risk for pregnancy-induced-hypertension, probably as a result of residual abnormalities in aortic compliance.1,5,24 Grade B2 Pulmonary stenosis Mild pulmonic stenosis, or pulmonic stenosis that has been alleviated by valvuloplasty or surgery, is well tolerated during pregnancy and fetal outcome is favorable.1,5 Grade B2 Although a woman with severe pulmonic stenosis may be asymptomatic, the increased hemodynamic load of pregnancy may precipitate right heart failure or atrial arrhythmias; such a patient should be considered for correction prior to pregnancy. Grade C5 Even during pregnancy, balloon valvuloplasty may be feasible if symptoms of pulmonary stenosis progress. Cyanotic heart disease: unrepaired and repaired In uncorrected or palliated pregnant patients with cyanotic congenital heart disease, such as tetralogy of Fallot, single ventricle etc., the usual pregnancy-associated fall in systemic vascular resistance and rise in cardiac output exacerbate right to left shunting, leading to increased maternal hypoxemia and cyanosis. A study examining the outcomes of 96 pregnancies in 44 women with a variety of cyanotic congenital heart defects reported a high rate of maternal cardiac events (32%, including one death), prematurity (37%) and a low livebirth rate (43%).25 Grade B4 The lowest livebirth rate (12%) was observed in mothers with arterial oxygen saturation 85%. Tetralogy of Fallot is the most common form of cyanotic congenital heart disease. Pregnancy risk is low in women who have had successful correction of tetralogy.1,4,5 Grade B2 However, residua and sequelae, such as residual shunt, right ventricular outflow tract obstruction, arrhythmias, pulmonary regurgitation, right ventricular systolic dysfunction, pulmonary hypertension (owing to the effects of a previous palliative shunt) or LV dysfunction (owing to previous volume overload), increase the likelihood of pregnancy complications and require independent consideration. Atrial repair (that is Mustard or Senning procedure) was developed for the surgical correction of complete transposition of the great arteries. The anatomic right ventricle supports the systemic circulation. Late adult complications following atrial repair include sinus node dysfunction, atrial arrhythmias and dysfunction of the systemic ventricle. In 43 pregnancies in 31 women described in recent reports, there was one late maternal death.26,27 Grade B4 There was a 14% incidence of maternal heart failure, arrhythmias or cardiac deterioration. There have been no studies of

Heart disease and pregnancy

pregnancy outcome in women who received the current repair of choice for complete transposition, the arterial switch procedure. However, in the absence of ventricular dysfunction, coronary obstruction or other important residua or sequelae, a good outcome is expected. Grade C5 The Fontan operation eliminates cyanosis and volume overload of the functioning systemic ventricle, but patients have a limited ability to increase cardiac output. In a review of 33 pregnancies in 21 women who were doing well after the Fontan operation there were 15 (45%) term pregnancies with no maternal mortality, although two women had cardiac complications and the incidence of first-trimester miscarriage was high (39%).28 Grade B4 As the 10 year survival following the Fontan operation is only 60–80% it is important that information regarding long-term maternal prognosis be discussed during preconception counseling. Marfan syndrome Life-threatening aortic complications of Marfan syndrome are due to medial aortopathy resulting in dilatation, dissection and valvular regurgitation. Risk is increased in pregnancy owing to hemodynamic stress and perhaps hormonal effects. Although older case reports suggested a very high mortality risk in the range of 30%, a subsequent study found an overall maternal mortality of 1% and fetal mortality of 22%.29 A prospective study of 45 pregnancies in 21 patients reported no increase in obstetric complications or significant change in aortic root size in patients with normal aortic roots. Importantly, in the eight patients with a dilated aortic root (40 mm) or prior aortic root surgery, three of nine pregnancies were complicated by either aortic dissection (2) or rapid aortic dilatation (1).30 Thus, patients with aortic root involvement should receive preconception counseling emphasizing their risk, and in early pregnancy should be offered termination. In contrast, women with little cardiovascular involvement and with normal aortic root diameter may tolerate pregnancy well, though there remains a possibility of dissection even without prior evidence of aortopathy. Grade B4 The likelihood of aortic dilatation increases with increasing maternal age, so that advice to complete families at a younger age is appropriate. Serial echocardiography should be used to identify progressive aortic root dilatation during pregnancy and for 6 months postpartum; prophylactic blockers should be administered.31 Grade C5 Congenitally corrected transposition of the great arteries Many adult patients will have had surgical interventions, primarily VSD closure and relief of pulmonic stenosis, sometimes requiring a valved conduit from the LV to the pulmonary artery. Potential problems in pregnancy include dysfunction of the systemic right ventricle and/or increased

systemic AV valve regurgitation, with heart failure, atrial arrhythmias and AV block. In two recent reports on 41 patients there were 105 pregnancies, with 73% live births and no maternal mortality, although seven patients developed heart failure, endocarditis, stroke or myocardial infarction.32,33 Grade B4 Eisenmenger syndrome and pulmonary vascular obstructive disease Maternal mortality in Eisenmenger syndrome is approximately 30% in each pregnancy.34 Grade B4 The majority of complications occur at term and during the first postpartum week. Preconception counseling should stress the extreme pregnancy-associated risks. Termination should always be offered to such patients, as should sterilization. The vasodilatation associated with pregnancy will increase the magnitude of right to left shunting in patients with Eisenmenger syndrome, resulting in worsening of maternal cyanosis and adverse effects on fetal outcome. Spontaneous abortion is common, intrauterine growth restriction is seen in 30% of pregnancies, and preterm labor is frequent. The high perinatal mortality rate (28%) is due mainly to prematurity. Pregnancy may accelerate the progression of pulmonary vascular disease by increasing the risk of in-situ thrombosis and/or thromboembolism; other mechanisms may be operative as well. Grade C5 A recent review of outcome of 125 pregnancies in patients with Eisenmenger syndrome, primary pulmonary hypertension and secondary pulmonary hypertension reported poor outcomes in all three groups.35 Grade B4 The maternal mortality observed in the various groups was 36%, 30% and 56%, respectively. The overall neonatal mortality was 13%. Mitral valve prolapse Isolated mitral valve prolapse has an excellent outcome in pregnancy36–38 Grade B4 and affects management only as a possible indication for endocarditis prophylaxis, or if severe mitral regurgitation has led to symptomatic deterioration or left ventricular dysfunction. Rheumatic heart disease Mitral stenosis is the most common rheumatic valvular lesion encountered during pregnancy. The hypervolemia and tachycardia associated with pregnancy exacerbate the impact of mitral valve obstruction. The resultant elevation in left atrial pressure increases the likelihood of atrial fibrillation. Thus, even patients with mild to moderate mitral stenosis who are asymptomatic prior to pregnancy, may develop atrial fibrillation and heart failure during the ante- and peripartum periods. Atrial fibrillation is a frequent precipitant of heart failure in pregnant patients with mitral stenosis, 855


owing primarily to uncontrolled ventricular rates; equivalent tachycardia of any cause may produce the same detrimental effect. Earlier studies examining a pregnant population comprised predominantly of women with rheumatic mitral disease showed that mortality rate increased with worsening antenatal maternal functional class.3 More recent studies found no mortality, but described substantial morbidity from heart failure and arrhythmia.1,5 The risk for complications was especially high in those women with moderate or severe mitral stenosis.1,5,39 Grade B2 Percutaneous mitral valvuloplasty should be considered in patients with functional class III or IV symptoms despite optimal medical therapy and hospitalization.40–42 Grade B4 Pregnant women whose dominant lesion is rheumatic aortic stenosis have a similar outcome to those with congenital aortic stenosis. Aortic or mitral regurgitation is generally well tolerated during pregnancy even if severe, although deterioration in maternal functional class has been observed. Peripartum cardiomyopathy Peripartum cardiomyopathy is a form of idiopathic dilated cardiomyopathy diagnosed by otherwise unexplained LV systolic dysfunction, confirmed echocardiographically, presenting during the last antepartum month or in the first 5 postpartum months.43 It usually manifests as heart failure, although arrhythmias and embolic events also occur. Many affected women will show improvement in functional status and ventricular function postpartum, but others may have persistent or progressive dysfunction. The relapse rate during subsequent pregnancies is substantial in women with evidence of persisting cardiac enlargement or LV dysfunction. However, pregnancy may not be risk free even in those with recovery of systolic function, as subclinical abnormalities may persist.44 In a recently published multicenter survey examining the outcomes of 60 pregnancies in women with peripartum cardiomyopathy diagnosed during a prior pregnancy, 44% of women with LV ejection fraction 0·50 developed symptoms of congestive heart failure during subsequent pregnancies, with an associated mortality rate of 19%. In contrast, symptoms of congestive heart failure developed in 21% of women with LV ejection fraction 0·50, and none of this group died (Figure 60.1).45 Grade B4 Hypertensive disorders in pregnancy Hypertensive disorders of pregnancy are the second most common cause of maternal mortality, accounting for 15% of all obstetric deaths.6 They also predispose to other complications, such as placental abruption, stroke, disseminated intravascular coagulation, renal and/or hepatic failure and congestive heart failure.46 The fetus is at increased risk for intrauterine growth restriction, prematurity and intrauterine death. 856

60

Maternal complication rates during first subsequent pregnancy (% women) LVEF 50% LVEF 50%

40

20

0 CHF

Death Number of predictors

Figure 60.1 The frequency of maternal heart failure (CHF) and death during the first subsequent pregnancy in women with peripartum cardiomyopathy as stratified by preserved (LVEF 50%) versus reduced left ventricular ejection fraction (LVEF 50%). In the group with preserved left ventricular ejection fraction there were no deaths during the first subsequent pregnancy. (From Elkayam et al.45)

Several guidelines and consensus documents have been developed which attempt to standardize definitions and criteria for diagnosis.46–49 Unfortunately, terminology and definitions vary slightly, but importantly in these documents this compromises clarity. The recommendations of the Canadian Hypertension Society and the Society of Obstetricians and Gynaecologists of Canada define hypertension in pregnancy as pre-existing hypertension (elsewhere called chronic hypertension, renal hypertension, underlying hypertension, essential hypertension or secondary hypertension), gestational hypertension without proteinuria and other adverse conditions (elsewhere called pregnancy-induced hypertension, transient hypertension) or gestational hypertension with proteinuria or other adverse conditions (elsewhere called preeclampsia, eclampsia, HELLP [hemolysis, elevated l iver enzymes, low platelets] syndrome, gestational proteinuric hypertension).46,48,49 The recent American guidelines set criteria for the diagnosis of hypertension in pregnancy as seated systolic blood pressure 140 and/or diastolic blood pressure 90 mmHg (using Korotkoff phase V (disappearance of sound) to determine diastolic pressure).6 The BP elevation should be noted on repeated measurements. Proteinuria in pregnancy is significant when there is 0·3 g protein in a 24 hour urine collection. Severe hypertension is defined as a systolic blood pressure 160 and/or a diastolic blood pressure 110 mmHg and severe proteinuria as a 24 hour urine protein excretion 2 g. Gestational hypertension may be superimposed on pre-existing hypertension. In the absence of proteinuria and other adverse conditions, gestational hypertension that resolves postpartum is called transient hypertension or benign gestational hypertension, whereas if it persists postpartum it is understood as pregnancy-induced unmasking of pre-existing (or chronic) hypertension.


The pathophysiology of gestational hypertension with proteinuria or other adverse conditions (pre-eclampsia) differs from other forms of hypertension. As a result of placental dysfunction, the normal cardiovascular adaptations to pregnancy (increased plasma volume and decreased peripheral resistance) do not occur. There is reduced perfusion to the placenta, liver, kidneys and brain. It is thought that endothelial dysfunction, perhaps a consequence of the decreased perfusion, results in excessive vasoactive toxins, which produce most if not all the manifestations of gestational hypertension. Thus, hypertension is but one effect, not a cause, of the clinical syndrome. Certain adverse conditions are associated with worse outcomes. Frontal headache, severe nausea and vomiting, visual disturbances, chest pain and shortness of breath, and right upper quadrant pain are significant symptoms. The components of the HELLP syndrome may be found, either individually or combined. Other adverse maternal manifestations are severe hypertension, severe proteinuria, hypoalbuminemia (18 g/l) oliguria, pulmonary edema and convulsions. Fetal compromise may be revealed by oligohydramnios, absent or reversed umbilical artery end-diastolic flow, and abnormalities in fetal biophysical profile. Intrauterine growth restriction, prematurity and placental abruption are the serious adverse fetoplacental consequences. Hypertrophic cardiomyopathy Hypertrophic cardiomyopathy is a disorder with distinct genetic abnormalities and a diverse clinical profile. Morphologically there is unexplained ventricular hypertrophy, which is usually asymmetric and predominantly involves the interventricular septum. Obstruction to left ventricular outflow is a common but not invariable feature. Diastolic function abnormalities are important determinants of the clinical manifestations. In patients with dynamic left ventricular outflow tract obstruction, increases in preload tend to reduce the severity of obstruction, whereas increases in contractility and decreases in afterload tend to worsen it. Diastolic dysfunction magnifies the preload dependence on cardiac output. As a consequence, pregnancy may be associated with worsening symptoms. Maternal outcome is often good, although at least two deaths have been reported, and serious complications (congestive heart failure, supraventricular tachyarrhythmias, ventricular tachycardia, syncope) may occur, especially in women who already have symptoms prior to pregnancy, and in those with substantial LV diastolic and/or systolic dysfunction.50,51 Grade B4 Fetal outcomes are good. Blockers may be used, as in the non-pregnant state. Dual chamber pacing may be of value in patients with symptoms refractory to medical therapy. Grade C5 The role of septal alcohol ablation or surgical myectomy during pregnancy has not been defined.

Coronary artery disease Symptomatic coronary artery disease (CAD) is an uncommon accompaniment of pregnancy. Major predisposing factors for atherosclerotic CAD include long-standing diabetes mellitus,52 familial hypercholesterolemia and tobacco abuse. In addition, some non-atherosclerotic causes of CAD, though also rare, are more frequent in or aggravated by pregnancy, such as coronary artery dissection, coronary artery embolism, vascular complications of vasoactive obstetric therapies (for example ergot derivatives, prostaglandins), and collagen vascular diseases. The long-term residua of childhood Kawasaki disease include coronary artery stenoses and aneurysms, which may become symptomatic during pregnancy. Cocaine abuse must be considered in any young person with an acute coronary event.53 Diagnosis of infarction may be confounded peripartum because of the release of CK-MB isoenzyme from the uterus.54 Because of the possibility of unusual causes of ischemia and infarction, coronary angiography should be considered early. Fetal exposure to radiation from routine coronary angiography is 5 mGy (500 mrad).55 Adverse fetal consequences of this amount of radiation are extremely small or negligible55 Grade B4 , and pregnancy should not be seen as a contraindication to a clinically necessary study.56 Thromobolysis is not contraindicated,53 but the diagnosis of coronary thrombosis as opposed to other causes of coronary occlusion cannot be routinely assumed; if immediately available, coronary angiography with the option of primary angioplasty can immediately confirm the diagnosis and thus increase the likelihood of providing appropriate therapy.

Management Risk stratification and counseling Risk stratification and counseling of women with heart disease is best accomplished prior to conception. The data required for risk stratification can be readily acquired from a thorough cardiovascular history and examination, 12-lead electrocardiogram (ECG) and transthoracic ECG. In patients with cyanosis, arterial oxygen saturation should be assessed by percutaneous oximetry. In counseling, the following areas should be considered: the underlying cardiac lesion; maternal functional status; the possibility of further palliative or corrective surgery; additional associated risk factors; maternal life expectancy and ability to care for a child; and the risk of congenital heart disease in the offspring. Defining the underlying cardiac lesion is an important part of stratifying risk and determining management. Review of prior catheterization and operative reports may be necessary to clarify the diagnosis. The nature of residua and sequelae should be clarified, especially ventricular function, pulmonary pressure, severity of obstructive lesions, persistence of shunts and the presence of hypoxemia. 857


Almost all patients can be stratified into low-, intermediateor high-risk groups (Box 60.1). Maternal functional status is widely used as a predictor of outcome, and most often defined by NYHA functional class. In a study of 482 pregnancies in women with congenital heart disease, cardiovascular morbidity was less (8% v 30%) and livebirth rate higher (80% v 68%) in mothers with NYHA functional class I than in the others.2 Grade B2 In two studies examining the outcomes of 851 pregnancies, poor functional status (NYHA II) or cyanosis, left ventricular systolic dysfunction, left heart obstruction, and history of cardiac events prior to pregnancy (arrhythmia, stroke or pulmonary edema) were independent predictors of maternal cardiac complications.1,5 Grade B2 Poor maternal functional class or cyanosis was also predictive of adverse neonatal events. Box 60.1 Maternal cardiac lesion and cardiac risk during pregnancy Low risk Left to right shunts Repaired lesions without residual cardiac dysfunction Isolated mitral valve prolapse without significant regurgitation Bicuspid aortic valve without stenosis Mild–moderate pulmonic stenosis Valvular regurgitation with normal ventricular systolic function Intermediate risk Unrepaired or palliated cyanotic congenital heart disease Uncorrected coarctation of the aorta Mitral stenosis Mild or moderate aortic stenosis Mechanical prosthetic valves Severe pulmonic stenosis Moderate to severe systemic ventricular dysfunction Systemic right ventricle or single ventricle Hypertrophic cardiomyopathy History of peripartum cardiomyopathy with no residual ventricular dysfunction Symptomatic arrhythmia Pre-existing hypertension Gestational hypertension without pre-eclampsia Stable coronary artery disease High risk New York Heart Association (NYHA) class III or IV symptoms Significant pulmonary hypertension with or without right to left shunt Marfan syndrome with aortic root or major valvular involvement Severe aortic stenosis History of peripartum cardiomyopathy with residual ventricular dysfunction Recent myocardial infarction or unstable angina Gestational hypertension with proteinuria or other adverse conditions (pre-eclampsia)

In a recently published prospective study the four independent risk factors described above (poor functional status or cyanosis, left ventricular systolic dysfunction, left heart obstruction, and history of cardiac events prior to 858

pregnancy) were incorporated into a revised risk index. The risk of a cardiac event (cardiac death, stroke, pulmonary edema or arrhythmia) during pregnancy increased with the number of predictors present during the antepartum evaluation. This risk index was derived using two thirds of the study sample and then validated in the remaining pregnancies. For each risk category there was excellent agreement between the expected and the observed rate of events in both the derivation and the validation set (Figure 60.2).1 Grade B2 The above-mentioned predictors were also predictive of the combined likelihood of cardiac event (as defined above), deterioration of maternal functional class during pregnancy, or need for an urgent cardiac intervention during the ante- or postpartum periods (Figure 60.3). This index, together with lesion-specific risk estimates, aids the risk stratification of women with heart disease at preconception counseling, and also during pregnancy.

80

Primary cardiac event rate (% pregnancies)

60

40

20

0 0 Predicted

1 Number of predictors Derivation n = 385

>1 Validation n = 214

Figure 60.2 The frequency of maternal cardiac complications (pulmonary edema, cardiac arrhythmia, stroke or cardiac death) as predicted by the risk index and observed in the derivation and validation groups, as a function of the number of cardiac predictors (n denotes number of pregnancies). (From Siu et al.1)

Further palliative or corrective surgery. Both maternal and fetal outcomes are improved by surgery to correct cyanosis, which should be undertaken prior to conception when possible.4 Grade B2 Similarly, patients with symptomatic obstructive lesions should undergo intervention prior to pregnancy.18 Grade B4 A systematic overview of the outcome of cardiovascular surgery performed during pregnancy reported a maternal and fetal mortality of 6% and 30%, respectively.57 Grade B4 Planning for valve replacement prior to pregnancy requires the need for ongoing anticoagulation with a mechanical valve to be weighed against the likelihood of early reoperation if a tissue valve is used. For aortic stenosis, an attractive alternative is


80

with congenital heart disease who reach reproductive age should be offered genetic counseling so that they are fully informed of the mode of inheritance and recurrence risk, as well as the prenatal diagnosis options available to them. Grade C5 Preventative strategies to decrease the incidence of congenital defects, such as preconception use of multivitamins containing folic acid, can be discussed at the time of such counseling.59 Grade A1d

Primary or secondary cardiac event rate (% pregnancies)

60

40

20

Antepartum management 0

0 Predicted

1 Number of predictors

>1

Derivation n = 385

Validation n = 214

Figure 60.3 The frequency of any primary or secondary cardiac events (deterioration in maternal functional class, need for urgent cardiac interventions during the ante- or postpartum periods, pulmonary edema, cardiac arrhythmia, stroke, death) as predicted by the risk index and observed in the derivation and validation groups, as a function of the number of cardiac predictors (n denotes number of pregnancies). (From Siu et al.1)

the pulmonary autograft. The lack of ideal choices once severe valve disease is present argues for completing families earlier, before the age-dependent progression of valve disease necessitates valve replacement surgery. Additional associated risk factors that may complicate pregnancy include a history of arrhythmia or heart failure, prosthetic valves and conduits, anticoagulant therapy, and the use of teratogenic drugs such as warfarin or angiotensinconverting enzyme inhibitors. Maternal life expectancy and ability to care for a child. A patient with limited physical capacity or with a condition that may result in premature death should be advised of her potential inability to look after her child. Women whose condition imparts a high likelihood of fetal complications, such as those with cyanosis or on anticoagulants, must be apprised of these added risks. The risk of recurrence of congenital heart disease in offspring should be addressed in the context of a 0·4–0·6% risk in the general population. The risk with an affected first degree relative increases about 10-fold. A multicenter study examining the offspring of patients with major congenital heart defects who survived cardiac surgery described an overall recurrence rate of 4% in the offspring.58 Grade B2 Obstructive left heart lesions have a higher recurrence rate. Certain conditions, such as Marfan syndrome and the 22q11 deletion syndromes, are autosomal dominant, conferring a 50% risk of recurrence in an offspring. Patients

Pregnant women with heart disease may be at particular risk for one or more of congestive heart failure; arrhythmias; or thrombosis, emboli and adverse effects of anticoagulants. Pre-existing and/or gestational hypertension/pre-eclampsia may also require management. When ventricular dysfunction is a concern, limitation of activity is helpful, and in severely affected women with class III or IV symptoms hospital admission by mid second trimester may be advisable. Gestational hypertension, hyperthyroidism, infection and anemia should be identified early and treated vigorously. For patients with functionally significant mitral stenosis, adrenergic blockers should be used to control heart rate. Digitalis, although a time-honored treatment for the same purpose, is often ineffective in blunting pregnancy-induced tachycardia. We also offer empiric therapy with adrenergic blockers to patients with coarctation and to Marfan patients. Grade C5

Arrhythmias Arrhythmias in the form of premature atrial or ventricular beats are common in normal pregnancy; sustained tachyarrhythmias have also been reported. In those with preexisting arrhythmias, pregnancy may exacerbate their frequency or hemodynamic severity. Pharmacologic treatment is usually reserved for patients with severe symptoms, or when sustained episodes are poorly tolerated in the presence of ventricular hypertrophy, ventricular dysfunction or valvular obstruction. Sustained tachyarrhythmias, such as atrial flutter or atrial fibrillation, should be treated promptly, avoiding teratogenic antiarrhythmic drugs. Digoxin and adrenergic blockers are the antiarrhythmic drugs of choice, in view of their known safety profiles.60 Grade B4 Quinidine, adenosine, sotalol and lidocaine are also “safe”, but published data on their use during pregnancy are more limited. Amiodarone is more problematic and standard texts classify it as contraindicated in pregnancy, although there are case reports describing its successful use with careful follow up; it is not teratogenic, but may impair neonatal thyroid function.61,62 Grade B4 Electrical cardioversion is safe in pregnancy. A recent report of 44 pregnancies in women with implantable cardioverter 859


defibrillators reported favorable maternal and fetal outcomes.63 Grade B4

Anticoagulation When a pregnant woman with a mechanical heart valve requires anticoagulation heparin and warfarin are used, but controversy continues as to which is better at different stages of pregnancy. Oral anticoagulation with warfarin is better accepted by patients and is effective. However, warfarin embryopathy may be produced during organogenesis, and fetal intracranial bleeding can occur throughout pregnancy. A recent study of 58 pregnancies reported that a daily warfarin dose of 5 mg was associated with no cases of embryopathy.22 Grade B4 Fetal intracranial hemorrhage during vaginal delivery is a risk with warfarin unless it has been stopped at least 2 weeks prior to labor. Adjusteddose subcutaneous heparin has no teratogenic effects as the drug does not cross the placenta, but heparin may cause maternal thrombocytopenia and osteoporosis. Claims of inadequate effectiveness of heparin in patients with mechanical heart valves have been countered by arguments that inadequate doses were used. In a systematic overview of prior studies examining the relationship between anticoagulation regimen and pregnancy outcome in women with prosthetic heart valves, the overall pooled maternal mortality was 2·9%.64 Grade B4 The use of oral anticoagulants throughout pregnancy was associated with the lowest rate of valve thrombosis/systemic embolism (4%). The use of unfractionated heparin between 6 and 12 weeks’ gestation only was associated with an increased risk of valve thrombosis (9%). Recent practice guidelines have favored the use of either warfarin plus low-dose aspirin during the entire pregnancy, or warfarin substituted by heparin only during the peak teratogenic period (6–12 weeks’ gestation).18 Grade B4 Low molecular weight heparin is easier to administer and has been suggested as an alternative to adjusted-dose unfractionated heparin. The adjunctive use of low-dose acetyl salicylic acid with heparin should also be considered.65,66 Grade C5 ASA in low dose is safe for the fetus, even at term67 Grade A1a , although high maternal doses may promote premature duct closure. Clinical trials are needed to better define the optimal anticoagulation strategy.

Eisenmenger syndrome If a woman with Eisenmenger syndrome does not accept counseling to terminate, or presents late in pregnancy, meticulous antepartum management is necessary, including early hospitalization, supplemental oxygen, and possibly empiric anticoagulation. Grade C5 The efficacy of nitric oxide therapy in these patients has yet to be demonstrated. 860

Hypertension in pregnancy Mild pre-existing hypertension may not require pharmacotherapy in pregnancy, as fetal outcomes are unaffected, maternal blood pressure falls lower than baseline during the first 20 weeks of gestation, and excessive lowering of maternal blood pressure may compromise placental perfusion, with no proven maternal benefit.6 Therapy should be initiated or reinstituted if moderate–severe hypertension develops (systolic BP 150–160; diastolic BP 100–110; or both), or there is target organ damage. It is not clear whether treatment of mild–moderate pre-existing (chronic) hypertension reduces the risk of developing superimposed gestational hypertension with proteinuria (pre-eclampsia). If treatment is indicated, drug therapy established as safe includes methyldopa, hydralazine, labetalol and other -blockers68 Grade B4 , and nifedipine.69 Grade B4 Diuretics are indicated for the management of volume overload in renal failure or heart failure, may be used as adjuncts in the management of pre-existing (chronic) hypertension, but should be avoided in gestational hypertension (pre-eclampsia), which is a volume contracted state.6,70 Grade A1c Angiotensinconverting enzyme inhibitors and angiotensin receptor blocking agents are contraindicated after the first trimester of pregnancy, and so should be stopped either before conception or in the first trimester as soon as pregnancy is diagnosed.71,72 Grade B4 Gestational hypertension with proteinuria (pre-eclampsia) is treated effectively only by delivery of the fetus and placenta. Delay in delivery to allow maturation of the fetus can often be accomplished if the syndrome is mild, the patient is under very close surveillance in a hospital or obstetric day unit, and pregnancy is terminated as soon as further benefit to the fetus is unlikely or maternal safety is compromised.6

Multidisciplinary approach and high-risk pregnancy units Women with heart disease who are at intermediate or high risk for complications should be managed in a high-risk pregnancy unit by a multidisciplinary team from obstetrics, cardiology, anesthesia and pediatrics (Box 60.2). Grade C5 When dealing with a complex problem the team should meet early in the pregnancy. At this time the nature of the cardiac lesion, the anticipated effects of pregnancy and potential problems should be explored. As it is often not possible for every member of the team to be at the patient’s bedside at a moment of crisis, it is helpful to develop and distribute widely a written management plan for foreseeable contingencies. Women with heart disease in the “low-risk” group can be managed in a community hospital setting. However, if there is doubt about the mother’s status or the risk, consultation at a regional referral center should be arranged.


Box 60.2 Management of pregnancy in women with heart disease All patients ● Define the lesion, the residua and the sequelae ● Assess functional status ● Determine predictors of risk: general and lesion specific ● Eliminate teratogens ● Arrange genetic counseling when relevant ● Consider consultation with a regional center ● Assess need for endocarditis prophylaxis during labor and delivery Intermediate and high-risk patients ● Arrange management at a regional center for high-risk pregnancy ● Consider antepartum interventions to reduce pregnancy risk ● Engage a multidisciplinary team, as appropriate ● Consider a multidisciplinary case conference ● Develop and disseminate a management plan ● Anticipate vaginal delivery in almost all cases, unless there are obstetric contraindications ● Consider early epidural anesthesia ● Modify labor and delivery to reduce cardiac work ● Plan postpartum monitoring

Labor and delivery Vaginal delivery is recommended, with very few exceptions. The only cardiac indications for cesarean section are aortic dissection, Marfan syndrome with dilated aortic root, and failure to switch from warfarin to heparin at least 2 weeks prior to labor. Grade C5 Preterm induction is rarely indicated, but once fetal lung maturity is assured a planned induction and delivery in high-risk situations will ensure the availability of appropriate staff and equipment. Although there is no consensus on the use of invasive hemodynamic monitoring during labor and delivery, we commonly utilize intra-arterial monitoring and often central venous pressure monitoring as well in cases where there are concerns about the interpretation and deleterious effects of a sudden drop in systemic blood pressure (for example in patients with severe aortic stenosis, pulmonary hypertension, or more than moderate systemic ventricular systolic dysfunction). Grade C5 The need for an indwelling pulmonary artery catheter is contentious and has not been studied during pregnancy. Its value has not been shown in several studies of unselected patients with heart disease monitored through non-cardiac surgical procedures. It may be considered when the information sought is not available otherwise and warrants the risk of the procedure; risk may be increased because of complex anatomy, such as atrial baffles, or in the setting of pulmonary hypertension, because of possible pulmonary infarction or rupture. Heparin anticoagulation is discontinued at least 12 hours prior to induction, or reversed with protamine if spontaneous

labor develops, and can usually be resumed 6–12 hours postpartum. Endocarditis prophylaxis is initiated at the onset of active labor when indicated. The American Heart Association recommendations state that delivery by cesarean section and vaginal delivery in the absence of infection do not require endocarditis prophylaxis except, perhaps, in patients at high risk.73 Grade B2 Although many centers with extensive experience in caring for pregnant women with heart disease utilize endocarditis prophylaxis routinely, there is no evidence to support this common practice. Epidural anesthesia with adequate volume preloading is the technique of choice. Epidural fentanyl is particularly advantageous in cyanotic patients with shunt lesions as it does not lower peripheral vascular resistance. In the presence of a shunt, air and particulate filters should be placed in all intravenous lines. Grade C5 Labor is conducted in the left lateral decubitus position to attenuate hemodynamic fluctuations associated with contractions in the supine position. Forceps or vacuum extraction will shorten the latter part of the second stage of labor and reduce the need for maternal expulsive effort. As hemodynamics do not approach baseline for many days after delivery, those patients at intermediate or high risk may require monitoring for a minimum of 72 hours postpartum. Grade C5 Patients with Eisenmenger syndrome require longer close postpartum observation, as mortality risk persists for 7 days or more. References 1.Siu SC, Sermer M, Colman JM et al. Prospective multicenter study of pregnancy outcomes in women with heart disease. Circulation 2001;104:515–21. 2.Whittemore R, Hobbins J, Engle M. Pregnancy and its outcome in women with and without surgical treatment of congenital heart disease. Am J Cardiol 1982;50:641–51. 3.McFaul P, Dornan J, Lamki H, Boyle D. Pregnancy complicated by maternal heart disease. A review of 519 women. Br J Obstet Gynaecol 1988;95:861–7. 4.Shime J, Mocarski E, Hastings D, Webb G, McLaughlin P. Congenital heart disease in pregnancy: short- and long-term implications. Am J Obstet Gynecol 1987;156:313–22. 5.Siu SC, Sermer M, Harrison DA et al. Risk and predictors for pregnancy-related complications in women with heart disease. Circulation 1997;96:2789–94. 6.Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy. Am J Obstet Gynecol 2000;183:S1–S22. 7.Elkayam U, Gleicher N. Hemodynamics and cardiac function during normal pregnancy and the puerperium. In: Elkayam U, Gleicher N, eds. Cardiac Problems in Pregnancy: Diagnosis and Management of Maternal and Fetal Disease, 3rd edn. New York: Wiley-Liss, 1998. 8.Rubler S, Damani P, Pinto E. Cardiac size and performance during pregnancy estimated with echocardiography. Am J Cardiol 1977;40:534–40.

861


9.Katz R, Karliner J, Resnik R. Effects of a natural volume overload state (pregnancy) on left ventricular performance in normal human subjects. Circulation 1978;58:434–41. 10.Robson S, Dunlop W, Moore M, Hunter S. Combined Doppler and echocardiographic measurement of cardiac output: theory and application in pregnancy. Br J Obstet Gynaecol 1987;94: 1014–27. 11.Vered Z, Poler S, Gibson P, Wlody D, Perez J. Noninvasive detection of the morphologic and hemodynamic changes during normal pregnancy. Clin Cardiol 1991;14:327–34. 12.Sadaniantz A, Kocheril A, Emaus S, Garber C, Parisi A. Cardiovascular changes in pregnancy evaluated by two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr 1992;5:253–8. 13.Geva T, Mauer M, Striker L, Kirshon B, Pivarnik J. Effects of physiologic load of pregnancy on left ventricular contractility and remodeling. Am Heart J 1997;133:53–9. 14.Robson S, Dunlop W, Boys R, Hunter S. Cardiac output during labour. BMJ 1987;296:1169–72. 15.Robson S, Hunter S, Moore M, Dunlop W. Haemodynamic changes during the puerperium: a Doppler and M-mode echocardiographic study. Br J Obstet Gynaecol 1987;94: 1028–39. 16.Hunter S, Robson SC. Adaptation of the maternal heart in pregnancy. Br Heart J 1992;68:540–3. 17.Zuber M, Gautschi N, Oechslin E, Widmer V, Kiowski W, Jenni R. Outcome of pregnancy in women with congenital shunt lesions. Heart 1999;81:271–5. 18.Bonow RO, Carabello B, de Leon AC Jr et al. ACC/AHA Guidelines for the management of patients with valvular heart disease: A report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients with Valvular Heart Disease). J Am Coll Cardiol 1998;32:1486–588. 19.Lao T, Sermer M, MaGee L, Farine D, Colman J. Congenital aortic stenosis and pregnancy – a reappraisal. Am J Obstet Gynecol 1993;169:540–5. 20.North RA, Sadler L, Stewart AW, McCowan LM, Kerr AR, White HD. Long-term survival and valve-related complications in young women with cardiac valve replacements. Circulation 1999;99:2669–76. 21.Dore A, Somerville J. Pregnancy in patients with pulmonary autograft valve replacement. Eur Heart J 1997;18:1659–62. 22.Vitale N, De Feo M, De Santo LS, Pollice A, Tedesco N, Cotrufo M. Dose-dependent fetal complications of warfarin in pregnant women with mechanical heart valves. J Am Coll Cardiol 1999;33:1637–41. 23.Deal K, Wooley CF. Coarctation of the aorta and pregnancy. Ann Intern Med 1973;78:706–10. 24.Beauchesne LM, Connolly HM, Ammash NM, Warnes CA. Coarctation of the aorta: outcome of pregnancy. J Am Coll Cardiol 2001;38:1728–33. 25.Presbitero P, Somerville J, Stone S, Aruta E, Spiegelhalter D, Rabajoli F. Pregnancy in cyanotic congenital heart disease. Outcome of mother and fetus. Circulation 1994;89:2673–6. 26.Clarkson P, Wilson N, Neutze J, North R, Calder A, BarrattBoyes B. Outcome of pregnancy after the Mustard operation for transposition of the great arteries with intact ventricular septum. J Am Coll Cardiol 1994;24:190–3.

862

27.Genoni M, Jenni R, Hoerstrup SP, Vogt P, Turina M. Pregnancy after atrial repair for transposition of the great arteries. Heart 1999;81:276–7. 28.Canobbio M, Mair D, van der Velde M, Koos B. Pregnancy outcomes after the Fontan repair. J Am Coll Cardiol 1996;28:763–7. 29.Pyeritz R. Maternal and fetal complications of pregnancy in the Marfan syndrome. Am J Med 1981;71:784–90. 30.Rossiter J, Repke J, Morales A, Murphy E, Pyeritz R. A prospective longitudinal evaluation of pregnancy in the Marfan syndrome. Am J Obstet Gynecol 1995;173:1599–606. 31.Shores J, Berger K, Murphy E, Pyeritz R. Progression of aortic dilatation and the benefit of long-term -adrenergic blockade in Marfan’s syndrome. N Engl J Med 1994;330:1335–41. 32.Connolly HM, Grogan M, Warnes CA. Pregnancy among women with congenitally corrected transposition of great arteries. J Am Coll Cardiol 1999;33:1692–5. 33.Therrien J, Barnes I, Somerville J. Outcome of pregnancy in patients with congenitally corrected transposition of the great arteries. Am J Cardiol 1999;84:820–4. 34.Gleicher N, Midwall J, Hochberger D, Jaffin H. Eisenmenger’s syndrome and pregnancy. Obstet Gynecol Surv 1979;34: 721–41. 35.Weiss B, Zemp L, Seifert B, Hess O. Outcome of pulmonary vascular disease in pregnancy: a systematic overview from 1978 through 1996. J Am Coll Cardiol 1998;31:1650–7. 36.Rayburn WF, Fontana ME. Mitral valve prolapse and pregnancy. Am J Obstet Gynecol 1981;141:9–11. 37.Tang LC, Chan SY, Wong VC, Ma HK. Pregnancy in patients with mitral valve prolapse. Int J Gynecol Obstet 1985; 23:217–21. 38.Chia YT, Yeoh SC, Lim MC, Viegas OA, Ratnam SS. Pregnancy outcome and mitral valve prolapse. Asia Oceania J Obstet Gynaecol 1994;20:383–8. 39.Hameed A, Karaalp IS, Tummala PP et al. The effect of valvular heart disease on maternal and fetal outcome of pregnancy. J Am Coll Cardiol 2001;37:893–9. 40.Mangione JA, Lourenco RM, dos Santos ES et al. Long-term follow-up of pregnant women after percutaneous mitral valvuloplasty. Catheter Cardiovasc Interv 2000;50:413–17. 41.Desai DK, Adanlawo M, Naidoo DP, Moodley J, Kleinschmidt I. Mitral stenosis in pregnancy: a four-year experience at King Edward VIII Hospital, Durban, South Africa. Br J Obstet Gynaecol 2000;107:953–8. 42.de Souza JAM, Martinez EE, Ambrose JA et al. Percutaneous balloon mitral valvuloplasty in comparison with open mitral valve commissurotomy for mitral stenosis during pregnancy. J Am Coll Cardiol 2001;37:900–3. 43.Pearson GD, Veille JC, Rahimtoola S et al. Peripartum cardiomyopathy: National Heart, Lung, and Blood Institute and Office of Rare Diseases (National Institutes of Health) workshop recommendations and review. JAMA 2000;283:1183–8. 44.Lampert MB, Weinert L, Hibbard J, Korcarz C, Lindheimer M, Lang RM. Contractile reserve in patients with peripartum cardiomyopathy and recovered left ventricular function. Am J Obstet Gynecol 1997;176:189–95. 45.Elkayam U, Tummala PP, Rao K et al. Maternal and fetal outcomes of subsequent pregnancies in women with peripartum cardiomyopathy. N Engl J Med 2001;344:1567–71.


46.Helewa M, Burrows R, Smith J, Williams K, Brain P, Rabkin S. Report of the Canadian Hypertension Society Consensus Conference: 1. Definitions, evaluation and classification of hypertensive disorders in pregnancy. Can Med Assoc J 1997;157:715–25. 47.Brown MA, Hague WM, Higgins J et al. The detection, investigation and management of hypertension in pregnancy: full consensus statement. Aust NZ J Obstet Gynaecol 2000;40:139–55. 48.Moutquin J, Garner P, Burrows R et al. Report of the Canadian Hypertension Society Consensus Conference: 2. Nonpharmacologic management and prevention of hypertensive disorders in pregnancy. Can Med Assoc J 1997;157:907–19. 49.Rey E, LeLorier J, Burgess E, Lange I, Leduc L. Report of the Canadian Hypertension Society Consensus Conference: 3. Pharmacologic treatment of hypertensive disorders in pregnancy. Can Med Assoc J 1997;157:1245–54. 50.Elkayam U, Dave R. Hypertrophic cardiomyopathy and pregnancy. In: Elkayam U, Gleicher N, eds. Cardiac Problems in Pregnancy, 3rd edn. New York: Wiley-Liss, 1998. 51.Benitez RM. Hypertrophic cardiomyopathy and pregnancy: maternal and fetal outcomes. J Maternal–Fetal Invest 1996; 6:51–5. 52.Gordon MC, Landon MB, Boyle J, Stewart KS, Gabbe SG. Coronary artery disease in insulin-dependent diabetes mellitus of pregnancy (class H): a review of the literature. Obstet Gynecol Surv 1996;51:437–44. 53.Roth A, Elkayam U. Acute myocardial infarction associated with pregnancy. Ann Intern Med 1996;125:751–62. 54.Leiserowitz GS, Evans AT, Samuels SJ, Omand K, Kost GJ. Creatine kinase and its MB isoenzyme in the third trimester and the peripartum period. J Reprod Med 1992;37:910–6. 55.Wagner LK, Lester RG, Saldana LR. Exposure of the pregnant patient to diagnostic radiations: A guide to medical management, 2nd edn. Madison, WI: Medical Physics Publishing, 1997. 56.Colletti PM, Lee K. Cardiovascular imaging in the pregnant patient. In: Elkayam U, Gleicher N, eds. Cardiac problems in pregnancy, 3rd edn. New York: Wiley-Liss, 1998. 57.Weiss BM, von Segesser LK, Alon E, Seifert B, Turina MI. Outcome of cardiovascular surgery and pregnancy: a systematic review of the period 1984–1996. Am J Obstet Gynecol 1998;179:1643–53. 58.Burn J, Brennan P, Little J et al. Recurrence risks in offspring of adults with major heart defects: results from first cohort of British Collaborative study. Lancet 1998;351:311–16.

59.Czeizel A. Reduction of urinary tract and cardiovascular defects by periconceptional multivitamin supplementation. Am J Med Genet 1996;62:179–83. 60.Chow T, Galvin J, McGovern B. Antiarrhythmic drug therapy in pregnancy and lactation. Am J Cardiol 1998;82:58I–62I. 61.Magee LA, Downar E, Sermer M, Boulton BC, Allen LC, Koren G. Pregnancy outcome after gestational exposure to amiodarone in Canada. Am J Obstet Gynecol 1995;172: 1307–11. 62.Bartalena L, Bogazzi F, Braverman LE, Martino E. Effects of amiodarone administration during pregnancy on neonatal thyroid function and subsequent neurodevelopment. J Endocrinol Invest 2001;24:116–30. 63.Natale A, Davidson T, Geiger M, Newby K. Implantable cardioverter-defibrillators and pregnancy: a safe combination? Circulation 1997;96:2808–12. 64.Chan WS, Anand S, Ginsberg JS. Anticoagulation of pregnant women with mechanical heart valves: a systematic review of the literature. Arch Intern Med 2000;160:191–6. 65.Turpie AG, Gent M, Laupacis A et al. A comparison of aspirin with placebo in patients treated with warfarin after heart-valve replacement. N Engl J Med 1993;329:524–9. 66.Ginsberg JS, Greer I, Hirsh J. Use of antithrombotic agents during pregnancy. Chest 2001;119:122S–131S. 67.CLASP: a randomised trial of low-dose aspirin for the prevention and treatment of pre-eclampsia among 9364 pregnant women. CLASP (Collaborative Low-dose Aspirin Study in Pregnancy) Collaborative Group. Lancet 1994;343:619–29. 68.Magee LA, Ornstein MP, von Dadelszen P. Fortnightly review: management of hypertension in pregnancy. BMJ 1999; 318:1332–6. 69.Magee LA, Schick B, Donnenfeld AE et al. The safety of calcium channel blockers in human pregnancy: a prospective, multicenter cohort study. Am J Obstet Gynecol 1996;174:823–8. 70.Collins R, Yusuf S, Peto R. Overview of randomised trials of diuretics in pregnancy. BMJ (Clin Res) 1985;290:17–23. 71.Hanssens M, Keirse MJ, Vankelecom F, Van Assche FA. Fetal and neonatal effects of treatment with angiotensin-converting enzyme inhibitors in pregnancy. Obstet Gynecol 1991; 78:128–35. 72.Piper JM, Ray WA, Rosa FW. Pregnancy outcome following exposure to angiotensin-converting enzyme inhibitors. Obstet Gynecol 1992;80:429–32. 73.Dajani A, Taubert K, Wilson W et al. Prevention of bacterial endocarditis: recommendations by the American Heart Association. JAMA 1997;277:1794–801.

863

61

Venous thromboembolic disease Clive Kearon, Jeffrey S Ginsberg, Jack Hirsh

There are three main aspects to the management of venous thromboembolism (VTE): diagnosis, prevention, and treatment. Prior to focusing on these, relevant aspects of the pathogenesis and natural history of VTE will be reviewed. Pathogenesis of VTE Virchow is credited with identifying stasis, vessel wall injury, and hypercoagulability as the pathogenic triad responsible for thrombosis. This classification of risk factors for VTE remains valuable. ●

●

●

864

Venous stasis. The importance of venous stasis as a risk factor for VTE is demonstrated by the fact that most deep vein thrombosis (DVT), associated with stroke, affect the paralyzed leg,1 and most DVT which are associated with pregnancy affect the left leg,2 the iliac veins of which are prone to extrinsic compression by the pregnant uterus and the right common iliac artery. Vessel damage. Venous endothelial damage, usually as a consequence of accidental injury, manipulation during surgery (for example, hip replacement), or iatrogenic injury, is an important risk factor for VTE. Hence, 75% of proximal DVT complicating hip surgery occur in the operated leg,3 and thrombosis is common with indwelling venous catheters.4,5 Hypercoagulability. A complex balance of naturally occurring coagulation and fibrinolytic factors, and their inhibitors, serve to maintain blood fluidity and hemostasis. Inherited, or acquired, changes in this balance predispose to thrombosis. The most important inherited biochemical disorders associated with VTE are due to defects in the naturally occurring inhibitors of coagulation: deficiencies of antithrombin, protein C or protein S, and resistance to activated protein C caused by factor V Leiden. The first three of these disorders are rare in the normal population (combined prevalence of 1%), have a combined prevalence of about 5% in patients with a first episode of VTE,6 and are associated with a 10- to 40-fold increase in the risk of VTE.7 The factor V Leiden mutation is common, occurring in about 5% of Caucasians and about 20% of patients with a first episode of VTE (that is, fourfold increase in VTE risk).7,8

●

Hyperhomocysteinemia, owing to hereditary and acquired factors, is also a risk factor for VTE.9 Elevated levels of a number of coagulation factors (I, II, VIII, IX, XI) are associated with thrombosis in a “dose-dependent” manner.10–12 It is probable that such elevations are often inherited, with strong evidence for this with factor VIII.10 A mutation in the 3 untranslated region of the prothrombin gene (G20210A), which is associated with about 25% increase in prothrombin levels, occurs in about 2% of Caucasians and about 5% of those with a first episode of VTE (that is, about a 2·5-fold increase in risk).7,8,13 Prothrombotic abnormalities of the fibrinolytic system have questionable importance. Acquired hypercoagulable states include estrogen therapy, antiphospholipid antibodies (anticardiolipin antibodies and/or lupus anticoagulants), systemic lupus erythematosus, malignancy, combination chemotherapy, and surgery.14 Patients who develop immunologicallyrelated heparin-induced thrombocytopenia also have a very high risk of developing arterial and venous thromboembolism.15 Unlike the congenital abnormalities, acquired risk factors are often transient, which has important implications for the duration of anticoagulant prophylaxis and treatment. Combinations of risk factors and risk stratification. The risk of developing VTE depends on the prevalence and severity of risk factors (Box 61.1).14 Accordingly, by assessment of these factors, surgical patients can be categorized as having a low, moderate, or high risk of VTE (Table 61.1).

Prevalence and natural history of VTE VTE is rare before the age of 16 years, probably because the immature coagulation system is resistant to thrombosis. However, the risk of VTE increases exponentially with advancing age (1·9-fold per decade), rising from an annual incidence of approximately 30/100 000 at 40 years, to 90/100 000 at 60 years, and 260/100 000 at 80 years.14,16 Clinically important components of the natural history of VTE are summarized in Box 61.2.17

Venous thromboembolic disease

Box 61.1 Risk factors for venous thromboembolisma ● Patient factors ● Previous VTEb ● Age over 40 ● Pregnancy, purpureum ● Marked obesity ● Inherited hypercoagulable state ● Underlying condition and acquired factors ● Malignancyb ● Estrogen therapy ● Cancer chemotherapy ● Paralysisb ● Prolonged immobility ● Major traumab ● Lower limb injuriesb ● Heparin-induced thrombocytopenia ● Antiphospholipid antibodies ● Type of surgery ● Lower limb orthopedic surgeryb ● General anesthesia 30 min a Combinations of factors have at least an additive effect on the risk of VTE. b Common major risk factors for VTE. Abbreviation: VTE, venous thromboembolism

Diagnosis of VTE Objective testing for DVT and pulmonary embolism (PE) is important because clinical assessment alone is unreliable, failure to diagnose VTE is associated with a high mortality, and inappropriate anticoagulation needs to be avoided. Diagnosis of DVT Venography is the criterion standard for the diagnosis of DVT.18,19 However, because of its invasive nature, technical demands, costs, and the risks associated with contrast media, non-invasive tests have been developed, of which venous ultrasound imaging (VUI) and, more recently, D-dimer testing, are the most important (Box 61.3). Clinical assessment Although clinical assessment cannot unequivocally confirm or exclude DVT, clinical evaluation with empiric assessment or a structured clinical model (Table 61.2), can stratify patients as having a low ( 10% prevalence), moderate (25% prevalence) or high (60% prevalence) probability

Table 61.1 Risk stratification for postoperative VTE, frequency of VTE without prophylaxis, and recommended methods of prophylaxis. Venographic DVTa Calf (%) Low risk less than 40 years and uncomplicated surgery and no additional risk factors

Proximal (%)

Pulmonary embolism Symptomatic (%)

Fatal (%) 0·01

Recommended prophylaxis

2

0·4

0·2

Early mobilization

Moderate risk more than 40 years or prolonged/complicated surgery or additional “minor” risk factors

20

5

2

0·5

Low-dose UFHb LMWH (3000 U daily)c GC stockingsd

High risk major surgery for malignancy or previous VTE or knee/hip surgery or heparin-induced thrombocytopenia

50

15

5

2

LMWH (3000 U per day)c Warfarin (INR 2–3)f Adjusted-dose UFHg IPC devicese

a

Asymptomatic DVT detected by surveillance bilateral venography. Low-dose UFH: 5000 U of subcutaneous unfractionated heparin preoperatively, and twice or three times daily postoperatively. c LMWH: subcutaneous low molecular weight heparin; higher doses (for example, 4000 U once daily with a preoperative start [Europe], or 3000 U twice daily with a postoperative start [North America]) are used in high-risk patients; in moderate-risk patients 3000 U daily with a preoperative start is used. d GC stockings: graduated compression stockings, alone or in combination with pharmacologic methods. e IPC devices: intermittent pneumatic compression devices, alone or in combination with graduated compression stockings and/or pharmacologic methods. f Warfarin: usually started postoperatively and adjusted to achieve an International Normalization Ratio (INR) of 2·0–3·0. g Adjusted-dose UFH: preoperative start with an adjusted, three times daily, dose to raise the activated partial thromboplastin time to the upper limit of the normal range. Abbreviations: DVT, deep vein thrombosis; VTE, venous thromboembolism b

865


Box 61.2 Natural history of venous thromboembolism (VTE) ● Clinical factors can identify high-risk patients14 ● VTE usually starts in the calf veins112 ● Over 80% of symptomatic DVTs are proximal19,112 ● Two thirds of asymptomatic DVT detected postoperatively by screening venography are confined to the distal (calf) veins19 ● About 20% of symptomatic isolated calf DVTs subsequently extend to the proximal veins, usually within a week of presentation113 ● PE usually arises from proximal DVT114 ● The majority (70%) of patients with symptomatic proximal DVT have asymptomatic PE (high probability lung scans in 40%),115 and vice versa43,116 ● Only one quarter of patients with symptomatic PE have symptoms or signs of DVT117 ● About 50% of untreated symptomatic proximal DVTs are expected to cause symptomatic PE83 ● About 10% of symptomatic PE are rapidly fatal118 ● About 30% of untreated symptomatic non-fatal PE will have a fatal recurrence74,119 ● The risk of recurrent VTE is lower if risk factors are reversible than if there is no apparent, or a persistent, risk factor90–93,96 Abbreviations: DVT, deep vein thrombosis; PE, pulmonary embolism

of DVT.20 Grade A Such categorization is useful in guiding the performance and interpretation of objective testing.20–22

Venous ultrasound imaging VUI has a sensitivity for proximal DVT of about 97% and a specificity of 94% in symptomatic patients, which, on average, translates into a positive predictive value of 97% and a negative predictive value of 98% for proximal DVT.19 The essential component of VUI is assessment of venous compressibility of the common femoral and popliteal veins (down to the calf vein trifurcation), with application of ultrasound probe pressure.19 Grade A VUI is much more difficult to perform and less accurate in the calf (sensitivity of 70%).19 For these reasons, and because isolated calf DVT is uncommon and of limited importance, VUI of the calf veins is often not performed. Instead, if DVT cannot be excluded by a normal proximal VUI in combination with other results (for example, low clinical probability or normal D-dimer [see Box 61.3]), a follow up VUI is performed after 1 week to detect extending calf vein thrombosis (2% of patients).19 If the second VUI examination is normal, the risk of symptomatic VTE during the next 6 months is less than 2%.19

Box 61.3 Test results which effectively confirm or exclude DVT (deep vein thrombosis) ● Diagnostic for first DVT ● Venography: intraluminal filling defect ● Venous ultrasound: non-compressible proximal veins at two or more of the common femoral, popliteal, and calf trifurcation sites19 ● Excludes first DVT ● Venography: all deep veins seen, and no intraluminal filling defects18 ● D-dimer: normal test, which has a very high sensitivity (98%) and at least a moderate specificity (40%)27 ● Venous ultrasound or impedance plethysmography: normal and (a) low clinical suspicion for DVT at presentation,29,120 or (b) normal D-dimer test, which has a moderately high sensitivity (85%) and specificity (70%) at presentation,29,120 or (c) normal serial testing (venous ultrasound at 7 days; impedance plethysmography at 2 and 7 days) ● Low clinical suspicion for DVT at presentation and a normal D-dimer test, which has a moderately high sensitivity (85%) and specificity (70%) at presentation29 ● Diagnostic for recurrent DVT ● Venography: intraluminal filling defect ● Venous ultrasound: (a) a new non-compressible common femoral or popliteal vein segment,19 or (b) a 4.0 mm increase in diameter of the common femoral or popliteal vein compared to a previous test19,37a ● Impedance plethysmography: (a) conversion of a normal test to abnormal121,122a (b) an abnormal test 1 year after diagnosis19a ● Excludes recurrent DVT ● Venogram: all deep veins seen and no intraluminal filling defects ● Venous ultrasound or impedance plethysmography: normal, or 1 mm increase in diameter of the common femoral or popliteal veins on venous ultrasound compared to a previous test, and remains normal (no progression of venous ultrasound) at 2 and 7 days19,37,121,122 ● D-dimer: normal test, which has a very high sensitivity (98%) and at least a moderate specificity (40%)27 a If other evidence is not consistent with recurrent DVT (for example, venous ultrasound, impedance plethysmography, clinical assessment, D-dimer), venography should be considered. (Adapted from Kearon et al.19)

866


Table 61.2 Clinical model for determining clinical suspicion of deep vein thrombosis (Wells et al.20) Variables Active cancer (treatment ongoing or within previous 6 months or palliative) Paralysis, paresis, or recent plaster immobilization of the lower extremities Recently bedridden 3 days or major surgery within 4 weeks Localized tenderness along the distribution of the deep venous system Entire leg swollen Calf swelling 3 cm asymptomatic side (measured 10 cm below tibial tuberosity) Pitting edema confined to the symptomatic leg Dilated superficial veins (non-varicose) Alternative diagnosis as likely or greater than that of DVT

Points 1 1

non-diagnostic on its own (see Box 61.4).28–32 As less sensitive D-dimer assays are more specific (70%), they yield fewer false-positive results. Specificity of D-dimer decreases with aging33 and with comorbid illness such as cancer.34 Consequently, D-dimer testing has limited value as a diagnostic test for VTE in hospitalized patients and is unhelpful in the early postoperative period.

1

Diagnosis of DVT in pregnancy

1

Pregnant patients with suspected DVT can generally be managed in the same way as non-pregnant patients, although except for serial impedance plethysmography19,35,36 diagnostic approaches have not been evaluated in this population. Grade B In pregnant patients with normal noninvasive tests who have a high clinical suspicion of isolated iliac or calf DVT, venography (a complete study or a limited study using abdominal shielding, respectively) should be considered. Alternatively, normal magnetic resonance imaging, a normal D-dimer, or normal Doppler ultrasound imaging of the iliac veins, are likely to helpful for excluding DVT.

1 1 1 1 2

Total points Pretest probability calculated as follows: High 2 Moderate 1 or 2 Low 1

Note: In patients with symptoms in both legs, the more symptomatic leg is used.

The accuracy of VUI is substantially lower if its findings are discordant with the clinical assessment22,23 and/or if abnormalities are confined to short segments of the deep veins;24 in about 25% of such cases, the results of venography differ with VUI or reveal calf vein thrombosis. The accuracy of VUI in asymptomatic postoperative patients who have a high risk for DVT is poor with a sensitivity for proximal DVT of only about 62%,19 and such screening is not recommended in patients who have received prophylaxis.25 Grade A

Diagnosis of recurrent DVT Persistent abnormalities of the deep veins are common following DVT.19,37 Grade B Therefore, diagnosis of recurrent DVT requires evidence of new clot formation. Tests that can diagnose or exclude recurrent DVT are noted in Box 61.3.19,37 Magnetic resonance imaging (MRI) A recent small but rigorous study suggests that direct MRI of thrombus is very accurate for the diagnosis of DVT, including thrombosis in the calf and pelvis, and in asymptomatic or pregnant patients.38 The technique does not require radiographic contrast and has the potential to differentiate acute from old thrombus. Diagnosis of PE

D-dimer blood testing D-dimer is formed when cross-linked fibrin is broken down by plasmin and levels are usually elevated with DVT and/or PE. Normal levels can help to exclude VTE but elevated Ddimer levels are non-specific and have low positive predictive value.26 D-dimer tests differ markedly as diagnostic tests for VTE. Grade A A normal result with a very sensitive (98%) D-dimer assay excludes VTE on its own.26,27 However, very sensitive D-dimer tests have low specificities (40%), which limits their usefullness because of high falsepositive rates.27 In order to exclude DVT and/or PE, a normal result with a less sensitive D-dimer assay (85%) needs to be combined with either a low clinical probability or another objective test that has negative predictive but is also

Pulmonary angiography is the criterion standard for the diagnosis of PE, but has similar limitations as venography.39 As with suspected DVT, clinical assessment is useful for categorizing probability of PE (Table 61.3 and Box 61.4).40 Grade A Ventilation–perfusion lung scanning The usual initial investigation in patients with suspected PE is a ventilation–perfusion lung scan. A normal perfusion scan excludes PE,41 but is found in a minority (10–40%) of patients.33,42–44 Perfusion defects are non-specific; only about one third of patients with defects have PE.42,45 The probability that a perfusion defect is due to PE increases 867


Table 61.3 Model for determining a clinical suspicion of pulmonary embolism (Wells et al 123)

Current evidence suggests that helical CT can be interpreted as follows: ●

Variables

Points

Clinical signs and symptoms of deep vein thrombosis (minimum leg swelling and pain with palpation of the deep veins) An alternative diagnosis is less likely than pulmonary embolism Heart rate 100 beats/min Immobilization or surgery in the previous 4 weeks Previous deep vein thrombosis/pulmonary embolism Hemoptysis Malignancy (treatment ongoing or within previous 6 months or palliative)

3·0 ●

3·0 1·5 1·5 1·5 1·0 1·0

Total points Pretest probability calculated as follows: High 6 Moderate 2–6 Low 2

with size and number, and the presence of a normal ventilation scan (“mismatched” defect).42,45 A lung scan with mismatched segmental or larger perfusion defects is termed “high probability”.45 A single mismatched defect is associated with a prevalence of PE of about 80%.46 Three or more mismatched defects are associated with a prevalence of PE of 90%.46 Lung scan findings are highly age dependent with a relatively high proportion of normal scans and a low proportion of non-diagnostic scans in younger patients.33 Lung scanning and clinical assessment Clinical assessment of PE is complementary to ventilation– perfusion lung scanning; a moderate or high clinical suspicion in a patient with a high probability lung scan is diagnostic (prevalence of PE of 90%); however, a low clinical suspicion with a high probability defect requires further investigation because the prevalence of PE with these findings is only about 50%.42,45 Grade A The prevalence of PE with subsegmental, matched, perfusion defects (“low probability” scan) and a low clinical suspicion is expected to be less than 10% (see below).27,30,42

●

Intraluminal filling defects in lobar or main pulmonary arteries are likely to be associated with a probability of PE of at least 85%, similar to a high-probability ventilation– perfusion scan.48 Intraluminal defects confined to segmental or subsegmental pulmonary arteries are non-diagnostic, and patients with such findings require further testing.48 A normal helical CT substantially reduces the probability of PE but does not exclude the diagnosis (that is, similar to a “low probability” ventilation–perfusion scan).47,48

Although this statement is largely based on extrapolation from experience with patients who have non-diagnostic lung scans, patients with helical CT scans that are not diagnostic for PE can be managed as outlined in Box 61.4. Grade C Box 61.4 Test results which effectively confirm or exclude pulmonary embolism (PE) ● Diagnostic for PE ● Pulmonary angiography: intraluminal filling defect ● Helical CT: intraluminal filling defect in a lobar or main pulmonary artery47,48 ● Ventilation–perfusion scan: high probability scan and moderate/high clinical suspicion42,43 ● Diagnostic for DVT: with non-diagnostic ventilation– perfusion scan or helical CT124 ● Excludes PE ● Pulmonary angiogram: normal39 ● Perfusion scan: normal41 ● D-dimer: normal test, which has a very high sensitivity (98%) and at least a moderate specificity (40%)27 ● Non-diagnostic ventilation–perfusion scan, or normal helical CT, and normal proximal VUI and (a) low clinical suspicion for PE30,50 a (b) normal D-dimer test, which has at least a moderately high sensitivity (85%) and specificity (70%)30,32 a ● Low clinical suspicion for PE and normal D-dimer, which has at least a moderately high sensitivity (85%) and specificity (70%)30,32 a If serial VUI (venous ultrasound imaging) is performed it is expected to become abnormal in 1–2% of these patients and reduce the frequency of symptomatic VTE (venous thromboembolism) during 3 months of follow up from 1.5% to 0.5%. (Adapted from Kearon40)

Helical (spiral) computerized tomography (CT)

D-dimer testing

Helical CT following intravenous injection of radiographic contrast can be used to visualize the pulmonary arteries. Although widely used to diagnose PE, the technique has yet to be definitively evaluated for this purpose.47,48 Grade B

As previously discussed when considering diagnosis of DVT, a normal D-dimer result, alone27 or in combination with another negative test,30,32 can be used to exclude PE (Box 61.4). Grade A

868


Box 61.5 Management of patients with non-diagnostic non-invasive tests for PE ● Serial VUI of the proximal veins after 1 and 2 weeks Suitable for most such patients,30,44 although pulmonary angiography is preferred for the subgroups outlined below. This approach can be supplemented with bilateral venography (for patients that might otherwise be considered for pulmonary angiography).116 ● Pulmonary angiography preferred option if: ● segmental intraluminal filling defect on helical CTa,b ● subsegmental intraluminal filling defect and high clinical suspicion ● high probability ventilation–perfusion scan and low clinical suspicionb ● symptoms are severe, post-test probability is high but non-diagnostic, and PE needs to be excluded from the differential diagnosis ● serial testing is not feasible (for example, scheduled for surgery, geographic inaccessibility) a A segmental intraluminal filling defect with a high clinical suspicion is likely to have a positive predictive value of 85% and could be considered diagnostic for PE. b Ventilation–perfusion scanning can be performed after these findings have been obtained with helical CT; or helical CT may be performed after these findings have been obtained with ventilation–perfusion scanning;47,48 If the second test is also non-diagnostic for PE, serial ultrasounds may be reconsidered. Abbreviations: DVT, deep vein thrombosis; LMWH, low molecular weight heparin; PE, pulmonary embolism; VTE, venous thromboembolism; VUI, venous ultrasound imaging, (Adapted from Kearon40)

evolving proximal DVT, the forerunner of recurrent PE. If serial VUI for DVT (two additional tests a week apart) is negative, the subsequent risk of recurrent VTE during the next 3 months is less than 1%,30,44,51 which is similar to that after a normal pulmonary angiogram.39 As an additional precaution, patients who have had PE and/or DVT excluded should routinely be asked to return for re-evaluation if symptoms of PE and/or DVT persist or recur. Diagnosis of PE in pregnancy Pregnant patients with suspected PE can be managed similarly to non-pregnant patients, with the following modifications: Grade B ●

● ● ●

VUI can be performed first and lung scanning performed if there is no DVT; patients with unequivocal evidence of DVT can be presumed to have PE. The amount of radioisotope used for the perfusion scan can be reduced and the duration of scanning extended. If pulmonary angiography is performed, the brachial approach with abdominal screening is preferable. In the absence of safety data relating to helical CT in pregnancy, this is discouraged (if it is necessary, abdominal screening should be used). Consistent with other young patients who are suspected of having PE, a high proportion of pregnant patients have normal scans and a small proportion have high probability scans.33,52 These recommendations are based on a belief that the risk of inaccurate diagnosis of suspected PE during pregnancy is greater than the risk of radioactivity to the fetus.52,53

Tests for DVT in patients with suspected PE

Algorithms for the diagnosis of PE

Testing for DVT is an indirect way to diagnose PE (see Box 61.4).49 VUI of the proximal veins is the usual method, although bilateral ascending venography, or CT or MRI of the legs at the same time as examination of the pulmonary veins, can also be used. Negative tests for DVT do not rule out PE but they reduce the probability, and suggest that the short-term risk of recurrent PE is low.49 Because the prevalence of PE is expected to be less than 5% in patients with a non-diagnostic lung scan, a low clinical suspicion of PE, and a normal VUI of the proximal veins, it is reasonable to exclude PE with these findings.27,30,44,50 Grade B

Local availability of methods of testing and differences among patient presentations influence the diagnostic approach to PE. A number of prospectively validated algorithms have been published, which emphasize the use of different initial non-invasive tests in conjunction with ventilation–perfusion lung scanning including: ● ● ●

structured clinical assessment and serial VUI;44 sensitive D-dimer assay, empiric clinical assessment, and single bilateral VUI;27 clinical assessment, moderately sensitive D-dimer assay and serial VUI.30

Management of patients with non-diagnostic combinations of non-invasive tests for PE (Box 61.5)

Prevention of VTE (Box 61.6)

Patients with non-diagnostic test results for PE at presentation have, on average, a prevalence of PE of 20%.42,49 Two management approaches are reasonable in such patients. The first is the performance of pulmonary angiography, which is usually definitive. The second is the withholding of anticoagulants and performance of serial VUI to detect

In a non-randomized trial, oral anticoagulation was shown to prevent PE in patients with fractured hips, without causing an unacceptable increase in bleeding.54 Grade A Subsequently, low-dose unfractionated heparin was shown to reduce postoperative DVT and fatal PE by two thirds.55,56 Further studies have demonstrated that the efficacy of 869


Box 61.6 Prevention and treatment of venous thromboembolism ● Primary prophylaxis with anticoagulants and/or mechanical methods should be used in hospitalized patients who have a moderate or high risk of VTE. Grade A ● Acute VTE (DVT and/or PE) should be anticoagulated with: ● Heparin (unfractionated or LMWH) for a minimum of 4–5 days. Grade A If unfractionated heparin is used, a dose of at least (a) 30 000 U/ day or 18 U/kg/h by intravenous infusion; or (b) 33 000 U/day, by twice daily, subcutaneous, injection, should be administered. Grade A Dose of unfractionated heparin should be adjusted to achieve “therapeutic” APTT results. Grade C ● Oral anticoagulation for 3–6 months, Grade A with a dose adjusted to achieve an INR of 2.0–3.0. Grade A Prolonged unfractionated heparin or LMWH at therapeutic, or near therapeutic, doses is a satisfactory alternative. Grade A Anticoagulation should be continued for longer than 3 months in patients with a first episode of idiopathic VTE, Grade B and when VTE is associated with a risk factor, for as long as such factors are active. Grade C See Box 61.5 for abbreviations

low-dose unfractionated heparin can be improved either by increasing the dose so as to minimally prolong the activated partial thromboplastin time (APTT),57 or by combining its use with graduated compression stockings58 or intermittent pneumatic compression devices.59 Meta-analyses support that, at doses that are associated with equivalent efficacy (odds ratio 1·03) following general surgery, low molecular weight heparins (LMWH) are associated with less bleeding (odds ratio 0·68) than low-dose unfractionated heparin.60 Grade A Used at higher dose than for general surgery, LMWH is more effective (odds ratio 0·83) that unfractionated heparin following orthopedic surgery and is associated with a similar frequency of bleeding.60 An additional 3 or 4 weeks of LMWH after hospital discharge further reduces the frequency of symptomatic VTE after orthopedic surgery (from 3·3% to 1·3%61). Warfarin (target INR 2–3 for about 7 to 10 days) is less effective that LMWH at preventing DVT detected by venography soon after orthopedic surgery,62 but appears to be similarly effective at preventing symptomatic VTE over a 3 month period.62,63 Grade A There is evidence that aspirin reduces the risk of postoperative VTE by one third.64 Grade A A study of over 17 000 patients, mostly following hip fracture repair, confirmed these findings, including a reduction in fatal PE (0·27% v 0·65%) during the month following surgery.65 However, as warfarin and LMWH are expected to be more effective (at least a two thirds 870

reduction in VTE), aspirin alone is not recommended during the initial postoperative period.62 It may be a reasonable alternative to LMWH or warfarin for the weeks following hospital discharge, particularly if patients do not have additional risk factors for VTE. Grade B Recently, hirudin66 (a direct thrombin inhibitor) and fondaparinux67,67a,67b (the pentasaccharide of heparin that binds antithrombin) have been shown to be more effective than LMWH following major orthopaedic surgery; fondaparinux may cause marginally more bleeding. Grade A The evidence that short-term prophylaxis (for example, low-dose unfractionated heparin) prevents clinically important VTE in immobilized medical patients is less convincing, partly because it has been less extensively studied in this population, and because there is concern that medical patients remain at high risk of VTE after prophylaxis is stopped.62,68 In addition to augmenting the efficacy of pharmacologic methods of prophylaxis, mechanical methods are effective on their own. Graduated compression stockings prevent postoperative VTE in moderate-risk patients (risk reduction of 68%),62,69 and intermittent pneumatic compression devices prevent postoperative VTE in high-risk orthopedic patients.62,70,71 The relative efficacy of graduated compression stockings and intermittent pneumatic compression devices is uncertain. No difference in efficacy was evident in neurosurgical patients;72 however, pneumatic compression devices are expected to be superior to graduated compression stockings in high-risk patients.62 Mechanical methods of prophylaxis should be used in patients who have a moderate or high risk of VTE if anticoagulants are contraindicated (for example, neurosurgical patients).62 Grade A Because postoperative fatal PE is rarely preceded by symptomatic DVT,55 prophylaxis is the best way to prevent it. Use of primary prophylaxis is strongly supported by cost effectiveness analyses, which indicate that it reduces overall costs in addition to reducing morbidity.73 Treatment of VTE Heparin therapy In 1960, Barritt and Jordan established that heparin (1·5 days) and oral anticoagulants (2 weeks) reduced the risk of recurrent PE and associated death.74 Based on expert opinion, 10–14 days of heparin therapy, and 3 months of oral anticoagulation became widely adopted in clinical practice. It was subsequently shown that 4 or 5 days of intravenous heparin is as effective as 10 days of therapy for the initial treatment of VTE.75,76 Comparatively recently, the need for an initial course of heparin therapy was verified.77 Many trials have established that weight-adjusted LMWH (without laboratory monitoring) is as safe and effective as adjusted-dose unfractionated heparin for the treatment of acute VTE;78 it can be used to treat patients without hospital admission79 and need only be


injected subcutaneously once daily.80 Danaparoid, hirudin, and argatroban can be used to treat heparin induced thrombocytopenia, with or without associated thrombosis.81,82

●

Oral anticoagulant therapy

●

A randomized trial of patients with DVT, comparing 3 months of warfarin (International Normalization Ratio (INR) 3·0–4·0) with low-dose heparin after initial treatment with full-dose intravenous heparin, established the necessity for prolonged oral anticoagulation after initial heparin therapy.83 Grade A Prolonged high-dose subcutaneous heparin84 and, subsequently, LMWH (50–75% of acute treatment dose) was subsequently shown to be equally effective.85 Grade A In the 1970s it was recognized that, because of differences in the responsiveness of thromboplastins to oral anticoagulants, a prothrombin time ratio of 2·0 reflected a much more intense level of anticoagulation in North America than in Europe. This prompted a comparison of two intensities of warfarin therapy (corresponding to mean INRs of 2·1 and 3·2) for the treatment of DVT.86 This study found that the lower intensity of oral anticoagulation was as effective as the higher intensity but caused less bleeding. The trials showing that heparin therapy could be reduced to 5 days also showed that warfarin could be started at the same time as heparin.75,76 A recent series of small studies support starting warfarin with the expected daily dose rather than a loading dose (for example, 5 mg v 10 mg),87,88 and managing over-anticoagulation without bleeding (for example, INRs 6) with small oral rather than subcutaneous doses of vitamin K (for example, 1 mg).89 During the last decade, a series of well-designed studies have helped to define the optimal duration of anticoagulation. Their findings can be summarized as follows: ●

●

●

●

●

Shortening the duration of anticoagulation from 390,91 or 692 months to 490,91 or 692 weeks results in a doubling of the frequency of recurrent VTE during 190,91 to 292 years of follow up. Grade A Patients with VTE provoked by a transient risk factor have a lower (about one third) risk of recurrence than those with an unprovoked VTE or a persistent risk factor.90–94 Three months of anticoagulation is adequate treatment for VTE provoked by a transient risk factor; subsequent risk of recurrence is 3% per patient-year.90,91,94–96 Grade A Three months of anticoagulation may not be adequate treatment for an unprovoked (“idiopathic”) episode of VTE; subsequent early risk of recurrence has varied from 5% to 25% per patient-year.92,95,97,98 After 6 months of anticoagulation, recurrent DVT is at least as likely to affect the contralateral leg; this suggests that “systemic” rather than “local” (including inadequate treatment) factors are responsible for recurrences after 6 months of treatment.99

●

●

●

●

●

●

●

There is a persistently elevated risk of recurrent VTE after a first episode; this appears to be 5–12% per year after 6 or more months of treatment for an unprovoked episode.92,95,98 Oral anticoagulants targeted at an INR of 2·5 are very effective (risk reduction 90%) at preventing recurrent unprovoked VTE after the first 3 months of treatment.97,100 Grade A Indefinite anticoagulation is an option for patients with a first unprovoked VTE who have a low risk of bleeding. Grade B A second episode of VTE does not necessarily indicate a high risk of recurrence or the need for indefinite anticoagulation.97 Risk of bleeding on anticoagulants differs markedly among patients depending on the prevalence of risk factors (for example, advanced age; previous bleeding or stroke; renal failure; anemia; antiplatelet therapy; malignancy; poor anticoagulant control).101 Risk of recurrence is lower (about half) following an isolated calf (distal) DVT; this favors a shorter duration of treatment.92,95 Grade B Risk of recurrence is higher with antiphospholipid antibodies (anticardiolipin antibodies and/or lupus anticoagulants),97,102 homozygous factor V Leiden103 cancer93 and, probably, antithrombin deficiency; these favor a longer duration of treatment. Grade B Heterozygous factor V Leiden and the G20210A prothrombin gene mutations do not appear to be clinically important risk factors for recurrence.103 Grade B Other abnormalities, such as elevated levels of clotting factors VIII, IX, XI, and homocysteine, and deficiencies of protein C and protein S, may be risk factors for recurrence; they have uncertain implications for duration of treatment.

Thrombolytic therapy Systemic thrombolytic therapy accelerates the rate of resolution of DVT and PE at the cost of around a fourfold increase in frequency of major bleeding, and about a 10-fold increase in intracranial bleeding.104–107 This can be life-saving for PE with hemodynamic compromise.106,108 Grade A Thrombolytic therapy may reduce the risk of the prothrombotic syndrome following DVT but this does not appear to justify its associated risks104,105 Catheter-based treatments (that is, thrombolytic therapy or removal of thrombus) require further evaluation before they can be recommended. Inferior vena caval filters A recent randomized trial demonstrated that a filter, as an adjunct to anticoagulation, reduced the rate of PE (asymptomatic and symptomatic) from 4·5% to 1·0% during the 871


12 days following insertion, with a suggestion of fewer fatal episodes (0% v 2%).109 However, after 2 years, patients with a filter had a significantly higher rate of recurrent DVT (21% v 12%) and a non-statistically significant reduction in the frequency of PE (3% v 6%). This study supports the use of vena caval filters to prevent PE in patients with acute DVT and/or PE who cannot be anticoagulated (that is, they are bleeding), but does not support more liberal use of filters. Grade A Patients should receive a course of anticoagulation if this subsequently becomes safe.

Treatment of VTE during pregnancy Unfractionated heparin and LMWH do not cross the placenta and are safe for the fetus, whereas oral anticoagulants cross the placenta and can cause fetal bleeding and malformations.110,111 Therefore, pregnant women with VTE should be treated with therapeutic doses of subcutaneous heparin (unfractionated heparin or, increasingly, LMWH) throughout pregnancy. Grade B Care should be taken to avoid delivery while the mother is therapeutically anticoagulated; one management approach involves stopping subcutaneous heparin 24 hours prior to induction of labor and switching to intravenous heparin if there is a high risk of embolism. After delivery, warfarin, which is safe for infants of nursing mothers, should be given (with initial heparin overlap) for 6 weeks and until a minimum of 3 months of treatment has been completed. Grade B

The future There are many questions relating to currently available antithrombotic agents and diagnostic techniques that need answering, and many new antithrombotic agents under development that will require clinical evaluation. In addition, future studies are expected to focus on clinical and genetic subgroups that may benefit from tailored management, such as different intensities or durations of prophylaxis or treatment. Thrombolytic therapy deserves further evaluation, particularly systemic therapy for severe PE without overt hemodynamic compromise (for example, with echocardiographic right ventricular dysfunction), and catheter-directed therapy for iliofemoral DVT. Safer thrombolytic regimens might also broaden indications. In order to provide clear directions for clinical management, future studies need to focus on clinically important outcomes (that is, symptomatic VTE, major bleeding).

References 1.Turpie AGG, Levine MN, Hirsh J et al. Double blind randomised trial of Org 10172 low-molecular-weight heparinoid

872

in the prevention of deep vein thrombosis in thrombotic stoke. Lancet 1987;1:523–6. 2.Ginsberg J, Brill-Edwards P, Burrows RF et al. Venous thrombosis during pregnancy: leg and trimester of presentation. Thromb Haemost 1992;67:519–20. 3.Cruickshank MK, Levine MN, Hirsh J et al. An evaluation of impedance plethysmography and 125I-fibrinogen leg scanning in patients following hip surgery. Thromb Haemost 1989;62: 830–4. 4.Bern MM, Lokich JJ, Wallach SR et al. Very low doses of warfarin can prevent thrombosis in central venous catheters. Ann Intern Med 1990;112:423–8. 5.Merrer J, De Jonghe B, Golliot F et al. Complications of femoral and subclavian venous catheterization in critically ill patients: a randomized controlled trial. JAMA 2001;286: 700–7. 6.Heijboer H, Brandjes PM, Buller HR, Sturk A, ten Cate JW. Deficiencies of coagulation-inhibiting and fibrinolytic proteins in outpatients with deep-vein thrombosis. N Engl J Med 1990;323:1512–16. 7.Kearon C, Crowther M, Hirsh J. Management of patients with hereditary hypercoagulable disorders. Ann Rev Med 2000; 51:169–85. 8.Emmerich J, Rosendaal FR, Cattaneo M et al. Combined effect of factor V Leiden and prothrombin 20210A on the risk of venous thromboembolism – pooled analysis of 8 case–control studies including 2310 cases and 3204 controls. Study Group for Pooled-Analysis in Venous Thromboembolism. Thromb Haemost 2001;86:809–16. 9.Cattaneo M. Hyperhomocysteinemia, atherosclerosis and thrombosis. Thromb Haemost 1999;81:165–76. 10.Rosendaal FR. High levels of factor VIII and venous thrombosis. Thromb Haemost 2000;83:1–2. 11.Meijers JC, Tekelenburg WL, Bouma BN, Bertina RM, Rosendaal FR. High levels of coagulation factor XI as a risk factor for venous thrombosis. N Engl J Med 2000;342: 696–701. 12.van Hylckama Vlieg A, van der Linden IK, Bertina RM, Rosendaal FR. High levels of factor IX increase the risk of venous thrombosis. Blood 2000;95:3678–82. 13.Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 3-untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996;88:3698–703. 14.Kearon C. Epidemiology of venous thromboembolism. Sem Vasc Med 2001;1:7–25. 15.Warkentin TE, Levine MN, Hirsh J et al. Heparin-induced thrombocytopenia in patients treated with low-molecularweight heparin or unfractionated heparin. N Engl J Med 1995;332:1330–5. 16.Anderson FA, Wheeler HB, Goldberg RJ et al. A populationbased perspective of the hospital incidence and case-fatality rates of deep vein thrombosis and pulmonary embolism. Arch Intern Med 1991;151:933–8. 17.Kearon C. Natural history of venous thromboembolism. Sem Vasc Med 2001;1:27–37. 18.Hull R, Hirsh J, Sackett DL et al. Clinical validity of a negative venogram in patients with clinically suspected venous thrombosis. Circulation 1981;64:622–5.


19.Kearon C, Julian JA, Newman TE, Ginsberg JS, for the McMaster Diagnostic Imaging Practice Guidelines Initiative. Non-invasive diagnosis of deep vein thrombosis. Ann Intern Med 1998;128:663–77. 20.Wells PS, Hirsh J, Anderson DR et al. A simple clinical model for the diagnosis of deep-vein thrombosis combined with impedance plethysmography: potential for an improvement in the diagnosis process. J Intern Med 1998;243:15–23. 21.Anand SS, Wells PS, Hunt D, Brill-Edwards P, Cook D, Ginsberg JS. Does this patient have deep vein thrombosis? JAMA 1998;279:1094–9. 22.Wells PS, Anderson DR, Bormanis J et al. Value of assessment of pretest probability of deep-vein thrombosis in clinical management. Lancet 1997;350:1795–8. 23.Wells PS, Hirsh J, Anderson DR et al. Accuracy of clinical assessment of deep-vein thrombosis. Lancet 1995;345: 1326–30. 24.Wells PS, Hirsh J, Anderson DR et al. Comparison of the accuracy of impedance plethysmography and compression ultrasonography in outpatients with clinically suspected deep vein thrombosis. A two center paired-design prospective trial. Thromb Haemost 1995;74:1423–7. 25.Robinson KS, Anderson DR, Gross M et al. Ultrasonographic screening before hospital discharge for deep venous thrombosis after arthroplasty: the Post-Arthoplasty Screening Study. A randomized, controlled trial. Ann Intern Med 1997;127: 439–45. 26.Lee AYY, Ginsberg JS. Laboratory diagnosis of venous thromboembolism. Bailliere’s Clin Haematol 1998;11:587–604. 27.Perrier A, Desmarais S, Miron MJ et al. Non-invasive diagnosis of venous thromboembolism in outpatients. Lancet 1999; 353:190–5. 28.Bernardi E, Prandoni P, Lensing AWA et al. D-dimer testing as an adjunct to ultrasonography in patients with clinically suspected deep vein thrombosis: prospective cohort study. BMJ 1998;317:1037–40. 29.Kearon C, Ginsberg JS, Douketis J et al. Management of suspected deep venous thrombosis in outpatients by using clinical assessment and D-dimer testing. Ann Intern Med 2001; 135:108–11. 30.Wells PS, Anderson DR, Rodger M et al. Excluding pulmonary embolism at the bedside without diagnostic imaging: management of patients with suspected pulmonary embolism presenting to the emergency department by using a simple clinical model and d-dimer. Ann Intern Med 2001;135: 98–107. 31.Anderson DR, Wells PS, Stiell I et al. Thrombosis in the emergency department: use of a clinical diagnosis model to safely avoid the need for urgent radiological investigation. Arch Intern Med 1999;159:477–82. 32.Ginsberg JS, Wells PS, Kearon C et al. Sensitivity and specificity of a rapid whole-blood assay for D-dimer in the diagnosis of pulmonary embolism. Ann Intern Med 1998; 129:1006–11. 33.Righini M, Goehring C, Bounameaux H, Perrier A. Effects of age on the performance of common diagnostic tests for pulmonary embolism. Am J Med 2000;109:357–61. 34.Lee A, Julian J, Levine M et al. Clinical utility of a rapid whole-blood D-dimer assay in patients with cancer who present with suspected acute deep venous thrombosis. Ann Intern Med 2000;131:417–23. 35.Hull RD, Raskob GE, Carter CJ. Serial impedance plethysmography in pregnant patients with clinically suspected deep-vein

thrombosis. Clinical validity of negative findings. Ann Intern Med 1990;112:663–7. 36. de Boer K, Buller HR, ten Cate JW, Levi M. Deep vein thrombosis in obstetric patients: diagnosis and risk factors. Thromb Haemost 1992;67:4–7. 37.Prandoni P, Cogo A, Bernardi E et al. A simple ultrasound approach for detection of recurrent proximal vein thrombosis. Circulation 1993;88:1730–5. 38.Fraser DG, Moody AR, Morgan PS, Martel AL, Davidson I. Diagnosis of lower-limb deep venous thrombosis: a prospective blinded study of magnetic resonance direct thrombus imaging. Ann Intern Med 2002;136:89–98. 39.Stein PD, Athanasoulis C, Alavi A et al. Complications and validity of pulmonary angiography in acute pulmonary embolism. Circulation 1992;85:462–8. 40.Kearon C. Diagnosis of pulmonary embolism. Can Med Ass J 2002 (in press). 41.Hull RD, Raskob GE, Coates G, Panju AA. Clinical validity of a normal perfusion lung scan in patients with suspected pulmonary embolism. Chest 1990;97:23–6. 42.The PIOPED investigators. Value of the ventilation perfusion scan in acute pulmonary embolism. JAMA 1990;263:2753–9. 43.Hull RD, Hirsh J, Carter CJ et al. Pulmonary angiography, ventilation lung scanning, and venography for clinically suspected pulmonary embolism with abnormal perfusion lung scan. Ann Intern Med 1983;98:891–9. 44.Wells PS, Ginsberg JS, Anderson DR et al. Use of a clinical model for safe management of patients with suspected pulmonary embolism. Ann Intern Med 1998;129:997–1005. 45.Hull RD, Hirsh J, Carter CJ et al. Diagnostic value of ventilation–perfusion lung scanning in patients with suspected pulmonary embolism. Chest 1985;88:819–28. 46.Stein PD, Henry JW, Gottschalk A. Mismatched vascular defects. An easy alternative to mismatched segmental equivalent defects for the interpretation of ventilation/perfusion lung scans in pulmonary embolism. Chest 1993;104:468–72. 47.Rathbun SW, Raskob GE, Whitsett TL. Sensitivity and specificity of helical computed tomography in the diagnosis of pulmonary embolism: a systematic review. Ann Intern Med 2000;132:227–32. 48.Perrier A, Howarth N, Didier D et al. Performance of helical computed tomography in unselected outpatients with suspected pulmonary embolism. Ann Intern Med 2001;135:88–97. 49.Kearon C, Ginsberg JS, Hirsh J. The role of venous ultrasonography in the diagnosis of suspected deep venous thrombosis and pulmonary embolism. Ann Intern Med 1998; 129:1044–9. 50.Perrier A, Miron MJ, Desmarais S et al. Using clinical evaluation and lung scan to rule out suspected pulmonary embolism: Is it a valid option in patients with normal results of lowerlimb venous compression ultrasonography? Arch Intern Med 2000;160:512–16. 51.Hull RD, Raskob GE, Ginsberg JS et al. A noninvasive strategy for the treatment of patients with suspected pulmonary embolism. Arch Intern Med 1994;154:289–97. 52.Chan WS, Ray JG, Murray S, Coady GE, Coates AL, Ginsberg JS. Suspected pulmonary embolism in pregnancy: Clinical presentation, results of lung scan, subsequent materenal and pediatric outcomes. Arch Intern Med 2002 (in press).

873


53.Ginsberg JS, Hirsh J, Rainbow AJ, Coates G. Risks to the fetus of radiologic procedures used in the diagnosis of maternal venous thromboembolic disease. Thromb Haemost 1989;61:189–96. 54.Sevitt S, Gallagher NG. Prevention of venous thrombosis and pulmonary embolism in injured patients. Lancet 1959;ii: 981–9. 55.Kakkar VV, Corrigan TP, Fossard DP. Prevention of fatal postoperative pulmonary embolism by low doses of heparin. An international multicenter trial. Lancet 1975;ii:45–51. 56.Collins R, Scrimgeour A, Yusuf S, Peto R. Reduction in fatal pulmonary embolism and venous thrombosis by perioperative administration of subcutaneous heparin. N Engl J Med 1988; 318:1162–73. 57.Leyvraz PF, Richard J, Bachmann F. Adjusted versus fixed dose subcutaneous heparin in the prevention of deep vein thrombosis after total hip replacement. N Engl J Med 1983;309:954–8. 58.Wille-Jorgensen P, Thorup J, Fischer A, Holst-Christensen J, Flamsholt R. Heparin with and without graded compression stockings in the prevention of thromboembolic complications of major abdominal surgery: a randomized trial. Br J Surg 1985;72:579–81. 59.Ramos R, Salem BI, De Pawlikowski MP, Coordes C, Eisenberg S, Leidenfrost R. The efficacy of pneumatic compression stockings in the prevention of pulmonary embolism after cardiac surgery. Chest 1996;109:82–5. 60.Koch A, Bouges S, Ziegler S, Dinkel H, Daures JP, Victor N. Low molecular weight heparin and unfractionated heparin in thrombosis prophylaxis after major surgical intervention: update of previous meta-analyses. Br J Surg 1997;84:750–9. 61.Eikelboom JW, Quinlan DJ, Douketis JD. Extended-duration prophylaxis against venous thromboembolism after total hip or knee replacement: a meta-analysis of the randomised trials. Lancet 2001;358:9–15. 62.Geerts WH, Heit JA, Clagett GP et al. Prevention of venous thromboembolism. Chest 2001;119:132S–75S. 63.Colwell CW Jr, Collis DK, Paulson R et al. Comparison of enoxaparin and warfarin for the prevention of venous thromboembolic disease after total hip arthroplasty. J Bone J Surg 1999;81-A:932–40. 64.“Antiplatelet trialists’ collaboration”. Collaborative overview of randomised trials of antiplatelet therapy-III: reduction in venous thrombosis and pulmonary embolism by antiplatelet prophylaxis among surgical and medical patients. BMJ 1994;308:235–46. 65.Prevention of pulmonary embolism and deep vein thrombosis with low dose aspirin: Pulmonary Embolism Prevention (PEP) trial. Lancet 2000;355:1295–302. 66.Eriksson BI, Wille-Jorgensen P, Kalebo P et al. A comparison of recombinant hirudin with a low-molecular weight heparin to prevent thromboembolic complications after total hip replacement. N Engl J Med 1997;337:1329–35. 67.Turpie AG, Gallus AS, Hoek JA. A synthetic pentasaccharide for the prevention of deep-vein thrombosis after total hip replacement. N Engl J Med 2001;344:619–25. 67a.Erikssen BI, Bauer KA, Lassen MR, Turpie AG. Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after hip-fracture surgery. N Engl J Med 2001;345:1298–304.

874

67b.Bauer KA, Eriksson BI, Lassen MR, Turpie AG. Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after elective major knee surgery. N Engl J Med 2001;345:1305–10. 68.Gårdlund for the Heparin Prophylaxis Group. Randomised, controlled trial of low-dose heparin for prevention of fatal pulmonary embolism in patients with infectious diseases. Lancet 1996;347:1357–61. 69.Wells PS, Lensing AWA, Hirsh J. Graduated compression stockings in the prevention of postoperative venous thromboembolism: a meta-analysis. Arch Intern Med 1994;154: 67–72. 70.Hull R, Delmore T, Hirsh J et al. Effectiveness of an intermittent pulsatile elastic stocking for the prevention of calf and thigh vein thrombosis in patients undergoing elective knee surgery. Thromb Res 1979;16:37–45. 71.Hull RD, Raskob GE, Gent M et al. Effectiveness of intermittent pneumatic leg compression for preventing deep vein thrombosis after total hip replacement. JAMA 1990;263: 2313–17. 72.Turpie AGG, Hirsh J, Gent M, Julian DH, Johnson J. Prevention of deep vein thrombosis in potential neurosurgical patients: a randomized trial comparing graduated compression stockings alone or graduated compression stockings plus intermittent pneumatic compression with control. Arch Intern Med 1989;149:679–81. 73.Salzman EW, Davies GC. Prophylaxis of venous thromboembollism: analysis of cost effectiveness. Ann Surg 1980;191: 207–18. 74.Barritt DW, Jordan SC. Anticoagulant drugs in the treatment of pulmonary embolism: a controlled trial. Lancet 1960;1: 1309–12. 75.Gallus AS, Jackaman J, Tillett J, Mills W, Sycherley A. Safety and efficacy of warfarin started early after submassive venous thrombosis or pulmonary embolism. Lancet 1986;2:1293–6. 76.Hull RD, Raskob GE, Rosenbloom D et al. Heparin for 5 days as compared with 10 days in the initial treatment of proximal venous thrombosis. N Engl J Med 1990;322:1260–4. 77.Brandjes DPM, Heijboer H, Buller HR, de Rijk M, Jagt H, ten Cate JW. Acenocoumarol and heparin compared with acenocoumarol alone in the initial treatment of proximal-vein thrombosis. N Engl J Med 1992;327:1485–9. 78.Dolovich LR, Ginsberg JS, Douketis JD, Holbrook AM, Cheah G. A meta-analysis comparing low-molecular-weight heparins with unfractionated heparin in the treatment of venous thromboembolism: examining some unanswered questions regarding location of treatment, product type, and dosing frequency. Arch Intern Med 2000;160:181–8. 79.Koopman MMW, Prandoni P, Piovella F et al. Treatment of venous thrombosis with intravenous unfractionated heparin administered in the hospital as compared with subcutaneous low-molecular-weight heparin administered at home. N Engl J Med 1996;334:682–7. 80.Couturaud F, Julian JA, Kearon C. Low molecular weight heparin administered once versus twice daily in patients with venous thromboembolism: a meta-analysis. Thromb Haemost 2001;86:980–4. 81.Hirsh J, Warkentin TE, Shaughnessy SG et al. Heparin and low-molecular-weight heparin: mechanisms of action,


pharmacokinetics, dosing, monitoring, efficacy, and safety. Chest 2001;119:64S–94S. 82.Chong BH, Gallus AS, Cade JF et al. Prospective randomized open-label comparison of danaparoid with dextran 70 in the treatment of heparin-induced thrombocytopenia with thrombosis. Thromb Haemost 2001;86:1170–5. 83.Hull R, Delmore T, Genton E et al. Warfarin sodium versus low-dose heparin in the long-term treatment of venous thrombosis. N Engl J Med 1979;301:855–8. 84.Hull R, Delmore T, Carter C et al. Adjusted subcutaneous heparin versus warfarin sodium in the long-term treatment of venous thrombosis. N Engl J Med 1982;306:189–94. 85.Hyers TM, Agnelli G, Hull RD et al. Antithrombotic therapy for venous thromboembolic disease. Chest 2001; 119:176S–93S. 86.Hull R, Hirsh J, Jay R et al. Different intensities of oral anticoagulant therapy in the treatment of proximal-vein thrombosis. N Engl J Med 1982;307:1676–81. 87.Harrison L, Johnston M, Massicotte MP, Crowther M, Moffat K, Hirsh J. Comparison of 5-mg and 10-mg loading doses in initiation of warfarin therapy. Ann Intern Med 1997; 126:133–6. 88.Crowther MA, Ginsberg JS, Kearon C et al. A randomized trial comparing 5 mg and 10 mg warfarin loading doses. Arch Intern Med 1999;159:46–8. 89.Crowther MA, Julian J, McCarty D et al. Treatment of warfarin-associated coagulopathy with oral vitamin K: a randomised controlled trial. Lancet 2000;356:1551–3. 90.Research Committee of the British Thoracic Society. Optimum duration of anticoagulation for deep-vein thrombosis and pulmonary embolism. Lancet 1992;340:873–6. 91.Levine MN, Hirsh J, Gent M et al. Optimal duration of oral anticoagulant therapy: a randomized trial comparing four weeks with three months of warfarin in patients with proximal deep vein thrombosis. Thromb Haemost 1995;74:606–11. 92.Schulman S, Rhedin A-S, Lindmarker P et al. A comparison of six weeks with six months of oral anticoagulant therapy after a first episode of venous thromboembolism. N Engl J Med 1995;332:1661–5. 93.Prandoni P, Lensing AWA, Cogo A et al. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med 1996;125:1–7. 94.Pini M, Aiello S, Manotti C et al. Low molecular weight heparin versus warfarin the prevention of recurrence after deep vein thrombosis. Thromb Haemost 1994;72:191–7. 95.Pinede L, Ninet J, Duhaut P et al. Comparison of 3 and 6 months of oral anticoagulant therapy after a first episode of proximal deep vein thrombosis or pulmonary embolism and comparison of 6 and 12 weeks of therapy after isolated calf deep vein thrombosis. Circulation 2001;103:2453–60. 96.Pinede L, Duhaut P, Cucherat M, Ninet J, Pasquier J, Boissel JP. Comparison of long versus short duration of anticoagulant therapy after a first episode of venous thromboembolism: a meta-analysis of randomized, controlled trials. J Intern Med 2000;247:553–62. 97.Kearon C, Gent M, Hirsh J et al. A comparison of three months of anticoagulation with extended anticoagulation for a first episode of idiopathic venous thromboembolism. N Engl J Med 1999;340:901–7.

98.Agnelli G, Prandoni P, Santamaria MG et al. Three months versus one year of oral anticoagulant therapy for idiopathic deep vein thrombosis. N Eng J Med 2001;345:165–9. 99.Lindmarker P, Schulman S. The risk of ipsilateral versus contralateral recurrent deep vein thrombosis in the leg. The DURAC Trial Study Group. J Intern Med 2000;247:601–6. 100.Schulman S, Granqvist S, Holmstrom M et al. The duration of oral anticoagulant therapy after a second episode of venous thromboembolism. N Engl J Med 1997;336:393–8. 101.Beyth RJ, Quinn LM, Landefeld S. Prospective evaluation of an index for predicting the risk of major bleeding in outpatients treated with warfarin. Am J Med 1998;105:91–9. 102.Schulman S, Svenungsson E, Granqvist S. Anticardiolipin antibodies predict early recurrence of thromboembolism and death among patients with venous thromboembolism following anticoagulant therapy. Am J Med 1998;104:332–8. 103.Lindmarker P, Schulman S, Sten-Linder M, Wiman B, Egberg N, Johnsson H. The risk of recurrent venous thromboembolism in carriers and non-carriers of the G1691A Allele in the coagulation factor V gene and the G20210A Allele in the prothrombin gene. Thromb Haemost 1999;81:684–9. 104.Hirsh J, Lensing A. Thrombolytic therapy for deep vein thrombosis. Int Angiol 1996;5:S22–S25. 105.Schweizer J, Kirch W, Koch R et al. Short- and long-term results after thrombolytic treatment of deep vein thrombosis. J Am Coll Cardiol 2000;36:1336–43. 106.Blackmon JR, Sautter RD, Wagner HN. Uokinase pulmonary embolism trial: phase I results. JAMA 1970;214:2163–72. 107.Dalen JE, Alpert JS, Hirsh J. Thrombolytic therapy for pulmonary embolism. Is it effective? Is it safe? When is it indicated? Arch Intern Med 1997;157:2550–6. 108.Jerjes-Sanchez C, Ramirez-Rivera A, de Lourdes Garcia M et al. Streprokinase and heparin versus heparin alone in massive pulmonary embolism: a randomized controlled trial. J Thromb Thrombolys 1995;2:227–9. 109.Decousus H, Leizorovicz A, Parent F et al. A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis. N Engl J Med 1998;338:409–15. 110.Ginsberg JS, Hirsh J, Levine MN, Burrows R. Risks to the fetus of anticoagulant therapy during pregnancy. Thromb Haemost 1989;61:197–203. 111.Ginsberg JS, Greer I, Hirsh J. Use of antithrombotic agents during pregnancy. Chest 2001;119:122S–31S. 112.Cogo A, Lensing AWA, Prandoni P, Hirsh J. Distribution of thrombosis in patients with symptomatic deep-vein thrombosis: Implications for simplifying the diagnostic process with compression ultrasound. Arch Intern Med 1993;153:2777–80. 113.Lagerstedt CI, Olsson CG, Fagher BO, Oqvist BW, Albrechtsson U. Need for long-term anticoagulant treatment in symptomatic calf-vein thrombosis. Lancet 1985;ii:515–18. 114.Moser KM, LeMoine JR. Is embolic risk conditioned by location of deep venous thrombosis? Ann Intern Med 1981;94: 439–44. 115.Moser KM, Fedullo PF, LittleJohn JK, Crawford R. Frequent asymptomatic pulmonary embolism in patients with deep venous thrombosis. JAMA 1994;27:223–5. 116.Kruit WHJ, de Boer AC, Sing AK, van Roon F. The significance of venography in the management of patients with clinically

875


suspected pulmonary embolism. J Intern Med 1991;230: 333–9. 117.Hull RD, Raskob GE, Coates G, Panju AA, Gill GJ. A new noninvasive management strategy for patients with suspected pulmonary embolism. Arch Intern Med 1989;149:2549–55. 118.Stein PD, Henry JW. Prevalence of acute pulmonary embolism among patients in a general hospital and at autopsy. Chest 1995;108:978–81. 119.Bell WR, Simon TL. Current status of pulmonary embolic disease: pathophysiology, diagnosis, prevention, and treatment. Am Heart J 1982;103:239–61. 120.Ginsberg J, Kearon C, Douketis J et al. The use of D-dimer testing and impedance plethysmographic examination in patients with clinical indications of deep vein thrombosis. Arch Intern Med 1997;157:1077–81.

876

121.Hull RD, Carter CJ, Jay RM et al. The diagnosis of acute recurrent deep vein thrombosis: a diagnostic challenge. Circulation 1983;67:901–6. 122.Huisman MV, Buller HR, ten Cate JW. Utility of impedance plethysmography in the diagnosis of recurrent deep-vein thrombosis. Arch Intern Med 1988;148:681–3. 123.Wells PS, Anderson DR, Rodger M et al. Derivation of a simple clinical model to categorize patients probability of pulmonary embolism: increasing the models utility with the SimpliRED D-dimer. Thromb Haemost 2000;83:416–20. 124.Turkstra F, Kiujer PMM, van Beek E Jr, Brandjes DPM, ten Cate JW, Buller HR. Diagnostic utility of ultrasonography of leg veins in patients suspected of having pulmonary embolism. Ann Intern Med 1997;126:775–81.

62

Peripheral vascular disease Jesper Swedenborg, Jan Östergren

Epidemiology The prevalence of lower extremity arterial occlusive disease as judged by history has been examined in several studies. Large cohorts of patients have been questioned about symptoms of intermittent claudication. This has mostly been done using a questionnaire initially designed by Rose.1 The method has an acceptable specificity but lacks sensitivity and, for obvious reasons, it does not detect asymptomatic arterial occlusive disease.2 The prevalence of peripheral arterial occlusive disease varies between studies, with high figures reported from Russia and Finland.3,4 With large and reliable studies, it is likely that the prevalence at the age of 60 is 3–6%.5 Most studies report a prevalence of less than 5% at 50 years. In order to detect lower extremity arterial occlusive disease more specifically, studies have been performed measuring ankle pressure with non-invasive techniques. In general it can be said that the prevalence of disease increases by a factor of 3 compared with studies based on questionnaires. There is a significant correlation between the ankle brachial pressure index (ABI) and the symptom of intermittent claudication, although the correlation is modest with r values between 0·1 and 0·2.6 Based on such objective methods, 11·7% of the population in the Framingham study had peripheral arterial disease. Thus assessment of peripheral arterial disease by the symptom of intermittent claudication underestimates the true prevalence,7 but the cut off points determining what is considered to be a pathologic ABI is of great importance for the estimation of the prevalence using objective methods.8 The prevalence is greatly influenced by age as pointed out in one of the major studies, the Framingham study.9 Other important factors are cigarette smoking and sex. Thus nonsmoking women in the age group 55–64 years showed a prevalence by history of 3·9% compared with smoking men in the age group 75–84 years where the prevalence was 14·5%. Additional factors increasing the risk are diabetes and fibrinogen levels.10 Few studies have examined the incidence of peripheral arterial occlusive disease by following normal subjects and determining when claudication appears. In the Framingham study, the yearly incidence increases from 0·2% in 45–55 year old men to 0·5% in 55–65 year old men.9 In the last

follow up after 38 years, the yearly rates were found to increase until the age 75 and then declined. The statistical analyses revealed that those with intermittent claudication were significantly older, had higher cholesterol levels, higher blood pressure, higher frequency of diabetes, and smoked more cigarettes.11 The Edinburgh artery study provides similar figures with an annual incidence of 1·8 per 1000 randomly selected patients from general practitioners.12

Long-term outcome The natural history of patients with lower extremity arterial disease has been studied regarding both the fate of the limb and mortality. Among patients with peripheral arterial occlusive disease, at most one in five will require surgical correction for their vascular disease13 and 2–5% will undergo amputation.5,9 The risk for amputation decreases if the patients can stop smoking.14 Patients with peripheral arterial occlusive disease have a decreased life expectancy compared with the normal population. This is almost solely explained by cardiovascular disease in general and coronary artery disease in particular. After 10 years, only 52% of claudicants are still alive.15 The relative risk of dying from cardiovascular disease and coronary heart disease (CHD) is reported to be 5–6 times that of the normal population over 10 years.16 The severity of the peripheral arterial occlusive disease is associated with the risk of dying, since the lower extremity arterial disease is a surrogate variable reflecting the severity of atherosclerosis affecting the coronary arteries.17 Smoking is also an important predictor of the risk of dying in these patient groups.18 The greatest threat to the patient with peripheral arterial disease is thus death from cardiac causes. Patients with peripheral arterial disease and concomitant three vessel coronary artery disease (CAD) have an improved survival after coronary artery bypass grafting (CABG).19 The natural course of intermittent claudication on the other hand is relatively benign in terms of limb survival as reflected by the low risk of amputation. This may, however, partly be explained by the fact that the mortality among patients with severe disease and high risk of amputation is considerably higher than for patients with mild disease. 877


Key points ● ●

● ●

The prevalence of lower extremity arterial occlusive disease is high: 3–5% in individuals over 50 years of age. Patients with peripheral arterial occlusive disease have approximately fivefold increased risk of dying from cardiovascular causes over 10 years. Mortality and morbidity are increased by smoking, hypertension, and the severity of the disease. Intermittent claudication itself has a relatively benign course as reflected by a low risk of amputation.

Key points ●

● ● ●

History (symptoms, smoking, other cardiovascular diseases) and physical examination (peripheral pulses, blood pressure) are essential. Screen for cardiovascular risk factors (cholesterol, glucose). Measurement of ABI is valuable in all patients. Duplex sonography and angiography are required only when invasive procedures are considered.

Investigation of the patient with peripheral vascular disease

Intermittent claudication

An adequate history and physical examination provide the basis for proper management of patients with peripheral vascular disease. The history should include a survey of relevant risk factors and possible symptoms of concomitant cardiovascular disease (for example, angina pectoris). Palpation of pulses and auscultation in the groin and over the femoral arteries may reveal signs of occlusion or stenoses in the vessels from the iliac artery down to the lower leg. The popliteal artery is best evaluated with the knee slightly elevated from the support and the tissue in the distal popliteal fossa pressed against the tibia. Palpation at this location is particularly important when a popliteal aneurysm is suspected. In cases with more severe ischemia, inspection may reveal a diminished growth of hair and nails, and distal ischemic ulcers often located on toes and heels. Elevation of the legs will cause a whitening of the most affected foot, which in the dependent position typically is more red than the contralateral one, owing to an increase of blood in the superficial venous plexa. Measurement of the ankle pressure is of value as a quantitative estimate of the degree of arterial insufficiency. This is easily done with a continous wave pen-doppler detecting the pulse either in the posterior tibial or the dorsal pedal artery when a blood pressure cuff around the ankle is slowly deflated from a suprasystolic pressure. By dividing the measured value with the brachial pressure the ABI is determined. An index below 0·9 is considered pathologic. In patients with diabetes mellitus, the ABI may be falsely elevated owing to sclerosis of the media of the arteries, which resists compression by the cuff. Further anatomic evaluation of the arterial system is needed only when invasive procedures are indicated. Duplex sonography is the method of choice, but in most cases has to be followed by angiography, when surgery is planned. All patients with peripheral vascular disease should have blood tests to detect other treatable risk factors such as blood lipids, blood or plasma glucose, and serum creatinine. Systemic blood pressure is also a treatable risk factor that should be measured.

Intermittent claudication is caused almost exclusively by atherosclerotic lesions in the arteries to the legs. The lesion causing the symptoms may be located above the inguinal ligament (the aorta, iliac artery, or the common femoral artery) or below, in such cases often in the distal part of the superficial femoral artery. Combinations of series of stenosis or occlusions also involving the popliteal and lower leg vessels are not uncommon. The evolution of the disease may be slow with gradual onset of symptoms but in many cases the occurrence of a thrombus in a severely stenosed area or overlying a ruptured atherosclerotic plaque may cause an acute onset of symptoms. The most common location of pain is in the calf, since the majority of vascular occlusions occur in the superficial femoral artery. When the main lesion is in the iliac region, pain and muscular dysfunction may also be located in the gluteal muscles and the thigh. The symptoms are caused by an inappropriate blood supply in relation to the metabolic needs of the muscles during exercise. When occlusion of the artery occurs gradually, collaterals, often from the deep femoral artery, may compensate for the limited arterial supply through the natural artery.

878

Pathophysiology

Therapy General measures The aim of therapy for intermittent claudication is twofold: ●

●

to reduce risk factors associated with the disease and thereby improving the long-term prognosis of the patient; to improve walking distance and thus the quality of life for the patient.

In the general management of the patient it is mandatory to screen for risk factors associated with atherosclerosis. Smoking should be stopped immediately as the risk for the patient with claudication for having an amputation in the

Peripheral vascular disease

future is reduced to virtually zero.14 Grade B Hyperlipidemia and hypertension should be treated according to guidelines outlined in other sections of this book. A meta-analysis of lipid lowering therapy in 698 patients with peripheral arterial disease indicated that active therapy reduced disease progression and the severity of claudication.20 Recently the Heart Protection Study including 20 000 patients with coronary or non-coronary artery disease or diabetes was reported, showing that simvastatin 40 mg/day reduced cardiovascular mortality and morbidity. The 24% decrease of vascular events was consistent in all subgroups including patients with peripheral vascular disease and regardless of cholesterol levels.21 Grade A Thus, a statin should be given as first-line therapy, but niacin could also be valuable since it increases serum HDL (high density lipoproteins) concentrations and lowers serum triglyceride concentrations, which are the most common lipid disturbances in patients with intermittent claudication. A fear of reducing distal perfusion pressures in patients with claudication by antihypertensive treatment has sometimes prevented doctors from instituting adequate treatment of hypertension. In particular, blockers have been considered by some to be contraindicated in this situation. Controlled studies have, however, shown that treatment of claudicants with blockers only reduces walking capacity marginally or not at all.22 Therefore, if strong indications, such as heart failure, or a previous myocardial infarction exist, blockers should also be used in claudicants. Grade A The HOPE study investigated the effect of the ACE inhibitor ramipril 10 mg/day compared with placebo.23 The study included 1715 patients with symptomatic peripheral vascular disease and 3099 patients with an ABI 0·9. These subgroups benefitted at least equally well as the entire study population from the treatment. The beneficial effect was seen even among patients who already had adequate blood pressure control. Treatment with an ACE inhibitor should thus be strongly considered in patients with peripheral arterial disease. Grade A If symptoms of increased ischemia of the legs occur during treatment for hypertension, this strengthens the indication for an invasive procedure in order to relieve the symptoms of leg ischemia. If this is not possible, the antihypertensive therapy should be reduced with caution. Since patients with intermittent claudication have an increased risk for major cardiovascular events because of their generalized atherosclerotic disease, antiplatelet therapy should be given prophylactically, preferably with aspirin, based on conclusions from meta-analysis.24 Grade A Although major studies on the effect of aspirin in patients with claudication are lacking, the effect in subgroups with claudication (n 3295; risk reduction from 11·8% to 9·7% over 27 months) seems to be equivalent to the reduction seen in the atherosclerotic population as a whole.24 The combination with dipyridamole may provide an additional

preventive effect,25 but so far only one study has shown an effect on major end points by this combination in the case of the secondary prevention of stroke.26 In 687 claudicants studied over a 7 year period, ticlopidine 250 mg 2/day reduced the need for vascular reconstructive surgery by 51% compared to placebo.27 In the same trial, the mortality rate was 29.1% lower (64 v 89 cases) in the ticlopidine group compared with the placebo group.28 The same dose of ticlopidine may also produce some increase in walking capacity in comparison with placebo.29 The disadvantage of this compound is the risk of adverse effects and the need for laboratory control of white blood cell counts. A better and safer alternative to ticlopidine for patients who cannot tolerate aspirin is clopidogrel, which was studied in the CAPRIE trial.30 In the 6452 patients with peripheral arterial disease, clopidogrel 75 mg/day showed a relative risk reduction of 23·8% in ischemic stroke, myocardial infarction, or death from other vascular causes compared to aspirin 325 mg/day.30 Exercise Patients with claudication should be instructed to walk as much as possible and, when pain occurs, they should try to walk despite the pain.31 Training by intensive walking on treadmill or outdoors has been shown to be as effective or even better than other programs of physical training and, in most cases, will improve walking capacity by 100–200%.31 In some cases the symptoms of claudication may even disappear completely. The optimal exercise program includes walking to near maximal pain for more than 30 minutes per session at least three times weekly during at least a 6 month period.32 Pharmacologic treatment to increase walking capacity Different pharmacologic agents have been evaluated for improvement of walking distance in addition to physical training. Most of these treatments have been inconsistent in their effect and of marginal benefit. Generally, vasodilators have not been shown to be effective. The agent so far most extensively studied has been pentoxifylline, which is available in most countries for the treatment of intermittent claudication. The patients most likely to respond are those with a history of claudication over 1 year and an ABI of 0·8.33 A meta-analysis of the pentoxifylline studies has shown an increase of 44 meters in maximal walking distance on the treadmill compared with placebo.34 The phosphodiesterase inhibitor cilostazol was approved in 1999 by the FDA for treatment of claudication. Cilostazol is primarily a platelet inhibitor and a vasodilator that has been shown to increase the walking distance compared with placebo and also with pentoxifylline.35 However, the use of the drug is hampered by the risk for worsening heart 879


failure.36 A randomized but open study37 indicates that prostaglandin E1 given intravenously may be more effective than pentoxifylline (60·4% compared with 10·5% increase in walking capacity), but further studies are needed to establish the role of prostaglandins in this context.

are methods that can assess tissue nutrition, thereby offering additional prognostic information in these patients.38 Patients with critical ischemia should be evaluated for possible vascular reconstructive surgery or endovascular treatment (see below).

Key points ● ● ●

● ●

Quit smoking! Regular exercise – walking until intolerable pain. Intervention against other cardiovascular risk factors; treat hypertension and institute a statin to all patients with a normal or high cholesterol level. Antiplatelet therapy and ACE inhibitor to be considered for all patients. Other pharmacologic therapy of very limited benefit.

Critical ischemia Pathophysiology When the distal pressure in the leg is too low to provide sufficient perfusion in order to meet the metabolic demands of the tissue, pain will also occur in the resting situation, particularly in the supine position when there is no contribution to distal pressures by hydrostatic forces. Subsequently ulcers in the apical parts of the extremity may develop owing to an insufficient nutritional blood flow in the skin. According to the European Consensus Document on chronic lower limb ischemia, critical ischemia is defined as “persistently recurring rest pain requiring regular analgesia for more than 2 weeks and/or ulceration or gangrene of the foot and toes in combination with an ankle systolic pressure less than 50 mmHg”. In the case of diabetes, where the measurements of ankle pressures are unreliable because of incompressible arteries, the absence of palpable pulses are sufficient.38 The definition has been criticized because many patients with critical limb ischemia according to the above definition still have an intact lower extremity after 1 year. This is exemplified by the findings in control groups of randomized trials regarding non-surgical treatment of critical limb ischemia.39 Furthermore, some patients who do not fit into this definition may lose their legs because of ischemia.40 A recent consensus document was made more practical. A patient with critical limb ischemia is defined as “a patient with chronic ischemic rest pain, ulcers and gangrene attributable to objectively proven arterial disease”.41 The crucial factor regarding tissue nutrition is the flow through the capillary bed, which is dependent not only on the pressure in the arteries but also on other factors, such as blood viscosity and distribution of flow between nutritional and non-nutritional vessels – that is, arteriovenous shunts. Intravital capillaroscopy and transcutaneous oxygen tension 880

General measures When invasive procedures to restore blood flow (see below) are not possible or have failed, several therapeutic measures should be considered. Optimization of the hemodynamic situation is one aim. Heart failure and edema should be treated vigorously. Lowering the foot end of the bed at night may improve distal perfusion pressure and relieve symptoms. Shoes should be well fitting to avoid the risk of pressure against the skin. Ulcers should be treated with care, and more often dry dressings are preferable in order not to moisturize intact skin around the ulcer area. Though not scientifically proven in this situation, anticoagulation may be of benefit. Thus, oral anticoagulants or low molecular weight heparin should be considered as an alternative or an addition to aspirin, since both arterial and venous thrombi are common in the severely ischemic leg.42 Warfarin has been shown to lower the risk of occlusion in femoropopliteal vein grafts.43 Pain should be treated by pharmacologic measures. Spinal cord stimulation could be used since this method has been shown to decrease pain possibly by increasing microvascular blood flow.44 The only pharmacologic agent so far convincingly shown to have a positive influence on the prognosis of patients with critical limb ischemia is a synthetic prostacyclin (Iloprost), which is given intravenously daily for a period of 2–4 weeks. In a meta-analysis, rest pain and ulcer size were found to improve in comparison with placebo and, more importantly, the probability of being alive with both legs still intact after 6 months was 65% in the Iloprost-treated group compared to 45% in the placebo-treated patients.39 Grade A Pentoxifylline has been shown to be of benefit in a short-term perspective as a pain reliever but no long-term trials have been performed.45 Spinal cord stimulation has been used to avoid amputations, but so far it has not benefitted patients with critical limb ischemia as a preventive treatment.46

Key points ● ● ● ● ● ● ●

Evaluate possibilities for revascularization. Optimize cardiac hemodynamics. Avoid hypotension – lower foot end of bed at night. Provide adequate pain relief. Optimize local skin and wound care. Consider anticoagulation or antiplatelet therapy. Consider Iloprost treatment when revascularization is not possible or has failed.


Surgical treatment of intermittent claudication and critical ischemia In this chapter both open surgery and endovascular treatment are considered. In the latter group percutaneous transluminal angioplasty (PTA) in combination with both thrombolysis and stenting are included. The major indications for reconstructive procedures for lower extremity ischemia are critical ischemia and claudication. Preoperative cardiac evaluation Since patients with peripheral vascular disease have a high frequency of cardiac comorbidity, the perioperative mortality and morbidity is dominated by cardiac problems. Many attempts have been made to identify patients with a high risk of perioperative cardiac complications. The rationale for such a strategy is to identify patients in need of coronary artery revascularization before the vascular procedure, and also to provide a basis for more intensive cardiac monitoring during peripheral vascular surgery. Although not specifically designed for peripheral vascular surgery, clinical risk scores according to Goldman47 or Detsky48 have been used. Further tests include ambulatory ECG, dipyridamole thallium scintigraphy, ejection fraction estimation by radionuclide ventriculography, and stress echocardiography. All these tests are effective in predicting perioperative cardiac mortality and morbidity, but dobutamine stress echocardiography seems to be most promising in a meta-analysis.49 Patients who have reversible defects on preoperative thallium scintigraphy are at a high risk of perioperative cardiac mortality and morbidity,50 and successful coronary revascularization decreases this risk following vascular surgery.51 Nevertheless, routine evaluation of all patients scheduled for peripheral vascular surgery with thallium scintigraphy is not warranted.52 The reason for this is that both coronary angiography and coronary revascularization add to the risk.53 Today it can therefore be concluded that patients with a low risk, as reflected by either absence of angina pectoris or only mild disease, do not benefit from further evaluation aiming at coronary angiography.54 Patients with high risk according to clinical scoring systems or careful history should be evaluated with dipyridamole thallium scintigraphy or dobutamin stress echocardiography. Grade B The use of bisoprolol, a 1 selective inhibitor, reduced the 30 day combined cardiac morbidity and mortality from 34% to 3·4% in high-risk patients undergoing peripheral vascular surgery.55 Whether other blockers have the same effect remains to be shown. Grade B Open surgical vascular reconstructions The vascular reconstructions for lower limb ischemia are mainly divided into supra- and infrainguinal reconstructions.

Suprainguinal vascular reconstructions In the aortoiliac segment, vascular reconstructive procedures were initially dominated by thromboendarterectomy (TEA); this, however, requires large dissections. After the introduction of bypass grafting with synthetic materials TEA was largely abandoned except for short localized lesions. The results of aortobifemoral bypass with Dacron grafts for arterial occlusive disease are usually good with 1 year patency rates in the range of 95%. The patency rates are influenced by the outflow bed, so that patients with a patent superficial femoral artery (SFA) have better patency rates than those with an occluded SFA. There are no prospective randomized trials comparing TEA and aortofemoral bypass. TEA is said to have lower long-term patency rates, and another disadvantage is that the surgical procedure is more extensive. Aortofemoral bypass with a synthetic graft, however, has the disadvantage of risk of infection. Although this is an infrequent complication, it is associated with major morbidity and mortality, since an infected graft has to be removed. During recent years the number of aortobifemoral reconstructions have declined owing to the more frequent use of endovascular methods, particularly PTA with or without stenting. Thus the extensive procedure of aortobifemoral bypass can be converted into a lesser procedure if at least one iliac artery can be opened with PTA. In such cases the contralateral leg can be revascularized with the aid of an extra-anatomic procedure – that is, femorofemoral bypass. The latter procedure has good patency rates, approximately 90% at 1 year and 65% at 5 years.56–58 In patients who are unfit for major surgery and where the iliac arteries cannot be opened up with endovascular procedures another extraanatomic bypass can be employed. In such patients axillobifemoral bypass can be used, but this type of extraanatomic bypass is a compromise, since it has lower patency rates than aortobifemoral bypass.59

Infrainguinal vascular reconstructions The standard procedure for infrainguinal occlusive disease is femoropopliteal bypass or bypass to the crural arteries. Bypass to the crural arteries is often performed in people with diabetes since their occlusive disease is in many cases more peripherally located than in non-diabetic patients with atherosclerosis. The most commonly used graft material is the saphenous vein but, if this is unavailable, arm veins or synthetic grafts may be used. In general it has been stated that use of autologous material is superior in infrainguinal reconstructions.60 Some randomized studies have failed to detect a difference in long-term patency between synthetic grafts and saphenous vein grafts. One study did not show a significantly different patency at 2 years follow up, but after 4 years there was a significant difference in favor of saphenous 881


vein grafts, 68% patency versus 47%.61 Grade A For bypass grafts with the lower anastomosis below the knee, autologous material is clearly preferred. This is particularly true when bypass procedures are done to the crural arteries where the use of synthetic grafts produces dismal results. When an autologous vein is used, the original procedure implies excision and reversal of the vein so that the blood can flow freely across the valves. The “in situ” technique, originally introduced by Hall, has, however, in recent years gained more popularity:62 the saphenous vein is left in its bed and the valves are destroyed by special instruments; tributaries are identified and tied off. Some prospective randomized trials have been performed comparing the two methods but no definitive advantage with either method has been shown.63 Therefore the personal preference of the surgeon often decides which method should be used. The advantage with the in situ method is that the larger end of the vein is anastomosed to the larger artery, and the smaller end of the vein to the small distal artery. With meticulous technique it is said that the vein is exposed to less trauma, but the valve destruction definitely induces some damage to the vein. In order to improve long-term patency rates, two methods have been employed: graft surveillance and pharmacologic treatment. Postoperative surveillance of vein grafts is used by many surgeons in order to detect a failing graft, defined as a graft with a developing stenosis that threatens to reduce the blood flow below a critical level. Only few randomized studies have been done examining the effect on long-term patency rates in surveillance programs identifying and treating critical graft stenosis. Conflicting results regarding the effectiveness of such programs have been obtained. One study reported a patency rate of 78% in an intensive surveillance program including duplex scanning of the graft after 3 years versus 53% without such a program.64 Other studies, however, have failed to demonstrate an advantage of duplex scanning over clinical surveillance with measurements of ankle pressure.65 Whether a graft surveillance program has a beneficial effect upon amputation rate also remains to be shown. The effect of antiplatelet therapy on total mortality has been studied in several trials and it seems to reduce cardiovascular mortality.66 There is only one trial that has studied the similar effects of oral anticoagulants, and this trial suggested that they both prevent graft occlusion and diminish the risk of cardiovascular death.43 Pharmacologic therapy seems to improve the patency rate for infrainguinal vascular reconstructions. Most centers use antiplatelet therapy with acetylsalicylic acid, and a meta-analysis of randomized trials has indicated that such treatment improves the patency rate.67 Grade A Oral anticoagulants are not used as widely as antiplatelet therapy but many surgeons use it selectively in patients where the prognosis for graft patency for some reason is bad. Whether antiplatelet therapy or oral anticoagulants differ in their effectiveness against graft 882

occlusion is not known. Only one study has addressed this question and no significant difference in graft patency was found between patients treated with warfarin or acetyl salicylic acid. Subgroup analysis, however, revealed that oral anticoagulants seemed to be more effective in patients receiving autologous grafts and antiplatelet therapy in those receiving synthetic grafts.68 Grade B Key points ●

● ●

●

For bilateral suprainguinal occlusions aortofemoral bypass is the standard procedure, but endovascular methods are used at an increasing rate. For unilateral suprainguinal occlusions femorofemoral bypass can be used. For infrainguinal occlusions saphenous vein bypass is the standard procedure, but synthetic grafts can be used if suitable veins are lacking. Bypass to infragenicular arteries using synthetic grafts produces inferior results compared to saphenous vein bypass.

Endovascular procedures Since the introduction of transluminal dilation by Dotter, this field has grown enormously.69 The introduction of PTA has resulted in more indications for endovascular procedures to some extent at the expense of open surgical reconstructions. Percutaneous transluminal angioplasty In common with other vascular reconstructive procedures the success rate of PTA is highly dependent upon various factors. In general it can be said that proximal lesions – that is, iliac lesions – have a better success rate than distal ones – that is, femoropopliteal lesions. The chance of a successful outcome is higher for stenoses rather than occlusions, irrespective of the site of the lesion. In common with surgical vascular reconstructions, the outflow determines the outcome also for PTA. Thus, in cases of a good outflow, the results are better than if the outflow is poor.70 In summary, the chance of success is much higher when a short iliac stenosis is dilated in a patient with patent superficial femoral and profunda femoris arteries than after dilation of a popliteal occlusion in a patient with occlusions of two out of three crural arteries. The indications for this procedure need to be considered. PTA of an iliac stenosis in a patient with claudication has a low risk and a high chance of success and may, therefore, be perfectly appropriate, even if the severity of the disease state is relatively mild, as compared with a patient with critical limb ischemia and a threat of amputation. On the other hand, a patient with an occluded popliteal artery and poor leg run-off with ischemic ulceration has a strong indication


for the procedure and, in such a patient, it may also be perfectly appropriate to make an attempt at PTA, even though the success rate is relatively low. For patients with critical ischemia, PTA of infrapopliteal vessels has also been performed successfully and could even be used for short occlusions.71,72 Subintimal angioplasty has been advocated.73 The method implies that a guidewire enters the subintimal space and then re-enters the vessel distal to the occlusion, and the subintimal space is then dilated with the balloon. In the femoropopliteal segment, occlusions longer than 20 cm can be treated, whereas intraluminal angioplasty is generally not advocated for occlusions longer than a few centimeters. Patency rates of approximately 60% at 3 years for femoropopliteal occlusions have been reported after subintimal angioplasty.74 The reported figures are patency rates for technically successful procedures, but in 20% the procedure cannot be performed. The method has also been used for infrapopliteal arteries.75 Subintimal angioplasty, if proven successful, could be a future alternative to femoropopliteal bypass. Formal comparisons in prospective randomized trials between PTA and surgery are relatively scarce. Such trials are difficult because, in order for a patient to be included, the lesion has to be suitable for PTA – that is, it should be either a stenosis or a short occlusion. Knowing that the treatment of a stenosis with PTA is relatively successful with less risk and shorter hospital stay, it is sometimes considered ethically questionable to include patients in a trial between PTA and surgery. In a study including 263 patients with lesions in the iliac, femoral, or popliteal arteries comparing bypass surgery and PTA, primary success favored surgery, while limb salvage favored PTA, but the differences were not statistically significant. After 4 years there was no significant difference in outcome.76 Randomized trials comparing angioplasty with nonsurgical treatment for intermittent claudication have, however, been performed, but they are relatively small and the results are to some extent contradictory. In one study the treadmill distances improved in both groups but were superior in those undergoing an exercise program, and after 6 years there was no benefit in treadmill walking distance after angioplasty.77 In another study, an improvement in ABI was shown 6 months following angioplasty, which could not be found in patients undergoing exercise programs. Significantly more patients were asymptomatic after 6 months in the angioplasty group compared with those treated with exercise programs. This study, however, had a shorter follow up, and the conservative treatment was not as active as in the study where no difference could be seen between exercise program and PTA.78 It can still be concluded that PTA is suitable for stenoses or short occlusions in claudicants, but few claudicants have discrete lesions suitable for PTA.

Stenting has been used at an increasing rate over the last few years. It is generally advised not to use stents in smaller arteries, and this implies that stents are used relatively seldom in the femoropopliteal region. Stents, however, are used in the iliac arteries after PTA, particularly when there is recoil or dissection. Several types of stents have been used, both self-expandable and balloon-expandable ones. Stenting below the inguinal ligament is not generally recommended. Thrombolysis Thrombolysis of peripheral arterial occlusive disease is recommended for acute arterial occlusions, but it also has a place in subacute occlusions. Thrombolysis should be intra-arterial and preferably the thrombolytic agent should be delivered into the clot, either with an end hole catheter or a catheter with multiple side holes. Today recombinant tissue plasminogen activator (rtPA) is used most commonly. Other thrombolytic agents are, however, being developed and have been tried for indications other than peripheral arterial occlusive disease. The dosage and rate of administration of thrombolytic agents varies in different reports and makes comparisons difficult. There are, however, some prospective randomized trials comparing surgery with intraarterial thrombolytic therapy. In one representative study, the mean duration of ischemia was almost 2 months and patients were included if the duration was less than 6 months. Overall the study favored surgery. Patients randomized to catheter-directed thrombolysis had significantly greater ongoing or recurrent ischemia, life-threatening hemorrhage, and vascular complications compared with surgical patients. Stratification by duration of ischemia, however, showed that patients treated within 14 days of onset of symptoms had an amputation rate after thrombolysis of 6% compared to 18% for those undergoing surgery. Patients treated with thrombolysis in this group also had a shorter hospital stay. In patients with acute ischemia the amputation-free survival at 6 months follow up was also better in those treated with thrombolysis.79 Further analysis of this material reveals that thrombolysis provides a reduction of the predetermined surgical procedure in 50–60% of the cases.80 Grade B Key points ● ● ● ● ●

PTA is more successful for stenoses than for occlusions. PTA is more successful for short than for long occlusions. PTA may be combined with stent if recoil occurs or if PTA produces dissection with intimal flaps. Thrombolysis should be performed by local intrathrombal administration of the drug. PTA may be preceded by thrombolysis in cases with recent occlusions.

883


Inflammatory vascular diseases – thromboangitis obliterans (Buerger’s disease) Temporal arteritis, Takayashu’s disease of the aortic arch, and several diseases affecting the arterioles and microcirculatory vessels have an inflammatory or immunogenic origin. In this chapter these diseases are not considered. Thromboangitis obliterans, or Buerger’s disease, also has an inflammatory component, although the pathophysiology is still not fully known. The major pathogenetic factor, tobacco smoke, has been clearly established, however, for a long time. The patient is usually a young or middle aged man with excessive smoking habits. The disease is segmental and affects both veins and arteries leading to recurrent thrombophlebitis and, in more severe cases, to multiple ulcerations of toes and fingers owing to occlusion of distal arteries. Larger arteries are often affected, which in part may be due to concomitant atherosclerotic disease. The treatment is based on total avoidance of tobacco smoke. Treatment with prostaglandins, especially the synthetic prostacyclin analog Iloprost (see critical ischemia above), has been shown to have positive effects regarding pain alleviation and healing of ulcers.81

References 1.Rose G. The diagnosis of ischaemic heart pain and intermittent claudication in field surveys. Bull WHO 1962;27:645–58. 2.Fowkes F. The measurement of atherosclerotic peripheral arterial disease in epidemiological surveys. Int J Epidemiol 1988;17:248–54. 3.Bothig S, Metelisa V, Barth W et al. Prevalence of ischaemic heart disease, arterial hypertension and intermittent claudication, and distribution of risk factors among middle-aged men in Moscow and Berlin. Cor Vasa 1976;18:104–18. 4.Heliovaara M, Karvonen W, Vilhunden R, Punsar S. Smoking, carbon monoxide and atherosclerotic diseases. BMJ 1978;I: 268–70. 5.Dormandy J, Heeck L, Vig S. The natural history of claudication: risk to life and limb. Semin Vasc Surg 1999;12: 123–37. 6.Feinglass J, McCarthy W, Slavensky R, Manheim L, Martin G. Effect of lower extremity blood pressure on physical functioning in patients who have intermittent claudication. J Vasc Surg 1996;24:503–12. 7.Criqui M, Fronek A, Barrett-Connor E, Klauber M, Gabriel S, Goodman D. The prevalence of peripheral arterial disease in a defined population. Circulation 1985;71:510–15. 8.Hiatt W. Effect of diagnostic criteria on the prevalence of peripheral arterial disease. The San Luis Valley Diabetes Study. Circulation 1995;91:1472–9. 9.Kannel W, McGee D. Update on some epidemiological features of intermittent claudication. J Am Geriat Soc 1985;33: 13–18. 10.Dormandy J, Heeck L, Vig S. Predictors of early disease in the lower limbs. Semin Vasc Surg 1999;12:109–17.

884

11.Murabito JM, D’Agostino RB, Silbershatz H, Wilson WF. Intermittent claudication. A risk profile from The Framingham Heart Study. Circulation 1997;96:44–9. 12.Leng GC, Lee AJ, Fowkes FG et al. Incidence, natural history and cardiovascular events in symptomatic and asymptomatic peripheral arterial disease in the general population. Int J Epidemiol 1996;25:1172–81. 13.McGrath M, Graham A, Hill D et al. The natural history of chronic leg ischaemia. Wld J Surg 1983;7:314–18. 14.Jonason T, Bergstrom R. Cessation of smoking in patients with intermittent claudication: effects on the risk of peripheral vascular complications, myocardial infarction and mortality. Acta Med Scand 1987;221:253–60. 15.Kallero K. Mortality and morbidity in patients with intermittent claudicationas defined by venous occlusion plethysmography: a ten year follow up. J Chron Dis 1981;34:445–62. 16.Criqui M, Langer R, Fronek A et al. Mortality over a period of 10 years in patients with peripheral arterial disease. N Engl J Med 1992;326:381–5. 17.Leng G, Fowkes F, Lee A, Dunbar J, Housley E, Ruckley C. Use of ankle brachial pressure index to predict cardiovascular events and death: a cohort study. BMJ 1996;313:1440–4. 18.Reunanen A, Takkunen H, Aromaa A. Prevalence of intermittent claudication and its effect on mortality. Acta Med Scand 1982;211:249–56. 19.Rihal S, Eagle K, Mickel C et al. Surgical therapy for coronary artery disease among patients with combined coronary artery and peripheral vascular disease. Circulation 1995;91:46–53. 20.Leng GC, Price JF, Jepson RG. Lipid-lowering for lower limb atherosclerosis. Cochrane Database Syst Rev 2000;2. 21.The Heart Protection Study Presentation at AHA, 2001. www.heart-protection.com 22.Hiatt W, Stoll S, Nies A. Effect of beta-adrenergic blockers on the peripheral circulation in patients with peripheral vascular disease. Circulation 1985;72:1226–31. 23.Yusuf S, Sleight P, Pogue J, Bosch J, Davies R, Dagenais G. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. The Heart Outcomes Prevention Evaluation Study Investigators. N Engl J Med 2000;342:145–53. 24.Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy-I: Prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. BMJ 1994; 308:81–106. 25.Hess H, Mietaschik A, Deichsel G. Drug induced inhibition of platelet function delays progression of peripheral occlusive arterial disease: a prospective double blind arteriographic controlled trial. Lancet 1985;1:416–19. 26.Diener F, Coccheri S, Libretti A et al. European stroke prevention study 2. Dipyridamole and acetylsalicylic acid in the secondary prevention of stroke. J Neurolog Sci 1996;143: 1–13. 27.Bergqvist D, Almgren B, Dickinson J. Reduction of requirement for leg vascular surgery during long-term treatment of claudicants patients with Ticlopidine: Results from the Swedish Ticlopidine Multicenter Study (STIMS). Eur J Vasc Endovasc Surg 1995;10:69–76.


28.Janzon L, Bergqvist D, Boberg J et al. Prevention of myocardial infarction and stroke in patients with intermittent claudication: effects of ticlopidine, results from STIMS, the Swedish Ticlopidine Multicenter Study. J Intern Med 1990;227: 301–8. 29.Balsano F, Coccheri S, Libretti A et al. Ticlodipine in the treatment of intermittent claudication. A 21-month double-blind trial. J Lab Clin Med 1989;114:84–91. 30.CAPRIE Steering Committee. A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischemic events (CAPRIE). Lancet 1996;348:1329–39. 31.Hiatt W, Wolfel E, Meire R, Regesteiner J. Superiority of treadmill walking exercise versus strength training for patients with peripheral arterial disease. Implications for the mechanism of the training response. Circulation 1994;90:1866–74. 32.Gardner A, Poehlman E. Exercise rehabilitation programs for the treatment of claudication pain. A meta-analysis. JAMA 1995;274:975–80. 33.Lindgärde F, Jelnes R, Björkman H et al. Conservative drug treatment in patients with moderately severe chronic occlusive peripheral arterial disease. Circulation 1989;80:1549–56. 34.Girolami B, Bernardi E, Prins MH et al. Treatment of intermittent claudication with physical training, smoking cessation, pentoxifylline, or nafronyl: a meta-analysis. Arch Intern Med 1999;159:337–45. 35.Dawson DL, Cutler BS, Hiatt WR et al. A comparison of cilostazol and pentoxifylline for treating intermittent claudication. Am J Med 2000;109:523–30. 36.Hiatt WR. Medical treatment of peripheral arterial disease and claudication. N Engl J Med 2001;344:1608–21. 37.Scheffler P, de la Hamette D, Gross J, Müller H, Schieffer H. Intensive vascular training in stage IIb of peripheral arterial occlusive disease. The additive effects of intravenous prostaglandin E1 or intravenous pentoxiphylline during training. Circulation 1994;90:818–22. 38.The European Working Group on Critical Leg Ischaemia. Second European consensus document on chronic critical leg ischaemia. Circulation 1991;84(Suppl. 4):1–22. 39.Loosemore T, Chalmers T, Dormandy J. A meta-analysis of randomized placebo control trials in Fontaine stages III and IV peripheral occlusive arterial disease. Int Angiol 1994;13: 133–42. 40.Thompson M, Sayers R, Varty K, Reid A, London M, Bell P. Chronic critical leg ischaemia must be redefined. Eur J Vasc Surg 1993;7:420–6. 41.Group TW. Management of peripheral Arterial Disease. J Vasc Surg 2000;31(1, part 2). 42.Conrad M. Abnormalities of the digital vasculature as related to ulceration and gangrene. Circulation 1968;49:1196–201. 43.Kretschmer G, Herbst F, Prager M et al. A decade of oral anticoagulant treatment to maintain autologous vein grafts for femoropopliteal atherosclerosis. Arch Surg 1992;127:1112–15. 44.Jivegård LE, Augustinson LE, Holm J et al. Effects of spinal cord stimulation (SCS) in patients with inoperable severe lower limb ischemia: a prospective randomised controlled study. Eur J Vasc Endovasc Surg 1995;9:421–5. 45.The European Study Group. Intravenous pentoxiphylline for the treatment of chronic critical limb ischemia. Eur J Vasc Endovasc Surg 1995;9:426–36.

46.Klomp HM, Spincemaille GH, Steyerberg EW, Habbema JD, van Urk H. Spinal-cord stimulation in critical limb ischaemia: a randomised trial. ESES Study Group. Lancet 1999;353:1040–4. 47.Goldman L, Caldera D, Nussbaum S. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297:845–50. 48.Detsky A, Abrams H, Forbath N, Scott J, Hilliard J. Cardiac assessment for patients undergoing noncardiac surgery: a multifactorial clinical risk index. Arch Intern Med 1986;146: 2131–4. 49.Mantha S, Roizen M, Barnard J, Thisted R, Ellis J, Foss J. Relative effectiveness of four preoperative tests for predicting adverse cardiac outcome after vascular surgery: a metaanalysis. Anaesth Analg 1994;79:422–33. 50.Eagle K, Singer D, Brewster D, Darling R, Mulley A, Boucher C. Dipyridamole-thallium scanning in patients undergoing vascular surgery. JAMA 1987;257:2185–9. 51.Hertzer N, Beven E, Young J et al. Coronary artery disease in peripheral vascular patients. A classification of 1000 coronary angiograms and results or surgical management. Ann Surg 1984;199:222–3. 52.Mangano D, London M, Tubau J et al. Dipyridamole thallium201 scintigraphy as a preoperative screening test. Circulation 1991;84:493–502. 53.Mason J, Owens D, Harris D, Ryan A, Cooke J, Hlatky M. The role of coronary angiography and coronary revascularization before noncardiac vascular surgery. JAMA 1995;273: 1919–25. 54.Wong T, Detsky A. Preoperative cardiac risk assesment for patients having peripheral surgery. Ann Intern Med 1992;116:743–53. 55.Poldermans D, Boersma E, Bax JJ et al. The effect of bisoprolol on perioperative mortality and myocardial infarction in high-risk patients undergoing vascular surgery. Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography Study Group. N Engl J Med 1999;341:1789–94. 56.Mason R, Smirnov V, Newton G, Giron F. Alternative procedures to aortobifemoral bypass grafting. J Cardiovasc Surg (Torino) 1989;30:192–7. 57.Becker GJ, Katzen BT, Dake MD. Noncoronary angioplasty. Radiology 1989;170:921–40. 58.Johnston KW. Iliac arteries: reanalysis of results of balloon angioplasty. Radiology 1993;186:207–12. 59.Swedenborg J, Bergmark C. Is there a place for primary axillofemoral bypass? In: Greenhalgh R, Fowkes F, eds. Trials and tribulations of vascular surgery. London: WB Saunders Company Ltd, 1996. 60.Michaels J. Choice of material above-knee femoropopliteal bypass graft. Br J Surg 1989;76:7–14. 61.Veith F, Gupta S, Ascer E et al. Six-year prospective multicenter randomised comparison of autologous saphenous vein and expaneded polytetrafluoroethylene grafts in infrainguinal arterial reconstruction. J Vasc Surg 1986;3:104–14. 62.Hall K, Rostad H. In situ vein bypass in the treatment of femoropopliteal atherosclerotic disease. A ten year study. Am J Surg 1978;136:158–61. 63.Moody A, Edwards P, Harris P. In situ versus reversed femoropopliteal vein grafts long-term follow-up of a prospective, randomized trial. Br J Surg 1992;79:750–2.

885


64.Lundell A, Lindblad B, Bergqvist D, Hansen F. Femoropoplitealcrural graft patency is improved by an intensive surveillance program: a prospective randomized study. J Vasc Surg 1996; 21:26–33. 65.Ihlberg L, Luther M, Alback A, Kantonen I, Lepantalo M. Does a completely accomplished duplex-based surveillance prevent vein-graft failure? Eur J Vasc Endovasc Surg 1999;18: 395–400. 66.Tangelder MJ, Lawson JA, Algra A, Eikelboom BC. Systematic review of randomized controlled trials of aspirin and oral anticoagulants in the prevention of graft occlusion and ischemic events after infrainguinal bypass surgery. J Vasc Surg 1999;30:701–9. 67.Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy-II: maintenance of vascular graft or arterial patency by antiplatelet therapy. BMJ 1994;308:159–68. 68.The Dutch Bypass Oral Anticoagulants or Aspirin Study. Efficacy of oral anticoagulants compared with aspirin after infrainguinal bypass surgery: a randomised trial. Lancet 2000;355:346–51. 69.Dotter C, Judkins M. Transluminal treatment of arteriosclerotic obstructions. Circulation 1964;30:654–70. 70.Johnston K, Rae M, Hogg-Johnston S et al. 5-year results of a prospective study of percutaneous transluminal angioplasty. Ann Surg 1987;206:403–12. 71.Dorros G, Lewin R, Jamnadas P et al. Below-the-knee angioplasty: Tibioperoneal vessels, the acute outcome. Catheter Cardiovasc Diag 1990;19:170–8. 72.Sivananthan U, Browne T, Thorley P et al. Percutaneous transluminal angioplasty of the tibial arteries. Br J Surg 1994;81:1282–5. 73.Bolia A, Miles K, Brennan J, Bell P. Percutaneous transluminal angioplasty of occlusions of the femoral and popliteal arteries by

886

subintimal dissection. Cardiovasc Interven Radiol 1990; 13:357–63. 74.London N, Srinivasan R, Naylor A et al. Subintimal angioplasty of femoropopliteal artery occlusions: the long-term results. Eur J Vasc Surg 1994;8:148–55. 75.Nydahl S, London N, Bolia A. Technical report: recanalisation of all three infrapopliteal arteries by subintimal angioplasty. Clin Radiol 1996;51:366–7. 76.Wolf G. Surgery or balloon angioplasty for peripheral vascular disease: a randomized clinical trial. Principal investigators and their Associates of Veterans Administration Cooperative Study Number 199. J Vasc Intervent Radiol 1993;4:639–48. 77.Perkins J, Collin J, Creasy T, Fletcher E, Morris P. Exercise training versus angioplasty for stable claudication. Long and medium results of a prospective, randomised trial. Eur J Vasc Endovasc Surg 1996;12:167–72. 78.Whyman M, Fowkes F, Kerracher E et al. Randomised controlled trial of percutaneous transluminal angioplasty for intermittent claudication. Eur J Vasc Endovasc Surg 1996;12: 167–72. 79.The STILE Investigators. Results of a prospective randomized trial evaluating surgery versus thrombolysis for ischemia of the lower extremity. The STILE Trial. Ann Surg 1994;220: 251–68. 80.Weaver F, Comerota A, Youngblood M et al. Surgical revascularization versus thrombolysis for nonembolic lower extremity native artery occlusions: Results of a prospective randomized trial. J Vasc Surg 1996;24:513–23. 81.Fiessinger J, Schäfer M. Trial of Iloprost versus aspirin treatment for critical limb ischemia of thromboangitis obliterans. Lancet 1990;335:556–7.

Part IV Clinical applications Ernest L Fallen, Editor

887


GRADE A

GRADE C


Level 5

GRADE B Level 2

Level 3 Level 4

888




63

Clinical applications of external evidence Ernest L Fallen, Salim Yusuf

External evidence derived from randomized clinical trials (RCTs) provides the practicing physician with a sound, rigorous and secure basis for making management decisions on individual patients. However, even vociferous advocates of evidence-based medicine will caution against the use of external evidence as the sole criterion for treating all patients. It is well to bear in mind that evidence obtained from clinical trials is derived from large population databases. More often than not, the entry criteria tend to define the boundaries of specified interest (for example, acute myocardial infarction), whereas exclusion criteria, such as age, sex, comorbid disease states etc., may well have denied entry to one’s individual patient now awaiting treatment. These exclusions may be based on concerns related to patient safety, lack of applicability, historical considerations or confounders (for example, significant non-cardiac illness) that can affect the evaluation of treatment. Nevertheless, the practicing physician is left inquiring, “Where can I find my patient within the trial’s data set?” Here is where interpretation and the application of external evidence requires a logical integration of overall trial results with a knowledge of biologic mechanisms, patient risk and clinical circumstances.

Evidence-based v patient centered medicine? Not an either/or choice Few would deny an approach to therapeutic decision making based on proven external evidence combined with clinical experience, knowledge of pathophysiology and sensitivity to individual patient needs. To marry the two effectively is to recognize, and hence to avoid, their respective limitations if either were to be applied alone.

Recognizing the limitations of external evidence For most RCTs, proving therapeutic efficacy necessitates certain constraints in patient selection. It is not uncommon that many patients in a physician’s practice would not have fulfilled the restrictive entrance criteria of most moderate-size

RCTs. For example, some RCTs have an age cut off that actually excludes more than half of all patients with the disorder. This by no means implies that the reputed benefit of the test drug is not applicable (effective) to the patients excluded, but it does beg the question. Entry criteria alone should never be the sole basis for denying a patient the benefit of proven therapy. Interpatient variability is inevitable in all RCTs and contributes much to the “random errors” seen in small and moderate-sized trials. However, the larger the trial the smaller the random error, and the more likely that benefit can be reliably extrapolated to some patients who do not necessarily qualify for entry.1,2 For example, one may observe that the benefits are consistent across different subgroups, suggesting that the results may be applicable beyond the boundaries of patient selection, whereas on the other hand there may emerge reliable evidence for a lack of benefit in certain subgroups. Evidence-based medicine that depends solely on external evidence is disease oriented rather than patient oriented. In other words, the verifiability of RCT data is often dependent on having a given diagnosis, as opposed to a clinical spectrum of risk associated with the diagnosis. This is the socalled “labeling” dilemma. For example, patients labeled as having “acute coronary syndrome” simply because they present with chest pain associated with non-ST segment elevation are often treated alike in an RCT, whereas the clinical expression of this entity may encompass a wide range from very low- to very high-risk patients. Translating external evidence based solely on a unified diagnosis into practice guidelines or clinical pathways has the unfortunate consequence of making management decisions dependent on a label rather than the presenting clinical circumstances and risk of the underlying disorder. Another nagging problem with the “bottom line” of clinical trials is the emphasis on primary end points that are measurable. Statistical dependency on hard data such as mortality rates, prespecified clinical outcome events, rehospitalizations etc. fails to acknowledge the significance of clinically relevant “soft” data, such as impact on symptoms, quality of life, psychosocial wellbeing, attitudes, economic realities and patient preferences. Finally, clinical trials all have finite time limits and, not uncommonly, the duration may be inadequate to 889


assess long-term benefits and risks, especially for any new drug. In such cases the information from RCTs may have to be supplemented by other sources of non-randomized evidence.

Recognizing the limitations of patient centered medicine Who would have guessed that aspirin could reduce the relative risk of death and adverse coronary events in postmyocardial infarction patients by 25%? Or that blockers would be so effective in class II–III chronic heart failure? Or that inotropic agents, despite improving hemodynamics and clinical wellbeing in patients with advanced heart failure, do so at the expense of shortened survival? Or that some antiarrhythmic agents, although they achieve cosmetic cleansing of so-called malignant ventricular ectopy from the electrocardiogram, are potentially hazardous? Previously held concepts of disease mechanisms as the basis for initiating new therapies or persisting with old therapies have been challenged by clinical trials results. And so, what is apparent as a “logical” management approach to a given clinical problem may commit even the most experienced physician to inappropriate prescribing practices. Patient centered medicine is not a concept that is firmly rooted in empirical medicine.3 It does not guarantee that a physician, feeling secure in his or her realm of expertise, will be kept abreast of therapeutic advances based on clinical trial results. Unfortunately, this can lead to a tendency to persist in outmoded approaches. Surely a cogent argument can be made to blend the positive features of patient centered and evidence-based approaches through a constant awareness of their respective limitations.

Some principles of application Knowing the person who has the disease is as important as knowing the disease that the patient has.4 Clinical decision making ought to incorporate the three following ingredients: (a) intelligent use of external evidence based on well established clinical trial results and epidemiologic data whenever available; (b) clinical expertise, knowledge of fundamental mechanisms of disease, and willingness to listen to the testimony of one’s patients; and (c) sensitivity to patients’ preferences, values, needs and beliefs. 1.

890

It is well to bear in mind that for any given diagnosis (label), patients at the greatest risk of a disease will usually derive the greatest benefit from an established treatment, as the absolute benefit usually increases with risk whereas harm due to the treatment remains comparatively fixed across the risk spectrum.5 Therefore, to

2.

3.

4.

avoid the hazard of labeling it is critical to risk stratify one’s patient. It is only after one has listened carefully to the testimony of the patient, performed a proper examination and conducted the relevant tests that one can formulate a degree of attributable risk. Remember, it is just as important to identify the patient at very low risk, thereby sparing him or her unnecessary aggressive investigation and/or therapy, as it is to identify the high-risk patient for whom aggressive treatment can be life saving. The absence of external evidence should not lead to therapeutic nihilism. Not all consensus recommendations are supported by grade A evidence. In fact, many consensus panel recommendations and clinical practice pathways are based on evidence that ranges from the use of clinical judgment, albeit under a cloud of uncertainty, to grade B through grade A evidence.6,7 When external evidence is lacking, one’s own clinical experience, knowledge of pathophysiology, reasoned judgment and awareness of the patient’s needs are indispensable substitutes. One should avoid using the trial entry criteria to determine whether a particular patient would benefit from the active treatment.2,5 Failure to qualify for entry is determined by many factors, few of which necessarily compromise the potential for therapeutic benefit. For example, if a trial’s age cut off was 65 years then a reasonable risk/benefit assessment can be done for those older by assessing whether, within the trial, age modified the treatment effect. One should always try to use the best available external evidence science has to offer, but never at the expense of ignoring the patient’s psychosocial conditions, beliefs, values and preferences. As medical decisions become more codified one should not fail to recognize and honor the importance of patient preferences.8 A patient’s medical decision based on his or her particular needs, preferences and beliefs should always be respected, as the patient is given the opportunity to hear the nature of the external evidence. Consensus recommendations are guidelines only. They represent an active process subject to continual review as new and as yet untested information emerges. When following any recommendation based on external evidence the physician should always exercise clinical judgment based on a close working interaction with the patient.

Section preview The following case reports stress the importance of knowing how and when to apply best external evidence, not just to know what that evidence is. They represent an attempt to put a clinical face on a statistical bottom line by illustrating

Clinical applications of external evidence

practical solutions in the application of evidence-based medicine to individual patient problems. These case studies are real life presentations in which therapeutic decisions are either clearly guided by external evidence or require extra clinical reasoning skills in concert with best available evidence. From the files of these distinguished consultant cardiologists two different cases are presented for each of the 11 clinical topics. The first case in each series represents a clinical scenario where the management decision is unequivocally substantiated by use of the external best evidence. The second case is more complex and represents a challenge to incorporate external evidence with reasoned judgment, an experienced examination, a sound knowledge of cardiovascular pathophysiology, and sensitivity to the patient’s needs and preferences.

2.Yusuf S, Wittes HJ, Probstfield J, Tyroler HA. Analysis and interpretation of treatment effects in subgroups of patients in randomized clinical trials. JAMA 1991;266:93–8. 3.Bensing J. Bridging the gap. The separate worlds of evidence-based and patient-centered medicine. Patient Educ Counseling 2000;39:17–25. 4.McCormick J. Death of the personal doctor. Lancet 1996; 348:667–8. 5.Glasziou PP, Irwig LM. An evidence based approach to individualising treatment. BMJ 1995;311:1356–9. 6.Fallen EL, Cairns J, Dafoe W et al. Management of the postmyocardial infarction patient: a consensus report. Can J Cardiol 1995;11:477–86. 7.Hayward RS, Wilson MC, Tunis SR et al. User’s guide to the medical literature. VIII. How to use clinical practice guidelines. A. Are the recommendations valid? JAMA 1995;274:570–4. 8.Kassirer JP. Incorporating patients’ preferences in medical decisions. N Engl J Med 1994;330:1895–6.

References 1.Yusuf S, Held P, Teo KK. Selection of patients for randomized controlled trials: Implications of wide or narrow eligibility criteria. Stat Med 1990;9:73–86.

891

64

Stable angina: choice of PCI v CABG v drugs Douglas A Holder

Case scenario 1

While vacationing in Puerto Rico, a 51 year old Canadian woman sustains an uncomplicated inferior myocardial infarction. Upon returning to Canada she experiences angina up to four to five times per day, relieved by one or two nitroglycerin sprays. Her risk factors include a history of smoking, a positive family history of coronary artery disease, and a total cholesterol of 5·4 mmol/l (LDL-C 4·1; HDL-C 1·2) and triglycerides 1·34 mmol/l. Her medications are diltiazem 240 mg CD daily; transdermal nitroglycerin 0·4 mg/h patch ON in am and OFF hs; enteric coated aspirin 325 mg daily; salbutamol and beclomethosone dipropionate puffs. She is unable to take blockers because of increased airways resistance secondary to chronic smoking. A decision is made to proceed with coronary angiography with the view to revascularization. Coronary angiography which reveals a 90% stenosis of the midthird of a dominant right coronary artery (RCA) (Figure 64.1). Angiographically, the dominant RCA stenosis, being severe, non-calcified, and discrete, is technically suitable for percutaneous coronary intervention (PCI). After the second inflation with a 2·5 mm balloon the procedure is complicated by acute closure due to a spiral dissection which extends down well below the original stenosis towards the crux (Figure 64.2). A 2·5/28 mm stent and a 2·5/24 stent are deployed to tack up the intima from the distal end of the dissection back to and including the original stenosis. This results in an angiographically excellent result (Figure 64.3). The patient makes an uneventful recovery and is discharged on clopidogrel 75 mg daily in addition to her previous medications. She is well for 2 months but then develops recurrent angina. An exercise (treadmill) thallium study is positive at 521_ minutes with angina; 1 mm ST depression and reversible inferior wall ischemia on scanning. A repeat coronary angiogram reveals a discrete restenosis at the juncture of the two stents (Figure 64.4). The patient and her husband decide to choose coronary artery bypass surgery (CABG) rather than repeat PTCA.

Figure 64.1 A 90% discrete stenosis in a dominant right coronary artery (RCA)

892

Figure 64.2 After the second balloon inflation, note the spiral dissection extending down toward the crux

Stable angina

Figure 64.3 Post-stent deployment revealing adequate patency of the RCA

Figure 64.4 Restenosis at the juncture of the distal and middle stents 2 months after angioplasty

Question

the BARI study of multivessel disease showed no survival benefit for either treatment over a 5 year follow up period.3 Thus, in making the therapeutic recommendation at this stage, prognosis was not the main issue. Symptom relief was. The ACME trial compared medical treatment to PTCA for single vessel left anterior descending disease with the end point being anginal frequency and treadmill time at 6 months of follow up.4 In this trial of 107 patients, 46% of medically treated patients were free of angina compared to 64% of PTCA patients (P 0·01) and there was an increase in treadmill time, 2·1 minutes over baseline in the PTCA group compared to 0·5 minutes in the medically treated group (P 0·0001). However, because of restenosis, those patients assigned to PTCA had a more frequent requirement for further procedures (16 patients required PTCA; 2 required CABG). In another trial comparing medical treatment to PTCA to left internal mammary artery (LIMA) grafting for proximal single vessel left anterior descending artery stenosis 80% there were no differences among the groups in mortality or infarction rates, but no patient needed further revascularization in the surgical group compared to 8/72 (11%) of those undergoing PTCA and 7/72 (10%) on medical treatment (P 0·019).5 Acknowledging that clinical trials that specifically apply to our patient are lacking, these studies nevertheless allow us to conclude that surgery would be an acceptable choice for achieving symptomatic relief, and that PTCA is intermediate between medical treatment and surgery in achieving symptomatic relief but at a cost of a higher likelihood of further revascularization in the future. The other considerations in comparing surgery to PTCA are higher initial mortality and morbidity with surgery, as well as the fact that the patency of a saphenous vein graft in the circulation is less than that of a mammary artery. The angiographic characteristics of the stenosis were consistent with a high likelihood of primary success with PTCA, and if restenosis did not occur then the time when CABG might have to be done could be delayed.

Is there evidence to support these therapeutic steps?

Comment There were three decision points where the patient was presented with options for therapeutic interventions. 1 Evidence to recommend initial PTCA The patient presented with postmyocardial infarction angina due to single vessel RCA stenosis. To date, randomized controlled trials (RCT) comparing CABG to medical treatment have shown no survival benefit from surgery because of the low prognostic risk associated with single vessel right coronary disease.1,2 There have been no direct comparisons of PTCA with CABG in this subset of single vessel disease, but

2 Evidence for the decision to employ a “bail-out” stent Acute closure due to spiral dissection is a recognized complication, occurring in 1–2% of patients undergoing PTCA. The risk of surgical mortality in the setting of an emergency operation is approximately twice that of an elective procedure. Current stent design almost always allows prompt deployment of a stent which quickly tacks up the dissection and relieves the ischemia.6 Thus, the need for immediate surgery is rare. 3 Evidence for the decision for coronary bypass surgery Unfortunately, the patient developed restenosis within the stented segment of the RCA. This was discrete, distal to the 893


site of the original plaque, and very unlikely to dissect with repeat dilation because of the stent buttressing the vessel wall. Thus, this stenosis would have been amenable to either repeat PTCA or coronary bypass surgery (CABG). The clinical advice was to offer repeat PTCA. However, when the substance of the earlier discussion of PTCA versus medical therapy versus CABG was again reviewed, the patient and her husband opted for surgery as a more “definitive” method of dealing with her problem. She subsequently underwent single vessel bypass without incident and continues with secondary prevention therapy. References 1.Alderman EL, Bourassa M, Cohen LSE. Ten year follow-up of survival and myocardial infarction in the randomized Coronary Artery Surgery Study. Circulation 1990;82:1629.

Case scenario 2

Conclusion

A 63 year old man tells the following story. Approximately 121_ years previously he experienced a feeling of “indigestion” while walking. This led to a symptom limited exercise test that was considered positive by ECG criteria but was not accompanied by any symptoms. His symptom of “indigestion” did not recur and he continued to pursue outdoor activities such as canoe tripping, camping, and walking without limitation. Risk factors include type II diabetes mellitus and hypercholesterolemia. He is a non-smoker. His current medications are humulin insulin 70/30; 20 units ac breakfast and 18 units ac supper, pravastatin 40 mg qhs (total cholesterol now is 4·47 mmol/l; LDL 2·87; HDL 1·12; TG 1·05), acebutolol 100 mg po bid, and enteric coated aspirin 325 mg po daily. A stress MUGA scan reveals the following. Exercise duration 11 minutes and 31 seconds; maximum heart rate 144 per minute; maximum blood pressure 170/84 and no angina. The ECG shows 4·5 mm downsloping ST-segment at maximum stress. Ejection fraction: baseline 69%; 200 kpm 73%; 400 kpm 66%; 600 kpm 61%; 800 kpm 58%; post-exercise recovery 67% indicating a fall in EF at higher workloads. Left ventricular wall motion analysis demonstrated hypokinesis of the septum, posterolateral, and inferior walls. Silent ischemia. Suggest referral for coronary angiography. Coronary angiography reveals a normal left main coronary artery trifurcating into left anterior descending, intermediate, and circumflex branches (Figure 64.5). There is a “left main equivalent” distribution of disease with stenoses involving the LAD 75%, intermediate 90%, and circumflex 90%. The RCA is diffusely diseased with a maximal narrowing of 60% in the midthird segment. LV systolic function is normal at rest.

Question Is conservative medical management optimal at this point?

Comment Here, the decision rests, in part, on whether clear evidence is available to offer sound advice to an asymptomatic patient with objective evidence of three vessel disease and exercise induced silent ischemia. If this patient had symptoms of 894

2.European Coronary Surgery Study Group. Long-term results of prospective randomized study of coronary artery bypass surgery in stable angina pectoris. Lancet 1982;ii:1173. 3.The Bypass Angioplasty Revascularization Investigation (BARI) Investigators. Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. N Engl J Med 1996;335:217–25. 4.Parisi AF, Folland ED, Hartigan P, for the Veterans Affairs ACME Investigators. A comparison of angioplasty with medical therapy in the treatment of single vessel coronary artery disease. N Engl J Med 1992;326:10. 5.Hueb WA, Bellotti G, deOliveira SA et al. A prospective randomized trial of medical therapy, balloon angioplasty or bypass surgery for single proximal left anterior descending coronary stenosis. J Am Coll Cardiol 1995;26:1600. 6.Lincoff AM, Topol EJ, Chapekis AT et al. Intracoronary stenting compared with conventional therapy for abrupt closure complicating coronary angioplasty: a matched case control study. J Am Coll Cardiol 1993;21:866–75.

classic angina pectoris the therapeutic decision for recommending coronary bypass surgery (CABG) with the expectation of symptom relief and improved prognosis could be substantiated. If the patient had a definite left main stenosis, or depressed LV function, the argument for CABG is strongly made because in this setting CABG improves prognosis even in the absence of symptoms.1,2 This patient is somewhat unusual in that not only is he asymptomatic but he is capable of a good workload (800 kpm of supine bicycle exercise). However, there is convincing evidence of significant ischemia at this level of work with ST depression

Stable angina

(A)

(B)

(C)

Figure 64.5 Multivessel coronary artery stenoses suggesting a “left main equivalent” with (A) LAD 75%, (B) intermediate 90%, and (C) circumflex 90%. The RCA is diffusely diseased

of 4.5 mm and a reduction in ejection fraction from 69% at rest to 58% with maximal effort. There is also evidence of LV wall motion abnormality in multiple sites consistent with the extent of coronary artery disease noted on the angiogram. The question therefore is one of prognosis rather than symptom relief. Common sense alone argues that his myocardium would benefit from revascularization. There is evidence to suggest that the long-term prognosis of patients with documented silent ischemia is similar to those with symptomatic angina pectoris,3,4 and thus our treatment should be aimed at the coronary disease substrate, rather than the clinical chest pain syndrome. Given the left main equivalent distribution as well as the diseased RCA, this patient was referred for consideration of coronary bypass surgery. The nature of the left coronary disease is also amenable to PCI although the procedure is somewhat riskier because of the proximal nature of the plaque in the LAD vessel. In addition, the likelihood of restenosis is significant because of the need for multivessel stenting in this diabetic patient. If the initial enthusiasm for coated stents is borne out with further observation then a PCI approach might well be a reasonable alternative in the future. Currently, the decision to undergo CABG is supported by evidence gleaned from a careful review of all major trials on bypass surgery for different severities of coronary artery disease.5

References 1.Alderman EL, Bourassa M, Cohen LSE. Ten year follow-up of survival and myocardial infarction in the randomized Coronary Artery Surgery Study (CASS). Circulation 1990; 82:1629. 2.European Coronary Surgery Study Group. Long-term results of prospective randomized study of coronary artery bypass surgery in stable angina pectoris. Lancet 1982;ii:1173. 3.Lotan C, Lokovitsky L, Gilon D et al. Silent myocardial ischemia during exercise testing. Does it indicate a different angiographic and prognostic syndrome? Cardiology 1994;85:407. 4.Marwick TH. Is silent ischemia painless because it is mild? J Am Coll Cardiol 1995;25:1513–15. 5.Yusuf S, Zucker D, Peduzzi P et al. Effect of coronary artery bypass surgery on survival. Lancet 1994;344:563–70.

895

65

Acute coronary syndromes George J Philippides

Case scenario 1

A 64 year old diabetic woman presents to the emergency ward with left arm discomfort, diaphoresis, nausea, and shortness of breath lasting 10 minutes. She has had similar episodes intermittently for the past 6 months, lasting up to 2 minutes. However, these episodes were usually brought on by heavy exertion and relieved by resting. The episodes have become more frequent over the past month and have been occurring with minimal exertion during the past 48 hours. Physical examination is within normal limits except for a blood pressure of 150/90 mmHg and a heart rate of 110. The initial ECG and the one taken an hour later after resolution of symptoms show no ischemic ST or T wave changes. Serum troponin and creatine kinase-MB isoenzyme (CK-MB) levels are not elevated.

Question What initial and further measures should be taken in this patient?

Comment This case represents a good example of unstable angina without “classic” chest pain. The initial evaluation should aim to answer two important questions: First, do the presenting symptoms and signs represent ischemic heart disease? Second, is this patient at significant risk for an adverse clinical outcome? While myocardial ischemia usually causes deep, poorly localized chest or arm discomfort, some patients may have no chest discomfort but present with epigastric, jaw, arm, or neck discomfort. Other “atypical” symptoms that may suggest angina even in the absence of chest pain include dyspnea, nausea, vomiting, and diaphoresis. If these symptoms are brought on by emotional stress or physical exertion and are relieved by rest or nitroglycerin, they should be considered equivalent to angina. Atypical angina is more common in women than men, and more prevalent in elderly people and in patients with diabetes. It is important to remember 896

that new-onset or worsening dyspnea on exertion is the most common “anginal equivalent”, especially in elderly patients. Although this patient does not complain of classic exertional chest pain, the diagnosis of an unstable coronary syndrome is likely given her history of diabetes and exertional dyspnea (Table 65.1). The recent change in pattern culminating in ischemic symptoms with minimal exertion is of concern. However the patient’s presentation lacks other “high risk” characteristics that are associated with an increased short-term risk of death or non-fatal myocardial infarction (MI) (Table 65.2). These include: rest pain for greater than 20 minutes, transient ST segment changes of greater than 0·5 mV, and elevated serum cardiac enzymes. Current published guidelines suggest that she should be admitted to the hospital in a unit that offers continuous rhythm monitoring and careful observation for recurrent ischemia. Initial medical management should include an aspirin, nitrates, a blocker, and antithrombotic therapy in the form of full dose, IV unfractionated heparin (UFH) or subcutaneous low molecular weight heparin (LMWH).1 Aspirin remains the gold standard for antiplatelet therapy. In unstable angina trials, aspirin therapy significantly reduced the relative risk of fatal or non-fatal MI by 60%.2 An oral loading dose of 160–325 mg/day non-enteric

Acute coronary syndromes: management issues

Table 65.1 Likelihood that signs and symptoms represent an ACS secondary to CAD Feature

High likelihood Any of the following:

Intermediate likelihood Absence of high-likelihood features and presence of any of the following:

Low likelihood Absence of high- or intermediate-likelihood features but may have:

Chest or left arm pain or discomfort as chief symptom reproducing prior documented angina Known history of CAD, including MI

Chest or left arm pain or discomfort as chief symptom Age 70 years Male sex Diabetes mellitus

Probable ischemic symptoms in absence of any of the intermediate likelihood characteristics Recent cocaine use

Examination

Transient MR, hypotension, diaphoresis, pulmonary edema, or rales

Extracardiac vascular disease

Chest discomfort reproduced by palpation

ECG

New, or presumably new, transient ST-segment deviation (0·05 mV) or T wave inversion (0·2 mV) with symptoms

Fixed Q waves Abnormal ST segments or T-waves not documented to be new

T-wave flattening or inversion in leads with dominant R waves Normal ECG

Cardiac markers

Elevated cardiac TnI, TnT, or CK-MB

Normal

Normal

History

Reproduced with permission. ACC/AHA Guidelines for the Perioperative Cardiovascular Evaluation for Noncardiac Surgery. J Am Coll Cardiol 1996;27:910–48. Copyright 1996 by the American College of Cardiology and American Heart Association, Inc. Abbreviations: CAD, coronary artery disease; CK-MB, creatine kinase-MB isoenzyme; ECG, electrocardiogram; MI, myocardial infarction; MR, mitral regurgitation; TnI, troponin I; TnT, troponin T

formulation should be given initially followed by 75–160 mg/day maintenance therapy. In patients who are aspirin intolerant, the adenosine diphosphate receptor antagonist clopidogrel can be substituted. This oral antiplatelet agent was shown in the Clopidogrel versus Aspirin in Patient at Risk of Ischemic Events (CAPRIE) trial to reduce the risk of cardiovascular events in patients with established vascular disease by 8·7% compared with patients treated with aspirin.3 Recently, the Clopidogrel in Unstable angina to prevent Recurrent Events (CURE) trial showed that treatment with clopidogrel in addition to aspirin reduced the risk of future ischemic events compared to aspirin therapy alone.4 A loading dose of 300 mg clopidogrel should be given followed by 75 mg/day orally. The widespread use of oral, topical, and IV nitrate preparations in unstable angina is based on their wellestablished physiologic and clinical effects, rather than on data from large-scale randomized trials. Nitroglycerin (NTG) reduces myocardial oxygen demand and improves regional myocardial blood flow by dilating epicardial coronary vessels. Patients with unstable symptoms can be given three 0·4 mg NTG tablets sublingually. Intravenous NTG should be initiated and titrated as needed in patients with refractory symptoms.5

Receptor antagonists should be started early in all patients who do not have contraindications. Intravenous boluses of metoprolol 5 mg or propranolol 1 mg can be given slowly over 1 to 2 minutes and repeated every 5 minutes until the target heart rate of 50–60 is reached. Oral therapy can be initiated 30–60 minutes later. An overview of several small studies of blocker therapy in unstable angina suggests a small, but significant, reduction in the risk of progression to myocardial infarction.6 Rate controlling, non-dihydropyridine calcium-channel blockers (verapamil or cardizem) can be used as initial therapy for patients with contraindications to blockers, for patients who continue to have ischemic symptoms despite treatment with nitrates and blockers, and for patients with variant angina.7 In general short-acting dihydropyridine calcium-channel blockers such as nifedipine should be avoided, and special care must be taken when using any calcium-channel blocker in patients with depressed left ventricular function. Antithrombotic therapy with unfractionated heparin (UFH) or a low molecular weight heparin (LMWH) should be started immediately to reduce the risk of myocardial infarction, death, or recurrent ischemia. The most recent guidelines from the American College of Cardiology recommend 897


Table 65.2 Short-term risk of death or non-fatal MI in patients with unstable anginaa Feature

High risk At least one of the following must be present:

Intermediate risk No high-risk feature but must have one of the following:

Low risk No high- or intermediaterisk feature but may have any of the following features:

History

Accelerating tempo of ischemic symptoms in preceding 48 h

Prior MI, peripheral or cerebrovascular disease, or CABG, prior aspirin use

Character of pain

Prolonged ongoing (20 minutes) rest pain

Prolonged (20 min) rest angina, now resolved, with moderate or high likelihood of CAD Rest angina (20 min) relieved with rest or or sublingual NTG

Clinical findings

Pulmonary edema, most likely due to ischemia New or worsening MR murmur S3 or new/worsening rales Hypotension, bradycardia, tachycardia Age 75 years

Age 70 years

ECG

Angina at rest with transient ST-segment changes 0·05 mV Bundle-branch block, new or presumed new Sustained ventricular tachycardia

T-wave inversions 0·2 mV Pathological Q waves

Normal or unchanged ECG during an episode of chest discomfort

Cardiac markers

Elevated (e.g. TnT or TnI 0·1 ng/ml)

Slightly elevated (e.g. TnT 0·01 but 0·1 ng/ml)

Normal

New-onset or progressive CCS Class III or IV angina the past 2 weeks without prolonged (20 min) rest pain but with moderate or high likelihood of CAD (see Table 65.1)

a

Estimation of the short-term risks of death and non-fatal cardiac ischemic events in unstable angina is a complex multivariable problem that cannot be fully specified in a table such as this; therefore, this table is meant to offer general guidance and illustration rather than rigid algorithms. Reproduced with permission. ACC/AHA Guidelines for the Perioperative Cardiovascular Evaluation for Noncardiac Surgery. J Am Coll Cardiol 1996;27:910–48. Copyright 1996 by the American College of Cardiology and American Heart Association, Inc. Abbreviations: CABG, coronary artery bypass graft; CAD, coronary artery disease; CCS, Canadian Cardiovascular Society; MI, myocardial infarction; MR, mitral regurgitation; NTG, nitroglycerine; TnI, troponin I; TnT, troponin T.

60 units of UFH followed by 12 units/kg/hour infusion. Alternatively, the low molecular weight heparin enoxaparin 1 mg/kg SQ 2/day can be used. Two trials that studied over 7000 patients showed a roughly 20% reduction in the rate of death, MI, and need for urgent revascularization in those treated with enoxaparin rather than UFH.8 The biggest advantage of LMWH over UFH is ease of administration as monitoring of activated partial thromboplastin time (APTT) is not required. Subsequent management depends on the patient’s clinical course. Repeat ECG and cardiac marker measurements should be performed 6–12 hours after the onset of symptoms. 898

Elevated serum cardiac enzymes9 or recurrent ischemic symptoms,10 despite treatment with aggressive pharmacotherapy as described above, would necessitate early coronary angiography with an eye toward percutaneous or surgical revascularization. In the absence of these “high-risk features”, the patient can safely undergo non-invasive testing after the doses of nitrates and blocker agents have been titrated and the heparin has been stopped. Patients with inducible ischemia or severely depressed left ventricular function (LVEF 35%) should also be considered for coronary angiography (Figure 65.1).


• Clinical suspicion of acute coronary syndrome • Physical examination • ECG • Blood samples

HIGH RISK • Elevated troponin • Refractory/recurrent ischemia • Hemodynamic instability

• No persistent ST-segment elevation

• GP IIb-IIIa inhibitor • Coronary angiography

• Early post-MI unstable angina

• Aspirin • Nitrates • β Blockers

LOW RISK

• UFH/LMWH • Clopidogrel

Assess risk

• Normal troponin on admission and 12 hours later

Stress test before discharge

Figure 65.1 Recommended strategy in unstable angina/non-ST elevation MI Abbreviations: UFH, unfractionated heparin; LMWH, low molecular weight heparin Adapted with permission from Bertrand ME et al. Management of acute coronary syndromes without persistent ST segment elevations: Recommendations of the Task Force of the European Society of Cardiology. Eur Heart J 2000;21:1424

References 1.Braunwald E, Antman EM, Beasly JW et al. ACC/AHA Guidelines for the management of patients with unstable angina and non-ST segment elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2000;36:970–1062. 2.Theroux P, Ouimet H, McCans J et al. Aspirin, heparin, or both to treat acute unstable angina. N Engl J Med 1988;319: 1105–11. 3.CAPRIE Steering Committee. A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischemic events. Lancet 1996;348:1329–39. 4.The Clopidogrel in Unstable Angina to Prevent, Recurrent Events Trial Investigators. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without STsegment elevation N Engl J Med 2001;345:494–502. 5.Conti CR. Use of nitrates in unstable angina pectoris. Am J Cardiol 1987;60:31H–34H. 6.Yusuf S, Wittes J, Friedman L. Overview of results of randomized clinical trials in heart disease, II: unstable angina, heart failure, primary prevention with aspirin, and risk factor modification. JAMA 1988;260:2259–63.

7.Held PH, Yusuf S, Furberg CD. Calcium-channel blockers in acute myocardial infarction and unstable angina: an overview. BMJ 1989;299:1187–92. 8.Antman EM, Cohen M, Radley D et al. Assessment of the treatment of effect of enoxaparin for unstable angina/ non-Q-wave myocardial infarction: TIMI 11B-ESSENCE meta-analysis. Circulation 1999;100:1602–8. 9.Cannon CP, Weintraub WS, Demopoulos LA et al. for the TACTICS-Thrombolysis in Myocardial Infarction 18 Investigators. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban. N Engl J Med 2001;344:1879–87. 10.Boden WE, O’Rourke RA, Crawford MH et al. Veterans Affairs Non-Q Wave Infarction Strategies in Hospital (VANQWISH). Outcomes in patients with acute non-Q wave myocardial infarction randomly assigned to an invasive as compared with a conservative management strategy. N Engl J Med 1998;338:1785–92. 11.Bertrand ME, Simoons ML, Fox KAA et al. Management of acute coronary syndromes: acute coronary syndromes without persistent ST segment elevation: Recommendations of the Task Force of the European Society of Cardiology. Eur Heart J 2000;21:1406–32.

899


Case scenario 2

A 77 year old male with a longstanding history of stable, class I angina is seen in the emergency ward with non-exertional precordial chest pain radiating to the left shoulder and dyspnea for 30 minutes. He has had similar episodes over the last year lasting up to 5 minutes but these were brought on by exertion and relieved with rest or sublingual nitroglycerin. On examination the blood pressure is 100/50, the heart rate is 120 bpm. The physical examination reveals rales to the mid lung fields bilaterally. The initial ECG is shown in Figure 65.2. The serum troponin level is 6·8.

I

aVR

V1

V4

II

aVL

V2

V5

III

aVF

V3

V8

RHYTHM STRIP: II 25 mm/sec: cm/mV

Figure 65.2

ECG of Case 2

Comment This patient presents with a classic acute coronary syndrome, and a diagnosis of non-ST-elevation MI (NSTEMI) is confirmed by an elevated serum troponin level. As usual, prompt treatment with a regimen of aspirin, heparin, nitrates, and blockers is mandated in order to decrease the risk for recurrent ischemic events. However several aspects of this patient’s clinical presentation suggest that he remains at high risk for death or myocardial infarction despite appropriate medical therapy. These “high-risk features” include advanced age, an accelerating tempo of ischemic symptoms over the preceding 48 hours, rest pain for greater than 20 minutes, evidence of pulmonary edema or hypoperfusion, new or worsening mitral regurgitation, dynamic ST changes 1 mm, and elevated serum cardiac markers.1–3 A review of the randomized clinical trials of GP IIb-IIIa receptor antagonists suggests that these agents should be administered, in addition to aspirin and heparin, in patients 900

with many of these high-risk clinical features, in patients with continuing ischemia, and in patients who are scheduled for percutaneous coronary intervention (PCI).4,5 Patients with elevated serum troponin levels appear to derive the greatest benefit from GP IIb-IIIa inhibitor therapy. While the benefits of GP IIb-IIIa inhibition appear to be greatest in patients undergoing PCI, several trials have shown that GP IIb-IIIa inhibitors are also effective in reducing the rate of ischemic events in the acute, “pre-cath” phase of medical management, and this benefit is maintained and heightened if a PCI is subsequently performed.6 Furthermore, similar to data from clinical trials of thrombolytic therapy in acute ST elevation MI, patients with unstable angina and NSTEMI who are treated the earliest after symptom onset with a GP IIb-IIIa inhibitor appear to derive the most benefit.7 Early coronary angiography should also be strongly considered. The Thrombolysis in Myocardial Infarction (TACTICS-TIMI-18) Trial randomized 2220 patients with


unstable angina or NSTEMI to an early invasive strategy, which included cardiac catheterization within 48 hours, and revascularization if appropriate versus a conservative strategy, in which catheterization was performed only if spontaneous or inducible ischemia was observed. All patients were treated with aspirin, heparin, and the glycoprotein IIb-IIIa inhibitor tirofiban. Early invasive therapy was associated with a significantly reduced rate of death or MI at 6 months (7·3% v 9·5%; OR, 0·74; P 0·05).8 Early coronary angiography in patients with acute coronary syndromes yielded similar reductions in ischemic events in the Fragmin and Fast Revascularization during Instability in Coronary Artery Disease (FRISC) II trial.9 Both trials found that the benefit of an early invasive strategy was greatest in intermediate- and high-risk patients with elevated troponin levels and/or ischemic ST changes on the admission electrocardiogram. Based on these recently published clinical trials and practice guidelines, this “high-risk” patient should be treated with an aggressive antithrombotic and antiplatelet regimen that includes an intravenous glycoprotein IIb-IIIa inhibitor in addition to aspirin and heparin. An early invasive strategy consisting of coronary angiography with an eye toward revascularization if necessary should also be pursued (Figure 65.1). Competing interest: the author has received educational grants for Aventis, and has spoken on unstable angina at events sponsored by Aventis. References 1.Braunwald E, Antman EM, Beasly JW et al. ACC/AHA guidelines for the management of patients with unstable angina and non ST-segment elevation myocardial infarction: a report of the

American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2000;36:970–1062. 2.Antman EM, Corbalan R, Huber K, Jaffe AS. Issues in early risk stratification for UA/NSTEMI. Eur Heart J Supplements 2001;3(Suppl. J):J6–J14. 3.Antman EM, Cohen M, Bernick P et al. The TIMI risk score for unstable Angina/non-ST-elevation MI: a method for prognostication and therapeutic decision making. JAMA 2000; 284:835–42. 4.The PRISM-PLUS Investigators. Inhibition of the platelet glycoprotein IIb/IIIa receptor with tirofiban in unstable angina and non-Q-wave myocardial infarction. N Engl J Med 1998; 338:1488–97. 5.The PURSUIT Trial Investigators. Inhibiton of platelet glycoprotein Ilb/IIIa with eptifibatide in patients with acute coronary syndromes. N Engl J Med 1998;339:436–43. 6.Boersma E, Akkerhuis KM, Theroux P et al. Platelet glycoprotein IIb/IIIa receptor inhibition in non-ST-elevation acute coronary syndromes: early benefit during medical treatment only, with additional protection during percutaneous coronary intervention. Circulation 1999;100:2045–8. 7.Bhatt DL, Marso SP, Houghtaling P et al. Does earlier administration of eptifbatide reduce death and MI in patients with acute coronary syndromes? Circulation 1998;98:1560–1. 8.Cannon CP, Weintraub WS, Demopoulos LA et al. for the TACTICS-Thrombolysis in Myocardial Infarction 18 Investigators. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban. N Engl J Med 2001;344:1879–87. 9.Fragmin and Fast Revascularization during Instability in Coronary artery disease (FRISC II) Investigators. Invasive compared with non-invasive treatment in unstable coronary-artery disease: FRISC II prospective randomised multicentre study. Lancet 1999;354:708–15.

901

66 Case 1

Acute myocardial infarction Bryan F Dias, Ernest L Fallen

A 68 year old woman with an 18 month history of chronic stable effort angina presents to the emergency room of a community hospital with a 2 hour history of increasing retrosternal chest pain associated with weakness, diaphoresis and nausea. Three applications of nitroglycerin spray 5 minutes apart fail to offer relief. She is a well controlled type 2 diabetic on glyburide 5 mg od. She also has esophageal acid reflux disease but clearly distinguishes her reflux symptoms from angina. There is no history of gastrointestinal bleeding. On examination she is pale, anxious and diaphoretic. Her blood pressure is 110/80 in both arms and her pulse is 70 and regular. Her neck veins are elevated 3 cm at 45 degrees, with a sustained hepatojugular reflux. She has a soft late crescendo apical systolic murmur. Her lungs are clear. An ECG is immediately ordered and reveals the following (Figure 66.1).

I

aVR

V1

V4

II

aVL

V2

V5

III

aVF

V3

V6

II

Figure 66.1 Twelve-lead ECG showing hyperacute ST segment elevation in leads II, III and AVF, signifying acute inferior wall ischemic injury/infarction

Question What is the proper course of action at this juncture?

Comment There is overwhelming evidence from major clinical trials that early intervention with thrombolytic therapy, combined 902

with aspirin for acute myocardial infarction, substantially reduces mortality and the risk of recurrent ischemic events.1–4 Does this case fit the entrance criteria? The symptoms of prolonged chest pain unresponsive to nitroglycerin, together with accompanying symptoms of nausea, weakness and diaphoresis, raise a strong suspicion of acute myocardial infarction. This is substantiated by the ECG findings of hyperacute ST segment elevation in leads II, III and AVF, signifying an acute inferior wall myocardial infarction.

Acute myocardial infarction

With an evolving infarction less than 3 hours from pain onset there is Grade A evidence that the treatment of choice is prompt coronary thrombolysis plus aspirin.1–4 The patient is therefore given aspirin 160 mg to chew while an intravenous line is inserted. After confirmation that the patient has no contraindication (recent bleed, stroke, trauma, surgery etc.) she is given streptokinase 1·5 million units intravenously over 1 hour. There is no evidence that rtPA (recombinant tissue plasminogen activator) alteplase or reteplase is superior to the less expensive streptokinase in patients with first-onset acute inferior myocardial infarction.5,6 The patient is monitored in the coronary care unit, where relief of anxiety and pain is achieved by administering oxygen, intravenous nitroglycerin and morphine as needed. She is prescribed an oral blocker (metoprolol 50 mg bid), which will be continued indefinitely.7–9 Because there is no major complication, such as congestive heart failure or intermittent chest pain, the patient will continue on aspirin and the blocker without the need for full-dose or low molecular weight heparin.5,6 There is also no echocardiographic evidence of significant left ventricular dysfunction (her estimated ejection fraction is 40%) to warrant an ACE inhibitor during the acute phase of the infarction.10 The community hospital where she is admitted does not have coronary angiographic facilities; the nearest hospital with such facilities is several hours away. If such facilities were immediately available this patient would be a candidate for primary coronary intervention (PCI), that is angioplasty with stent.11

Case 2

References 1.GISSI trial. Effectiveness of intravenous thrombolytic treatment in acute myocardial infarction. Lancet 1986;1:397–401. 2.ISIS II trial. Randomized trial of intravenous streptokinase, oral aspirin, both or neither among 17,187 cases of suspected acute myocardial infarction. Lancet 1988;ii:349–60. 3.FTT Collaborative group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomized trials of more than 1000 patients. Lancet 1994;343:311–22. 4.The GUSTO Investigators. An international randomized trial comparing four thrombolytic strategies for acute myocardial infaction. N Engl J Med 1993;329:673–82. 5.Cairns J, Armstrong P, Belenkie I et al. Canadian Cardiovascular Society Consensus Conference on coronary thrombolytics – 1994 update. Can J Cardiol 1994;10:517–29. 6.Collins R, Peto R, Baigent C, Sleight P. Aspirin, heparin and fibrinolytic therapy in suspected acute myocardial infarction. N Engl J Med 1997;336:847–60. 7.BHAT Study Group. A randomized trial of propanolol in patients with acute myocardial infarction. JAMA 1982;247:1707–14. 8.The Norwegian Multicenter Study Group. Six year follow up on Timolol after acute myocardial infarction. N Engl J Med 1985;313:1055–8. 9.Yusuf S, Peto R, Lewis JA et al. Beta blockade during and after myocardial infarction: a review of randomized trials. Prog Cardiovasc Dis 1985;27:335–71. 10.Syed M, Borzak S, Jafri SM et al. ACE Inhibition after myocardial infarction with special reference to the ISIS-4 Trial. Prog Cardiovasc Dis 1996;39:201–6. 11.Ryan TJ et al. ACC/AHA guidelines for the management of patients with acute myocardial infarction. J Am Coll Cardiol 1999;34:890–909.

A 66 year old man presents to the emergency room with severe chest pain. He has had two previous coronary artery bypass operations (2 and 12 years ago) and a history of two previous myocardial infarctions prior to his last bypass. For the past year his functional status has been stable, with Canadian Cardiovascular Society class I angina. He is now 3 hours following the onset of retrosternal chest pain unrelieved by sublingual nitroglycerin (0·6 mg 3). The pain is characterized as 10/10 severity, crushing in nature, with radiation to the jaw. Associated symptoms include dyspnea, diaphoresis and weakness. These symptoms are similar to those he had with the onset of his previous infarctions. Physical examination reveals an acutely ill-looking man. He is in painful distress, pale and clammy. His blood pressure is 100/80, equal in both arms. His pulse is 90 bpm and regular. Lung fields are clear. His neck veins are not abnormally elevated. There are no murmurs or extra heart sounds. His ECG is shown in Figure 66.2.

Comment Question What action would you now take?

Here is an example where strict adherence to clinical pathway algorithms based on the results of clinical trials can be 903


I

aVR

V1

V4

II

aVF

V2

V5

III

aVF

V3

V6

II

40 40Hz Hz 10 10 mm mm mV mV 25 25 mm/s mm/s 2433 2433 C C

Figure 66.2 Twelve-lead ECG showing widespread ST segment depression with left axis deviation. There are no Q waves nor ST segment elevation, except in lead AVR.

misleading. In this case the widespread ST segment depression led to the initial diagnosis of subendocardial ischemia, manifested as either unstable angina or non-ST segment elevation (NSTEMI) infarction (so-called acute coronary syndrome, or ACS). The use of thrombolysis for unstable angina or NSTEMI infarction has been studied but results show a trend towards no benefit or possible harm.1,2 The patient was therefore given aspirin 325 mg and full-dose intravenous heparin, starting with a 7units/kg bolus followed by 1200 units/h IV. Within the next hour neither sublingual nor intravenous nitroglycerin had any effect on the pain, which persisted at 10/10 severity. Similarly, an intravenous blocker (propanolol 5 mg IV 3 every 10 min) and morphine (10 mg IV over 30 min) had little effect. The pain persisted unabated, and serial ECGs showed ongoing ischemic changes but no evidence of injury (ST elevation). This patient initially received the standard treatment for the diagnosis of acute coronary syndrome (ACS), namely aspirin and an antithrombotic (unfractionated heparin in this case3–5), and yet these measures, plus intensification of the antianginal therapy, failed to relieve his symptoms. One could make a case for a different antithrombotic/antiplatelet approach. Studies examining low molecular weight heparin (LMWH) versus unfractionated heparin (UHF) in patients presenting with ACS have shown equivalency.6,7 As for an additional antiplatelet agent, clopidogrel, an adenosine 904

diphosphate receptor antagonist, when combined with aspirin in patients with ACS, showed a relative risk reduction of death, myocardial infarction or stroke by 18% compared to aspirin alone in the CURE Study.8 The use of glycoprotein IIb/IIIa receptor antagonists (eptifibatide or tirofiban) has been shown to be effective in acute ischemic situations when added to aspirin and heparin.9,10 However, there is a strong suspicion that this patient is actually suffering an acute myocardial infarction with total coronary artery occlusion. He had experienced identical symptoms with his previous infarctions; he is acutely ill with diaphoresis, weakness and restlessness – symptoms that are characteristic of myocardial necrosis, as opposed to reduced perfusion. On reflection the ECG changes may be construed as misleading in view of his chronic history of multiple ischemic insults. A decision was therefore made to proceed immediately with thrombolytic therapy. Within 45 minutes following IV infusion of rtPA the patient’s pain had completely abated and his ECG normalized, with only persistent T wave negativity in the anterior leads. Subsequent investigations revealed a peak creatinine kinase of 2969 with a strongly positive CKMB fraction. The patient went on to an uneventful recovery and was discharged from hospital on day 7. Although the evidence from clinical trials would not necessarily support the routine use of thrombolytic agents based on the ECG changes seen in this case, here is an example

Acute myocardial infarction

where exclusive reliance on a test (ECG) without appreciating clear clinical signs of myocardial necrosis, due probably to an occlusive thrombus, can result in misdirected therapy. A case could be made for primary angioplasty should facilities be available. However, in view of the probability of extensive three-vessel coronary disease and multiple blocked bypass grafts it would be more prudent to consider invasive investigation and intervention on an elective basis.

References 1.Freeman MR, Langer A, Wilson RF, Morgan CD, Armstrong PW. Thrombolysis in unstable angina: A randomized double blind trial of tPA and placebo. Circulation 1992;85:150–7. 2.The TIMI-IIIB Investigators. Effects of tissue plasminogen activator and a comparison of early invasive and conservative strategies in unstable angina and non-Q infarction. Results of the TIMI-IIIB Trial. Circulation 1994;89:1545–56. 3.Theroux P, Ouimet H, McCans et al. Aspirin, heparin or both to treat unstable angina. N Engl J Med 1988;319:1105–11. 4.Cairns JA, Gent M, Singer J et al. Aspirin, sulfinpyrozone, or both in unstable angina. Results of a Canadian multicenter trial. N Engl J Med 1985;313:1369–75.

5.Braunwald E, Antman EM, Beasley JW et al. ACC/AHA guidelines for the management of patients with unstable angina and non-ST elevation myocardial infarction. J Am Coll Cardiol 2000;36:970–1062. 6.Eikelboom JW, Anand SS, Malmberg K et al. Unfractionated heparin and low molecular weight heparin in acute coronary syndrome without ST elevation: a metaanalysis. Lancet 2000; 355:1936–42. 7.Antman EM, McCabe CH, Gurfunkel EP et al. Enoxaparin prevents death and cardiac ischemic events in unstable angina/non-Q wave infarction. Results of the thrombolysis in myocardial infarction (TIMI IIB trial). Circulation 1999; 100:1593–601. 8.The CURE Investigators. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST elevation. N Engl J Med 2001;345:494–502. 9.The Pursuit trial Investigators. Inhibition of platelet glycoprotein IIb/IIIa with eptifibatide in patients with acute coronary syndromes. N Engl J Med 1998;339:436–43. 10.PRISM Study Investigators. A comparison of aspirin plus tirofiban with aspirin plus heparin for unstable angina. N Engl J Med 1998;339:1498–505.

905

67

Postmyocardial infarction: preventive measures Ernest L Fallen

Case 1

A 53 year old sedentary, overweight male chartered accountant, previously symptom free, was admitted to hospital 6 days ago with an acute anteroseptal infarction. He received thrombolysis with rtPA (total dose 100 mg). The acute phase of his illness was complicated by frequent isolated ventricular ectopic beats (10/h) and mild cardiac decompensation (Killip class IIa). His risk factors include a positive family history, a 30 pack year smoking history, mild hypertension and obesity (body mass index BMI 29). His lipid status is unknown, except for a single LDL-C 3·0 mmol/l at the time of hospital admission. He is now ready for discharge and is free of heart failure and chest pain. His ECG shows a sinus rhythm at 65 bpm with qS waves and T inversion in leads V1–V3 without ST elevation. His echocardiogram reveals severe hypokinesis of the septum and apex, but no other regional wall abnormalities. There is no left ventricular dilatation and his estimated ejection fraction is 45%. There is no valvular abnormality and no endocardial thrombus. On physical examination he is in sinus rhythm with a blood pressure of 122/80 and a supine respiratory rate of 16 breaths per minute. His lungs are clear. There is a soft S4 but no murmurs. He has good peripheral pulses, no neck vein distension and no arterial bruits.

Question

Secondary prophylaxis

What advice would you now give him?

To reduce his risk of recurrent ischemic events and death he should continue indefinitely on enteric coated aspirin 81–325 mg/day1 and a blocker.2,3 In view of his reduced LV function a number of clinical trials support the use of an ACE inhibitor.4–6

Comment Here is a case where a physician’s advice to an otherwise recalcitrant patient is strongly fortified by clear evidence that preventive strategies post acute myocardial infarction yield favorable outcomes. The patient is now chest pain and failure free. He has one non-modifiable risk factor (family history) and several modifiable ones (smoking, sedentary lifestyle, obesity, and a propensity for hypertension). His lipid status is unknown, although the LDL-C of 3·0 on the first day of his infarct raise suspicion of a hyperlipidemic state. Evidence from clinical trials ought to persuade the patient that a preventive strategy of pharmacologic prophylaxis, combined with risk factor modification and lifestyle changes, can reduce his likelihood of suffering a major event in the foreseeable future. 906

Risk stratification For risk stratification he should be scheduled for a symptomlimited exercise test in 2–4 weeks, with a view to (a) assessing his exercise capacity; (b) determining whether he is high, intermediate or low risk; and (c) ruling out severe underlying ischemia that might warrant early coronary angiography. For instance, a low-risk patient is one who can achieve more than 8 METS of exercise with no chest pain or ST segment changes. A high-risk patient would be one who either experiences angina or significant ST segment depression at a low workload, or whose blood pressure either fails to rise or decreases during exercise.

Postmyocardial infarction: preventive measures

Risk factor modification/lifestyle changes A proper rehabilitation program7 should include (a) an exercise program; (b) nutrition counseling; (c) smoking cessation strategies; and (d) maintenance of ideal weight (BMI 25). He should have his lipids checked (total cholesterol, LDL-C, HDL-C and triglycerides) in about 4–6 weeks. Efforts should be made to ensure that his LDL-C is less than 2·6 mmol/l and his total cholesterol to HDL-C ratio is less than 4·5 (see Chapters 12, 13 and 36). In summary, using evidence-guided recommendations8,9 the patient’s chance of avoiding recurrent ischemic events and returning to a satisfactory quality of life is significantly enhanced by adherence to a program consisting of daily aspirin, blocker, a statin and an ACE inhibitor, combined with regular exercise and cessation of smoking. Our patient has shown signs of depression, which is another significant risk factor, but it has not yet been confirmed with certainty that antidepressants reduce the risk of postinfarction ischemic events. References 1.ISIS 2 Trial. Randomized trial of intravenous streptokinase and aspirin, both or neither among 17,187 cases of suspected acute myocardial infarction. Lancet 1988;ii:349–60.

Case 2

2.Pederson TR (for the Norwegian Multicenter Study Group). Six year follow up of the Norwegian Multicenter Study on Timolol after acute myocardial infarction. N Engl J Med 1985;313:1055–8. 3.Yusuf S, Peto R, Lewis JA et al. Beta blockade during and after myocardial infarction: a review of the randomized trials. Prog Cardiovasc Dis 1985;27:335–71. 4.ISIS-4 Study Group. A randomized factorial trial assessing early oral captopril, oral mononitrate, oral magnesium sulphate in 58 050 patients with suspected acute myocardial infarction. Lancet 1995;345:669–85. 5.The SOLVD Investigators. Effect of enalapril on survival inpatients with reduced left ventricular ejection fraction and congestive heart failure. N Engl J Med 1991;325:293. 6.The Heart Outcomes Prevention Evaluation Study investigators (HOPE). Effects of an angiotensin converting enzyme inhibitor, ramipril, on cardiovascular events in high risk patients. N Engl J Med 2000;342:145–53. 7.Oldridge NB, Guyatt GH, Fischer ME et al. Cardiac rehabilitation after myocardial infarction. Combined experience of randomized clinical trials. JAMA 1988;260:945–50. 8.Fallen EL, Cairns JA, Dafoe W, Frasure-Smith N et al. Management of the post myocardial infarction patient: a consensus report – revision of 1991 CCS guidelines. Can J Cardiol 1995;11:477–86. 9.Ryan TJ et al. 1999 Update ACC/AHA guidelines for the management of patients with acute myocardial infarction. J Am Coll Cardiol 1999;34:890–9.

A 66 year old woman sustains an acute inferoposterior myocardial infarction for which she receives streptokinase (1·5 million units over 1 hour) and aspirin to chew. She makes a satisfactory recovery. The only complications are persistent but asymptomatic sinus bradycardia (45–50/min), papillary muscle dysfunction with moderate mitral regurgitation, and an estimated left ventricular ejection fraction of 0·35. There are no overt signs of heart failure or ventricular irritability. She is angina free post MI. Her risk factors include heavy smoking (40 pack years); severe peripheral vascular disease with asymptomatic carotid bruits (80% stenosis of the right internal carotid artery), and intermittent claudication of the right leg. She has elevated triglycerides, at 6·5 mmol/l. Her previous total cholesterol (3 months ago) was 5·8 mmol/l and HDL-C was 0·98. She is 22 years post hysterectomy and bilateral oophorectomy. That operation was complicated by phlebitis and a pulmonary embolus. Her only medication at the time of hospital admission was an H2 receptor antagonist because of a history of upper gastrointestinal bleeding 12 years ago and episodic epigastric burning. There is mild renal impairment, with a creatinine of 180. There is no family history of heart disease. However, her mother died of breast cancer at age 65. Physical examination at the time of hospital discharge shows a resting heart rate of 48 bpm. Her blood pressure is 160/85. Her urea is 9·5 and creatinine 180. An ECG shows sinus rhythm but with a first-degree atrioventricular block of 0·24 seconds.

Question What advice would you now give her?

Comment This case raises several issues regarding the risk versus benefit of routine prophylactic strategies which, for most post-MI 907


patients, are supported by strong evidence from well designed clinical trials. For instance, one would wish to be cautious about reflexly prescribing a blocker. Remember, she has persistent bradycardia with partial AV block, symptomatic peripheral vascular disease, hyperlipidemia and impaired LV function. On the other hand, blockers have been shown to be especially effective in reducing mortality in post-MI patients with mild to moderate LV dysfunction.1 Here the benefit of a blocker probably outweighs the overall risk because it has the combined effect of protecting the ischemic myocardium, reducing the risk of sudden cardiac death and stabilizing blood pressure. In view of her LV dysfunction, she would also be a candidate for an ACE inhibitor, but she is a non-diabetic with an elevated creatinine, peripheral vasculopathy and persistent hypertension, probably due to renovascular disease. Here an ACE inhibitor need not be denied if careful and frequent monitoring of renal function is carried out over the first month or so post discharge. As for hormone replacement therapy with estrogen, the alleged benefit on endothelial function has not so far been substantiated by carefully designed clinical trials.2 Besides, the patient is asking some hard questions regarding the history of breast cancer in the family and her previous history of phlebitis and pulmonary embolus. Finally, the history of gastrointestinal bleeding raises caution with respect to aspirin prophylaxis. The clinical trials unfortunately have not provided satisfactory guidelines on how to weigh the risk versus the benefit of postinfarction prophylaxis for patients with certain potential hazards, such as this case. Although this case does not easily qualify for entrance criteria to most of the large post-MI clinical trials, it is not a reason to deny her useful protective measures as indicated above.

908

As for rehabilitative measures, although it would be difficult for her to achieve a high level of exercise training using leg exercise, there is evidence that a well supervised upper arm exercise program may be useful.3 If she is intolerant of an ACE inhibitor perhaps a combination of hydralazine and nitroglycerin4 might serve to improve circulatory function, control her hypertension and improve her heart rate. She should continue on aspirin as long as she also continues with an H2 blocker or, better still, reduce the dose of aspirin to 81 mg a day and add clopidogrel.5 As for her lipids, she probably would benefit from a fibrate rather than a statin, although there are no clinical trials that have compared the two, head to head, for this particular type of patient.

References 1.Beta Blocker Heart Attack Study Group. A randomised trial of propanolol in patients with acute myocardial infarction. JAMA 1982;247:1707–14. 2.Hulley S, Grady D, Bush T et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in post menopausal women. HERS Research group. JAMA 1998;280:605–13. 3.Ghilarducci LE, Holly RG, Amsterdam EA. Effects of high resistance training in coronary artery disease. Am J Cardiol 1989;64:866–70. 4.Cohn JN, Archibald DG, Ziesche S et al. Effect of vasodilator therapy on mortality in chronic congestive heart failure: results of a Veterans Administration Cooperative Study. N Engl J Med 1986;314:1547–52. 5.The CURE Investigators. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST segment elevation. N Engl J Med 2001;345:494–502.

68

Metabolic risk and secondary prevention of coronary disease Jacques Genest Jr

Case 1

The patient is a 55 year old man post acute myocardial infarction. His initial treatment consisted of thrombolytic therapy within 4 hours of presentation. He was in Killip class I and 4 days after his myocardial infarction a submaximal stress test did not reveal ischemia at 9·2 METS. He was initially treated with aspirin 325 mg od, metoprolol 50 mg po bid and simvastatin 40 mg po qhs. Six months after his infarction his metabolic profile revealed a total cholesterol of 4·20, plasma triglyceride of 1·2 mmol/l, HDL cholesterol of 1·2 mmol/l and an LDL cholesterol of 2·45 mmol/l. The patient had stopped smoking and his body mass index was 26·7·

Question Is his metabolic risk profile adequate for the treatment he is on?

Comment The results of the Heart Protection Study1 have dramatically altered the way in which clinicians should be identifying and treating lipoprotein disorders in high-risk individuals. The currently accepted strategy supported by the National Cholesterol Education Program (Adult Treatment Panel III) in the USA2 and the Canadian Guidelines for the Diagnosis and Treatment of Dyslipoproteinemias,3 as well as the European Task Force for the Prevention of Cardiovascular Disease, is an evaluation of global cardiovascular risk based on the patient’s gender, age, total (or LDL) cholesterol, HDL cholesterol, cigarette smoking, blood pressure and diabetes. In patients at high risk for cardiovascular disease or those

Case 2

with established atherosclerotic cardiovascular disease, some notable differences between European, Canadian and US guidelines have emerged. From a strategy of reaching target values (for example, LDL cholesterol less than 2·5 mmol/l in Canada) the Heart Protection Study shows that baseline levels of LDL or total cholesterol do not markedly influence the positive impact of treatment. And so, for our patient the widespread use of the statin class of drugs post acute coronary event has now become standard therapy. Although some lipid specialists may disagree with initiating a statin drug without knowledge of prior basal measurements of serum lipids, this class of drugs has proved safe and highly efficacious in improving outcome. Obtaining a lipoprotein profile within 24 hours of hospital admission remains sound clinical practice, but the initiation of treatment as soon as possible after diagnosis, and verification 6 weeks to 6 months later, is useful. In this case, the LDL cholesterol is below the recommended target and it remains to be determined whether allowing the LDL to exceed 2·5 mmol/l should invite further intervention.

A 55 year old woman, recently postmenopausal, presented to the emergency department with retrosternal squeezing chest pain of 30 minutes’ duration. The ECG showed anterolateral ST segment depression and the troponins were slightly elevated. Urgent cardiac catheterization revealed an 85% proximal circumflex stenosis, which was successfully dilated and stented. A lipid profile done in hospital showed her total cholesterol to be 4·1 mmol/l, plasma triglyceride 2·6 mmol/l, HDL cholesterol 0·62 mmol/l and an LDL cholesterol 2·3 mmol/l. Her body mass index was 27·2, her blood pressure was 150/86 and her fasting glucose was 5·2 mmol/l.

909


Question What would be optimum metabolic management for this patient?

Comment This case represents a therapeutic challenge. The Veterans Affairs HDL intervention trial (VA-HIT) showed that patients with a low HDL cholesterol benefited from the drug gemfibrozil 600 mg po bid.4 The recently released HATS trial showed that low-dose simvastatin with niacin greatly improved both HDL and the total cholesterol to HDL ratio, and was associated with a decrease in the progression of atherosclerosis.5 The Health Protection Study (HPS), however, showed that in patients whose total cholesterol

Case 3

A 60 year old man with hyperglycemia (fasting blood sugar 7·6 mmol/l) treated with diet also has moderate hypertension (150/92) and an abnormal lipid profile. His total cholesterol is 6·4 mmol/l, plasma triglyceride 3·2 mmol/l, HDL cholesterol 0·8 mmol/l and LDL cholesterol 4·15 mmol/l. He has a positive exercise tolerance test at 8·2 METS using the Bruce protocol and his body mass index is 29·8 with a waist circumference of 104 cm. He has a sedentary lifestyle and his wife reports that he snores loudly at night.

Question What is the approach to this patient with multiple metabolic risk factors?

Comment Since the last edition of Evidence Based Cardiology a major rethink has taken place in preventive cardiology. Landmark clinical trials have shown unequivocally that patients with atherosclerotic disease who have multiple metabolic risk factors, such as diabetes and hyperlipidemia, have a risk of future cardiovascular events not dissimilar to those who have already suffered an acute coronary event. Therefore, the previously narrow definition of secondary prevention has been widened to include individuals with multiple risk factors where the rate of hard cardiovascular events exceeds 20% in 10 years. This “high-risk strategy” centers on the identification of global cardiovascular risk, as seen in this patient. This patient has the typical manifestations of the metabolic syndrome, consisting of abdominal obesity, increase in abdominal girth, low HDL cholesterol, elevated plasma triglycerides, hyperglycemia (insulin resistance) and hypertension. These patients tend to have small dense LDL particles and elevated apolipoprotein B levels. They may 910

was higher than 3·5 mmol/l simvastatin 40 mg/day conferred a dramatic benefit.1 The optimal treatment of highrisk patients with very low HDL cholesterol and mild to moderate triglycerides is still an issue of intense debate. Our patient has an HDL that is markedly below the fifth percentile for age- and gender-matched subjects. The fibric acid group of medications has been shown in clinical trials to decrease cardiovascular mortality4 and are an appropriate option. The combination of a low-dose statin with niacin is also an appropriate choice here, and yet the use of a statin as seen in the HPS1 and 4S6 studies appears to confer benefit. Each of these therapies has been shown to be superior to placebo alone. The use of postmenopausal estrogen (partly because of its effect on increasing HDL cholesterol levels) has been questioned in view of the negative findings in the HERS trial.7

also have elevated procoagulant factors, as well as an altered tPA : PAI-1 ratio and endothelial dysfunction. Although these patients represent some of the highest cardiovascular risk, current algorithms, including the Framingham Heart Study, may not provide a true estimation of risk because obesity, plasma triglycerides and other factors are not included in the cardiovascular risk calculations. Nevertheless, the treatment of these patients is multifactorial. Clearly, blood pressure control with appropriate medications is warranted. Although fibric acid derivatives will lower plasma triglyceride and raise the HDL cholesterol somewhat,4 the preferred medication in this case would be a statin at a moderately high dose.1 Although statins have a modest effect on HDL per se (unless, as in this case, there is hypertrigyceridemia) they significantly improve the total cholesterol to HDL ratio. In addition they do lower plasma triglyceride levels. Overall, the patient is urged to engage in a concerted effort to reduce all the metabolic risk factors. This can be helped by making the patient aware of the clustering of his risk profile and the potential effect on his cardiovascular health. The patient needs to lose weight, change diet and exercise regularly. The loss of even a few kilograms of intra-abdominal fat will significantly decrease fasting blood glucose levels and insulin resistance. It will also reduce his triglyceride and help increase his HDL cholesterol levels, as well as lower his

Metabolic risk and secondary prevention of coronary disease

blood pressure. This constitutes a great therapeutic challenge which can best be achieved by a multidisciplinary team approach, including a dietician and exercise rehabilitation specialists. References 1.Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering in 20536 high-risk individuals: a randomized placebo-controlled trial. Lancet 2002;340:7–12. 2.Executive summary of the third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation and Treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001; 285:2486–97. 3.Fodor JG, Frohlich JJ, Genest JJ Jr et al. Recommendations for the management and treatment of dyslipidemia. Report of

the Working Group on Hypercholesterolemia and Other Dyslipidemias. Can Med Assoc J 2000;162:1441–7. 4.Rubins HB, Robins SJ, Collins D et al. Gemfibrozil for the secondary prevention of coronary heart disease in men with low levels of high density lipoprotein cholesterol. N Engl J Med 1999;341:410–18. 5.Brown BG, Zhao XQ, Chait A et al. Simvistatin and niacin, antioxidant vitamins or the combination for prevention of coronary disease. N Engl J Med 2001;345:1583–92. 6.Ballantyne CM, Olsson AJ, Cook TJ et al. Influence of low highdensity lipoprotein cholesterol and elevated triglyceride on coronary heart disease events and response to simvistatin therapy in 4S. Circulation 2001;104:3046–51. 7.Barrett-Connor E. Looking for the pony in the HERS data. Heart and estrogen/progestin replacement study. Circulation 2002;105:902–3.

911

69

Peripheral vascular disease with suspect coronary artery disease Peter C Spittell

Case 1

Over the past several years a 74 year old man has noticed progressive left leg discomfort with walking. Standing still provides complete relief. His walking distance has gradually decreased to less than one block in the past year. He is a non-smoker, non-diabetic, and has no prior cardiac history. His medications include captopril and dyazide for hypertension. On examination, his blood pressure is 140/80 mmHg (both arms) and his resting pulse is 65 bpm and regular. There is a right carotid bruit and reduced femoral pulses with bilateral bruits. Pulses are non-palpable below the femoral level on the left. On the right, the popliteal and posterior tibial pulses are mildly reduced and the dorsalis pedis pulse is absent. Elevation pallor is grade III (pallor is less than 30 seconds) on the left and grade 0 (no pallor in 60 seconds) on the right. The remainder of the physical examination is normal. A resting ECG reveals normal sinus rhythm with a non-specific T wave abnormality. The chest radiograph is normal. Routine hematology and chemistry values are normal. The resting ankle : brachial systolic pressure indexes (ABI, normal 0·9) are 1·0 on the right, 0·6 on the left. After walking 124 yards (113 m) on a treadmill (10% incline, 1·5 mph) and developing left hip, thigh and calf claudication, his postexercise ABIs are 0·3 on the right and 0·2 on the left, consistent with moderate (right) and moderately severe (left) peripheral arterial occlusive disease, respectively. Carotid ultrasound demonstrates a large amount of atheromatous plaque in the right carotid bulb associated with a 70–99% stenosis.

Question What would be the most appropriate management at this point?

Comment Here, one seeks evidence to guide the management of an elderly patient with extensive peripheral vascular disease who may or may not have concomitant coronary artery disease. The finding of an asymptomatic high grade carotid stenosis (60%) warrants consideration of prophylactic carotid endarterectomy if the patient’s general health is good. A 60% diameter reducing stenosis (carotid ultrasound) is considered as an indication for surgical intervention in asymptomatic patients who are in a low-risk surgical category: using a ratio of 3·2 for the peak systolic velocity at the site of narrowing divided by that from the carotid artery, 912

a sensitivity of 92% and a specificity of 86% can be achieved. This gives a positive predictive value of 85%, a negative predictive value of 93% and an overall accuracy of 89%.1 The evidence shows that carotid endarterectomy in patients with an asymptomatic carotid stenosis 60% in severity significantly reduces the risk of ipsilateral stroke, perioperative stroke or death compared to medical therapy alone.2 In view of the patient’s age and widespread peripheral vascular disease it is prudent to assess his perioperative cardiac risk. Dobutamine stress echocardiography was performed and was negative for ischemia. This test is associated with a high negative predictive value.3 The patient underwent right carotid endarterectomy without complications. Postoperatively, following a discussion with the patient regarding the natural history of intermittent claudication and indications for restoration of pulsatile flow,4 he elected a conservative treatment program which included aspirin 325 mg/day, a walking program for intermittent claudication, and foot care and protection.

Peripheral vascular disease with suspect coronary artery disease

References 1·Edwards JM et al. Duplex ultrasound criteria for determining 50% and 60% internal carotid artery stenosis: implications for screening examinations. Noninvasive vascular laboratory and vascular imaging. In: Young JR, Olin JW, Bartholomew JR, eds. Peripheral vascular diseases, 2nd edn. St Louis: Mosby, 1996. 2·Executive Committee for the Asymptomatic Carotid Atherosclerosis (ACAS) Study. Endarterectomy for asymptomatic carotid artery stenosis. JAMA 1995;273:1421–8.

Case 2

3.Eichelberger JP, Schwarz KQ, Black ER, Green R, Ouriel K. Predictive value of dobutamine echocardiography before noncardiac vascular surgery. Am J Cardiol 1993;72:602–7. 4.McDaniel MD, Cronenwett JL. Basic data related to the natural history of intermittent claudication. Ann Vasc Surg 1989; 3:273–7.

A 52 year old man presents with progressive exertional right calf discomfort over the past year. Standing still provides complete relief but his walking distance has gradually shortened to one block. He notes similar but less severe discomfort in the left calf. His claudication is severely limiting his lifestyle. He is taking amlodipine for systemic hypertension, has smoked tobacco for the past 34 years and has a history of hyperlipidemia treated by diet. Blood glucose is normal. He denies a history of angina pectoris or diabetes mellitus. On examination his blood pressure is 140/90 mmHg (both arms), resting pulse is 85 bpm and regular. A grade 1/4 systolic murmur at the cardiac apex and a right carotid bruit are present. A small abdominal aortic aneurysm is also palpable. The popliteal, posterior tibial and dorsalis pedis pulses are absent on the right. On the left, the popliteal and posterior tibial pulses are moderately reduced. The dorsalis pedis pulse is absent. Elevation pallor is grade III (pallor in less than 30 seconds) and grade I (pallor in less than 60 seconds) on the right and left, respectively. A resting ECG shows normal sinus rhythm without other abnormalities. Routine hematology and chemistry values are normal. Resting ankle:brachial systolic pressure index (ABI) is 0·5 on the right and 0·6 on the left (ABI normal 0·9). After walking 282 yards (258 m) on a treadmill (10% incline, 2 mph) and developing right calf claudication, his postexercise ABIs are 0·2 on the right, 0·7 on the left, consistent with severe peripheral arterial occlusive disease on the right and mild disease on the left. Carotid ultrasound reveals a right external carotid artery stenosis with mild atherosclerotic disease in the right carotid bulb, but is otherwise normal. Abdominal ultrasound confirms a small abdominal aortic aneurysm (2·5 cm). After a discussion of peripheral arterial occlusive disease with the patient, including its natural history, prognosis, treatment and goals of treatment, he elects to pursue restoration of pulsatile flow.

Question How would you proceed at this point?

Comment This case is somewhat more complex insofar as there is more suspicion of concomitant coronary artery disease. Therefore, with the decision to pursue vascular surgery, an assessment of the patient’s perioperative cardiac risk is warranted.1 Although he has no history of angina pectoris he has a number of cardiovascular risk factors (male gender,

tobacco, hypertension, hyperlipidemia). Furthermore, his intermittent claudication limits his activity and may prevent him from experiencing exertional angina. To further assess his perioperative cardiac risk, dobutamine stress echocardiography was performed and demonstrated normal left ventricular size and function, but was positive for ischemia (new regional wall motion abnormalities in the mid and apical anterior wall). A positive dobutamine stress echocardiogram has a positive predictive value of 35% for a perioperative cardiac event. When ischemia occurs at a significantly lower heart rate during dobutamine stress (70% of the age-corrected maximal heart rate) the positive predictive value increases to 53%.2 913


Medical therapy was elected (suspect single vessel disease with normal left ventricular function), and it was felt safe to proceed with peripheral revascularization. Peripheral angiography demonstrated a high grade stenosis involving a short segment of the distal right superficial femoral artery. Percutaneous transluminal angioplasty of the stenosis was performed. Although no well designed large scale clinical trial has yet been done to determine the efficacy of percutaneous angioplasty for this type of patient, percutaneous angioplasty was none the less felt to be indicated because the prognosis for limb loss in patients with intermittent claudication is related mostly to the severity of disease, as assessed by ankle pressure measurements, at the time of study entry.3 After the procedure, a normal dorsalis pedis pulse was restored on the right. The intermittent claudication completely resolved following the procedure. He was placed on aspirin (325 mg/day), advised to stop smoking and instructed in foot care and protection. Amlodipine was continued for

914

treatment of his hypertension, and the importance of adequate control of hypertension was discussed. Ultrasound of the abdominal aorta in 1 year was recommended to re-evaluate his abdominal aortic aneurysm. Additional instruction on dietary therapy for hyperlipidemia was given and a follow-up lipid profile was arranged in 3 months.

References 1.Eagle KA, Brundage BH, Chaitman BR et al. Guidelines for perioperative cardiovascular evaluation for noncardiac surgery. ACC/AHA Task Force report. Circulation 1996;93:1278–317. 2.Poldermans D, Arnese M, Fioretti PM et al. Improved cardiac risk stratification in major vascular surgery with dobutamine stress echocardiography. J Am Coll Cardiol 1995;26:648–53. 3.McDaniel MD, Cronenwelt JL. Basic data related to the natural history of intermittent claudication. Ann Vasc Surg 1989; 3:271–7.

70 Case 1

Heart failure Michael M Givertz

A 58 year old man with non-insulin dependent diabetes mellitus and coronary artery disease presents to your office with progressive dyspnea on exertion and lower extremity edema. He is status post anterior myocardial infarction 15 years ago. Following cardiac rehabilitation, he returned to work full-time as an electrical engineer and remained asymptomatic without angina or heart failure until 1 year ago, when he started to “slow down”. Over the last several months he has noted increasing shortness of breath with usual daily activities, and more recently the onset of bilateral lower extremity edema. He can no longer play with his grandchildren because of fatigue. He denies chest pain, palpitations or lightheadedness. He sleeps comfortably on one pillow and has had no recent change in his weight or appetite. His current medications include captopril 12·5 mg tid, furosemide 20 mg qd, glyburide 5 mg qd and aspirin. On physical examination he is a well-nourished older-appearing man who appears comfortable lying supine. Blood pressure is 130/70 mmHg and heart rate is 100 beats per minute and regular. Jugular venous pressure is 8 cmH2O. Lungs are clear bilaterally. Cardiovascular examination reveals a displaced PMI, grade 2 over 6 holosystolic murmur at the apex, and soft S3 gallop. The liver edge is palpable 2 cm below the right costal margin, and there is 1 pitting edema to the mid-calves bilaterally. His feet are warm with intact distal pulses. Laboratory tests are notable for a serum sodium of 136 mmol/l, creatinine of 1·8 mg/dl, and hemoglobin A1c of 6·2%. His electrocardiogram reveals normal sinus rhythm with a non-specific intraventricular conduction delay and old anterior myocardial infarction. A transthoracic echocardiogram demonstrates moderate left ventricular dilation with anteroapical akinesis, left ventricular ejection fraction of 25%, mild right ventricular dysfunction, and 2 mitral regurgitation.

Question How would you manage this patient?

Comment To summarize, this is a middle-aged man with diabetes and an ischemic cardiomyopathy who presents with New York Heart Association functional class III heart failure. There is both clinical and laboratory evidence of left ventricular systolic dysfunction and volume overload, without clinical evidence of decreased cardiac output. In addition, his echocardiogram shows findings consistent with left ventricular remodeling without intracavitary thrombus formation. There is no clear precipitant to his recent clinical decompensation. He has been compliant with his medications and fluid and salt restriction, and does not drink alcohol. His blood sugars have been well controlled and he has had no recent infection. A complete blood count and thyroid

function tests should be checked to rule out anemia and hyperthyroidism, respectively. Although we have not identified a precipitating factor in this case, it is reasonable to consider treatment of the underlying cause of heart failure.1 The patient is status post myocardial infarction many years ago, and may have developed recurrent ischemia, either silent or with dyspnea as an anginal equivalent. An exercise imaging study should be performed to rule out reversible ischemia. If this is negative, an assessment of myocardial viability with low-dose dobutamine stress echocardiography, resting thallium scintigraphy or positron emission tomography should be considered.2 Several studies have shown that patients with ischemic cardiomyopathy and viable myocardium have significant improvement in left ventricular function following surgical revascularization.3,4 However, in the absence of angina or inducible ischemia, the superiority of surgery over medical therapy in prolonging survival in patients with ischemic cardiomyopathy remains unproven. A National Institutes of Health-sponsored trial (STICH) is currently under way to test this hypothesis. 915


The overall goals in the management of heart failure are to eliminate symptoms, improve quality of life and prolong survival. Non-pharmacologic management of heart failure should be reviewed with the patient and his family. The importance of salt and fluid restriction and daily monitoring of home weights should be reinforced. Once euvolemia has been achieved (see below), a submaximal exercise program (for example, walking, stationary bicycle) should be encouraged. Recent studies suggest that exercise training may result in improvement in symptoms and functional capacity, improved blood flow and skeletal muscle metabolism, and reduced hospitalizations for heart failure.5 The effect of exercise training on survival in heart failure remains unknown. Pharmacologic therapy should be optimized according to recent consensus guidelines.6 The patient is currently taking captopril, an angiotensin converting enzyme (ACE) inhibitor, at a relatively low dose. Several large prospective randomized controlled trials have demonstrated the beneficial effects of ACE inhibitors on exercise tolerance, salt and water balance, symptoms, neurohormonal activation, quality of life and survival in patients with chronic heart failure.7 Furthermore, in this patient with mild diabetic nephropathy, ACE inhibitor therapy may slow the progression of renal dysfunction; also, as demonstrated in the SAVE study, captopril reduces the risk of recurrent myocardial infarction and stroke in patients with post-MI left ventricular dysfunction.8 The optimal dosing of ACE inhibitors remains controversial. One prospective study (ATLAS) demonstrated the superiority of high-dose versus low-dose ACE inhibitor therapy in patients with chronic heart failure without increased toxicity.9 Current guidelines recommend increasing the dose of captopril to 50 mg tid as blood pressure and renal function tolerate. For the treatment of systemic and pulmonary venous congestion, more aggressive diuresis is warranted. The daily dose of furosemide should be increased until the required response is achieved (for example, the absence of jugular venous distention, hepatomegaly and edema). If this strategy is not effective, combination therapy with a thiazide diuretic should be tried. Adequate diuresis with careful attention to weight and renal function will generally result in improved symptoms and may slow the progression of chamber dilation by reducing ventricular filling pressures. However, diuretic therapy may also cause renal dysfunction, electrolyte depletion and neurohormonal activation. It should be emphasized that there have been no randomized controlled trials demonstrating the long-term efficacy and safety of diuretic therapy in patients with heart failure.

Follow up 1 The patient undergoes exercise echocardiography, which is negative for ischemia, and resting thallium scintigraphy demonstrates no significant viability of the anterior and 916

apical walls. He diureses 8 lb on an increased dose of oral furosemide, and captopril is titrated to 37·5 mg tid. Jugular venous distention and lower extremity edema resolve, but blood pressure, heart rate and renal function are unchanged. Despite adjustment of vasodilator and diuretic therapy, there is no change in his exertional dyspnea. Question He remains moderately symptomatic despite treatment with an ACE inhibitor and diuretic. What is the next step?

The next step is to initiate therapy with a adrenergic antagonist. Traditionally, blockers were contraindicated in the treatment of heart failure because of concern about negative inotropic effects leading to clinical deterioration. In the late 1970s and early 1980s, small uncontrolled trials suggested a beneficial effect of blockers in patients with dilated cardiomyopathy. Subsequent randomized controlled trials have demonstrated that blockers improve symptoms and cardiac function, and reduce morbidity and morality in patients with chronic heart failure due to left ventricular systolic dysfunction.10 Consensus guidelines recommend the use of blockers, in addition to ACE inhibitors and diuretics, in the management of patients with mild to moderate heart failure,6 and more recent data suggest their safety and efficacy in patients with severe heart failure.11 Because euvolemia has been achieved, blocker therapy can be initiated safely at a low dose and titrated gradually at regular intervals (for example, every 1–2 weeks). Renal insufficiency should not prevent the initiation of blocker therapy, as the patient’s renal function has remained stable on ACE inhibitor therapy. However, renal function should be followed during blocker titration.

Follow up 2 Carvedilol, a non-selective blocker with 1 blocking properties, is initiated at a dose of 3·125 mg bid. One week later the patient returns to clinic with complaints of increased dyspnea on exertion and recurrent lower extremity edema. His clinical evaluation is consistent with worsening heart failure, a known adverse effect of blocker therapy, and this responds to a doubling of the furosemide dose for 2 days. Over the next 2 months carvedilol is titrated to a target dose of 25 mg bid. During this period, symptoms of worsening heart failure occur on one additional occasion. Close monitoring of symptoms and weight, and adjustments in diuretic dosing, enable the patient to achieve a target dose of blocker therapy. After 6 months, a follow-up echocardiogram reveals a decrease in the left ventricular end-diastolic dimension and an increase in the left ventricular ejection fraction to 35%. This time-dependent reverse remodeling of the left ventricle has been demonstrated with several different blockers, including metoprolol12 and carvedilol.13

Heart failure

Question After remaining clinically stable for 1 year, he develops worsening heart failure. What is the next step in management?

For persistent or recurrent heart failure symptoms despite treatment with an ACE inhibitor, blocker and diuretic, several adjunctive therapies may be considered. Digoxin is an oral positive inotropic agent with antiadrenergic effects that has been shown to be safe and effective in patients with symptomatic heart failure. Although the DIG trial showed no difference in survival in heart failure patients treated with digoxin versus placebo, there were fewer deaths attributable to progressive heart failure in the digoxin-treated

Case 2

group.14 Another class of medications that may provide symptomatic benefit in heart failure are the organic nitrates.15 These may be used to reduce both systemic and pulmonary venous congestion. As with diuretics, it should be remembered that nitrates alone have not been shown to reduce morbidity or mortality in patients with chronic heart failure. If, despite these therapies, the patient develops severe heart failure, other medical and surgical options may be considered. The aldosterone antagonist spironolactone has been shown to prolong survival in patients with severe heart failure,16 and more recently biventricular pacing has been approved for patients with symptomatic LV dysfunction and intraventricular conduction delay.17 For refractory heart failure, mechanical cardiac assist and cardiac transplantation may be considered in selected cases.18

A 74 year old woman is referred to you by her primary care physician for evaluation of “newonset heart failure”. Her cardiac risk factors are positive for hypertension, diabetes and hypercholesterolemia. She was told that she had a myocardial infarction many years ago, but details are not available for review. Four years ago, an echocardiogram obtained as part of a preoperative evaluation for laparoscopic cholecystectomy revealed a left ventricular ejection fraction of 45% and moderate mitral regurgitation. She was treated briefly with an angiotensin converting enzyme inhibitor, but this was changed to an angiotensin receptor antagonist because of cough. She remained stable until 2 months ago, when she presented with progressive dyspnea on exertion, associated with dull chest pain and pedal edema. The addition of digoxin and furosemide resulted in a 5 lb weight loss and improvement in symptoms. Her past medical history is significant for osteoarthritis, peptic ulcer disease and hypothyroidism. Her medications include losartan 50 mg qd, furosemide 40 mg qd, digoxin 0·25 mg qd, pravastatin 40 mg qhs, levothyroxine 100 micrograms qd, and ibuprofen 400 mg tid. She does not smoke cigarettes or drink alcohol. On physical examination she appears anxious but in no respiratory distress. Blood pressure is 140/80 mmHg, heart rate is 58 beats per minute and regular, and weight is 148 pounds. Jugular venous distention and hepatojugular reflux are absent. Lungs are clear to auscultation bilaterally. Cardiovascular examination reveals a non-palpable PMI, S3 and S4 gallops, and a soft systolic ejection murmur at the apex without radiation. Her abdomen is non-tender, without hepatosplenomegaly. Her extremities are warm and without edema. Laboratory tests reveal a serum sodium of 141 mmol/l, potassium 4·4 mmol/l, BUN 25 mg/dl, creatinine 1·4 mg/dl, digoxin level 1·6 ng/ml, TSH 2·1 mU/l, and hematocrit 38%. Chest radiography shows cardiomegaly without congestion. Her 12-lead electrocardiogram reveals sinus bradycardia and left bundle branch block. An echocardiogram demonstrates mild left ventricular enlargement, left ventricular ejection fraction of 20% with inferior and apical akinesis, normal right ventricular size and function, and mild thickening of the mitral valve leaflets with 2–3 central mitral regurgitation. Right heart catheterization reveals a right atrial pressure of 9 mmHg, pulmonary artery systolic and diastolic pressures of 58 and 24 mmHg, respectively, mean pulmonary capillary wedge pressure of 26 mmHg, and cardiac output of 4·8 l/min. Angiography reveals no significant coronary artery disease, and ventriculography confirms severe left ventricular dysfunction with 3 mitral regurgitation.

Question What is your assessment of this patient?

In summary, this is an elderly woman with hypertension, diabetes and a non-ischemic dilated cardiomyopathy who presents for further management. The cause of her left ventricular systolic dysfunction remains incompletely defined. 917


Although she has a high likelihood of having ischemic heart disease based on her cardiac risk factors, history of “myocardial infarction”, left bundle branch block on electrocardiogram and regional wall motion abnormalities on echocardiogram, coronary angiography demonstrates no significant coronary artery disease. This case highlights the lack of specificity of the history, ECG and echocardiogram for diagnosing ischemic heart disease in the presence of severe left ventricular systolic dysfunction.19 Not infrequently, patients are told by their physician that they may have had a “heart attack” based on the presence of a left bundle branch block. In the evaluation of heart failure, cardiac catheterization is indicated not only to define coronary anatomy, but also to assess hemodynamics and, in this case, to determine the severity of mitral regurgitation. If ischemic heart disease does not explain this patient’s LV dysfunction, what are the other possible causes? There is no history of myocarditis, anemia or significant alcohol use. She has a history of hypothyroidism, but is maintained on thyroid hormone replacement therapy and has no clinical or biochemical evidence of active thyroid disease. Chronic mitral regurgitation (MR) may result in heart failure and left ventricular remodeling, and it is often difficult to distinguish primary mitral valvular disease causing advanced heart failure from MR secondary to left ventricular dilation (for example, following myocardial infarction). In this case, moderate LV dysfunction was noted 4 years ago in association with moderate MR, suggesting that myocardial disease is the primary process. Other echocardiographic findings that make primary MR less likely are the lack of structural abnormalities of the mitral valve or subvalvular apparatus, normal left atrial size, and the absence of pulmonary hypertension or right ventricular dysfunction. Alternatively, the presence of a left bundle branch block and known LV dysfunction make the diagnosis of idiopathic dilated cardiomyopathy likely,20 although recent studies have shown that up to 20% of these cases may be familial.21 In general, idiopathic dilated cardiomyopathy is a diagnosis of exclusion.22 A “bedside” clinical assessment should be performed on all patients with chronic heart failure to assess the presence or absence of congestion and hypoperfusion.23 In addition, functional status should be determined both for prognostic reasons and to assess the response to therapy. This patient presents with exertional dyspnea and chest pain; pedal edema has resolved with diuresis. Although her examination does not suggest the presence of pulmonary or systemic venous congestion, right heart catheterization is notable for elevated right and left heart filling pressures and moderate pulmonary hypertension with adequate systemic perfusion. Importantly, this case demonstrates the limited reliability of physical signs24 and chest radiography25 for estimating hemodynamics in patients with chronic heart failure. Some findings, such as an S3 gallop or cardiomegaly, are highly sensitive but lack the specificity to be of diagnostic value; 918

other findings, such as pulmonary rales and jugular venous distention, are highly specific but insensitive. The clinical utility of other non-invasive tools to assess cardiac filling pressures, such as myocardial tissue Doppler imaging or plasma B-type natriuretic peptide (BNP) levels, remains unproven. To better define her functional capacity, exercise treadmill testing with continuous gas exchange analysis (also termed cardiopulmonary exercise testing) should be considered. Numerous studies have demonstrated the independent prognostic value of peak oxygen consumption in patients with heart failure,26 and cardiopulmonary exercise testing may be used to assess the response to therapy. If the patient is unable to undergo maximal exercise testing, a submaximal test such as the 6-minute walk test may provide an estimate of peak functional capacity, and is better tolerated.27 Quality of life assessment, although commonly used in clinical research protocols and for quality improvement analyses, is not a standard of care in clinical practice. Question How would you manage this patient?

Despite the presence of significant mitral regurgitation, there is no proven role for mitral valve repair or replacement in the treatment of patients with dilated cardiomyopathy, although there are surgical proponents of this approach.28 Appropriate initial medical therapy would be to increase diuresis, which can be accomplished with higher doses of furosemide. Without the assistance of a pulmonary artery catheter or a reliable non-invasive tool to measure cardiac filling pressures,29 clinical end points such as decreased dyspnea with stable renal function should be targeted. As previously noted, the long-term efficacy and safety of diuretics in the treatment of heart failure remain unknown. Ibuprofen, a non-steroidal anti-inflammatory agent that may contribute to fluid retention in heart failure,30 should be discontinued. Neurohormonal antagonist therapy should be maximized. The patient is taking a moderate dose of losartan, an angiotensin receptor blocker (ARB), in place of an ACE inhibitor, which was discontinued because of cough. Based on available clinical trials data31 and according to consensus guidelines,6 ACE inhibitors rather than ARBs continue to be the agents of choice for blockade of the renin–angiotensin system (RAS) in heart failure. Cough, although not uncommon with ACE inhibitors, is not an absolute contraindication to ACE inhibitor therapy and requires discontinuation of medication in less than 5% of patients.31 In addition, cough may be due to the heart failure itself, rather than the ACE inhibitor prescribed to treat heart failure. In the present case it would be reasonable to rechallenge the patient with an ACE inhibitor and titrate to a target dose (for example, lisinopril 40 mg qd).9 Although there has been increasing interest in the combined use of an ACE inhibitor and

Heart failure

ARB for maximal RAS blockade, preliminary data from the Val-HeFT study raises concerns about this approach in patients who are also taking a blocker.32 In this case, blocker therapy should be initiated and titrated to target doses (for example, carvedilol 25 mg bid, Toprol-XL 200 mg qd) to achieve important survival benefits.10 Sinus bradycardia and left bundle branch block are not contraindications to blocker therapy, but require close monitoring. If symptomatic bradycardia occurs, implantation of a permanent pacemaker should be considered rather than discontinuation of the blocker. Digoxin may be contributing to bradycardia in this case. Although the serum digoxin level of 1·6 ng/ml is within the normal range, the level may not correlate with symptoms or electrocardiographic evidence of digoxin toxicity.33 During blocker titration it would be reasonable to decrease the digoxin dose to 0·125 mg qd and maintain serum digoxin levels in the low normal (0·5–1·0 ng/ml) range. However, the role of monitoring digoxin levels in patients with chronic heart failure remains controversial.

Follow up The patient responds well to an increase in her furosemide dose with improved exertional tolerance and resolution of chest pain. Arrangements are made to initiate blocker therapy on an outpatient basis. On the evening prior to her scheduled visit, she has a witnessed syncopal event without prodromal symptoms while standing in the bathroom. An ambulance is called and the emergency medical technicians find her lying on the floor, awake but mildly confused, with spontaneous respirations and pulse. Her blood pressure is 100/70 mmHg. A rhythm strip obtained on route to the hospital reveals sinus tachycardia at a rate of 108 beats per minute with occasional premature ventricular complexes and a 3-beat run of non-sustained ventricular tachycardia. In the emergency room she is alert and oriented, without focal neurologic deficits, postural hypotension or evidence of gastrointestinal bleeding. She has no sustained arrhythmias on telemetry. Question What would you do next?

The evaluation and management of syncope has been well described34 and is beyond the scope of this chapter. Syncope in patients with symptomatic LV dysfunction is particularly concerning, as it may be caused by ventricular arrhythmias and is associated with an increased risk of sudden death. Among survivors of sustained ventricular tachycardia associated with syncope and LV failure, the implantable cardioverter defibrillator (ICD) has been shown to improve survival compared with antiarrhythmic drug therapy.35

Furthermore, in patients with coronary artery disease, reduced ejection fraction and asymptomatic non-sustained VT, the ability to induce ventricular tachycardia at the time of electrophysiologic study is an indication for ICD placement.36 Finally, Moss et al recently demonstrated that in patients with a prior myocardial infarction and advanced LV dysfunction, prophylactic implantation of a defibrillator improves survival.37 However, the role of ICD therapy in patients with syncope and heart failure due to non-ischemic cardiomyopathy is less clear. Some would advocate placement of an ICD in this patient based on observational data showing improved outcomes compared with medical therapy alone.38 Alternatively, evidenced-based cardiologists will await the results of primary prevention trials of sudden cardiac death in the broad population of patients with LV dysfunction.39 References 1.Givertz MM, Colucci WS, Braunwald E. Clinical aspects of heart failure; high-output heart failure; pulmonary edema. In: Braunwald E, Zipes DP, Libby P, eds. Heart Disease: A Textbook of Cardiovascular Medicine. Philadelphia: WB Saunders, 2001. 2.Bonow RO. The hibernating myocardium: implications for management of congestive heart failure. Am J Cardiol 1995; 75:17A–25A. 3.Tamaki N, Kawamoto M, Tadamura E et al. Prediction of reversible ischemia after revascularisation. Perfusion and metabolic studies with positron emission tomography. Circulation 1995;91:1697–705. 4.Bax JJ, Comel JH, Visser FC et al. Prediction of recovery of myocardial dysfunction after revascularization. Comparison of fluorine-18 fluorodeoxyglucose/thallium-201, SPECT, thallium-201 stress-reinjection SPECT and dobutamine echocardiography. J Am Coll Cardiol 1996;28:558–64. 5.Coats AJ. Exercise training for heart failure: coming of age. Circulation 1999;99:1138–40. 6.Hunt SA, Baker DW, Chin MH et al. ACC/AHA guidelines for the evaluation and management of chronic heart failure in the adult: executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (committee to revise the 1995 Guidelines for the Evaluation and Management of Heart failure). J Am Coll Cardiol 2001;38:2101–13. 7.Garg R, Yusuf S. Overview of randomized trials of angiotensinconverting enzyme inhibitors on mortality and morbidity in patients with heart failure. Collaborative Group on ACE Inhibitor Trials. JAMA 1995;273:1450–6. 8.Loh E, Sutton MS, Wun CC et al. Ventricular dysfunction and the risk of stroke after myocardial infarction. N Engl J Med 1997;336:251–7. 9.Packer M, Poole-Wilson PA, Armstrong PW et al. Comparative effects of low and high doses of the angiotensin-converting enzyme inhibitor, lisinopril, on morbidity and mortality in chronic heart failure. Circulation 2000;100:2312–18. 10.Bristow MR. Beta-adrenergic receptor blockade in chronic heart failure. Circulation 2000;101:558–69.

919


11.Packer M, Coats AJ, Fowler MB et al. Effect of carvedilol on survival in severe chronic heart failure. N Engl J Med 2001; 344:1651–8. 12.Hall SA, Cigarroa CG, Marcoux L et al. Time course of improvement in left ventricular function, mass and geometry in patients with congestive heart failure treated with beta- adrenergic blockade. J Am Coll Cardiol 1995;25:1154–61. 13.Doughty RN, Whalley GA, Gamble G et al. Left ventricular remodeling with carvedilol in patients with congestive heart failure due to ischemic heart disease. Australia–New Zealand Heart Failure Research Collaborative Group. J Am Coll Cardiol 1997;29:1060–6. 14.The Digitalis Investigation Group. The effect of digoxin on mortality and morbidity in patients with heart failure. N Engl J Med 1997;336:525–33. 15.Elkayam U, Karaalp IS, Wani OR et al. The role of organic nitrates in the treatment of heart failure. Prog Cardiovasc Dis 1999;41:255–64. 16.Pitt B, Zannad F, Remme WJ et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized Aldactone Evaluation Study Investigators. N Engl J Med 1999;341:709–17. 17.Abraham WT, Fischer WG, Smith AL et al. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002;346: 1845–53. 18.Hunt SA, Frazier OH. Mechanical circulatory support and cardiac transplantation. Circulation 1998;97:2079–90. 19.Costanzo MR, Augustine S, Bourge R et al. Selection and treatment of candidates for heart transplantation. A statement for health professionals from the Committee on Heart Failure and Cardiac Transplantation of the Council on Clinical Cardiology, American Heart Association. Circulation 1995; 92:3593–612. 20.Dec GW, Fuster V. Idiopathic dilated cardiomyopathy. N Engl J Med 1994;331:1564–75. 21.Michels VV, Moll PP, Miller FA et al. The frequency of familial dilated cardiomyopathy in a series of patients with idiopathic dilated cardiomyopathy. N Engl J Med 1992;326:77–82. 22.Felker GM, Thompson RE, Hare JM et al. Underlying causes and long-term survival in patients with initially unexplained cardiomyopathy. N Engl J Med 2000;342:1077–84. 23.Givertz MM, Stevenson LW, Colucci WS. Hospital management of heart failure. In: Antman Em, ed. Cardiovascular Therapeutics: A Companion to Braunwald’s Heart Disease, 2nd edn. Philadelphia: WB Saunders, 2002. 24.Stevenson LW, Perloff JK. The limited reliability of physical signs for estimating hemodynamics in chronic heart failure. JAMA 1989;261:884–8. 25.Chakko S, Woska D, Martinez H et al. Clinical, radiographic, and hemodynamic correlations in chronic congestive heart failure: conflicting results may lead to inappropriate care. Am J Med 1991;90:353–9.

920

26.Mancini DM, Eisen H, Kussmaul W et al. Value of peak exercise oxygen consumption for optimal timing of cardiac transplantation in ambulatory patients with heart failure. Circulation 1991;83:778–86. 27.Lipkin DP, Scriven AJ, Crake T et al. Six minute walking test for assessing exercise capacity in chronic heart failure. BMJ 1986;292:653–5. 28.Bolling SF, Pagani FD, Deeb GM et al. Intermediate-term outcome of mitral reconstruction in cardiomyopathy. J Thorac Cardiovasc Surg 1998;115:381–6. 29.Sanders GP, Mendes LA, Colucci WS et al. Noninvasive methods for detecting elevated left-sided cardiac filling pressure. J Card Fail 2000;6:157–64. 30.Page J, Henry D. Consumption of NSAIDs and the development of congestive heart failure in elderly patients: an underrecognized public health problem. Arch Intern Med 2000; 160:777–84. 31.Pitt B, Poole-Wilson PA, Segal R et al. Effect of losartan compared with captopril on mortality in patients with symptomatic heart failure: randomised trial – the Losartan Heart Failure Survival Study ELITE II. Lancet 2000;355:1582–7. 32.Cohn JN, Tognoni G, Valsartan Heart Failure Trial Investigators. A randomized trial of the angiotensin-receptor blocker valsartan in chronic heart failure. N Engl J Med 2001;345:1667–75. 33.Hauptman PJ, Kelly RA. Digitalis. Circulation 1999;99: 1265–70. 34.Kapoor WN. Syncope. N Engl J Med 2000;343:1856–62. 35.The Antiarrhythmics versus Implantable Defibrillators (AVID) Investigators. A comparison of antiarrhythmic-drug therapy with implantable defibrillators in patients resuscitated from near-fatal ventricular arrhythmias. N Engl J Med 1997;337: 1576–83. 36.Moss AJ, Hall WJ, Cannom DS et al. Improved survival with an implanted defibrillator in patients with coronary disease at high risk for ventricular arrhythmia. Multicenter Automatic Defibrillator Implantation Trial Investigators. N Engl J Med 1996;335:1933–40. 37.Moss AJ, Zaveha W, Hall WJ et al. Prophylactic implantation of a defibrillator in patients with myocardiol infarction and reduced ejection fraction. N Engl J Med 2002;346:877–83. 38.Fonarow GC, Feliciano Z, Boyle NG et al. Improved survival in patients with nonischemic advanced heart failure and syncope treated with an implantable cardioverter-defibrillator. Am J Cardiol 2000;85:981–5. 39.Klein H, Auricchio A, Reek S et al. New primary prevention trials of sudden cardiac death in patients with left ventricular dysfunction: SCD-HEFT and MADIT-II. Am J Cardiol 1999; 83:91D–7D.

71

Atrial fibrillation Michael Klein

Case 1

A 74 year old woman who had been followed with annual examinations requested an interim office visit because of exertional shortness of breath and palpitations. Review of her medical records confirmed the absence of asthma, diabetes or cardiovascular disease and the presence of easy bruising when using aspirin on a regular basis. Recent blood pressures have been 150–160/70–80 mmHg and the possibility of drug treatment for systolic hypertension had been discussed, but she opted for a healthy lifestyle, including recreational exercise three to four times a week. On this occasion her BP was 156/72. The lungs were clear and there was no ankle edema. Cardiac examination showed variable intensity S1, A2 P2, No S3 and a 1/6 mitral systolic ejection murmur. Thyroid examination was normal and thyroid chemistries were TSH 4·0, total T4 7·2 micrograms/dl, total T3 1·2 ng/ml. ECG confirmed atrial fibrillation with a ventricular rate of 96/min, normal voltage and QRS interval, and a QRS axis of minus 45 degrees, signifying left anterior fascicular delay.

Question What should be done now?

Comment Three issues need to be reviewed with the patient: the most likely cause of the atrial fibrillation; the desirability of cardioversion, with restoration and maintenance of sinus rhythm; and the rationale for lifelong anticoagulation therapy. In this case, lacking any clinical features of coronary, hypertensive, valvular or myopathic heart disease, the atrial fibrillation is most likely to be due to electrical and, perhaps, mechanical remodeling of the atria. Heterogeneous electrical properties throughout the atria have evoked the trigger to provoke atrial fibrillation and the substrate to sustain it.1,2 Moreover, the persistence of atrial fibrillation undermines effective heart rate control, resulting in exertional dyspnea and palpitations, and is associated with thrombus formation, especially in the left atrial appendage.3 Consequently, there is an increased risk of thromboembolic stroke, particularly when atrial fibrillation converts back to sinus rhythm. Careful appraisal of the patient’s story suggested that her symptoms had been present for several days. Three options for abrogating the atrial fibrillation were reviewed with her: (1) anticoagulation and antiarrhythmic drug usage with interim transthoracic echocardiography (TTE) to quantify left ventricular systolic function, left atrial size, and search

for atrial thrombus; then, elective cardioversion in 3–4 weeks, if drug therapy did not remit the atrial fibrillation; (2) 3–4 weeks of warfarin anticoagulation and interim TTE, followed by elective cardioversion with an INR 2·0–3·0 to minimize the risk of thromboembolic stroke;4 (3) anticoagulation with heparin, expedited hospitalization for transesophageal echocardiography (TTE), and facilitated cardioversion, if there were no evidence of atrial thrombi. The patient was assured that TEE would provide a clearer view of the left atrial appendage than TTE;5 could quantify left atrial appendicular inflow and outflow blood velocity, enhancing clot identification and possibly improving stroke risk estimate;6 and had been carefully documented in a randomized clinical trial to be an effective and safe alternative strategy for guiding cardioversion of atrial fibrillation, with an embolic event risk 0·8% compared to conventional 3–4-week anticoagulation with TTE of 0·5%.7 Additionally, TEE-guided cardioversion could reduce the rate of recurrent atrial fibrillation during the year following restoration of sinus rhythm.8 Because the electrical remodeling and reverse remodeling processes that accompany atrial fibrillation and its correction occur over several days,9 other mechanisms, such as reverse mechanical or structural remodeling of the atria, would be beneficially involved in the prevention of atrial fibrillation by prompt cardioversion.1 The patient understood that the longer atrial fibrillation persisted the more difficult it would be to restore sinus rhythm and prevent recurrence of this disorder, and opted for facilitated cardioversion. 921


Follow up The TEE showed normal LV wall thickness, size and function (LVEF 60%); the LA was normal in size (38 mm) and contained no clot even in its appendage. That same morning cardioversion was successfully accomplished with a 100 J transthoracic countershock restoring sinus rhythm. The patient was discharged on low molecular weight heparin for several days, and anticoagulation with warfarin with an intended INR of 2·0–3·0 was continued. The absence of diabetes, heart failure or prior thrombotic stroke or transient ischemic attack suggested a low future risk for atrial fibrillation related stroke,10 as did the normal LV and LA parameters on echocardiography.11 Because population-based studies have showed the attributable risk of stroke for atrial fibrillation to increase substantially in the eighth and ninth decades,12 plans were made for long-term anticoagulation and warfarin surveillance via an anticoagulation clinic. Lowdose blocker and angiotensin converting enzyme inhibitor therapy was given to lower systolic pressure.13 Agents of these two antihypertensive classes were also selected to blunt adrenergically mediated triggers for atrial fibrillation1 and to suppress angiotensin II contributions to an arrhythmogenic electrical dispersion substrate via regional increases in L-type Ca currents (I Ca, L), regional decreases in transient outward potassium currents (It0), and atrial fibrosis and atrial myocyte hypertrophy.14,15

References 1.Allessie MA, Boyden PA, Camm AJ et al. Pathophysiology and prevention of atrial fibrillation. Circulation 2001;103: 769–77. 2.Fareh S, Villemaire C, Nattel S. Importance of refactoriness heterogenity in the enhanced vulnerability to atrial fibrillation induction caused by tachycardia-induced atrial electrical remodeling. Circulation 1998;98:2202–9.

Case 2

922

3.Aberg H. Atrial fibrillation. A study of atrial thrombosis and system embolism in a necropsy material. Acta Med Scand 1996;195:373–9. 4.Laupacis A, Albers G, Dalen J, Dunn MI, Jacobson AK, Singer DE. Antithrombotic therapy in atrial fibrillation. Chest 1998; 144(Suppl.):579s–89s. 5.Manning WJ, Weintraub RM, Waksmonski CA et al. Accuracy of transesophageal echocardiography for identifying left atrial thrombi: A prospective intraoperative study. Ann Intern Med 1995;123:817–22. 6.Atrial fibrillation investigators. Atrial fibrillation risk factors for embolization and efficacy of antithrombotic therapy. Arch Intern Med 1994;154:1149–57. 7.Klein AL, Grimm RA, Murray RD et al. For the assessment of cardioversion using transesophageal echocardiography investigators. Use of transesophageal echocardiography to guide cardioversion in patients with atrial fibrillation. N Engl J Med 2001;344:1411–20. 8.Silverman DI, Manning WJ. Strategies for cardioversion of atrial fibrillation: time for a change? N Engl J Med 2001;344:1468–9. 9.Yu WC, Lee SH, Tai CT et al. Reversal of atrial electrical remodeling following cardioversion of longstanding atrial fibrillation in man. Cardiovasc Res 1999;42:470–6. 10.Stroke Prevention in Atrial Fibrillation Investigators. Prevention of thomboembolism in atrial fibrillation: I. Clinical features of patients at risk. Ann Intern Med 1992; 116:1–5. 11.Stroke Prevention in Atrial Fibrillation Investigators. Predictors of thromboembolism in atrial fibrillation: II. Echocardiographic features of patients at risk. Ann Intern Med 1992;116:6–12. 12.Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: The Framingham Study. Stroke 1991;22:983–8. 13.Kaplan NM. New issues in the treatment of isolated systolic hypertension. Circulation 2000;102:1079–81. 14.Sadoshima J, Isumo S. The cellular and molecular response of cardiac myocytes to mechanical stress. Annu Rev Physiol 1997;59:551–71. 15.Goette A, Arndt M, Röcken C et al. Regulation of angiotensin II receptor subtypes during atrial fibrillation in humans. Circulation 2000;101:2678–81.

A 69 year old man had a history of chronic coronary heart disease. He had sustained two prior myocardial infarctions which cumulatively had resulted in congestive heart failure. Persistent exertional angina prompted cardiac catheterization, which identified three-vessel severe (70% luminal diameter narrowing) obstructive coronary artery disease and severe LV systolic dysfunction (LVEF 30%). Three-vessel coronary bypass surgery was successfully completed, with perioperative use of amiodarone to suppress atrial fibrillation (AF). Medical therapy with angiotensin converting enzyme inhibitor, glycoside, diuretic, statin, aspirin and blocker was continued for chronic heart failure. During outpatient cardiac rehabilitation monitoring sinus rhythm with rare atrial and ventricular premature beats was confirmed and no exertional angina was observed. Exercise capacity improved. One year later the patient complained of exertional fatigue. Repeat ECG showed AF with a ventricular rate of 90 at rest, left ventricular hypertrophy, and old anteroseptal and inferior myocardial infarction patterns. On examination there was no evidence of pulmonary congestion or peripheral edema and the BP was 116/70.

Atrial fibrillation

Question What is the best course of action?

AF after 1 month of amiodarone therapy, electrical cardioversion was successfully accomplished using a biphasic waveform device. Amiodarone was then reduced to 200 mg daily and aspirin reduced to 81 mg daily as INR-adjusted warfarin was continued.

Comment The principal hazard from atrial fibrillation would be enhanced thromboembolic stroke risk. Warfarin anticoagulation can reduce this risk by about two thirds;1 although it has not been conclusively proved that restoration of sinus rhythm will significantly reduce cardiovascular mortality,2 in this case resumption of coordinated atrial contraction (atrial transport and booster pump function) would enhance left ventricular filling by 25%, fortifying left ventricular stroke output via a Starling mechanism while tending to lower left atrial pressure. Symptoms of exertional fatigue and shortness of breath would thereby be improved. Warfarin was initiated, and 3 weeks after attaining an INR of between 2·0 and 3·0 transthoracic electrical cardioversion using a damped sign wave monophasic waveform was attempted. Countershocks of up to 400 J were unsuccessful in restoring sinus rhythm, however. Question What should be done now?

Further comment The patient was informed that antiarrhythmic drug therapy with amiodarone could reduce the heterogeneity of regional atrial electrical properties seen in AF, and that it had been successfully used to suppress AF at the time of his coronary bypass surgery. He was also informed that clinical trial evidence had indicated a significantly lower frequency of AF in this setting (22·5% v 38·0%).3 Additional clinical data had also showed the value of amiodarone in the conversion to and maintenance of sinus rhythm in chronic heart failure patients with AF.4 In the event that AF still persisted after amiodarone therapy then transthoracic cardioversion could again be attempted, this time using a rectilinear biphasic waveform device. These devices have greater efficacy and require less countershock energy in restoring sinus rhythm.5 They utilize impedance compensation, ensuring a constant current delivery to depolarize the heart, and are especially advantageous in patients with high impedance.6 The patient concurred with these plans. Amiodarone therapy 400 mg tid for 1 week, followed by 400 mg daily, was established. Digoxin dosage was reduced from 0·125 mg daily to three times a week to accommodate the amiodarone reduction in digoxin excretion. Warfarin daily dosage was also modified downward to allow for its interaction with amiodarone. The INR was maintained between 2·0 and 3·0. With persisting

Follow up The patient was now taking seven drugs for his chronic heart failure. The quartet of agents (ACE inhibitor, glycoside, diuretic, blocker) fortified the heart mechanically and mitigated excess adrenergic stimulation, thereby reducing triggering mechanisms for atrial fibrillation. They also provided a framework for interdicting the excess adrenergic stimulation that accompanies chronic heart failure and which is deleterious to the progressive LV dilatation and hypertrophy that accompanies this condition. The aspirin and warfarin were deployed to minimize the likelihood of recurrent heart attacks and both ischemic7 and thromboembolic stroke,8 associated with atrial fibrillation. The aspirin was also deployed to reduce mortality risk.9 Amiodarone was utilized to maintain sinus rhythm. Post hoc analysis of clinical trial data provided an additional rational for combined blocker/amiodarone usage: significant reductions in sudden and non-sudden cardiac deaths.10 The complex drug regimen was justified in this case as atrial fibrillation did not recur, a crucial matter, as randomized clinical trial data have indicated that AF is associated with increased mortality in heart failure patients (CHF)11 and is an independent predictor of mortality in CHF patients with implanted cardioverter defibrillators.12 In the future more advanced dual channel implantable defibrillators may allow additional strategies for coping with AF in this seriously ill population.13 References 1.Hirsh J, Dalen JE, Anderson DR et al. Anticoagulants. Mechanism of action, clinical effectiveness, and optimal therapeutic range. Chest 1998;114:445s–69s. 2.Wyse DG, Anderson JL, Antman EM et al. Atrial Fibrillation Follow-Up Investigation of Rhythm Management – the AFFIRM study design. Am J Cardiol 1997;79:1198–202. 3.Giri S, White CM, Dunn AB et al. Oral amiodarone for prevention of atrial fibrillation after open heart surgery, the Atrial Fibrillation Suppression Trial (AFIST): a randomized placebocontrolled trial. Lancet 2001;357:830–6. 4.Deedwania PC, Singh BN, Ellenbogen K, Fisher S, Fletcher F, Singh SN. Spontaneous conversion and maintenance of sinus rhythm by amiodarone in patients with heart failure and atrial fibrillation: Observations from the Veterans Affairs Congestive Heart Failure Survival Trial Antiarrhythmic Therapy (CHFSTAT). Circulation 1998;98:2574–9. 5.Mittal S, Ayati S, Stein KM et al. Transthoracic cardioversion of atrial fibrillation. Comparison of rectilinear biphasic versus

923


damped sign wave monophasic shocks. Circulation 2000; 101:1282–7. 6.Kerber RE, Martins JB, Kienzle MG et al. Energy, current, and success in defibrillation and cardioversion: clinical studies using an automated impedance-based method of energy adjustment. Circulation 1998;77:1038–46. 7.Cairns JA, Theroux P, Lewis HD Jr, Ezekowitz M, Meade TW, Sutton GC. Antithrombotic agents in coronary artery disease. Chest 1998;114:611s–33s. 8.Albers GW, Easton JD, Sacco RL, Teal P. Antithrombotic and thrombolytic therapy for ischemic stroke. Chest 1998;114: 683s–98s. 9.Gum PA, Thamilarasan M, Watanabe J, Blackstone EH, Lauer MS. Aspirin use and all-cause mortality among patients being evaluated for known or suspected coronary artery disease. A propensity analysis. JAMA 2001;286:1187–94. 10.Boutitie F, Boissel J-P, Connolly SJ et al. and the EMIAT and CAMIAT investigators. Amiodarone interaction with -blockers. Analysis of the merged EMIAT (European

924

Myocardial Infarct Amiodarone Trial) and the CAMIAT (Canadian Amiodarone Myocardial Infarction Trial) databases. Circulation 1999;99:2268–75. 11.Dries DL, Exner DW, Gersh BJ et al. Atrial fibrillation is associated with an increased risk for mortality and heart failure progression in patients with asymptomatic and symptomatic left ventricular systolic dysfunction: A retrospective analysis of the Solvd trials. Studies of Left Ventricular Dysfunction. J Am Coll Cardiol 1998;32:695–703. 12.Antiarrhythmics Versus Implantable Defibrillators (AVID) Investigators. A comparison of antiarrhythmic drug therapy with implantable defibrillators in patients resuscitated from near-fatal ventricular arrhythmias. N Engl J Med 1997;337: 1576–84. 13.Friedman PA, Dijkman B, Warman EN et al. for the Worldwide Jewel AF Investigators. Atrial therapies reduce atrial arrhythmia burden in defibrillator patients. Circulation 2001;104:1023–8.

72

Ventricular dysrhythmias: pharmacologic v non-pharmacologic treatment L Brent Mitchell

Case scenario 1

At the age of 60 years, Mr D presents a therapeutic choice in the management of a patient who has had ventricular tachycardia (VT). Two years previously he had experienced an acute anterolateral myocardial infarction but did not receive thrombolysis. When post-MI angina continued, he underwent cardiac catheterization which demonstrated three vessel coronary artery disease and compromised LV function. Accordingly, he underwent three vessel coronary artery bypass surgery. Thereafter, Mr D was free of any potential cardiovascular symptoms until now, when he presents to the Emergency Department after the sudden onset, while performing minor car maintenance, of presyncope followed by diaphoresis and dyspnea. He is found to have VT at a rate of 175 beats per minute with a right bundle branch “block” pattern, a left axis deviation QRS morphology and 2 : 1 retrograde VA block (Figure 72.1). His systolic BP is 85 mmHg. Under general anesthesia this rhythm is cardioverted with a 50 J QRS synchronous D/C shock to another VT with a right bundle branch “block” configuration and right axis deviation QRS morphology (Figure 72.2). A second 200 J QRS synchronous D/C shock restores normal sinus rhythm. His initial postconversion evaluation reveals no transient or reversible causes of VT (such as acute myocardial infarction, electrolyte disturbance or proarrhythmic drug effect). On a treadmill exercise test Mr D exercises for 4 minutes, reaching stage II of the standard Bruce protocol. The end point was dyspnea at a maximum heart rate of 153 beats per minute (target heart rate 136 beats per minute). The blood pressure response is flat. There is no evidence of reversible myocardial ischemia or exercise-related arrhythmia. A 24 hour ambulatory ECG shows sinus rhythm within the physiologic rate range and only rare isolated premature ventricular beats (two on the 24 hour recording). A cardiac catheterization demonstrates patent coronary bypass grafts and no new native coronary artery lesions. His left ventricular angiogram reveals an anteroapical LV aneurysm. A catheter electrophysiologic study shows normal sinus nodal, atrial, and AV nodal electrophysiology. The HV interval is prolonged to 60 msec. Programmed ventricular stimulation induces sustained VT (Figure 72.3) that matches the initial presenting VT and could be pace terminated. The mechanism of the VT is not bundle branch re-entry. Intravenous procainamide is administered (total dose 1 g). Thereafter, VT is no longer inducible.

Question: What treatment, if any, should now be applied?

Comment For this case, current available “best” evidence does not define a single preferred therapeutic approach. This 60 year old male presents with hypotensive VT in the setting of

stable atherosclerotic heart disease on the background of a previous anterolateral wall myocardial infarction and a left ventricular aneurysm. After ruling out a transient or reversible cause for his VT and optimizing the therapy of underlying structural heart disease, an electrophysiologic study demonstrated the persistence of a substrate for VT that was suppressed by intravenous procainamide. Viable alternatives for the treatment of this man’s high future risk of VT recurrence include (a) standard antiarrhythmic drug therapy individualized by the Holter monitoring 925


I

aVR

V1

V4

II

aVL

V2

V5

III

aVF

V3

V6

II

25 mm/s 10 mm/mV

~0·15 Hz–40 Hz

12625

Figure 72.1 Presenting ventricular tachycardia of right bundle branch “block” pattern with left axis deviation QRS morphology and 2:1 retrograde VA block

I

aVR

V1

V4

II

aVL

V2

V5

III

aVF

V3

V6

II

25 mm/s 10 mm/mV

~0·15 Hz–40 Hz

12626

Figure 72.2 Ventricular tachycardia with right bundle branch “block” configuration and right axis deviation QRS morphology after low energy cardioversion

approach; or (b) the electrophysiologic study approach; or (c) empiric amiodarone therapy; or (d) implantation of a tiered therapy implantable cardioverter defibrillator (ICD); or (e) surgical/transcatheter ablative therapy. In addition, grade B evidence supports the concomitant use of ancillary blocker therapy where possible.1 Mr D’s clinical circumstance is a frequent clinical scenario, and such patients have been the most commonly recruited subjects of clinical trials evaluating treatment for life threatening VT. Nevertheless, there is still uncertainty as 926

to the most appropriate initial form of therapy. Randomized clinical trials relative to the treatment of this patient population include the Calgary study;2 ESVEM: Electrophysiologic Study versus Electrocardiographic Monitoring Study;3 CASCADE: Cardiac Arrest in Seattle: Conventional versus Amiodarone Drug Evaluation;4 the Antiarrhythmics versus Implantable Defibrillator (AVID) trial;5 the Cardiac Arrest Study Hamburg (CASH);6 and the Canadian Implantable Defibrillator Study (CIDS).7 It should be noted that each of these trials compares one therapy to another therapy. To

Ventricular dysrhythmias

50 mm/s

500 ms

21–1

26-aVF

27-V1

32-V6

1-RVA

2-pHIS

Figure 72.3 Ventricular tachycardia of right bundle branch “block” configuration and right axis deviation QRS morphology induced by program stimulation

date, ethical considerations have precluded comparisons of one form of therapy to no therapy. Both the Calgary study and the ESVEM trial compared the long-term outcome of patients treated with standard antiarrhythmic drug therapy selected by the Holter monitoring approach to that selected by the electrophysiologic study approach. In a small population of drug naive patients with inducible VT, the Calgary study reported superiority of therapy selected by the electrophysiologic study approach. Furthermore, such therapy was associated with low VT recurrence and sudden death probabilities in the Calgary trial. In contrast, in a larger population of drug resistant patients with either inducible VT or VF, the ESVEM trial reported equivalence of therapy selected by either of the approaches. Furthermore, such therapy was associated with a high VT/VF recurrence rate in the ESVEM trial. Mr D’s clinical situation was most comparable to the patients in the Calgary trial who had arrhythmia substrates for which the electrophysiologic study performs best (inducible VT in the setting of stable atherosclerotic heart disease). Furthermore, Mr D did not have sufficient spontaneous ventricular arrhythmia on his Holter monitor to provide an index to guide the Holter monitoring approach. Accordingly, Mr D had standard antiarrhythmic drug therapy – procainamide in this case – selected by the electrophysiologic study approach. Empiric amiodarone therapy was considered. The most impressive data supporting the use of empiric amiodarone in

a patient such as Mr D emerges not from randomized clinical trials but rather from descriptions of excellent long-term outcomes of patients who had failed other therapy and then received empiric amiodarone. Some would argue that CASCADE demonstrated the superiority of empiric amiodarone over standard antiarrhythmic drug therapy in this setting. However, we must recall that CASCADE included only patients with out-of-hospital VF cardiac arrests. Furthermore, in CASCADE, empiric amiodarone appeared superior to antiarrhythmic therapy selected by either the electrophysiologic study approach or the Holter monitoring approach (rather than the better of these two approaches). Finally, many of the patients in the standard antiarrhythmic therapy limb of CASCADE received therapy that was predicted to be (and presumably was) ineffective. Nevertheless, this report and the concerns regarding the inefficacy of standard antiarrhythmic therapy in the ESVEM trial have allowed empiric amiodarone to emerge as the “gold standard” pharmacologic therapy to which ICD therapy is being compared in ongoing trials. Accordingly, this therapeutic choice would also have been appropriate. Electrosurgery/transcatheter ablation therapy was considered. Ablative therapies have had their greatest success in patients with atherosclerotic heart disease and previous myocardial infarction. This is particularly true if another indication exists for open heart surgery, such as a need for a coronary revascularization procedure. Nevertheless, this 927


approach exposes the patient to an important surgical mortality and may not be appropriate for patients with a single episode of VT who do not need coronary revascularization and have other therapeutic alternatives. Of course, had the electrophysiologic study demonstrated a VT that required the participation of the right bundle branch (bundle branch re-entry), then ablation of the right bundle branch would have been a preferred form of therapy. The use of an implantable automatic cardioverter defibrillator (ICD) was considered. Once reserved for patients with VT/VF resistant to other therapy, the long-term results of ICD therapy have been impressive. Furthermore, the Data and Safety Monitoring Board of the AVID trial recently recommended that the study be terminated after slightly more than 1000 patients had been enrolled as a statistically significant advantage had emerged relative to all-cause mortality in favour of the ICD over antiarrhythmic drug therapy, consisting of empiric amiodarone for the vast majority of patients with a few having received sotalol, that was predicted to be effective by either the electrophysiologic study approach or the Holter monitoring approach. However, preliminary costing analysis has suggested that the ICD may not be an economically competitive strategy, with a cost per year of life saved of approximately $130 000. Nevertheless, patients such as Mr D frequently receive an ICD. Of note, the results of the Multicenter Automatic Defibrillator Implantation Trial (MADIT),8 which suggested the superiority of ICD therapy over “conventional” therapy, are not relevant to patients such as Mr D. All the patients enrolled in MADIT had demonstrated drug-resistant VT by virtue of continued VT induction after the administration of IV procainamide.

Outcome Mr D had therapy initiated with oral procainamide. On Procan-SR 1000 mg q 6 hours, his procainamide level was 26 mol/l (therapeutic range 17–43 mol/l) and his Case scenario 2

928

NAPA level was 19 mol/l. A drug assessment electrophysiologic study was performed and no VT/VF was inducible. In follow up, Mr D’s procainamide dosage has been altered to maintain the procainamide/NAPA levels that were predicted to be effective, requiring dosages as high as 1250 mg q 6 hours and as low as 750 mg q 6 hours. He has now been receiving this therapy for 6 years without arrhythmia recurrence. References 1.Szabo BM, Crijns HJGM, Wiesfeld ACP et al. Predictors of mortality in patients with sustained ventricular tachycardias or ventricular fibrillation and depressed left ventricular function: importance of beta blockade. Am Heart J 1995;130:281–6. 2.Mitchell LB, Duff HJ, Manyari DE, Wyse DG. A randomized clinical trial of the noninvasive and invasive approaches to drug therapy of ventricular tachycardia. N Engl J Med 1987; 317:1681–7. 3.Mason JW, Electrophysiologic Study versus Electrocardiographic Monitoring Investigators. A comparison of electrophysiologic testing with Holter monitoring to predict antiarrhythmic drug efficacy for ventricular tachyarrhythmias. N Engl J Med 1993;329: 445–51. 4.The CASADE Investigators. Randomized antiarrhythmic drug therapy in survivors of cardiac arrest (the CASADE study). Am J Cardiol 1993;72:280–7. 5.AVID Investigators. Antiarrhythmics versus Implantable Defibrillator (AVID): rationale, design and methods. Am J Cardiol 1995;75:470–5. 6.Siebels J, Cappato R, Ruppel R et al, CASH Investigators. ICD versus drugs in cardiac arrest survivors: preliminary results of the cardiac arrest study Hamburg. PACE 1993;16:552–8. 7.Connolly SJ, Gent M, Roberts RS et al, CIDS Investigators. Canadian Implantable Defibrillator Study (CIDS): study design and organization. Am J Cardiol 1993;72:103F–108F. 8.Moss AJ, Hall J, Cannom DS, Doubert JP et al, Multicenter Automatic Defibrillator Implantation Trial Investigators (MADIT). Improved survival with an implanted defibrillator in patients with coronary disease at high risk for ventricular arrhythmia. N Engl J Med 1996;335:1933–40.

Miss M presents a therapeutic choice in the management of a propensity to ventricular tachycardia (VT) at the age of 17 years. When Miss M was 3 years old her father died suddenly during sleep at the age of 28 years. The father’s autopsy documented the presence of a right ventricular cardiomyopathy. When Miss M was 13 years of age her sister died suddenly during sleep at the age of 18 years. The sister’s autopsy documented the presence of a right ventricular cardiomyopathy. Understandably, Miss M’s mother became alarmed and Miss M was referred for her initial evaluation. That evaluation included clinical examination, a 2D echocardiographic examination, a biventricular radionuclide angiogram, a Holter monitoring examination, and a treadmill exercise test. All were entirely normal. Over the next three years, annual clinical examinations, Holter monitoring examinations, and echocardiographic examinations remained normal. However, her annual evaluation at the age of 17 years included a Holter examination showing frequent ventricular premature beats (14 VPB/hour) that were complex to the level of four beat salvos of consecutive ventricular beats. Accordingly, a search for evidence of right ventricular structural heart disease and for a propensity to ventricular tachyarrhythmia was undertaken.

Ventricular dysrhythmias

An echocardiographic examination shows questionable right ventricular enlargement. A biventricular radionuclide angiogram shows normal right and left ventricular size and function at rest but the right ventricle becomes mildly and diffusely hypokinetic with supine bicycle exercise. Cardiac catheterization is performed and the right ventricular angiogram demonstrates two dyskinetic right ventricular segments. A treadmill exercise test precipitates a ventricular triplet. Finally, a transvenous catheter electrophysiologic study is performed. Triple ventricular extra stimuli applied during right ventricular pacing at a rate of 150 bpm initiates a polymorphic VT that then stabilizes into sustained monomorphic VT at a rate of 230 bpm. The monomorphic VT has a left bundle branch “block” configuration and normal frontal plane QRS axis morphology.

Question: What treatment, if any, should now be applied?

Comment Although this high-risk state is rare, there is persuasive evidence in the literature that should help guide therapy. The studies performed in this 17 year old female indicate both structural right ventricular disease and a propensity to VT with a QRS morphology consistent with a right ventricular “origin”. One must conclude that Miss M has developed the same arrhythmogenic right ventricular dysplasia (ARVD) that had affected her father and her sister. The natural history of this disorder as defined by her father and her sister strongly suggests that Miss M is at high risk of sudden death. In this setting, Miss M’s viable therapeutic alternatives include individualized antiarrhythmic drug therapy selected by (a) the Holter monitoring approach or (b) the electrophysiologic study approach; or (c) empiric amiodarone; or (d) electrosurgery/transcatheter ablation; or (e) placement of an implantable automatic cardioverter defibrillator (ICD). Miss M’s clinical circumstance is unusual and has not been the subject of specific randomized clinical trials. The closest patient population studied to date is that of the Multicenter Automatic Defibrillator Implantation Trial (MADIT),1 which suggested that early use of an ICD was superior to “conventional” therapy. However, the study population of MADIT (patients with remote myocardial infarction, compromised left ventricular function, and inducible sustained VT/VF that could not be suppressed by intravenous procainamide) was unrelated to that of Miss M. Accordingly, a therapeutic decision was made without definitive clinical trial data. Individualized antiarrhythmic drug therapy was considered. Both approaches to individualized antiarrhythmic drug therapy have been validated in patient populations dominated by atherosclerotic heart disease. Furthermore, such therapy is more prone to failure when the underlying structural heart disease is cardiomyopathic rather than

atherosclerotic. Nevertheless, the factor that most recommended an alternative approach was that if a spontaneous episode of VT/VF subsequently occurred, it would be impossible to distinguish between a failure of drug therapy to prevent an episode of VT/VF that was inevitable and a proarrhythmic drug response that precipitated an episode of VT/VF that was not otherwise going to occur. Empiric blocker therapy was considered. Steinbeck et al 2 have published evidence suggesting that empiric blocking therapy is as effective as antiarrythmic drug therapy selected by the electrophysiologic approach for the prevention of VT/VF. Careful scrutiny of their results shows that empiric blocking therapy is actually more effective than is standard drug therapy predicted to be ineffective by the electrophysiologic study approach, but is less effective than is a standard drug therapy predicted to be effective by the electrophysiologic study approach. Nevertheless grade B evidence supports the concomitant use of ancillary blocking therapy where possible.3 Empiric amiodarone was considered. Empiric amiodarone has prophylactic efficacy in patient populations at high risk of VT/VF who have not yet experienced a spontaneous VT/VF episode. Such trials include that of Grupo de Estudio de la Sobrevida en la Insuficiencia Cardiaca en Argentina (GESICA),4 CHF-STAT,5 the Canadian Amiodarone Myocardial Infarction Arrhythmia Trial (CAMIAT),6 and the European Myocardial Infarct Amiodarone Trial (EMIAT).7 Furthermore, GESICA and CHF-STAT suggested that empiric amiodarone is particularly effective when the underlying structural heart disease is cardiomyopathic rather than atherosclerotic. Finally, the proarrhythmic potential of amiodarone is very low (approximately 1%), rendering the probability of a future need to distinguish drug failure from proarrhythmia vanishingly small. Nevertheless, the factor that most recommended an alternative approach was the expected adverse effect profile of amiodarone in one so young who might require therapy for more than 50 years. Electrosurgery/transcatheter ablation therapy was considered. Ablative therapies have had their greatest success in patients with atherosclerotic heart disease and previous myocardial infarction.8 The probability of long-term success 929


is compromised in patients with a cardiomyopathy, especially when the natural history of that cardiomyopathy is progression with new lesion formation. Although a right ventricular disconnection electrosurgical procedure has been developed for the treatment of patients with arrhythmogenic right ventricular dysplasia, the long-term consequences of right ventricular failure are frequently devastating. Nevertheless, the factor that most recommended an alternative approach was consideration of the surgical risk in a person who has not yet had a spontaneous episode of VT/VF. These considerations favored the implantation of an ICD. ICD therapy provides excellent protection from sudden death in patients whose structural heart disease is atherosclerotic or cardiomyopathic.9 Furthermore, the proarrhythmia potential of an ICD is low, thereby reducing concern as to a future need to distinguish between therapeutic failure and therapy-related proarrhythmia. Finally, the availability of single lead transvenous ICD conformations allows the therapy to be instituted with a low-risk surgical procedure.

Outcome Miss M had her ICD implanted when she was 17 years of age. The procedure was uncomplicated and her convalescence was unremarkable. Three years later, her ICD reached end of battery life indicators. During these years she did not have a spontaneous episode of VT/VF and had received no therapies from her device. A new ICD impulse generator was implanted when Miss M was 20 years of age. One year later, 4 years after her initial presentation, Miss M was playing baseball (at bat) when she suddenly felt marked presyncope followed by a shock from her ICD. She then felt well and completed her turn at bat. Subsequent interrogation of her ICD showed that the pre-event rhythm was sinus tachycardia at 162 beats per minute that gave way to a ventricular tachycardia at 300 beats per minute. These rhythm assessments were augmented by the availability of ICD

930

stored intracardiac electrograms. She is presently undergoing genetic counseling relative to her desire to conceive a child – evidence of a satisfactory quality of life.

References 1.Moss AJ, Hall J, Cannom DS et al. Multicenter Automatic Defibrillator Implantation Trial Investigators. Improved survival with an implanted defibrillator in patients with coronary disease at high risk for ventricular arrhythmia. N Engl J Med 1996;335:1933–40. 2.Steinbeck G, Anderson D, Bach P et al. A comparison of electrophysiologically guided antiarrythmic drug therapy with beta blocker therapy in patients with symptomatic sustained ventricular arrythmias. N Engl J Med 1992;327:987–92. 3.Szabo BM, Crijns HJGM, Weisfeld ACP et al. Predictors of mortality in patients with sustained ventricular tachycardias or ventricular fibrillation and depressed left ventricular function: importance of beta blockade. Am Heart J 1995;130:281–6. 4.Doval HC, Nul DR, Grancelli HO et al, GESICA Investigators. Randomized trial of low-dose amiodarone in severe congestive heart failure. Lancet 1994;344:493–8. 5.Singh SN, Fletcher RD, Fisher SG et al, CHF-STAT Investigators. Amiodarone in patients with congestive heart failure and asymptomatic ventricular arrhythmia. N Engl J Med 1995;333:77–82. 6.Cairns JA, Connolly SJ, Roberts R, Gent M, CAMIAT (Canadian Amiodarone Myocardial Infarction Arrhythmia Trial) Investigators. Randomized trial of outcome after myocardial infarction in patients with frequent or repetitive ventricular premature depolarizations: CAMIAT. Lancet 1997;349:675–82. 7.Julian DG, Camm AJ, Frangin G et al, EMIAT Investigators. Randomized trial of effect of amiodarone on mortality in patients with left ventricular dysfunction after recent myocardial infarction: EMIAT. Lancet 1997;349:667–74. 8.Levy S. Non-medical therapy of ventricular tachyarrhythmias. Eur Heart J 1989;10(Suppl. E):48–52. 9.Gillis AM. The current status of the implantable cardioverter defibrillator. Annu Rev Med 1996;47:85–93.

73

Bradyarrhythmias: choice of pacemaker John A Boone

A 66 year old man had a VVIR pacemaker implanted because of syncope owing to complete heart block. He was previously well except for one documented and, possibly, a second undocumented episode of atrial fibrillation. An ECG taken prior to pacemaker insertion showed sinus bradycardia. When he presented with transient complete heart block it was believed he would do well with VVIR pacing. However, he now presents with a form of “pacemaker syndrome” characterized by a combination of fatigue and the presence of cannon A waves from intermittent VA conduction (Figure 73.1). Furthermore, episodes of paroxysmal atrial fibrillation became more frequent and, interestingly, he was less symptomatic when in atrial fibrillation, presumably due to the absence of cannon waves. He was anticoagulated with warfarin. As he was in sinus rhythm more often than in atrial fibrillation, and as he was symptomatic with asynchronous pacing, the VVIR pacemaker was removed and his pacing was “upgraded” to DDDR pacing (Figure 73.2).

Case scenario 1

Question I

II

III

aVR

V1

aVL

V2

Should this patient have had a DDD type pacemaker inserted from the start?

Comment

aVF

V3

Figure 73.1 VVI pacing with ventricular dissociation (arrows)

I

aVR

II

aVL

V2

III

aVF

V3

Figure 73.2 Pacing pacemaker

following

V1

implantation

of

DDDR

With established symptoms of syncope in the presence of documented complete heart block there is little doubt about the need for pacemaker implant for this patient. Rather, one seeks evidence to support the choice of the optimum pacemaker mode. The treatment of symptomatic acquired complete heart block by cardiac pacing became established practice in the 1960s. Following the introduction of permanent cardiac pacing – initially VOO followed by VVI (see Chapter 42 for description of pacemakers) – published experience clearly documented that patients with complete heart block and syncope had an improved survival.1–3 Although one might have felt intuitively that “physiologic” pacing was indicated from the start in this patient, a recent trial that randomized patients to receive either VVI (R) or physiologic pacing failed to show the benefit expected from dual chamber pacing, this despite the observations from several older studies that demonstrated a benefit of atrial synchrony over VVI pacing4–7 in evaluating cardiac output, exercise capacity, and feeling of wellbeing. However, the increase in cardiac output with exercise is mostly achieved by an increase in heart rate rather than by 931


AV synchrony.8–10 It was on this basis, namely, to enable adequate tracking of the patient’s physical activity, that VVIR pacing was originally chosen in this patient. However, factors other than exercise capacity may determine suitability of a particular pacing mode, as exemplified by this patient, wherein VVI pacing caused a form of pacemaker syndrome – that is, a clinical state wherein stroke volume is reduced by virtue of asynchrony between atrial transport function and ventricular systole. The incidence of pacemaker syndrome in VVI pacing is not known but estimates vary from 0·1% to 5%.11,12 Other benefits attributed to DDD pacing include prevention of atrial fibrillation, prevention of embolic stroke and other systemic emboli as well as protection from congestive heart failure and early mortality. The data suggesting these benefits are mostly from studies that are retrospective and non-randomized.13,14 However, when these data are used to calculate annual event rates, there is a two thirds risk reduction for atrial fibrillation and a one third reduction for death in patients who have received DDD pacing. One prospective randomized trial that compared atrial to ventricular pacing found significantly less atrial fibrillation and thromboemboli in the atrial-paced patients compared with those receiving ventricular pacing, but there was no significant difference in congestive heart failure or mortality. Future prospective randomized trials may or may not confirm these findings.

References 1.Friedberg CK, Donoso E, Stein WB. Nonsurgical acquired heartblock. Ann NY Acad Sci 1964;111:833–47. 2.Donmoyer TL, DeSanctis RW, Austen WG. Experience with implantable pacemakers using myocardial electrodes in the management of heartblock. Ann Thorac Surg 1967;3:213–27. 3.Edhag O, Swahn A. Prognosis of patients with complete heartblock or arrhythmic syncope who are not treated with artificial

Case scenario 2

932

pacemakers: a long-term follow-up study of 101 patients. Acta Med Scand 1976;200:457–63. 4.Connolly SJ, Kerr CR, Gent et al. Effect of physiologic pacing versus ventricular pacing on the risk of stroke and death due to cardiovascular causes. Canadian Trial of Physiologic Pacing Investigators (CTOPP). N Engl J Med 2000;342:1385–91. 5.Perrins EJ, Morley CA, Chan SL, Sutton R. Randomized control trial of physiological and ventricular pacing. Br Heart J 1983;50:112–17. 6.Yee R, Benditt DG, Kostuk WJ et al. Comparative functional effects of chronic ventricular demand and atrial synchronous ventricular inhibited pacing. PACE 1984;7:23–8. 7.Rediker DE, Eagle KA, Homma S et al. Clinical and hemodynamic comparison of VVI versus DDD pacing in patients with DDD pacemakers. Am J Cardiol 1988;61:323–9. 8.Fananapazir L, Bennett DH, Monks P. Atrial synchronized ventricular pacing: contribution of the chronotropic response to improved exercise performance. PACE 1983;6:601–8. 9.Ehrsson SK. Influence of heart rate and atrioventricular synchronization on maximal work tolerance in patients treated with artificial pacemakers. Acta Med Scand 1983;214: 311–15. 10.McMeekin JD, Lautner D, Hanson S, Gulamhusein SS. Importance of heart rate response during exercise in patients using atrial ventricular synchronous and ventricular pacemakers. PACE 1990;13:59–68. 11.Rosenqvist M, Brandt J, Schuller H. Long-term pacing in sinus node disease: the effects of stimulation mode on cardiovascular morbidity and mortality. Am Heart J 1988;1126:16–22. 12.Santini M, Alexidou G, Ansalone G et al. Relation of prognosis of sick-sinus syndrome to age, conduction defects and modes of permanent cardiac pacing. Am J Cardiol 1990;65: 729–35. 13.Hesselson AB, Parsonnet V, Bernstein AD, Bonavita GJ. Deleterious effects of long-term single chamber ventricular pacing in patients with sick-sinus syndrome: the hidden benefits of dual chamber pacing. J Am Coll Cardiol 1992;19: 1542–9. 14.Andersen HR, Thuesen L, Bagger JP, Vesterlund T, Thomsen PEB. Prospective randomized trial of atrial versus ventricular pacing in sick-sinus syndrome. Lancet 1994;344:1524–8.

A 30 year old woman complains of sudden shortness of breath and exhaustion occurring during moderate to severe exertion. She is fine at the beginning of exercise, but if she continues she feels as if she had “hit a brick wall”. She is a member of a womens softball team and the example she gives is when she hits what she believes to be a home-run she would run past first base without difficulty, but beyond that she would suddenly be incapable of running and would barely make it to second base. On one occasion she lost consciousness during exertion. Her history otherwise is unremarkable and clinical examination is completely normal. The ECG shows sinus rhythm with a first degree AV block and a PR interval of 0·38 seconds. An echocardiogram is normal. An exercise stress test is performed according to the Bruce protocol. Upon completion of Stage III her heart rate reaches 166 beats per minute with a PR interval of 0·34 seconds (Figure 73.3). The P waves are superimposed on the terminal portion of the QRS. At this point she experiences sudden exhaustion and the test is terminated.

Bradyarrhythmias: choice of pacemaker

II I

At onset of exercise aVR

II

V1 aVR aVR

I

At 6 min of exercise aVR

V5 At 9 min of exercise aVR

aVR

V1 V5

I

II

V5

V1 At onset of exercise

Figure 73.3 Exercise ECG prior to pacemaker implant. Arrows indicate P waves. Note the almost fixed first degree block such that, at the peak of exercise, atria and ventricles depolarize spontaneously

Question What advice would you now give her?

Comment In this rather unusual circumstance, a search for evidence from clinical trials to guide therapy would be better served by reliance on clinical judgment and a sound knowledge of electromechanical physiology. This woman’s symptoms were the result of an almost fixed first degree AV block, such that with exercise atrial contraction would frequently occur against closed AV valves during the period of ventricular contraction. The consequence of this is that the atria “empty” in a retrograde fashion during atrial systole (producing the equivalent of “cannon waves”) resulting in severely reduced filling volumes. The result is a reduced stroke volume and a fall in cardiac output during heart rate acceleration. This patient received implantation of a DDD pacemaker with a good clinical result. The device was programmed to a nominal AV delay of 200 msec, thus permitting normal

At 6 min of exercise

At 9 min of exercise

Figure 73.4 Exercise ECG after DDD pacemaker implant. Arrows indicate P waves. Note the DDD pacemaker mode maintains a physiologic AV interval

ventricular filling. A repeat exercise test demonstrated ventricular pacing “tracking” sinus rhythm with a more physiologic PR interval (Figure 73.4). The original published ACC/AHA Task Force guidelines for the implantation of cardiac pacemakers in 1991 state that there is no evidence to support pacemaker implantation in patients with isolated first degree AV block, and thus assign this condition to a class III recommendation – that is, there is general agreement that pacemakers are not necessary. The revised guidelines, however,1 acknowledge the usefulness of pacemaker implantation for patients with symptomatic first degree AV block. This wise decision is largely based on physiologic need substantiated by the comparison of atrioventricular versus ventricular pacemaker activity on cardiac function as well as the relative low prevalence of these types of cases in the population. This physically active patient clearly benefitted from the pacemaker.

References 1.Barold SS. ACC/AHA guidelines for implantation of cardiac pacemakers. PACE 1993;16:1221.

933

74

Valvular heart disease: timing of surgery Adrian P Banning, Brian B Gribbin

Case 1

A 74 year old man is referred to the cardiology clinic with a 3 month history of worsening breathlessness. Three years earlier he was referred for cardiology review after a systolic murmur had been detected at a routine medical examination. At that time he was asymptomatic, despite an active lifestyle and a Doppler echocardiogram that had shown evidence of a calcified aortic valve with a peak Doppler instantaneous gradient of 70 mmHg (mean gradient 40 mmHg), left ventricular hypertrophy and dynamic left ventricular systolic function. In the absence of symptoms he was advised to avoid sudden or strenuous exercise and of the need for endocarditis prophylaxis. A 6 month follow-up appointment was arranged for clinical assessment and repeat Doppler echocardiography. Subsequently he defaulted from all medical follow up. He now presents with severe exertional dyspnea, orthopnea and paroxysmal nocturnal dyspnea. He has no risk factors for coronary artery disease and there is no history of exertional chest pain or presyncope. Examination reveals a slow upstroke carotid pulse, blood pressure of 115/75, elevated venous pressure, sustained left ventricular impulse, soft aortic closure sound and an ejection systolic murmur. Examination of the lungs reveals a right-sided pleural effusion and bibasal crepitations. An ECG shows sinus rhythm with voltage criteria of severe left ventricular hypertrophy and a left ventricular strain pattern (Figure 74.1). Doppler echocardiography demonstrates a hypertrophied, dilated left ventricle with severe global impairment of systolic function, and an ejection fraction estimated at 0·20. There is evidence of moderate mitral regurgitation. The aortic valve is heavily calcified with restricted opening and a Doppler peak gradient of 40 mmHg (mean gradient 32 mmHg). The continuity equation measures the aortic valve area at 0·5 cm2.

Question Was the initial management of this man appropriate, and what is the appropriate management now?

Comment This case presents two major issues calling for evidence to support the correct timing of surgical intervention for aortic stenosis: (a) the asymptomatic patient with severe aortic stenosis; and (b) the symptomatic patient with severe aortic stenosis but with a reduced transvalvular gradient, presumably due to impaired left ventricular function. The incidence of sudden death is increased in patients with severe aortic stenosis. Fortunately, this rarely occurs without premonitory symptoms, and in the elderly in particular the risk of sudden death in an asymptomatic patient 934

is less than the risk of valve replacement. Thus, following careful clinical and echocardiographic assessment, asymptomatic elderly patients with severe aortic stenosis can be managed conservatively with regular but close outpatient review at least every 6 months.1 However, any genuine deterioration in exercise capacity must be declared and followed by early surgical assessment.2 Aortic valve replacement should always be considered in symptomatic patients, as their mortality rates with medical management are 50% at 3 years and 90% at 10 years.3,4 Survival curves have shown that the interval from onset of symptoms to death is approximately 2 years in patients with heart failure, 3 years in patients with syncope, and 5 years in patients with angina.5,6 Despite the increased incidence of sudden death, the principal cause of death is progressive heart failure. In the absence of a myocardial infarct or atrial fibrillation, the concomitant development of severe heart failure with

Valvular heart disease: timing of surgery

aVR

VI

V4

II

aVL

aVRV2

V5

III

aVF

V3

V6

RHYTHM STRIP: 11 25 mm/sec: 1 cm mV

LOC 00000–0000 00000 0000

40

12011

Figure 74.1 Case 1. A 12-lead ECG showing sinus rhythm with voltage criteria of severe left ventricular hypertrophy and a left ventricular strain pattern

a fall in the Doppler peak instantaneous gradient (from 70 mmHg to 40 mmHg) indicates critical aortic stenosis in this patient. Although left ventricular function is poor, valve replacement surgery is the treatment of choice, the alternative of balloon valvoplasty being only a temporary remedy.7 Poor preoperative left ventricular function should never be a contraindication to valve replacement surgery, although those in congestive cardiac failure face an increased perioperative risk,8 as do those with coronary artery disease.9 However, the majority of surviving patients will experience functional improvement and a reduction in their NYHA classification.9 Mitral insufficiency secondary to dilation of the left ventricle is common in patients with “end-stage” aortic stenosis. Following successful valve replacement, the degree of mitral regurgitation can be expected to improve and mitral valve surgery is rarely necessary unless the mitral valve is structurally abnormal or the mitral regurgitation is severe. When 2D echocardiography shows a heavily calcified aortic valve with restricted opening and impaired left ventricular function, peak instantaneous gradients of less than 50 mmHg should be regarded as indicating significant stenosis until proved otherwise. Applying the continuity equation to measure the aortic valve area is recommended,10,11 and

cardiac catheterization need only be performed when coronary arteriography is necessary12 and in those few patients in whom doubt remains despite careful echocardiographic assessment. References 1.Selzer A. Changing aspect of the natural history of valvular aortic stenosis. N Engl J Med 1987;317:91–8. 2.Otto CM, Burwash IG, Legget ME et al. Prospective study of asymptomatic valvular aortic stenosis: clinical, echocardiographic and exercise predictors of outcome. Circulation 1997;95:2262–70. 3.Frank S, Johnson A, Ross J. Natural history of valvular aortic stenosis. Br Heart J 1973;35:41–6. 4.Rapaport E. Natural history of aortic and mitral valve disease. Am J Cardiol 1975;35:221–7. 5.Ross J, Braunwald E. Aortic stenosis. Circulation 1968; 38(Suppl 5):V61–V67. 6.Olesen KH, Warburg E. Isolated aortic stenosis – the late prognosis. Acta Med Scand 1958;160:437–46. 7.Bernard Y, Etievent J, Mourand JL et al. Long-term results of percutaneous aortic valvuloplasty compared with aortic valve replacement in patients more than 75 years old. J Am Coll Cardiol 1992;20:796–801.

935


8.Obadia JF, Eker A, Rescigno G et al. Valvular replacement for aortic stenosis in NYHA class III and IV. Early and long term results. J Cardiovasc Surg 1995;36:251–6. 9.Connoly HM, Oh JK, Orszniak TA et al. Aortic valve replacement for aortic stenosis with severe left ventricular dysfunction: prognostic indicators. Circulation 1997;95:2395–400. 10.Richards KL, Cannon RS, Miller JF, Crawford MH. Calculation of aortic valve area by Doppler echocardiography: a direct application of the continuity equation. Circulation 1986;73:964–9.

Case 2

A 32 year old man with no previous medical history presents with severe exertional breathlessness and some orthopnea. Examination reveals a collapsing pulse, a blood pressure in the right arm of 120/45 mmHg, a volume overloaded and laterally displaced apex, an early diastolic murmur and a third heart sound. Femoral pulses and distal lower limb pulses are barely palpable. Transthoracic echocardiography demonstrates a dilated left ventricle (LVEDD 7·9 cm, LVESD 5·9 cm) with severe global impairment of systolic function and an ejection fraction estimated at less than 0·2. The aortic valve is lightly calcified, bicuspid, and there is a broad jet of severe aortic reflux. Doppler interrogation of the descending aorta confirms coarctation with an estimated gradient of 30 mmHg. Magnetic resonance imaging (Figure 74.2) confirms a normal ascending aorta, coarctation in the upper descending aorta distal to the left subclavian artery, and some enlargement of the left internal mammary artery.

Question What are the pharmacological and surgical management options for this man?

Comment Here is a therapeutic challenge where one must strengthen what little external evidence exists with a combination of clinical judgment, experience, and a sound knowledge of cardiovascular pathophysiology. There are two issues to address: the aortic valve disease and the coarctation of the aorta. A bicuspid aortic valve is commonly associated with coarctation of the aorta. When the valvular disease is significant, aortic stenosis is more common than aortic insufficiency, although a combination may occur. Coarctation results in a high vascular resistance and, when present, the combination of coarctation and dominant aortic regurgitation results in both a large volume and pressure load on the left ventricle. The insidious onset of severe aortic insufficiency may be well tolerated for many years. In asymptomatic patients with isolated aortic insufficiency, vasodilation using nifedipine has been shown to lengthen the period before valve replacement is necessary.1 In a patient with coarctation of the aorta, the elevated fixed afterload is unlikely to respond to vasodilator treatment and distal perfusion could 936

11.Zoghbi WA, Farmer KL, Soto JG, Nelson JG, Quinones MA. Accurate noninvasive quantification of stenotic aortic valve area by Doppler echocardiography. Circulation 1986;73:452–9. 12.Hall RJC, Kadushi OA, Evemy K. Need for cardiac catheterization in assessment of patients for valve surgery. Br Heart J 1983;49:268–75.

be compromised. Treatment with nifedipine is also best avoided in patients with impaired left ventricular function. When patients with aortic insufficiency do develop symptoms this is usually a reflection of left ventricular dysfunction and valve replacement is advised.2 When left ventricular dysfunction is mild and prompt surgery is performed, the benefits are maximal. However, if surgery is delayed until symptoms or left ventricular dysfunction are established, the prognostic and symptomatic benefits of surgery can be limited.2 Therefore, evidence of significant left ventricular dilation (end-systolic dimension 5·5 cm)3,4 or a reduction in the resting left ventricular ejection fraction5 is usually considered sufficient reason to recommend valve replacement, even in the absence of symptoms. In this patient, recovery of left ventricular function following valve surgery is likely to be limited if the coarctation is significantly obstructive. Doppler assessment of the severity of the coarctation is complicated by the valvar and myocardial dysfunction, but a gradient of 30 mmHg suggests significant but not critical obstruction. In adults, severe aortic coarctation is usually accompanied by increased collateral flow through enlarged branches of subclavian arteries. The presence of an enlarged internal mammary artery in this patient also suggests that the coarctation is likely to be hemodynamically significant. The risk of paraplegia during surgical repair of aortic coarctation is low, but this is enhanced when clamping of the left subclavian artery is necessary. As the coarctation does not involve the left subclavian artery in this

Valvular heart disease: timing of surgery

If expertise is available, balloon dilation of the coarctation is an alternative, but in the absence of this expertise initial surgical repair of the coarctation is probably the initial management of choice,6 although no well conducted comparative studies are available. Reducing afterload in this way, together with the introduction of an ACE inhibitor, is likely to reduce the degree of aortic regurgitation and improve left ventricular function. Subsequent aortic valve replacement could then be performed at a reduced risk. If, after successful coarctation surgery, the left ventricle remains severely compromised, cardiac transplantation could be considered.

References

Figure 74.2 Case 2. Sagittal T1-weighted magnetic resonance image of the descending aorta. There is a concentric narrowing of the upper descending aorta which does not involve the left subclavian artery.

patient, the risk of coarctation surgery is determined mainly by his left ventricular impairment. Combined surgery attempting to replace the aortic valve and repair the coarctation could be considered as a single procedure. In practice, surgery could not be performed easily through the same incision (left thoracotomy for the upper descending aorta and median sternotomy for the aortic valve) and a protracted procedure could have a detrimental effect on the already compromised left ventricle.

1.Scognamigilo R, Rahimitoola SH, Fasoli G, Nistri S, Dalla Volta S. Nifedipine in asymptomatic patients with severe aortic regurgitation and normal left ventricular function. N Engl J Med 1994;331:689–94. 2.Bonow RO, Rosing DR, Kent KM, Epstein SE. Timing of operation for aortic regurgitation. Am J Cardiol 1982;50:325–36. 3.Stone PH, Clark RD, Goldschlager N, Selzer A, Cohn K. Determinants of prognosis of patients with aortic regurgitation who undergo aortic valve replacement. J Am Coll Cardiol 1984;3:1118–26. 4.Henry WL, Bonow RO, Borer JS et al. Observations on the optimum time for operative intervention for aortic regurgitation: 1. Evaluation of the results of aortic valve replacement in symptomatic patients. Circulation 1980;61:471–83. 5.Bonow RO. Radionuclide angiography in the management of asymptomatic aortic regurgitation. Circulation 1991; 84(Suppl I):296–302. 6.Cohen M, Fuster V, Steele PM, Driscoll D, McGoon DC. Coarctation of the aorta. Long-term follow-up and prediction of outcome after surgical correction. Circulation 1989; 80:840–5.

937

Page links created automatically - some may be erroneous, disregard ones formed not from page numbers

Index

For abbreviations used in subentries, please refer to the glossary page xxi v denotes differential diagnosis or comparisons.

adducin, hypertension 155, 156 linolenic acid, cardiovascular disease and 312 tocopherol see vitamin E ADMIT, diabetic patients 140 AASK 153, 154 Abate and Pace Trial (APT) 563–564 ABCA1 gene mutations, coronary atherosclerosis 295 abciximab (7E3) acute coronary syndromes 363, 411–413 adjunctive therapies, trials 435–436 AMI, v coronary stents 448 combination trials 470–471 comparative studies 366–367 comparative trials 364, 433, 460–461, 462, 463, 469–470 with percutaneous coronary interventions 468–469 coronary restenosis prevention 379 efficacy 436 PTCA 364–366, 368 trials 468 unstable angina 363, 411–413 Abciximab Before Direct Angioplasty and Stenting in Myocardial Infarction Regarding Acute and Long-Term Follow-Up (ADMIRAL) study 468–469 abdominojugular reflux test 18 Aboriginal populations 269–271, 273 disease burden 269 geographic variations 270 prevention/treatment approaches 270–271 risk factors 270 temporal trends 270 abscesses, perivalvular 820 acarbose, diabetes mellitus 167 ACAS study 847 ACCORD study 167 ACE inhibitors see angiotensin converting enzyme (ACE) inhibitors acetylsalicylic acid (ASA) see aspirin (acetylsalicylic acid) ACIP (Asymptomatic Cardiac Ischemia Pilot) study 342 ACME (Angioplasty Compared with Medicine) trial 343 ACP Journal Club 3–4, 43 ACST trial 848 ACTC gene, mutations in dilated cardiomyopathy 292 actin ACTC mutations in dilated cardiomyopathy 292 idiopathic dilated cardiomyopathy 684–685 mutations in familial hypertrophic cardiomyopathy 291 ACTION study 332 Action to Control Cardiovascular Risk in Diabetes (ACCORD) study 167

active dissemination, clinical practice changes 82 acupuncture, smoking cessation 117–118, 118 acute coronary syndrome (ACS) 397–425 see also myocardial infarction (MI); unstable angina atypical presentation 401 classification 399 clinical presentation 398 comorbidities 399–400 definitions 397–399 diabetes 401–402, 419 diagnosis 400 ejection fraction 401 elderly 401 etiology 405–406 gender 401 historical perspective 397 incidence 399 inflammation markers 404 long-term management, integrated approach 507–512 management 406–419, 896–901 acute therapy 406–418 chronic therapy 419 invasive therapy 415 natural history 399–401 pathophysiology 404–406 prognosis 399, 400, 401 risk scores 403–404 risk stratification 401–402 Acute Myocardial Infarction-Streptokinase (AMI-SK) study 462, 464 Acute Myocardial Infarction Study of Adenosine (AMISTAD) trial 483 ACUTE pilot study 552–553 adenosine AMI 483 pregnancy 859 adenosine deaminase (ADA), pericardial fluid 741, 742 adenosine diphosphate (ADP) 362 antagonists 411–413 adherence see compliance ADMIRAL study 468–469 adolescents aortic stenosis 783–784 hypertrophic cardiomyopathy 705, 707 pacemaker insertion 592–593, 595 smoking prevention 110, 114 ADOPT-A study 559 adrenaline see epinephrine adrenergic agents, cardiac arrest 637 adrenergic atrial fibrillation 521 advanced glycation end product (AGE) proteins 164 “Adventist” diet 310 Adventist Health Study 317 AFASAK study 548, 549, 550, 551

AFCAPS see Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS) AFFIRM study 533, 534, 556 AFIST study 536–537 African-Americans 272–273, 273 -2 adrenergic receptor 155 disease burden 272 geographic variations 272–273 idiopathic dilated cardiomyopathy 686 prevention/treatment approaches 273 risk factors 272 temporal trends 272 African Blacks 271 African-Caribbeans, -2 adrenergic receptor 155 AF study 529 age blood pressure and 147 coronary artery disease risk and 24 female mortality and 244, 245 fibrinolytic therapy and 437–438 heart failure and 643 LV dysfunction prognosis and 651 pacemaker insertion and 588 peripheral vascular disease and 877 prosthetic valve selection and 814 serum cholesterol risk and 123 of starting smoking 103–104 venous thromboembolism and 864 AIMS (APSAC Intervention Mortality Study) 430 AIRE study 478, 480, 510, 664 cost-effectiveness analysis 665 Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS) 301 inflammation 226 women 140 alcohol (ethanol) consumption 261 blood pressure effects 149, 150 cardioprotective effects 318 cardiovascular disease relationship 318 HDL cholesterol and 127 moderate, definition 318 aldactone idiopathic dilated cardiomyopathy 696 sudden cardiac death prevention 580 aldosterone, in heart failure 661–662 aldosterone receptor blockers, in heart failure 661–662 ALFA study 521 ALIVE study 510, 529, 530, 579, 638 allele heterogeneity 287 allergic reactions, fibrinolytic agents 432 ALLHAT 153, 154 allopurinol, Chagas’ disease 725 alpha-tocopherol see vitamin E Alpha-tocopherol, Beta-carotene Cancer Prevention (ATBC) study 220 alpha-tropomyosin mutations 703

939


Index

alteplase see tissue-type plasminogen activator (tPA, alteplase) American College of Cardiology/American Heart Association (ACC/AHA) guidelines AMI management 437 direct PTCA 452 early CABG 451 antithrombotic therapy guidelines 471–472 heparin guidelines 460 pacemaker implantation 588, 589–593 American College of Chest Physicians (ACCP) anticoagulation in valve replacement 833, 906 antithrombotic therapy guidelines 471–472 American College of Sports Medicine 179 American Heart Association (AHA) infective endocarditis prophylaxis 827 obesity classification 231 postmenopausal hormone therapy 255 rheumatic fever prevention 752 American Indians see Aboriginal populations American Stop Smoking Intervention Study for Cancer Prevention (ASSIST) program 110 AMI see myocardial infarction (MI), acute amiloride 661 aminoglycosides 824 amiodarone 528 atrial fibrillation 525, 534 decision analysis 61, 65 paroxysmal 522, 526, 527, 529, 569 persistent 532 postinfarction 498 post-operative 537, 539 prevention 527 rate/rhythm control 533 cardiac arrest 638–639 Chagas’ disease 728–729 heart failure 671 hypertrophic cardiomyopathy 711, 713 idiopathic dilated cardiomyopathy 697 postinfarction patients 511, 513 postinfarction ventricular premature beats 497 sudden death survivors 581 supraventricular tachycardia 568 ventricular arrhythmias implantable cardioverter defibrillators v 582 non-sustained 579 sustained 578 ventricular tachycardia due to 540 Amiodarone Reduction in Coronary Heart (ARCH) study 536 AMI-SK study 462, 464 AMISTAD 483 amlodipine heart failure 334, 663 hypertension 152, 154 idiopathic dilated cardiomyopathy 695 AMPK gene mutation, Wolff–Parkinson–White syndrome 290 amputations, lower limb 877 analgesics, in AMI 477 aneurysm(s) abdominal aorta 125–126 left ventricular 492–493 ventricular 721 aneurysmectomy, left ventricular 493 angina antioxidants in prevention 219–222 aortic regurgitation 776 aortic stenosis 769 Braunwals class III 367 CABG v PTCA v medical therapy 339–359 clinical diagnosis 16, 24

940

exercise training and 173 postinfarction 330, 333–334, 498–499, 512 post PTCA, calcium antagonists 334 Prinzmetal’s variant 333 stable effort 329, 330, 335 ACE inhibitors 334–335 aspirin 67, 484 blockers 331–332 calcium antagonists 331–332 decision analysis 67 diuretics 335 management case studies 892–895, 900–901 nitrates 332 revascularization recommendations 355 statin therapy 335 therapy choice 892–895 unstable see unstable angina (UA) Angina With Extremely Serious Operative Mortality Evaluation (AWSOME) trial 353 angiography see also radionuclide angiography; venography coronary see coronary angiography coronary restenosis definition 371–372 fibrinolytic therapy studies 469–470 angiopeptin 383 coronary restenosis prevention 381, 383 angioplasty percutaneous transluminal coronary see percutaneous transluminal coronary angioplasty (PTCA) v medical therapy 469 Angiorad Radiation for REStenosis Trial (ARREST) 382, 638 Angiorad Radiation Therapy for In-Stent Restenosis Intra-Coronary Trial (ARTISTIC) 382 angiotensin converting enzyme (ACE), DD genotype 289, 703 angiotensin converting enzyme (ACE) inhibitors see also specific agents acute coronary syndromes 418 aortic regurgitation 776 aortic stenosis 771 asymptomatic LV dysfunction 654 atrial fibrillation 537, 540 Chagas’ disease 727 contraindications 481, 666 coronary restenosis prevention 381 diuretics with 661 efficacy, evidence 34 effort angina 334–335 heart failure 334, 664–666, 671 clinical perspective 665–666 cost-effectiveness 665 documented value 665–666 exercise capacity and 665 hemodynamic effects 665 postinfarction 480–481, 481, 509–510, 513 prevention 646–647, 653, 665–666 survival trials 664–665 hypertension 151–152, 152, 153–154, 304 obesity with 234 hypertrophic cardiomyopathy 712 idiopathic dilated cardiomyopathy 694 myocardial infarction 478, 480–481, 509–510 aspirin interaction 481 cost effectiveness 60–62, 64 nitrate interaction 482 prevention 222–223, 305–306 recommendations 481

myocarditis 690 peripartum cardiomyopathy treatment 686–687 postinfarction left ventricular dysfunction 489–490 postinfarction patients 513 pregnancy 860 stroke prevention 843, 844 sudden cardiac death prevention 580 angiotensin II receptor antagonists atrial fibrillation 537, 540 heart failure 666 idiopathic dilated cardiomyopathy 694 myocarditis 690 in pregnancy 860 angiotensinogen gene (AGT) 154 animal models cardiac arrest management 637 coronary restenosis 372–373 prevention 379 diet and cardiovascular disease fatty acid types 313 methodological issues 309 genetic 296 soy consumption and CHD 317 anistreplase (anisoylated plasminogen streptokinase activator complex, APSAC) 427, 428 comparative trials 432–434 efficacy 429, 430 risks 432 ankle brachial pressure index (ABI) 877, 879 ankle edema 16 Annals of Internal Medicine 44 annuloplasty ring 811 antepartum care 859–860 antiarrhythmic agents see also specific agents arrhythmias due to 569 risk factors 569 atrial fibrillation 522–542, 552 see also atrial fibrillation Chagas’ disease 728–729 class I atrial fibrillation after cardiac surgery 535, 538–540 classification 577 paroxysmal atrial fibrillation 526, 527 persistent atrial fibrillation 530, 531 sudden death prevention 579 ventricular arrhythmias due to 540, 541 class II 577 class III 577 atrial fibrillation after cardiac surgery 538–540 paroxysmal atrial fibrillation 527, 528–529 sudden death prevention 579 ventricular arrhythmias due to 540, 541 ventricular arrhythmias treatment 578, 579 class IV 577 classification 577 heart failure 671–672 documented value 672 hypertrophic cardiomyopathy 710–711, 712–713 with pacing 559 postinfarction ventricular premature beats 496–497 pregnancy 859–860 prophylactic postinfarction 511 research evidence 7–8 supraventricular tachycardia 567–569, 571 tolerability and safety 540–541 ventricular arrhythmias 629


Index

implantable cardioverter defibrillators combined 583 implantable cardioverter defibrillators v 582 non-sustained 578–579 sustained 578 ventricular rate control during therapy 530–532 ventricular tachycardia due to 540–541 Antiarrhythmics Versus Implantable Defibrillators (AVID) Trial see AVID trial antibiotics see antimicrobial agents anticoagulants see also heparin; warfarin acute coronary syndromes 413–416 atrial fibrillation 548, 549–550 cardioversion 552–553, 553 Chagas’ disease 728 contraindications 814 coronary restenosis prevention 378 coronary stent recipients 368 intramuscular injection and 754 MI secondary prevention 508–509 peripheral vascular disease 880 postinfarction left ventricular thrombi 493 pregnancy 860, 861, 872 prosthetic valve recipients 811–812, 814, 832–836 secondary prevention of stroke 840 stroke prevention 845 venous thromboembolism prophylaxis 870 venous thromboembolism therapy 871, 872 antidiabetic agents, weight gain with 235 Antihypertensive and Lipid Lowering Treatment to Prevent Heart Disease Trial, elderly 140 antihypertensive drugs 150–156 see also specific drugs/drug groups choice in obesity and hypertension 234 cost-effectiveness 156–157, 303–304 costs 149 heart failure prevention 653 obesity reduction v in hypertension 233 peripheral vascular disease 880 pregnancy 860 stroke prevention 150–151, 151, 843–844 anti-inflammatory approaches, coronary restenosis prevention 379 anti-ischemic drugs 329–338 see also blockers; calcium antagonists; nitrates ACE inhibitors as 334–335 acute coronary syndromes 406, 407–408 diuretics as 335 safety and efficacy 329–331 statins as 335 antimicrobial agents acute rheumatic fever 754–755 infective endocarditis 822–824 prophylactic aortic stenosis 769 prosthetic valve recipients 827–828 rheumatic heart disease 753–754, 755 streptococcal pharyngitis 752–754, 755 antineoplastic agents, coronary restenosis prevention 381 antioxidants 219–222, 314 cardiovascular disease relationship 314 coronary restenosis prevention 381 epidemiological studies 219–222 flavonoids as 315 folate as 314–315 randomized clinical trials 219–222 vitamins C and E as 314 Antioxidant Supplementation in Atherosclerosis Prevention (ASAP) study 222 antiphospholipid syndrome 871

antiplatelet therapy see also aspirin; ticlopidine acute coronary syndromes 408–410 AMI 468–470 atrial fibrillation 548, 549, 550–558, 553 infrainguinal vascular reconstructions 882 mechanism of action 409 peripheral vascular disease 879–880 postinfarction patients 484, 508–509 prosthetic valve recipients 834–835 secondary prevention of stroke 840 stroke prevention 844–845 Antiplatelet Trialists’ Collaboration 410, 484 antiproliferative agents, coronary restenosis prevention 380–382 Antithrombin-Argatroban in Acute Myocardial Infarction (ARGAMI)-2 study 465–466, 467 antithrombin III (ATIII) 364, 456 deficiency 864 antithrombotic therapy see also thrombin acute coronary syndromes 408–411 adjunctive therapies recommendations 471–472 trials 436–437 AMI 456–476 atrial fibrillation 548–555 cardiogenic shock (postinfarction) 491 coronary restenosis prevention 378 future work 472–473 new agents 378–379 prosthetic valve recipients 832–836 PTCA 362–368 Antithrombotic Therapy in Acute Coronary Syndromes (ATACS) study, in postinfarction patients 508 Antithrombotic Trialists’ Collaboration, in postinfarction patients 508 anti-TNF- antibodies, in myocarditis 693 antitrypanosomal agents 724–726 antituberculous chemotherapy 737, 742, 745–780 antiviral therapy, myocarditis 690 anxiety cardiac rehabilitation and 174 CHD risk and 189, 198–202 syncope 625 aorta abdominal, aneurysms 125–126 coarctation clinical diagnosis 20 pregnancy 854 dissection, clinical diagnosis 20 aortic regurgitation 774–779 aortic valve replacement and 777–779 prognosis 778 recommendations 779 complicating balloon valvuloplasty 785–786, 786 etiology 774 management 776–779 natural history 774 prognosis 775 rheumatic 774 syphilitic 774 aortic root, in Marfan syndrome and pregnancy 855 aortic stenosis 767–781 clinical diagnosis 19, 20 complications 769 congenital 784 critical 768, 783 degenerative calcific (senile) 767, 782 etiology 767–769 exercise testing 773

grading 768–769, 769 management 769–773, 792–793 mild, or moderate 768 natural history/prognosis 767–769, 782 pregnancy 853–854 rheumatic 767, 782 severe 768–769 surgery 783 decision analysis 67 indications 769–773 timing 782 syncope 624 aortic valve area grading 768–769, 769 normal 767 bicuspid 782 bioprosthetic, pregnancy 854 commissurotomy 773, 783 disease 767–784 complications 769 pressure gradient, in stenosis 767 repair aortic regurgitation 777–779 aortic stenosis 772, 773, 774, 783 replacement 811–814 after balloon valvuloplasty 789–790 antithrombotic therapy 833 aortic regurgitation 777–779 aortic stenosis 769–773, 783 balloon valvuloplasty as bridge 771, 791 balloon valvuloplasty v 790 bioprostheses 812–814 decision analysis 61 timing 782 restenosis 787 balloon aortic valvuloplasty 789 mechanism 787–788 valve replacement 789–780 stenosis see aortic stenosis aortic valvuloplasty balloon 782–795 aortic valve surgery after 789–790 aortic valve surgery v 790 as bridge to replacement 771, 791 cardiogenic shock 791 complications 785–786, 786 development 783–784 follow up results 787–788 initial results 785 mechanism 784–785 patients with low output/gradient 791–792 predictors of outcome 788–789 pregnancy 792 prior to non-cardiac surgery 790 recommended use 793 repeat 789 retrograde femoral approach 784 specific indications 790–792 technical aspects 785 surgical 772, 773, 783 aortofemoral bypass surgery 881 apical impulse 17–18, 21 apnea, sleep 610–611 apolipoprotein A1 127 genetic control of levels 294–295 apolipoprotein B 121 APSAC see anistreplase APSIS study 331, 332 APT (Abate and Pace Trial) 563–564 Arabs 267–268 disease burden 267 prevention 268

941


Index

Arabs continued risk factors 267–268 temporal trends 267 arachidonic acid, dietary, cardiovascular disease relationship 312 ARCH study 536 ARGAMI-2 study 465–466, 467 argatroban, in acute coronary syndromes 415 Argentine Randomized Trial of PTCA Versus Coronary Artery Bypass Surgery in Multivessel Disease (ERACI) 347, 348 Argentine Randomized Trial of PTCA Versus Coronary Artery Bypass Surgery in Multivessel Disease (ERACI) II 352 arginine vasopressin (AVP) see vasopressin Aries 43 Arizona, tobacco control interventions 110–112 ARREST trial 382, 638 arrhythmias see also bradyarrhythmias; supraventricular tachycardia; ventricular arrhythmias; individual arrhythmias antepartum management 859–860 antiarrhythmic drug-induced 540–541, 569 diagnosis, pacemakers for 611 exercise training 176 hypertrophic cardiomyopathy 705, 706 included in supraventricular tachycardia 567 ischemic 329, 330, 334 postinfarction 512–513 syncope 622–624, 629 arrhythmogenic right ventricular dysplasia (ARVD) 293–296 gene loci 294 natural history 294 Arterial Disease Multiple Intervention Trial (ADMIT), diabetic patients 140 Arterial Revascularization Therapy Study (ARTS) 352 ARTISTIC trial 382 ARTS 352 ASAP study 222, 529 ascorbic acid see vitamin C Asian-Americans 273, 274 ASIST study 332 ASPECT 558 aspirin (acetylsalicylic acid) ACE inhibitor interaction 481 acute coronary syndromes 408–410, 417, 419 acute rheumatic fever 754–755 adjunctive therapies 438–439 trials 436–437 AMI 468 atrial fibrillation 548, 550, 553 decision analysis 56–58, 61, 65–66 v warfarin 550–551 comparative trials 462 coronary restenosis prevention 378 efficacy, evidence 34 infrainguinal vascular reconstructions 882 ischemic stroke management 842 low-dose, pregnancy 860 mechanism of action 844 pericarditis after MI 495 peripheral vascular disease 879 postinfarction left ventricular thrombi 493 prosthetic valve recipients 833, 835 PTCA 363, 368 secondary prevention of MI 484, 508–509 side-effects 844 stable angina 484 stroke prevention 468, 844, 845, 848 effective dose 844 ASSENT-1 trial 460 ASSENT-2 trial 433, 434–435, 460, 461

942

ASSENT-3 trial 433, 435–436, 438, 439, 447, 460–461, 463, 465, 470–471 adverse effects 462, 463 re-infarction rates 445 ASSENT PLUS study 462, 464 ASSET (Anglo-Scandinavian Study at Early Thrombolysis) 430 ASSIST (American Stop Smoking Intervention Study for Cancer Prevention) program 110 asymmetric septal hypertrophy (ASH) 705–706, 707 Asymptomatic Carotid Atherosclerosis Study (ACAS) 847 Asymptomatic Carotid Surgery Trial (ACST) 848 asystole 634 ATACS study, in postinfarction patients 508 ATBC (Alpha-Tocopherol Beta-Carotene Cancer Prevention) Study 220, 221 atenolol AMI 479 atrial fibrillation 534 effort angina 331–332 vasovagal syndrome 602, 628 atherectomy, coronary see coronary atherectomy AtheroGene Study 226 atherosclerosis acute coronary syndromes 405–406 candidate genes 295 ABCA1 gene 295 CYBA gene 296 carotid artery 846, 847 Chlamydia pneumoniae 227 exercise training and 175 homocysteinemia relationship 314–315 oxidative stress hypothesis 219 polygenic inheritance 294 Atherosclerosis Risk in Communities (ARIC) study, homocysteinemia 225 atherosclerotic plaque coronary angioplasty mechanism of action 373 rupture, during PTCA 360 unstable angina 456 athletes heart 707–708 heart rate 596 ATLAS trial 664–665 atorvastatin 123–124, 127–128, 131 combined therapy 138 cost effectiveness 141 efficacy 132, 133 pleiotropic effects 132 postinfarction patients 511 toxicity 133 Atorvastatin Versus Revascularization Treatment (AVERT) pleiotropic effects 132 PTCA v medical therapy 344 ATP(Adult Treatment Panel reports) I and ATP II 130 ATP (Adult Treatment Panel reports) III 130 cholesterol classification 131 multiple risk factors 131 atria see also left atrium enlargement, in atrial fibrillation 521–522 thrombus, in atrial fibrillation 552 atrial fibrillation (AF) 9, 519–547, 567, 570 ACE inhibitor benefits 537, 540 adrenergic tone 521 alternative-site pacing 558 antiarrhythmic therapy 522–542, 552 drug selection 523 guidelines 522

initiation indications 541–542 prophylactic 526–530 strategies 522 summary of drugs 524 tolerability and safety 540–541 antithrombotic therapy 548–555 atrial-based pacing 557–561 with antiarrhythmic drugs 558 see also individual types atrial implantable defibrillator see atrial implantable defibrillator atrial size progression 521–522 candidate genes 521 catheter ablation 569, 570–571 see also atrioventricular (AV) conducting system catheter-based Maze procedure 561 classification 520 clinical diagnosis 20–21 clinical impact 521–522 decision analysis in management 56–58, 61, 65 definition 519–520 dualsite right pacing 558–560 safety 560 electrical cardioversion 536, 537 factors modulating 520–521 familial 521 gene associated 290 hemodynamic consequences 521 hypertrophic cardiomyopathy 705, 706, 707, 712 inhospital v out-of-hospital therapy 541 lone 548 management case studies 921–924 mitral regurgitation 761 mortality 521 natural history 520–521 new onset (recent onset) 519 non-pharmacologic therapies 556–566 benefits 564 classification 557 see also individual therapies non-rheumatic 548 pacing 606–610 algorithms 608 alternative site 609 atrial overdrive 607 atrial rate support 606–607 biatrial 608 dual site 609 post-operative 537 rate-adaptive 607–610 paroxysmal 522, 570–571 acute conversion 522–526 chronicity 526 conversion rates 522, 525 drug therapy 522–524, 568, 569 focal ablation 571 natural history 526–527 outcome of antiarrhythmic drugs 528 prevention 526–530 spontaneous conversion 522 ventricular rate control 530–532 pathophysiology 520–521 permanent 520 drug therapy 524 “upstream” therapy evidence 537, 540 persistent 519 antiarrhythmic drugs 530, 531 DC cardioversion 532 drug therapy 524 recurrence prevention 532 “upstream” therapy evidence 537, 540


Index

post-cardiac surgery 534–537 atrial pacing 537 management options 536 postinfarction 497–498 pregnancy 855–856 rate control 532–534 post-cardiac surgery 535 therapeutic agents 534, 535 v rhythm control 532–534 rheumatic 548 single-site pacing 556–558 efficacy 556 stroke prevention strategy 564 stroke risk 548, 845 thromboembolic complications 522 time course 519 vagally-mediated 520–521 Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) 533, 534, 556 Atrial Fibrillation Investigators 549–550 Atrial Fibrillation Suppression Trial (AFIST) 536–537 atrial flutter 567 catheter ablation 570 paroxysmal, drug therapy 568 atrial implantable defibrillator 560–561 combination therapies 561 atrial natriuretic factor (ANF), plasma 649–650, 652 atrial overdrive, pacing 607 atrial repair, pregnancy 854 atrial septal defects (ASD) complicating balloon mitral valvuloplasty 800, 801 pregnancy 853 Atrial Septal Pacing Efficacy Clinical Trial (ASPECT) 558 atrial tacharrhythmias see also atrial fibrillation antiarrhythmic drug selection 523 blocker therapy 535 rate control therapy 534 atrial tachycardia 567, 570, 611 management 570 atrioventricular (AV) block 594–600, 606, 629 complete (CHB), pacemaker insertion 589, 594–595 congenital complete pacemaker insertion 592, 595 syncope 622–623 first degree, pacemaker insertion 589, 594 pacemaker mode selection 597 postinfarction 498 pacemaker insertion 590, 596 second degree 594–595 Mobitz type H (advanced) 594–595, 622 Mobitz type I (Wenckebach) 594, 622 pacemaker insertion 589, 594–595 syncope 622–623 atrioventricular (AV) conducting system, catheter ablation 561–564 see also catheter ablation efficacy 563 rate control strategies 563–564 rhythm control 561–563 safety 563 substrate ablation 563 trigger ablation 561–563 atrioventricular nodal re-entrant tachycardia (AVNRT) 567, 570 ablation 570 atrioventricular re-entrant tachycardia (AVRT) 567 antidromic, ablation 569

orthodromic, catheter ablation 569 atropine, in AMI 477 ATTEST study 557–558 audit criteria 74–75 application and applicability 75, 76–77 validity 75–18 process-of-care 73–78 audit-feedback studies, clinical practice changes 82–83 Australia–New Zealand Heart Failure Group 650 Australian–New Zealand (ANZ) study 669 autoantibodies idiopathic dilated cardiomyopathy 683–684, 695 myocarditis 683–684 autoimmunity Chagas’ heart disease 723 idiopathic dilated cardiomyopathy 684 myocarditis 682–684 automated external defibrillators (AEDs) 634 autonomic dysfunction Chagas’ heart disease 722–723 syncope 621–622, 628 autosomal dominant disorders 287–288, 292 autosomal recessive disorder 292 AVERT trial pleiotropic effects 132 PTCA v medical therapy 344 AVID trial 497, 513, 581, 629 postinfarction patients 512, 513 AWESOME trial 352 axillobifemoral bypass surgery 881 azathioprine idiopathic dilated cardiomyopathy 695 myocarditis 691, 692, 693 pericarditis 736 azimilide paroxysmal atrial fibrillation prevention 527, 528, 529, 530 ventricular arrhythmias, non-sustained 579 AzimiLide post Infarct survival Evaluation Trial (ALIVE) 510, 529, 530, 579, 638 -2 adrenergic receptor antagonists see blockers genetic polymorphisms 155 hypertension 154–155 blockers see also specific agents acute coronary syndromes 332–333, 407–408, 418, 419 AMI 334, 479–480 atrial fibrillation 534 paroxysmal 526 paroxysmal, prevention 530 post-cardiac surgery 535 Chagas’ disease treatment 727 combination with calcium antagonists 332 contraindications 331, 479, 509 cost-effectiveness 54, 304 efficacy, evidence 34 effort angina 331–332 heart failure 667–668, 668–671 documented value 670 drug titration/intolerance 670–671 hemodynamic effects 669 neuroendocrine effects 669 prevention 653 quality of life effects 669 survival effects 669–670 heart failure due to 541 hypertension 150–153, 154, 304, 653 obesity with 234 hypertrophic cardiomyopathy 711, 712

in pregnancy 857 idiopathic dilated cardiomyopathy 696 MI secondary prevention 305, 509 myocardial ischemia prevention 654 myocarditis 690 obesity and hypertension 234 peripartum cardiomyopathy treatment 687 peripheral vascular disease 879 postinfarction angina 334 postinfarction patients 513 postinfarction ventricular premature beats 497 pregnancy 859, 860 safety concerns 330–331 sudden cardiac death prevention 579 supraventricular tachycardia 568 syncope 629 threatened MI 333 unstable angina 332–333, 408 ventricular arrhythmias implantable cardioverter defibrillators v 582 non-sustained 579 ventricular fibrillation 496 -carotene see beta-carotene -Carotene and Retinol Efficacy Trial (CARET) 221 -myosin heavy chain mutations 703, 709 BAATAF study 548, 549 bacteria, causing myocarditis 682 balloon angioplasty, percutaneous see percutaneous transluminal coronary angioplasty (PTCA) balloon flotation catheters 488, 489 balloon valvuloplasty see aortic valvuloplasty; mitral valvuloplasty Balloon versus Optimal Atherectomy Trial (BOAT) 353, 376 BARI (Bypass Angioplasty Revascularization Investigation) study 167, 346, 347, 348–350, 349–350 patient profiles 348 Bartonella spp endocarditis 818, 819 batimastat, coronary restenosis prevention 382 Batista operation 727–728 Bayes’ theorem 26–27 beclafibrate 136 bedrest acute coronary syndromes 407 acute rheumatic fever 754 AMI 478 behavior, type A see type A behavior behavioral therapy, obesity 238 BENESTENT study 376 benzathine penicillin G, intramuscular 752, 754, 756 benznidazole 724, 725 BEST ( blocker Evaluation Survival trial) 670, 697 postinfarction patients 517 Beta Blocker in Heart Attack Trial (BHAT) 305, 480 beta-carotene 219 epidemiological studies 221–222 bezafibrate 136 dosage 137 Bezafibrate Infarction Prevention (BIP) trial, in postinfarction patients 511 BHAT (Beta Blocker in Heart Attack Trial) 305, 480 bias see also randomized controlled clinical trials (RCTs) regression dilution 122–123 bifascicular block pacemaker insertion 589, 596 syncope 622–623

943


Index

bile acid sequestering agents (resins) 135–136 adverse reactions 136 clinical use 136 dosage 135–136 mechanism of action 135 results 136 Biochemical Markers in Acute Coronary Syndromes (BIOMACS)-II study 462 BIOMACS-II study 462, 464 bioprosthetic valves 811–812, 812–814 antithrombotic therapy 832, 833 balloon dilation 802–803 factors in selection 811–812, 814 homograft 813–814 mechanical valves v 814 stentless 813 biopsy endomyocardial, idiopathic dilated cardiomyopathy 695 myocarditis 688, 692 primary acute pericardial disease 737 tuberculous pericarditis 741 BIP trial, in postinfarction patients 511 birth size, midlife cardiovascular disease and 96, 279, 280 birthweight CHD association 279 hypertension and type 2 diabetes link 281–282 income in adult life and CHD link 283 bisoprolol heart failure 669 idiopathic dilated cardiomyopathy 696 paroxysmal atrial fibrillation prevention 530 bivalirudin (Hirulog) acute coronary syndromes 415, 417 adjunctive therapies, trials 436–437 comparative trials 464–465 coronary restenosis prevention 378 PTCA 368 bleeding/hemorrhage fibrinolytic therapy associated 431–432, 447 gastrointestinal see gastrointestinal hemorrhage post-PTCA 364 prosthetic valve recipients 812, 813, 832–836 warfarin-treated atrial fibrillation 549, 551–552 blood cultures, in infective endocarditis 819 blood pressure (BP) 146–160 see also hypertension age-related changes 147 calcium supplements effect 316 cardiac rehabilitation and 174 classification 146 clinical v prevention norms 97 disturbances of control 621–622 measurement accuracy 19, 20 factors affecting accuracy 20–21 obesity relationship 232 optimal level of treated 157 pregnancy 853, 856 reduction, stroke prevention 843–844 regulation 155 risk continuum 97–98 trends in developing countries 95 BOAT study 352, 376 body mass index (BMI) children hypertension and type 2 diabetes link 281–282, 283 later CHD link 280–281, 281 obesity classification 231, 232

944

Asians and Caucasians 232 obesity definition 231 risk status assessment 231 body size, children, later CHD link 280–281 bone marrow depression, ticlopidine induced 364 brachial artery approach, balloon aortic valvuloplasty 785 brachytherapy coronary restenosis prevention 382 intracoronary 352–353 bradyarrhythmias cardiac pacemakers 587–618, 629 case studies 931–933 sleep apnea 610–611 syncope 621–624, 629 bradycardias 594 pacing 560 post-cardiac transplantation 610 bradycardia-tachycardia syndrome 596, 622 Braunwals class III angina 367 BRIE trial 382 British Medical Journal 44 British Pacing and Electrophysiology Group (BPEG) recommendations 588 British Regional Heart Study 114 homocysteinemia 225 British Union Provident Association Study, homocysteinemia 225 Brugada syndrome 290 bucindolol heart failure 670 idiopathic dilated cardiomyopathy 697 postinfarction patients 517 bumetanide 660 bundle branch block fibrinolytic therapy and 430, 438 left (LBBB), in idiopathic dilated cardiomyopathy 686, 688 MI 596 pacemaker insertion 596 postinfarction 498 syncope 622–623 bupropion, smoking cessation 117 Bypass and Angiography Revascularization Investigation (BARI II) trial 167 see also BARI (Bypass Angioplasty Revascularization Investigation) study CABG see coronary artery bypass grafting CABG-Patch trial 582 CABRI study 347 patient profiles 348 CADILLAC trial 445, 469 CAFA study 548, 549 cafedrine, vasovagal syndrome 628 calcium antagonists (channel blockers) see also specific agents AMI 331, 478, 482 angina after PTCA 334 atrial fibrillation 534 combination with blockers 332 effort angina 331–332 heart failure 663, 664 prevention 653 hypertension 153–154, 304 obesity with 234 hypertrophic cardiomyopathy 712 idiopathic dilated cardiomyopathy 695–696 myocarditis 691 paroxysmal atrial fibrillation 526 postinfarction angina 333–334 postinfarction patients 509 Prinzmetal’s variant angina 333

safety concerns 330–331 threatened MI 333 unstable angina 331, 332–333, 408 calcium-sensitizing drugs 668 calcium supplements, effect on blood pressure 316 California, tobacco control interventions 110–112 Cambridge Heart Antioxidant Study (CHAOS) 220, 309–310 CAMIAT study 497, 511, 512–513, 579, 671 postinfarction patients 512 Canada, appropriateness of service use 76, 77–78 Canada Institute for Scientific and Technical Information 41 Canadian Amiodarone Myocardial Infarction Arrhythmia Trial (CAMIAT) see CAMIAT study Canadian Cooperative Study 844 Canadian Implantable Defibrillator Study (CIDS) 497, 581, 583 Canadian Multicenter Trial 409–410 postinfarction patients 508 Canadian Trial of Atrial Fibrillation (CTAF) 527 Canadian Trial of Physiologic Pacing (CTOPP) 596, 598, 599 cancer antioxidants in prevention 221–222 calcium antagonists and 331 endometrial, postmenopausal hormone therapy 255 low serum cholesterol and 126 tobacco use and 103 candesartan, in heart failure 669 “candy wrapper effect,” coronary stents 382 CAPRICORN trial 480, 540, 670 MI secondary prevention 483 myocardial ischemia prevention 654 postinfarction patients 509 CAPRIE trial 411, 897 postinfarction patients 508 captopril AMI 480 aortic regurgitation 776 cost-effectiveness 304, 305–306, 665 decision analysis 61, 64 heart failure 664, 666 myocarditis 690 postinfarction patients 509 Captopril-Digoxin Multicenter Research Group trial 659–660 Captopril Prevention Project (CPP), randomized clinical trials 35 CAPTURE (c7E3 Fab Antiplatelet Therapy in Unstable Refractory Angina) trial 364–366, 379, 404, 412–413 carbohydrates cardiovascular disease relationship 313 glycemic index 313 Carbomedic prosthetic valve 812, 833 carcedilol, in MI secondary prevention 483 cardiac actin (ACTC) gene, mutations in dilated cardiomyopathy 292 cardiac arrest drug treatment 636–639 emergency medical services 61, 66 rhythms causing 634 Cardiac Arrest in Seattle: Conventional versus Amiodarone Drug Evaluation (CASCADE) study 578 Cardiac Arrest Study Hamburg 497 cardiac arrhythmias see arrhythmias


Index

Cardiac Arrhythmias Suppression Trial (CAST) see CAST Cardiac Care Network 84 cardiac catheterization aortic stenosis 767 constrictive pericarditis 740 hypertrophic cardiomyopathy 707 mitral regurgitation 761 mitral stenosis 798 cardiac channelopathies, inheritance and mutations causing 290 cardiac failure see heart failure cardiac glycosides see also digitalis/ digoxin heart failure 659–660 cardiac hypertrophy, familial hypertrophic cardiomyopathy 291, 292 Cardiac Insufficiency Bisoprolol Study (CIBIS) see CIBIS Cardiac Insufficiency Bisoprolol Study II (CIBIS II) see CIBIS II trial cardiac output low atrial fibrillation 521 balloon aortic valvuloplasty and 791–792 mitral stenosis 763 cardiac rehabilitation see rehabilitation, cardiac Cardiac Resynchronization for Heart Failure (CARE-HF) trial, cardiac resynchronization therapy 606 cardiac resynchronization therapy (CRT), dilated cardiomyopathy 603–606 cardiac surgery see also other individual conditions atrial fibrillation after 534–537 Chagas’ disease 727–728 endomyocardial fibrosis 729, 758 hypertrophic cardiomyopathy 711–712 pregnancy 858–859 cardiac tamponade see tamponade cardiac transplantation balloon aortic valvuloplasty as bridge 792 Chagas’ disease 727 decision analysis 61, 66–67 exercise training 175–176 heterotropic 610 myocarditis 694 pacemaker insertion 592–593, 610 cardiogenic shock adjunctive intravenous heparin 459–461 aortic stenosis 771 balloon aortic valvuloplasty 791 clinical features and prognosis 491 direct PTCA 445–447 management 491 postinfarction 491 cardiomegaly, clinical diagnosis 17–18 cardiomyopathies 718–732 arrhythmogenic right ventricular see arrhythmogenic right ventricular dysplasia (ARVD) dilated see dilated cardiomyopathy epidemiological transition 92–93 hypertrophic see hypertrophic cardiomyopathy (HCM) peripartum (PPCM) see peripartum cardiomyopathy (PPCM) restrictive 729 v constrictive pericarditis 739 cardiomyoplasty, in Chagas’ disease 727 cardioprotection alcohol 318 dietary potassium 316 cardiopulmonary resuscitation 634–640 decision aid 638, 639

defibrillation see defibrillation drug treatment 636–639 termination 639 cardiothoracic ratio, LV dysfunction prognosis and 652 cardiovascular disease (CVD) 91–102 age of death in developing countries 92 diet association 309–325 see also diet epidemic 284, 321 evolution 92–94 intervention strategies 96–99 mechanisms of acceleration 94–96 epidemiological transition 92–93 ethnic variations 259–260, 260–274 family history 289 genetics see genetics, of cardiovascular disorders global burden 91–92, 259–260 infections 227 inflammation 226–227 intrauterine influences 96 polygenic inheritance 289 pregnancy see pregnancy projections 94 risk factors see risk factors single gene disorders 287–288, 290–293 see also hypertrophic cardiomyopathy, familial; long QT syndrome Cardiovascular Disease Life Expectancy Model, cost effectiveness 304–305 Cardiovascular Health Study, left ventricular (LV) dysfunction 645 cardiovascular history 14 cardiovascular physiology, pregnancy 853 cardiovascular services appropriateness of use 75, 76, 77–78 process-of-care studies 73–78 resources 46, 47 cardioversion anticoagulant therapy 552–553, 553 atrial fibrillation 537 post-operative 536 persistent atrial fibrillation 532 pregnancy 859–860 CardLine 43 CARE-HF, cardiac resynchronization therapy 606 CARET 221 CARE (Cholesterol and Recurrent Events) trial 511 cost analysis 302–303 diabetic patients 140 elderly 139 inflammation 226 pleiotropic effects 132 women 140 cariporide, in AMI 483–484 carotenoids, cardiovascular disease relationship 314 Carotid and Vertebral Artery Transluminal Angioplasty Study (CAVATAS) 847 carotid artery disease, syncope 624 Carotid Artery Stenosis with Asymptomatic Narrowing: Operation Versus Aspirin (CASANOVA) 847 carotid artery surgery 846 carotid endarterectomy asymptomatic disease 847–848 indications 848 secondary prevention of stroke 840 stroke prevention 846–848 symptomatic atherosclerotic disease 846–847 carotid sinus syndrome 621 diagnosis 621

pacing 591, 600–601, 628 Carpentier-Edwards bioprosthetic valves 813 carteolol, in myocarditis 690 carvedilol heart failure 669, 670–671 idiopathic dilated cardiomyopathy 696 Carvedilol Post-Infarct Survival Control in Left Ventricular Dysfunction (CAPRICORN) see CAPRICORN trial Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS) study see COPERNICUS trial CASANOVA trial 847 CASCADE study 578 case management 99 case studies 890–891 CASH study 497, 581 CASS (Coronary Artery Surgery Study) 339–340, 343 CAST (Cardiac Arrhythmia Suppression Trial) 52, 334, 496–497, 527, 569, 578, 671 antiarrhythmic drugs in ventricular arrhythmias 578–579 catheter ablation 571–572 see also atrioventricular (AV) conducting system focal, paroxysmal atrial fibrillation 571 linear 571 risks 572 supraventricular tachycardia 569–570 Catheter Ablation Registry 563 catheterization, cardiac see cardiac catheterization causal associations, methodological issues 309–311 see also diet, cardiovascular disease association CAVATAS trial 847 CAVEAT 352 CAVEAT-I 376 CAVEAT-III 352 C-CAT 352, 376 CD4 + T lymphocytes, in Chagas’ heart disease 724 CD11b blocking antibody, coronary restenosis prevention 379 CD-ROM textbooks 44 central nervous system modulators, in heart failure 671 central venous pressure (CVP), clinical assessment 18, 21 cephalosporins, in streptococcal pharyngitis 753 cerebrovascular disease (CBVD) see also stroke African-Americans 272 Chinese 263 estrogen replacement therapy and 252–254 clinical trials 252–254 observational studies 252 primary prevention 252–253 secondary prevention 253–254 ethnic variations 260, 273–274 Europeans 260–262 global burden 91–92 Hispanics 268–269 Japanese 262 native North Americans 269 syncope 624–625 cerivastatin 126, 131 cesarean section 861 Chagas’ heart disease 718–729 acute phase 719, 720–721, 723–724 chronic phase 719, 720–721, 721, 724–725 clinical classification 722 clinical features 721–722 clinical studies 725, 726 epidemiology 718–719

945


Index

Chagas’ heart disease continued indeterminate phase 719–720, 721 management 724–757 natural history/prognostic factors 719–721 pathophysiology/pathogenesis 722–724 CHAMP study, in postinfarction patients 509 CHAOS (Cambridge Heart Antioxidant Study) 220, 309–310 charges, v costs 48–49 CHD see coronary artery disease chemotherapy, coronary restenosis prevention 381 chest pain see also angina Chagas’ heart disease 721 coronary artery disease risk 24 diagnostic usefulness 15–16, 21 hypertrophic cardiomyopathy 705 infective endocarditis 818 chest radiography constrictive pericarditis 738–739 tuberculous pericarditis 740, 744 CHF-STAT trial 527, 529, 534, 697 childbirth, management in heart disease 861 children aortic stenosis 783–784 body mass index hypertension and type 2 diabetes link 281–282 later CHD link 280–281 growth, CHD association 279–281 hypertrophic cardiomyopathy 705, 707, 711 idiopathic dilated cardiomyopathy 688–689 mother’s ability to take care (in heart disease) 859 pacemaker insertion 592–593, 595 rheumatic fever prevention 752–753 Chinese 263–264 Americans 274 cardiovascular disease epidemic 95, 263 cardiovascular mortality 260 disease burden 263 migrants 264 prevention approaches 264 risk factors 263 Chinese Captopril Study-1 (CCS-1) 480 Chlamydia pneumoniae atherosclerosis 227 cardiovascular disease 227 cholesterol see also high density lipoprotein (HDL) cholesterol; hypercholesterolemia; low density lipoprotein (LDL) cholesterol abdominal aortic aneurysm and 125–126 African-Americans 272 ATP III classification 131 CHD relationship 121–124, 310, 311 continuum of risk 97–98, 122–123 size of effect 123 speed of reversal/consistency 123–124 Chinese 263 clinical v prevention norms 97 contentious issues 126 dietary, plasma cholesterol relationship 311 dietary fat and 124, 311 diseases other than CHD and 124–126 heart failure risk 648 Hispanics 269 impact of therapy 132, 133 Japanese 262, 263 low hemorrhagic stroke risk 124–125 safety 126 lowering therapy see lipid-lowering therapy native North Americans 270 peripheral arterial disease and 125

946

reduction, nut consumption 317 as screening test 127 serum 121 stroke and 124–125 total 121 Lyon Heart Study 312–313 trends in developing countries 95 Cholesterol and Recurrent Events (CARE) trial see CARE cholestyramine, combination therapies 138 chordae tendinae artificial 813 role in LV function 760 spontaneous rupture 759 chronic fatigue syndrome 628 chronic obstructive airways disease (COPD) 16–17 CIBIS 540 idiopathic dilated cardiomyopathy 696 CIBIS II trial 540, 669, 670 postinfarction patients 512 CIDS (Canadian Implantable Defibrillator Study) study 497, 581, 583 cigarettes modification 114 smoking see smoking cilostazol, in peripheral vascular disease 879 ciprofibrate 136 dosage 137 CLASSICS 363–364 clinical assessment 14–22, 23–26 critical appraisal of literature 15 diagnosis 24 prediction of patient outcome 26 screening 23–26 strategies to locate literature 14–15 usefulness 15–22 clinical expertise 8–9 clinical guidelines 81–83 Clinical Outcomes from the Prevention of Postoperative Arrhythmia (COPPA) II study 535, 538 clinical practice 71–88 assessing 71, 73–81 changing 81–85 audits 82–84 incentives/disincentives 81–82 data primary v secondary 72 quality 72–73 outcome studies see outcome studies process-of-care studies 73–78 descriptive 73–74, 78–79 use for policy inferences 77–78 utilization reviews/clinical audits 73–78 process-outcome relationships 71–81 clinical trials, randomized see randomized controlled clinical trials (RCTs) clinofibrate 136 clofibrate 136 efficacy 137 randomized clinical trials 35 clonidine, in heart failure prevention 653 clopidogrel 364 acute coronary syndromes 411, 417, 419 adjunctive therapies 439 coronary stent recipients 363–364 postinfarction patients 508–509 stroke prevention 844–845, 848 Clopidogrel Aspirin Stent International Cooperative Study (CLASSICS) 363–364 Clopidogrel in Unstable angina to prevent Recurrent Events (CURE) trial see CURE trial

Clopigogrel versus Aspirin in Patient at Risk of Ischemic Events (CAPRIE) trial see CAPRIE trial CMV infections, cardiovascular disease 227 coagulation dietary fat and 126 estrogen effects 244–245 mechanism 413 venous thromboembolism and 864 coarctation of aorta see aorta, coarctation Cochrane Controlled Trials Registry (CCTR) 43 Cochrane Database of Systematic Reviews (CDSR) 43 Cochrane Library 43 Cochrane Review Methodology Database (CRMD) 43 cohort studies 72 cholesterol/CHD relationship 121–124, 122 colchicine, coronary restenosis prevention 381 in pericarditis 736 colesevelam 135–136 clinical use 139 combined therapy 139 colestipol 135 combination therapies 138 collagen, in thrombus formation 362, 406 Combination Hemotherapy And Mortality Prevention (CHAMP) study 509 COMET trial 670 COMMIT (Community Intervention Trial for Smoking Cessation) 109–110 Committee on Valvular Heart Disease 768–769 community interventions preventive 98–99 tobacco control 108–112, 116 Community Intervention Trial for Smoking Cessation (COMMIT) 109–110 COMPANION trial, cardiac resynchronization therapy 606 Comparison of Medical Therapy, Pacing and Defibrillation in Chronic Heart Failure (COMPANION) trial 606 “compensatory growth,” growth and CHD development 282 compliance exercise 177 methods of improving 177–178 multiple drug therapies 513 computed tomography (CT) constrictive pericarditis 739–740 pulmonary embolism 868, 869 conducting system see atrioventricular (AV) conducting system conduction disturbances 588, 594 see also atrioventricular (AV) block postinfarction 498, 590, 596 syncope 622–623 confounding outcomes report cards 83 outcome studies 80–81 randomized clinical trials 37 congenital heart disease pregnancy 853–855 risk in offspring of mothers with heart disease 859 congestive heart failure see also heart failure prevention 643–658 congestive heart failure, obesity with 236–237 Congestive Heart Failure Survival Trial of Antiarrhythmic Therapy (CHF-STAT) 527, 529, 534, 697 CONSENSUS I 664 CONSENSUS II 480, 490 Consumer Price Index 52


Index

CONTAK-CD, cardiac resynchronization therapy 606 contrast media, low v high osmolality 60, 63 Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications (CADILLAC) trial 445, 469 COPERNICUS trial 540, 670 postinfarction patients 517 COPPA study 535, 538 coronary angiography acute coronary syndromes 416–418 aortic stenosis 769 appropriateness of use 75, 76 cardiogenic shock (postinfarction) 491 costs 47–48 preoperative, decision analysis 66 coronary angioplasty see also percutaneous transluminal coronary angioplasty (PTCA) cardiogenic shock (postinfarction) 491 mechanisms of action 373 postinfarction angina 498 coronary artery abrupt closure after PTCA 360–370 mechanism 361–362 blood flow, in idiopathic dilated cardiomyopathy 689 dissection, during PTCA 360 elastic recoil in restenosis 373, 374–376 injury to wall, restenosis 375, 376–377, 380 left anterior descending, occlusion 492 postinfarct patency effects of fibrinolytics 429 effects of heparin 457 fibrinolytic therapy v direct PTCA 445–447 remodelling in coronary restenosis 383 restenosis see coronary restenosis spasm, during PTCA 360, 368 coronary artery bypass grafting (CABG) 339–359 acute coronary syndromes 416–418 AMI 451 aortic valve replacement with 769–770 appropriateness of use 75, 76, 77–78 cost-effectiveness 53–54, 54 economic aspects 47, 49–50 exercise training after 171–175 indications 339, 340 medical therapy v 339–343, 355 case studies 892–895 current recommendations 355 database studies 353–354 limitations of randomized trials 342–343, 354–355 single vessel disease 346 mortality 340–342 outcomes measurement 83–84 postinfarction angina 499 postmenopausal hormone therapy 248 preoperative, decision analysis 61, 66 survival 341 Coronary Artery Bypass Graft Patch Trial (CABG-Patch) 582 coronary artery disease (CAD; CHD) see also angina; myocardial infarction aortic stenosis with 770–771 CABG v PTCA v medical therapy 339–359 candidate genes 295 causes, “destructive model” 279 diagnosis clinical 16–17, 24 incremental value of tests 23–33 value of stress tests 23, 26–28, 30 “epidemics” in Western countries 284 epidemiological transition 92 estrogen replacement therapy and 245–251

clinical trials 246–247, 249–251 observational studies 245–246, 248–249 primary prevention 245–248 secondary prevention 248–251 ethnic variations 259–260, 260–274, 273 exercise training 171–175 fetal origins 279–286 see also fetal origins, coronary heart disease genetics 294–296 global burden 91–92 growth in children and 279–281 hazard ratios in relation to birthweight and growth 280 incidence increase 279 left anterior descending (LAD), CABG v medical therapy 341, 346, 348–349, 355 left main, CABG v medical therapy 340, 341 LV dysfunction prognosis and 652 management case studies 912–914 mitral regurgitation 763 mortality decline, polyunsaturated fatty acid consumption 313 multivessel current recommendations 355 PTCA v CABG 3, 346, 348–350 non-invasive screening for severe CHD 24–26, 28–29, 30–31 obesity 236 weight reduction effect 236 observational studies 245 pathogenesis adult living standards and 283 fetal mechanisms 282–283 “fetal origins hypothesis” 96, 284 undernutrition and current-day fetal effects 283–284 peripheral vascular disease and 877 prediction of outcome 26, 29, 31 pregnancy 857 prevention 96–99 antihypertensive therapy 150–152, 150–156 antioxidants role 314 decision analysis 63–64, 65–66 folate 314–315 see also prevention psychosocial factors 181–218 reduced risk, dietary fiber effect 314 risk factors see risk factors serum cholesterol and 97, 121–124 single vessel current recommendations 355 PTCA v CABG 346 three vessel CABG v medical therapy 340, 341 current recommendations 355 tobacco as risk factor 104–106 women 244–245 Coronary Artery Surgery Study (CASS) 339–340 limitations 343 coronary atherectomy glycoprotein Ilb/IIIa receptor blockers 366–367 rotational 352 coronary atherosclerosis see atherosclerosis coronary care units (CCUs), in AMI 61, 66 Coronary Drug Project, intention to treat analysis 35 coronary heart disease (CHD) see coronary artery disease (CAD; CHD) Coronary Heart Disease (CHD) Policy Model, cost effectiveness 301–302, 305

coronary perfusion pressure, vasopressin in cardiac arrest and 637 coronary restenosis 361, 371–394 animal studies 372–373 characteristics 372 clinical trials design 373 definitions (angiographic) 371–372 extracellular factors/cytokines involved 377 future prospects 384 healing mechanisms after angioplasty causing 373 incidence 371 mechanism 361–362 methods of studying 372–373 new etiologies 383 pathobiologic events 373–383 phases 373–383 phase I (elastic recoil) 373, 374–376 phase II (thrombus formation) 374, 376–379 phase III (neointimal proliferation) 374, 379–383 sequence of events 375 prediction 372 prevention after PTCA 361, 363 molecular approaches 383 peripheral vascular disease 879 of phase I restenosis 375–376 of phase II restenosis by anti-inflammatories 379 of phase II restenosis by antithrombotic drugs 378–379 of phase III 380–383 postinfarction patients 484 probucol 361 prosthetic valve recipients 834–835 PTCA complications 368 therapeutic approaches 376 PTCA and 348, 360–361 variables/factors associated 372 coronary revascularization see myocardial revascularization coronary stents 355 acute coronary syndromes 416–418 AMI 449 anticoagulation regimens 368 antithrombotic therapy 368 CABG v 351 coatings/covering strategies 381–382 foreign body reaction to 377 indication 350 local drug delivery 351–352 phosphorylcholine-coated 381 polymer-coated 381 PTCA v 351 radioactive 382 rapamycin-coated 382 restenosis prevention 376 coronary thrombus coronary restenosis prediction and 372 formation in coronary restenosis 376–378 prevention 378 coronary vascular disease (CVD), prevention 96–99 Corpus Christi Heart Project (CCHP) 268 corridor procedure 551 corticosteroids acute rheumatic fever 754–755 idiopathic dilated cardiomyopathy 695 myocarditis 691–694, 692–693 pericarditis 736 tuberculous pericarditis 742–744, 745, 746 cost-benefit analysis 51

947


Index

cost-effectiveness (CE) 300–308 see also specific drugs/therapies analysis 51–55, 300 decision analysis 58–60, 62 measuring effectiveness 52–53 new technologies 62–63 specific clinical products 63–66 treatment strategies 66–68 benchmarks 53, 300 calculation 53 diagnostic tests 29–31, 54–55 patient selection and 53–54 prevention (cardiovascular diseas) 300–308 ratio 53, 62 selected therapies 54 cost-minimization analysis 51, 53 costs 47–51 average 48 cost-effectiveness analysis 52 estimation 49–50 international perspectives 50–51 marginal 48, 49 staff 48 supply 47–48 v charges 48–49 counseling genetic 290 heart disease in pregnancy 857–859 physician, cost-effectiveness 54 smoking cessation 116, 118–119, 303 COX-1 and COX-2 410 Coxiella burnetii, infective endocarditis 823 Coxsackie B3 myocarditis 690, 691, 693 Coxsackie virus infections idiopathic dilated cardiomyopathy 684 myocarditis 681–682 Cox’s linear proportional hazards model 29 C-reactive protein (CRP) 226 acute coronary syndromes 404, 405 creatine kinase, postinfarction 489 critical limb ischemia 880 medical treatment 880 surgical treatment 881–883 Crohn’s disease 692 cross-sectional studies 72 cross-subsidization 49 CTAF trial 527, 529 CTG repeats, expansion, dilated cardiomyopathy 293 CTOPP 556, 598, 599 Cuban-Americans see Hispanics CURE trial 411, 897 acute coronary syndromes 417 postinfarction patients 508 cyanide toxicity 662 cyanotic heart disease, pregnancy 854–855 CYBA gene 295–296 mutations and coronary atherosclerosis 296 cyclin-dependent kinases, smooth muscle cell proliferation 380 cyclo-oxygenase-1 410 cyclo-oxygenase-2 410 cyclophosphamide, in myocarditis 693 cyclosporin myocarditis 691, 692, 693 statin interactions 139 cytochrome P450 system, statin toxicity 133 cytokalasin B 383 cytokines, platelet aggregation releasing 377 cytomegalovirus infections, cardiovascular disease 227 cytoskeletal proteins, mutations in dilated cardiomyopathy 292

948

cytoskeleton, in idiopathic dilated cardiomyopathy 684 -sacroglycan gene, in idiopathic dilated cardiomyopathy 684 DAAF trial 526 dairy products, CHD relationship 318 dalteparin 461–462 acute coronary syndromes 414, 416, 417 comparative trials 464 postinfarction patients 508–509 DANAMI trial 499, 512 Danish Investigations of Arrhythmia and Mortality ON Dofetilide in Congestive Heart Failure (DIAMOND-CHF) 526, 527–528, 529 Danish Study 599 pacemaker mode selection 597 Danish Trial in Acute Myocardial Infarction (DANAMI) 499, 512 DANPACE trial 599, 600 DAPPAF trial 560 DASH diet trial 310, 316, 319–320 dairy products 318 diet types and composition 319, 320 fruit and vegetable consumption 317 low sodium 319–320 data see information Database of Abstracts of Reviews of Effectiveness (DARE) 43 DAVIT-II trial 333, 482, 509 DDAFF study 529 D-dimers, blood levels 865, 867, 868 pulmonary embolism 868 venous thrombosis 867 decision analysis 29–30, 56–70, 71 applications in cardiology 60–68 examples 56–60 modeling 6 new technologies 60, 62–63 specific clinical products 60–61, 63–66 treatment strategies 61–62, 66–68 decision-making, evidence-based medicine 890 decision node 56 decision tree 56, 57, 58–59, 59 evaluation 57–58, 59 folding back 58 deep vein thrombosis see venous thromboembolism defibrillation efficacy data 635 monophasic v biphasic 634–636 public access 634 waveforms 634–635 biphasic 634–635 damped sinusoidal (MDS) 634, 636 guideline 636 Gurvich biphasic 635 monophasic 634 rectilinear biphasic 635 transthoracic biphasic 635 truncated exponential biphasic 635 truncated exponential monophasic 634, 636 Defibrillator in Acute Myocardial Infarction Trial (DINAMIT) 583 defibrillators automated external (AEDs) 634 implantable cardioverter see implantable cardioverter defibrillators (ICDs) delivery (childbirth), management 861 demographic transition 92 dental infections, cardiovascular disease 227 dental procedures 827

depression after stroke 839 cardiac rehabilitation and 174 CHD risk and 189, 190–197 postinfarction 189, 499 treatment to prevent CHD 212 desmin, mutations in dilated cardiomyopathy 293 “destructive model,” coronary artery disease aetiology 279 developed countries cardiovascular disease burden 91–92, 284 projected cardiovascular mortality 94 projected tobacco-related mortality 106–107, 108 rheumatic fever 751 developing countries cardiovascular disease burden 91–92 cardiovascular disease epidemic 94–96 early age of deaths 92 epidemiological transition 92–93, 259 need for evidence based medicine 100 projected cardiovascular mortality 94 projected tobacco-related mortality 106–107 rheumatic fever 751 tuberculous pericarditis 735, 740–741, 765 “developmental plasticity,” growth and CHD development 282 dexfenfluramine 239 Diabetes Control and Complications Trial (DCCT) 163, 165, 306 diabetes mellitus 161 acute coronary syndromes 401–402, 419 Arabs 267–268 CABG v PTCA 355 cardiovascular disease risk 163 Chinese 263 classification 161, 162 complications 161–162 cost-effectiveness of intervention 306 glucose levels and 163, 167 coronary artery disease prediction 24–25 epidemiological transition 93 heart failure risk 648 hypertension 154 lipid lowering therapy 140 LV dysfunction prognosis and 652 native North Americans 270 nicotinic acid therapy 134 obesity with 234–235 weight loss effect on glycemic control 235 peripheral arterial disease 879, 880 prevalence 162 revascularization 350 type 1 (IDDM) 161 type 2 (NIDDM) 161 African-Americans 272 birthweight link 281–282, 283 ethnic variations 262 Hispanics 269 obesity and effect of weight loss 235 South Asians 265 Diabetes Mellitus Insulin Glucose Infusions in Myocardial Infarction (DIGAMI) 166 Diabetes Prevention Study, Finnish, weight loss effect in diabetes 235 Diabetes Reduction Assessment with ramipril and rosiglitazone Medications (DREAM) study 167 diabetic nephropathy 161, 163 diabetic neuropathy autonomic 627 peripheral 163 diabetic retinopathy 161, 163


Index

diagnosis, MEDLINE search strategies 42 diagnostic tests 23–33 see also clinical assessment approaches to assessing 26–29 clinical significance 29–31 cost-effectiveness 29–31, 54–55 diagnosis 26–28 familial hypertrophic cardiomyopathy 296–297 incremental value 23–33 prognosis 29 screening 28–29 DIAMOND-CHF trial 526, 527–528, 529 DIAMOND MI trial 579 DIAMOND study 540, 579 diastolic function, in hypertrophic cardiomyopathy 704 diet “Adventist” 310 antioxidant rich 219 blood pressure effects 149–150, 150 cardiovascular disease association 309–325 antioxidants 314 calcium and magnesium 316 carbohydrates 313 CHD risk factors affecting 310–311 cholesterol levels 310 dietary fiber 313–314 exposure variables 310 fats 311–313 flavonoids and phytochemicals 315 folate 314–315 food consumption data collection 310 food items and groups 316–318 lag time effect 310 methodology of causal associations 309–311 other variables 310–311 outcome variables 309–310 patterns and composite interventions 318–320 policy implications 321 potassium effect 316 recommendations 320 sodium effect 315–316 study design issues 309 see also fatty acids DASH see DASH diet trial developing countries 95 epidemiological transition 92–93 Europeans 261 high carbohydrate, effect on HDL and LDL cholesterol 313 Japanese 319 low-calorie, weight loss in obesity 237 low sodium 316, 319–320 Mediterranean see Mediterranean diet postinfarction patients 507 “prudent” v “Western” 319 recommendations, fatty acid intake 313 serum cholesterol and 124 unhealthy 311 vegetarian 319 very low-calorie, weight loss in obesity 237 Diet and Reinfarction Trial 317 Dietary Approaches to Stop Hypertension see DASH diet trial dietary behavior, unhealthy 311 dietary fiber 313 cardiovascular disease relationship 313–314 composition 313–314 hypertension and CHD risk reduction 314 DIGAMI 166 digitalis/digoxin

aortic regurgitation 776–777 atrial fibrillation 534 paroxysmal 526 post-cardiac surgery 535, 537 postinfarction 498 heart failure 659–660 acute effects 659 chronic therapy 659–660 documented value 660 hypertrophic cardiomyopathy 712 idiopathic dilated cardiomyopathy 694 postinfarction left ventricular dysfunction 490 pregnancy 859 supraventricular tachycardia 568 Digitalis in Acute Atrial Fibrillation (DAAF) 526 Digitalis Investigation Group (DIG) 38 digoxin see digitalis/digoxin Digoxin Investigators Group (DIG) study 660 dihomogammalinolenic acid (DHGLA), cardiovascular disease and 312 dilated cardiomyopathy 681–702 clinical features 292, 688 familial 685–686 genetics 292–293 mutations 292–293 idiopathic (IDC) 292, 684–690, 694–698 clinical presentation 688 epidemiology/natural history 685–687 myocarditis v 690 pathogenesis 684–685 peripartum 856 prognosis 688–689 treatment 694–698 inheritance 292 pacing 591–592, 603–606 alternative 603–604 conventional 603 multisite 603–604 temporary 603 pathogenesis 291, 293, 684–685 diltiazem acute coronary syndromes 408 AMI 482 atrial fibrillation 534 post-cardiac surgery 535 heart failure 663 postinfarction angina 333–334 postinfarction patients 510 unstable angina 332–333, 408 DINAMIT trial 583 dipyridamole coronary restenosis prevention 378 stroke prevention 844, 845 valve replacement 834 directional atherectomy catheter 352 Disability Adjusted Life Years (DALY) lost to cardiovascular disease 91, 94, 259 projected smoking-related losses 107 rank changes 107 tobacco as cause 108 disabled, physically, exercise training 176–177 discounting 52 disopyramide Chagas’ disease 728 heart failure due to 541 hypertrophic cardiomyopathy 711 paroxysmal atrial fibrillation 526, 528 persistent atrial fibrillation 531 supraventricular tachycardia 568 vasovagal syndrome 628 ventricular arrhythmias, sustained 578 diuretics antianginal effects 335 heart failure 660–662

acute effects 660–662 chronic effects 661 clinical management 661–662 documented value 661 survival effects 661 hypertension 150–151, 152–154, 304 heart failure prevention 150–151, 151, 653 hypertrophic cardiomyopathy 712 potassium sparing 661 pregnancy 860 dobutamine heart failure 667 idiopathic dilated cardiomyopathy 689 preoperative, in aortic stenosis 772, 773 docosahexaenoic acid (DHA), cardiovascular disease relationship 312 doctors see physicians dofetilide atrial fibrillation 525 paroxysmal 522, 525–526, 529 post-operative 537, 540 prevention 527–528, 530 supraventricular tachycardia 568 ventricular arrhythmias, non-sustained 579 ventricular tachycardia due to 540 dopamine, in heart failure 667 Doppler echocardiography acute mitral regurgitation after MIs 494 aortic regurgitation 776–777 aortic stenosis 767, 768–769 hypertrophic cardiomyopathy 707 mitral regurgitation 760–761 ventricular septal rupture after MIs 495 Doppler velocity, familial hypertrophic cardiomyopathy diagnosis 296–297 double-balloon technique, aortic valvuloplasty 785 DREAM study 167 Dressler’s syndrome 496 dronedarone, paroxysmal atrial fibrillation prevention 527 drugs inducing myocarditis 681, 683 inducing syncope 627 see also specific drugs/drug groups drug users, intravenous 824 dual chamber pacing, in heart failure 673 Dual-Site Atrial Pacing for Prevention of Atrial Fibrillation Trial (DAPPAF) 560 dysautonomias Chagas’ heart disease 723 syncope 627, 628 dysbetalipoproteinemia, familial (type III, remnant removal disease) 137 dysglycemia 163, 164, 166–167 dyslipidemia, obesity with 235–236 dyspnea diagnostic usefulness 16–17 hypertrophic cardiomyopathy 705 paroxysmal nocturnal 16, 705 dystrophin gene, mutations in dilated cardiomyopathy 293 Eastern Europe, CVD mortality 261 EAST study 347, 349–350 patient profiles 348 ECG see electrocardiogram echocardiography aortic stenosis 767 cardiac tamponade 737, 738–739 constrictive pericarditis 739–740, 744–745 Doppler see Doppler echocardiography hypertrophic cardiomyopathy 705–706 infective endocarditis 819–821

949


Index

echocardiography continued left ventricular (LV) dysfunction 644–645, 652 mitral stenosis 797–798 pericardial effusion 737 syncope 626 transesophageal see transesophageal echocardiography ECLA Glucose-Insulin-Potassium (GOK) trial 483 eclampsia 856 economics see also costs general concepts 47 health 46–55 international perspectives 50–51 edema, in heart failure 660 efegatran, in acute coronary syndromes 415 effectiveness, measuring 52–53 effective orifice area (EFA), prosthetic valves 812 egg consumption, CHD association 318 eicosapentaenoic acid (EPA), cardiovascular disease relationship 312 Eisenmenger syndrome, pregnancy 855, 860, 861 ejection fraction (EF) acute coronary syndromes 401 aortic valve surgery and 771–773, 773, 777–779 balloon aortic valvuloplasty and 787 exercise training and 175 LV dysfunction prognosis and 652 mitral regurgitation 761–762, 763, 809–810 mitral stenosis 763 population studies 644–646 ECG see electrocardiogram elastic recoil, phase I coronary restenosis 373, 374–376 elastin, reduced in arteries, low birthweight and CHD link 282 elderly acute coronary syndromes 401 aortic stenosis 767, 769, 771, 783 balloon aortic valvuloplasty 784, 786, 788–789 cardiac pacing 596 exercise training 173–174, 176 fibrinolytic therapy 437–438 hypertrophic cardiomyopathy 708–709 lipid lowering therapy 139–140 mitral regurgitation 762 stroke risk in atrial fibrillation 550–552, 553 electrical alternans 737 electrocardiogram (ECG) 12-lead, acute coronary syndromes prognosis 403 ambulatory monitoring 626 arrhythmogenic right ventricular dysplasia 293–294 athletes heart 708 Chagas’ heart disease 719, 720, 721 coronary artery disease 24–25, 26 decision analysis 68 mathematical correction 27 monitoring guidelines 178 QT prolongation antiarrhythmic drugs causing 541 see also long QT syndrome ST segment see ST segment T wave abnormalities see T wave abnormalities electroencephalogram (EEG), in syncope 627 Electrophysiological Study Versus Electrocardiographic Monitoring (ESVEM) study 578

950

electrophysiological testing syncope 627 ventricular arrhythmias 578 ELITE studies 666 idiopathic dilated cardiomyopathy 694 ELITE II study 666 idiopathic dilated cardiomyopathy 694 embolism see pulmonary embolism; systemic embolism; thromboembolism EMERALD study 527, 529, 530 EMERAS collaborative group 431 emergency medical services 61, 66 EMIAT study 497, 510, 512–513, 579 postinfarction patients 512 Emory Angioplasty v Surgery Trial (EAST) see EAST study enalapril as anti-ischemic drug 334 aortic regurgitation 776 cost-effectiveness 54, 665 heart failure 664–665, 666, 669 idiopathic dilated cardiomyopathy 694 myocarditis 690 encainide heart failure 671 research evidence 7 encephalomyocarditis (ECM), murine 690–691, 693 endarterectomy see carotid endarterectomy endocarditis 817 infective see infective endocarditis non-bacterial thrombotic (NBTE) 817 endometrial cancer, postmenopausal hormone therapy 255 endomyocardial biopsy, in idiopathic dilated cardiomyopathy 695 endomyocardial disease 728–729 endomyocardial fibrosis (EMF) 718, 729, 757–8 epidemiology/natural history 729, 758 surgical management 729, 758 symptoms and signs 729, 758 endothelin-1 (ET-1), in coronary restenosis 380 endovascular procedures, in peripheral vascular disease 882–883 endpoints see outcomes Enhanced Suppression of the Platelet IIb/IIIa Receptor with Integrillin Therapy (ESPIRIT) 364 eniporide, in AMI 483–484 enoxaparin 364, 462–463 acute coronary syndromes 414, 417–419 adjunctive therapies 470 trials 435–436 combination trials 470–471 comparative trials 364, 433, 462–463, 463, 464 efficacy 436 postinfarction patients 508 regimen selection 438 unstable angina treatment 367 Enoxaparin and Thrombolysis Reperfusion for Acute Myocardial Infarction Treatment (ExTRACT-TIMI-25) trial 463 enoximone 667 ENRICHD trial 212 enterococcal endocarditis 824 ENTIRE-TIMI 23 436, 464, 470 environmental influences 260 epidemiological transition 92–94, 259 “arrested” 93 definition 92 demographic changes due to 94–95 “early” 93–94 early and late adopters 94

four phase model 92–94 variations 93–94 epidural anesthesia, childbirth 861 epilepsy, syncope 625 EPILOG trial 364–366, 379 epinephrine, cardiac arrest 636–637 vasopressin comparison 637 EPISTENT trial 364, 365, 366 eplerenone, in heart failure 662 epoprostenol, in heart failure 663 eptifibatide (integrelin) 364 acute coronary syndromes 411–413, 417 adjunctive therapies 470 combination trials 469 comparative studies 366–367 PTCA/atherectomy 366 ERACI study 347, 348 ERACI II study 352 ERAFT trial 527 ERA trial 249 ERBAC 352 ERK1 and ERK2, familial hypertrophic cardiomyopathy 292 erythrocyte sedimentation rate (ESR), in infective endocarditis 819 erythromycin rheumatic fever prevention 753, 754, 756 statin interactions 139 E-selectin 133, 377 ESETCID 685, 687, 693 esmolol, in atrial fibrillation 534 post-cardiac surgery 535 ESPIRIT trial 247, 366, 412 postinfarction patients 511 ESPRIM 482 ESSENCE 414 acute coronary syndromes 417 mortality 401 estradiol 251–252 estrogen see postmenopausal hormone therapy Estrogen in the Prevention of Reinfarction Trial (ESPIRIT) see ESPIRIT trial Estrogen Replacement and Atherosclerosis (ERA) trial 249 ESVEM study 578 ethacrynic acid 660 ethanol consumption see alcohol (ethanol) consumption ethnic groups 259–278 see also specific groups cardiovascular disease rates 96 definition 259 diabetes prevalence 162, 262 heart failure 643 hypertension rates 147, 262 idiopathic dilated cardiomyopathy and 686 interpretation of studies in 259 multiple, studies of 273–274 etilefrine, vasovagal syndrome 628 etiofibrate 136 etiology, MEDLINE search strategies 42 Etude en Activité Liberale sur le Fibrillation Auriculaire (ALFA) 521 EUROPA trial ACE inhibitors 581 postinfarction patients 510 Europe 260–262 cardiovascular mortality 91, 93, 260–262 prevention approaches 261–262 risk factors 261 European and Australian Multicenter Evaluative Research on Atrial Fibrillation Dofetilide (EMERALD) study 527, 529, 530 European Atrial Fibrillation Trial (EAFT) 549, 550, 551


Index

European Belgian–Netherlands Stent Trial (BENESTENT) 376 European Carotid Surgery Trial (ECST) 846 European Coronary Surgery Study (ECSS) 339–340 European Myocardial Infarction Amiodarone Trial (EMIAT) see EMIAT study European Pacing in Cardiomyopathy (PIC) study 603 European Recurrence of Atrial Fibrillation Trial (ERAFT) 527 Europeans 260–262 European Society Task Force on Syncope Evaluation 619 European Stroke Prevention Study 2 (ESPS 2) 845 European Study of Epidemiology and Treatment of Cardiac Inflammatory Disease (ESETCID) 685, 687, 693 Evaluation of Losartan in the Elderly Study (ELITE) see ELITE studies Evaluation of Platelets IIb/IIIa Inhibitor for Stenting (EPISTENT) trial 364, 365, 366 Evaluation of the Safety and Cardioprotective Effects of Eniporide (ESCAMI) trial 484 evidence, external 889–891 see also evidence-based medicine evidence based cardiology about diagnosis, finding current 40–45 basic model 4 clinical expertise 8–9 clinical prediction tools 5 clinical state/circumstances 4–6 decision analytic modeling 6 definition 3–13 developing countries, need for 99–100 evolving model 4 example 4–5, 9, 10, 11 general approach 4–12 history 3 limitations 10–12 patient communication 8 patients’ preferences/actions 6–7 randomized controlled trials 5, 7–8 research evidence 7–8 contradictions 8 hierarchy 7 limitations 7 value 7 variations 9 Evidence-Based Cardiovascular Medicine 43, 44 evidence-based medicine application principles 890 case studies and 890–891 limitations 889–890 patient-centred medicine v 889 Evidence-Based Medicine 3–4, 43 exercise 170–180 alternatives to monitored training 179–179 arrhythmias 176 benefit evidence 171, 173 cardiac transplant patients 175–176 CHD 171–175 clinical/physiologic outcomes 171–175 cost-effectiveness 54, 55, 304–305 effect on dietary studies of CHD 311 elderly 173–174, 176 evidence for benefits 170–180 exercise monitoring guidelines 178 heart failure 175 hypertrophic cardiomyopathy 705, 706, 710 intermittent claudation therapy 879 morbidity/mortality reduction 171 myocarditis 688

obesity, dyslipidemia management 236 pericardial effusion 737 physically disabled 176–177 postinfarction 304–305, 499 recommendations for adults 170–171 risk stratification 178, 179 safety issues 171, 177 signal averaged (SAECG) 625–626, 710 syncope 626 tuberculous pericarditis 740–741, 744 weight loss with, effect on blood pressure 233–234 exercise stress testing aortic stenosis 773 incremental value 28–29, 30 mathematical correction 27 extracellular matrix, coronary restenosis 379–380 ExTRACT-TIMI-25 trial 463 ezetimibe 137 combined therapy 138 factor v Leiden 864, 871 factor VII, dietary fat and 126–127 factor IX, thrombus formation 406 factor X inhibition 456, 466–468 thrombus formation 406 false negative results, randomized clinical trials 36 familial dilated cardiomyopathy 685–686 familial hypercholesterolemia, genetics 287 familial hypertrophic cardiomyopathy see hypertrophic cardiomyopathy family history, cardiovascular disease 289 fascicular block pacemaker insertion 589, 595–596 syncope 622–623 FASTER trial 470 fat, body see also lipid(s) abdominal distribution, diabetics 235 fats, dietary cardiovascular disease association 311–313 coagulation and 126 developing countries 95 Mediterranean diet 318, 319 recommendations and policy 320, 321 serum cholesterol effects 124 total consumption and recommendations 313 fatty acids cardiovascular disease relationship 311–313 cis-unsaturated 124 dietary recommendations 313 long chain saturated 124 monounsaturated, cardiovascular disease and 312, 313 omega-3 507 polyunsaturated see polyunsaturated fatty acids (PUFAs) saturated, cardiovascular disease relationship 311–312, 313 trans-fatty acids 124, 312, 313 felodipine, in heart failure 663 femoral artery approach, balloon valvuloplasty 784, 785, 798 clinical assessment 878 femorofemoral bypass surgery 881 femoropopliteal bypass surgery 881 femoropopliteal-crural grafts 881 fenfluramine 239 fenofibrate 136 dosage 137 fetal origins, coronary heart disease 279–286

biologic mechanisms 282–283 birthweight and weight gain link 279–281 current evidence from undernutrition 283–284 growth, hypertension and type 2 diabetes link 281–282 impact of maternal nutrition 284 responses to adult living standards and 283 strength of effects 283 “fetal origins hypothesis” 96, 284 fever, fibrinolytic agent induced 432 fibric acid derivatives 135, 136–137 adverse reactions 137 clinical use 137 dosage 137 mechanism of action 137 results 137 fibrinogen acute coronary syndromes 404 blood levels 456 as marker 226 South Asians 265–266 fibrinolysis, mechanism of action 458 Fibrinolytic and Aggrastat ST-Elevation Resolution (FASTER) trial 470 fibrinolytic (thrombolytic) therapy 75, 426–443 adjunctive heparin 457 adjunctive therapies 432, 438–439 trials 435–437 adverse effects 447 agent selection 438–439 AMI contraindications 444 coronary stents v 448 PTCA v 445–448 time to treatment 448–449 combination trials 469 comparative trials 432–435 contraindications 431–432 current use 438 decision analysis 61, 65 early 430 early studies 426–427 efficacy 429–431 coronary artery patency rates 429 evidence 34 mortality rates 429–431 impact 426 indications/guidelines 452 late 431 pathophysiology 426 postinfarction 490–491 left ventricular thrombi 493 right ventricular infarction/failure 492 postinfarction angina 498 pregnancy and 857 procoagulant state after 456–457 rationale 426 risks 431–432 time to treatment 430–431 venous thromboembolism therapy 871 Fibrinolytic Therapy Trialists’ (FTT) Collaborative Group 430, 438 fibrinopeptide A 457 fibroblast growth factor, basic (b-FGF) 377 FIDL cholesterol see high density lipoprotein (HDL) cholesterol filters, vena cava 871–872 Finland 99, 259, 260 sodium excretion and CHD 315 fish consumption, CHD and 317 fish oils 312, 378 flavonoids, cardiovascular disease relationship 315

951


Index

flecainide 528 atrial fibrillation 525 paroxysmal 522, 527, 569 persistent 531, 532 prevention 527 atrial tachycardia 570 Chagas’ disease 728 heart failure 671 research evidence 7 supraventricular tachycardia 568 Flecainide Multicenter Atrial Fibrillation Study 527 FLORIDA, in postinfarction patients 511 Florida, tobacco control interventions 110–112 flosequinan 663 fluorine-18 imaging, in hypertrophic cardiomyopathy 707 fluvastatin 131 cost effectiveness 141 efficacy 132, 133 postinfarction patients 511 toxicity 133 folate, cardiovascular disease relationship 314–315 folic acid supplementation 224, 226, 314, 315 fondaparinux 870 Fontan operation 855 foods consumption, data collection 310 glycemic index 313 items and groups, CVD risk and 316–318 Forrester classification, myocardial infarction 488 Fragmin and Fast Revascularization during Instability in Coronary Artery Disease (FRISC) II trial see FRISC II trial Fragmin during Instability in Coronary Artery Disease (FRISC) study, in postinfarction patients 508 Fragmin in Acute Myocardial Infarction (FRAMI) study 461–462, 464 Framingham Heart Study, hypertension as risk factor 647 Framingham Study antihypertensive therapy 304 atrial fibrillation 548 blood lipids 64, 97 obesity and congestive heart failure 236–237 peripheral vascular disease 877 syncope 619–620 FRAMI study 461–462, 464 France 259, 261 Friedreich’s ataxia 707 FRISC II trial 416–417, 417, 901 mortality 401 postinfarction patients 508–509, 512 FRISC trial, in postinfarction patients 508 fruit and vegetables 315, 316–317 see also vegetables DASH diet trial 319 fruits 149–150, 219 fungi, causing myocarditis 682 furosemide (frusemide) heart failure 660–661, 661 postinfarction left ventricular dysfunction 489 GABI study 347 patient profiles 348 gallbladder disease, postmenopausal hormone therapy 250–251, 255 gastric bypass, Roux-en-Y, for obesity 240 gastrointestinal hemorrhage aspirin plus oral anticoagulants 834–835 calcium antagonists and 331

952

gastroplasty, banded 240 gemfibrozil 136 cost effectiveness 142 efficacy 137 postinfarction patients 511 toxicity 133, 136 gender differences cardiovascular disease 244–245 Chagas’ heart disease 720–721 coronary artery disease 24 heart failure 643, 648 idiopathic dilated cardiomyopathy 686 LV dysfunction prognosis 651–652 peripheral vascular disease 884 gene(s) 287 candidate CHD and myocardial infarction 295 familial atrial fibrillation 521 susceptibility 289 hypertension 294 therapy 296 coronary restenosis prevention 383 gene-environment interactions, growth and CHD development 282 general practitioners, smoking cessation advice 115, 116 genetic counseling 290 genetic diagnosis cardiovascular disease 287 familial hypertrophic cardiomyopathy 296–297 genetic factors cardiovascular disease 96 hypertrophic cardiomyopathy 703 Marfan syndrome 859 genetics, of cardiovascular disorders 287–299 see also inheritance animal models 296 arrhythmogenic right ventricular dysplasia 293–296 atrial fibrillation 521 clinical trials alternative based on 309 coronary artery disease 294–296 dilated cardiomyopathy 292–293 familial hypertrophic cardiomyopathy 154–155, 287, 291–292 family history and 289 hypertension 294 polygenic inheritance 289, 294 single gene disorders 290–293 inheritance patterns 287–289 mutations causing 287 see also hypertrophic cardiomyopathy, familial; long QT syndrome therapeutic prospects 296 geographic variations see also urban-rural differences African-Americans 272–273 cardiovascular disease burden 91–92, 260 clinical practice 73, 74 native North Americans 270 German Cardiovascular Prevention Study 109 GESICA study 671, 697, 728–729 Giant Cell Myocarditis Treatment trial 692 GISSI (Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico) 429, 430, 495, 497 fish oils effect 312 GISSI-1, subgroup analysis, inappropriate 36 GISSI-2 trial 432–434, 458, 459 GISSI-3 trial 220, 478, 480, 481, 496, 509 GISSI-Prevenzione trial 507, 580 Global Burden of Diseases study 94

Global Registry of Acute Coronary Events (GRACE) 399 Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO)-1 trial see GUSTO-1 trial glomerulonephritis, immune complex 819 glucocorticoids, coronary restenosis prevention 381 glucose see also hyperglycemia abnormalities 161–169 direct toxic effects 164 levels cardiovascular disease risk and 163–164, 166 diabetic complications and 161–162, 163, 167 as modifiable risk factor 165–166 tolerance Hispanics 269 impaired (IGT) 161–162, 167 South Asians 265 glycemic index, foods 313 glyceryl trinitrate see nitroglycerin glycoprotein (GP) IIb/IIIa receptor 362 antibodies, coronary restenosis prevention 379 platelet activation in coronary restenosis 377 thrombus formation 406 glycoprotein (GP) Ilb/IIIa receptor inhibitors acute coronary syndromes 411–413, 417, 419 adjunctive therapies, trials 435–436, 436, 468–470 AMI 447, 449 efficacy, troponin levels 403 fibrinolytic agents and 432 mechanism of action 468 PTCA 364–366 Glycoprotein Receptor Antagonist Patency Evaluation (GRAPE) 468 GRACE 399 gradings, definitions 2, 90, 328, 396, 518, 576, 642, 734, 750, 838, 888 GRAPE 468 growth CHD pathogenesis and 282–283 children CHD association 279–281 hypertension and type 2 diabetes link 281–282 growth factors, in coronary restenosis 377, 380 prevention 382–383 growth hormone, in idiopathic dilated cardiomyopathy 695 GUARd During Ischemia Against Necrosis (GUARDIAN) trial 484 guidelines, clinical 71 guidelines, online 44 Guillain-Barré syndrome 621–622 GUSTO-I trial 432, 433, 434, 459–460, 460, 461, 491, 497 outcome studies 78, 79 re-infarction rates 445 GUSTO-II 400 acute coronary syndromes 403 GUSTO-IIA trial 460 GUSTO-IIb 367–368, 415, 445, 446–447, 460, 464, 467 cost-effectiveness 450–451 mortality 401 GUSTO-III 433, 434 re-infarction rates 445 GUSTO-IV 412–413 mortality 401


Index

GUSTO-V 447, 461, 465, 470–471 adverse effects 462, 463 facilitated PCI 449 re-infarction rates 445 GUSTO-V AMI 433, 435, 436, 438 HACEK group organisms 818, 819, 826 Haemophilus spp. endocarditis 820 HALT-MI trial 483 Hancock porcine valve 813 HART-II study 462–463, 464 HDL-Atherosclerosis Treatment Study (HATS) 222 health care costs 46 developed countries 100–101 developing countries 100 outcomes see outcomes quality see quality of care Health Care Information Service 41 health economics 46–55 Health Professionals Follow-up Study CHD risk and social support 206 diabetes and obesity 234–235 dietary fiber and hypertension inverse risk 314 “prudent” v “Western” diets 319 health services, research 71–73 health transition 92 Heart and Estrogen/Progestin Study (HERS) 247, 249, 250–251 cerebrovascular disease 253–254 postinfarction patients 511 venous thromboembolism 254 heart block see atrioventricular (AV) block heart disease see also cardiovascular disease (CVD) prevention, ACE inhibitors 653 heart failure see also left ventricular (LV) dysfunction ACE inhibitors see angiotensin converting enzyme (ACE) inhibitors angiotensin II receptor antagonists 666 antiarrhythmic drugs causing 541 antiarrhythmic drug therapy 671 aortic regurgitation 776 aortic stenosis 769, 771–18, 773, 783 blockers 668–671 cardiac glycosides (digoxin) 659–660 Chagas’ disease 719, 720, 721, 726–726 congestive see congestive heart failure diastolic 644 diuretics 660–662 epidemiology 643–644 exercise training 175 hypertension and 647 idiopathic dilated cardiomyopathy 685, 686, 688 inotropic drugs 666–668 management 659–680 case studies 915–920 MEDLINE search strategies 41–43 mitral regurgitation 761, 763 myocarditis 685 new cardiovascular events 148 pathophysiology 649–651 postinfarction 488–491 ACE inhibitors 480, 481, 489–490, 509–510 biochemical markers 489 calcium antagonists 333 inotropic agents 490 management 489–491 pathophysiology 488 prognostic markers 488–489

reperfusion therapy 490–491 treatment 512–513 prevention 652–654 ACE inhibitors 646–647 right ventricular, postinfarction 491–492 risk factors 646–648, 653 heart murmurs see murmurs, systolic Heart Outcomes Prevention Evaluation (HOPE) 220, 334–335, 480, 580, 648, 843 cost effectiveness 305–306 heart failure prevention 646–648 hypertension as risk factor 647 left ventricular hypertrophy 647, 654 MI prevention 222 postinfarction patients 510 vitamin E supplements and CHD 314 Heart Protection Study (HPS) 133, 222 elderly 139–140 postinfarction patients 511 women 140 heart rate clinical assessment 20–21 heart failure risk 648 variability (HRV), in hypertrophic cardiomyopathy 710 heart sounds, in hypertrophic cardiomyopathy 705 heart valves see valves Helicobacter pylori, cardiovascular disease 227 HELLP syndrome 856, 857 helminths, causing myocarditis 682 Helsinki, growth in boys and CHD association 279–280, 282 Helsinki Heart Study, fibric acid derivatives 136 HELVETICA trial 367 hemodynamics balloon aortic valvuloplasty and 785, 786 exercise training and 175 hypertrophic cardiomyopathy 704 mitral regurgitation 758–759 hemorrhage see bleeding/hemorrhage heparin acute coronary syndromes 414, 417, 419 adjunctive therapies 439 trials 435–436, 436–437 AMI 457–463 infarct artery patency and 457–458 intravenous 458–459, 458–461 intravenous v control 458–459 intravenous v subcutaneous 459–460 meta-analyses of trials 459 recent work 460–461 subcutaneous (SC) 458 combination trials 470–471 comparative studies 433 comparative trials 462, 464 efficacy 436 fibrinolytic agents and 432 low molecular weight (LMWH) acute coronary syndromes 414 adjunctive therapies 435, 438, 439, 461–463, 470 AMI 447 comparative trials 464 in PTCA 367 mechanism of action 456, 870–871 postinfarction left ventricular thrombi 493 postinfarction patients 508–509 pregnancy 860, 861 PTCA 367, 368 stroke prevention 845 therapeutic guidelines 472 venous thromboembolism prophylaxis 870 venous thromboembolism therapy 870–871

Heparin and Aspirin Reperfusion Therapy (HART)-II study 462–463 hepatic toxicity, statins 133 hepatitis C 682 HERO-1 465, 467 HERO-2 436–437, 461, 465, 467, 471 adverse effects 462, 463 herpes simplex virus infections, cardiovascular disease 227 HERS see Heart and Estrogen/Progestin Study (HERS) Hertfordshire, UK, growth in boys and CHD association 279–280, 282 hibernating myocardium 329 hierarchical statistical modeling, outcome studies 80–81 hierarchy of evidence 7 high density lipoprotein C (HDL-C), genetic control of levels 294–295 high density lipoprotein (HDL) cholesterol 121 alcohol intake effect 318 CHD and 127 gender differences 244–245 high carbohydrate diet effect 313 impact of therapy 132 lipid lowering therapy and 133 high risk groups cholesterol lowering therapy 127–128 prevention strategies using 98 value of identifying 31 high-risk units, pregnancy 860 HINT study 333, 407 hirudin acute coronary syndromes 414–415 AMI 464 coronary restenosis prevention 378, 379 fibrinolytic agents and 432 PTCA 367–68 venous thromboembolism prophylaxis 870 Hirudin for the Improvement of Thrombolysis (HIT)-4 study 464 Hirudin in a European Trial v Heparin in the Prevention of Restenosis after PTCA (HELVETICA) 367 hirulog see bivalirudin Hirulog and Early Reperfusion or Occlusion (HERO)-2 trial see HIRO-2 His bundle ablation see atrioventricular (AV) conducting system, catheter ablation Hispanic Health and Nutrition Examination (HHANES) 268 Hispanics 268–269, 273, 274 disease burden 268 geographic variations 269 prevention/treatment strategies 269 risk factors 269 temporal trends 269 history cardiovascular 14 usefulness 15–17 HIT-4 study 464, 467 HLA associations, idiopathic dilated cardiomyopathy 684–685 HMG-CoA reductase inhibitors (statins) 131–135 acute coronary syndromes 418 adverse reactions 133 as anti-ischemic drugs 335 clinical use 133 combined therapy 138–139 coronary restenosis prevention 381 cost-effectiveness 300–303 primary prevention 141–142, 300–302 secondary prevention 302–303, 513

953


Index

HMG-CoA reductase inhibitors (statins) continued dosage 131–133 future work 142 HDL cholesterol effects 127 mechanism of action 131 pleiotropic effects 132 policies for use 127–128 postinfarction patients 510–511 safety 126 serum cholesterol reduction 133 speed of effect 123–124 stroke prevention 125, 845, 848 homocysteine 223–226 see also hyperhomocysteinemia folate effect 314–315 South Asians 265–266 homocysteinemia, atherosclerosis relationship 314–315 homocystinuria 223–226 HOPE see Heart Outcomes Prevention Evaluation (HOPE) hormone replacement therapy (HRT) see postmenopausal hormone therapy hospital discharge abstracts 73 hostility, CHD risk and 183–189 HPS study see Heart Protection Study (HPS) HSV infections, cardiovascular disease 227 hydralazine aortic regurgitation 776–777 heart failure 662–663, 663 idiopathic dilated cardiomyopathy 694 peripartum cardiomyopathy treatment 687 pregnancy 860 hydrochlorothiazide 304, 660–661 hypercholesterolemia see also cholesterol Arabs 268 familial 127 lipid lowering therapy see lipid-lowering therapy therapeutic goals 133 hypercoagulable states 864 hypereosinophilic syndrome 728–729 hyperglycemia see also diabetes mellitus mechanisms of cardiovascular effects 164–165 non-diabetic range cardiovascular disease risk 163–164 South Asians 265–266 hyperhomocysteinemia 223–226 epidemiological studies 224 pathogenesis of atherosclerosis 223–224 randomized clinical trials 224, 226 venous thromboembolism risk 864 hyperinsulinemia 165 hyperlipidemia familial combined 137 heart failure risk 648 peripheral vascular disease 125, 879 prediction of coronary artery disease 24–25 South Asians 265 stroke risk 124–125 therapy see lipid-lowering therapy hypersensitivity myocarditis 681–682, 683 hypersomnolence 610–611 hypertension 146–60 see also blood pressure African-Americans 272 Arabs 268 birthweight link 281–282 calcium supplements effect 316 cardiovascular risk 147, 149, 222 Chinese 263 classification 146–147 definition 146 development, model involving reduced nephron number 282

954

diabetes mellitus 154 dietary approach 319–320 see also DASH diet trial dietary minerals effect 315–316 disease burden 147, 149 epidemiological transition 93 essential 155 ethnic variations 147, 262 genes and mutations associated 294 genetics 154–155, 294 gestational 856, 860 heart failure and 647 Hispanics 269 hypertrophic cardiomyopathy v 707–708 isolated systolic 146, 150 DASH diet trial 319 LV dysfunction prognosis and 652 management, alcohol consumption cessation 318 obesity association 232–234 pharmacogenetics 155–156 polygenic inheritance 289, 294 pregnancy see pregnancy prevalence 147 prevention 149–150 reduced risk, dietary fiber effect 314 salt-sensitive, genetics 155 sodium intake and 315 transient, in pregnancy 856 treatment 150–156, 157 cost-effectiveness 156–157, 303–304 by exercise and weight loss 233–234 heart failure prevention 653 by obesity reduction 232–234 stroke prevention 843–844 see also antihypertensive drugs unanswered questions 157 West Indians 271 hypertriglyceridemia 126–127, 137 heart failure risk 648 hypertrophic cardiomyopathy (HCM) 703–717 see also left ventricular hypertrophy (LVH) differential diagnosis 707–709 elderly 708–709 epidemiology 704–705 familial 291–292 animal model 296 autosomal dominant inheritance 288 genetic diagnosis 296–297 genetics 287 genotype/phenotype correlations 291–292 molecular genetics 291–292 pathogenesis 292 pathology 291 symptoms 291 genetics 287, 291–292, 703 identification, high risk patient 710 incidence 704 incomplete penetrance in adults 708 investigations 705–707 management high risk patient 710–711 supraventricular arrhythmias 712–713 symptomatic patients 711–713 mouse model 296, 703 natural history 705 obstructive 711–712 pacing 591, 602–603, 712 pathology 703 pathophysiology 704 physical examination 705 pregnancy 857 risk stratification 709–711 by mutation 292

sporadic 291 symptoms 705 syncope 624–625, 629, 705 hyperventilation, syncope 625 hypnosis, smoking cessation 117–118, 118 hypobetalipoproteinemia, heterozygous familial 127 hypoplastic left heart syndrome, decision analysis 67–68 hypotension AMI 477 fibrinolytic agents inducing 432 orthostatic 17, 621–622, 628, 629 pregnancy 854 hypoventilation, obesity with 237 hysteria, syncope 625 ibopamine 667 ibuprofen, pericarditis after MI 495 ibutilide atrial fibrillation 525 paroxysmal 522, 524, 528 post-operative 537, 540 ventricular arrhythmias due to 541 ICDs see implantable cardioverter defibrillators IDENT 152, 153 idiopathic dilated cardiomyopathy 694 iliac arteries bypass surgery, cholesterol reduction 123–124 percutaneous transluminal angioplasty 882–883 stenting 883 iloprost coronary restenosis prevention 379 peripheral vascular disease 880 thromboangitis obliterans 884 imaging techniques constrictive pericarditis 739–740 syncope 626, 627 immune globulin (IgG) idiopathic dilated cardiomyopathy 695 myocarditis 693 immune-mediated damage, in Chagas’ heart disease 723–724 immunoabsorption, in idiopathic dilated cardiomyopathy 695 immunosuppressants idiopathic dilated cardiomyopathy 694–695 myocarditis 691–693 peripartum cardiomyopathy treatment 687 IMPACT 110, 366, 379, 578 IMPACT-AMI study 469 IMPACT-II trial 379 impetigo 751 implantable cardioverter defibrillators (ICDs) antiarrhythmic drugs combined with 583 atrial, in atrial fibrillation 560–561 cost-effectiveness analysis 58–60 costs 583 decision analysis 60, 62–63 efficacy assessment 581 hypertrophic cardiomyopathy 711 for inducible ventricular tachycardia/ fibrillation 582 mechanism of action 580 mortality 580 postinfarction patients 512–513 postinfarction ventricular premature beats 497 prophylactic trials 582–583 sudden cardiac death prevention 582–583 sudden death survivors 581 syncope 629 treatment trials 580–583 ventricular arrhythmias 580–583 ventricular fibrillation 580


Index

implantable loop recorders, syncope 621 imvastatin, cost analysis 302 incremental analysis 58 indamide, stroke prevention 844 India cardiovascular disease epidemic 95 disease burden 264–265 risk factors 265–266 temporal trends 265 urban-rural differences 266, 267 indomethacin, pericarditis after MI 495 infections acute pericarditis 735 cardiovascular disease 227 idiopathic dilated cardiomyopathy 684 myocarditis 681–684 infective endocarditis (IE) 817–831 antimicrobial therapy bacteriostatic agents 822–824 combination 823–824 optimal duration 824 outpatient parenteral (OPAT) 824 clinical features 818 complications 818 culture negative 818, 819 diagnosis 817–822 blood culture methods 819 criteria 821, 822, 823, 824 transesophageal echocardiography 819–890 epidemiology 817–818 indications for surgery 825–826 mitral valve repair 811 native valve 817 pathophysiology 817 prevention 827–828 during labor 861 prosthetic valve 817–818 timing of valve replacement 826–827 inferior vena cava filters 871–872 inflammation cardiovascular disease 226–227 Chagas’ heart disease 723 coronary restenosis 376–378 process during coronary restenosis 377 inflammation markers, acute coronary syndromes 404 inflammatory vascular disease 880 inflation 52 information, sources 40–45, 72 specialized 43 inheritance see also genetics dilated cardiomyopathy 292 family history of cardiovascular disease and 289 hypertrophic cardiomyopathy 291 Mendelian patterns 288 polygenic, of cardiac disease 289, 294 single gene disorders 287–289 Initiatives to Mobilize for the Prevention and Control of Tobacco (IMPACT) program see IMPACT INJECT trial 434 inogatran, in acute coronary syndromes 415 inotropic drugs cardiogenic shock (postinfarction) 491 documented value 668 heart failure 666–668 idiopathic dilated cardiomyopathy 697 postinfarction left ventricular dysfunction 490 right ventricular infarction/failure 492 Inoue technique see mitral valvuloplasty, balloon insulin excess production 245 insufficient production 164–165 resistance syndrome 141

weight gain 235 insurance databases 73 InSync ICD, cardiac resynchronization therapy 606 InSync study, cardiac resynchronization therapy 603 integrative reports 71 integrelin see eptifibatide integrilin, coronary restenosis prevention 379 Integrillin and Low-Dose Thrombolytics in Acute Myocardial Infarction (INTROAMI) study 470 Integrillin and Tenectplase in Acute Myocardial Infarction (INTEGRITI) trial 470 integrins see also glycoprotein (GP) Ilb/IIIa receptor platelet adhesion 362, 406 vascular remodelling in coronary restenosis 383 INTEGRITI trial 470 “intention to treat” analysis 35 intercellular adhesion molecule-1 (ICAM-1) 226, 404 INTERCEPT trial 509 interferon-alpha 693, 695 interferon-gamma, pericardial fluid 741 Inter-Health Program 271 interleukin-1, as marker 226 interleukin-6 acute coronary syndromes 404 as marker 226 interleukin-10, recombinant, coronary restenosis prevention 379 intermittent claudication 878–880 epidemiology 877 medical therapy 878–880 natural history 878–879 pathophysiology 878 surgical treatment 881–883 internal jugular vein 18 International Mexiletine and Placebo Antiarrhythmic Coronary Trial see IMPACT International Stroke Trial (IST-3) 842 Internet 44 interpretability fibrinolytic therapy and 431–432, 434 interventricular septum asymmetric hypertrophy (ASH) 706, 707 myotomy-myectomy 711–712 intra-aortic balloon pumping, in mitral regurgitation 760 intracranial hemorrhage (ICH) direct PTCA v fibrinolytic therapy 446–447 serum cholesterol and 124–125 see also stroke warfarin-treated atrial fibrillation 549–550, 550, 551–552 INTERSALT study 315, 316 interstitial fibrosis, familial hypertrophic cardiomyopathy 291, 292 intervention studies 72 interventricular septum, rupture, postinfarction 494–495 InTIME-II 433, 434, 460–461 intra-aortic balloon pumping, cardiogenic shock (postinfarction) 491 intracerebral hemorrhage 842 intracoronary brachytherapy 352–353 intrauterine influences, midlife cardiovascular disease 96 intravascular ultrasound, vascular remodelling in coronary restenosis 383 Intravenous NPA for Treatment of Infarcting Myocardium Early (InTIME)-II study 433, 434, 460–461

INTROAMI study 470 irbersartan, in atrial fibrillation 537, 540 Isbartan Diabetic Nephropathy Trial (IDENT) 152 ischemia critical see critical limb ischemia definition 329 lower limb 885 see also intermittent claudication ischemic heart disease see coronary artery disease ischemic stroke see stroke ischemic ulcers 878, 880 ISIS-1 479 ISIS-2 429, 430, 459 postinfarction patients 508 subgroup analysis drawbacks 36 ISIS-3 432–434, 458, 459 ISIS-4 478, 480, 481–482, 483, 496, 510 isoflavones 317 isosorbide dinitrate acute coronary syndromes 407 heart failure 663 idiopathic dilated cardiomyopathy 694 ISSUE trial 621 Italy 261 JAMA 44 Japanese 260, 262, 273 Americans 274 cardiovascular mortality 91, 93 disease burden 262 migrants 96, 259, 262 prevention/treatment approaches 262 risk factors 262 Japanese diet 319 Jervell and Lang-Nielsen syndrome 290 job, strain, CHD risk and 189, 203–205, 206 journals keeping up to date 44 online 44 specialized 43 jugular venous pressure “a” wave clinical assessment 18 hypertrophic cardiomyopathy 705 ketorolac, in pericarditis 737 Kawasaki disease 857 Kenyan Luo Migration Study 315 Killip classification 489, 490 Kussmaul’s sign 492 labetalol, in pregnancy 860 labor, management 861 lamin A/C gene mutation in atrial fibrillation 521 mutations in dilated cardiomyopathy 292–293 laminin, in idiopathic dilated cardiomyopathy 684–685 The Lancet 44 lanoteplase (nPA) 429 comparative trials 433, 434, 460 LASAF trial 550 LATE study 431 latrofiban, in acute coronary syndromes 413 “law of diminishing returns” 47 LDL cholesterol see low density lipoprotein (LDL) cholesterol leaflets, smoking cessation 119 learning opportunities 40 left anterior hemiblock idiopathic dilated cardiomyopathy 686 pacing 596 post-MI 596

955


Index

left atrium enlargement, in mitral regurgitation 759 ischemic 330 mitral regurgitation 759, 762–763, 763, 809–810 myocarditis 688 postinfarction, treatment 510 pressure, in mitral regurgitation 759 thrombus, in mitral stenosis 797, 798 left bundle branch block (LBBB) acute coronary syndromes 403 idiopathic dilated cardiomyopathy 686, 688 left posterior hemiblock 596 left-to-right shunts, in pregnancy 853 left ventricle (LV) aneurysm, postinfarction 492–493 dilation 652 aortic regurgitation 776, 778 mechanism 650–651 dimension aortic regurgitation 776 end-diastolic volume 652 enlargement, clinical diagnosis 17–18 heart failure and 652 mitral regurgitation 761–762, 763, 809–810 remodeling 650–651 free wall rupture postinfarction 495 pregnancy 853 thrombi, postinfarction 493 left ventricular (LV) dysfunction see also heart failure ACE inhibitors 222–223, 480, 481, 665 aortic regurgitation 775–776, 776 aortic valve surgery and 771–772, 773, 777–779, 783 asymptomatic, treatment 643–658 ACE inhibitors 654, 717 balloon aortic valvuloplasty and 786–787, 788–789, 791–792 CABG v medical therapy 340, 341 clinical diagnosis 17–18 diastolic 644, 646 echocardiology 644–645 epidemiology 644–646, 704–6 idiopathic dilated cardiomyopathy 688–689, 689 management approach 654–655 pathophysiological abnormalities 649–651, 709–713 postinfarction 488–491 ACE inhibitors 489–490 biochemical markers 489 inotropic agents 490 management 489–491 mortality 489 mortality reduction by ACE inhibitors 489–490 pathophysiology 489 prognostic markers 488–489 reperfusion therapy 490–491 prevalence 645 prognosis 775 prognostic factors 651–652, 713–716, 713–716 risk factors 645 screening 646, 704 left ventricular function see also ejection fraction aortic valve replacement and 770–771, 771–772 balloon aortic valvuloplasty and 786–787, 788–789 mitral regurgitation 761–762 role of mitral valve apparatus 759–760

956

left ventricular hypertrophy (LVH) see also hypertrophic cardiomyopathy aortic valve replacement and 770–771, 777–8 concentric 707 differential diagnosis 707–709 genetics 287 heart failure and 647 hypertensive 707–709 hypertrophic cardiomyopathy 703, 706, 707 mitral regurgitation 758 left ventricular mechanical assist devices (LVADs), in heart failure 672–673 left ventricular outflow tract obstruction, pregnancy 853–854 left ventricular pseudoaneurysm 493 left ventriculectomy, partial 727–728 LeukArrest, in AMI 483 leukocyte adhesion, in AMI 483 levels of evidence, gradings 2, 90, 328, 396, 518, 576, 642, 734, 750, 838, 888 levosimendan 668 lidocaine (lignocaine) AMI 478–479 cardiac arrest 638–639 in pregnancy 859 life expectancy, maternal 859 LIFE study, left ventricular hypertrophy 647 lifestyles adult living standards and CHD development 283 blood pressure lowering modifications 149–150 developing countries 95 epidemiological transition 93 developing countries 94–95 epidemiological transition 92–93 likelihood ratio (LR) 15 LIMIT-2 478, 482–483 linear proportional hazards model 29 lipid lowering therapy see lipid-lowering therapy obesity management 237–238 postinfarction 499 stroke prevention 843 lignocaine see lidocaine (lignocaine) LIMIT-2 478 LIMIT AMI 483 Limitation of Myocardial Injury following Thrombolysis in Acute Myocardial Infarction (LIMIT AMI) 483 linoleic acid, cardiovascular disease and 312 lipid(s) 121–129 see also cholesterol abnormalities in obesity 235–236 blood atherogenic effects 311 cardiac rehabilitation and 174 epidemiological transition 93 estrogen effects 244–245 as screening tests 127 CHD relationship 310 dietary see fats, dietary lipid-lowering agents 130–145 see also individual drugs/drug groups coronary restenosis prevention 381 secondary prevention of stroke 840 stroke prevention 845, 848 lipid-lowering therapy 130–145 combination 137–139 contentious issues 126 cost-effectiveness 300–303 heart failure prevention 653 policies 127–128 postinfarction patients 127–128, 510–511 risk factors 131 specific groups 139–141

speed of effect 123–124 stroke risk reduction 125 women 140 LIPID study 132, 511 Lipoprotein Coronary Atherosclerosis Study (LCAS) 296 lisinopril AMI 480 hypertension, obesity with 234 liver function, poor infant weight gain and CHD link 282–283 locus (gene) heterogeneity 287 Loffler endocarditis 729 logistical regression analysis 28–29, 30 long QT syndrome 290, 623–624 autosomal dominant inheritance 288 drug-induced 624 mutations associated 290 pacing 610 treatment, mexiletine 290 Long-term Intervention with pravasatin in ischemic heart disease (LIPID) study 511 loop-diuretics, in heart failure 660–661, 661 losartan animal model of hypertrophic cardiomyopathy 296 heart failure 666 hypertension 152 idiopathic dilated cardiomyopathy 694 myocarditis 690 Losartan Intervention For Life (LIFE) study, left ventricular hypertrophy 647 lovastatin 131 clinical use 139 combined therapy 138, 139 cost-effectiveness 54, 141, 302–303, 303 efficacy 132, 133 toxicity 133 low birthweight CHD development mechanisms 282 hypertension and type 2 diabetes link 281–282, 283 low-calorie diet, weight loss in obesity 237 low density lipoprotein (LDL) cholesterol 121 classification guidelines 131 gender differences 244–245 high carbohydrate diet effect 313 impact of therapy 132 lipid lowering therapy and 133, 137 Lyon Heart Study 312–313 oxidative modification 219 small dense particles (phenotype B) 141 Lp(a) African-Americans 272 Mansfield Balloon Aortic Valvuloplasty Registry 786, 787, 789 marginal costs 48, 49 Markov (state transition) models 58 South Asians 265–266 Lyon Diet Heart Study 310, 312–313 Lyon Heart Study 507 Mac-1 function 379 monoclonal antibody in coronary restenosis prevention 379 MACE trial 847 MADIT trial 497, 513, 582, 583 MADIT II trial 497, 582, 583, 629 postinfarction patients 513 magnesium AMI 478, 482–483 dietary, cardiovascular disease and 316 intravenous, ventricular fibrillation 496


Index

magnesium sulphate, atrial fibrillation, post-operative 537 magnetic resonance imaging (MRI) constrictive pericarditis 739–740 venous thrombosis 867 major histocompatibility complex (MHC) genes, in dilated cardiomyopathy 684–685 Male Health Professionals’ Study 220 MARCATOR trial 381 Marfan syndrome genetics 859 pregnancy 855 Massachusetts, tobacco control interventions 110–112 MASS study 342, 346, 347 patient profiles 348 MATE trial 416 Mauritius 99 MAVID trial 629 Mayo Asymptomatic Carotid Endarterectomy (MACE) trial 847 Maze procedure, catheter-based 551 MDPIT study 333–334 mechanical prosthetic valves 811–812, 812 antithrombotic therapy 832–834 bioprostheses v 814 factors in selection 811–812, 814 Medical Matrix 40–41, 44 medical subject headings (MeSH) 41 Medicine, Angioplasty, or Surgery Study (MASS), CABG see MASS study Mediterranean diet 310, 318–319 components 318–319 n-3 fatty acids 312–313 MEDLINE 14, 40–43, 44 search strategies 42 Medtronic AT 500 pacemaker, atrial pacing 557–558 Medtronic-Hall prosthetic valve 812 antithrombotic therapy 833 megacolon 719 megaesophagus 719 megakaryocytes 361 “mendelian randomization” 309 Mendelian transmission 287, 288 MERCATOR trial 381 MERIT-HF 669–670 idiopathic dilated cardiomyopathy 696 postinfarction patients 512 meta-analayses see systematic overviews metabolic risk, management case studies 909–911 metabolic syndrome (syndrome X) 141, 235–236 metalloproteinases, in coronary restenosis 380 metformin 166, 235 methionine synthase 224 methotrexate, in myocarditis 693 methylcobalamin (vitamin B12) 223–224, 224, 226 methyldopa, in pregnancy 860 methyl-tetrahydrofolate reductase (MTHFR) 224 metoprolol 667 acute coronary syndromes 407 AMI 480 effort angina 331 heart failure 669, 671 idiopathic dilated cardiomyopathy 696 myocarditis 690 atrial fibrillation 534 post-operative 536 heart failure 669–670, 670 sudden death survivors 581 Metoprolol CR/XL Randomized Intervention Trial in Congestive Heart Failure (MERIT-HF) see MERIT-HF

Metoprolol in Dilated Cardiomyopathy trial 696 mevastatin 131 Mexican-Americans see Hispanics mexiletine long QT syndrome 290 ventricular arrhythmias, non-sustained 578 M-HART study 212, 499 microalbuminuria, heart failure risk 648 microvascular disturbances, in Chagas’ heart disease 723 Middle Eastern Crescent 95 midodrine, vasovagal syndrome 628 migraine, syncope 625 migrant groups 96, 260 Chagas’ heart disease 718 Chinese 264 South Asians 265, 266–267, 267 milk consumption, coronary mortality 318 milrinone 667 mineralocorticoid receptor mutation, hypertension 294 minerals, effect on blood pressure and cardiovascular disease 315–316 Minnesota Health Project 109 MIRACLE ICD, cardiac resynchronization therapy 606 MIRACL trial 418 anti-ischemia drugs 335 cardiac resynchronization therapy 605–606 pleiotropic effects 132 postinfarction patients 511 MITI registry 447, 450 mitochondria, genome and mutations 289 mitochondrial abnormalities 289 mitochondrial disease 707 mitral commissurotomy closed 796 balloon valvuloplasty v 802–803 costs 804 open 764, 796, 807 balloon valvuloplasty v 803, 804 mitral regurgitation 758–763, 809–810 acute 758–759, 761 clinical features and prognosis 494 management 494 postinfarction 494 asymptomatic 762 chronic 761–762 compensated 759 complicating balloon valvuloplasty 801–802 as outcome predictor 801–802 contraindicating balloon valvuloplasty 798 etiology 758 far advanced 763 indications for surgery 760–763, 809–810 ischemic 763 medical therapy 763 pathophysiology 758–760 rheumatic 758 severity assessment 760–761 surgical objectives 758 timing of surgery 758, 809–810 valve repair see mitral valve repair mitral stenosis 763–764 etiology and pathophysiology 763–764 pregnancy 855–856 rheumatic 763 surgery 796 indications 764 timing 764 Wilkins-Weyman score 797 mitral valve calcification 709 disease 758–766

prolapse, pregnancy 855 mitral valve repair 809–810 LV function effects 760 replacement v 810–811 timing 758, 762, 809–810 mitral valve replacement (MVR) 811–814 antithrombotic therapy 833 bioprostheses 812–814 hypertrophic cardiomyopathy 711–712 mitral regurgitation 763, 809–810 v repair 810–811 mitral stenosis 763–764 young women 764 mitral valve surgery apparatus, importance 759–760 disease, indications for surgery 758–766 flail leaflet 761, 762 postinfarction 494 restenosis 799 balloon mitral valvuloplasty 802 severe calcification 798 surgery, timing 758, 809–810 systolic anterior motion (SAM) 704, 707 mitral valvotomy see mitral commissurotomy mitral valvuloplasty 796–808 balloon 764, 796–808 bioprosthetic valves 802 complications 801 contraindications 798 costs 804 cylindrical balloon techniques 799 development 796–797 Inoue technique 796–797, 798–799 v double balloon methods 800 long-term follow-up 799 mechanisms 796–797 mild mitral stenosis 802 mitral restenosis 802 open surgical commissurotomy v 803, 804 pregnancy 802 pre-procedure evaluation 797–798 single v double cylindrical balloons 799 techniques 798–799 transesophageal echocardiography during 801 v closed surgical commissurotomy 802–803, 803 Mode Selection Trial in Sinus Node Dysfunction (MOST) 556, 598–599, 599–600 molecular genetics see genetics, of cardiovascular disorders molecular mimicry hypothesis 682–683 MONICA project 399 WHO 260, 271 monorail double-balloon techniques 801 monounsaturated fatty acids, cardiovascular disease and 312, 313 Montreal Heart Attack Readjustment Trial (M-HART) 212, 499 moricizine 671 morphine, in AMI 477 mortality cost-effectiveness analysis 52 ethnic groups 260 global cardiovascular (CVD) 91–92 MOST 556, 598–599, 599–600 motivation, smoking cessation 116 MOXCON 671 moxonidine, in heart failure 671 M-PATHY 603 M protein, streptococcal 751–752 MRFIT (Multiple Risk Factor Intervention Trial) 97, 98, 121–123, 124–125, 149, 183 homocysteinemia 225

957


Index

Multicenter Automatic Defibrillator Implantation Trial see MADIT trial Multicenter Automatic Defibrillator Implantation Trial II (MADIT-II) see MADIT II trial Multicenter InSync Randomized Clinical Evaluation (MIRACLE), cardiac resynchronization therapy 605–606 Multicenter Study of Pacing Therapy for Hypertrophic Cardiomyopathy (M-PATHY) 603 Multicenter Unsustained Tachycardia Trial (MUSTT) 582 Multiple Risk Factor Intervention Trial (MRFIT) see MRFIT multiple system atrophy 627 Multisite Stimulation in Cardiomyopathy (MUSTIC) trial 605 murmurs, systolic clinical assessment 18–19 hypertrophic cardiomyopathy 705 Mustard procedure, pregnancy after 854 MUSTT (trial) 582, 629 mutations missense, single gene disorders due to 287 point, single gene disorders due to 287 MVP study 372 MYBP-C, mutations in familial hypertrophic cardiomyopathy 291 Mycobacterium tuberculosis culture 740–741 DNA detection 741–742 Mycoplasma-associated pericarditis 735 myectomy, chemical, in hypertrophic cardiomyopathy 711–712 myocardial contractile reserve, in dilated cardiomyopathy 689 myocardial contractility, exercise training and 175–176 myocardial infarction (MI) acute (AMI) 329, 330, 335, 477–487 antithrombotic therapy 456–476 blockers 334, 479–480 calcium antagonists 331, 478, 482 clinical diagnosis 15–16, 16 coronary care unit (CCU) admission 61, 66 general management 477–479 magnesium therapy 478, 482–483 management case studies 902–905 mechanical reperfusion strategies 444–455 nitrates 478, 481–482 other adjunctive treatments 479 pain relief 477 thrombolytic therapy see fibrinolytic (thrombolytic) therapy CABG 342 candidate genes 295 complicating balloon valvuloplasty 785, 786 complications 488–506, 512–513 acute mitral regurgitation 494 angina and myocardial ischemia 498–499 atrial fibrillation 497–498 cardiac thromboembolism 493 cardiogenic shock 491 conduction disturbances 498 Dressler’s syndrome 496 free wall rupture 495 heart block 498 left ventricular aneurysm 492–493 left ventricular dysfunction/failure 488–491 pericardial effusion and tamponade 495–496 pericarditis 495 pseudoaneurysm 493 psychosocial 499

958

right ventricular infarction and failure 491–492 treatment see individual complications ventricular fibrillation 496 ventricular premature beats 496–497 ventricular septal rupture 494–495 ventricular tachycardia (non-sustained) 496–497 ventricular tachycardia (sustained) 496 conduction disturbances 590, 596 decision analysis 68 depression after 189 fish oils effect (GISSI) 312 Forrester classification 488 Killip classification 489, 490 long-term management 507–512 complications 148, 512–513 implantable cardioverter defibrillators 513 integrated approach 513–514 limitations of evidence 513 mortality 488, 489 new cardiovascular events 148 non-0 wave 456 see also unstable angina non-ST-segment elevation (NSTEMI) classification 398–399 see also acute coronary syndrome (ACS) obesity as predictor of mortality 236 postinfarction exercise 171–175 postinfarction management 906–908 pregnancy 857 previous history 24–25 Q wave, pericarditis after 495 renin-angiotensin system and 222–223 right ventricular 491–492 secondary prevention 507–513 ACE inhibitors 222–223, 305–306 antiarrhythmic agents 510 anticoagulants 508–509 antioxidants 221–222 antiplatelet therapy 484, 508–509 blockers 305, 479–480, 509 calcium antagonists 509 cardiac rehabilitation 171–175, 304–305, 508 diet/dietary supplements 507 hormone replacement therapy 511 integrated approach 513–514 lipid lowering agents 127–128, 510–511 management case studies 906–908 nitrates 509 psychosocial interventions 206, 212 PTCA 511 smoking cessation 114–115, 118, 303, 507–508 smoking and 106 ST segment elevation (STEMI) 488, 489 threatened 329, 330, 333 vitamin E effect (CHAOS study) 309–310 Myocardial Infarction Triage and Intervention Investigation registry 450 direct angioplasty v fibrinolytic therapy 447 myocardial ischemia see also angina clinical spectrum 329, 330 diagnosis 16–17 exercise training and 175 hypertrophic cardiomyopathy 704 postinfarction 498–499 prevention in heart failure 653–654 recurrent, coronary restenosis definition 372 syncope 623, 629 Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL) trial see MIRACL trial

myocardial revascularization see also coronary artery bypass grafting (CABG); coronary atherectomy; percutaneous transluminal coronary angioplasty (PTCA) AMI 444–455 indications 339 peripheral vascular disease 881–882 recommendations for stable angina 355 myocarditis 681–698 Chagas’ disease 721–722 clinical presentation 687 Dallas criteria 687 definition 681 drug induced 681, 683 epidemiology/natural history 685–687 giant cell 691–9, 692 hypersensitivity 681, 683 idiopathic dilated cardiomyopathy v 690 immunopathogenesis 681–684 murine models 690–691, 693 peripartum cardiomyopathy 686 prognosis 688 systemic disease associated 681, 683 toxic 683 treatment 690–694, 697–698 viral 681–684 Myocarditis Treatment Trial 684, 685, 691 myocardium hibernating 329, 330, 490 rupture, blockers and 479 myocytes, cardiac, familial hypertrophic cardiomyopathy pathogenesis 292 myosin idiopathic dilated cardiomyopathy 684–685 mutations atrial fibrillation 521 dilated cardiomyopathy 292 hypertrophic cardiomyopathy 291, 296 myosin binding protein-C mutations 703, 709 familial hypertrophic cardiomyopathy 291 myotomy-myectomy, septal (SMM) 711–712 NADPH oxidase, CYBA gene action 295–296 nadroparin, in acute coronary syndromes 414 naloxone 477 NASCET trial 846 NASPE prospective catheter ablation registry 572 National Cholesterol Education Panel (NCEP) 130, 302 National Cholesterol Education Program (NCEP), Adult Treatment Panel, metabolic syndrome 236 National Health and Nutrition Examination Survey (NHANES) 222 National Heart Lung and Blood Institute (NHLBI), Balloon Valvuloplasty Registry 786, 787, 801 National High Blood Pressure Education Program 147, 149 National Institute of Neurologic Disease and Stroke (NINDS) Acute Stroke Studies 842 National Investigators Collaborating on Enoxaparin (NICE) 367 National Library of Medicine (NLM) 40–43 National Registry of Myocardial Infarction-2 479–480 National Registry of Myocardial Infarction-3 248 National Tobacco Control Program (NCTP) 110 native North Americans see Aboriginal populations neointima coronary restenosis 380 animal models 373 hyperplasia inhibition 380–382, 383


Index

hyperplasia, remodelling in coronary restenosis 383 proliferation in coronary restenosis 374 nephrons, reduced number at birth, hypertension and 282 neurohormones asymptomatic LV dysfunction 649–650 blocker effects 669 LV dysfunction prognosis and 652 neurological studies, in syncope 627–628 neurologic disorders, syncope 624–625 New England Journal of Medicine 44 New York Heart Association (NYHA) functional class LV dysfunction prognosis and 652 mitral valve surgery and 764 NICE (National Investigators Collaborating on Enoxaparin) 367 nicorandil acute coronary syndromes 408 AMI 484 as anti-ischemic drug 335 unstable angina 408 nicotine addiction 115 assessment 118, 119 replacement therapy 116, 119 adverse effects 117 cost-effectiveness 303 safety 117, 119 nicotinic acid (niacin) 134–135, 135 adverse reactions 134 clinical use 134–135 combined therapy 138 dosage 134 mechanism of action 134 results 134 nifedipine acute coronary syndromes 407–408 AMI 331, 482 aortic regurgitation 776–777 effort angina 331, 332 heart failure 663 postinfarction angina 333 pregnancy 860 safety concerns 331 threatened MI 333 unstable angina 331, 332–333 nifurtimox 724, 725 NINDS trial 842 nitrates see also isosorbide dinitrate; nitroglycerin; nitroprusside ACE inhibitor interaction 482 acute coronary syndromes 407 AMI 478, 481–482 effort angina 332 heart failure 662–663 peripartum cardiomyopathy treatment 687 postinfarction left ventricular dysfunction 489 postinfarction patients 509 unstable angina 332–333 nitroglycerin acute coronary syndromes 407 AMI 477, 481–482 heart failure 662–663 unstable angina 332–333 nitroprusside AMI 481–482 heart failure 662 nitrous oxide, inhaled 477 non-steroidal anti-inflammatory drugs (NSAIDs) acute rheumatic fever 755 Dressler’s syndrome 496 mechanism of action 410

myocarditis 691 pericarditis 736 after MI 495 non-ST-segment elevation myocardial infarction see acute coronary syndrome (ACS); myocardial infarction (MI) Noonan syndrome 707 norepinephrine (noradrenaline) blocker effects 669 plasma (PNE), in LV dysfunction 649–650, 652 therapeutic use 666–667 North American Indians see Aboriginal populations North American Recurrence of Atrial Fibrillation Trial (RAFT) 527 North American Symptomatic Carotid Endarterectomy Trial (NASCET) 846 North American Vasovagal Pacemaker Study 602–603 North Karelia Project 99, 225 numbers needed to treat (NNT), cardiovascular incidents 148 nurses, smoking cessation advice 117–118, 119, 303 Nurses’ Health Study 220, 245–246 cerebrovascular disease 252 diabetes and obesity 234–235 dietary fiber and CHD inverse risk 314 folate effect on CHD 314–315 saturated fats and CHD 311–312 nut consumption, cardiovascular disease relationship 317 nutrition cardiovascular disease association 311–313 see also diet epidemiological transition 92–93 fetal 96 maternal, impact on fetal development and later CHD 284 OARS study 376 OASIS 400, 414, 416 OASIS-2 415, 416 mortality 401 obesity 231–243 African-Americans 272 Arabs 267 classification in Caucasians and Asians 232 congestive heart failure 236–237 coronary artery disease 236, 313 definition 231 diabetes mellitus with 234–235 dyslipidemia with 235–236 epidemiology 231 heart failure risk 648 Hispanics 269 as hypertension risk factor 232–234 hypertension with 232–234 antihypertensive agent choice 234 hypoventilation syndrome 237 native North Americans 270 sleep apnea 237 South Asians 265 treatment algorithm 233, 238 behavioral 238 orlistat or sibutramine 234, 239 pharmacotherapy 239 surgery 239–241 see also weight, loss in obesity OBIS study 696 observational studies outcomes assessment 78–81

randomized clinical trials v 38 obstructive sleep apnea, obesity precipitating 237 oestrogen see estrogen oEStrogen in the Prevention of Re-Infarction Study (ESPIRIT) see ESPIRIT trial oleic acid, cardiovascular disease and 312 olive oil, Mediterranean diet 318 omega-3 fatty acids see polyunsaturated fatty acids (PUFAs), n-3 Omnicarbon prosthetic valve 812 OPSITE, cardiac resynchronization therapy 605 Optimal Pacing SITE (OPSITE), cardiac resynchronization therapy 605 Oregon, tobacco control interventions 110–112 Oregon Tobacco Prevention and Education Program 111–112 organ transplant patients, lipid lowering therapy 138–139 orlistat 240 hypertension reduction in obese 234 mechanism of action 239 obesity treatment 239 sibutramine comparison 239 side-effects and contraindications 240 weight loss in diabetes 235 orofiban, in acute coronary syndromes 413 orthopnea 16–17 orthostatic hypotension 17, 621–622, 628 treatment 629 OSIRIS study 458, 459 osteopontin, in coronary restenosis 380 outcomes (endpoints) anti-ischemic drugs 329–330 cost-effectiveness analysis 52–53 decision analysis 57, 59 categories 73 diagnostic test evaluation 23 economic aspects 47 psychosocial variables 182 PTCA and CABG studies 344–345 quality of care studies 73, 78–79 studies see outcome studies outcomes report cards 83–84 outcome studies 78–81 competing process factors 79 confounding 80–81 GUSTO-1 78, 79 hierarchical statistical modeling 80–81 non-randomized 80–81 process/outcome hypothesis 79 process-outcome relationships 78–81 propensity scores 80 quality of care 78–81, 79 risk-adjustment algorithms 80 selection bias 80 types 78–8180 overhead, hospital 48 overviews, systematic see systematic overviews oxidative stress see also antioxidants hyperglycemia 244–5 pathogenesis atherosclerosis 219 oxygen therapy, in AMI 477–478 p22phox protein 295, 296 PAC-A-TACH trial 599, 600 Pacemaker Atrial Tachycardia (PAC-A-TACH) Trial 599, 600 pacemakers 587–618 bradyarrhythmias 587–618, 629 cardiac 587–618 Chagas’ disease 729 choice, case studies 931–933 conventional indications 588, 594–598

959


Index

pacemakers continued current practice 587–588 diagnostic use 611 goals 587 hypertrophic cardiomyopathy 591, 602–603, 712 mode selection trials 597–600 neurally mediated syncope 591, 601, 628–629 new indications 600–611 Pacemaker Selection in the Elderly (PASE) trial 597–598, 599 pacemaker syndrome 596 without a pacemaker 594 pacing dual chamber 673 heart failure 672–673 Pacing Therapy in Congestive Heart Failure (PATH-CHF) trials 605, 606 paclitaxel, coronary restenosis prevention 381 PACMAN, cardiac resynchronization therapy 606 PAFAC study 521, 529 PAFIT-3 trial 522 PAFT trial 527 pain relief see analgesics Palmaz-Schatz stent 376 palpitations, supraventricular tachycardia 567 PAMI trial 445 pancreatitis, postmenopausal hormone therapy 255 panel methods 75 papillary muscle, rupture, postinfarction 494 Papworth HRT and Survival Enquiry (PHASE) 247 PARAGON, mortality 401 PARAGON-A study, acute coronary syndromes 403 Parkinson’s disease, syncope 621 paroxysmal supraventricular tachycardia (PSVT) 527, 568 partial left ventriculectomy 727–728 PASE trial 597–598, 599 passive diffusion, clinical practice changes 81 PATAF 551 patent ductus arteriosus (PDA), pregnancy 853 PATH-CHF trial, cardiac resynchronization therapy 605 PATH-CHF-II trial, cardiac resynchronization therapy 606 patient-centred medicine evidence-based medicine v 889 limitations 890 patient communication 8 patient compliance 177–178 patients, adherence see compliance PEACE trial ACE inhibitors 581 postinfarction patients 510 pedigree, inheritance of cardiac disease 289 penicillin acute rheumatic fever 754 infective endocarditis 824 rheumatic fever prophylaxis 754 streptococcal pharyngitis 752–753 PENTALYSE study 466, 468 pentasaccharide 466, 468 acute coronary syndromes 414 pentoxiphylline 879, 880 “penumbra,” ischemic stroke 842 percutaneous coronary intervention (PCI) see also percutaneous transluminal coronary angioplasty (PTCA) adjunctive therapy 360–370 in AMI cost effectiveness 450–451

960

facilitated 450 with glycoprotein (GP) IIb/IIIa receptor inhibitors 449 resource use 450–451 restenosis after see coronary restenosis percutaneous left atrial appendage transcatheter occlusion (PLAATO) 564 percutaneous metal mitral commissurotomy 800 percutaneous transluminal angioplasty (PTA) peripheral arterial disease 882–883 subintimal 883 percutaneous transluminal coronary angioplasty (PTCA) 882–883 see also coronary angioplasty; percutaneous coronary intervention (PCI) acute complications 360 acute coronary syndromes 416–418 adjunctive therapy 360–370 adverse effects 348–349 angina after, calcium antagonists 334 appropriateness of use 76–77 CABG v 75, 355 case studies 892–895 current recommendations 355, 389 database studies 353–354 multivessel disease 346, 348–350 single vessel disease 346 theoretic aspects 344–345 chronic complications 361–362 chronic coronary artery disease 339–359 coronary stenting v 344 direct (in AMI) 444–455 coronary stents v 449 cost-effectiveness 450–451 fibrinolytic therapy v 445–448 recommendations 452 resource use 450–451 time to treatment 448–449 economic aspects 49–50 high-risk patients 344–345 indications 339 low-risk patients 345 mechanisms of action 373, 383 medical therapy v 343–344, 355 theoretic aspects 344–345 moderate-risk patients 344–345 mortality postinfarction 488 observational studies 353–354 phase I restenosis prevention 375 postinfarction patients 511–512 prevention of restenosis after PTCA 344 primary (in AMI), recommendations 438 restenosis after see coronary restenosis stents use after 376 performance index (PI), prosthetic valves 812 pericardial disease 735–748 primary acute 735–737 pericardial effusion 737–738 diagnosis 737, 741 postinfarction 495–496 treatment 737, 743 tuberculous 740–744 diagnosis 740–742 pericardial knock 745 pericardiectomy constrictive pericarditis 738–740, 745–746 recurrent pericarditis 736 pericardiocentesis primary acute pericardial disease 737 tuberculous pericarditis 740–741 pericarditis acute 735–770 diagnosis 735 etiology 736

primary 735–736 treatment 735–736 constrictive 738–740 diagnosis 738–740 endomyocardial fibrosis v 758 restrictive cardiomyopathy v 739 treatment 740 Idiopathic relapsing 736 postinfarction 495 tuberculous 740–746 tuberculous constrictive 738, 744–746 diagnosis 741, 742, 744–745 effusive 746 treatment 745–746 perindopril postinfarction patients 510 stroke prevention 843–844 Perindopril Protection Against Recurrent Stroke Study (PROGRESS) see PROGRESS trial peripartum cardiomyopathy (PPCM) 686–687, 856 familial 686 mortality 686 subsequent pregnancies 686 treatment 686–687 peripheral vascular disease 877–886 see also critical limb ischemia; intermittent claudication epidemiological transition 93 epidemiology 877 hyperlipidemia 125, 879 investigations 878 lipid lowering therapy 140–141 long term outcome 877 management case studies 912–914 preoperative cardiac evaluation 881 surgical treatment see vascular surgery suspected coronary disease with 912–914 thrombolysis 883 personality, CHD risk and 183–189 Pharmacological Intervention in Atrial Fibrillation trial (PIAF) 533 pharyngitis, group A streptococcal (GAS) 751–752, 753 PHASE (Papworth HRT and Survival Enquiry) 247 “phenotypic plasticity,” growth and CHD development 282 phosphodiesterase inhibitors 667 phosphorylcholine, coronary stent coating 381 photodynamic therapy, coronary restenosis prevention 383 physical activity see exercise physical examination 14 usefulness 17–21 physicians see also clinical practice smoking cessation advice 115, 116, 118, 119, 308 Physicians’ Health Study (PHS) 221 homocysteinemia 224 phytochemicals, cardiovascular disease relationship 315 PIAF trial 533 pimobendan 668 piretanide 660–661 PLAC study 64 plasmin 427 plasminogen activation factor (PAI-1), in South Asians 265–266 plasminogen activators 427 see also reteplase (rPA); tenecteplase (TNK-tPA); tissue-type plasminogen activator (tPA, alteplase) recombinant tissue (rt-PA), in ischemic stroke 842


Index

single chain urokinase-type (scuPA) 427, 428 TNK-plasminogen activator (TNK-PA), ICH risk 432 platelet activating factor 362, 406 platelet-derived growth factor (PDGF) 377 platelets see also antiplatelet therapy activation 377 adhesion 362, 406, 409 aggregation 362, 406, 409 coronary restenosis 376–378 antiaggregatory strategies 362–368, 409 IIb/IIIa receptor inhibitors see glycoprotein (GP) IIb/IIIa receptor inhibitors thrombus formation 361–362, 405–406, 463 pneumatic compression devices 870 Poland 261 policosanol 138 polymerase chain reaction (PCR) myocarditis 981 tuberculosis diagnosis 741–742 polyunsaturated fatty acids (PUFAs) biologic effects 312 cardiovascular disease and 312–313 n-3 312, 317, 319, 507 sudden cardiac death prevention 580 n-6 312, 319 popliteal artery clinical assessment 879 percutaneous transluminal angioplasty 882–883 population based interventions lipid lowering therapy 127–128 prevention 98–99 tobacco control 108–112, 116 population growth 95 Portuguese Salt Trial 316 positron emission tomography (PET), hypertrophic cardiomyopathy 706–707 Post-Coronary Artery Bypass Graft (CABG) trial cost analysis 302 elderly 139 postinfarction see myocardial infarction (MI), postinfarction postmenopausal hormone therapy 244–258 adverse effects 250–252, 255 angioplasty 248 cardiovascular disease and 244–245 cerebrovascular disease and 252–254 clinical trials 252–254 observational studies 252 primary prevention 252–253 secondary prevention 253–254 coronary artery bypass grafting 248 long-term use 255 postinfarction patients 511 treatment recommendations 255 unstable angina 249–250 venous thromboembolism 254 clinical trials 254 observational studies 254 post-test probability 26–27 post-test referral bias 27 postural tachycardia syndromes (POTS) 628 syncope 621 potassium intake, hypertension and cardiovascular disease 316 supplementation 149–150 potassium channel, mutation in long QT syndrome 290 potassium channel blockers arrhythmias due to 569 supraventricular tachycardia 568 PPP study 220

PRAISE studies 334, 695 PRAISE I study 663 PRAISE II study 663 idiopathic dilated cardiomyopathy 695 pravastatin 131 acute coronary syndromes 418 as anti-ischemic drug 335 cost-effectiveness 63–64, 141–142, 302–303 decision analysis 61 diabetic patients 140 efficacy 132 pleiotropic effects 132 postinfarction period 511 toxicity 133 women 140 Pravastatin Inflammation CRP Evaluation (PRINCE) 132, 226 prazosin, in heart failure 663 prednisolone idiopathic dilated cardiomyopathy 695 pericardial effusion treatment 742–744 pre-eclampsia 856, 857 management 860 mutation associated and spironolactone effect 294 pregnancy 856 antepartum management 859–860 antiarrhythmics 859–860 anticoagulants 860, 861 antihypertensive drugs 860 aortic stenosis 853–854 arrhythmia management 859–860 atrial fibrillation 855–856 balloon valvuloplasty 792, 802 blood pressure 853, 856 cardiac surgery 858–859 cardiovascular physiology 853 congenital heart disease 853–855 coronary artery disease 857 cyanotic heart disease 854–855 Eisenmenger syndrome 855, 860, 861 heart disease and 853–867 management 857–861 risk stratification and counseling 857–859 high-risk units 860 hypertensive disorders 854, 856–857 gestational 856, 860 management 860 pre-existing 856 hypertrophic cardiomyopathy 857 left-to-right shunts 853 left ventricular outflow tract obstruction 853–854 Marfan syndrome 855 maternal functional status in risk stratification 858 maternal life expectancy 859 mineralocorticoid receptor mutation causing hypertension 294 mitral valve prolapse 855 multidisciplinary approach 860 myocardial infarction 857 nutrition, effect on birthweight and later CHD 284 palliative surgery (for heart disease) 858–859 pulmonary embolism, diagnosis 869 pulmonary hypertension 853 pulmonary stenosis 854 pulmonary vascular obstructive disease 855 rheumatic heart disease 855–856 transposition of great arteries 854, 855 venous thrombosis 865 diagnosis 867 PRESTO trial 379

pretest probability 24, 26–27, 55 prevention, CHD 219–230 see also antihypertensive drugs; lipid-lowering therapy; smoking, cessation African-Americans 273 African Blacks 271 Arabs 268 Chinese 264 combined strategies 98–99 of coronary restenosis see coronary restenosis cost-effectiveness 300–308 emerging approaches 219–230 ethnic variations 273 Europeans 261–262 high risk approach 98 Hispanics 269 hypertension 149–150 Japanese 262–5 native North Americans 270–271 paradox 98 physical activity/exercise in 170–180 population approach 98–99 primordial 98–99 psychosocial factor modification 206, 212 South Asians 267 Prevention of Atrial Fibrillation after Cardioversion (PAFAC) 521, 529 Prevention of Restenosis with Tranilast and Its Outcome (PRESTO) 379 PREVENT study 331–332 Primary Prevention Project (PPP) study 220 PRIME-II study 667 PRINCE 226 PR interval, prolonged 594 PRISM, mortality 401 PRISM-PLUS 412–413 acute coronary syndromes 417 mortality 401 TIMI risk scores 404 proarrhythmias, antiarrhythmic drug-induced 540–541, 569 probability decision analysis 57, 59 post-test 26–27 pretest 24, 26–27, 55 proband 289 probucol coronary restenosis prevention 362 hyperlipidemia 222 Probucol Quantitative Regression Swedish Trial (PQRST) 222 procainamide, in atrial fibrillation 525 paroxysmal 525, 526 persistent 531 post-cardiac surgery 535–536 post-operative 538 process of care 73 procoagulant state, after fibrinolytic therapy 456–457 PROFILE trial 663–664 progestins, CHD and 252, 253 prognosis, MEDLINE search strategies 42 PROGRESS trial 152, 843 hypertension as risk factor 647 MI prevention 223 projected smoking-related losses 107 PROMISE trial 667 propafenone 528 atrial fibrillation 525 paroxysmal 522, 569 persistent 531 post-cardiac surgery 535, 538 prevention 527

961


Index

propafenone continued sudden death survivors 581 supraventricular tachycardia 568 ventricular arrhythmias, sustained 578 Propafenone Atrial Fibrillation Trial (PAFT) 527 Propafenone in Atrial Fibrillation Italian Trial (PAFIT-3) 522 propensity scores, outcome studies 80 propranolol acute coronary syndromes 407–408 atrial fibrillation 534 post-operative 536 heart failure 669 threatened MI 333 Prospective Pravastatin Pooling (PPP) Project, pleiotropic effects 132 Prospective Randomized Amlopidine Survival Evaluation (PRAISE) trial see PRAISE studies prostacyclin aspirin effects 410–411 heart failure 664 peripheral vascular disease 880 synthesis 362 thromboangitis obliterans 884 prostacyclin analogs, coronary restenosis prevention 378 prosthetic valves 811–814 see also bioprosthetic valves; mechanical prosthetic valves choice 811–812, 814 effective orifice area (EFA) 812, 813 infective endocarditis 817–818 performance index (PI) 812 regurgitation 812, 813 young women 764 protein, glycation 164 protein C deficiency 864, 871 protein S deficiency 864, 871 proteinuria, gestational hypertension with 856 proteoglycans, in coronary restenosis 380 prothrombinase complex, in thrombus formation 463 proto-oncogenes, smooth muscle cell proliferation 380 protozoal infections, causing myocarditis 682 prourokinase (scuPA) 427, 428 PROVED trial 660 PROVE trial 558 P-selectin, in coronary restenosis 377, 379 pseudoaneurysm 493 postinfarction 493 Pseudomonas cepacia, infective endocarditis 823 PSVT trial 527 psychiatric disorders, syncope 624–625 psychological well-being, cardiac rehabilitation and 174 psychosocial complications, myocardial infarction 499 psychosocial factors 181–218 as CHD risk factors 181 causal association 181–182 definition 182 evidence for effect on CHD 181–218 mechanisms of effect on CHD 182, 183 modification to prevent CHD 206, 212 studies 182 bias 213 design 182 size effects 182, 212–213 summaries 213 systematic review method 182 psychosocial interventions, postinfarction 499 PTCA see percutaneous transluminal coronary angioplasty

962

“publication bias” 37 Puerto Rican-Americans see Hispanics pulmonary angiography 867, 869 pulmonary artery pressure, measurement, postinfarction 488 pulmonary congestion mitral regurgitation 758–759, 759–760 mitral stenosis 763 pulmonary edema, in heart failure 660 pulmonary embolism (PE) 865 Chagas’ heart disease 721 diagnosis 867–868 algorithms 869 clinical suspicion 868 pregnancy 869 differential diagnosis 865, 869 estrogen replacement therapy and 254 prophylaxis 865, 869 risk factors 865 syncope 623 treatment 869 women 244, 345 pulmonary hypertension mitral stenosis 763, 764 pregnancy 853 pulmonary stenosis, pregnancy 854 pulmonary valve, allograft 813–814 pulmonary vascular obstructive disease, pregnancy 855 pulmonary vein triggers, AV ablation 562–563 pulseless electrical activity (PEA), cardiac arrest 634 pulses clinical assessment 20–21 peripheral vascular disease 879 pulsus paradoxus 738 PURSUIT trial 412–413 acute coronary syndromes 417 risk scores 403 mortality 401 Q fever 823 quality-adjusted life years (QALYs) 52–53 quality-adjusted survival 57 quality of care outcome studies 78–81, 79 process studies 73–78 quality of life, in cost-effectiveness analysis 52–53 quinapril, in heart failure 666 quinidine atrial fibrillation 525 paroxysmal 526, 528 persistent 530, 531, 532 post-cardiac surgery 536 decision analysis 61, 65 pregnancy 859 supraventricular tachycardia 568 RACE 556 race see ethnic groups RACE study 533–534 RADIANCE trial 660 radiation coronary restenosis prevention 382 inducing constrictive pericarditis 738 radiofrequency (RF) ablation cost-effectiveness 54 supraventricular tachycardias 629 ventricular tachycardia 629 radionuclide angiography exercise, incremental value 30 hypertrophic cardiomyopathy 706–707

radiotherapy, external beam, coronary restenosis prevention 382 RAFT trial 527 RALES trial 580, 661 heart failure 661–662 idiopathic dilated cardiomyopathy 696 postinfarction patients 517 raloxifene 251 ramipril see also AIRE study cost effectiveness 305–306 diabetes mellitus 167 heart failure 664, 665, 666 MI prevention 223 postinfarction patients 510 with left ventricular dysfunction 490 stroke prevention 843 sudden cardiac death prevention 580 RAND group 75, 77–78 Randomized Aldactone Evaluation Study (RALES) see RALES trial randomized controlled clinical trials (RCTs) 5, 7–8, 34–39, 71–72, 889 compliance 35 confounding 37 false negative results 36 large scale 37–38 limitations 889–890 minimizing bias 34–37 data-dependent emphasis 35–36 intention to treat analysis 35 moderate 34–37 proper randomization 34–35 see also bias minimizing random errors 36–37 observational studies v 38 random errors and 889 subgroup analysis, inappropriate 36 uncertainty principle 37–38 Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart failure (REMATCH) 672 Randomized Evaluation of Strategies for Left Ventricular Dysfunction (RESOLVD) study, diabetic patients 648 Randomized Intervention Treatment of Angina (RITA) see RITA trials ranolazine, as anti-ischemic drug 335 rapamycin, restenosis therapy 361 rapamycin-coated coronary stents 382 rapeseed oil 319 RAPID 2 study 434 RAte Control versus Electrical cardioversion (RACE) 533–534 RAVEL study 361 reagudization, Chagas’ disease 726 receiver operating characteristic (ROC) curves 28–29 recombinant tissue plasminogen activator (rt-PA), ischemic stroke 842 recommendations, gradings 2, 90, 328, 396, 518, 576, 642, 734, 750, 838, 888 Reducing Tobacco Use 111–112 Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan (RENAAL) Study 152 referral bias, post-test 27 regression dilution bias 122 rehabilitation alternative delivery approaches 178–179 cardiac 172, 508 cost-effectiveness 304–305 exercise training 171–175 patient compliance 177–178 relative risks (RR), cardiovascular incidents 148


Index

REMATCH 672 remodeling, left ventricle 650–651 RENAAL Study 152 renin activity, plasma (PRA) 649–650, 652 renin-angiotensin system 222–223 drugs affecting see angiotensin converting enzyme (ACE) inhibitors; angiotensin II receptor antagonists heart failure 660, 664–666 ventricular remodeling and 651 reperfusion damage 483–484 reperfusion therapy, postinfarction 490–491 strategies 444–455 reports, integrative 71 resistance-training 171, 173–174 RESOLVD trial 666, 669 diabetic patients 648 RESOLVE trial 379 resources cardiovascular care 46, 47 “law of diminishing returns” 47 rest see also bedrest pain, in limb ischemia 880 restenosis see coronary restenosis RESTORE study 366, 379 resuscitation, cardiopulmonary see cardiopulmonary resuscitation resynchronization therapy, in heart failure 673 reteplase (rPA) 427, 428 adjunctive therapies, trials 435 combination trials 470–471 comparative trials 433, 434, 462, 469–470 efficacy 444 ICH risk 432 regimen selection 438 rethrombosis 463 retrograde non-transseptal technique 800–801 revascularization, myocardial see myocardial revascularization reviparin 367 rhabdomyolysis fibric acid derivatives 137 statins 133 rheumatic fever 751–757 acute management 754–755, 756 clinical features 752 epidemiology 751 pathogenesis 751–752 prevention 752–754 primary 752–753, 756 secondary 753–754, 756 rheumatic heart disease (RHD) 751–757 aortic valve 767, 774, 782 atrial fibrillation 548 epidemiological transition 92–93 global burden 92 mitral valve 758–759, 763–764 pregnancy 855–856 ribaviron, in myocarditis 690 rickettsial infections, causing myocarditis 682 right atrial linear compartmentalization, AV ablation 563 right ventricle (RV) Chagas’ disease 721 dysfunction, in myocarditis 683 dysplasia, arrhythmogenic see arrhythmogenic right ventricular dysplasia (ARVD) hypertrophy 703 mitral stenosis 763 right ventricular infarction/failure 491–492 clinical features and prognosis 491–492 management 492 risk comprehensive cardiovascular (total) 98

continuum 97–98 multiplicative 98 pyramid 97, 98 relative see relative risks (RR) thresholds 97 risk-adjustment algorithms, outcome studies 80 risk factors, CHD 279, 310 African-Americans 272 African Blacks 271 Arabs 267–268 causality criteria 96–97 Chinese 263 clinical v prevention norms 97 developing countries 95 diet 309–325 see also diet emerging 165–166, 219–230 epidemiological transition 92–93, 94 ethnic variations 96, 273 Europeans 261 evolving concept 96–98 Hispanics 269 hyperglycemia associated 165–166 Japanese 262 multiple 98 intervention programs 306 native North Americans 270 obesity 236, 313 plasma cholesterol levels see cholesterol sodium/salt intake 315–316 South Asians 265–266 weight loss effect 234, 236 West Indians 271 risk scores, acute coronary syndromes 403–404 risk stratification acute coronary syndromes 401–402 exercise 178, 179 heart disease in pregnancy 857–859 hypertrophic cardiomyopathy 292 ventricular arrhythmias and sudden death 577 RITA trials 343–344, 346, 347 patient profiles 348 rosiglitazone, diabetes mellitus 167 Ross procedure 814 rosuvastatin 133–134 rotational atherectomy 352 Rotterdam Study, homocysteinemia 225 Roux-en-Y gastric bypass, obesity 240 rPA see reteplase rural areas see urban-rural differences Russian Federation 260 cardiovascular disease 260–261 4S study see Scandinavian Simvastatin Survival Study SAFE PACE trial 601 SAFE PACE 2 trial 601 SAFIRE-D trial 526, 529, 530 salicylates see also aspirin (acetylsalicylic acid) in acute rheumatic fever 754–755 salt intake see sodium intake San Antonio Heart Study 269 saphenous vein infrainguinal vascular reconstructions 881–882 situ technique 881–882 sarcomeric proteins cardiac, gene mutations 703 familial hypertrophic cardiomyopathy pathogenesis 292 genes coding 291 saruplase 428–429 saturated fatty acids, cardiovascular disease relationship 311–312, 313

SAVE (Survival and Ventricular Enlargement) trial 334, 478, 480, 494, 509–510, 664 cost-effectiveness 64, 305–306 heart failure prevention 646 MI prevention 665 myocardial ischemia prevention 654 Scandinavian Simvastatin Survival Study (4S) 122, 125, 510–511 cost analysis 302 cost-effectiveness analysis 63 elderly 139 heart failure rates 648, 653 stroke rates 125 women 140 scarring, balloon aortic valvuloplasty and 787, 789 SCATI study 458 SCD-HeFT trial 583, 697 school-based smoking prevention programs 110 Scientific American Medicine (SAM) 44 scopalamine, vasovagal syndrome 628 Scottish Heart Health Study 183 SCPS (Skin Cancer Prevention Study) 221 screening asymptomatic LV dysfunction 644–646, 646, 704 blood lipids in 127 severe coronary artery disease 24–26, 28–29, 30–31 SCRIP (Stanford Coronary Risk Intervention Program) 306 SCRIPPS trial 382 Secondary Prevention with Antioxidants in Endstage renal disease (SPACE) trial 221 SECURE trial 220–221 seizures, syncope and 625 selectins, in acute coronary syndromes 404 selection bias, outcome studies 80 selective serotonin reuptake inhibitors (SSRIs), vasovagal syndrome 628 Senning procedure, pregnancy 854 sensitivity, analysis 57, 58, 60 septal myotomy-myectomy (SMM) 711–712 serotonin 362 serum amyloid A (SAA) acute coronary syndromes 404 as marker 226 Seven Countries study 126, 260, 262 CHD risk and plasma cholesterol 310 saturated fats and CHD 311–312 trans fatty acids and CHD 312 sex differences see gender differences SHEP 148, 150 shock, cardiogenic see cardiogenic shock SHOCK trial 491, 494, 495 sibrafiban, in acute coronary syndromes 413 sibutramine 240 hypertension reduction in obese 234 mechanism of action 239 obesity treatment 239 orlistat comparison 239 side-effects and contraindications 240 weight loss in diabetes 235 sick sinus syndrome 530 cardiac pacing 597 SIDS study 497 significance, clinical, diagnostic tests 29–31 simvastatin 123–124, 127–128, 131, 510–511 animal model of hypertrophic cardiomyopathy 296 combined therapy 138 cost-effectiveness 63–64, 141 decision analysis 60 efficacy 132, 133

963


Index

simvastatin continued heart failure prevention 648, 653 hypertension 222 toxicity 133 single-intervention studies, clinical practice changes 82 single nucleotide polymorphisms (SNP) 289 ABCA1 gene and atherosclerosis 295 sinoatrial disease see sinus node dysfunction Sino-MONICA project 263 sinus node dysfunction (SND) 594, 596–600, 629 pacemaker mode selection 597–600 pacing indications 590, 597 post-cardiac transplantation 610 syncope 597, 622 sinus tachycardia 628 Sirolimus, coronary stent coating 382 Skin Cancer Prevention Study (SCPS) 221 sleep apnea 610–611 obesity and 237 Smart Artery Radiation Therapy trial (SMART) 382 SMART trial 382 SMILE study 510 smoking see also tobacco African-Americans 272 African Blacks 271 Arabs 268 assessment 118 cessation 114–120 bupropion 117 cardiac rehabilitation and 174 clinical practice 118–119 clinics 118 community interventions 108–112, 116 cost-effectiveness 303 evidence basis 115–116 evidence of benefits 114–115, 115, 116 nicotine replacement 116, 117–118, 119 physician advice 115, 116, 118, 119, 303 postinfarction 114–115, 118, 303, 507–508 practical assistance 118–119 process 116 review of studies 117–118, 118 secondary prevention of stroke 840, 843 as CHD risk factor 104–106 Chinese 263 current burden 103–104 developing countries 95 effect on dietary studies of CHD 311 epidemiological transition 92–93, 93 future projections 106–108 global burden 103–104 HDL cholesterol and 127 heart failure risk 647 Hispanics 269 Japan 263 LV dysfunction prognosis and 652 mortality 109 native North Americans 270 nature 115 passive, reducing 114 peripheral vascular disease and 877, 879 prediction of coronary artery disease 24–25 prevalence 105 prevention 110–112, 114–115 process 116 relapse 116 South Asians 265, 266 stroke risk 106 thromboangitis obliterans 884 smooth muscle cells (SMC)

964

activation in phase III coronary restenosis 379–380 coronary restenosis 377 phase III proliferation 374, 379–380 prevention 379 platelet interactions 377, 378 proliferation 380 control 377, 378, 380 inhibition 383 prevention 380–382 SNC5A gene, mutation in long QT syndrome 290 social class, dietary behavior link 311 social functioning, exercise training and 174–175 social supports buffer theory 206 CHD risk and 206, 207–211 interventions to prevent CHD 212 societal perspective 52 socioeconomic status African-Americans 272 cardiovascular disease (CVD) risk 93 CVD prevalence 270 sodium, excretion and CHD relationship 315 sodium channel, SNC5A mutation in long QT syndrome 290 sodium channel blockers 290 arrhythmias due to 569 supraventricular tachycardia 568 sodium intake cardiovascular disease relationship 315–316 diet low in, trial 316, 319–320 hypertension association 315–316 Japanese diet 319 reduction 149–150 DASH diet trial 319–320 SOLVD (Studies of Left Ventricular Dysfunction) angina prevention 334 aspirin therapy 481 diabetic patients 648 heart failure prevention 646, 651–652, 665 heart failure therapy 664, 671 myocardial ischemia prevention 654 neuroendocrine changes 649–650 postinfarction patients 512 sotalol 528, 578 atrial fibrillation paroxysmal 526, 527, 530 persistent 532 post-operative 536, 538, 539 prevention 527, 530 heart failure 671 postinfarction period 511 pregnancy 859 sudden death survivors 581 supraventricular tachycardia 568 ventricular arrhythmias non-sustained 579 sustained 578 South Asians 264–267, 273 disease burden 264, 265 migrant 265, 266–267, 267 migrant groups 265 prevention strategies 267 risk factors 265–266 temporal trends 265 urban-rural differences 266 soy consumption 317 SPACE trial 221 SPAF (Stroke Prevention in Atrial Fibrillation) studies 527, 548, 549, 550–551, 551 SPEED trial 435, 460, 468, 469–470 SPINAF study 548, 549

spinal cord stimulation 880 SPIRIT trial 845 spirochetes, causing myocarditis 682 spironolactone Chagas’ disease 727 heart failure 661–662 hypertension associated 294 postinfarction patients 517 SPNIDDM study 167 STAF trial 533, 534, 556 Stages of Change model 116 standard agents 427–429 Stanford Five-City Project 109 Staphylococcus aureus endocarditis 817, 824 staphylokinase (SAK) 429 STARC trial 383 Starr-Edwards prosthetic valve 812 STAT-CHF study 697 state transition (Markov) models 58 statins see HMG-CoA reductase inhibitors STE-AMI trial 437 STEMI (ST segment elevation) see ST segment Stenotrophomonas maltophilia, infective endocarditis 823 Stenting versus Internal Mammary Artery (SIMA) study 352 Stent in Restenosis Study (STRESS) see STRESS study Stent PAMI 445 stents comparative studies 469 intracoronary see coronary stents peripheral arteries 883 Stent Versus Thrombolysis for Occluded Coronary Arteries in Patients With Acute Myocardial Infarction (STOPAMI) 469 steroids see corticosteroids STIMA study 352 St Jude prosthetic valve 813 antithrombotic therapy 833 stockings, graduated compression 870 Stokes-Adams attacks 595 STOP-AF trial 599 STRATAS study 376 Strategies for Patency Enhancement in the Emergency Department (SPEED) trial 435, 460, 468, 469–470 Strategies of Treatment of Atrial Fibrillation (STAF) 533, 534, 556 strength training 171, 173–174 recommendations 174 safety 174 streptococcal endocarditis 817, 824 streptococcal group A (GAS) pharyngeal infections 751–752, 752 Streptococcus pneumoniae endocarditis 817 streptokinase (SK) 427–428 combination trials 469 comparative trials 61, 65, 432–434, 462, 465, 469 cost-effectiveness 450 direct PTCA v 445–446 efficacy 429, 431, 444, 460 intracoronary 368 procoagulant state after 456–457 risks 432 therapeutic guidelines 472 stress see also psychosocial factors management 206, 212 STRESS study 376 stress tests see also exercise stress testing incremental value 23, 26–28, 30 preoperative, peripheral vascular disease 881


Index

stroke atrial fibrillation 548, 845 Chagas’ heart disease 728 Chinese 263 complicating balloon valvuloplasty 786, 801 current trends 839, 841 diagnosis 839 direct PTCA v fibrinolytic therapy 446–447 epidemiological transition 93 estrogen replacement therapy and 252 global burden 91–92, 839 hemorrhagic fibrinolytic therapy and 431–432 serum cholesterol and 124–125 high alcohol consumption link 318 home-based care 841 ischemic antiplatelet therapy 844–845 risk of treatment 842 “therapeutic window” 842 mortality 839 new cardiovascular events 148 prevention 843–848 anticoagulants 845 antihypertensive therapy 150–151, 151, 843–844 antiplatelet therapy 844–845, 848 atrial fibrillation 549–553 carotid endarterectomy 846–848 cholesterol-lowering agents 845, 848 decision analysis 56–58, 61, 65–66 HMG-CoA reductase inhibitors 125 secondary 840 prosthetic valve recipients 834–835 recurrence 846 sequelae 839 serum cholesterol and 124–125 smoking and 106 treatment 839–852, 848 costs 840 effectiveness 840 women 244, 345 Stroke Prevention Atrial Fibrillation (SPAF I) see SPAF Stroke Prevention in Reversible Ischemic Trial (SPIRIT) 845 stroke services 841–842, 848 ischemic, treatment 842–843 organization 841 stroke units 841–842, 848 cost-effectiveness 840 stroke volume, in aortic stenosis 767 Strong Heart Study 269 ST segment depression disease/disorders see acute coronary syndrome (ACS) hypertrophic cardiomyopathy 705 elevation 456 disease/disorders see acute coronary syndrome (ACS) fibrinolytic therapy 429, 430 idiopathic ventricular fibrillation 290 left ventricular aneurysm 492–493 postinfarction 488, 489 right ventricular infarction 492 Studies of Left Ventricular Dysfunction (SOLVD) study see SOLVD Study to Evaluate Carotid Ultrasound Changes with Ramipril and Vitamin E (SECURE) trial 220–221 Study to Prevent Non-Insulin-Dependent Diabetes Mellitus (SPNIDDM) study 167 subclavian steal syndrome 624

subgroup analysis 513 sudden cardiac death 577–586 African-Americans 272 aortic stenosis 769 Chagas’ heart disease 720, 721 familial hypertrophic cardiomyopathy 291 heart failure 671 hypertrophic cardiomyopathy 705, 709–710 idiopathic dilated cardiomyopathy 689 implantable cardioverter defibrillation for survivors 581 inducible ventricular tachycardia/fibrillation and 582 myocarditis 685 pharmacologic interventions to prevent 577–580 blockers 579 miscellaneous agents 580 postinfarction, prevention 512–513 prevention by implantable cardioverter defibrillators 58–60, 62–63, 582–583 resuscitation see cardiopulmonary resuscitation Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) 583, 697 suicide, low serum cholesterol and 126 sulfinpyrazone comparative trials 409–410 postinfarction patients 484 stroke prevention 844 sulfonamides, in infective endocarditis 823 sulphonylureas, weight gain 235 superstatins 133–134 see also HMG-CoA reductase inhibitors (statins) supraventricular tachycardia (SVT) 567–574 see also specific types ablation 569–570, 571–572 risks 572 accessory pathway-mediated, ablation 569–570 causes/mechanism 567, 568 clinical features 567 drug therapy 567–569, 571 efficacy 568–569 trials 568 hypertrophic cardiomyopathy 705, 706, 707, 712–713 management recommendations 572 paroxysmal 527, 568 radiofrequency ablation 629 syncope 624, 629 terminology and arrhythmias included 567 therapeutic options 567 surgery see also individual conditions acute mitral regurgitation after MIs 494 cardiac see cardiac surgery left ventricular aneurysm 493 left ventricular wall rupture postinfarction 495 obesity treatment 239–241 pregnancy 858–859 prior aortic valvuloplasty 790 vascular see vascular surgery venous thromboembolism risk 864, 865 survival cost-effectiveness analysis 52–53 decision analysis 57 Survival and Ventricular Enlargement (SAVE) trial see SAVE (Survival and Ventricular Enlargement) trial Survival with Oral d-sotalol (SWORD) study 497, 511, 597 SVT see supraventricular tachycardia (SVT) SWIFT trial 511 SWORD study 497, 511, 579

SYDIT study, vasovagal syndrome pacing 602 sympathetic system, in LV dysfunction 650 Symptomatic Atrial Fibrillation Investigative Research on Dofetilide (SAFIRE-D) 526, 529, 530 syncope 619–633 aortic regurgitation 776 aortic stenosis 769 cardiovascular/cardiopulmonary disease 624, 629 classification of causes 620–625 cost-effectiveness issues 630 definition 619 diagnostic evaluation 17, 625–628 diagnostic pacemaker 611 drug-induced 627 epidemiology 619–620 fascicular block 595–596 hypertrophic cardiomyopathy 624–625, 629, 705 implantable loop recorders 621 neurally mediated 620–621, 627 pacing 591, 601, 628–629 neurocardiogenic, pacing 600 Parkinson’s disease 621 postural tachycardia syndromes 621 sinus node disease 596, 622 situational 17 treatment 628–629 vasovagal see vasovagal syndrome Syncope And Falls in the Elderly - Pacing And Carotid sinus Evaluation (SAFE PACE) 2 trial 601 Syncope Diagnosis and Treatment (SYDIT) study, vasovagal syndrome pacing 602 syndrome X (metabolic syndrome) 141, 235–236 SYNPACE trial, vasovagal syndrome pacing 602 Synthetic Pentasaccharide as an Adjunct to Fibrinolysis in ST-Elevation Acute Myocardial Infarction (PENTALYSE) study 466, 468 syphilis 774 Syrian hamsters, cardiomyopathic 695–696 systematic overviews (meta-analyses) 71 advantages 36 incomplete ascertainment 37 MEDLINE searching 41 publication bias 37 small scale 37 trial selection 37 unreliability 37 Systematic Trial of Pacing to Prevent Atrial Fibrillation (STOP-AF) trial 599 systemic diseases, causing myocarditis 681, 683 systemic embolism see also thromboembolism atrial fibrillation 548, 552 Chagas’ heart disease 728, 750 infective endocarditis 818 Syst-Eur, hypertension treatment 152 systolic ejection time, in aortic stenosis 767 tachycardia accessory pathway-mediated 569–570 atrial 567, 570, 611 atrioventricular nodal re-entrant 567, 570 atrioventricular re-entrant 567, 570 diagnostic pacing 611 narrow complex 567 sinus 628 supraventricular see supraventricular tachycardia (SVT) ventricular see ventricular tachycardia tachycardia-bradycardia syndrome 596, 622

965


Index

TACTICS-TIMI-18 trial 417, 900–901 mortality 401 postinfarction patients 512 TACTICS trial 417 TIMI risk scores 404 tamponade ablation risk 572 complicating balloon valvuloplasty 785, 786, 801 echocardiographic features 737, 738–739 pericardial effusion causing 737, 738–739 postinfarction 495–496 syncope 623–624 treatment 737 Tangier disease 295 TARGET 364 TEE see transesophageal echocardiography tenecteplase (TNK-tPA) 427, 428 adjunctive therapies 470 trials 435–436 combination trials 470–471 comparative trials 433, 434–435, 460, 462, 463 efficacy 436, 444 regimen selection 438 tetracyclines, in infective endocarditis 823 tetralogy of Fallot, pregnancy 854 TexCAPS see Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS) textbooks 43–44 online 44 thallium-201 imaging, incremental value 28–29, 30–31 thiazolidinediones, weight gain 235 thiocyanate toxicity 662 third heart sound, clinical assessment 18 three-dimensional mapping, AV ablation 563 thrifty gene 95–96 thrombin see also antithrombotic therapy direct inhibitors 465–466 AMI 464–466 mechanism of action 456 meta-analysis 466 PTCA 367–368 generation during thrombolysis 456–457 indirect inhibitors mechanism of action 456 PTC A 367 see also heparin mechanism of action 463 thrombus formation 362, 406 thromboangitis obliterans 884 thrombocytopenia, heparin causing in pregnancy 860 thromboembolism see also pulmonary embolism; stroke; systemic embolism atrial fibrillation 522 cardiac, postinfarction 493 Chagas’ heart disease 721, 728 complicating balloon valvuloplasty 801 prosthetic valve recipients 811–812, 813, 814, 832–836 venous see venous thromboembolism thromboendarterectomy (TEA) 881 Thrombolysis and Angioplasty in Myocardial Infarction (TAMI)-3 study 457 Thrombolysis in Myocardial Infarction (TIMI) see TIMI (various trials) Thrombolysis in Myocardial Infarction (TACTICS-TIMI-18) trial see TACTICS-TIMI-18 trial thrombolytic therapy see fibrinolytic (thrombolytic) therapy

966

thrombosis dietary fat and 126 venous see venous thrombosis thromboxane A2 (TXA2) 362, 409, 410 inhibition by aspirin 844 inhibitors 410–411 thromboxane antagonists, coronary restenosis prevention 378 thrombus see also coronary thrombus atrial, in atrial fibrillation 552 formation 361, 405–406 left atrial, in mitral stenosis 797, 798 prosthetic valve 811–812 thymomodulin idiopathic dilated cardiomyopathy 695 myocarditis 693 Tianjin trial 316 TIBET study 332 Ticlid or Plavix Post-Stent (TOPPS) trial 363–364 ticlopidine acute coronary syndromes 411 adjunctive therapies 439 coronary restenosis prevention 378 coronary stent recipients 368 peripheral vascular disease 879–880 prosthetic valve recipients 835 stroke prevention 844 tilt-table testing, head up 621, 625, 627 TIMI-2B trial 401, 414, 508 TIMI-2 trial 479, 492, 499, 511 TIMI-3 registry 403, 416 TIMI-5 trial 464 TIMI-9A trial 460 TIMI-9B trial 415, 460, 464, 467 TIMI-9 trial 432 TIMI-14 trial 468, 469 TIMI risk score acute coronary syndromes 403–404 validation 404 timolol, in AMI 480 tinzaparin, in acute coronary syndromes 417 tirofiban 364 acute coronary syndromes 412–413, 417 comparative studies 366–367 coronary restenosis prevention 379 PTCA/atherectomy 366–367 Tirofiban And ReoPro Give similar Efficacy Trial (TARGET) 366 tissue Doppler velocity, familial hypertrophic cardiomyopathy diagnosis 296–297 tissue-type plasminogen activator (tPA, alteplase) 427, 428 adjunctive therapies 439, 469 AMI coronary stents v 448 direct PTCA v 445–446 bolus/infusion 428 comparative trials 61, 65, 432–434, 434, 460 cost-effectiveness 54 current use 438 efficacy 429, 430, 431, 444, 460 ICH risk 432 intra-arterial 883 intracoronary 368 mechanism of action 427 mutant see reteplase procoagulant state after 456–457 risks 432 titin, mutations in familial hypertrophic cardiomyopathy 291 TNK-plasminogen activator (TNK-PA), ICH risk 432 tobacco 103–113

control community based 108–112, 116 strategies 114 environmental exposure (ETS), reducing 110–112, 114 nature 115 toxic myocarditis 683 TRACE (Trandolapril Cardiac Evaluation) study 480 use see smoking TOHP II trial 232 tolerance 171 ACE inhibitors and 664 mitral regurgitation 762 peripheral vascular disease 879–880 women 176 TONE trial (trial of non-pharmacologic interventions in the elderly 232–233 TOPPS trial 363–364 toresemide 650 torsade de pointes 541 antiarrhythmic drug-induced 540–541, 579 long QT syndrome 623–624 tPA see tissue-type plasminogen activator (tPA, alteplase) TRACE (Trandolapril Cardiac Evaluation) study 478, 510, 537, 664 trandolapril atrial fibrillation 537 heart failure 664, 666 postinfarction patients 510 Trandolapril Cardiac Evaluation (TRACE) study 478, 510, 537, 664 tranilast, coronary restenosis prevention 379, 382 transesophageal echocardiography (TEE) during balloon mitral valvuloplasty 801 cardioversion in atrial fibrillation 552 mitral regurgitation 761 mitral stenosis 797 trans fatty acids (t-FAs) 124, 312, 313 transforming growth factor- (TGF-), in coronary restenosis 377 transgenic mice, hypertrophic cardiomyopathy 296 transgenic rabbits, hypertrophic cardiomyopathy 296–297 transient ischemic attacks (TIAs) diagnosis 839 stroke risk 549–550, 551, 553 transplantation heart see cardiac transplantation lipid lowering therapy after 138–139 transposition of great arteries, pregnancy 854, 855 transseptal techniques, balloon valvuloplasty 785, 798 transvenous pacing, postinfarction 498 trapidil 383 treadmill exercise testing, incremental value 27–28, 30 treatment cost-effectiveness 53–54 MEDLINE search strategies 42 Trial of Non-Pharmacologic Interventions in the Elderly (TONE) 232–233 Trials of Hypertension Prevention II (TOHP II) 232 triglycerides dietary, CHD relationship 311 gender differences 344–345 impact of therapy 132 lipid lowering therapy and 133, 137 serum 126–127


Index

trimetazidine, as anti-ischemic drug 335 tropomyosin, in idiopathic dilated cardiomyopathy 684–685 troponin-I, postinfarction 489 troponin-T mutations 703, 709 postinfarction 489 prognosis in acute coronary syndromes 402–403 Trypanosoma cruzi infection 718–719 tubercle bacilli, culture 740–741 tuberculin skin testing 741 tuberculosis 735, 740, 765 tuberculous pericarditis 740–746 tumor necrosis factor (TNF) acute coronary syndromes 404 as marker 226 myocarditis 693 Turner syndrome 854 T wave abnormalities acute coronary syndromes 403 hypertrophic cardiomyopathy 705 twin studies, polygenic inheritance of cardiac disease 289 type A behavior CHD risk and 183–189 modification 212 ulcerative colitis 692 ulcers, ischemic 879, 880 ultrasound lower limb ischemia 880 venous (VUI) United States (USA), ethnic variations 273 UKPACE trial 596, 599, 600 UKPDS 165–166 UK Prospective Diabetes Study 235 UK-TIA 550 ultrasound, in syncope 626 uncertainty principle, randomized clinical trials 37–38 undernutrition, CHD pathogenesis 283–284 United Kingdom, appropriateness of service use 76 United Kingdom Pacing and Cardiovascular Events (UKPACE) trial 596, 599, 600 United Kingdom Prospective Diabetes Study (UKPDS) 165–166 United States (USA) appropriateness of service use 76, 77–78 cardiovascular mortality 91–92, 260 ethnic variations 273–274 unstable angina (UA) 329, 330, 332–333, 335, 397–425 see also acute coronary syndrome (ACS) acute phase, antithrombotic therapy 364 blockers 332–333 Braunwald classification 398 calcium antagonists 331, 332–333 definitions 397–398 historical perspective 397 incidence 399 management 406–409 algorithm 899 case studies 896–899 nitrates 332–333 pathophysiology 456 PCI 364 postmenopausal hormone therapy 249–250 prognosis 399 subacute phase limitations of evidence 513 PTCA and CABG 514 treatment 367

urban-rural differences African Blacks 271 Arabs 268 cardiovascular disease (CVD) risk 93 China 263, 264 South Asia 266, 267 urogenital procedures 827 urokinase (UK) 427, 428 intracoronary 368 utilization review (clinical audit) 73–78 VACSDM 166 vagus nerve atrial fibrillation mediated 520–521 permanent 520 VA-HIT trial gemfibrozil 136 postinfarction patients 511 statin cost effectiveness 142 Val-HeFT study 666 validity, audit criteria 75 Valsartan Heart Trial Investigators, idiopathic dilated cardiomyopathy 694 valves prosthetic see prosthetic valves replacement antithrombotic therapy 832–836 infective endocarditis 826–827 see also aortic valve surgery infective endocarditis 826–827 rheumatic heart disease 755 valvular heart disease case studies 934–937 rheumatic see rheumatic heart disease vancomycin 824 VANQWISH trial 416, 511–512 vascular adhesion molecule-1 (VCAM-1) 226 acute coronary syndromes 404 vascular disease inflammatory 880 peripheral see peripheral vascular disease vascular injury platelet aggregation 362 venous thromboembolism 864 vascular remodelling, chronic, coronary restenosis etiology 383 vascular surgery 881–883 endovascular procedures 882–883 open reconstructions 881–882 infrainguinal 881–882 suprainguinal 881 preoperative cardiac evaluation 881 preoperative coronary revascularization 61, 66 VASIS trial 621, 629 vasoconstrictor drugs, syncope treatment 628 vasodilators aortic regurgitation 776–777 heart failure 662–664 acute therapy 662–664 documented value 664 long term therapy 662–663 survival effects 663 idiopathic dilated cardiomyopathy 694 mitral regurgitation 761, 763 vasoflux 466 Vasoflux International Trial for Acute Myocardial Infarction Lysis (VITAL) 466 vasopressin cardiac arrest 637–638 plasma 649 Vasovagal International Study (VASIS) group, vasovagal syndrome pacing 602

Vasovagal Pacemaker Study (VPS III), vasovagal syndrome pacing 602 Vasovagal Syncope and Pacing (SYNPACE) trial, vasovagal syndrome pacing 602 vasovagal syndrome 17, 621 pacing 601, 628 vegetables 149–150, 219 see also fruit and vegetables vegetarian diet 319 vegetations, echocardiographic detection 820–821 venography 865, 866 venous damage 864 venous stasis 864 venous thromboembolism (VTE) 864–876 see also pulmonary embolism diagnosis 865 estrogen replacement therapy and 254 clinical trials 254 observational studies 254 future research 872 natural history 864, 866 pathogenesis 864 postmenopausal hormone therapy 250–251, 255 prevalence 864 prevention 869–870 recurrence 871 risk factors 864, 865 treatment 870–872 pregnancy 872 venous thrombosis clinical assessment 865–866 deep see venous thromboembolism (VTE) diagnosis 865, 866, 869 clinical model 867 pregnancy 865 estrogen replacement therapy 254 natural history 864 pregnancy 867 recurrent 866, 867, 871 diagnosis 865 venous ultrasound imaging (VUI) 865, 866–867, 869 VENTAK-CHF, cardiac resynchronization therapy 606 ventilation perfusion scans 867–868 ventricular aneurysms Chagas’ disease 721 left 492–493 ventricular arrhythmias antiarrhythmic drug-induced 530–532, 540–541 Chagas’ disease 721, 728–729 deaths and risk stratification 577 heart failure 671 implantable cardioverter defibrillators 580–583 life-threatening 577–586 management case studies 925–930 non-sustained, antiarrhythmic drugs 578–579 pharmacologic v non-pharmacologic treatment 925–930 postinfarction 512–513 sustained antiarrhythmic drugs 578 implantable cardioverter defibrillators 581 syncope 623–624, 629 ventricular fibrillation (VF) aortic regurgitation 776 cardiac arrest 634 drug treatment 636, 637, 638 defibrillation 634 hypertrophic cardiomyopathy 710, 711

967


Index

ventricular fibrillation (VF) continued idiopathic, gene mutation associated 290 implantable cardioverter defibrillators 580, 581 inducible, implantable cardioverter defibrillators 582 management 496 postinfarction 496 prophylactic lidocaine 478–479 ventricular premature beats, postinfarction 496–497 ventricular proarrhythmia, antiarrhythmic drug-induced 540–541 ventricular septal defect (VSD), pregnancy 853 ventricular septal rupture, postinfarction 494–495 ventricular tachycardia (VT) antiarrhythmic agents 629 antiarrhythmic drug-induced 530–532, 540–541 aortic regurgitation 776 cardiac arrest 634 drug treatment 636, 637, 638 catecholaminergic (stress-induced), mutations associated 294 control during paroxysmal atrial fibrillation 530–532 hypertrophic cardiomyopathy 705, 706, 707, 710, 711 implantable cardioverter defibrillators 580, 581, 629 inducible, implantable cardioverter defibrillators 582 pacemaker insertion 591 postinfarction 496 surgical and catheter ablation 629 syncope 623–624, 629 ventriculectomy, left partial 727–728 verapamil AMI 482 atrial fibrillation 534 post-cardiac surgery 535 effort angina 331 hypertrophic cardiomyopathy 711, 712 idiopathic dilated cardiomyopathy 695–696 myocarditis 691 postinfarction angina 333 postinfarction patients 509 supraventricular tachycardia 568 unstable angina 333, 408 very low-calorie diets, weight loss in obesity 237 vesnarinone heart failure 668 myocarditis 691 Veterans Administration (VA), carotid endarterectomy trial 847 Veterans Administration Cooperative Study in Diabetes Mellitus (VACSDM) 166 Veterans Administration (VA) Coronary Artery Bypass Surgery Cooperative Study Group 339–340 Veterans Administration Study 409 aspirin post-MI 508 Veterans Affairs HDL Intervention Trial (VA-HIT) see VA-HIT trial

968

Veterans Affairs Non-Q wave Infarction Strategies in-hospital (VANQWISH) trial 416, 511–512 V-HeFT studies 663, 694 viral infections acute pericarditis 735 idiopathic dilated cardiomyopathy 674 myocarditis 681–684 vital capacity, heart failure and 648 VITAL trial 466 vitamin B6 224, 226 vitamin B12 224, 226 vitamin C (ascorbic acid) 219 antioxidant activity, cardiovascular disease relationship 314 epidemiological studies 222 randomized clinical trials 222 vitamin E (alpha-tocopherol) 219–221 antioxidant activity, cardiovascular disease relationship 314 effect on myocardial infarction (CHAOS study) 309–310 epidemiological studies 219–221 randomized clinical trials 220 volume, chronic expansion, in syncope 628 von Willebrand factor 362 thrombus formation 362, 406 VPS1 study 621 VPS III study, vasovagal syndrome pacing 602 waist circumference, obesity definition 231 walking capacity, in peripheral vascular disease 879–880 warfarin acute coronary syndromes 415–416 atrial fibrillation 548, 549–553 aspirin v 550–551 cardioversion 552–553, 553 decision analysis 53–58, 61, 65 hemorrhage risk 551–552 coronary restenosis prevention 378 intramuscular injection and 754 peripheral vascular disease 880 postinfarction left ventricular thrombi 493 postinfarction patients 508–509 pregnancy 860 prosthetic valve recipients 832–836 stroke prevention 845, 848 venous thromboembolism therapy 871 Warfarin-Aspirin Recurrent Stroke Study (WARSS) 845 Warfarin-Aspirin Symptomatic Intracranial Disease study 845 WARIS II trial, in postinfarction patients 509 WARSS trial 845 Washington Radiation for In-Stent Restenosis-Saphenous Vein Graft trial (WRIST-SVG) 382 Weibel–Palades bodies 379 weight body see obesity cardiac rehabilitation and 175 gain, antidiabetic agents 235 gain in children CHD development mechanisms 282 later CHD association 279–280

loss, blood pressure effects 149–150, 150 loss in obesity algorithm 233 coronary artery disease improvement 236 diabetic control improvement 235 dyslipidemia management 236 effect on coronary risk factors 234 effect on hypertension 233–234 goals 237 interventions for 237, 238 lifestyle intervention 237–238 trends in developing countries 95 weight training 171, 173–174 WEST (Women’s Estrogen for Stroke Trial) 251 cerebrovascular disease 253 venous thromboembolism 254 West Indies 271 West of Scotland Coronary Prevention study (WOSCOPS) 301, 302, 335 elderly 139 pleiotropic effects 132 statin cost effectiveness 141–142 WHI see Women’s Health Initiative (WHI) WISDOM 247 cerebrovascular disease 253 coronary heart disease prevention 248, 318, 319 Wolff–Parkinson–White syndrome 532, 569 AMPK gene mutation 290 catheter ablation 569 Wolff–Parkinson–White (WPW) syndrome 707 treatment, decision analysis 60, 63 women see also estrogen; gender differences; pregnancy cardiovascular disease 244–245 exercise training 176 lipid lowering therapy 140 mitral valve replacement 764 smoking prevalence 104–105 Women’s Estrogen for Stroke Trial (WEST) see WEST Women’s Health Initiative (WHI) 140, 247, 255–256 cerebrovascular disease 253 coronary heart disease prevention 318 elderly 140 venous thromboembolism 254 Women’s Intervention Study of long-Duration Oestrogen after the Menopause (WISDOM) see WISDOM work return to, exercise training and 174 strain, CHD risk and 189, 203–205, 206 World Wide Web 644 WOSCOPS see West of Scotland Coronary Prevention study (WOSCOPS) WRIST-SVG trial 382 xamoterol 667 xemilofiban, in acute coronary syndromes 413 X-linked inherited disorders 288–289 characteristic features 289 ZWOLLIE trial 445

Evidence based cardiology

Concise Cardiology: An Evidence-Based Handbook

Evidence-Based Nephrology (Evidence-Based Medicine)

Evidence-Based Respiratory Medicine (Evidence-Based Medicine)

Evidence-based Orthopedics (Evidence-Based Medicine)

Evidence-Based Rheumatology (Evidence-Based Medicine)

Evidence-Based Health Economics (Evidence-Based Medicine)

Evidence-Based Hypertension (Evidence-Based Medicine)

Evidence-Based Oncology (Evidence-Based Medicine)