Cardiovascular Disease in the Elderly:Third Edition, Revised and Expanded

We dedicate this third edition to our elderly patients and to the memory of Donald D. Tresch, M.D., who coedited with Dr. Aronow the first and second editions of this book. At the time of his death, Dr. Tresch was Professor of Medicine (Cardiology and Geriatrics) and Director of the Cardiology Fellowship Training Program at the Medical College of Wisconsin in Milwaukee. Dr. Tresch was an outstanding teacher, clinician, and researcher who made numerous important original research contributions to the field of cardiovascular disease in the elderly. Dr. Tresch was also a very compassionate person and perfect gentleman who was very encouraging and supportive to his patients, his students, and his colleagues. The geriatric cardiology community misses him and recognizes his passing as the loss of a valuable mentor, scholar, and friend. All beginnings are joyous. Our elderly patients show us that endings may also be joyous and they never fail to teach us that life, even with the struggles of aging and disease, may be full of meaning and joy.

iii

Series Introduction

Marcel Dekker, Inc., has focused on the development of various series of beautifully produced books in different branches of medicine. These series have facilitated the integration of rapidly advancing information for both the clinical specialist and the researcher. My goal as editor-in-chief of the Fundamental and Clinical Cardiology series is to assemble the talents of world-renowned authorities to discuss virtually every area of cardiovascular medicine. In the current monograph, Cardiovascular Disease in the Elderly: Third Edition, Revised and Expanded, Wilbert S. Aronow and Jerome L. Fleg have edited a much-needed and timely book. Future contributions to this series will include books on molecular biology, interventional cardiology, and clinical management of such problems as coronary artery disease and ventricular arrhythmias. Samuel Z. Goldhaber

v

Foreword

Publication of the third edition of Cardiovascular Disease in the Elderly reflects increasing appreciation of the close association between aging and cardiovascular disease (CVD), which rises exponentially as a cause of morbidity and mortality through middle and old age, accounting for nearly half of all adult deaths. All of the chapters in this edition reflect this sobering reality of increasing importance in our aging society, contributed by authors who are not only acknowledged experts in cardiovascular physiology and disease in general but also possess specific expertise in the emerging discipline of geriatric cardiology. The opening section on aging changes in the cardiovascular system introduces the subject appropriately. Its chapters document the declining homeostatic reserve of the aging heart and vasculature, whether measured at the cellular and organ system level (Chapter 1), non-invasively in the aging patient (Chapter 2), or at necropsy (Chapter 3). This decrease in physiological resilience, coupled with the myriad of comorbidities that are the norm in elderly patients, makes pharmacological management of CVD in both preventive and therapeutic modes an exercise of excruciating complexity and judgment (Chapter 4). The array of CVD risk factors, notably hypertension (Chapter 5), hyperlipidemia (Chapter 6), and diabetes (Chapter 7), and their increased incidence and prevalence in the elderly converge to place the older patient at exponentially escalating risk with advancing age (Chapter 8). Fortunately, however, each of these risk indices appears amenable to management with a combination of lifestyle and pharmacological interventions. Moreover, the evidence base supporting the efficacy of such strategies grows with each passing year. But to what age, we must now ask? Or, indeed, regardless of age? These age-related processes also demand appreciation of special subtleties in diagnosis (Chapter 10) and management of CVD in the elderly patient: angina (Chapter 11) and acute myocardial infarction (Chapter 12) may present in atypical (e.g., painless) fashion or as frank heart failure (Chapter 22). Thyroid disease, itself increasingly common in aging patients, may mask or aggravate CVD (Chapter 21). Arrhythmias abound in the absence or presence of ASCVD (Chapters 23–25). Stroke (Chapter 26), syncope (Chapter 27), and pulmonary embolism (Chapter 29) are ever-present risks to the health and independence in older persons. Yet bold interventions are increasingly undertaken in these arenas, with remarkable success in even the oldest patients—aortic valve replacement (Chapter 27) being a notable example. vii

viii

Foreword

Thus the third edition documents advances in all dimensions of vascular biology, cardiology, and cardiovascular surgery in elderly patients. Such remarkable progress, however, barely manages to keep pace with the rising age of patients cared for by cardiologists and indeed all physicians who care for older adults. Thus it is also encouraging that the current edition reflects increasing acceptance of the content of this book by an expanding readership and the emergence of a new subdiscipline of cardiology—geriatric cardiology. This development takes place as a growing number of clinicians identify themselves as geriatric cardiologists and the membership of the Society of Geriatric Cardiology grows apace. Although this sub-subspecialty has yet to achieve formal recognition through specialized Residency Review Committee (RRC)—accredited training and certification by the American Board of Internal Medicine (ABIM) as a domain of Added Qualifications, as has become the case with Cardiac Electrophysiology and Interventional Cardiology—this, too, may come to pass in the not-too-distant future as the base of knowledge and skills in geriatric cardiology continue to expand. These developments could not be more timely. Not only is the American (and global) population growing older with each passing year but also—in no small part attributable to the triumphs of modern preventive and therapeutic cardiology—the number of persons who do not succumb to premature disease but instead survive with clinical or subclinical CVD into advanced old age is increasing at an even greater pace! While age-adjusted cardiovascular mortality rates have declined over the past four decades, overall cardiovascular deaths have remained relatively constant. However, these deaths have occurred at an increasingly advanced average age. A by-product of this deferral of CVD death into old age has been a narrowing in the age-adjusted sex ratio in CVD mortality, which historically has disproportionately prematurely affected men. As a result, with more and more women surviving into old age (above 75 and 80), the number of women dying of CVD now equals or exceeds that of men, though still at a somewhat higher mean age. On the other hand, the average longevity of men is increasing faster in men than in women, and the gender gap in the surviving elderly has narrowed progressively over the past quarter century. This has translated directly into shorter average duration of widowhood, which exerts a major social and economic impact in an aging society. Thus by escaping CVD death in middle age, men are more likely to accompany their wives into old age, and the mutual support rendered to each other by aging spouses has been a major factor in decreasing utilization of nursing homes, historically populated primarily by widows. While of course this is enormously good news for aging couples and society at large, the implications of these trends for the patient clienteles of practicing physicians—not only cardiologists but all who care for adult patients—underscore the importance of books like this one. Thus as more and more patients survive or escape clinical CVD in middle age, more and more of those in their 80s and 90s will come under the care of both generalists and cardiologists. Perhaps then all (except perhaps pediatric) cardiologists will become (at least part time) geriatric cardiologists, and Cardiovascular Disease in the Elderly will become core reading for every cardiologist and a reference text for all physicians who share in the challenge to provide excellent geriatric medical care. As this upward shift in the average age of patients cared for by cardiologists continues over the next several decades, many of the special skills and insights of geriatricians—physicians who have received specialized additional training in geriatrics—will also become essential to the expert practice of cardiology. Such specialized training appropriately focuses upon comprehensive evaluation and management of the most complex,

Foreword

ix

challenging, and vulnerable patients, most of whom are in their 80s and 90s. What are the goals of such training? First, given the major burden of cardiovascular disease among the elderly patients they care for, geriatricians must acquire sophisticated appreciation of all of the issues presented in this volume at the level of the skilled generalist. However, geriatricians must also come to understand and manage with equivalent expertise all of the diseases, disabilities, and functional impairments that are common in patients of such advanced age, including cancer, renal disease, osteoporosis, osteoarthritis, pulmonary disease, infections, diabetes, dementia, depression; the ‘‘geriatric syndromes’’ of falls, incontinence, immobility, impaired homeostasis, susceptibility to iatrogenesis; and palliative and end-of-life care. In certain cases geriatricians will serve as consultants to other specialists (often cardiologists) in the evaluation and management of their frail elderly patients undergoing diagnostic evaluation or treatment in hospital or office settings. In other cases geriatricians will serve as primary care clinicians for frail and complicated patients, with consultation and in collaboration with specialists and subspecialists. In yet other cases, as supervisors and administrators, geriatricians will oversee or provide medical input to the care of groups of the most frail and dependent patients, especially those in long-term care settings. However, it cannot be overstated that in the future—as at present and in the past—the vast bulk of the care of elderly patients will be provided by generalists and specialists without specialized geriatric training at the fellowship level. Geriatricians have no monopoly on expert care of elderly patients, nor do they aspire to such. Stated in the most concrete and practical terms, it remains clear that the provision of geriatric medical care will extend well beyond the capacity of those who have received advanced geriatric training. Currently fellowship-trained geriatricians comprise less than 1% of practicing American physicians, a fraction that remains constant even as the numbers and average age of elderly patients continue to grow with every passing year. Moreover, less than 1% of current U.S. medical graduates pursue fellowship training in geriatrics—in spite of its having been compressed into a single year following completion of accredited training in internal medicine, family practice, or a subspecialty of internal medicine. Therefore, given the ‘‘age wave’’ of babyboomers now passing through the American population, certified geriatricians will be challenged to the maximum to leverage their professional efforts. This will be accomplished by focusing principally in the academic arena in advancing the aging knowledge base through research and broadly disseminating their expertise through the education in geriatrics and gerontology of all physicians, especially the other 99% of medical graduates (including future cardiologists as students, residents, and fellows), physicians who will in turn provide the vast bulk of direct geriatric medical care for the indefinite future. In their own generally limited academic practices, geriatricians will appropriately concentrate upon the care of the most frail and complex patients, those with the panoply of physical, psychological, and social problems enumerated above, including those on their final pathway toward death. This brings us back to the principal goal of this book: to prepare all physicians— including but by no means limited to cardiologists—to better prevent, diagnose, and treat CVD in their aging and elderly patients. As required for optimal management, such knowledge can be supplemented by reference to companion textbooks of geriatrics, internal medicine, psychiatry, or other relevant specialty disciplines. Inevitably this demands of the practitioner both command of the expanding relevant knowledge base in each of these domains and also skill in managing their complex interactions in the prototypical frail

x

Foreword

elderly patient. Fortunately the possibility of remaining abreast of the increasing knowledge base is facilitated by the exploding field of information technology, the management of which becomes an additional essential skill for all modern practitioners. Likewise fortunately, continuing growth in the diagnostic and therapeutic potential of interventions to deal with problems of their aging patients permits an optimistic, ‘‘can-do’’ attitude in the application of those skills. Thus two medical disciplines, cardiology and geriatrics, approach each another and begin to merge as the patients cared for in each domain become more alike in age, medical morbidity, and complexity. Here a challenge emerges to build bridges of collaboration and mutual education and training between the two disciplines, as well as to identify problems that remain to be solved through clinically informed and creative research. One gap in the essential knowledge base has been a product of the longstanding under-representation of the kinds of patients cared for by geriatricians in the studies that constitute the principal evidence base of the practicing cardiologist. To put this challenge another way, randomized clinical trials have by design generally excluded the kinds of frail patients cared for by geriatricians: the demented, the immobile, the incontinent, the unstable, those with life expectancy limited by competing conditions (e.g., cancer), those in precarious homeostatic balance at grave risk to rapid demise (classically illustrated by the patient with heart failure), and the dying. Furthermore, in clinical trials of cardiovascular interventions, patients with premature CVD have generally been over-represented (typically clinical trials have enrolled patients of average age 65 or below). Hence interpretation of the results of such studies has required significant extrapolation to the care of ‘‘truly geriatric’’ patients in their 80s and 90s (though to be fair, to date each such trial has suggested encouraging continued efficacy of interventions in participants above 75 and even beyond 80 years of age). Nevertheless, intervention trials specifically targeting the frail elderly are indicated with increasing urgency, lest such patients either fail to receive appropriate treatment because of a presumed lack of efficacy (e.g., as in the current under-use of statins) or receive inappropriate treatments of presumed (but specifically untested) efficacy based on trials of less representative elderly participants. What is the most evident link between cardiology and geriatrics, the point at which the cardiologist exclaims, ‘‘Aha, I get it now?’’ Here I would point to frailty as the defining syndrome that most challenges both fields. Frailty is becoming better defined—most recently according to the five criteria of ‘‘shrinking’’ (unintentional weight loss); exhaustion; weakness; slow walking; and low physical activity, as specifically investigated in the landmark Cardiovascular Health Study (CHS) [1]. Thus frail subjects meeting this definition can be selectively recruited for targeted CVD intervention studies, which can be designed to capitalize on increasing understanding of this syndrome at the most basic as well as the most practical level. In many studies it appears to be pathophysiologically linked with CVD and other common afflictions of the elderly through the process of inflammation, which seems ubiquitous among the frail. Can, for instance, anti-inflammatory agents (which include the statins) prevent or retard CVD in the frail elderly? Thus conjoint continuing challenges to education and training, research, and practice are shared by cardiologists and geriatricians. We predict with great confidence that gaps of communication and cooperation will progressively narrow with each passing year and that more effective models of collaborative research and training between these disciplines will be documented in each successive edition of Cardiovascular Disease in the Elderly.

Foreword

xi

REFERENCE 1.

Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, Seeman T, Tracy R, Kop WJ, Burke G, McBurnie MA. Frailty in older adults: evidence for a phenotype. J Gerontol 2001; 56A:M146–M156.

William R. Hazzard, M.D. Professor, Division of Gerontology and Geriatric Medicine, School of Medicine, University of Washington, Seattle, Washington, U.S.A.

Preface

Although only a few years have elapsed since the publication of the second edition of Cardiovascular Disease in the Elderly, the ever-accelerating pace of advances in the understanding and treatment of heart and vascular disorders dictates an update of that edition. Many important clinical trials addressing disorders common to older patients have been published since then. Approximately 34 million Americans are currently 65 years or older. Because of the ‘‘coming of age’’ of the post–World War II baby boomers, this number is expected to reach 75 million by 2040, representing more than 20% of the U.S. population. The implications of the demographic shift for the prevalence and socioeconomic burden of cardiovascular disease for our society are enormous. Conditions common in the elderly such as acute myocardial infarction, atrial fibrillation, and heart failure can be anticipated to reach epidemic proportions. Hopefully, improvements in prevention and treatment of antecedent risk factors and continued advances in the treatment of these disorders will ameliorate their impact. Because the elderly represent the majority of cardiovascular patients, it is crucial that practitioners be aware of advances in geriatric cardiology, particularly those that reduce morbidity and enhance survival and quality of life. The primary goal of this third edition is to provide clinicians with a comprehensive, easy-to-read information source incorporating the latest knowledge on the epidemiology, pathophysiology, and management of cardiovascular disease in older persons. A recurring theme throughout the new edition is that aging per se lowers the threshold for the development of various cardiac, vascular, and metabolic disorders and alters their clinical manifestations. For example, the marked age-associated increase in hypertension prevalence can be linked to a progressive stiffening of the arterial tree that begins in early adulthood. A second example is the striking reduction of early diastolic filling rate observed with normative aging, which appears to lower the threshold for development of diastolic heart failure. Understanding these interactions between the aging process and development of cardiovascular disease should help practitioners in managing their older cardiac patients. As with the previous editions, the target audience includes all clinicians who treat older cardiovascular patients, not just cardiologists. Indeed, a large proportion of these patients are managed by general internists, geriatricians, or family physicians, who we hope will find the current edition useful in their day-to-day decision-making. In this third edition, all of the prior chapters have been revised to incorporate the most current information available. Evidence-based cardiovascular medicine is emphasized, xiii

xiv

Preface

particularly recent multicenter studies with sizeable numbers of older patients. Sixteen of the 32 chapters have new authorship, allowing fresh perspectives on these topics. A chapter on diabetes and cardiovascular disease has been added to emphasize the major importance of this condition as a contributor to cardiovascular morbidity and mortality in the elderly. The prior chapter on electrocardiography has been incorporated into an expanded chapter on normal aging changes; similarly, information on mitral annular calcium is now included in the chapter on mitral valve disease (Chapter 18). Finally, the table of contents has been organized into parts, for easier location of specific topics. As in the two prior editions, all of the chapter contributors are widely acknowledged experts in geriatric cardiology. Their many years of personal experience allow them to summarize and synthesize their respective topics with unique insights that are highly beneficial to the reader. We greatly appreciate their expertise and diligence. We hope you, the reader, will find this third edition useful in your day-to-day care of older cardiovascular patients, helping them to enjoy both longer and fuller lives. Wilbert S. Aronow Jerome L. Fleg

Contents

Series Introduction Samuel Z. Goldhaber Foreword William R. Hazzard Preface Contributors

Part I

v vii xiii xix

Aging Changes in the Cardiovascular System

1. Normal Aging Changes of the Cardiovascular System Jerome L. Fleg and Edward G. Lakatta 2. Echocardiographic Measurements in Elderly Patients Without Clinical Heart Disease Julius M. Gardin and Cheryl K. Nordstrom

1

43

3. Morphological Features of the Elderly Heart William Clifford Roberts

69

4. Cardiovascular Drug Therapy in the Elderly William H. Frishman, Angela Cheng-Lai, and Wilbert S. Aronow

95

Part II Coronary Artery Disease: Risk Factors and Epidemiology 5. Systemic Hypertension in the Elderly William H. Frishman, Mohammed J. Iqbal, and Gregory Yesenski

131

6. Hyperlipidemia in the Elderly: When Is Drug Therapy Justified? John C. LaRosa

153

7. Diabetes Mellitus and Cardiovascular Disease in the Elderly Gabriel Gregoratos and Gordon Leung

163

8. Epidemiology of Coronary Heart Disease in the Elderly Pantel S. Vokonas and William B. Kannel

189 xv

xvi

Part III

Contents

Coronary Artery Disease

9. Pathophysiology of Coronary Artery Disease in the Elderly Roberto Corti, Antonio Ferna´ndez-Ortiz, and Valentin Fuster

215

10. Diagnosis of Coronary Artery Disease in the Elderly Wilbert S. Aronow and Jerome L. Fleg

251

11. Angina in the Elderly Wilbert S. Aronow and William H. Frishman

273

12. Therapy of Acute Myocardial Infarction Michael W. Rich

297

13. Management of the Older Patient After Myocardial Infarction Wilbert S. Aronow

329

14. Surgical Management of Coronary Artery Disease in the Elderly Edward A. Stemmer and Wilbert S. Aronow

353

15. Percutaneous Coronary Intervention in the Elderly Charles L. Laham, Mandeep Singh, and David R. Holmes, Jr.

389

16. Exercise Training in the Older Cardiac Patients Philip A. Ades

405

Part IV

Valvular Heart Disease

17. Aortic Valve Disease in the Elderly Wilbert S. Aronow and Melvin B. Weiss 18. Mitral Regurgitation, Mitral Stenosis, and Mitral Annular Calcification in the Elderly Melvin D. Cheitlin and Wilbert S. Aronow 19. Endocarditis in the Elderly Thomas C. Cesario and David A. Cesario

Part V

417

443 477

Cardiomyopathies and Heart Failure

20. Cardiomyopathies in the Elderly John Arthur McClung, Wilbert S. Aronow, Robert Belkin, and Michael Nanna

489

21. Thyroid Heart Disease in the Elderly Steven R. Gambert, Charles R. Albreht, and Myron Miller

511

22. Heart Failure Dalane W. Kitzman and Jerome L. Fleg

529

Contents

Part VI

xvii

Arrhythmias and Conduction Disorders

23. Supraventricular Tachyarrhythmias in the Elderly Wilbert S. Aronow and Carmine Sorbera

559

24. Ventricular Arrhythmias in the Elderly Wilbert S. Aronow and Carmine Sorbera

585

25. Bradyarrhythmias and Cardiac Pacemakers in the Elderly Anthony D. Mercando

607

Part VII

Cerebrovascular Disease

26. Cerebrovascular Disease in the Elderly Patient Jesse Weinberger

625

27. Evaluation of Syncope in the Elderly Patient Scott Hummel and Mathew S. Maurer

653

Part VIII

Miscellaneous Topics

28. Anticoagulation in the Elderly Jonathan L. Halperin

677

29. Acute Pulmonary Embolism in the Elderly Paul D. Stein and Fadi Kayali

695

30. Peripheral Vascular Disease in the Elderly Roy M. Fujitani, Ganesha B. Perera, Ian L. Gordon, and Samuel E. Wilson

707

31. Perioperative Cardiovascular Evaluation and Treatment of Elderly Patients Undergoing Noncardiac Surgery Rita A. Falcone, Roy C. Ziegelstein, and Lee A. Fleisher

765

32. Ethical Decisions and Quality of Life in Older Patients with Cardiovascular Disease Lofty L. Basta and Henry D. McIntosh

811

Index

825

Contributors

Philip A. Ades University of Vermont College of Medicine, Fletcher-Allen Health Care, Burlington, Vermont, U.S.A. Charles R. Albreht Sinai Hospital of Baltimore, Johns Hopkins University School of Medicine, Baltimore, Maryland, U.S.A. Wilbert S. Aronow Westchester Medical Center/New York Medical College, Valhalla, and Mount Sinai School of Medicine, New York, New York, U.S.A. Lofty L. Basta Clearwater Cardiovascular and Interventional Cardiovascular Consultants and Project GRACE, Clearwater, Florida, U.S.A. Robert Belkin Westchester Medical Center/New York Medical College, Valhalla, New York, U.S.A. David A. Cesario University of California, Irvine, Irvine, California, U.S.A. Thomas C. Cesario University of California, Irvine, Irvine, California, U.S.A. Melvin D. Cheitlin University of California, San Francisco, San Francisco, California, U.S.A. Angela Cheng-Lai Montefiore Medical Center, Bronx, New York, U.S.A. Roberto Corti University Hospital Zurich, Zurich, Switzerland Rita A. Falcone Johns Hopkins University School of Medicine, Baltimore, Maryland, U.S.A. Antonio Ferna´ndez-Ortiz San Carlos University Hospital, Madrid, Spain Jerome L. Fleg National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland, U.S.A. xix

xx

Contributors

Lee A. Fleisher Johns Hopkins University School of Medicine, Baltimore, Maryland, U.S.A. William H. Frishman Westchester Medical Center/New York Medical College, Valhalla, New York, U.S.A. Roy M. Fujitani University of California–Irvine Medical Center, Orange, California, U.S.A. Valentin Fuster Zena and Michael A. Wiener Cardiovascular Institute, The Mount Sinai School of Medicine, New York, New York, U.S.A. Steven R. Gambert Sinai Hospital of Baltimore, Johns Hopkins University School of Medicine, Baltimore, Maryland, U.S.A. Julius M. Gardin St. John Hospital and Medical Center, Detroit, Michigan, U.S.A. Ian L. Gordon University of California–Irvine Medical Center, Orange, California, U.S.A. Gabriel Gregoratos University of California, San Francisco, San Francisco, California, U.S.A. Jonathan L. Halperin The Zena and Michael Wiener Cardiovascular Institute, Mount Sinai Medical Center, New York, New York, U.S.A. David R. Holmes, Jr. Mayo Graduate School of Medicine, Mayo Clinic, Rochester, Minnesota, U.S.A. Scott Hummel College of Physicians & Surgeons, Columbia University, New York, New York, U.S.A. Mohammed J. Iqbal New York Medical College/Westchester Medical Center, Valhalla, New York, U.S.A. William B. Kannel Boston University Medical Center, Boston, Massachussettes, U.S.A. Fadi Kayali St. Joseph Mercy Oakland, Pontiac, Michigan, U.S.A. Dalane W. Kitzman Wake Forest University School of Medicine, Winston-Salem, North Carolina, U.S.A. Charles L. Laham Mayo Graduate School of Medicine, Mayo Clinic, Rochester, Minnesota, U.S.A. Edward G. Lakatta National Institute on Aging, National Institutes of Health, Bethesda, Maryland, U.S.A.

Contributors

xxi

John C. LaRosa State University of New York Health Science Center at Brooklyn, Brooklyn, New York, U.S.A. Gordon Leung University of California, San Francisco, California, U.S.A. Mathew S. Maurer College of Physicians & Surgeons, Columbia University, New York, New York, U.S.A. John Arthur McClung Westchester Medical Center/New York Medical College, Valhalla, New York, U.S.A. Henry D. McIntosh Lakeland, Florida, U.S.A. Anthony D. Mercando College of Physicians and Surgeons, Columbia University New York, and Albert Einstein College of Medicine and Montefiore Medical Center, Bronx, New York, U.S.A. Myron Miller Sinai Hospital of Baltimore, Johns Hopkins University School of Medicine, Baltimore, Maryland, U.S.A. Michael Nanna Albert Einstein College of Medicine, Bronx, New York, U.S.A. Cheryl K. Nordstrom St. John Hospital and Medical Center, Detroit, Michigan, U.S.A. Ganesha B. Perera University of California–Irvine Medical Center, Orange, California, U.S.A. Michael W. Rich Washington University School of Medicine and Barnes-Jewish Hospital, St. Louis, Missouri, U.S.A. William Clifford Roberts Baylor Heart and Vascular Institute, Baylor University Medical Center, Dallas, Texas, U.S.A. Mandeep Singh Mayo Graduate School of Medicine, Mayo Clinic, Rochester, Minnesota, U.S.A. Carmine Sorbera New York Medical College, Valhalla, New York, U.S.A. Paul D. Stein St. Joseph Mercy Oakland, Pontiac, Michigan, U.S.A. Edward A. Stemmer University of California at Irvine, Irvine, and Veterans Affairs Medical Center, Long Beach, California, U.S.A. Pantel S. Vokonas Boston University Medical Center, Boston, Massachusetts, U.S.A. Jesse Weinberger The Mount Sinai School of Medicine, New York, New York, U.S.A.

xxii

Contributors

Melvin B. Weiss Westchester Medical Center/New York Medical College, Valhalla, New York, U.S.A. Samuel E. Wilson University of California–Irvine Medical Center, Orange, California, U.S.A. Gregory Yesenski St. Barnabas Hospital–Cornell Medical School, Bronx, New York, U.S.A. Roy C. Ziegelstein Johns Hopkins University School of Medicine, Baltimore, Maryland, U.S.A.

1 Normal Aging of the Cardiovascular System Jerome L. Fleg National Heart, Lung, and Blood Institute Bethesda, Maryland, U.S.A.

Edward G. Lakatta National Institute on Aging Bethesda, Maryland, U.S.A.

INTRODUCTION The dramatic increase in life expectancy that has occurred during the past century in developed countries has markedly altered their population composition. Between 1900 and 1990, the U.S. population increased threefold while the subgroup z65 years increased tenfold. Currently, 13% of Americans are in this age group; by 2035, nearly one in four individuals will be z65 years old. The most rapidly growing segment of elders comprises those aged 85 years or older, whose numbers more than quadrupled between 1960 and 2000 (Table 1) [1]. Accompanying this aging of the population has been an exponential rise in the prevalence of cardiovascular (CV) disease. More than 80% of CV deaths occur in persons z65 years old. Despite these parallel trends, it must be emphasized that aging per se is not synonymous with the development of CV disease. This chapter will delineate those changes in the CV system that are thought to represent the effects of age per se in the absence of CV disease. This is by no means a simple task, given the multiple factors that blur their separation. It is important, however, to develop norms of CV structure and function in the elderly to facilitate the accurate diagnosis of CV disease in this age group. The enhanced CV risk associated with age indicates important interactions between mechanisms that underlie aging and those that underlie diseases. The nature of these interactions is complex and involves not only mechanisms of aging but also multiple defined and undefined (e.g., genetic) risk factors. The role of specific age-associated changes in CV structure and function has, and largely continues to be, unrecognized by those who shape medical policy and, until recently, not considered in most epidemiological studies of CV disease. Yet quantitative information on age-associated alterations in CV structure and function is essential to define and target the specific characteristics of CV aging that render it such a major CV risk factor. Such information is also required to differentiate between the 1

2 Table 1

Fleg and Lakatta Actual and Projected Growth of the Elderly American Population z65 Years

z85 Years

Year

Total population (all ages)

Number

% of total

Number

% of z65

1960 1980 1990 2000 2020 2040

179,323 226,546 248,710 274,634 322,742 369,980

16,560 25,550 31,079 34,709 53,220 75,233

9.2 11.3 12.5 12.6 16.5 20.6

929 2,240 3,021 4.259 6,460 13,552

5.6 8.8 9.7 12.3 12.1 18.0

Source: From Ref. 1.

limitations of an elderly person that relate to disease and those that are within expected normal limits.

METHODOLOGICAL ISSUES Multiple methodological issues must be addressed in the attempt to define ‘‘normal’’ aging. Because the population sample from which norms are derived will strongly influence the results obtained, it should be representative of the general population. For example, neither nursing home residents nor a seniors running club would yield an appropriate estimate of maximal exercise capacity that could be applied to the majority of elders. The degree of screening used to define a normal population can profoundly influence the results. In clinically healthy older adults, a resting electrocardiogram (ECG), echocardiogram, or exercise perfusion imaging will often identify silent CV disease, especially coronary artery disease (CAD). If several such screening tests are used, only a small proportion of the older population may qualify as normal, limiting the applicability of findings to the majority of elderly individuals. In addition, the inclusion limits chosen for body fatness, blood pressure, smoking status, and other constitutional or lifestyle variables will significantly influence the normal values for measuring CV variables. Additional methodological factors can affect the definitions of normal aging. Crosssectional studies, which study individuals across a wide age range at one time point, may underestimate the magnitude of age-associated changes because, in such studies, older normal persons represent ‘‘survival of the fittest.’’ True age-induced changes are better estimated by longitudinal studies, in which given individuals are examined serially over time. A reality of aging research, however, is that most data are derived from cross-sectional studies because they are easier to perform. Even longitudinal studies have their limitations— changes in methodology or measurement drift over time and development of disease in previously healthy persons. In addition, secular trends such as the downward drift in serum cholesterol or increasing obesity of Americans can alter age-related longitudinal changes. In the current chapter, emphasis will be given to data obtained from communitydwelling samples screened for the absence of clinical and, in some cases, subclinical CV disease and major systemic disorders. A sizeable portion of the data presented derive from the authors’ studies over the past three decades in community-based volunteers from the Baltimore Longitudinal Study of Aging (BLSA).

Normal Aging

3

UNSUCCESSFUL VASCULAR AGING AS THE ‘‘RISKY’’ COMPONENT OF AGING Intimal Medial Thickening Age-associated changes in arterial properties of individuals who are otherwise considered healthy may have relevance to the exponential increase in CV disease. Cross-sectional studies in humans have found that wall thickening and dilatation are prominent structural changes that occur within large elastic arteries during aging [2]. Postmortem studies indicate that aortic wall thickening with aging consists mainly of intimal thickening, even in populations with a low incidence of atherosclerosis [3]. Noninvasive measurements made within the context of several epidemiological studies indicate that the carotid wall intimal medial (IM) thickness increases nearly threefold between 20 and 90 years of age, which is also the case in BLSA individuals rigorously screened to exclude carotid or coronary arterial disease (Fig. 1A). It has been argued that the age-associated increase in IM thickness with aging in humans represents an early stage of atherosclerosis [4]. Indeed, excessive IM thickening at a given age predicts silent CAD (Fig. 1B). Since silent CAD (as defined in Fig. 1B) often progresses to clinical CAD, it is not surprising that increased IM thickness (a vascular endpoint) predicts future clinical CV disease. A plethora of other epidemiological studies of individuals not initially screened to exclude the presence of occult CV disease, have indicated that increased IM thickness is an independent predictor of future CV events. Note in Figure 1C that the degree of risk varies with the degree of vascular thickening, and that the risk gradation among quintiles of IM thickening is nonlinear, with the greatest risk occurring in the upper quintile [5]. Thus, those older persons in the upper quintile of IM thickness may be considered to have aged ‘‘unsuccessfully’’ or to have ‘‘subclinical’’ vascular disease. The potency of IM thickness as a risk factor in older individuals equals or exceeds that of most other, more ‘‘traditional,’’ risk factors. Note, however, in Figure 1B that the difference in IM thickness between those older and younger persons without evidence of CAD far exceeds the difference between older persons free of coronary disease, and those with disease. The phenomenon of age-dependent IM thickening has been noted in the absence of atherosclerosis, both in laboratory animals and in humans [3]. In other words, the ‘‘subclinical disease’’ of excessive IM thickening is not necessarily ‘‘early’’ atherosclerosis. Rather, ‘‘subclinical disease’’ is strongly correlated with arterial aging. Interpreted in this way, the increase in IM thickness with aging is analogous to the intimal hyperplasia that develops in aortocoronary saphenous vein grafts, which is independent of atherosclerosis, but predisposes for its later development [6]. Age-associated endothelial dysfunction, arterial stiffening, and arterial pulse pressure widening can also be interpreted in the same way. Combinations of these processes occurring to varying degrees determine the overall vascular aging profile of a given individual (i.e., the degree of ‘‘unsuccessful’’ vascular aging). In Western societies, additional risk factors (e.g., hypertension, smoking, dyslipidemia, diabetes, diet, or heretofore unidentified genetic factors) are required to interact with vascular aging (as described above) to activate a preexisting atherosclerotic plaque. According to this view, atherosclerosis that increases with aging is not a specific disease, but an interaction between atherosclerotic plaque and intrinsic features related to vascular aging modulated by atherosclerotic risk factors. Evidence in support of this view comes from studies in which an atherogenic diet caused markedly more severe atherosclerotic

4

Fleg and Lakatta

Figure 1 (A) The common carotid intimal-medial thickness in healthy BLSA volunteers as a function of age. (From Ref. 4) (B) Common carotid artery intimal-medial thickness as a function of age, stratified by coronary artery disease (CAD) status. CAD-1: subset with positive exercise ECG but negative thallium scan. CAD-2: subset with concordant positive exercise ECG and thallium scan. (From Ref. 4) (C) Common carotid intimal-medial thickness predicts future cardiovascular events in the Cardiovascular Health Study. (From Ref. 5.)

Normal Aging

Figure 1

5

Continued.

lesions in older versus younger rabbits and nonhuman primates despite similar elevations of serum lipids [7,8]. Hence, it is possible that atherosclerosis occurring at younger ages may be attributable not only to exaggerated traditional CV risk factors, but also to accelerated aging of the vascular wall. Of course, the traditional risk factors may themselves accelerate aging of the vascular wall. Studies in various populations with clinically defined vascular disease have demonstrated that pharmacological F lifestyle (diet, physical activity) interventions can retard the progression of IM thickening [9–14]. Arterial Pressure, Arterial Stiffness, and Endothelial Dysfunction Systolic, Diastolic, and Pulse Pressure Arterial pressure is determined by the interplay of peripheral vascular resistance and arterial stiffness; the former raises both systolic and diastolic pressure to a similar degree, whereas the latter raises systolic but lowers diastolic pressure. An age-dependent rise in average systolic blood pressure across adult age has been well documented (Fig. 2, lower right panel). In contrast, average diastolic pressure (Fig. 2, lower left panel) was found to rise until about 50 years of age, level off from age 50 to 60, and decline thereafter [15]. This decline in diastolic pressure is due to the impaired ability of the stiffer aorta to expand in systole and thereby augment diastolic pressure by the release of stored blood during diastole. Thus, pulse pressure (systolic minus diastolic), which is a useful hemodynamic indicator of conduit artery vascular stiffness, increases with age (Fig. 2, upper right panel). The age-dependent changes in systolic, diastolic, and pulse pressures are consistent with the notion that in younger people, blood pressure is determined largely by peripheral vascular resistance, while in older individuals it is determined to a greater extent by central conduit vessel stiffness.

6

Fleg and Lakatta

Figure 2 Change in blood pressure with age in the Framingham Study, stratified by systolic blood pressure at baseline visit. Thick lines represent entire study cohort; thin lines represent cohort after deaths; nonfatal myocardial infarction and heart failure have been excluded. (From Ref. 15.)

Because of the decline in diastolic pressure in older men and women in whom systolic pressure is increased, isolated systolic hypertension emerges as the most common form of hypertension in individuals over the age of 50 [15]. Isolated systolic hypertension, even when mild in severity, is associated with an appreciable increase in cardiovascular disease risk [16,17]. Based on long-term follow-up of middle-aged and older subjects, however, Framingham researchers have found pulse pressure to be a better predictor of coronary disease risk than systolic or diastolic blood pressures [18]. A subsequent Framingham investigation found that pulse pressure was especially informative of coronary risk in older subjects (Fig. 3) [19]. This is because of the ‘‘J’’- or ‘‘U’’-shaped association between diastolic pressure and coronary risk. Thus, consideration of the systolic and diastolic pressures jointly (which are reflected in pulse pressure) are preferable to consideration of either value alone.

Normal Aging

7

Figure 3 Influence of pulse pressure on risk of coronary heart disease (CHD) for given systolic blood pressure strata in the Framingham study. All estimates are adjusted for age, sex, body mass index, cigarettes smoked per day, glucose intolerance, and total cholesterol/HDL ratio. (From Ref. 19.)

As the definition of ‘‘disease’’ continues to evolve, we may find that many subjects who were formerly thought to be healthy may no longer be considered free from disease. For example, if systolic pressure z140 mmHg is now considered to be hypertension, and if hypertension is considered to be a disease, then individuals with a systolic pressure between 140 and 160 mmHg, who a decade ago were thought to be disease- free, are now identified as having CV disease. Large studies have shown that individuals who manifest modest elevations in systolic and pulse pressures are more likely to develop clinical disease or die from it [16,17]. There are compensatory mechanisms to normalize blood pressure that fail with advancing age. For example, endothelial dysfunction becomes apparent by about the sixth decade, approximately the time when pulse pressure begins to rise appreciably. Thus, impaired endothelial function with aging may be a mechanism that not only permits arterial pulse pressure to rise but also underlies the importance of pulse pressure as a risk factor for cardiovascular events and death, even after accounting for systolic pressure [15,18–27]. Abnormalities in endothelial function have also been identified as one of the earliest pathophysiological manifestations of atherosclerosis, diabetes, and hypertension [28]. The age-associated increase in IM thickening and endothelial dysfunction are accompanied by both luminal dilatation and a reduction in arterial compliance or distensibility with an increase in vessel stiffness [2]. Pulse-wave velocity (PWV), a relatively convenient and noninvasive index of vascular stiffening, increases with age in both men and women (Fig. 4B) in parallel with systolic blood pressure (Fig. 4A). PWV is determined in part by the intrinsic stress/strain relationship (stiffness) of the vascular wall, and by the mean arterial pressure. Increased PWV has traditionally been linked to structural alterations in the vascular media, such as increased collagen content, covalent cross-linking of the collagen, reduced elastin content, elastin fracture, and calcification. Prominent age-

8

Fleg and Lakatta

Figure 4 Bar graphs of systolic blood pressure, aortofemoral pulse wave velocity, and augmentation index in young (mean age 29 years), older (mean age 67 years) sedentary BLSA volunteers, and older (mean 67 years) endurance athletes. The age-associated increase in each of these arterial stiffness indices is attenuated in the athletes. (A) Older sedentary differ from young sedentary at p < 0.01; (B) older sedentary differ both other groups at p 0.05) among 143 normal subjects aged 20 to 80 [35]. Among 85 of 117 healthy volunteers with adequate tracings, Klein et al. found that those aged 50 years and older had increased pulmonary venous peak systolic flow velocity (71 F 9 vs. 48 F 9 cm/s; p < 0.01), decreased diastolic flow velocity (38 F 9 vs. 50 F 10 cm/s; p < 0.01), and increased peak atrial reversal flow velocity (23 F 4 vs. 19 F 4 cm/s; p < 0.01) compared with those younger than 50 years (36). Klein et al. later reported the difference in duration of pulmonary venous atrial reversal flow and transmitral A-wave flow, previously related to LV end-diastolic pressures among patients with heart disease [37], to be unrelated to aging in 72 healthy volunteers aged 23 to 84 years (r =
0.16) [38].

60

Gardin and Nordstrom

Figure 17 Data for mitral A/E ratio is depicted versus age in years. Note the positive slope of the relationship, mitral A/E ratio = 0.016 (age) + 0.13. (Reprinted from Ref. 25, with permission.)

EPIDEMIOLOGICAL STUDIES IN THE ELDERLY The Framingham study has shown by M-mode echocardiography that (1) LV hypertrophy is a powerful, independent predictor for mortality and morbidity from coronary heart disease, and (2) increased left atrial dimension is associated with an increased risk of stroke [39–41]. However, no previous multicenter population-based study has evaluated specifically in the elderly predictive value for coronary heart disease and stroke of echocardiographic imaging and Doppler techniques.

Figure 18

Natural history of LV filling. (Reprinted from Ref. 34, with permission.)

Echocardiographic Measurements

61

The Cardiovascular Health Study (CHS) is a multiyear prospective epidemiological study of 5201 men and women older than 65 years recruited from four U.S. field centers: Davis, California; Hagerstown, Maryland; Winston-Salem, North Carolina; and Pittsburgh, Pennsylvania [42]. The main objectives of incorporating echocardiography were to determine whether echocardiographic measurements, or changes in these measurements, are (1) correlated with traditional risk factors for coronary heart disease and stroke and (2) independent predictors of morbidity and mortality from coronary heart disease and stroke. Echocardiographic measurements obtained include those related to global and segmental LV systolic and diastolic structure and function and left atrial and aortic root dimension [43]. For each subject, a baseline two-dimensional (2-D), M-mode (2-D directed), and Doppler echocardiogram was recorded on super-VHS tape using a standard protocol and equipment. All studies were sent to a reading center (University of California, Irvine), where images were digitized and measurements made using customized computer algorithms. M-mode measurements were made according to American Society of Echocardiography convention [44]. LV mass was derived from the formula described by Devereux et al. LV mass ðgÞ ¼ 0:80  1:04½ðVSTd þ LVIDd þ PWTdÞ3
ðLVIDdÞ3 þ 0:6; where VSTd is ventricular septal thickness at end-diastole, LVIDd is LV internal dimension at end-diastole, and PWTd is LV posterior wall thickness at end-diastole [45]. Calculated data and images were stored on optical disks to facilitate retrieval and future comparisons in longitudinal studies. Quality-control measures included standardized training of echocardiography technicians and readers, technician observation by a trained echocardiographer, periodic blind duplicate readings with reader review sessions, phantom studies, and quality-control audits (43).

Left Ventricular Mass A number of initial cross-sectional associations with baseline echocardiographic data have been described in the Cardiovascular Health Study. M-mode measurements (2-D directed) of LV mass could not be made in 34% of CHS participants, and this was highly related to age (29% in the 65- to 69-year vs. 50% in the 85+ age group; p < 0.001), white race, male gender, and history of hypertension, diabetes, or CHD. LV mass was found to be significantly higher in men than women, even after multivariate adjustments, and modestly increased with aging [46,47]. LV mass increased less than 1 g per year increase in age for both men and women. Of interest, across all CHS age subgroups, the difference in weightadjusted LV mass by sex was greater in magnitude than the difference related to clinical CHD (Fig. 19). This may relate to the well-known relatively high prevalence of subclinical manifestations of cardiac disease (e.g., coronary disease) in elderly individuals without clinical disease. After adjustment for traditional CVD risk factors, baseline LV mass was also related to 6- to 7-year incidence of CHD, CHF, and stroke among CHS participants initially free of prevalent disease (48). The highest quartile of LV mass conferred a hazards ratio of 3.36 for incident CHF, compared with the lowest quartile. Further, eccentric and concentric LV hypertrophy conferred higher hazards ratios compared with normal LV geometry for both incident CHF (2.95 and 3.32, respectively) and CHD (2.05 and 1.61, respectively). In comparison, the Framingham study, in a 4-year follow-up of its original cohort (mean age

62

Gardin and Nordstrom

Figure 19 Cardiovascular Health Study data: Bar graph shows weight-adjusted mean left ventricular (LV) mass displayed by sex and disease status group across 5-year age intervals. Data are computed using the lower age end point and the mean weight for each age category (65 to 69, 70 to 74, 75 to 79, 80 to 84, 85+ years). Weight-adjusted LV mass was significantly associated with sex, disease status, and age. Of interest, within each age group, the magnitude of the sex effect exceeded that of the effect of the disease (e.g., clinical coronary heart disease [CHD]). HTN indicates hypertension.

69 years) found even higher relative risks (7.8 in men and 3.4 in women) for CHD events in the highest versus the lowest quartile of risk factor-adjusted LV mass [40]. Left Ventricular Wall Motion In this regard, 4.3% of participants with hypertension but no clinical heart disease, and 1.9% of participants with neither clinical heart disease nor hypertension had LV segmental wall motion abnormalities, suggesting silent coronary disease [47]. Multivariate analyses revealed male sex and presence of clinical CHD (both p < 0.001) to be independent predictors of LV akinesis or dyskinesis. Subclinical Disease A new method of classifying subclinical disease at the baseline examination in the Cardiovascular Health Study included measures of ankle–brachial blood pressure, carotid artery stenosis and wall thickness, ECG and echocardiographic abnormalities, and positive response to the Rose Angina and Claudication Questionnaire (see Table 2) [49]. Participants were followed for an average of 2.4 years (maximum, 3 years). For participants without evidence of clinical cardiovascular disease at baseline, the presence of subclinical disease compared with no subclinical disease was associated with a significantly increased risk of incident total coronary heart disease including CHD deaths and nonfatal MI and angina pectoris for both men and women. For individuals with subclinical disease, the increased risk of total coronary heart disease was 2.0 for men and 2.5 for women, and the increased risk of total mortality was 2.9 for men and 1.7 for women. The increased risk changed little after adjustment for other risk factors, including lipoprotein levels, blood pressure, smoking, and diabetes. Consequently, the measurement of subclinical disease provides an approach for identifying high-risk older individuals who may be candidates for more active intervention to prevent clinical disease.

Echocardiographic Measurements

63

Table 2 Criteria for Clinical and Subclinical Disease in the Cardiovascular Health Study Clinical disease criteria Atrial fibrillation or pacemaker History of intermittent claudication or peripheral vascular surgery History of congestive heart failure History of stroke, transient ischemic attack, or carotid surgery History of coronary artery bypass graft surgery or percutaneous transluminal coronary angioplasty History of angina or use of nitroglycerin History of myocardial infarction

Subclinical disease criteria Ankle-arm index l 0.9 mmHg Internal carotid wall thickness >80th percentile Common carotid wall thickness >80th percentile Carotid stenosis >25% Major ECG abnormalitiesa Abnormal ejection fraction on echocardiogram Abnormal wall motion on echocardiogram Rose questionnaire claudication positive Rose questionnaire angina positive

a

According to Minnesota Code, ventricular conduction defects (7-1, 7-2, 7-4); major Q/QS wave abnormalities (11, 1-2); left ventricular hypertrophy (high-amplitude R waves with major or minor ST-T abnormalities) (3-1, 3-3, and 4-1 to 4-3 or 5-1 to 5-3); isolated major ST/T wave abnormalities (4-1, 4-2, 5-1, 5-2). Source: From Ref. 37, with permission of the author and the American Heart Association.

‘‘Healthy’’ Subgroup By excluding CHS participants with either LV ejection fraction or wall motion abnormalities from the group of participants with neither clinical heart disease (including CHD) nor hypertension, a subgroup that was apparently free of clinical heart disease and hypertension was defined. The ‘‘healthy’’ subgroup consisted of 516 men and 773 women, with 339 and 569, respectively, having available echo measurements of LV mass. Preliminary analysis indicated that after adjustment for weight, no adjustment for height was necessary. Likewise, after adjustment for weight, age had only a modest effect on LV mass. Therefore, for simplicity, the reference equations were not expressed as a function of age. There was no evidence of an interaction between sex and weight. Since the baseline CHS cohort race was not a significantly independent predictor of LV mass, the reference equations are not race-specific. The expected LV mass (g) derived from the CHS healthy subgroup can be calculated from the following equations: Men: 16:6  ½Weight ðkgÞ0:51 Women: 13:9  ½Weight ðkgÞ0:51 If the ratio of observed-to-expected LV mass is between 0.69 and 1.47, the patient’s LV mass should not be considered larger than expected given his or her weight. This prediction procedure classified 28% of the men and 18% of the women with clinical CHD as outside the range of expected LV mass measurements. Exclusions of obese (n = 220) participants from this healthy group had a negligible effect on the reference equations. Left Ventricular Diastolic Filling Pulsed Doppler transmitral flow velocities were analyzed as part of the baseline examination in the Cardiovascular Health Study [43]. Early diastolic LV Doppler (transmitral)

64

Gardin and Nordstrom

peak filling velocity decreased, and peak late diastolic (atrial) velocity increased with increasing age in multivariate analyses (all p < 0.001) [50]. Early and late diastolic peak filling velocities were both significantly higher in women than in men, even after adjustment for body surface area (or height and weight). In multivariate models in the entire cohort and a healthy subgroup (n = 703), gender, age, heart rate, and blood pressure were most strongly related to early and late diastolic transmitral peak velocities. Early and late diastolic peak velocities both increased with increases in systolic blood pressure, and decreased with increases in diastolic blood pressure ( p < 0.001). Doppler transmitral velocities were compared among health status subgroups. In multiple regression models adjusted for other covariates, and in analysis-of-variance models examining differences across subgroups adjusted only for age, the subgroup with CHF had the highest early diastolic peak velocities (50). All clinical disease subgroups had higher late diastolic peak velocities than did the healthy subgroup, with CHF and hypertensive subgroups having the highest age-adjusted means. The hypertensive subgroup had the lowest ratio of early-to-late diastolic peak velocity, and men with CHF had the highest ratio. Borderline and definite isolated systolic hypertension were positively associated with LV mass ( p < 0.001) and with increases in transmitral late peak flow velocity and decreases in the ratio of early-to-late diastolic peak flow velocity (51). These findings are consistent with previous reports that hypertensive subjects exhibit an abnormal relaxation pattern, whereas CHF patients develop a pattern suggestive of an increased early diastolic LA–LV pressure gradient. In summary, echocardiographic imaging and Doppler flow recording have provided convincing evidence of structural and functional cardiac changes related to aging. These noninvasive ultrasound techniques have provided important insights into the cardiovascular epidemiology of aging, as well as the clinical evaluation of elderly patients with suspected cardiovascular disease.

ACKNOWLEDGMENT The authors acknowledge the expert assistance of Kathleen Steiner in the preparation of this manuscript.

REFERENCES 1. 2. 3.

4. 5. 6.

Feigenbaum H. Echocardiography. 5th ed. Philadelphia: Lea & Febiger, 1994:1–133. Talano JV, Gardin JM. Textbook of Two-Dimensional Echocardiography. New York: Grune & Stratton, 1983:1–64,325–355. Gardin JM, Henry WL, Savage DD, Ware JH, Burn C, Borer JS. Echocardiographic measurements in normal subjects: evaluation of an adult population without clinically apparent heart disease. J Clin Ultrasound 1979; 7:439–447. Pombo JE, Troy BL, Russell RD Jr, Left ventricular volumes and ejection fraction by echocardiography. Circulation 1971; 43:480–490. Henry WL, Gardin JM, Ware JH. Echocardiographic measurements in normal subjects from infancy to old age. Circulation 1980; 62:1054–1061. Gardin JM, Savage DD, Ware JH, Henry WL. Effect of age, sex and body surface area on echocardiographic left ventricular wall mass in normal subjects. Hypertension 1987; 9:1136–1139.

Echocardiographic Measurements 7. 8. 9. 10. 11. 12. 13. 14.

15. 16.

17.

18. 19.

20.

21.

22.

23.

24.

25. 26.

27.

65

Roberts WC, Perloff JK. Mitral valve disease. A clinicopathologic survey of the conditions causing the mitral valve to function abnormally. Ann Intern Med 1972; 77:939–975. Krovetz J. Age-related changes in size of the aortic valve annulus in man. Am Heart J 1975; 90:569–574. Demir M, Acarturk E. Clinical characteristics influence aortic root dimension and blood flow velocity in healthy subjects. Angiology 2001; 52:457–461. Derman U. Changes of the mitral echocardiogram with aging and the influence of atherosclerotic risk factors. Atherosclerosis 1972; 15:349. Borer JS, Bacharach SL, Green MV, et al. Evaluation of clinical utility of real-time computerbased radionuclide cineangiography. Proc IEEE Comp Soc 1977. Strandell T. Heart rate, arterial lactate concentration and oxygen uptake during exercise in old men compared with young men. Acta Physiol Scan 1964; 60:197–213. Brandfonbrener M, Landowne M, Shock NW. Changes in cardiac output with age. Circulation 1955; 12:557–566. Granath A, Strandell T. Relationships between cardiac output, stroke volume and intracardiac pressures at rest and during exercise in supine position and some anthropometric data in healthy old men. Acta Med Scand 1964; 176:447–466. Gerstenblith G, Frederikson J, Yin FCP, et al. Echocardiographic assessment of a normal adult aging population. Circulation 1977; 56:273–278. Shub C, Klein AL, Zachariah PK, Bailey KR, Tajik AJ. Determination of left ventricular mass by echocardiography in a normal population: effect of age and sex in addition to body size. Mayo Clin Proc 1994; 69:205–211. Savage DD, Levy D, Dannenberg AL, Garrison RJ, Castelli WP. Association of echocardiographic left ventricular mass with body size, blood pressure and physical activity (The Framingham Study). Am J Cardiol 1990; 65:371–376. Dannenberg AL, Levy D, Garrison RJ. Impact of age on echocardiographic left ventricular mass in a healthy population (The Framingham Study). Am J Cardiol 1989; 64:1066–1068. Gardin JM, Iseri LT, Elkayam U, Tobis J, Childs W, Burns CS, Henry WL. Evaluation of dilated cardiomyopathy by pulsed Doppler echocardiography. Am Heart J 1983; 106:1057– 1065. Elkayam U, Gardin JM, Berkley R, Hughes C, Henry WL. The use of Doppler flow velocity measurements to assess the hemodynamic response to vasodilators in patients with heart failure. Circulation 1983; 67:377–383. Gardin JM, Davidson DM, Rohan MK, Butman S, Knoll M, Garcia R, Dubria S, Gardin SK, Henry WL. Relationship between age, body size, gender and blood pressure and Doppler flow measurements in the aorta and pulmonary artery. Am Heart J 1990; 113:101–1093. Gardin JM, Burns CS, Childs WJ, Henry WL. Evaluation of blood flow velocity in the ascending aorta and main pulmonary artery of normal subjects by Doppler echocardiography. Am Heart J 1984; 107:310–319. Rodeheffer RJ, Gersenblith G, Becker LC, Fleg JL, Weisfeldt ML, Lakatta EG. Exercise cardiac output is maintained with advancing age in healthy human subjects: cardiac dilation and increased stroke volume compensate for a diminished heart rate. Circulation 1984; 69:203–213. Lazarus M, Dang TY, Gardin JM, Allfie A, Henry WL. Evaluation of age, gender, heart rate and blood pressure changes and exercise conditioning on Doppler measured aortic blood flow acceleration and velocity during upright treadmill testing. Am J Cardiol 1988; 62:439–443. Rushmer RF. Initial ventricular impulse: a potential key to cardiac evaluation. Circulation 1964; 29:268–283. Noble M, Tenchard D, Guz A. Left ventricular ejection in conscious dog. I. Measurement and significance of the maximum acceleration of blood from the left ventricle. Circ Res 1966; 19:139–147. Jewitt D, Gabe I, Millis C, Maurer B, Thomas M, Shillingford J. Aortic velocity and acceleration measurements in the assessment of coronary heart disease. Eur J Cardiol 1974; 1:299–305.

66

Gardin and Nordstrom

28. Miyatake K, Okamoto M, Kinishita N, et al. Augmentation of atrial contribution to left ventricular inflow in aging as assessed by intracardiac Doppler flowmetry. Am J Cardiol 1984; 53:586–589. 29. Gardin JM, Rohan MK, Davidson DM, Dabestani A, Sklansky M, Garcia R, Knoll ML, White DB, Gardin SK, Henry WL. Doppler transmitral flow velocity parameters: relationship between age, body surface area, blood pressure and gender in normal subjects. Am J Noninvas Cardiol 1987; 1:3–10. 30. Dabestani A, Johnston WD, Tobis JM, et al. Relation between Doppler transmitral diastolic flow and left ventricular filling [abstr]. J Am Coll Cardiol 1984; 3:612. 31. Rokey R, Kuo LK, Zoghbi WA, et al. Determination of parameters of left ventricular diastolic filling with pulsed Doppler echocardiography: comparison with cineangiography. Circulation 1985; 71:543–550. 32. Mantero A, Gentile F, Azzollini M, Barbier P, Beretta L, Casazza F, Corno R, Faletra F, Giagnoni E, Gualtierotti C, Lippolis A, Lombroso S, Mattioli R, Morabito A, Ornaghi M, Pepi M, Pierini S, Todd S. Effect of sample volume location on Doppler-derived transmitral inflow velocity values in 288 normal subjects 20 to 80 years old: an echocardiographic, twodimensional color Doppler cooperative study. J Am Soc Echocardiogr 1998; 11:280–288. 33. Gardin JM, Rohan MK, Davidson D, Dabestani A, Sklansky MI, Garcia R, Knoll ML, White DB, Henry WL. Doppler transmitral flow velocity parameters: relationship between age, body surface area, blood pressure and gender in normal subjects. Am J Noninvas Cardiol 1987; 1:3– 10. 34. Appleton CP, Hatle LK. The natural history of LV filling abnormalities: assessment by twodimensional and Doppler echocardiography. Echocardiography 1992; 9:437–457. 35. Gentile F, Mantero A, Lippolis A, Ornaghi M, Azzollini M, Barbier P, Beretta L, Casazza F, Corno R, Faletra F, Giagnoni E, Gualtierotti C, Lombroso S, Mattioli R, Morabito A, Pepi M, Todd S, Pezzano A. Pulmonary venous flow velocity patterns in 143 normal subjects aged 20 to 80 years old. An echo 2D colour Doppler cooperative study. Eur Heart J 1997; 18:148–164. 36. Klein AL, Burstow DJ, Tajik AJ, Zacharia PK, Bailey KR, Seward JB. Effects of age on left ventricular dimensions and filling dynamics in 117 normal persons. Mayo Clin Proc 1994; 69: 212–224. 37. Rossvoll O, Hatle LK. Pulmonary venous flow velocities recorded by transthoracic Doppler ultrasound: relation to left ventricular diastolic pressures. J Am Coll Cardiol 1993; 21:1687– 1696. 38. Klein AL, Abdalla I, Murray RD, Lee JC, Vandervoort P, Thomas JD, Appleton CP, Tajik AJ. Age independence of the difference in duration of pulmonary venous atrial reversal flow and transmitral A-wave flow in normal subjects. J Am Soc Echocardiogr 1998; 11:458–465. 39. Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP. Prognostic implications of echocardiographically determined left ventricular mass in the Framingham Heart Study. N Engl J Med 1990; 322:1561–1566. 40. Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP. Left ventricular mass and incidence of coronary heart disease in an elderly cohort: The Framingham Heart Study. Ann Intern Med 1989; 110:101–107. 41. Benjamin E, Levy D, Plehn J, Belanger A, D’Agostino R, Wolf P. Left atrial size: an independent risk factor for stroke [abstr]. Circulation 1989; 80(suppl 2):615. 42. Fried L, Borhani N, Enright P, Furberg C, Gardin JM, Kronmal R, Kuller L, Manolio T, Mittelmark M, Newman A, O’Leary D, Psaty B, Rautaharju P, Tracy R, Weiler P, for the CHS Collaborative Research Group. Cardiovascular Health Study: design and rational. Ann Epidemiol 1991; 1:263–276. 43. Gardin JM, Wong ND, Bommer W, Klopfenstein HS, Smith V-E, Tabatznik B, Siscovick D, Lobodzinski S, Anton-Culver H, Manolio TA. Echocardiographic design of a multicenter investigation of free-living elderly subjects: The Cardiovascular Health Study. J Am Soc Echocardiogr 1992; 5:63–72.

Echocardiographic Measurements

67

44. Sahn DJ, DeMaria A, Kisslo J, Weyman A, for the Committee on M-mode Standardization of the American Society of Echocardiography. Recommendations regarding quantitation in Mmode echocardiography: results of a survey of echocardiographic methods. Circulation 1978; 58:1072–1083. 45. Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I, Reichek N. Echocardiographic assessment of left ventricular hypertrophy: comparison with necropsy findings. Am J Cardiol 1986; 57:450–458. 46. Gardin JM, Siscovick D, Anton-Culver H, Smith V-E, Klopfenstein S, Bommer W, Fried LP, O’Leary D, Manolio TA. Age and gender relations of left ventricular mass and wall motion in elderly subjects with no clinical heart disease: The Cardiovascular Health Study (CHS) [abstr]. Circulation 1992; 84(4):I–672. 47. Gardin JM, Siscovick D, Anton-Culver H, Lynch JC, Smith V-E, Klopfenstein HS, Bommer WJ, Fried L, O’Leary D, Manolio TA. Sex, age and disease affect echocardiographic left ventricular mass and systolic function in the free-living elderly: The Cardiovascular Health Study. Circulation 1995; 91:1739–1748. 48. Gardin JM, McClelland R, Kitzman D, Lima JA, Bommer W, Klopfenstein HS, Wong ND, Smith VE, Gottdiener J. M-mode echocardiographic predictors of six- to seven-year incidence of coronary heart disease, stroke, congestive heart failure, and mortality in an elderly cohort (the Cardiovascular Health Study). Am J Cardiol 2001; 87:1051–1057. 49. Kuller LH, Shemanski L, Psaty BM, Borhani NO, Gardin JM, Haan MN, O’Leary DH, Savage PJ, Tell GS, Tracy R. Subclinical disease as an independent risk factor of cardiovascular disease. Circulation 1995; 92:720–726. 50. Gardin JM, Arnold AM, Bild DE, Smith V-E, Lima JAC, Klopfenstein HS, Kitzman DW. Left ventricular diastolic filling in the elderly: The Cardiovascular Health Study. Am J Cardiol 1998; 82(3):345–351. 51. Psaty BM, Furbert CD, Kuller LH, Borhani NO, Rautaharju PM, O’Leary DH, Bild DE, Robbins J, Fried LP, Reid CL. Isolated systolic hypertension and subclinical cardiovascular disease in the elderly. Initial findings from the Cardiovascular Health Study. JAMA 1992; 268:1287–12913.

3 Morphological Features of the Elderly Heart William Clifford Roberts Baylor Heart and Vascular Institute, Baylor University Medical Center, Dallas, Texas, U.S.A.

INTRODUCTION As in any body tissue or organ, changes take place in the cardiovascular system as life progresses. Some of these changes allow easy identification of the very elderly heart when examining autopsy cardiac specimens as unknowns. The ‘‘normal’’ elderly heart has relatively small ventricular cavities and relatively large atria and great arteries. The ascending aorta and left atrium, in comparison with the relatively small left ventricular cavity, appear particularly large. The coronary arteries increase in both length and width; the former, particularly in association with the decreasing size of the cardiac ventricles, results in arterial tortuosity. (The young river is straight and the old one winding.) The leaflets of each of the four cardiac valves thicken with age, particularly the atrioventricular valves; these have a smaller area to occupy in the ventricles because of the diminished size of the latter. Histological examination discloses large quantities of lipofuscin pigment in myocardial cells; some contain mucoid deposits (‘‘mucoid degeneration’’). These changes appear to affect all population groups of elderly individuals regardless of where they reside on the earth or their level of serum lipids. An elevated systemic arterial pressure appears to both accelerate and amplify these normal expected cardiac changes of aging. Both the aorta and its branches and the major pulmonary arteries and their branches enlarge with age. Because the enlargement is in both the longitudinal and the transverse dimensions, the aorta, as the coronary arteries, tends to become tortuous. This process is further amplified as the vertebral bodies become smaller and the height becomes somewhat shorter. The major pulmonary arteries appear to be too short and have too low a pressure to dilate longitudinally. The amount of information at necropsy in very elderly persons is relatively sparse. In 1983, Waller and Roberts [1] described some clinical and necropsy findings in 40 American patients aged 90 years and over. In 1988, Lie and Hammond [2] reported findings at necropsy in 237 patients aged 90 and older. In 1991, Gertz and associates [3] described composition of coronary atherosclerotic plaques in 18 persons z90 years of age. In 1993, Roberts [4] described cardiac necropsy findings in an additional 53 patients z90 years of age. In 1995, Shirani et al. [5] described cardiac findings at necropsy in 366 Americans aged 69

70

Roberts

80 to 89 years, and in 1998 Roberts [6] described cardiac findings at necropsy in six centenarians. Also, in 1998 Roberts and Shirani [7] compared cardiovascular findings in 490 patients studied at necropsy in three age categories: 80–89; 90–99; and z100 years. This chapter expands on that previous comparative study.

METHODS Patients The files of the Pathology Branch, National Heart, Lung, and Blood Institutes, National Institutes of Health, Bethesda, Maryland, from 1959 to 1993, and those of the Baylor University Medical Center, Dallas, Texas, beginning January 1993, were searched for all accessioned cases of patients aged z80 years of age. Of 511 such cases found, adequate clinical information was available in 490 necropsy patients and they are the subject of this chapter. The clinical and cardiac morphological records, photographs, and postmortem radiographs, histological slides, and the initial gross description of the heart were reviewed. All 490 hearts were originally examined by WCR, who recorded gross morphological abnormalities in each case. Sources of Patients Of the 490 cases, the hearts in 412 were obtained from 12 Washington, DC, area hospitals, and the hearts in the other 78 cases, from hospitals outside that area, including 25 from Baylor University Medical Center. Of the 490 hearts, 37 (8%) were examined in 1970 or before; 153 (31%), from 1971 through 1980; 244 (50%), from 1981 through 1990; and 56 (11%) from 1991 through 1997. Definitions Sudden coronary death was defined as death within 6 h from the onset of new symptoms of myocardial ischemia in the presence of morphological evidence of significant atherosclerotic coronary artery disease (z1 major epicardial coronary artery narrowed >75% in cross-sectional area by atherosclerotic plaque). Most patients who died suddenly did so outside a hospital; a few, however, died shortly after admission to an emergency room. Sudden, out-of-hospital death also occurred in some patients with cardiac disease other than atherosclerotic coronary artery disease. In each case, an underlying cardiac disease generally accepted to cause sudden death was present at necropsy. Acute myocardial infarction was defined as a grossly visible left ventricular wall lesion confirmed histologically to represent coagulation-type myocardial necrosis. Ischemic cardiomyopathy was defined as chronic congestive heart failure associated with a transmural healed myocardial infarct and a dilated left ventricular cavity. Cardiac Morphological Data Hearts were fixed in 10% phosphate-buffered formalin for 3 to 15 days before examination. They were ‘‘cleaned’’ of parietal pericardium and postmortem intracavity clot, and the pulmonary trunk and ascending aorta were incised approximately 2 cm cephalad to the

Morphological Features

71

sinotubular junction. Heart weight was then measured on accurate scales by WCR (Lipsaw scale before 1971, accurate to 10 g, and Mettler P1210 scale after 1971, accurate to 0.1 g). Heart weight was considered increased if it was z350 g in women and >400 g in men. Most hearts were studied by cutting the ventricles transversely at approximately 1-cm-thick intervals from apex to base parallel to the atrioventricular groove posteriorly. In each heart, the sizes of cardiac ventricular cavities (determined by gross inspection), presence of left ventricular necrosis (acute myocardial infarct) and fibrosis (healed myocardial infarct), status of the four cardiac valves, and the maximal degree of cross-sectional luminal narrowing in each of the three major (left anterior descending, left circumflex, and right) epicardial coronary arteries were recorded.

RESULTS Number of Patients in Each of the Three Groups Certain clinical and necropsy cardiac findings in the 490 cases are summarized in Table 1 and illustrated in Figures 1 through 19. The 490 patients were divided into three groups: the octogenarians (80–89 years) (n = 391[80%]); the nonagenarians (90–99 years) (n = 93[19%]); and the centenarians (z100 years) (n = 6[1%]); 248 (51%) were women and 242 (49%) were men.

CLINICAL FINDINGS The clinical manifestations of the various cardiac disorders probably represent minimal numbers: many patients apparently were unable to provide much clinical information, many came to the hospital from nursing homes, and many had varying degrees of dementia. Nevertheless, angina pectoris was noted in the records of 137 (35%) of the 391 octogenarians, in 5 (5%) of the 93 nonagenarians, and in none of the centenarians. A history of a hospitalization for an illness compatible with acute myocardial infarction was present in 78 (20%) of the 391 octogenarians; in 18(19%) ofthe 93 nonagenarians; and innone ofthe 6 octogenarians. Chronic congestive heart failure was present in 36% (140/391) of the octogenarians; in 25% (23/93) of the nonagenarians; and in none of the 6 centenarians. A history of systemic hypertension was present in 44% (174/391) of the octogenarians; in 54% (50/93) of the nonagenarians; and in none of the centenarians. Diabetes mellitus was present in 14% (56/391) of the octogenarians; in 9% (8/93) of the nonagenarians; and in none of the centenarians.

CAUSES OF DEATH The causes of death in the 490 patients are summarized in Table 2. A cardiac condition was the cause of death in 51% (198/391) of the octogenarians; in 32% (30/99) of the nonagenarians; and in none of the centenarians. A noncardiac but vascular condition was responsible for death in 13% (52/391) of the octogenarians and in 20% (19/93) of the nonagenarians. A noncardiac and a nonvascular condition was responsible for death in 36% (141/391) of the octogenarians; in 47% (44/93) of the nonagenarians; and in all of the centenarians.

72 Table 1

Roberts Certain Clinical and Necropsy Findings in 490 Patients Aged 80 to 103 Years Age group (years)

Variable 1. 2. 3. 4. 5. 6. 7. 8. 9.

Mean age (Years) Male: female Angina pectoris Acute myocardial infarction Chronic congestive heart failure Systemic hypertension (history) Diabetes mellitus Atrial fibrillation Heart weight (g): range(mean) Men Women 10. Cardiomegaly Men > 400 g Women > 350 g 11. Cardiac calcific deposits None Present Coronary arteries Aortic valve cusps Heavy (stenosis) Mitral annulus Heavy Papillary muscle 12. Numbers of patients with 0,1,2, or 3 major (right, left anterior descending, left circumflex) coronary arteries # >75% in cross-sectional area 0 1 2 3 Mean 13. Number of major coronary arteries (3/patient) # >75% in cross-sectional area by plaque 0 1 2 3 Totals 14. Left ventricular necrosis and/or fibrosis Necrosis only Fibrosis only Both 15. Ventricular cavity dilatation Neither One Right ventricle Left ventricle Both 16. Cardiac amyloidosis (massive)

80–89 (n = 391)

90–99 (n = 93)

z100 (n = 6)

84 F 4 194(50%): 97(50%) 137(35%) 78(20%) 140(36%) 174(44%) 56(14%) 57(15%) 185–900(449) 230–830(493) 185–900(409)

93 F 4 52(56%): 41(44%) 5(5%) 18(18%) 23(25%) 50(54%) 8(9%) 35(38%) 220–660(420) 285–660(436) 220–630(406)

102 2/4 0 0 0 0 0 0 240–410(328) 335&410(372) 240–385(306)

103/154(67%) 133/165(81%)

27/49(55%) 23/42(55%)

43(11%) 348(89%) 304(78%) 164(42%) 43(11%) 146(37%) 52(13%) 37(9%)

3(3%) 90(97%) 89(96%) 59(63%) 8(9%) 42(45%) 11(12%) 42(45%)

159(41%) 9 67 = 71 232ð59%Þ ; 94 1.7

33(35%) 9 20 = 32 60ð65%Þ ; 8 1.5

0/477 67/201 142/213 282/282 491/1173(42%)

0/99 20/60 64/96 24/24 108/279(39%)

1/2 1/4 0 6(100%) 5 5 0 2 0 6

29 2= 2 4 ; 0 1.5

0/6 2/6 4/6 0 6/18(33%)

54/(14%) 101(26%) 37(9%)

10(11%) 20(21%) 5(5%)

0 2 0

222(57%) 84(21%) 42 42 85(22%) 14(4%)

58(62%) 4(4%) 2 2 31(34%) 8(9%)

4 2 2 2 0 0

Figure 1 Tortuous and heavily calcified coronary arteries in a 95-year-old woman (SH #A80-74) who never had symptoms of cardiac dysfunction and who died from a perforated gastric ulcer. Top left: postmortem radiogram of the excised right (R), left main (LM), left anterior descending (LAD), and left circumflex (LC) coronary arteries. Top center: radiogram of a portion of the heart after removing the walls of the atria and most of the walls of the ventricles. MVA = mitral valve annulus. Top right and lower panels: photomicrographs of coronary arteries at sites of maximal narrowing by calcified atherosclerotic plaques. (Movat stains; magnification  17, reduced 40%). This case illustrates extensive calcific deposits in the coronary arteries without significant luminal narrowing.

Morphological Features 73

74

Roberts

Figure 2 Coronary arteries in a 95-year-old man (SH #A79-77) who never had symptoms of cardiac dysfunction and who died from cancer. Top: aorta and coronary arteries from above. Bottom: radiograms of aortic root (Ao) and excised coronary arteries that contain calcific deposits. FD = first diagonal, LAD = left anterior descending, LC = left circumflex, LM = left main, R = right.

Morphological Features

75

Figure 3 Heart and coronary arteries in a 93-year-old woman who was hospitalized with worsening chronic congestive heart failure. Top left: postmortem radiogram showing calcific deposits in the right (R), left anterior descending (LAD), and left circumflex (LC) coronary arteries and in the mitral valve annulus (MVA) and aortic valve. Top right: view of right (RV) and left (LV) ventricles and ventricular septum (VS) showing a transmural scar (arrows). Bottom: coronary arteries at sites of maximal narrowing. LM = left main. (Movat stains; magnification F 17, reduced 40%.)

Figure 4 This 91-year-old woman (LRC #3) had a ‘‘typical’’ clinical acute myocardial infarct, which was fatal. Left: postmortem radiogram showing calcific deposits in the right (R), left main (LM), left anterior descending (LAD), and left circumflex (LC) coronary arteries. Right, view of the left and right (RV) ventricles showing transmural necrosis and fibrosis (arrows) and dilated ventricles. VS = ventricular septum.

76

Roberts

Figure 5 This 90-year-old woman (SH #A82-1) had no symptoms of cardiac dysfunction and died from a ruptured fusiform abdominal aortic aneurysm. Left: postmortem radiogram of the entire heart showing calcific deposits in the left anterior descending (LAD), left circumflex (LC), and left main (LM) coronary arteries. Right: view of heart cut in an anteroposterior fashion (Mmode or long-axis two-dimensional echocardiographic view) showing a dilated left atrium (LA) and left ventricle (LV). A and P = anterior and posterior mitral valve leaflets; Ao = aorta; RV = right ventricle; VS = ventricular septum.

CARDIAC NECROPSY FINDINGS Cardiac findings at necropsy are tabulated in Table 1. Heart Weight The mean heart weights were largest in the octogenarians and smallest in the centenarians (449 g vs. 420 g vs. 328 g). Heart weight was increased (>400 g in men; >350 g in women) in 131 (64%) of the 205 men and in 157(74%) of the women, and the percent was highest in the octogenarians. Calcific Deposits in the Heart Calcific deposits were present in the heart at necropsy in 444 (91%) of the 490 patients and were most common in the coronary arteries—in all cases being located in atherosclerotic plaques and not in the media—81% (398/490); in the aortic valve cusps in 47% (228/490)— heavy enough to result in aortic stenosis in 10% (51/490); mitral valve annulus in 39% (190/ 490)—very heavy deposits in 13% (63/490), and in the apices of 1 or both left ventricular papillary muscles in 17% (85/490). The cardiac calcific deposits were more frequent and heavier in the nonagenarians than in the octogenarians.

Figure 6 This 97-year-old man (GT # 69A-585) had complete heart block and had a pacemaker inserted when he was 91 years old. Left: radiogram of heart at necropsy showing calcific deposits in the left anterior descending (LAD), left circumflex (LC), and right (R) coronary arteries and a pacemaker wire in the right ventricle (RV). LV = left ventricle. Right: electrocardiogram.

Morphological Features 77

78

Roberts

Figure 7 This 96-year-old woman (DCGH #68A80) had chronic congestive heart failure from dilated cardiomyopathy. Her total serum cholesterol level was 108 mg/dl. Left: postmortem radiogram of the right (R), left main (LM), left anterior descending (LAD), left circumflex (LC), first diagonal (FD), and left obtuse marginal (OM) coronary arteries showing no calcific deposits. Right: right coronary artery section devoid of calcium or narrowing. (Movat stain; magnification  16, reduced 23%.)

Numbers of Patients with Narrowing of 1 or More Major Epicardial Coronary Arteries Among the 490 patients, 194 (40%) had none of the three major (right, left anterior descending, and left circumflex) epicardial coronary arteries narrowed by plaque >75% in cross-sectional area; 89 (18%) had one artery so narrowed; 105 (21%) had two arteries so narrowed; and 94 (19%) had all three arteries so narrowed. The percent of patients in each of the three groups with no arteries and 1, 2, and 3 arteries narrowed >75% was similar. Numbers of Major Coronary Arteries (Three/Patient) Narrowed >75% in Cross-Sectional Area by Plaque Among the 490 patients, a total of 1470 major epicardial coronary arteries were examined: 865 (59%) arteries had insignificant (75% in cross-sectional area. The percent of arteries significantly narrowed was similar in each of the three groups (42% vs. 39% vs. 33%). Acute and Healed Myocardial Infarcts Grossly visible foci of left ventricular wall (includes ventricular septum) necrosis (acute infarcts) without associated left ventricular scars (healed infarcts) were found in 64 (13%)

Morphological Features

79

Figure 8 Right (R), left main (LM), left anterior descending (LAD), and left circumflex (LC) coronary arteries at sites of maximal narrowing in a 103-year-old woman (GT #70A301). She never had evidence of cardiac dysfunction and died from complications of a duodenal ulcer. The coronary arteries are devoid of calcium and of significant narrowing. (Elastic-van Gieson’s stain; magnification  16, reduced 31%.)

patients; foci of left ventricular fibrosis without associated necrosis were observed in 123 patients (25%); and foci of both necrosis and fibrosis were found in 42 patients (9%). Thus, a total of 229 (47%) of the 490 patients had grossly visible evidence of acute or healed myocardial infarcts or both. The percents of patients with myocardial lesions of ischemia were similar among the octogenarians and nonagenarians. Ventricular Cavity Dilatation One or both ventricular cavities were dilated (by gross inspection) in 218 (44%) of the 490 patients and no significant differences were observed in the three groups. Cardiac Amyloidosis Grossly visible amyloid (confirmed histologically) in ventricular and atrial myocardium as well as in atrial mural endocardium was present in 22 patients (4%). In these 22 patients, the amyloidosis was symptomatic and fatal. A number of other patients who had no gross

80

Roberts

Figure 9 Mean percentage cross-sectional area narrowing by atherosclerotic plaques in 1789 segments (5 mm) of the four major epicardial coronary arteries in 36 necropsy patients aged 90 years and older, 10 of whom had angina pectoris or acute myocardial infarction, or both, and 26 of whom did not. The mean percentage of narrowing for each 5-mm segment differs significantly in the group with clinical evidence of myocardial ischemia (55%) and the group without such evidence.

evidence of cardiac amyloidosis had small foci in the heart on histological study. These minute deposits did not cause symptoms of cardiac dysfunction. In the 22 patients with fatal cardiac amyloidosis, deposits of amyloid also were present in several other body organs at necropsy.

ELECTROCARDIOGRAPHIC FINDINGS Information on electrocardiograms was available on 30 of the 99 patients aged z90 years. Of the 30 patients, three had electrocardiograms recorded during acute myocardial infarction. The electrocardiograms in all three, however, disclosed only atrial fibrillation and complete left bundle branch block, and these findings in these patients were known to be present before the fatal acute myocardial infarction. The electrocardiograms in the other 27 patients were not recorded during periods of acute myocardial infarction. Of the 30 patients on whom electrocardiographic information was available, 8 had clinical evidence of heart disease and 22 did not; the findings are summarized in Table 2. The total 12-lead QRS voltage ranged from 82 to 251 mm (mean 151; 10 mm = 1 mV). In the 19 women, the total voltage ranged from 82 to 251 mm (mean 158) and, in the 5 men, from

Morphological Features

81

Figure 10 Mean percentage cross-sectional area (XSA) narrowing by atherosclerotic plaques in the 5-mm segments of the right, left main, left anterior descending, and left circumflex coronary arteries in the 36 patients in whom the four major coronary arteries in their entirety were available for examination. The 36 patients were classified into 10 with and 26 without clinical evidence of myocardial ischemia. For each of the four categories of narrowing, a highly significant difference in the amount of narrowing was observed.

105 to 154 mm (mean 101; p75% in cross-sectional area by plaque and the percent of patients in each of the three age groups and the percent of coronary arteries significantly narrowed in each of the three age groups was similar.

REFERENCES 1.

Waller BF, Roberts WC. Cardiovascular disease in the very elderly. Analysis of 40 necropsy patients aged 90 years or over. Am J Cardiol 1983; 51:403–421. 2. Lie JT, Hammond Pl. Pathology of the senescent heart: anatomic observations on 237 autopsy studies of patients 90 to 105 years old. Mayo Clinic Proc 1988; 63:552–564. 3. Gertz SD, Malekzadeh S, Dollar AL, Kragel AH, Roberts WC. Composition of atherosclerotic plaques in the four major epicardial coronary arteries in patients z90 years of age. Am J Cardiol 1991; 67:1228–1233. 4. Roberts WC. Ninety-three hearts z90 years of age. Am J Cardiol 1993; 71:599–602. 5. Shirani J, Yousefi J, Roberts WC. Major cardiac findings at necropsy in 366 American octogenarians. Am J Cardiol 1995; 75:151–156. 6. Roberts WC. The heart at necropsy in centenarians. Am J Cardiol 1998; 81:1224–1225. 7. Roberts WC, Shirani J. Comparison of cardiac findings at necropsy in octogenarians, nonagenarians, and centenarians. Am J Cardiol 1998; 82:627–631. 8. Fries JF. Aging, natural death, and the compression of morbidity. N Engl J Med 1980; 303:, 130– 135. 9. Lachman AS, Spray TL, Kerwin DM, Shugoll GI, Roberts WC. Medial calcinosis of Monckeberg. A review of the problem and a description of a patient with involvement of peripheral, visceral, and coronary arteries. Am J Cardiol 1977; 63:615–622. 10. Hammer WJ, Roberts WC, deLeon AC Jr, ‘‘Mitral stenosis’’ secondary to combined ‘‘massive’’ mitral annular calcific deposits and small, hypertrophied left ventricles. Am J Med 1978; 64:371– 376. 11. Roberts WC. Qualitative and quantitative comparison of amounts of narrowing by atherosclerotic plaques in the major epicardial coronary arteries at necropsy in sudden coronary death, transmural acute myocardial infarction, transmural healed myocardial infarction and unstable angina pectoris. Am J Cardiol 1989; 4:324–328.

4 Cardiovascular Drug Therapy in the Elderly William H. Frishman Westchester Medical Center/New York Medical College, Valhalla, New York, U.S.A.

Wilbert S. Aronow Westchester Medical Center/New York Medical College, Valhalla, and Mount Sinai School of Medicine, New York, New York, U.S.A.

Angela Cheng-Lai Montefiore Medical Center, Bronx, New York, U.S.A.

Cardiovascular disease is the greatest cause of morbidity and mortality in the elderly, and cardiovascular drugs are the most widely prescribed drugs in this population. Since many cardiovascular drugs have narrow therapeutic windows in the elderly, the incidence of adverse effects from using these drugs is also highest in the elderly. The appropriate use of cardiovascular drugs in the elderly requires knowledge of age-related physiological changes, the effects of concomitant diseases that alter the pharmacokinetic and pharmacodynamic effects of cardiovascular drugs, and drug interactions. PHARMACOKINETIC CONSIDERATIONS IN THE ELDERLY Absorption Age-related physiological changes that may affect absorption include reduced gastric secretion of acid, decreased gastric emptying rate, reduced splanchnic blood flow, and decreased mucosal absorptive surface area (Table 1). Despite these physiological changes, the oral absorption of cardiovascular drugs is not significantly affected by aging, probably because most drugs are absorbed passively [1]. Bioavailability There are almost no data available for age-related changes in drug bioavailability for routes of administration other than the oral route [2]. The bioavailability of cardiovascular drugs depends on the extent of drug absorption and on first-pass metabolism by the liver and/or 95

Reduced glomerular filtration, renal tubular function, and renal blood flow

Result

Accumulation of renally cleared drugs

Accumulation of metabolized drugs

Increased Vd of highly lipid-soluble drugs Decreased Vd of hydrophilic drugs Changed % of free drug, Vd, and measured levels of bound drugs

Reduced tablet dissolution and decreased solubility of basic drugs Decreased absorption for acidic drugs Less opportunity for drug absorption

Abbreviations: GI, gastrointestinal; ACE, angiotensin converting enzyme. Source: Adapted from Ref. 157.

Excretion

Metabolism

Distribution

Reduced gastric emptying rate Reduced GI mobility, GI blood flow, absorptive surface Decreased total body mass. Increased proportion of body fat Decreased proportion of body water Decreased plasma albumin, disease-related increased a1-acid glycoprotein, altered relative tissue perfusion Reduced liver mass, liver blood flow, and hepatic metabolic capacity

Reduced gastric acid production

Physiological change

Physiological Changes with Aging Potentially Affecting Cardiovascular Drug

Absorption

Process

Table 1

" propranolol, nitrates, lidocaine, diltiazem, warfarin, labetalol, verapamil, mexiletine digoxin, ACE inhibitors, antiarrhythmic drugs, atenolol, sotalol, nadolol

" digoxin and ACE inhibitors " disopyramide and warfarin, lidocaine, propranolol

# h-blockers, central a-agonists

Drugs affected

96 Frishman et al.

Cardiovascular Drug Therapy

97

the wall of the gastrointestinal (GI) tract. In the elderly, the absolute bioavailability of drugs such as propranolol, verapamil, and labetalol is increased because of reduced first-pass hepatic extraction [3]. However, the absolute bioavailability of prazosin in the elderly is reduced [4]. Drug Distribution With aging there is a reduction in lean body mass [5] and in total body water [6], causing a decrease in volume of distribution (Vd) of hydrophilic drugs. This leads to higher plasma concentrations of hydrophilic drugs such as digoxin and angiotensin converting enzyme (ACE) inhibitors with first dose in the elderly [7]. The increased proportion of body fat that occurs with aging also causes an increased Vd for lipophilic drugs. This leads to lower initial plasma concentrations for lipophilic drugs such as most beta-blockers, antihypertensive drugs, and central alpha-agonists. The level of alpha1-acid glycoprotein increases in the elderly [8]. Weak bases such as disopyramide, lidocaine, and propranolol bind to alpha1-acid glycoprotein. This may cause a reduction in the free fraction of these drugs in the circulation, a decreased Vd, and a higher initial plasma concentration [9]. In the elderly, there is also a tendency for plasma albumin concentration to be reduced [10]. Weak acids, such as salicylates and warfarin, bind extensively to albumin. Decreased binding of drugs such as warfarin to plasma albumin may result in increased free-drug concentrations, resulting in more intense drug effects [11]. Half-Life The half-life of a drug (or of its major metabolite) is the length of time in hours that it takes for the serum concentration of that drug to decrease to one-half of its peak level. This can be described by the kinetic equation: t1/2 = 0.693Vd/Cl, where t1/2 is directly related to drug distribution and inversely to clearance. Therefore, changes in Vd and/or Cl due to aging, as previously mentioned, can affect the half-life of a drug. In elderly patients, an increased halflife of a drug means a longer time until steady-state conditions are achieved. With a prolonged half-life of a drug, there may be an initial delay in maximum effects of the drug and prolonged adverse effects. Table 2 lists the pharmacokinetic changes, routes of elimination, and dosage adjustment for common cardiovascular drugs used in the elderly. Drug Metabolism Decreased hepatic blood flow, liver mass, liver volume, and hepatic metabolic capacity occur in the elderly [12]. There is a reduction in the rate of many drug oxidation reactions (phase 1) and little change in drug conjugation reactions (phase 2). These changes in the elderly may result in higher serum concentrations of cardiovascular drugs metabolized in the liver, including propranolol, lidocaine, labetalol, verapamil, diltiazem, nitrates, warfarin, and mexiletine. Drug Excretion With aging there is a reduction in the total numbers of functioning nephrons and thereby a parallel decline in both glomerular filtration rate and renal plasma flow [13,14]. The age-

98

Frishman et al.

Table 2 Pharmacokinetic Changes, Route of Elimination, and Dosage Adjustment of Selected Cardiovascular Drugs in the Elderly Drug

T 1/2

Primary route(s) of elimination

VD

CI

Alpha-Adrenergic Agonists Centrally Acting Clonidine -

-

-

Hepatic/renal

Guanabenz

-

-

-

Hepatic

Guanfacine

"

-

#

Hepatic/renal

Methyldopa

-

-

-

Hepatic

"

"*

Hepatic

Alpha1-Selective Adrenergic Antagonists Peripherally Acting Doxazosin "

Dosage adjustment

Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response

Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response

Prazosin

"

-

-

Hepatic

Terazosin

"

-

-

Hepatic

" NS

-

# #

Renal Renal

Enalapril

-

-

-

Renal

Fosinopril Lisinopril

"

NS

#

Hepatic/renal Renal

Moexipril

-

-

-

Hepatic/renal

Perindopril

-

-

#

Renal

Quinapril

-

-

-

Renal

Ramipril

-

-

-

Renal

Trandolapril

-

-

-

Hepatic/renal

No adjustment needed Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response No adjustment needed Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response No adjustment needed

NS "

-

# -

Hepatic/renal Hepatic/fecal/renal Hepatic Hepatic Renal/biliary Hepatic/biliary Hepatic

No No No No No No No

Angiotensin Converting Enzyme Inhibitors Benazepril Captopril

Angiotensin-II Receptor Blockers Candesartan Eprosartan Irbesartan Losartan Olmesartan Telmisartan Valsartan

adjustment adjustment adjustment adjustment adjustment adjustment adjustment

needed needed needed needed needed needed needed

Cardiovascular Drug Therapy

99

Table 2 Continued Drug

Primary route(s) of elimination

T 1/2

VD

CI

Class I Disopyramide

"

-

#

Renal

Flecainide

"

"

#

Hepatic/renal

Lidocaine

"

"

NS

Hepatic

Mexilitine

-

-

-

Hepatic

Moricizine

-

-

-

Hepatic

Procainamide

-

-

#

Renal

Propafenone

-

-

-

Hepatic

Quinidine

"

NS

#

Hepatic

Tocainide

"

-

#

Hepatic/renal

Class III Amiodarone

-

-

-

Hepatic/biliary

Bretylium

-

-

-

Renal

Dofetilide

-

-

-

Hepatic/renal

Ibutilide Sotalol

-

-

-

Hepatic Renal

Dosage adjustment

Antiarrhythmic Agents Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response

Class II (see b-Blockers) Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response No adjustment needed Adjust dose based on renal function

Class IV (see Calcium Channel Blockers) Other Antiarrhythmics Adenosine

-

-

-

Atropine

-

-

-

Anticoagulants Heparin

-

-

-

Ardeparin Argatroban Bivalirudin

-

-

-

Erythrocytes/vascular endothelial cells Hepatic/renal

No adjustment needed Use usual dose with caution

Antithrombotics Hepatic/reticuloendothelial system Renal Hepatic/biliary Renal/proteolytic cleavage

Use usual dose with caution Use usual dose with caution Use usual dose with caution Adjust dose based on renal function

100

Frishman et al.

Table 2 Continued Drug

Primary route(s) of elimination

Dosage adjustment

T 1/2

VD

CI

NS " "

NS -

NS # #

Renal Renal Renal Renal Renal

-

-

-

Renal

NS

NS

NS

Antiplatelets Aspirin Abciximab Anagrelide Clopidogrel Dipyridamole Eptifibatide Ticlopidine Tirofiban

NS "

-

# # #

Hepatic/renal Unknown Hepatic/renal Hepatic Hepatic/biliary Renal/plasma Hepatic Hepatic

Use Use Use Use Use Use Use Use

usual usual usual usual usual usual usual usual

dose dose dose dose dose dose dose dose

with with with with with with with with

caution caution caution caution caution caution caution caution

Thrombolytics Alteplase Anistreplase Reteplase Streptokinase

-

-

-

Use Use Use Use

usual usual usual usual

dose dose dose dose

with with with with

caution caution caution caution

Tenecteplase Urokinase Chlorothiazide

-

-

-

Hepatic Unknown Hepatic Circulating antibodies/ reticuloendothelial system Hepatic Hepatic Renal

Chlorthalidone

-

-

-

Renal

Hydrochlorothiazide

-

-

#

Renal

Hydroflumethiazide

-

-

-

Unknown

Indapamide

-

-

-

Hepatic

Methyclothiazide

-

-

-

Renal

Metolazone

-

-

-

Renal

Polythiazide

-

-

-

Unknown

Quinethazone

-

-

-

Unknown

Trichlormethiazide

-

-

-

Unknown

Dalteparin Danaparoid Enoxaparin Fondaparinux Lepirudin Tinzaparin Warfarin

Hepatic

Use usual dose with caution Use usual dose with caution Use usual dose with caution Use usual dose with caution Adjust dose based on renal function Use usual dose with caution Initiate at lowest dose; titrate to response

Use usual dose with caution Use usual dose with caution Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response

Cardiovascular Drug Therapy

101

Table 2 Continued Drug

Primary route(s) of elimination

Dosage adjustment

T 1/2

VD

CI

Potassium-Sparing Amiloride

-

-

#

Renal

Initiate at lowest dose; titrate to response

Eplerenone Spironolactone

-

-

-

Hepatic/renal

Triamterene

"

-

-

Hepatic/renal

Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response

-

-

Hepatic/biliary

Use usual dose with caution

Human B-Type Natriuretic Peptide Nesiritide -

-

Cellular internalization and lysosomal proteolysis/proteolytic cleavage/renal filtration

Use usual dose with caution

Inotropic and Vasopressor Agents Inamrinone -

-

Hepatic/renal

Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response No adjustment necessary Adjust based on renal function Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response

Endothelin Receptor Antagonist Bosentan -

Digoxin

"

#

#

Renal

Dobutamine

-

-

-

Hepatic/tissue

Dopamine

-

-

-

Renal/hepatic/plasma

Epinephine

-

-

-

Isoproterenol

-

-

-

Sympathetic nerve endings/plasma Renal

Metaraminol

-

-

-

Unknown

Methoxamine

-

-

-

Unknown

Midodrine Milrinone

-

-

-

Tissue/hepatic/renal Renal

Norepinephrine

-

-

-

Sympathetic nerve endings/plasma Hepatic/intestinal

BAS Cholestyramine

-

-

-

Colestipol

-

-

-

Not absorbed from GI tract Not absorbed from GI tract

Phenylephrine Lipid-Lowering Agents

No adjustment needed No adjustment needed

102

Frishman et al.

Table 2 Continued Drug Colesevelam

Primary route(s) of elimination

Dosage adjustment

T 1/2

VD

CI

-

-

-

Not absorbed from GI tract

No adjustment needed

-

-

Renal

Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response

SCAI Ezetimibe b-Adrenergic Blockers Nonselective without ISA Nadolol NS Propranolol

"

NS

#

Hepatic

Timolol

-

-

-

Hepatic

NS

#

Renal

b1 Selective without ISA Atenolol "

Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Use at usual dose with caution Initiate at lowest dose; titrate to response

Betaxolol

-

-

-

Hepatic

Bisoprolol

-

-

-

Hepatic/renal

Esmolol

-

-

-

Erythrocytes

NS

NS

NS

Nonselective with ISA Carteolol

-

-

-

Renal

Penbutolol Pindolol

-

-

-

Hepatic Hepatic/renal

Selective with ISA Acebutolol

"

#

-

Hepatic/biliary

Initiate at lowest dose; titrate to response

Dual Acting Carvedilol

-

-

-

Hepatic/biliary

Labetalol

-

-

NS

Hepatic

Initiate at lowest dose; titrate to response No adjustment needed

Calcium Channel Blockers Amlodipine "

-

#

Hepatic

Metoprolol

Hepatic

Bepridil Diltiazem

"

NS

#

Hepatic Hepatic

Felodipine

-

NS

#

Hepatic

Isradipine

-

-

-

Hepatic

NS

-

-

Hepatic

Nicardipine

Initiate at lowest dose; titrate to response Use usual dose with caution Initiate at lowest dose; titrate to response

Initiate at lowest dose; titrate to response Use usual dose with caution Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response No adjustment needed

Cardiovascular Drug Therapy

103

Table 2 Continued Drug

Primary route(s) of elimination

T 1/2

VD

CI

Nifedipine

"

NS

#

Hepatic

Nimodipine

-

-

-

Hepatic

Nisoldipine

-

-

-

Hepatic

Verapamil

"

NS

#

Hepatic

Loop Bumetanide

-

NS

-

Renal/hepatic

Ethacrynic Acid

-

-

-

Hepatic

Furosemide

"

NS

#

Renal/hepatic

Torsemide

-

-

-

Hepatic

Thiazides Bendroflumethiazide

-

-

-

Renal

Benzthiazide

-

-

-

Renal

FADS Clofibrate

-

-

-

Hepatic/renal

Fenofibrate Gemfibrozil Nicotinic Acid

-

-

-

Renal Hepatic/renal Hepatic/renal

" -

– -

-

Hepatic/biliary Hepatic Hepatic/fecal Hepatic

-

-

-

Hepatic/fecal

-

-

-

Hepatic/renal

Guanethidine

-

-

-

Hepatic/renal

Mecamylamine

-

-

-

Renal

Reserpine

-

-

-

Hepatic/fecal

Dosage adjustment Initiate at lowest dose; titrate to response Use usual dose with caution Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response

Diuretics

HMG CoA Reductase Inhibitors Atorvastatin Fluvastatin Lovastatin Pravastatin Simvastatin Neuronal and Ganglionic Blockers Guanadrel

Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Adjust based on renal function No adjustment necessary No adjustment necessary No adjustment necessary

No adjustment necessary No adjustment necessary No adjustment necessary Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response

Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response

104

Frishman et al.

Table 2 Continued Drug

Primary route(s) of elimination

Dosage adjustment

T 1/2

VD

CI

-

-

-

Hepatic/renal

Initiate at lowest dose; titrate to response

Vasodilators Alprostadil

-

-

-

Pulmonary/renal

Cilostazol Diazoxide

-

-

-

Hepatic/renal Hepatic/renal

Epoprostenol

-

-

-

Hepatic/renal

Fenoldopam Hydralazine

-

-

-

Hepatic Hepatic

ISDN

-

-

-

Hepatic

NS -

-

NS -

Hepatic Renal

Minoxidil

-

-

-

Hepatic

Nitroglycerin

-

-

-

Hepatic

Nitroprusside

-

-

-

Papaverine

-

-

-

Hepatic/renal/ erythrocytes Hepatic

Initiate at lowest dose; titrate to response No adjustment necessary Initiate at lowest dose; titrate to response Initiate at usual dose with caution No adjustment necessary Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response No adjustment necessary Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Initiate at lowest dose; titrate to response Use usual dose with caution

Pentoxifylline Sildenafil

-

-

# -

Hepatic/renal Hepatic/fecal

Trimethaphan

ISMN Isoxsuprine

Initiate at lowest dose; titrate to response Use usual dose with caution Initiate at lowest dose; titrate to response

* Increase in Cl is small compared to increase in VD; T 1/2, half-life; VD, volume of distribution; Cl, clearance; ", increase; #, decrease; -, no information or not relevant; NS, no significant change; LMWH, low-molecular-weight heparin; ISA, intrinsic sympathomimetic activity; GI, gastrointestinal; BAS, bile acid sequestrants; SCAI, selective cholesterol absorption inhibitors; FADS, fibric acid derivatives; HMG CoA, hydroxymethylglutaryl coenzyme A; ISDN, isosorbide dinitrate; ISMN, isosorbide mononitrate. Source: From Ref. 158.

related decline in renal function is likely the single most important physiological change causing pharmacokinetic alterations in the elderly. The change in renal function with aging is insidious and poorly characterized by serum creatinine determinations, although serum creatinine measurements remain one of the most widely used tests for gauging renal function. To estimate renal function from a serum creatinine value requires its being indexed for muscle mass, which is difficult in even the most skilled hands. Creatinine is a byproduct of creatine metabolism in muscle and its daily production correlates closely with muscle mass. Thus, the greater the muscle mass, the higher the ‘‘normal serum creatinine.’’ For example, in a heavily muscled male, a serum creatinine value of 1.4 mg/dL might be considered normal, although such a value may be considered grossly abnormal in an individual with

Cardiovascular Drug Therapy

105

less muscle, such as an aged individual. A safer way to estimate renal function in the elderly is by use of a urine-free formula such as the Cockcroft–Gault formula [15]: Creatinine clearance ðmL=minÞ ¼

ð140
ageÞ  body weight ðkgÞ 72  Screat ðmg=dLÞ

For females, the results of this equation can be multiplied by 0.85 to account for the small muscle mass in most females. It should be appreacited that creatinine clearance is reciprocally related to serum creatinine concentrations, such that a doubling of serum creatinine represents an approximate halving of renal function. The axiom that glomerular filtration rate is reciprocally related to serum creatinine is most important with the first doubling of serum creatinine. For example, a serum creatinine value of 0.6 mg/dL in an elderly subject doubles to 1.2 mg/dL and with this doubling creatinine clearance falls from 80 cc/min to about 40 cc/min. The reduced clearance of many drugs primarily excreted by the kidneys causes their half-life to be increased in the elderly. Cardiovascular drugs known to be excreted by the kidney, via various degrees of filtration and tubular secretion, include digoxin, diuretics, ACE inhibitors, antiarrhythmic medications (bretylium, disopyramide, flecainide, procainamide, tocainide) and the beta-blockers (atenolol, bisoprolol, carteolol, nadolol, and sotalol). Typically, a renally cleared compound begins to accumulate when creatinine clearance values drop below 60 cc/min. An example of this phenomenon can be seen with ACE inhibitors [16] wherein accumulation begins early in the course of renal functional decline. Moreover, ACE inhibitor accumulation in the elderly is poorly studied in the case of many of the ACE inhibitors, particularly as relates to the ‘‘true level of renal function’’ when an otherwise healthy elderly subject undergoes formal pharmacokinetic testing. Thus, it has not been uncommon for elderly subjects with serum creatinine values as high as 2.0 mg/dL to be allowed entry into a study whose primary purpose is to determine the difference in drug handling of a renally cleared compound in young versus elderly subjects.

PHARMACODYNAMICS There are numerous physiological changes with aging that affect pharmacodynamics with alterations in end-organ responsiveness (Table 3). Increased peripheral vascular resistance is the cause of systolic and diastolic hypertension in the elderly [17]. Inappropriate sodium intake and retention may contribute to increased arteriolar resistance and/or plasma volume. Cardiac output, heart rate, renal blood flow, glomerular filtration rate, and renin levels decline with aging. Increased arterial stiffness, resulting from changes in the arterial media and an increase in arterial tonus and arterial impedance, increases systolic blood pressure, and contributes to a widened pulse pressure. Maintenance of alpha-adrenergic vasoconstriction with impaired beta-adrenergic-mediated vasodilation may be an additional contributory factor to increased peripheral vascular resistance. The cardiovascular response to catecholamines as well as carotid arterial baroreflex sensitivity are both decreased in the elderly. Left ventricular (LV) mass and left atrial dimension are increased, and there is a reduction in both the LV early diastolic filling rate and volume [17]. The pharmacodynamic, chronotropic, and inotropic effects of beta-agonists and betablockers on beta1-adrenergic receptors are diminished in the elderly [18–20]. The density of beta-receptors in the heart is unchanged in the elderly, but there is a decrease in the ability of

106 Table 3

Frishman et al. Characteristics of the Elderly Relative to Drug Response

Physiological changes Decreased cardiac reserve Decreased LV compliance due to thickened ventricular wall, increased blood viscosity, decreased aortic compliance, increased total and peripheral resistance Decreased baroreceptor sensitivity Diminished cardiac and vascular responsiveness to h-agonists and antagonists Suppressed renin-angiotensin-aldosterone system Increased sensitivity to anticoagulant agents Concurrent illnesses Multiple drugs Sinus and AV node dysfunction

Changes in response Potential for heart failure Decrease of cardiac output

Tendency to orthostatic hypotension Decreased sensitivity to h-agonists and antagonists Theoretically decreased response to ACE inhibitors, but not observed Increased effects of warfarin Increased drug–disease interactions Increased drug–drug interactions Potential for heart block

Abbreviations: LV, left ventricular; ACE, angiotensin converting enzyme; AV, atrioventricular. Source: Adapted from Ref. 157.

beta-receptor agonists to stimulate cyclic adenosine monophosphate production [21]. There are also age-related changes in the cardiac conduction system, as well as an increase in arrhythmias in the elderly. In the Framingham Study, the prevalence of atrial fibrillation was 1.8% in persons 60 to 69 years old, 4.8% in those 70 to 79 years old, and 8.8% in those 80 to 89 years old [22]. In a study of 1153 elderly patients (mean age 82 years), the prevalence of atrial fibrillation was 13% [23]. In elderly patients with unexplained syncope, a 24-h ambulatory electrocardiogram (ECG) should be obtained to rule out the presence of second- or third-degree atrioventricular block or sinus node dysfunction with pauses >3 not seen on the resting ECG. These phenomena were observed in 21 of 148 patients (14%) with unexplained syncope [24]. These 21 patients included 8 with sinus arrest, 7 with advanced second-degree atrioventricular block, and 6 with atrial fibrillation with a slow ventricular rate not drug induced. Unrecognized sinus node or atrioventricular node dysfunction may become evident in elderly persons after drugs such as amiodarone, beta-blockers, digoxin, diltiazem, procainamide, quinidine, or verapamil are administered. Clinical use of these drugs in the elderly, therefore, must be carefully monitored. USE OF CARDIOVASCULAR DRUGS IN THE ELDERLY Digoxin Digoxin has a narrow toxic–therapeutic ratio, especially in the elderly [25]. Decreased renal function and lean body mass may increase serum digoxin levels in this population. Serum creatinine may be normal in elderly persons despite a marked reduction in creatinine clearance, thereby decreasing digoxin clearance and increasing serum digoxin levels. Older persons are also more likely to take drugs that interact with digoxin by interfering with bioavailability and/or elimination. Quinidine, cyclosporin, itraconazole, calcium preparations, verapamil, amiodarone, diltiazem, triamterene, spironolactone, tetracycline, erythromycin, propafenone, and propantheline can increase serum digoxin levels. Hypokalemia,

Cardiovascular Drug Therapy

107

hypomagnesemia, hypercalcemia, hypoxia, acidosis, acute and chronic lung disease, hypothyroidism, and myocardial ischemia may also cause digitialis toxicity despite normal serum digoxin levels. Digoxin may also cause visual disturbances [26], depression, and confusional states in older persons, even with therapeutic blood levels. Indications for using digoxin are slowing a rapid ventricular rate in patients with supraventricular tachyarrhythmias such as atrial fibrillation, and treating patients with congestive heart failure (CHF) in sinus rhythm associated with abnormal LV ejection fraction that does not respond to diuretics and ACE inhibitors, or in patients unable to tolerate ACE inhibitors or other vasodilator therapy. Digoxin should not be used to treat patients with CHF in sinus rhythm associated with normal LV ejection fraction. By increasing contractility through increasing intracellular calcium ion concentration, digoxin may increase LV stiffness and increase LV filling pressures, adversely affecting LV diastolic dysfunction. Since almost half the elderly patients with CHF have normal LV ejection fractions [27,28], LV ejection fraction should be measured in all older patients with CHF, so that appropriate therapy may be given [29]. Many older patients with compensated CHF who are in sinus rhythm and are on digoxin may have digoxin withdrawn without decompensation in cardiac function [30,31]. Therapeutic levels of digoxin do not reduce the frequency or duration of episodes of paroxysmal atrial fibrillation detected by 24-h ambulatory ECGs [32]. In addition, therapeutic concentrations of digoxin do not prevent the occurrence of a rapid ventricular rate in patients with paroxysmal atrial fibrillation [32,33]. Many elderly patients are able to tolerate atrial fibrillation without the need for digoxin therapy because the ventricular rate is slow as a result of concomitant atrioventricular nodal disease. Some studies have suggested that digoxin may decrease survival after acute myocardial infarction (MI) in patients with LV dysfunction [34,35]. Leor et al. [36] showed that digoxin may exert a dose-dependent deleterious effect on survival in patients after acute MI, although other studies have not confirmed this finding [37,38]. Eberhardt et al. [39] demonstrated in the Bronx Longitudinal Aging Study that digoxin use in the elderly without evidence of CHF was an independent predictor of mortality. The results of the Digitalis Investigators Group Trial demonstrated that digoxin could be used in older subjects with CHF, but in lower doses than that previously employed in clinical practice [40]. Diuretics The Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure recommended as initial drug treatment of hypertension thiazidelike diuretics or beta-blockers because these drugs had been demonstrated to reduce cardiovascular morbidity and mortality in controlled clinical trials [41]. Moreover, the results of the Systolic Hypertension in the Elderly (SHEP) trial specifically show the safety and efficacy of a diuretic and beta-blocker in the treatment of isolated systolic hypertension in the elderly [42]. In the elderly, a blanket recommendation for the starting medication in the treatment of hypertension is ill-advised, in part because of the presence of comorbid conditions. For example, older hypertensive patients with systolic heart failure and a reduced LV ejection fraction [43–45] or in those elderly with a normal LV ejection fraction [46], therapy should include both a diuretic and an ACE inhibitor. Loop diuretics remain first-line drug therapy in the treatment of patients with decompensated CHF. Diuretics are multifaceted in their effect in CHF. First, they effect a reduction in plasma volume by triggering a time-dependent natriuretic response. This drop in plasma volume reduces venous return and thereby decreases ventricular filling pressures.

108

Frishman et al.

These volume changes facilitate relief of congestive symptomatology, such as peripheral and/or pulmonary edema. Intravenous loop-diuretic therapy has also been shown to increase central venous capacitance, which may further contribute to improvement in congestive symptomatology. Both loop and thiazidelike diuretics undergo a mixed pattern of renal/hepatic elimination with the component of renal clearance being responsible for diuresis [47]. Age-related decreases in renal function may reduce the efficacy of conventional doses of diuretics in elderly patients. This ‘‘renal function-related resistance’’ can be easily overcome, if recognized, by careful upward titration of the diuretic dose. Resistance to diuretic effect in CHF may also derive from a pattern of variable and unpredictable absorption, particularly with the loop diuretic furosemide. This issue is resolvable with the use of a predictably absorbed loop diuretic, such as torsemide [48]. A thiazidelike diuretic, such as hydrochlorothiazide, may be used in the occasional older patient with mild CHF. However, thiazidelike diuretics have diminished effectiveness at conventional doses when the glomerular filtration rate falls below 30 mL/min; accordingly, older patients with moderate-to-severe CHF should be treated with a loop diuretic, such as furosemide. Older patients with severe CHF or concomitant significant renal insufficiency may need combination diuretic therapy employing a loop diuretic together with the thiazidelike diuretic metolazone [47]. The slowly and erratically absorbed form of metolazone (ZaroxylynR) is the preferred form when combination therapy is being considered. Nonsteroidal anti-inflammatory drugs (NSAIDs) may decrease both the antihypertensive and natriuretic effect of loop diuretics [47]. This is a particular problem when loop diuretics are being employed t manage CHF-related congestive symptomatology [49]. A final consideration is the sometimes insidious manner by which NSAIDs can interact with diuretics in that several commonly used NSAIDs are now available over-the-counter. Serum electrolytes need to be closely monitored in older patients treated with diuretics. Hypokalemia and/or hypomagnesemia, both of which may precipitate ventricular arrhythmias and/or digitalis toxicity, can occur with diuretic therapy [50]. Hyponatremia is not uncommon in the elderly treated with diuretics, particularly when thiazidelike diuretics are being employed [51]. Older patients with CHF are especially sensitive to volume depletion with dehydration, hypotension, and prerenal azotemia occurring in the face of excessive diuretic effect. Older patients with CHF and normal LV ejection fraction should receive diuretics more cautiously. Beta-Adrenergic Blockers Beta-blockers are used in various cardiovascular disorders, with resultant beneficial and adverse effects [52]. Beta-blockers are very effective antianginal agents in older, as well as younger, patients. Combined therapy with beta-blockers and nitrates may be more beneficial in the treatment of angina pectoris than either drug alone [52]. Diuretics or beta-blockers have been recommended as initial drug therapy for hypertension in older persons because these drugs have been shown to decrease cardiovascular morbidity and mortality in controlled clinical trials [41,53]. Beta-blockers are especially useful in the treatment of hypertension in older patients who have had a prior MI, angina pectoris, silent myocardial ischemia, complex ventricular arrhythmias, supraventricular tachyarrhythmias, or hypertrophic cardiomyopathy. Teo et al. [54] analyzed 55 randomized controlled trials that investigated the use of beta-blockers in patients after MI. Mortality was significantly decreased (19%) in patients receiving beta-blockers compared with control patients. In the Beta-Blocker Heart Attack Trial (BHAT), propranolol significantly decreased total mortality by 34% in patients aged

Cardiovascular Drug Therapy

109

60 to 69 years, and insignificantly reduced total mortality by 19% in patients aged 30 to 59 years [55]. In the Norwegian Timolol Study, timolol significantly decreased total mortality by 43% in postinfarction patients aged 65 to 75 years, and significantly reduced total mortality by 34% in postinfarction patients90%), in decreasing the average number of ventricular premature complexes per hour (>70%), and in abolishing silent ischemia. Multivariate Cox regression analysis showed that propranolol caused a significant 47% decrease in sudden cardiac death, a significant 37% reduction in total cardiac death, and an insignificant 20% decrease in total death [61]. Univariate Cox regression analysis showed that the reduction in mortality and complex ventricular arrhythmias in elderly patients with heart disease taking propranolol was due more to an anti-ischemic effect than to an antiarrhythmic effect [62]. Table 4 also shows that there was a circadian distribution of sudden cardiac death or fatal MI, with the peak incidence occurring from 6 a.m. to 12 a.m. (peak hour 8 a.m. and secondary peak around 7 p.m.) in patients treated with no

110

Frishman et al.

Table 4 Effect of h-Blockers on Mortality in Elderly Patients with Complex Ventricular Arrhythmias and Heart Disease Age (yrs)

Mean follow-up (mos)

BHAT (55)

60-69 (33%)

25

Hallstrom (65)

62 (mean)

Aronow et al. (61)

62-96 (mean 81)

29

Aronow et al. (62)

62-96 (mean 81)

29

Aronow et al. (66)

62-96 (mean 81)

29

CAST (67)

66-79 (40%)

12

Study

108

Results Compared with placebo, propranolol reduced sudden cardiac death by 28% in patients with complex VA and 16% in patients without VA. Reduced incidence of death or recurrent cardiac arrest in patients treated with h-blockers vs. no antiarrhythmic drug (adjusted relative risk 0.62). Compared with no antiarrhythmic drug, propranolol caused a 47% significant decrease in sudden cardiac death, a 37% significant reduction in total cardiac death, and a 20% insignificant decrease in total death. Among patients taking propranolol, suppression of complex VA caused a 33% reduction in sudden cardiac death, a 27% decrease in total cardiac death, and a 30% reduction in total death; abolition of silent ischemia caused a 70% decrease in sudden cardiac death, a 72% reduction in total cardiac death, and a 69% decrease in total death. Incidence of sudden cardiac death or fatal myocardial infarction was significantly increased between 6 A.M. and 12 A.M. with peak hour at 8 A.M. and secondary peak at 7 P.M. in patients with no antiarrhythmic drug; propranolol abolished the circadian distribution of sudden cardiac death or fatal myocardial infarction. Patients on h blockers (30% of study group) had a significant reduction in all-cause mortality of 43% at 30 days, 46% at 1 year, and 33% at 2 years: in patients on h-blockers, arrhythmic death or cardiac arrest was significantly reduced by 66% at 30 days, 53% at 1 year, and 36% at 2 years; multivariate analysis showed h-blockers to be an independent factor for reduced arrhythmic death or cardiac arrest by 40% and for all-cause mortality by 33%.

Abbreviations: BHAT, Beta Blocker Heart Attack Trial; VA, ventricular arrhythmias; CAST, Cardiac Arrhythmia Suppression Trial. Source: Reproduced from Ref. 159.

Cardiovascular Drug Therapy

111

antiarrhythmic drug [66]. Propranolol abolished this circadian distribution of sudden cardiac death or fatal MI [66]. In a retrospective analysis of data from the Cardiac Arrhythmia Suppression Trial (CAST), Kennedy et al. [67] found that 30% of patients with an LV ejection fraction V40% were receiving beta-blockers. Forty percent of these 1735 patients were between 66 and 79 years of age. Patients on beta-blockers had a significant reduction in all-cause mortality of 43% within 30 days, 46% at 1 year, and 33% at 2 years. Patients receiving beta-blockers also had a significant decrease in arrhythmic death or cardiac arrest of 66% at 30 days, 53% at 1 year, and 36% at 2 years. Multivariate analysis showed that beta-blockers were an independent factor for reducing arrhythmic death or cardiac arrest by 40%, for decreasing all-cause mortality by 33%, and for reducing the occurrence of new or worsening CHF by 32%. On the basis of these data (55,61,62,65–67), beta-blockers can be utilized in the treatment of older and younger patients with ventricular tachycardia or complex ventricular arrhythmias associated with ischemic or nonischemic heart disease, and with normal or abnormal LV ejection fraction, if there are no absolute contraindications to the drugs. Beta-blockers are also useful in the treatment of supraventricular tachyarrhythmias in older and younger patients [68,69]. If a rapid ventricular rate associated with atrial fibrillation persists at rest or during exercise despite digoxin therapy, then verapamil [70], diltiazem [71], or a beta-blocker [72] should be added to the therapeutic regimen. These drugs act synergistically with digoxin to depress conduction through the atrioventricular junction. The initial oral dose of propranolol is 10 mg q6h, which can be increased to a maximum of 80 mg q6h if necessary. Beta-blockers have been demonstrated to reduce mortality in older persons with New York Heart Association Class II–IV CHF and abnormal LV ejection fraction treated with diuretics and ACE inhibitors with or without digoxin [73–77]. Beta-blockers have also been shown to reduce mortality in older persons with New York Heart Association Class II–III CHF and normal LV ejection fraction treated with diuretics plus ACE inhibitors [78]. Numerous drug interactions have been reported with beta-blockers in the elderly [52]. Recently, quinidine, a known inhibitor of CYP2D6, was shown to decrease the hepatic metabolism of topically applied ophthalmic timolol, with resultant exaggeration of the betablocking effect of timolol [79]. ACE Inhibitors ACE inhibitors are effective antihypertensive agents. A meta-analysis of 109 treatment studies showed that ACE inhibitors are more effective than other antihypertensive drugs in decreasing LV mass [80]. Older hypertensive patients with CHF associated with abnormal [43–45] or normal [46] LV ejection fraction, LV hypertrophy, or diabetes mellitus, should initially be treated with an ACE inhibitor. ACE inhibitors reduce mortality in patients with CHF associated with abnormal LV ejection fraction [43–45]. The Survival and Ventricular Enlargement (SAVE) trial [81] and the combined Studies of Left Ventricular Dysfunction (SOLVD) treatment and prevention trials [82] also demonstrated that ACE inhibitors such as captopril or enalapril should be standard therapy for most patients with significant LV systolic dysfunction with or without CHF. In addition, ACE inhibitor therapy has been shown to be beneficial in the treatment of elderly patients (mean age 80 years) with CHF caused by prior MI associated with normal LV ejection fraction [46]. High-dose ACE inhibitor therapy remains standard-of-care in the management of CHF. Low-dose ACE inhibitor therapy has been studied in CHF, but with

112

Frishman et al.

less favorable results. For example, a recent trial compared low (2.5–5.0 mg/day) with highdose lisinopril (32.5–35.0 mg/day), with the latter being associated with a more significant reduction in mortality and all-cause hospitalization rate [83]. Treatment with ACE inhibitors should be initiated in elderly patients in low doses after correction of hyponatremia or volume depletion. It is important to avoid overdiuresis before beginning therapy with ACE inhibitors since volume depletion may cause hypotension or renal insufficiency when ACE inhibitors are begun or when the dose of these drugs is increased to full therapeutic levels. After the maintenance dose of ACE inhibitor is reached, it may be necessary to increase the dose of diuretics. The initial dose of enalapril is 2.5 mg daily and of captopril is 6.25 mg TID. The maintenance doses are 5 to 20 mg daily and 25 to 50 mg TID, and the maximum doses are 20 mg BID and 150 mg TID, respectively. Older patients at risk for excessive hypotension should have their blood pressure monitored closely fo the first 2 weeks of ACE inhibitor or angiotensin II receptor blocking therapy, and thereafter whenever the dose of ACE inhibitor or diuretic is increased. Renal function should be monitored in patients on ACE inhibitors to detect increases in blood urea nitrogen and serum creatinine, especially in older patients with renal artery stenosis. A rise in serum creatinine in an ACE inhibitor–treated congestive heart failure patient is not uncommonly the result of ACE inhibitor–induced alterations in renal hemodynamics. There is no specific rise in serum creatinine where corrective actions need be taken although logic would suggest the greater the increment in serum creatinine the more important the intervention. Typically, reducing or temporarily discontinuing diuretics and/or liberalizing sodium intake are sufficient measures to return renal function to baseline. Not uncommonly, though, the administered ACE inhibitor is either stopped or the dose reduced. In most instances, ACE inhibitor therapy can be safely resumed as long as careful attention is paid to patient volume status [84]. Potassium-sparing diuretics or potassium supplements should be carefully administered to patients receiving ACE inhibitor therapy because of the attendant risk of hyperkalemia. In this regard, the recently concluded Randomized Aldactone Evaluation Study (RALES) showed that in persons with severe CHF treated with diuretics, ACE inhibitors, and digoxin compared with placebo, spironolactone 25 mg/ day did not carry an excessive risk of hyperkalemia, while resulting in a significant reduction in mortality and hospitalization for CHF [85]. Angiotensin-II Receptor Blockers Angiotensin-II receptor blockers (ARBs) are the newest class of antihypertensive drugs to be approved. They have been studied fairly extensively in hypertension [86–88], diabetic nephropathy [89], and CHF [90], with results comparable to those seen when these disease states are treated with ACE inhibitors. Although the published experience with these drugs in the elderly is limited, the drugs appear to be safe if used with similar precautions as those recommended for ACE inhibitors, as described above [86,88]. These drugs are noteworthy in that they have a much more favorable side-effect profile and, in particular, are not associated with cough, a fairly common side effect with ACE inhibitor therapy [87]. Likewise, in the Losartan Heart Failure Survival Study (ELITE II), losartan was associated with fewer adverse effects than was captopril [91]. Outcomes studies are supportive of ARBs, such as losartan and irbesartan, being superior to conventional non-ACE-inhibitor– based therapy in decreasing end-stage renal failure event rates in patients with type II diabetic nephropathy [92,93]. In CHF the hope that ARBs are more effective therapy than ACE inhibitors has not been realized, as of yet, although additional studies are underway to establish the positioning of ARB in current heart failure regimens.

Cardiovascular Drug Therapy

113

Nitrates Nitrates are effective therapies for older individuals; however, caution should always be used because of the associated dangers of orthostatic hypotension, syncope, and falls, especially if the treatment is combined with diuretics and other vasodilators. Recently, it was shown that nitrate headaches are less frequent in older patients and in individuals with renal dysfunction [94]. Calcium Channel Blockers Calcium channel blockers are effective antihypertensive and antianginal drugs in older patients. Verapamil [70] and diltiazem [71] are especially valuable in treating hypertensive patients who also have supraventricular tachyarrhythmias. However, recent reports have suggested an increased mortality risk with calcium channel blockers, especially with the use of short-acting dihydropyridines in older subjects [95–97]. With the use of longer-acting calcium blockers, such as the dihydropyridine nitrendipine, a strong mortality benefit was seen in patients with isolated systolic hypertension [98], although many were receiving concurrent beta-blocker therapy. In contrast, nisoldipine was shown to be less effective in protecting against cardiovascular mortality in diabetic patients with hypertension when compared to an enalapril-treated group [99]. Verapamil improved exercise capacity, peak LV filling rate, and a clinicoradiographic heart failure score in patients with CHF, normal LV ejection fraction, and impaired LV diastolic filling [100]. However, calcium channel blockers such as verapamil, diltiazem, and nifedipine may exacerbate CHF in patients with associated abnormal LV ejection fraction [101]. In addition, some calcium channel blockers have been shown to increase mortality in patients with CHF and abnormal LV ejection fraction after myocardial infraction [102]. Therefore, calcium channel blockers such as verapamil, diltiazem and nifedipine may be used to treat older patients with CHF associated with normal LV ejection fraction, but are contraindicated in treating older patients with CHF associated with abnormal LV ejection fraction. Amlodipine and felodipine are two vasculospecific dihydropyridine agents that appear to be safer in patients having CHF, although neither of these drugs should be used to exclusion of ACE inhibitors in the CHF patient. The age-associated decrease in hepatic blood flow and hepatic metabolic capacity may result in higher serum concentrations of verapamil, diltiazem, and nifedipine [103]. Therefore, these drugs should be given to older persons in lower starting doses and titrated carefully. Alpha-Adrenergic Blockers Alpha-adrenergic blockers are effective treatments for patients with hypertension and have become first-line treatments for males with symptomatic prostatism. Caution should be exercised when using these agents because of a significant incidence of postural hypotension, especially in patients receiving diuretics or other vasodilator drugs [104,105]. A more selective alpha-blocker, tamsulosin, has become available, which improves prostatism symptoms without having vasodilator effects [106]. Recently, the National Heart, Lung and Blood Institute announced that doxazosin was being withdrawn from the ALLHAT trial after an interim analysis showed a 25% greater rate of a secondary endpoint, combined cardiovascular disease, in patients on doxazosin than in those on chlorthalidone, largely

114

Frishman et al.

driven by the increased risk of CHF [107]. These findings have cast a shroud over the use of doxazosin in the elderly, particularly if it is being contemplated as monotherapy in an elderly hypertensive. Lidocaine I.V. lidocaine may be used to treat complex ventricular arrhythmias during acute myocardial infarction [68]. Lidocaine toxicity is more common in the elderly and older patients should be monitored for dose-related confusion, tinnitus, paresthesias, slurred speech, tremors, seizures, delirium, respiratory depression, and hypotension. Older patients with CHF or impaired liver function are at increased risk for developing central nervous system adverse effects from lidocaine [108]. In these patients, the loading dose should be decreased by 25 to 50%, and any maintenance infusion should be initiated at a rate of 0.5 to 2.5 mg/ min, with the patient monitored closely for adverse effects. The dose of lidocaine should also be reduced if the patient is receiving beta-blockers [109] or cimetidine, since these drugs reduce the metabolism of lidocaine. Other Antiarrhythmic Drugs The use of antiarrhythmic drugs in the elderly is extensively discussed elsewhere [69,110]. In the CAST I trial, encainide and flecainide significantly increased mortality in survivors of myocardial infarction with asymptomatic or mildly symptomatic ventricular arrhythmias, when compared to placebo [111]. In the CAST II trial, moricizine insignificantly increased mortality when compared to placebo [112]. Akiyama et al. [113] found that older age increased the likelihood of adverse events, including death, in patients treated with encainide, flecainide, or moricizine in these two studies. In a retrospective analysis of the effect of empirical antiarrhythmic treatment in 209 cardiac arrest patients who were resuscitated out of hospital, Moosvi et al. [114] found that the 2-year mortality was significantly lower for patients treated with no antiarrhythmic drug than for patients treated with quinidine or procainamide. Hallstrom et al. [65] showed an increased incidence of death or recurrent cardiac arrest in patients treated with quinidine or procainamide versus no antiarrhythmic drug. In a prospective study of 406 elderly subjects (mean age 82 years) with heart disease (58% with prior myocardial infarction) and asymptomatic complex ventricular arrhythmias, the incidence of sudden cardiac death, total cardiac death, and total mortality were not significantly different in patients treated with quinidine or procainamide or with no antiarrhythmic drug [115]. In this study, quinidine or procainamide did not reduce mortality in comparison with no antiarrhythmic drug in elderly patients with presence versus absence of ventricular tachycardia, ischemic or nonischemic heart disease, and abnormal or normal LV ejection fraction. The incidence of adverse events causing drug cessation in elderly patients in this study was 48% for quinidine and 55% for procainamide. A meta-analysis of six double-blind studies of 808 patients with chronic atrial fibrillation who underwent direct current cardioversion to sinus rhythm demonstrated that the 1-year mortality was significantly higher in patients treated with quinidine than in patients treated with no antiarrhythmic drug [116]. In the Stroke Prevention in Atrial Fibrillation Study, arrhythmic death and cardiac mortality were also significantly increased in patients receiving antiarrhythmic drugs compared with patients not receiving antiarrhythmic drugs, especially in patients with a history of CHF [117].

Cardiovascular Drug Therapy

115

Teo et al. [54] analyzed 59 randomized controlled trials comprising 23,229 patients which investigated the use of class I antiarrhythmic drugs after MI. Patients receiving class I antiarrhythmic drugs had a significantly higher mortality than patients receiving no antiarrhythmic drugs. None of the 59 trials demonstrated that a class I antiarrhythmic drug decreased mortality in postinfarction patients. Therefore, it is currently not recommended that class I antiarrhythmic drugs be used for the treatment of ventricular tachycardia or complex ventricular arrhythmias associated with heart disease. Amiodarone is very effective in suppressing ventricular tachycardia and complex ventricular arrhythmias. However, there are conflicting data about the effect of amiodarone on mortality [118–124]. The Veterans Administration Cooperative Study comparing amiodarone versus placebo in heart failure patients with malignant ventricular arrhythmias was recently completed and showed that amiodarone was very effective in decreasing ventricular tachycardia and complex ventricular arrhythmias, but it did not affect mortality [124]. The incidence of adverse effects from amiodarone has been reported to approach 90% after 5 years of treatment [125]. In the Cardiac Arrest in Seattle: Conventional Versus Amiodarone Drug Evaluation Study, the incidence of pulmonary toxicity was 10% at 2 years in patients receiving an amiodarone dose of 158 mg daily [126]. On the basis of these data, one should reserve the use of amiodarone for the treatment of life-threatening ventricular tachyarrhythmias or in patients who cannot tolerate or who do not respond to beta-blocker therapy. Amiodarone is also the most effective drug for treating refractory atrial fibrillation in terms of converting atrial fibrillation to sinus rhythm and slowing a rapid ventricular rate. However, because of the high incidence of adverse effects caused by amiodarone, amiodarone should be used in low doses in patients with atrial fibrillation when life-threatening atrial fibrillation is refractory to other therapy [127]. Lipid-Lowering Drugs The safety of lipid-lowering drugs, specifically 3-hydroxy-3-methylglutaryl-coenzyme A (HMG CoA) reductase inhibitors, was demonstrated in the Cholesterol Reduction in Seniors Program (CRISP) [128]. Based on CRISP, the Antihypertensive Lipid Lowering Heart Attack Trial (ALLHAT), a large primary prevention trial is now in progress using pravastatin in older subjects to determine whether it reduces morbidity and mortality [129]. In the Scandinavian Simvastatin Survival Study [130], 4444 men and women with coronary artery disease were treated with double-blind simvastatin or placebo. At 5.4 years follow-up, patients treated with simvastatin had a 35% decrease in serum low-density lipoprotein cholesterol, a 25% reduction in serum total cholesterol, an 8% increase in serum high-density lipoprotein cholesterol, a 34% decrease in major coronary events, a 42% reduction in coronary deaths, and a 30% decrease in total mortality. In patients aged 65 to 70 years, simvastatin reduced all-cause mortality 35%, coronary artery disease mortality 43%, major coronary events 34%, nonfatal MI 33%, any atherosclerosis-related endpoint 34%, and coronary revascularization 41% [131]. The absolute risk reduction for both allcause mortality and coronary artery disease mortality was approximately twice as great in persons aged 65 to 70 years compared to persons younger than 65 years [131]. In the Cholesterol and Recurrent Events Trial [132], 4159 men and women aged 21 to 75 years (1283 aged 65–75 years) with MI, serum total cholesterol levels140/90 mmHg [1,18]. Higher levels of either systolic or diastolic blood pressure are associated with an increased risk of cardiovascular morbidity or mortality. Examples include coronary artery disease, congestive heart failure, renal insufficiency, peripheral vascular disease, and stroke. Hypertension in the elderly poses an important problem for two reasons: (1) the proportion of elderly in the population is steadily increasing, and (2) the incidence of hypertension—and therefore the risk of cardiovascular disease—increases with age [19]. In the past, there was a tendency to deemphasize elevated blood pressure readings in the elderly and to downplay the need for therapeutic intervention. Increasing blood pressure was viewed as part of the normal aging process, and blood pressure measurements that would dictate treatment in young adults were often considered satisfactory or borderline in elderly patients. It was believed that the benefits of antihypertensive therapy might not be realized in patients with

Randomized placebo-controlled double-blind Randomized placebo-controlled single-blind Randomized placebo-controlled single-blind

Randomized placebo-controlled double-blind Randomized non-blind

Study Design

4396

4736

1627

884

840

39

43

z60 65–74

37

70–84

31

30

z60 60–79

Males (yr)

Age (yr)

5.8

4.5

2.1

4.4

4.7

Duration (yr)

160–209

160–219

160 Other Age 50–69 yr BP: diastolic >110 or >100 (sitting) if treated or other risk factors present

NORDIL 5 years 10,881 pts

PREDICT 4–4.5 years 8000–8500 pts

Age z 55 yr BP: 90–109/140–179 Other

SISH 2 years 171 pts

Age >55 yr BP: systolic 140–159

STOP-2 4 years 6600 pts

Age 70–84 yr BP >180/105 (sitting)

Therapy

Endpoints

1st: amlodipine, lisinopril, doxazosin, or chlorthalidone + pravastatin 2nd: reserpine, clonidine, atenolol, or hydralazine at physician’s discretion 1st: verapamil, atenolol, or HCTZ 2nd: HCTZ or beta blocker for achievement of BP goal 3rd: moexipril 1st: felodipine + aspirin or placebo 2nd: captopril, enalapril, ramipril, atenolol, meto prolol, or propranolol, as needed 1st: long-acting nifedipine or HCTZ-amiloride 2nd: atenolol or enalapril

1st: fatal/nonfatal MI 2nd: morbidity, mortality, cost, benefits of cholesterol reduction

1st: diltiazem, diuretic, or beta blocker 2nd: ACE inhibitor, beta blocker, or diuretic 3rd: ACE inhibitor, beta blocker, or diuretic

1st: diltiazem or chlorthalidone 2nd: open-label therapy 3rd: open-label therapy 1st: felodipine or placebo

1st: felodipine, isradipine, atenolol, metoprolol, pindolol, enalapril, lisinopril, or HCTZ-amiloride

1st: fatal/nonfatal MI, stroke; sudden cardiac death 2nd: cardiovascular disease, number and time of deaths 1st: fatal/nonfatal MI, stroke; achievement of BP goal 2nd: benefits of aspirin

1st: fatal/nonfatal MI, stroke; congestive heart failure 2nd: mortality 1st: fatal/nonfatal MI, stroke; sudden cardiac death 2nd: mortality, ischemic heart disease or attack, atrial fibrillation, congestive heart failure, renal disease 1st: fatal/nonfatal cardiovascular disease

1st: compare effects on EKG measurements of ventricular wall thickness and ventricular function 1st: fatal/nonfatal MI, stroke; sudden cardiac arrest 2nd: cardiovascular disease, cost, institutionalization, tolerability of therapy

Systemic Hypertension Table 3

141

Continued

Trial Syst-Eur 5 years 3000 pts Syst-China STONE

Entry criteria

Therapy

Endpoints

Age z 60 yr BP: systolic 160–219 or diastolic 60 years)

11.9 3.0 9.5 20.0 — —

12.1

Men

4.6 4.9 4.9 15.2 — —

16.0

Women

Elderly Patients with hypertension (%) (age >65 years)

16 40 36 23 —

Angina and/or MI: 20

Residents of nursing homes (%) (age 65–74 yrs)

Abbreviations: MI, myocardial infarction; CHF, congestive heart failure; SHEP, Systolic Hypertension in the Elderly Program, Syst-Eur, Systolic Hypertension in Europe; Syst-China, Systolic Hypertension in the Elder Chinese, —, data not available. Source: From Ref. 90.

Men

Disease

General population (%)

Table 4 Incidence of Comorbid Conditions Observed in the General Elderly Population, Hypertension Trial Participants, Elderly Patients with Hypertension, and Nursing Home Residents with Hypertension

Systemic Hypertension 143

144

Frishman et al.

Angiotensin Converting Enzyme Inhibitors Angiotensin converting enzyme (ACE) inhibitors block the conversion of angiotensin-I to angiotensin-II both systemically and locally in arterioles to lower total peripheral vascular resistance. Blood pressure is reduced without reflex stimulation of heart rate and cardiac output. As aging occurs, angiotensin levels are lower and, theoretically, ACE inhibitors should not be as effective as other therapies. However, multiple studies have shown otherwise. The Captopril Prevention Project (CAPPP) compared captopril to a beta-blocker in 10,925 older patients with diastolic blood pressure >100 mmHg [79]. Patients receiving captopril had a lower rate of cardiovascular death; patients on beta-blockers had a lower stroke rate. Other studies previously mentioned (ALLHAT and STOP-2) [39,43] showed that ACE inhibitors can reduce the risk of stroke and coronary heart disease as well as other antihypertensive drugs, although in ALLHAT there was less of a reduction, especially in Black patients. ACE inhibitors have been shown to reduce mortality in patients who have already had a stroke [80] and have been shown to reduce mortality in patients with heart failure and poor LV function [81]. The drugs appear to reduce the progression of renal disease in type 1 diabetic patients with proteinuria [82]. The HOPE trial evaluated the effects of ramipril in 9297 patients over the age of 55 who were both normotensive and hypertensive in a placebo-controlled trial [83]. Subjects had a previous history of stroke, coronary disease, peripheral vascular disease, and diabetes mellitus without heart failure. After 5 years, the endpoints of MI, stroke, or cardiovascular death were evaluated. Ramipril was associated with a lower blood pressure than placebo, and reduced the risk of the combined primary outcome by 25%, MI by 22%, stroke by 33%, cardiovascular death by 37%. Given all the available data, ACE inhibitors should be considered as drugs of choice in elderly hypertensive patients with heart failure and/or diabetes mellitus. The main adverse effects of ACE inhibitors include excessive hypotension, chronic dry cough, and, rarely, angioedema or rash. Renal failure can develop in those with renal artery stenosis. Hyperkalemia can occur in patients with renal insufficiency or those taking potassium supplements. Therefore, these agents must be used carefully in patients with renal impairment. Rarely neutropenia or agranulocytosis can occur; therefore, close monitoring is suggested the first few months of therapy.

Angiotensin Receptor Blockers The angiotensin receptor blockers (ARB) are a relatively new class of drugs employed in the treatment of hypertension. These agents work selectively at the AT1-receptor subtype, the receptor that mediates all the known physiological effects of angiotensin-II that are believed to be relevant to cardiovascular and cardiorenal homeostasis. A large-scale open-label clinical experience trial (ACTION study) showed the effectiveness of the ARB candesartan cilexetil alone or in combination with other agents in 6465 patients with a mean age of 58 who had either combined or isolated systolic hypertension [84]. The LIFE study compared losartan to atenolol in 9193 patients aged 55 to 80 years with essential hypertension and LVH [44]. The study showed a substantially reduced rate of stroke in the losartan-treated group despite comparably reduced blood pressure reduction in both the losartan and atenolol treatment groups. A primary composite outcome of death,

Systemic Hypertension

145

stroke, and cardiovascular mortality showed a significant benefit in favor of losartan. There was also a greater effect on LVH regression with losartan compared to atenolol. Overall, safety and efficacy trials have shown that ARBs are similar to other agents in reducing blood pressure and are well tolerated. ARBs have also been found in recently completed clinical trials to be useful in protecting the kidney in the setting of type 2 diabetes [44], both in established diabetic nephropathy with proteinuria and in microalbuminuria diabetic patients [82]. The drugs have also been shown to be useful in reducing mortality and morbidity in patients with congestive heart failure [82]. In elderly patients, ARBs can be considered a first-line treatment in hypertensive patients with type 2 diabetes, and as an alternative to ACE inhibitors in hypertensive patients with heart failure who cannot tolerate ACE inhibition. A study is being done to look at the efficacy of the ARB, candesartan, for left ventricular diastolic dysfunction. Alpha-Adrenergic Blocking Agents In large comparative clinical trials, the efficacy and safety of alpha-blockers have been well documented [46]. Doxazosin 2 to 8 mg daily was one of the drugs used in the ALLHAT trial, and that arm of the study was discontinued after an interim analysis showed a 25% greater rate of cardiovascular events compared to chlorthalidone, largely driven by congestive heart failure [85]. Based on this study, alpha-blockers should not be considered as first-line monotherapy treatment for hypertension, but as part of a combination regimen to provide maximal blood pressure control. These drugs are a treatment of choice for symptomatic prostate hypertrophy and caution should always be used because of their potent hypotensive actions. Centrally Acting Agents Centrally acting agents (e.g., clonidine) are not first-line treatments in the elderly because many patients experience troublesome sedation. They can be used as part of a combination regimen to maximize blood pressure control. Other Antihypertensive Drugs Under consideration for approval in hypertension is eplerenone, a selective aldosterone antagonist. Aldosterone antagonists as diuretics have been combined with thiazides as part of an antihypertensive regimen. Drugs in this class should be used with caution in elderly patients with renal disease. Combination Therapy Treating hypertension with combination therapy provides more opportunity for creative solutions to a number of problems. Five issues in combined therapy—some practical, some speculative—are listed in Table 5 [86]. The most obvious benefit of drug combinations is the enhanced efficacy that fosters their continued widespread use. Theoretically, some drug combinations might produce synergistic effects that are greater than would be predicted by summing the efficacies of the component drugs. More commonly, combination therapy achieves a little less than the sum of its component drug efficacies. In contrast, some combinations of drugs produce

146

Frishman et al.

Table 5 Rationale for Combination Drug Therapy for Hypertension Increased antihypertensive efficacy Additive effects Synergistic effects Reduced adverse events Low-dose strategy Drugs with offsetting actions Enhanced convenience and compliance Prolonged duration of action Potential for additive target organ protection Source: From Ref. 86 with permission.

offsetting interactions that weaken rather than strengthen their antihypertensive effects, as previously seen with agents affecting peripheral-neuronal actions. For instance, guanethidine and reserpine produced this offsetting interaction; but since these agents are used now only rarely, this issue need not be considered further. A second benefit of combination therapy concerns the avoidance of adverse effects. When patients are treated with two drugs, each drug can be administered in a lower dose that does not produce unwanted side effects but still contributes to overall efficacy. A third issue concerns convenience. Obviously, the multiple drugs of a combination regimen could be confusing and distracting to elderly patients, and could lead to poor treatment compliance. On the other hand, a well-designed, combination pill that incorporates logical doses of two agents could enhance convenience and improve compliance. Further potential value of combination treatment may result from the effects that two drugs have on each other’s pharmacokinetics. Although this has not been studied well, there might be situations where the clinical duration of action of the participating drugs becomes longer when used in combination than when they are administered as monotherapies. Finally, it is interesting to consider the attributes of such agents as ACE inhibitors, ARBs, and calcium channel blockers, which exhibit anti-growth or anti-atherosclerotic actions in addition to their blood pressure–lowering properties. Is it possible that combinations of these newer agents may provide even more powerful protective effects on the circulation?

CONCLUSION A meta-analysis of nine trials with more than 15,000 elderly subjects reported that allcause mortality, stroke mortality, and coronary mortality were reduced by lowering blood pressure [87]. Elderly patients with hypertension, especially isolated systolic hypertension, are the demographic group most likely to have uncontrolled hypertension. Despite clear evidence in elderly patients that morbidity can be reduced with antihypertensive drugs, clinicians are less aggressive with pharmacotherapy than they could be. The elderly also have concomitant diseases that often warrant major reductions in systolic blood pressure. Decreasing blood pressure in this age groups may be difficult. It has to be done carefully and although all of the guidelines suggest that the goal of systolic blood pressure should be 75 years of age) are uncertain and debatable. The FTT overview showed no statistically significant survival benefit in patients over the age of 75 years. Presumably, this was due to the increased incidence of hemorrhagic stroke. Two more recent population-based analyses suggest that short-term mortality from STEMI is not reduced and may even be increased in patients over 75 years who receive thrombolytic therapy [109,110]. Thus, the use of this form of reperfusion therapy in very elderly patients with and without diabetes remains controversial. For patients up to age 75, however, there is clear evidence from multiple randomized controlled trials that TL therapy affords a significant survival benefit. Primary Percutaneous Coronary Intervention Acute success rates with (PPCI) are comparable in the presence or absence of diabetes, although the restenosis and long-term mortality rates are higher in diabetic patients [111,112]. Despite the well-known limitations of PPCI (lack of widespread availability and treatment delays), accumulating evidence suggests that this form of reperfusion therapy may well be superior to TL in improving clinical outcomes if performed expeditiously by experienced operators [113]. Data comparing TL with PPCI in elderly diabetic patients are limited. A recent observational study comparing early and late outcomes of PPCI and TL therapy of diabetic patients with STEMI concluded that PPCI was associated with reduced early and late adverse outcomes when compared to TL therapy [114]. However, mortality was similar in both groups. In this study over one-third of the patients were 65 years or older. From the limited data available, it is reasonable to conclude that PPCI may be preferable to TL in the elderly diabetic patient with STEMI, especially for patients over the age of 75.

Diabetes Mellitus

173

Management of Unstable Angina and Non-ST-Elevation Myocardial Infarction (UA/NSTEMI) Beginning with the TIMI III study of early invasive versus early conservative management of UA, subgroup analyses of several randomized controlled trials have demonstrated a consistent, modest benefit of early invasive therapy in elderly patients (>65 years of age) and in diabetics. In the FRISC II trial [115] the reduction in the composite endpoint of death and nonfatal myocardial infarction among those assigned to the early invasive group resulted entirely from the benefit realized by the over-65-year subgroup. In this trial, the overall results of the two strategies were similar for both diabetics and nondiabetics. The more recent TACTICS-TIMI 18 study compared outcomes of patients with ACS treated with the glycoprotein IIb/IIIa inhibitor tirofiban and assigned to either an early invasive strategy and revascularization if clinically indicated or to a more conservative strategy in which cardiac catheterization was undertaken only if the patient demonstrated spontaneous or provoked ischemia [116]. Although there was no clear survival benefit, the results demonstrated a 3.1% absolute and 33% relative risk reduction of the combined endpoint of death, nonfatal MI, and rehospitalization for ACS at 6 months in patients assigned to an early invasive strategy. Subgroup analysis demonstrated that elderly patients over the age of 65 years had a 4.6% absolute reduction in the combined endpoint compared to a 2.9% reduction for patients under the age of 65. Similarly, diabetic patients realized a 7.6% absolute reduction compared to 2.2% for nondiabetic patients. Based on these and other trials, current thinking is that an early invasive strategy is the preferred method of managing ‘‘high-risk’’ patients with UA/NSTEMI. Since both advancing age and diabetes increase the risk of adverse outcomes, it is reasonable to conclude that an early invasive strategy is appropriate in the elderly diabetic patient with UA/NSTEMI. Adjunctive Therapies in Acute Coronary Syndromes In almost all respects, the adjunctive therapy of ACS is the same for patients with and without diabetes mellitus [117]. However, specific contraindications and risks must be recognized and will be mentioned below. Aspirin The beneficial effects of aspirin in ACS have been conclusively demonstrated in many studies. In the ISIS-2 study, aspirin was shown to be as efficacious in older as in younger patients with myocardial infarction [118]. Consequently, it is generally accepted that aspirin in doses of 162 to 325 mg should be administered to all patients with ACS and should be continued indefinitely if there are no major contraindications (e.g., bleeding peptic ulcer). Antithrombotic Therapy Intravenous unfractionated or low-molecular-weight heparin (LMWH) should be administered to all patients with ACS regardless of age and diabetic state unless there are major contraindications (e.g., intracranial bleed). Recent studies have shown that LMWH are superior to unfractionated heparin in patients with UA/NSTEMI and especially so in elderly patients [119,120]. Concerns regarding excess bleeding when patients who received LMWH undergo an invasive procedure have not been substantiated. Other Antiplatelet Therapy Platelet aggregability is enhanced in patients with diabetes mellitus. Several studies have shown that glycoprotein IIb/IIIa inhibitors, when added to aspirin, improve outcomes in high-risk patients with UA/NSTEMI, especially those undergoing PCI. A review of the results of four abciximab trials [121] reveals that this agent is safe and effective in patients over the age of 70 years, although the net benefit

174

Gregoratos and Leung

declines slightly with increasing age. Tirofiban and Eptifibatide have also been shown to be effective agents in this setting. Clopidogrel was shown in the CURE trial to improve long-term outcomes when administered for 3 to 9 months to patients with UA/NSTEMI who did not undergo PCI [122]. This improvement in outcomes was demonstrated across a number of subgroups, including patients over the age of 65 years and patients with diabetes mellitus. However, an excess of major bleeding complications was reported: 2.7% in the placebo group versus 3.7% in the clopidogrel group. Furthermore, limited data are available on the safety of adding heparin and a glycoprotein IIb/IIIa inhibitor in patients receiving aspirin and clopidogrel. Data on the efficacy of long-term administration of clopidogrel are limited and, therefore, such decisions should be individualized. Beta-Adrenoreceptor Blockers Beta-blockers have been found to be effective in improving outcomes when used for acute coronary syndromes in patients with diabetes. Several studies have consistently demonstrated an average twofold greater reduction in the relative risk of mortality as a result of beta-blocker therapy after myocardial infarction in patients with diabetes compared to those without diabetes. Despite these findings, diabetics with coronary artery disease are undertreated with beta-blocker drugs. Reviewing a large cohort of patients from the National Cooperative Cardiovascular Project, Chen et al. reported that even after adjusting for demographics and clinical factors, patients with diabetes were less likely to receive beta-blockers at discharge following acute myocardial infarction with an odds ratio for insulin-treated diabetics of 0.88 and 0.93 for non-insulin-treated diabetics [123]. In the same study, beta-blockers were found to be associated with lower 1-year mortality for insulin-treated diabetics (hazard ratio 0.87), for non-insulin-treated diabetics (hazard ratio 0.77), and for nondiabetics (hazard ratio 0.87). Furthermore, beta-blocker therapy was not significantly associated with increased 6-month readmission rates for diabetic complications. The authors concluded that beta-blocker therapy following acute myocardial infarction was associated with a lower 1-year mortality rate for elderly diabetic patients to a similar extent as for nondiabetics without increased risk of diabetic complications. It is evident from the above and several other randomized controlled trials that beta-blockers are important agents and should be used in all elderly diabetic patients in the absence of major contraindications (severe bradycardia with heart rate 160 mmHg with diastolic BP V95 mmHg) accounts for nearly two-thirds of the total prevalence of hypertension in both older men and women [16]. Combined hypertension characterized by abnormal elevations of both systolic and diastolic blood pressures accounts for less than one-third of the total prevalence. Isolated diastolic hypertension is considerably less prevalent in older men and women accounting for less than 15% of the prevalence. Blood lipids, including serum total cholesterol, also vary with age across the lifespan and trends appear to be different in women as compared to men. Figure 5 shows longitudinal trends in average levels of serum cholesterol in the Framingham Study as a function of age. Note that levels in men tend to peak in early middle age and then slowly decline with advancing age. Levels in women peak later in middle age and remain relatively high until advanced age before a decline occurs. The practical implication of this trend is the relatively high prevalence of hypercholesterolemia likely to be encountered in comparable populations of older women. Corresponding age trends for specific lipoprotein–cholesterol subfractions in the Framingham Study are illustrated in Figure 6. Trends for low-density lipoprotein cholesterol (LDL), in general, are similar to those for serum total cholesterol. Mean levels of high-density lipoprotein cholesterol (HDL) are consistently higher in women than in men across the age span, but tend to decline slightly after menopause. Trends in men remain essentially unchanged throughout life. Mean values of very-low-density lipoproteins (VLDL) tend to be higher in men than women, but trends in both tend to rise throughout middle age and then remain relatively stable thereafter. Presumably, these trends reflect fluctuations in body weight occurring at corresponding ages in the sexes. The prevalence of a number of cardiovascular risk factors in older subjects of the Framingham Study is presented in Table 3. Data are arrayed for each decade of age from 65 to 94 years and separately for men and women.

Epidemiology of CHD

195

Figure 5 Average age trends in serum cholesterol levels in cross-sectional and longitudinal (cohort) data. Framingham Study, Biennial exams 1–16.

Figure 6 Average age trends in lipoprotein cholesterol subfractions. Framingham Study. (Reproduced by permission.)

196 Table 3

Vokonas and Kannel Percentage Prevalence of Cardiovascular Risk Factors in the Elderly Definite hypertension

Hypercholesterolemia

Glucose intolerance

Age (years)

Men

Women

Men

Women

Men

Women

65–74 75–84 85–94

42.1 45.5 20.7

48.9 61.1 64.8

16.6 9.7 9.4

39.7 36.3 18.4

29.5 30.9 33.3

17.5 26.0 29.0

Obesity

Cigarette smoking

ECG–LVH

Age (years)

Men

Women

Men

Women

Men

Women

65–74 75–84 85–94

52.2 35.4 7.7

47.8 39.3 30.0

21.9 13.9 9.4

23.0 8.5 3.2

5.5 6.0 10.4

4.2 7.8 13.0

Definite hypertension: BP >160/95 mmHg; hypercholesterolemia: serum cholesterol >250 mg/dL; glucose intolerance: blood glucose >120 mg/dL, glycosuria, or diabetes mellitus; Obesity: relative weight >120%; cigarette smoking: any (Exam 15); ECG–LVH: evidence of left ventricular hypertrophy on electrocardiogram. Source: Framingham Study, Exam 16.

As would be expected from age trends observed earlier, hypercholesterolemia appears to be quite prevalent in older women. Glucose metabolism becomes progressively impaired with advancing age, resulting in increasing prevalence of glucose intolerance in both older men and women [17]. Body weight tends to decrease at far-advanced age. Weight loss, however, represents a reduction primarily in lean body mass (i.e., muscle and bone tissue) rather than adipose tissue. The result is an alteration of body composition in older persons with higher proportional adiposity per unit of weight [18]. Despite the tendency for lower body weights at advanced age, obesity appears to be well represented in both older men and women in this cohort. The prevalence of cigarette smoking appears to decrease with advancing age. This can be attributed not only to higher mortality rates in smokers but also to discontinuation of cigarettes because of health problems or concerns in older persons. The prevalence of ECG left ventricular hypertrophy (LVH) increases with advancing age in both older men and women. From the foregoing analysis, it is clear that the majority of established risk factors for CHD in middle-aged persons are prevalent in the elderly. What remains is the task of marshaling evidence to address the question of whether or not such risk factors continue to remain operative in the development of CHD in older persons. This constitutes the third and final perspective of interactions alluded to earlier (i.e., associations between specific risk factors and CHD in older persons) which will be reviewed next.

ASSOCIATION BETWEEN SPECIFIC RISK FACTORS AND CHD IN THE ELDERLY Associations for a number of specific risk factors and CHD are summarized in Table 4. Putative risk factors are listed in the column on the left. Standardized, age-adjusted

Epidemiology of CHD Table 4

197

Associations Between Specific Risk Factors and Incidence of CHD Bivariate standardized regression coefficients (age-adjusted) Ages 35–64

Risk factors Systolic pressure Diastolic pressure Serum cholesterol Cigarettes Blood glucose Vital capacity Relative weight

Ages 65–94

Men

Women

Men

Women

0.338c 0.321c 0.322c 0.259c 0.043
0.112a 0.190c

0.418c 0.363c 0.307c 0.095 0.206c
0.331c 0.264c

0.401c 0.296c 0.121
0.017 0.166c
0.127 0.177b

0.286c 0.082 0.213c
0.034 0.209c
0.253c 0.124a

Significant at ap < 0.05; bp < 0.01; cp < 0.001. Source: Framingham Study, 30-year follow-up.

(bivariate) logistic regression coefficients are categorized for younger and older subjects of the Framingham cohort and also arrayed separately for men and women. Regression coefficients are derived using a logistic regression model to mathematically relate the level of a risk factor or its categorical value to the development of CHD events. The magnitude of the coefficient and also the level of statistical significance reflects the strength of the association between the specific risk factor and CHD for the age group and sex under consideration. A cursory inspection of the results indicates that the majority of significant associations between risk factors and CHD apparent in younger men and women remain significant in older age groups but not consistently in both sexes. Systolic blood pressure, for example, demonstrates strong risk associations for CHD in younger men and women and maintains strong associations in both older men and women. The risk association for diastolic blood pressure, in contrast, appears to lose significance in older women. Serum total cholesterol demonstrates strong risk associations for CHD in younger men and women; however, the strength of this risk association clearly weakens in older men but remains significant in older women. Cigarette smoking modeled using this methodology fails to show significant risk associations in either older men or women. Blood glucose levels, as well as other parameters reflecting impaired glucose metabolism, including glucose intolerance and diabetes mellitus, show strong risk associations for CHD, in both older men and women. Vital capacity demonstrates strong negative risk associations, particularly in older women. Of interest, significant risk associations between CHD and body weight expressed as Metropolitan Relative Weight are maintained in both older men and women. A number of these risk associations will be characterized in detail below. Similar data for associations between specific risk factors and CHD incidence can be derived using an alternative regression methodology, the Cox proportional hazards model [19]. Blood Pressure and Hypertension The relation between blood pressure and CHD, particularly in older persons, represents one of the most striking risk associations in data from the Framingham Study.

198

Vokonas and Kannel

Risk relations between CHD incidence and systolic blood pressure in men and women are shown in Figure 7. CHD incidence, expressed as age-adjusted annual rate of CHD events per 1000, appears on the ordinates of left and right panels, respectively. Systolic blood pressure appears on the abscissas of both panels. Two risk relations are illustrated in each panel, one for younger men and women, aged 35 to 64, another for older men and women, aged 65 to 94. While overall incidence rates for CHD in women are substantially lower than those at corresponding blood pressures in men, the same underlying risk associations are observed. Note that CHD risk for systolic blood pressure rises with increasing pressure in all age groups and that this rise is even more striking in older than in younger persons. These trends are interpreted as marked increases in relative risk indicating that progressively higher levels of blood pressure confer additional risk for CHD. Although these trends do not establish causality, they serve to emphasize an important role for blood pressure in the extended sequence of pathophysiological events that result in manifest CHD. Also note that risk relations in older men and women are configured well above those of younger persons in each sex indicating substantially higher incidence rates for CHD at similar blood pressures. This is interpreted as an increase in absolute risk for CHD in older men and women suggesting a substantially higher burden of disease at all levels of blood pressure in older as compared to younger persons, even at normal or low blood pressures. Risk relations between CHD incidence and diastolic blood pressure in men and women are shown in Figure 8. Again, risk appears to rise with increasing blood pressure in all age groups (i.e., increased relative risk) and incidence rates at the same level of blood pressure are higher in older than in younger persons (i.e., increased absolute risk). Although there is a minor deviation from the nearly linear trend for the relation between

Figure 7 Risk of CHD by level of systolic blood pressure according to specified age groups in men and women. Framingham Study, 30-year follow-up.

Epidemiology of CHD

199

Figure 8 Risk of CHD by level of diastolic blood pressure according to specified age groups in men and women. Framingham Study, 30-year follow-up.

diastolic blood pressure and CHD incidence in older men, the relation maintains strong statistical significance. In contrast, the deviation in the relation corresponding to the fourth level of diastolic blood pressure (95–104 mmHg) in older women yields a statistically insignificant result using the logistic regression model, despite an apparent upper curvilinear trend. Although the discontinuities in these curves remain unexplained, these findings suggest a more consistent and reliable role for systolic as compared to diastolic blood pressure as a predictor for CHD in both elderly men and women. Risk gradients for CHD that, in general, are similar in direction and magnitude to those suggested earlier are observed when individuals are classified according to hypertensive status instead of absolute levels of blood pressure (Table 5). For all age and sex

Table 5

Risk of CHD by Hypertensive Status According to Age and Sex Average annual age-adjusted rate per 1000 Coronary heart disease 35–64a years

Hypertensive status Normal (160/95 mmHg)

65–94a years

Men

Women

Men

Women

8 15 21

3 7 10

14 28 41

11 16 22

a All trends significant at p < 0.001. Source: Framingham Study, 30-year follow-up.

200

Vokonas and Kannel

groups considered, the overall risk of CHD is two to three times higher in subjects with definite hypertension compared with normotensives, while risk is intermediate for those with mild hypertension. Absolute risk is two to three times higher in older subjects, both in men and women, and risk is nearly always higher in men than women, regardless of age. Similar patterns of risk attributable to hypertension have been observed specifically for cerebrovascular events, congestive heart failure, and peripheral vascular disease [15]. When considered alone, isolated systolic hypertension also confers substantial risk for CHD and other cardiovascular disease outcomes. Previous data from randomized clinical trials have established a strong case for the efficacy of treating combined elevations of systolic and diastolic blood pressure in older hypertensives [20,21], although considerable uncertainty remained regarding the treatment of isolated systolic hypertension. The findings of the Systolic Hypertension in the Elderly Program (SHEP) have served to dispel much of this uncertainty [22]. This study documented impressive reductions in total numbers of fatal and nonfatal strokes in the active treatment group as compared to the placebo group. Statistically significant reductions in nonfatal myocardial infarctions plus coronary death, as well as combined CHD and total cardiovascular disease outcomes, were also noted in the group on active treatment. There appeared to be little or no evidence in this study that lowering of either systolic or diastolic blood pressure resulted in an increased risk of CHD events or mortality, particularly at the lower end of the distribution for blood pressure, the so-called J-shaped curve. Five other recent clinical trials of drug therapy for hypertension in the elderly that included patients with isolated systolic hypertension also showed beneficial effects [23–27]. In addition to substantial reductions in cerebrovascular events and congestive heart failure, the majority of intervention studies of drug therapy for hypertension in older persons to date have consistently demonstrated either beneficial trends or significant reductions in CHD events and mortality. Such findings appear to be considerably less prominent in clinical trials experiences derived from predominantly middle-aged hypertensives [20,21,28]. Details regarding clinical evaluation and treatment of the elderly hypertensive patient appear in Chapter XX. Blood Lipids The risk of CHD in older persons attributable to serum lipids represents an area of considerable uncertainty and controversy. Relations between CHD incidence and serum total cholesterol in the Framingham Study are shown in Figure 9. Separate trends are plotted in younger and older subjects for men and women, respectively. Note that absolute risk is substantially increased in older as compared to younger men. Note also that relative risk clearly rises over the entire distribution of serum total cholesterol in younger as well as older men. The break in continuity of the risk relation at the fourth level of the distribution in older men, however, yields results that narrowly miss statistical significance both in bivariate (age-adjusted) and multivariate estimates using the logistic regression model, whereas the association remains strongly predictive in younger men. Although incidence rates for CHD at corresponding levels of cholesterol are lower in women than those in men, the same overall pattern pertains (i.e., an increase in both absolute and relative risk). In this instance, however, statistical significance is maintained in both younger and older women. The finding that serum total cholesterol loses strength as a risk factor for CHD, particularly in older men of the Framingham cohort, has resulted in serious misinter-

Epidemiology of CHD

201

Figure 9 Risk of CHD by level of serum cholesterol according to specified age groups in men and women. Framingham Study, 30-year follow-up.

pretation by some authors who have used this information to argue that serum lipids do not represent important risk factors for CHD in the elderly and thus justify neither detection nor treatment [29,30]. This view is both extreme and unreasonable, since several other studies have clearly validated the role of serum total cholesterol as a predictor of CHD events in older men [31–37] and also older women [31,33,37]. Despite these findings, however, a recent meta-analysis encompassing data from 22 U.S. and international cohort studies concluded that serum cholesterol did indeed lose strength as a risk predictor for CHD mortality in older men and also in older women [38], a finding consistent with original observations made in data from the Framingham Study. Cholesterol emerges as a significant predictor of CHD death when steps are taken to adjust for factors related to fraility or other comorbid conditions that serve to confound this risk association [39,40]. Focusing on serum cholesterol as the sole measure of risk for CHD attributable to serum lipids should now be considered obsolete based on our current understanding of lipoprotein subfractions and the availability of standardized laboratory methods to measure them in clinical practice. Substitution of either LDL or HDL cholesterol in the regression model fully restores statistical predictability for the risk relation between serum lipids and CHD [41]. HDL cholesterol, in particular, has emerged as an important lipid moiety that adds substantial precision to assessing coronary risk at limited additional cost [42]. Construction of a serum cholesterol/HDL ratio provides a highly accurate characterization of CHD risk in older men and women in the Framingham Study, which is illustrated in Figure 10. Indeed, data from Framingham as well as other studies confirm the overall reliability of the cholesterol/HDL ratio in assessing CHD risk in younger and older persons and in men as well as women [33,41,43,44]. The rationale for this approach is

202

Vokonas and Kannel

Figure 10 Risk of CHD by total/HDL cholesterol ratios among men and women aged 50 to 90 years. Framingham Study, 26-year follow-up.

that the ratio reliably captures the effect of a dynamic equilibrium of lipid transport into and out of body tissues, possibly including the intima of blood vessels. Several studies have suggested that serum triglycerides may be important predictors for CHD in either older men or older women, but not consistently in both sexes [33,41, 43,45]. Despite these observations, the present consensus holds that elevated levels of serum triglycerides represent a risk marker for obesity, glucose intolerance, and low HDL levels, all of which confer risk for CHD and, to the extent possible, deserve preventive attention. Data from intervention studies using dietary measures or drug therapy demonstrate the benefit of lipid alteration in reducing risk of CHD events particularly in middle-aged men [46,47]. Systematic information from clinical trials regarding the efficacy and safety of treating lipid abnormalities in the elderly, despite their prevalence in this population, is not yet available. A strong case for the widespread application of drug therapy to reduce risk of CHD in older persons as an element of primary prevention is not warranted at the present time. Of considerable interest, information from four large clinical trials makes a compelling case for reducing elevated or even average levels of LDL cholesterol with drugs in patients with established CHD: angina pectoris or following myocardial infarction [48– 52]. These effects of drug therapy in the secondary prevention of CHD events (i.e., in persons with preexisting CHD) appear to be beneficial in both middle-aged and older patients of both sexes. Nearly all such recent clinical trials have also demonstrated the benefit of statin drug therapy in reducing the subsequent incidence of stroke events in treated patients [53]. Current management of hypercholesterolemia for an older person considered to be at risk should consist of a highly individualized approach beginning with appropriate dietary measures and weight control before initiating a trial of specific drug therapy, preferably at lower doses, to achieve a carefully monitored lipid-lowering effect [54].

Epidemiology of CHD

203

Cigarette Smoking Cigarette smoking fails to demonstrate strong risk associations for total CHD events in either older men or older women using the logistic regression methodology indicated in Table 4. Significant risk associations, however, are discerned between cigarette smoking and death due to CHD [10]. One explanation for this phenomenon is that smoking may be more closely related to lethal events than to outcomes composed of combinations of morbid and lethal events [55]. Another explanation is that the cross-sectional pooling approach used in the analysis may classify long-term smokers who have discontinued cigarettes for relatively brief periods as nonsmokers, in effect diluting the strength of the association between the risk factor and the outcome [56]. Because of these difficulties, the most reliable approach to assessing risk associations between cigarette smoking and cardiovascular morbidity or mortality is to model these events prospectively for defined categories of smoking behaviors (e.g., current smoker, former smoker, and never smoker) and for longer time intervals. Such approaches yield strong risk associations between cigarette smoking and a broad array of cardiovascular outcomes, including CHD, stroke, and peripheral arterial disease, even in older men and women. These observations have been documented using data from Framingham as well as other studies [33,57–59]. Reducing the risk of cardiovascular disease is not the only reason to encourage discontinuation of cigarettes in older persons. Cigarettes contribute to the development of chronic bronchitis and obstructive lung disease as well as lung cancer and other malignancies. These conditions also exact a heavy toll in terms of disability and death in the elderly. Thus, the clinician can make a compelling case that it is never too late to stop smoking and extend appropriate advice and encouragement to assist patients in their effort to discontinue cigarettes.

Glucose Tolerance and Diabetes Mellitus Impaired glucose metabolism is not only highly prevalent in the elderly, but also confers substantial risk for CHD as well as other cardiovascular events in both older men and women. Various measures of glucose tolerance are employed and nearly all demonstrate significant risk associations with CHD in the Framingham Study [10]. These measures include blood glucose levels, glycosuria, and the composite risk categories designated as glucose intolerance and diabetes mellitus. Although diabetes mellitus confers enhanced risk for both younger and older men, overall risk increases dramatically for both younger and older women [60] (Fig. 11). Similar patterns of risk are noted for coronary and cardiovascular mortality [10,60]. Diabetes also emerges as an important risk factor in the development of congestive heart failure, particularly in older women with insulin-dependent diabetes mellitus [61]. Presumably, the microvascular disease that is unique to diabetes, as well as other mechanisms, serves to produce progressive damage to heart muscle, ultimately resulting in compromised ventricular function and heart failure. There is little evidence that control of hyperglycemia, either by oral hypoglycemic agents or insulin, effectively forestalls either the development or complications of cardiovascular disease [60,62], although encouraging trends in this regard were identified in the recently completed Diabetes Control and Complications Trial [63]. Available evidence would, therefore, suggest that there is more to be gained in reducing risk by correcting associated cardiovascular risk factors in persons with diabetes than by attention confined to early detection and control of hyperglycemia.

204

Vokonas and Kannel

Figure 11 Age-adjusted incidence rates for CHD based on diabetic status classified according to age and sex. Solid bar = diabetic; open bar = nondiabetic; RR = risk ratio diabetic/nondiabetic. (From Ref. 55, with permission.)

Left Ventricular Hypertrophy Left ventricular hypertrophy as determined by the electrocardiogram (ECG–LVH) emerges as a strong risk factor for CHD in older men and women (Fig. 12). Marked increases in CHD incidence are noted for voltage criteria for LVH alone with additional risk conferred by definite LVH which, in addition to voltage criteria, includes repolarization (ST and T wave) abnormalities consistent with LVH. These electrocardiographic findings presumably reflect abnormalities of myocardial structure and function related to early compromise of the underlying coronary circulation that antedate the development of clinical manifestations of CHD [64–66]. In this context, left ventricular hypertrophy (LV mass) as determined by echocardiography has emerged as an extremely potent independent predictor of CHD as well as other cardiovascular disease events especially in older persons [67,68]. Body Weight Of considerable interest is the finding that increased body weight represents a significant risk marker for CHD even at advanced age. The association between body weight and risk for CHD in older subjects of the Framingham Study is illustrated in Figure 13. Note that while CHD risk rises more strikingly

Epidemiology of CHD

Figure 12

205

Risk of CHD according to ECG-LVH status. Framingham Study, 30-year follow-up.

Figure 13 Risk of CHD according to body weight in men and women aged 65 to 94 years. Framingham Study, 30-year follow-up.

206

Vokonas and Kannel

with increasing body weight in older men, trends in older women clearly indicate enhanced risk at higher body weights. Progressive increases in body weight occurring earlier in life correlate closely with changes in the levels of several risk factors considered to be more directly related to pathogenesis of atherosclerosis [69]. These include increases in blood pressure, serum cholesterol, and triglycerides and blood glucose with a reduction in HDL cholesterol. These findings serve to emphasize the need to incorporate measures ultimately designed to either control or, if necessary, gradually reduce body weight as part of risk management even in older persons. When coupled with appropriate dietary measures, weight control would be particularly useful in the initial management of elderly patients with hypertension, dyslipidemia, and diabetes or combinations of these conditions. Vital Capacity Vital capacity is a simple and sensitive clinical technique for assessing compromise of pulmonary or cardiac function, or both, in individuals of all ages. In contrast to inconsistent risk associations with CHD found in younger subjects, vital capacity emerges as an important risk factor in the elderly, particularly with respect to risk for major coronary events [2,10]. Vital capacity normally declines with advancing age, an effect that is strongly exacerbated by cigarette smoking and obesity [70]. Diminished vital capacity in older people is also correlated with loss of muscular strength as assessed by hand grip, presumably reflecting overall neuromuscular debility [71]. Low vital capacity remains a strong predictor of cardiovascular mortality, as well as mortality from all causes. Measurement of this parameter, therefore, should be considered as part of the standard risk assessment of all older patients. Heart Rate Recent data from the Framingham Study demonstrates that resting heart rate confers additional risk for CHD in men, both in younger as well as older men, although no similar risk association is observed in either younger or older women [2,10]. An explanation for this finding is not readily apparent. Heart rate in men may be a more sensitive indicator of underlying cardiac dysfunction than in women or the net impact of sympathetic nervous system influences in precipitating CHD events is greater in men. High heart rates may also reflect physical deconditioning. Physical Activity Accumulating evidence now suggests that lifetime vigorous physical activity may forestall CHD in the elderly [72–74]. Previously reported data from Framingham indicated that overall mortality, including coronary mortality, was inversely related to level of physical activity in middle-aged men [75]. A benefit for exercise in older men was also suggested; however, the levels of exercise involved were quite modest. Although regular physical activity in the elderly is desirable and should be strongly encouraged, it would be unwise to place undue emphasis on this approach alone in attempting to reduce the risk for CHD. Prevalent CHD An important predictor of CHD events at all ages is the presence of antecedent CHD. It is therefore of interest that several of the risk factors associated with the initial development of

Epidemiology of CHD

207

CHD in the elderly not only continue to have strong correlations with prevalent disease but also maintain predictive associations for new CHD events in older persons with preexisting disease [33,35,76–78]. Serum lipids, cigarette smoking, and diabetes appear to be more prominent in this context than hypertension. Blood pressure may fall substantially following myocardial infarction, especially extensive anterior myocardial infarction which carries a poorer prognosis, thereby confounding the relation between hypertension and CHD morbidity and mortality [79]. Blood pressure, however, reemerges as a significant predictor of recurrent CHD events in long-term survivors of myocardial infarction [80]. These findings serve to emphasize the critical role of controlling risk factors in older persons with established CHD. Effective treatment of hypertension, discontinuation of cigarettes, and adequate control of hyperglycemia are important measures in the preventive clinical management of such individuals. The potential benefit of lipid-lowering drugs in older persons with established CHD and hypercholesterolemia was alluded to earlier. Other Risk Factors Several hematological or hemostatic factors have been described as risk variables in the Framingham Study. Hematocrit appeared to contribute to CHD in younger men and women in this analysis, but not in older persons [10]. White blood cell count, which was strongly correlated with the number of cigarettes smoked per day, hematocrit, and vital capacity, was also associated with enhanced risk for CHD and other cardiovascular endpoints in older men, both in smokers and nonsmokers, but only in women who smoked [81]. These data were consistent with reports from other studies [82]. Plasma fibrinogen showed strong risk associations for CHD and other cardiovascular disease outcomes in men, including older men [83], similar to findings from other studies [84]. Significant risk associations, however, were not apparent in older women. An extensive array of psychosocial, occupational, dietary and other factors have been described as putative risk parameters for CHD in the Framingham Study [9,85]; however, only limited information is available regarding specific associations of these factors with CHD in older persons. Although family history of CHD is strongly related to early development of CHD (before age 60) in both men and women in the Framingham Study, predictive associations also remained signficant for the late onset of CHD [86].

CHD RISK PROFILES IN THE ELDERLY Although associations between a specific risk factor and CHD can be considered in isolation as a single relation, in many instances combinations of several risk factors may constitute the observed risk profile, especially in older persons. Risk of CHD, in such instances, can be reliably estimated by synthesizing a number of risk factors into a composite score, based on a multiple logistic function [87,88]. Risk factors are assessed by standard clinical procedures (smoking history, blood pressure, and electrocardiogram) and by routine laboratory studies (serum total cholesterol, HDL cholesterol, and blood glucose). This type of composite index permits detection of individuals at relatively high risk, either on the basis of marked elevation of a single factor or because of marginal abnormalities of several risk factors. This multivariate risk scenario is illustrated in Figure 14, which characterizes the risk of CHD at two predefined levels of serum cholesterol and then considers changes in the

208

Vokonas and Kannel

Figure 14 Risk of CHD of two levels of serum cholesterol according to specified levels or categories of other risk factors for 70-yr-old women in the Framingham Study.

levels or values of other risk factors toward worsening risk, as indicated in the table below. Note that risk increases progressively with the additional impact of other risk factors for both categories of serum cholesterol, even in instances where a factor such as cigarette smoking has a relatively weak risk association with CHD when considered alone. The major risk factors, including age, when taken together, explain only a limited proportion of the variance of CHD incidence in younger as well as older persons [87,88]. It is likely that other major risk attributes exist among both the young and the elderly, but are yet to be identified. It must be emphasized, however, that the factors that have already been delineated do identify high-risk subgroups of the elderly population that should be targeted for preventive management.

PERSPECTIVES FOR PRIMARY PREVENTION OF CHD IN THE ELDERLY An important principle of prevention is the concept that measures limiting the effect of known risk factors should be initiated as early in life as possible to minimize the subsequent development of disease in both the young and the elderly. It is unreasonable to conclude, however, that modifications of risk factors initiated at advanced age are likely to be ineffective in reducing the toll of related disease in older persons. A simple, but important, observation in this context is that the incidence of CHD in the elderly varies widely within distributions of continuous risk factors such as blood pressure or serum cholesterol. Incidence also appears to differ markedly for values of categorical variables such as the presence or absence of left ventricular hypertrophy. In addition, data presented earlier indicate that the incidence of CHD, as well as other cardiovascular disease events, is not only substantially higher in populations of older persons but also continues to rise with advancing age across the lifespan. Because incidence is usually higher in older persons, the beneficial effect of a given intervention such as treatment of hypertension or hypercholesterolemia, as assessed by reduction in relative risk, may be similar to or lower than that observed in younger persons [2,20,89]. In many such instances, however, the same effect measured as a difference in absolute risk, reflecting reduction in numerical toll

Epidemiology of CHD

209

of disease outcomes, actually remains higher in the elderly. A corollary of this effect is that a smaller number of older as compared to younger people must be treated to yield the same benefit in terms of numbers of disease events prevented. These considerations are more relevant today than ever before. Recently, a marked and progressive decline in mortality due to coronary and cardiovascular disease has occurred in the United States and several other industrialized nations [90]. Age-specific trends indicate decreasing mortality due to CHD and cardiovascular disease, both in the elderly and in younger adults of both sexes. Similar trends in cardiovascular mortality have been identified in the Framingham population [91]. At the same time, the prevalence of several coronary risk factors, such as untreated hypertension, elevated serum cholesterol levels, and cigarette smoking, has diminished in the population at large, including the elderly, while impressive improvements have occurred in the diagnosis and treatment of CHD. Although the available information supports the contention that both of these potentially beneficial effects have contributed to the observed decline in mortality from CHD, the present consensus gives greater weight to the success of widespread primary preventive strategies, resulting in lowered levels of major risk factors that contribute to disease rather than to improved diagnosis and treatment of established disease [92]. In this context, hypertension clearly emerges as the dominant, potentially remediable, risk factor for both CHD and cerebrovascular disease morbidity and mortality in the elderly. Hypertension is highly prevalent in the aged, easily detected, and can be corrected by the careful application of appropriate measures, including drug therapy. As mentioned earlier, direct evidence from clinical trials has already established the efficacy of antihypertensive measures in reducing the frequency of both stroke and CHD events in elderly hypertensives. Available information also makes a compelling case for discontinuation of cigarettes at all ages, including the elderly. Information is needed, however, regarding the feasibility and effectiveness of nonpharmacological approaches such as diet and weight reduction in treating older persons with hypertension and lipid abnormalities both as initial therapy and as adjuncts to specific drug therapy. Also, as a matter of critical importance, the efficacy and safety of drug therapy for the treatment of lipid abnormalities in the elderly for the purpose of primary prevention should be established in large, well-designed clinical trials. Lack of such information severely limits our confidence in extending what may be a potentially beneficial preventive measure to greater numbers of older persons.

ACKNOWLEDGMENTS The authors wish to thank Claire Chisholm for her invaluable assistance in preparing this manuscript. This work was supported by the Cooperative Studies Program/ERIC of the Department of Veterans Affairs and the Massachusetts Veterans Epidemiology and Research Information Center (MAVERIC), the Visiting Scientist Program of the Framingham Heart Study, and grant nos. NO1-HV-92922, NO1-HV52971, and 5T32-HL-0737413 of the National Institutes of Health.

REFERENCES 1. 2.

Kannel WB, Gordon T. Evaluation of cardiovascular risk in the elderly: the Framingham Study. Bull NY Acad Med 1978; 54:573–591. Kannel WB, Vokonas PS. Primary risk factors for coronary heart disease in the elderly: the

210

3. 4. 5. 6. 7.

8. 9. 10.

11. 12. 13. 14.

15. 16. 17. 18. 19.

20. 21. 22.

23.

24. 25.

26.

Vokonas and Kannel Framingham study. In: Wenger NK, Furberg CD, Pitt E, eds. Coronary Heart Disease in the Elderly. New York: Elsevier Science Publishing, 1986:60–92. Kannel WB, Vokonas PS. Demographics of prevalence, incidence and management of coronary heart disease in the elderly and in women. Ann Epidemiol 1992; 2:5–14. Rich M. Acute myocardial infarction in the elderly. Cardio 1990; 7:79–90. Forman DE, Bernal JLG, Wei JY. Management of acute myocardial infarction in the very elderly. Am J Med 1992; 93:315–326. Wenger NK. Cardiovascular disease in the elderly. Curr Probl Cardiol 1992; 10:615–690. Kannel WB, Doyle JT, Shepard RJ, Stamler J, Vokonas PS. Prevention of cardiovascular disease in the elderly. 18th Bethesda Conference: Cardiovascular Disease in the Elderly. J Am Coll Cardiol 1987; 19(suppl A):25A–28A. Lakatta EG. Cardiovascular function in later life. Cardiovasc Med 1985; 10:37–40. Kannel WB, Sytkowski PA. Atherosclerosis risk factors. Pharmac Ther 1987; 32:207–235. Cupples LA, D’Agostino RB. Section 34. Some risk factors related to the annual incidence of cardiovascular disease and death using pooled repeated biennial measurements: Framingham Heart Study, 30-year follow-up. In: Kannel WB, Wolf PA, Garrison RJ, eds. The Framingham Study: An Epidemiological Investigation of Cardiovascular Disease. Bethesda, MD: National Heart, Lung, and Blood Institute publication No. (NIH) 87-2703, 1987. Kannel WB, Abbott RD. Incidence and prognosis of unrecognized myocardial infarction: an update from the Framingham Study. N Engl J Med 1984; 11:1144–1147. Learoyd BM, Taylor MG. Alterations with age in the viscoelastic properties of human arterial walls. Circ Res 1966; 18:278–292. Maddocks I. Possible absence of essential hypertension in two complete Pacific Island populations. Lancet 1976; 2:327–331. Avolio AP, Deng FQ, Li WG, Luo YF, Huang ZD, Xing LF, O’Rourke MF. Effects of aging on arterial distensibility in populations with high and low prevalence of hypertension: comparison between urban and rural communities in China. Circulation 1985; 71:202–210. Vokonas PS, Kannel WB, Cupples LA. Epidemiology and risk of hypertension in the elderly: The Framingham Study. J Hypertens 1988; 8(suppl I):53–59. Wilking SV, Belanger AJ, Kannel WB, D’Agostino RB, Steel K. Determinants of isolated systolic hypertension. JAMA 1988; 260:3451–3455. DeFronzo RA. Glucose intolerance and aging. Diabetes Care 1981; 4:493–501. Borkan GA, Norris AH. Fat redistribution and the changing body dimensions of the adult male. Human Biol 1977; 49:495–514. Harris T, Cook EF, Kannel WB, Goldman L. Proportional hazards analysis of risk factors for coronary heart disease in individuals 65 or older. The Framingham Heart Study. J Am Geriatr Soc 1988; 36:1023–1028. Applegate WB. Hypertension in elderly patients. Ann Intern Med 1989; 110:901–915. Mulrow CD, Cornell JA, Herrero CR, Kadri A, Farnett L, Aguilar C. Hypertension in the elderly: implications and generalizability of randomized trials. JAMA 1994; 272:1932–1938. SHEP Cooperative Research Group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension: final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA 1991; 265:3255–3264. Dahlof B, Lindholm LH, Hausson L, Schersten B, Ekbom T, Wester PO. Morbidity and mortality in the Swedish Trial in Old Patients with Hypertension (STOP-Hypertension). Lancet 1991; 338:1281–1285. MRC Working Party. Medical Research Council trial of treatment of hypertension in older adults: principal results. Br Med J 1992; 304:405–412. Staessen JA, Fagard R, Lutgarde T and others for the Systolic Hypertension in Europe (SystEur) Trial Investigators. Randomised double-blind comparison of placebo and active treatment for older patients with isolated systolic hypertension. Lancet 1997; 350:757–764. Lindholm LH, Hansson L, Ekbom T, Dahlof B, Lanke J, Linjer E, Schersten B, Wester PO,

Epidemiology of CHD

27.

28. 29. 30.

31. 32.

33.

34. 35.

36.

37.

38.

39.

40. 41. 42.

43. 44.

211

Hedner T, de Faire U. Comparison of antihypertensive treatments in preventing cardiovascular events inelderly diabetic patients; results from the Swedish Trial in Older Patients with Hypertension-2, STOP Hypertension-2 Study Group. J Hypertens 2000; 18(11):1671–1675. Kjeldsen SE, Dahlof B, Devereux RB, Julius S, Aurup P, Edelman J, Beevers G, de Faire U, Fyhrquist F, Ibsen H, Kristianson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Snapinn S, Wedel H; LIFE (Losartan Intervention for Endpoint Reduction) Study Group. Effects of losartan on cardiovascular morbidity and mortality in patients with isolated systolic hypertension and left ventricular hypertrophy: a Losartan Intervention for Endpoint Reduction (LIFE) substudy. JAMA 2002; 288(12):1491–1498. Strandgaard S, Haunso S. Why does anti-hypertensive treatment prevent stroke but not myocardial infarction? Lancet 1987; 2:658–661. Hulley SB, Newman TB. Cholesterol in the elderly. Is it important? (Editorial). JAMA 1994; 272:1372–1374. Garber AM, Browner WS, for the American College of Physicians. Guidelines for using serum cholesterol, high density lipoprotein cholesterol, and triglyceride levels as screening tests for preventing coronary heart disease in adults. Ann Intern Med 1996; 124:515–517. Barret-Connor E, Suarez L, Khaw KT, Criqui MH, Wingrad DL. Ischemic heart disease factors after age 50. J Chronic Diseases 1984; 37:903–908. Stamler J, Wentworth D, Neaton JD, for the MRFIT Research Group. Is the relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? Findings in 356,222 primary screenees of the Multiple Risk Factor Intervention Trial (MRFIT). JAMA 1986; 256:2835–2838. Aronow WS, Herzig AH, Etienne F, D’Alba P, Ronquillo J. 41-month follow-up of risk factors correlated with new coronary events in 708 elderly patients. J Am Geriatric Soc 1989; 37:501– 506. Benfante R, Reed D. Is elevated serum cholesterol level a factor for coronary heart disease in the elderly. JAMA 1990; 263:393–396. Pikkanen J, Linn S, Heiss G, Suchidran CM, Leon A, Rifkin BM, Tyroler HA. Ten-year mortality from cardiovascular disease in relation to cholesterol level among men with and without preexisting cardiovascular disease. N Engl J Med 1990; 322:1700–1707. Sorkin JD, Andres R, Muller DC, Baldwin HL, Fleg JL. Cholesterol as a risk factor for coronary heart disease in elderly men. The Baltimore Longitudinal Study of Aging. Ann Epidemiol 1992; 2:59–67. Keil JE, Sutherland E, Knapp RG, Gazes PC. Serum cholesterol-Risk factor for coronary heart disease mortality in younger and older blacks and whites. The Charleston Heart Study, 1960–1988. Ann Epidemiol 1992; 2:93–99. Manolio TA, Pearson TA, Wenger NK, Barret-Connor E, Payne GH, Harlan WR. Cholesterol and heart disease in older persons and women: review of an NHLBI Workshop. Ann Epidemiol 1992; 2:161–176. Corti MC, Guralnik JM, Salive ME, Harris T, Ferrucci L, Glynn RJ, Havlik RJ. Clarifying the direct relation between total cholesterol levels and death from coronary heart disease in older persons. Ann Intern Med 1997; 126:753–760. Jacobs D, Blackburn H, Higgins HM, et al. Report of the Conference on Low Blood Cholesterol: mortality associations. Circulation 1992; 86:1045–1060. Castelli WP, Wilson PWF, Levy D, Anderson K. Cardiovascular risk factors in the elderly. Am J Cardiol 1989; 63:12H–19H. Corti MC, Guralnik JM, Salive ME, Harris T, Field TS, Wallace RB, Berkman LF, Seeman TE, Glynn RJ, Hennekens CH, Havlik RJ. HDL cholesterol predicts coronary heart disease mortality in older adults. JAMA 1995; 274:539–544. Castelli WP, Anderson K, Wilson PWF, Levy D. Lipids and risk of coronary heart disease. The Framingham Study. Ann Epidemiol 1992; 2:23–28. Hong MK, Romm PA, Reagan K, Green CE, Rackley CE. Total cholesterol/HDL ratio is the

212

45. 46. 47.

48.

49.

50.

51.

52.

53.

54. 55. 56. 57. 58. 59.

60.

61. 62. 63.

Vokonas and Kannel best predictor of anatomic coronary artery disease among elderly patients [Abstr]. J Am Coll Cardiol 1991; 17:151A. Gotto AM Jr. Hypertriglyceridemia: risks and perspectives. Am J Cardiol 1992; 70:19H–25H. Levine GN, Keaney JF Jr, Vita JA. Cholesterol reduction in cardiovascular disease: clinical benefits and possible mechanisms. N Engl J Med 1995; 332:512–519. Shepherd J, Cobbe SM, Ford I, Isles CG, Lorimer AR, MacFarlane PW, McKillop JH, Packard CJ, for the West of Scotland Coronary Prevention Study Group. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. N Eng J Med 1995; 333:1301–1307. Scandinavian Simvastatin Survival Study Group. Randomized trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994; 344:1383–1389. Miettinen TA, Pyorala K, Olsson AG, Musliner TA, Cook TJ, Faergeman O, Berg K, Pederson T, Kjekshus J, for the Scandinavian Simvastatin Study Group. Cholesterol-lowering therapy in women and elderly patients with myocardial infarction or angina pectoris: findings from the Scandinavian Simvastatin Survival Study (4S). Circulation 1997; 96:4211–4218. Sacks FM, Pfeffer MA, Moye LA, Moye LA, Rouleau JL, Rutherford JD, Cole TG, Brown L, Warnica JW, Arnold JMO, Wun CC, Davis BR, Braunwald E, for the Cholesterol and Recurrent Events Trial Investigators. The effects of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med 1996; 335:1001– 1009. Downs JR, Clearfield M, Weis S, Whitney E, Shapiro DR, Beere PA, Langendorfer A, Stein EA, Kruyer W, Gotto AM Jr. Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels: results of AFCAPS/TexCAPS Air Force/ Texas Coronary Atherosclerosis Prevention Study. JAMA 1998; 279(20):1615–1622. 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; 339:1349–1357. Hebert PR, Gaziano JM, Chan KS, Hennekens CH. Cholesterol lowering with statin drugs, risk of stroke, and total mortality. An overview of randomized trials. JAMA 1997; 278(4):313– 321. Aronow WS. Pharmacologic therapy of lipid disorders in the elderly. Am J Geriatr Cardiol 2002; 11(4):247–256. Muller JE, Abela GS, Nesto RW, Toffler GH. Triggers, acute risk factors and vulnerable plaques: the lexicon of a new frontier. J Am Coll Cardiol 1994; 23:809–813. Cupples LA, D’Agostino RB, Anderson K, Kannel WB. Comparison of baseline and repeated measure covariate techniques in the Framingham Heart Study. Stat Med 1988; 7:205–218. Kannel WB, Higgins M. Smoking and hypertension as predictors of cardiovascular risk in population studies. J Hypertens 1990; 8(suppl 5):S3–S8. Benfante R, Reed D, Frank J. Does smoking have an independent effect on coronary heart disease incidence in the elderly? Am J Publ Health 1991; 81:897–899. LaCroix AZ, Lang J, Scherr P, Wallace RB, Cornoni-Huntley J, Berkman L, Curb JD, Evans D, Hennekens CH. Smoking and mortality among older men and women in three communities. N Engl J Med 1991; 324:1619–1625. Vokonas PS, Kannel WB. Diabetes mellitus and coronary heart disease in the elderly. Aronow W, Tresch D, eds. Clinics in Geriatric Medicine. Vol. 12. No. 1. Philadelphia: W.B. Saunders Co., 1996:69–78. Kannel WB, Hjortland M, Castelli WB. Role of diabetes in congestive heart failure: the Framingham Study. Ann Intern Med 1970; 72:813–822. Wilson PWF, Cupples L, Kannel WB. Is hyperglycemia associated with cardiovascular disease? The Framingham Study. Am Heart J 1991; 121:586–590. Diabetes Control and Complications Trial Research Group. The effect of intensive treatment

Epidemiology of CHD

64. 65. 66.

67.

68.

69.

70. 71. 72. 73. 74.

75.

76. 77.

78. 79. 80.

81. 82.

83. 84.

213

of diabetes in the development and progression of long-term complications in insulindependent diabetes mellitus. N Eng J Med 1993; 329:986–997. Kannel WB, Dannenberg AL, Levy D. Population implications of electrocardiographic left ventricular hypertrophy. Am J Cardiol 1987; 60:851–931. Aronow WS. Left ventricular hypertrophy. J Am Geriatr Soc 1992; 40:71–80. Levy D, Salomon M, D’Agostino RB, Belanger AJ, Kannel WB. Prognostic implications of baseline electrocardiographic features and their serial changes in subjects with left-ventricular hypertrophy. Circulation 1994; 90:1786–1793. Aronow WS, Koenigsberg M, Schwartz KS. Usefulness of echocardiographic left ventricular hypertrophy in predicting new coronary events and atherothrombotic brain infarction in patients over 62 years of age. Am J Cardiol 1988; 61:1130–1132. Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP. Left ventricular mass and incidence of coronary disease in an elderly cohort. The Framingham Heart study. Ann Intern Med 1989; 110:101–107. Borkan GA, Sparrow D, Wisnieski C, Vokonas PS. Body weight and coronary risk: patterns of risk factor change associated with long-term weight change. The Normative Aging Study. Am J Epidemiol 1986; 124(3):410–419. Sparrow D, Weiss ST. Pulmonary system. In: Rowe JW, Besdine R, eds. Geriatric Medicine. Boston: Little Brown, 1988:266–275. Kannel WB, Hubert H, Lew EA. Vital capacity as a predictor of cardiovascular disease: the Framingham Study. Am Heart J 1983; 34:1022–1033. Donahue RP, Abbott RD, Reed DM, Yano K. Physical activity and coronary heart disease in middle-aged and elderly men: the Honolulu Heart Program. Am J Publ Health 1988; 78:683–685. Berlin JA, Colditz GA. A meta-analysis of physical activity in the prevention of coronary heart disease. Am J Epidemiol 1990; 132:612–628. Paffenbarger RS Jr, Hyde RT, Wing AL, Lee IM, Jung DL, Kampert JB. The association of changes in physical activity level and other lifestyle characteristics with mortality among men. N Engl J Med 1993; 328:538–545. Kannel WB, Belanger AJ, D’Agostino RB, Israel I. Physical activity and physical demand on the job and risk of cardiovascular disease and death: the Framingham Study. Am Heart J 1986; 112:820–825. Wong ND, Wilson PWF, Kannel WB. Serum cholesterol as a prognostic factor after myocardial infarction: the Framingham Study. Ann Intern Med 1991; 115:687–697. Simons LA, Friedlander Y, McCallum J, Simons J, Powell I, Heller R, Berry G. The Dubbo Study of health of elderly: correlates of coronary heart disease at study entry. J Am Geriatric Soc 1991; 39:584–590. Applegate WB, Hughes JP, Vander-Zwaag R. Case-control study of coronary heart disease risk factors in the elderly. J Clin Epidemiol 1991; 44:409–415. Kannel WB, Sorlie PD, Castelli WP, McGee DL. Blood pressure and survival after myocardial infarction. The Framingham Study. Am J Cardiol 1980; 45:326–330. Wong ND, Cupples LA, Ostfeld AM, Levy D, Kannel WB. Risk factors for long-term coronary prognosis after initial myocardial infarction. The Framingham Study. Am J Epidemiol 1989; 130:469–480. Kannel WB, Anderson K, Wilson PWF. While blood cell count and cardiovascular disease. JAMA 1992; 267:1253–1256. de L, Labry O, Campion EW, Glynn RJ, Vokonas PS. White blood cell count as a long term predictor of mortality: results over 18 years from the Normative Aging Study. J Clin Epidemiol 1990; 43:153–157. Kannel WB, Wolf PA, Castelli WP, D’Agostino RB. Fibrinogen and risk of cardiovascular disease. The Framingham Study. JAMA 1987; 258:1183–1186. Ernst E. Fibrinogen: its emerging role as a cardiovascular risk factor. Angiology 1994; 45:87– 93.

214

Vokonas and Kannel

85. Kannel WB, Eake ED. Psychosocial and other features of coronary heart disease: insights from the Framingham Study. Am Heart J 1986; 112:1066–1073. 86. Schildkraut JM, Myers RH, Cupples LA, et al. Coronary risk associated with age and sex of parental heart disease in the Framingham Study. Am J Cardiol 1989; 64:555–559. 87. Gordon T, Kannel WB. Multiple risk functions for predicting heart disease: the concept, accuracy and application. Am Heart J 1982; 103:1031–1039. 88. Chambless LE, Dobson AJ, Patterson CC, Raines B. On the use of logistic risk score in predicting risk of coronary heart disease. Stat Med 1990; 9:385–396. 89. Malenka DJ, Baron JA. Cholesterol and coronary heart disease. The importance of patientspecific attributable risk. Arch Intern Med 1988; 148:2247–2252. 90. Walker WJ. Changing U.S. life style and declining vascular mortality-a retrospective. N Engl J Med 1983; 308:649–651. 91. Sytkowski PA, Kannel WB, D’Agostino RB. Changes in risk factors and the decline in mortality from cardiovascular disease. The Framingham Heart Study. N Engl J Med 1990; 322:1635–1641. 92. Goldman L. Cost-effectiveness perspectives in coronary heart disease. Am Heart J 1990; 119:733–739.

9 Pathophysiology of Coronary Artery Disease in the Elderly Roberto Corti University Hospital Zurich, Zurich, Switzerland

Antonio Ferna´ndez-Ortiz San Carlos University Hospital Madrid, Spain

Valentin Fuster Zena and Michael A. Wiener Cardiovascular Institute The Mount Sinai School of Medicine, New York, New York, U.S.A.

INTRODUCTION Atherosclerotic involvement of coronary artery arteries with subsequent coronary heart disease is the leading cause of death and disability in the elderly. Approximately half of all deaths of persons older than 65 years of age are the result of coronary artery disease and, of all coronary death, 80% occur in patients 65 years of age or older. Heart failure (which has enormous socioeconomic impact) is the terminal stage of all cardiac disease (Fig. 1) and coronary artery disease is the most frequent cause of heart failure. Despite crucial advances in our knowledge of the pathological mechanisms and the availability of effective diagnostic and treatment modalities, coronary atherothrombosis remains the most frequent cause of ischemic heart disease. Plaque disruption with superimposed thrombosis is the main cause of unstable angina, myocardial infarction, and sudden death. New findings have recently introduced exciting concepts that could have major impact on the treatment of atherothrombotic disease [1–3]. We will discuss the mechanisms that lead to the development of atherothrombosis and those responsible for the acute coronary syndromes, as well as some of the concepts derived from in vivo observations using new imaging technologies (such as high-resolution magnetic resonance imaging). Although risk factor modification and advances in therapeutic modalities have resulted in a welcome decline in cardiovascular mortality during the past two decades, recent statistics indicate that this decline is less notable in the elderly, the fastest growing segment of society. In fact, with the aging of the population, the overall prevalence, mortality, and cost of coronary artery disease will increase (Fig. 2). It has been projected that cardiovascular disease will climb from the second most common cause of death (29% of all deaths in 1990) to the first, with more than 36% of all deaths by 2020. According to current estimates [4], 61,800,000 Americans have one or more types of cardiovascular 215

216

Corti et al.

Figure 1 Course of coronary heart disease beginning with endothelial dysfunction and ending in heart failure.

Figure 2 Prevalence of coronary heart disease by age and sex in the U.S. (1988-1994). (Reprinted from Ref. 4.)

Pathophysiology of CAD

217

disease. Of these, 29,700,000 are male and 32,100,000 are female, 24,750,000 are estimated to be age 65 and older. Cardiovascular disease claimed 958,775 lives in the United States in 1999. This is 40.1% of all deaths or 1 of every 2.5 deaths. Cardiovascular disease was about 60% of ‘‘total mention mortality,’’ which means that of the more than 2,000,000 deaths from all causes, CVD was listed as a primary or contributing cause on about 1,391,000 death certificates. Since 1900, cardiovascular disease has been the No. 1 killer in the United States every year except 1918. Therefore, efforts to increase our knowledge of the pathophysiology of the atherosclerotic involvement of coronary arteries aimed to reduce mortality, disability, and cost associated with coronary heart disease are an important public health priority. This chapter will provide a pathophysiological explanation for the broad spectrum of ischemic coronary syndromes based on the most recent cell and molecular biology studies of the arterial wall.

INITIATION OF ATHEROSCLEROTIC LESIONS Atherothrombosis is a systemic disease of the vessel wall that is characterized by the tendency to accumulate lipids, inflammatory cells, smooth muscle cells (SMC), and extracellular matrix within the subendothelial space, and to progress in an unpredictable way to different stages. The molecular and biological mechanisms involved in the initiation and progression of atheroclerotic disease have been extensively studied during the past decades leading to the introduction of new concepts that form the basis for research efforts in the field of vascular biology [2,5]. The past two decades have elucidated the pivotal role of the endothelium in the process of atherogenesis. The endothelium is a monocellular layer that lines the luminal surface of the whole vascular tree. By means of its magnitude, with an estimated surface of more than 150 m2, and geographical location, it modulates vascular perfusion, permeability, and blood fluidity. The endothelim is a dynamic autocrine/paracrine organ that regulates contractile, secretory, and mitogenic activities in the vessel wall and hemostatic processes within the vascular lumen. The endothelium responds to hemodynamic and blood-borne signal by synthesizing and/or releasing a variety of agents. The endothelial dysfunction, defined as reduced synthesis and/or release of nitric oxide (NO), is now considered the earliest pathological signal of atherosclerosis. Interestingly, systemic risk factors and local risk factors (Table 1) may trigger endothelial dysfunction and lead to initiation of atherosclerotic lesions.

Table 1

Risk Factors for the Development of Atherosclerosis

Systemic risk factors

Local risk factors

Hypercholesterolemia Diabetes mellitus Arterial hypertension Cigarette smoke Systemic inflammatory disease Chronic infections (?) Low shear stress Arterial bifurcation and trifurcations Vascular bending points Vascular curvatures

218

Corti et al.

Systemic Atherogenic Risk Factors The initiation of atherosclerotic plaque has been correlated with the presence of a number of cardiovascular risk factors (such as hyperlipidemia, diabetes mellitus, hypertension, aging, and others), leading to endothelial dysfunction. Severe endothelial dysfunction in the absence of obstructive coronary artery disease has been shown to be associated with increased cardiac events supporting the concept that coronary endothelial dysfunction plays a pivotal role in the initiation and progression of coronary atherosclerosis [6]. Interestingly, atherosclerogenesis has been recognized as an inflammatory disease [7], characterized by accumulation of inflammatory cells within the vessel wall. Recently, endothelial dysfunction has been reported in patients with chronic inflammatory diseases, such as in rheumatoid arthritis, and endothelial function improves in these patients with anti-tumor necrosis factor–alpha [8]. More interestingly, treatment with potent anti-inflammatory drugs (selective COX-2 inhibitors) in patients with severe CAD and endothelial dysfunction leads to a significant improvement of the endothelium-dependent vasodilation and reduces low-grade chronic inflammation and oxidative stress in coronary artery disease [9]. Thus potent anti-inflammatory drugs, such as selective COX-2 inhibitors, have the potential to beneficially impact outcome in patients with cardiovascular disease. There is general agreement that hypercholesterolemia is an important causative factor in atherogenesis and that correcting it can strikingly reduce the risk of coronary heart disease. For instance, increased plasma levels of low-density lipoprotein (LDL) cholesterol induce endothelial dysfunction, the initial step in atherogenesis, whereas correction of this metabolic disorder by lipid apheresis or treatment with lipid-lowering drugs can normalize endothelial dysfunction [10–12]. In addition, recent observations have highlighted the role of high-density lipoproteins (HDL) and reverse cholesterol transport, as they are responsible for the removal of free cholesterol from the blood [13,14]. A low plasma level of HDL has been associated with increased cardiovascular risk, but only recently has this been recognized as a major risk factor requiring adequate treatment [15]. Several experimental studies have demonstrated the potential antiatherogenic properties of HDL which prevents plaque formation and even induces plaque regression [16]. More recently, it has been demonstrated that HDL administration restores normal endothelial function by increasing bioavailability of nitric oxide in hypercholesterolemic patients [17]. How cholesterol interacts with the cells of the arterial wall to initiate the atherogenic process has been only recently elucidated. Experimental studies have suggested a direct injurious effect of elevated levels of LDL cholesterol, in particular its oxidative derivate, on the endothelium. Two distinct mechanisms that link oxidative stress with impairment in endothelium-dependent vasodilation have been described. First, lysolecithin, a product formed as a consequence of lipid peroxidation of LDL particles, may be involved in the development of abnormal arterial vasomotion [10,11]. Second, hypercholesterolemia may be a stimulus to augmented generation of superoxide radicals by the endothelium, which directly inactivates nitric oxide and also increases the subsequent oxidation of LDL particles by the formation of peroxynitrate. Chronic hyperglycemia may also impair the function of the endothelium, resulting in abnormal vasomotion. Extended exposure to hyperglycemia results in the glycation (i.e., the nonenzymatic conjugation of glucose) of extracellular matrix proteins, which leads to the formation of advanced glycosylation end products [18]. Accumulation of such end products in the vessel wall loads to increased vessel stiffness, lipoprotein binding, macro-

Pathophysiology of CAD

219

phage recruitment, secretion of platelet-derived growth factor, and proliferation of vascular smooth muscle cells (SMC). Other forms of endothelial injury such those that can be induced by chemical irritants in tobacco smoke or circulating vasoactive amines may potentiate chronic endothelial injury favoring accumulation of lipids and monocytes into the intima. The endothelial dysfunction is characterized by increased permeability, reduced synthesis and release of nitric oxide, and overexpression of adhesion molecules (such as intracellular adhesion molecule-1, vascular cell adhesion molecule-1, and selectins) and chemoattractants (such as monocyte chemoattractant protein-1, macrophage colony stimulating factor, IL-1/-6, and IFN-a/-g). Expression of endothelial adhesion molecules is induced by a variety of stimuli such as the classical cardiovascular risk factors (hyperilemia, diabetes, smoke, etc.), facilitating the recruitment and internalization of circulating monocytes and LDL cholesterol (Fig. 3) [5,19,20]. In fact, these molecules allow the internalization (homing) of inflammatory cells and bone marrow–derived vascular progenitor cells in the subendothelial space. Monocyte chemoattractant protein-1 (MCP-1) is a potent chemokine, produced by the endothelial and smooth muscle cells (SMCs), which causes the directed migration of leukocytes. Macrophage colony stimulating factor (M-CSF) is an activator that can cause the expression of scavenger receptors on macrophages and a co-mitogen that leads to the proliferation of macrophages into the forming plaque. The lipid material accumulated within the subendothelial space becomes oxidized and triggers an inflammatory response that induces the release of different chemotactic and proliferative growth factors. In response to these factors, SMC

Figure 3 Endothelial dysfunction and initiation of atherosclerotic plaques. Expression of endothelial adhesion molecules is induced by a variety of stimuli such as the classic cardiovascular risk factors (hyperilemia, diabetes, smoke, etc.). These molecules allow the internalization (homing) of inflammatory cells and bone marrow–derived vascular progenitor cells in the subendothelial space. Monocyte chemoattractant protein-1 (MCP-1) is a potent chemokine, produced by the endothelial and smooth muscle cells (SMCs), which causes the directed migration of leukocytes. Macrophage colony stimulating factor (M-CSF) is an activator that can cause the expression of scavenger receptors on macrophages and a co-mitogen that led to the proliferation of macrophages into the forming plaque. (Reprinted from Ref. 143.)

220

Corti et al.

and macrophages will activate, migrate, and proliferate resulting in the thickening of the arterial wall. Local Atherogenic Risk Factors Despite exposure of different areas of the endothelial surface to the same injurious stimuli concentration, spontaneous atherosclerotic lesions only develop in certain locations. Clearly, there must be local factors that modulate the impact of hypercholesterolemia and other risk factors on the vessel wall, determining the location and possibly the rate of atherosclerosis progression. Atherosclerotic plaques are located more frequently near bifurcations and in curvatures of the vessels. It is believed that disturbances in the pattern of local blood flow are also responsible for development of atherosclerotic lesions. Recent studies have confirmed the importance of hemodynamic shear stress in endothelial dysfunction, in atherogenesis, remodeling, and restenosis after angioplasty and thrombus formation. The endothelium, by virtue of its location interfacing between the blood and the vascular wall, responds rapidly and sensitively to biochemical and biomechanical stimuli in order to maintain constant rheological conditions. In numerous experiments, shear stress has been shown to actively influence vessel wall remodeling. Specifically, chronic increases in blood flow, and consequently shear stress, lead to expansion of the luminal radius such that mean shear stress is returned to its baseline level. Conversely, decreased shear stress resulting from lower flow or blood viscosity induces a decrease in internal vessel radius. Interestingly, two contradictory hypotheses were advanced more than 30 years ago to explain the typical distribution of atherosclerotic lesions at the vessel bifurcation. The first implicated high shear stress via endothelial injury and denudation. The second proposed low shear stress as an atherogenic factor [5]. More conclusive evidence for the direct effect of hemodynamic forces on endothelial structure and function derives from in vitro studies with cultured monolayers subjected to various rheological conditions [21]. Importantly, recognition of the modulator effect of the changes of shear stress on cultured cell monolayers and on gene modulation led to further research under more complex flow conditions. Several endothelial targets have been suggested to act as shear stress-responsive elements [21–23], which may be activated by changes in shear stress conditions leading to the various signaling cascades, and resulting in highly coordinated and orchestrated mechanotransduction messages with the ultimate goal of restoring physiological conditions. Despite the systemic nature of the major cardiovascular risk factors, atherosclerosis is a geometric focal disease preferentially affecting the outer edges of vessel bifurcation [24] and is characterized by low and oscillatory shear stress as observed in the early 1970s [25]. Atherosclerotic lesions colocalize with regions of low shear stress throughout the arterial tree, from the carotid artery bifurcation [26,27] to the coronary [28], infrarenal, and femoral artery vasculatures [29]. The rate of atherosclerosis progression in patients with coronary artery disease was found, by serial quantitative coronary angiography, to correlate inversely with shear stress magnitude, even when corrected for other cardiovascular risk factors such as lipoproteins levels [30]. The interaction between rheology and the functional phenotype of endothelium plays a critical role in the atherosclerotic process. The early stage of atherosclerosis is characterized by an increased permeability of the vessel wall to blood-borne cells and proteins (macrophages, lymphocytes, LDL-lipoprotein, etc.). Caro et al. [31] first proposed a shear stress–dependent transfer of lipids into the vessel wall. Interestingly,

Pathophysiology of CAD

221

cultured endothelial cells exposed to physiological laminar shear stress elongate and align in the direction of blood flow, while those exposed to low shear stress (or cultured in static condition) do not change their shape or orientation, presenting with polygonal shape (Fig. 4). All these alterations may explain changes in endothelial cell permeability for atherogenic lipoprotein particles [24,28,30].

PROGRESSION OF ATHEROSCLEROTIC LESIONS Atherogenesis is a dynamic process initiating at the level of the blood–endothelial surface, leading to infiltration of blood-derived products and cells into the subendothelial space, with progress actively involving the intimal–medial interface and the media itself in advanced disease. Based on the pathological and clinical characteristics of atherosclerotic plaques, the American Heart Association Committee on Vascular Lesions [32] has developed a standardized classification of distinct plaques simplified by Fuster (Fig. 5) [33]. The homing of inflammatory cells into the subendothelial space, together with increased lipid accumulation and increased connective tissue synthesis, leads to atheroma formation. However, in some instances, a much faster process ensues and thrombosis associated with a ruptured plaque appears to be responsible (see plaque disruption). Ross [7] has recently stated that every step of the molecular and cellular responses leading to atherosclerosis is an inflammatory process. For instance, over the past decade, there has

Figure 4 Effect of shear stress on endothelial phenotype and activity. In physiological shear stress conditions (left panel) endothelial cells are elongated and align in the direction of blood flow and their function is mainly characterized by the production of vasodilators as well as substances with antithrombotic, anticoagulant, and antioxidant properties. In low-flow conditions (such as turbulent flow, right panel) cultured endothelial cells show a random pattern and produce substances favoring cell apoptosis and proliferation inflammation, vasoconstriction, thrombosis and atherogenesis. (Reprinted from Ref. 144.)

222

Corti et al.

Figure 5 Schematic summary of the different stages that characterize the progression of atherosclerotic plaques. (Reprinted from Ref. 145.)

been increasing evidence that this inflammatory response may, in part, be elicited by the oxidation of phospholipids contained in LDL, which are responsible for the endothelial expression of adhesion molecules and the induction of endothelial and SMC production of chemoattractants and chemokines [34]. More importantly, Sata et al. [35] recently reported that bone marrow cells, including hematopoietic stem cells, give rise to most of the SMCs that contribute to arterial remodeling observed in the experimental model of postangioplasty restenosis, graft vasculopathy, and atherosclerotic plaque growth. This new finding provides new and intriguing insights into the pathogenesis and are the basis for the development of new therapeutic strategies for vascular disease, by targeting mobilization, homing, differentiation, and proliferation of bone marrow–derived vascular progenitor cells [35]. Endothelial dysfunction alters the normal homeostatic properties of the endothelium, thereby initiating inflammatory cell activation. Continued inflammation results in increased numbers of macrophages and lymphocytes migrating from the blood into the lesion. Activation of these cells leads to release of hydrolytic enzymes, cytokines, chemokines, and growth factors causing overall growth of the plaque and ultimately leading to cellular necrosis in advanced lesions. While the artery initially compensates by dilating (remodeling), at some point it can no longer do so and the lesion protrudes into the lumen compromising the blood flow. Endothelial dysfunction is considered the precursor that initiates the atherosclerotic process and is characterized by the increased expression of adhesion molecules (e.g., selectins, VCAMs, and ICAMs) that participate in ‘‘homing’’ and infiltrating monocytes

Pathophysiology of CAD

223

[7,11]. The process of leukocyte binding and extravasation is critically dependent on both selection and immunoglobulin adhesion molecules on endothelial cells interacting with integrin receptors on leukocytes. Immunohistochemical analysis of human atherosclerotic coronary lesions shows expression of intercellular adhesion molecule-1 (ICAM-1) on endothelial cells, macrophages, and smooth muscle cells within the plaque (Table 2). The extracellular portion of these molecules can be enzymatically cleaved and detected in serum and is referred to as soluble intercellular adhesion molecules (sICAM). Increased concentration of sICAM-1 was detected in patients with acute coronary syndromes (ACS) [36]. Increased concentrations of sP-selectin were detected after a chest pain episode in patients with unstable angina but not in those with stable angina or control subjects [37]. Concentration of sICAM-1, sVCAM-1, and sP-selectins remain increased throught the first 72 h after presentation in unstable angina and non-Q-wave myocardial infarction, whereas the concentration of sE-selectin falls during this time [38]. Evidence indicates prognostic value of adhesion molecules. In fact, adverse outcome has been associated with increased concentration of sVCAM-1 in ACS. These observations confirm the critical pathogenic role of inflammation in acute coronary syndrome. The recruitment of monocytes seems to be induced by intimal LDL accumulation and its oxidative modifications [39]. The monocytes migrate into the subendothelium, where they transform into macrophages (Fig. 3). This differentiation process includes the up-regulation of various receptors such as CD36 and the scavenger receptor A (SR-A), which are responsible for the internalization of oxidized LDL. The expression of scavenger receptors is regulated by peroxisome proliferator-activated receptor-g, a nuclear transcription factor that controls a variety of cellular functions whose ligands include oxidized fatty acids, and cytokines such as tumor necrosis factor-a and interferon-g. Native LDL is not taken up by macrophages rapidly enough to generate foam cells, and so it was proposed that LDL undergoes ‘‘modification’’ in the vessel wall [40]. It has been shown that trapped LDL does undergo ‘‘modification,’’ including oxidation, lipolysis, proteolysis, and aggregation, and that such modifications contribute to inflammation and foam-cell formation. Indeed, oxidized LDL is recognized by the scavenger receptor of the macrophages that enable the intracellular accumulation of lipids. Oxidized LDL was demonstrated in vivo to induce endothelial dysfunction and ultimately to alter the barrier function of the endothelium [41,42]. LDL undergoes oxidative modification as a result of interaction with reactive oxygen species, including products of 12/15-lipoxygenase [20]. Lipoxygenases insert molecular oxygen into polyenoic fatty acid, which are likely to be transferred across the cell membrane to ‘‘seed’’ the extracellular LDL. High-density

Table 2

Adhesion Molecules Expressed in Atherosclerotic Lesions

Family

Molecule (CD nomenclature)

Selectins

E-selectin (ELAM-1, CD62E) P-selectin (CD62P, PAGEM) L-selectin (CD62L) ICAM-1 (CD54) ICAM-2 ICAM-3 (CD50) VCAM-1 (CD106) PECAM-1 (CD31)

Immunoglobulins

Expressed by Endothelial Endothelial Leukocytes Endothelial Endothelial Leukocytes Endothelial Endothelial

cells cells cells, leukocytes cells, platelets cells, smooth muscle cells cells, platelets, leukocytes

224

Corti et al.

lipoprotein (HDL) is considered strongly protective against atherosclerosis. Several mechanisms may explain the protective effect of HDL. HDL plays a pivotal role in the removal of excess cholesterol from the peripheral tissue through the reverse cholesterol transport mechanism. HDL may also protect by inhibiting lipoprotein oxidation and improving endothelial dysfuction. Recently, low HDL cholesterol was recognized as an independent risk factor, suggesting the possible need for therapy in patients with low HDL plasma levels [43]. Exogenous administration of recombinant HDL was found to restore endothelial function in patients with hypercholesterolemia and to induce regression of atherosclerotic plaque in cholesterol-fed rabbits [16,17]. Once they are lipid-enriched, macrophages transform into foam cells leading to the formation of fatty streaks (Fig. 6). These activated macrophages release mitogens and chemoattractants that perpetuate the process by recruiting additional macrophages as well as vascular smooth muscle cells from the media into the injured intima (Fig. 3). This process may eventually compromise the vascular lumen. The classic hypothesis of atherosclerosis initiation has been challenged by a recent observation demonstrating in the murine model that the majority of smooth muscle cells present in the subendothelial space of atherosclerotic lesions, as well as in graft vasculopathy and restenosis following arterial injury, derive from undifferentiated bone marrow cells and do not express replication of smooth muscle cells present in the media [35]. This important information could explain why antiproliferative agents did not completely succeed in the effort to control atherosclerotic disease and could led to the development of new therapeutic approaches.

Figure 6 Gross pathology of early atherosclerotic lesions in the abdominal aorta: the fatty streaks.

Pathophysiology of CAD

225

The recruited smooth muscle cells and macrophages significantly contribute to plaque growth, not only by increasing their number, but also by synthesizing extracellular matrix components. The ratio between smooth muscle cells and macrophages plays an important role in plaque vulnerability because of their lytic activity. Activated macrophages within atheroma are able to elaborate a number of matrix metalloproteinases (MMPs) and elastolytic enzymes, including certain nonmetalloenzymes such as cathepsins S and K (Table 3) [44]. MMPs are an ever-expanding family of endopeptidases with common functional domains and mechanisms of action discovered because of their ability to degrade extracellular matrix components [45]. These enzymes facilitate plaque disruption and thrombus formation by digesting the extracellular matrix [5]. Enzymes of the MMPs family can attack interstitial collagen fibrils, molecules ordinarily exceedingly resistant to proteolytic degradation [46]. Recent work has localized the potent elastases, cathepsins S and K, in macrophages and smooth muscle cells located in complicated human atherosclerotic lesions [47]. Thus, members of several proteinase families may participate in the degradation of structurally important components of the extracellular matrix. All protease families have endogenous inhibitors both tissue inhibitors of the metalloproteinase families (TIMPs) and the endogenous inhibitor of cathepsin S, cystein, have been localized in human atheroma [45]. The balanced production and degradation of extracellular matrix components may determine the changes in vascular wall shape seen in the process of vascular remodeling, plaque growth, and plaque disruption. The rate of progression of atherosclerotic lesions is variable and poorly understood. Atherogenesis and lesion progression may be expected to be linear. However, coronary angiographic studies have demonstrated that progression in humans is neither linear nor predictable. An analysis of various studies unexpectedly revealed that over 75% of acute fatal myocardial infarcts occur in areas supplied by mildly stenosed (10 mmHg in diastolic blood pressure upon standing. In the vast majority of patients with OH, the drop in blood pressure is detected within 2 min of assuming upright posture [63]. However, in certain patients, there is a delayed orthostatic intolerance and blood pressure progressively falls over 15 to 45 min. This response is termed a dysautonomic response to upright posture and can be observed during TTT. Orthostatic hypotension is a common finding in the elderly and is frequently not associated with symptoms [64]. In one prospective study of 223 patients with syncope, OH was detected in 69 patients (31%), but was common in patients for whom other probable causes of syncope were assigned following evaluation [63]. Syncope should not be attributed to orthostatic changes alone unless orthostatic symptoms are present or systolic blood pressure declines to less than 90 mmHg upon postural change [65]. Decreased intravascular volume and adverse effects of drugs are the most common causes of symptomatic OH in the elderly [64]. The status of the patient’s hydration preceding the syncopal event should be assessed as well as their daily intake of salt and alcohol. The

Evaluation of Syncope

667

medications most commonly contributing to orthostasis in the elderly are nitrates, diuretics, antihypertensives, anti-Parkinsonian drugs, antidepressants, and antipsychotics. Certain situational factors may predispose elderly patients to orthostasis, notably the redistribution of circulatory volume to the splanchnic circulation following a meal. An evaluation of 47 community-dwelling elders and 3 patients admitted for syncope evaluation indicated that symptomatic hypotension during TTT occurred earlier and increased in incidence from 12 to 22% in the postprandial state [66]. Orthostatic hypotension may also be an important manifestation of autonomic dysfunction resulting from a wide variety of diseases and drugs. The Shy-Drager syndrome is associated with autonomic failure and involvement of the corticospinal, extrapyramidal, and cerebellar tracts, including a Parkinson-like syndrome. Parkinson’s disease itself can be associated with autonomic dysfunction. Lesions of the spinal cord caused by cervical trauma and other disorders, such as transverse myelitis, syringomyelia, and various tumors, may result in severe OH. Orthostatic hypotension may also be a manifestation of peripheral neuropathy that involves sympathetic fibers. This may be associated with such diseases as diabetes mellitus, amyloidosis, porphyria, chronic alcoholism, and pernicious anemia. Neurally Mediated Disorders of Blood Pressure Control This collection of disorders involves a neural reflex mediated by the nucleus tractus solitarius, an area in the brainstem that regulates blood pressure. The receptors for the afferent limb of the reflex are contained in several possible locations; examples include the left ventricle (vasovagal syncope), carotid sinus body (carotid sinus hypersensitivity), pharynx, or trachea (cough syncope), and the bladder (postmicturition syncope). In susceptible individuals, stimulation of the reflex results in withdrawal of sympathetic vasomotor tone and vasodilatation with resultant syncope. When accompanied by bradycardia, the term ‘‘vasovagal’’ is used while the term ‘‘vasodepressor’’ is employed for reflex reactions that are not accompanied by significant bradycardia. Vasovagal Syncope The ‘‘simple faint’’ has traditionally been considered a disease of the young patient. Early studies of elderly institutionalized and community-dwelling patients indicated that syncope was neurally mediated in only 1 to 5% of cases [1,6]. At the time, neurally mediated syncope was diagnosed with the classical features of a precipitating event followed by nausea, pallor, and sweating. The elderly more commonly have syncope without a prodrome or with retrograde amnesia for the event, making diagnosis by history alone problematic. The more frequent use of TTT has suggested that neurally mediated syncope is much more common in the elderly than previously believed, and likely constitutes a significant fraction of those previously carrying an ‘‘unknown’’ diagnosis. Carotid Sinus Hypersensitivity Carotid sinus hypersensitivity (CSH) has recently been described as a relatively underappreciated cause of unexplained falls and syncope in the elderly, with a prevalence in elderly adults presenting to the emergency department with unexplained falls of 34% in one trial [67]. Attacks may be precipitated by factors that exert pressure on the carotid sinus (tight collars, shaving, sudden turning), a history of which is obtained in one-quarter of patients with this syndrome. Syncope from CSH occurs predominantly in men, 70% of whom are

668

Hummel and Maurer

over the age of 50; the majority of patients have coronary artery disease or hypertension. Other predisposing factors include cervical pathology such as enlarged lymph nodes, scars resulting from surgery to the neck, carotid body tumors, parotid, thyroid, and head and neck tumors. Carotid sinus hypersensitivity can manifest as bradycardia or vasodilatation and is classified according to its predominant hemodynamic manifestation. Cardioinhibitory CSH is defined as sinus pauses greater than 3 s, while vasodepressor CSH is denoted as a fall in systolic blood pressure greater than 50 mmHg; if both mechanisms are noted, the disorder is termed mixed-type CSH. The diagnosis is made with carotid sinus massage (CSM), which some investigators consider positive only if symptoms are also reproduced during the examination [68]. Tilt-Table Testing In TTT, patients are secured to a motorized table with a footboard and brought quickly to an upright position at an angle of 60j to 80j while undergoing ECG and beat-to-beat blood pressure monitoring. Various protocols have been evaluated for the provocation of neurally mediated syncope through TTT, either in the drug-free state or with various pharmacological agents (nitroglycerin, isoproterenol, adenosine). In general, sensitivity of TTT increases with the use of pharmacological agents, longer duration of TTT, and a greater angle of head uptilt. However, with the addition of pharmacological agents, the specificity declines [69,70]. In patients with negative EPS, rates of hypotension on upright TTT (without pharmacological provocation) of 27 to 67% are reported as compared to approximately 10% in controls without syncope [71–74]. A retrospective evaluation of TTT in 352 subjects with unexplained syncope (including 133 patients >65 years of age and 43 patients >80 years of age) showed an age-related decline in positive TTT. However, a surprisingly high proportion of elderly patients with unexplained syncope had a positive TTT (37% of patients aged 65 and older, and 23% of patients aged 80 and older) [75]. These data demonstrate that a large proportion of elderly patients may have neurally mediated syncope that may be difficult to diagnose based on clinical grounds alone. Upright TTT can confirm the diagnosis in these patients and can help reassure patients that they have an excellent long-term prognosis in regard to mortality. Carotid Sinus Massage Many physicians have been hesitant to perform CSM because of the theoretical risk of stroke. However, after exluding patients with clinical history of stroke or TIA and/or audible carotid artery bruits, the risk of neurological complications was 0.17 to 0.45% in two studies on the safety of the maneuver [68,76]. Patients with history of VF or VT should be excluded, as ventricular dysrhythmias during CSM have been reported, albeit rarely, in this setting. Carotid sinus massage should be performed on one side at a time by applying gentle digital pressure for 5 to 10 s at the bifurcation of the carotid artery, below the angle of the jaw at the level of the cricothyroid cartilage. The procedure was initially performed only in the supine position, but CSM in the upright position (usually performed during a TTT session) increases sensitivity [77]. The distinction between cardioinhibitory and vasodepressor types is important so the test should be performed with continuous beat-to-beat blood pressure monitoring in addition to continuous ECG if possible. Dual-chamber pacing has been shown to prevent falls in patients with cardioinhibitory CSH [67], but the benefits in mixedtype and vasodepressor forms of CSH are less clear.

Evaluation of Syncope

669

Primary Neurological Syncope As previously noted, primary neurological causes of syncope are relatively rare. Despite this finding, examination of current clinical practice indicates that specific neurological testing is performed in approximately half of syncope evaluations [12,41]. Seizures and Electroencephalography In patients with a clinical history compatible with a seizure or in whom the diagnosis is suspected after initial evaluation, EEGs may be beneficial in confirming the diagnosis. However, EEG should not be performed routinely in the evaluation of patients with syncope. In one study of 99 patients referred for EEG because of syncope, near-syncope, or related complaints, 15 had abnormal findings, but in only one patient was the final diagnosis or treatment of the syncope affected by the EEG. In addition, almost all patients with abnormal EEGs had clinical histories compatible with seizures or focal neurological findings on examination [78]. Strokes, Transient Ischemic Attacks, and Specific Testing In the elderly, cerebrovascular atherosclerosis and impaired cerebrovascular autoregulation secondary to hypertension can contribute to cerebral hypoperfusion in a variety of circumstances. However, syncope is generally not a primary manifestation of cerebrovascular disease unless disease of the posterior circulation is present. When secondary to vertebrobasilar disease, syncope is almost always accompanied by other neurological symptoms traced to ischemia in the same territory, such as vertigo and ataxia [79,80]. The common practice of cerebrovascular imaging via ultrasonography or other methods has not been extensively studied, but in the absence of suggestive history probably contributes little to diagnosis [81]. Brain imaging in the evaluation of syncope yields helpful information only in the setting of a suggestive history or abnormal findings on neurological examination. A review of 195 patients estimated the diagnostic yield of head CT at 4%; in all of these patients, a witnessed seizure or focal neurological examination was present [17]. In the setting of syncope and suspected trauma, head CT may be indicated to exclude a subdural hematoma. Suggested Diagnostic Algorithm Determining the underlying cause and risk stratification for subjects with syncope relies on the performance of a complete history, physical examination (including orthostatic vital signs), and ECG. Approximate one-third of patients have a clear diagnosis evident at presentation. Another third of patients have a particular etiology highly suggested after initial evaluation, and directed testing can be performed in this group to confirm the suspected diagnoses. Primary neurological causes of syncope are rare and specific neurological tests rarely are definitive in the absence of a suggestive history and physical examination. A suggested strategy for the diagnosis of syncope in the elderly patient is detailed below and in Figure 1. In the remaining patients with unexplained syncope after initial evaluation, cardiac testing, specifically with consideration of EPS, should be performed in patients with structural heart disease. For patients with recurrent unexplained syncope without structural heart or cardiac conduction system disease (or patients with structural heart disease and

670

Hummel and Maurer

Figure 1 Suggested diagnostic algorithm for elderly patients with syncope. (Adapted from Ref. 82.)

negative cardiac testing), TTT to evaluate for orthostatic hypotension and neurally mediated causes is appropriate. This should be performed prior to long-term ECG monitoring in order to identify the significant percent of subjects with neurally mediated syncope and resultant bradycardia who do not require pacing. If the etiology of syncope is still undetermined following directed cardiac and autonomic testing, long-term ECG monitoring should be employed. The ILR has excellent potential to obtain a diagnosis in these patients even if external monitoring is negative, particularly if episodes are separated by long time intervals.

CONCLUSIONS Syncope is common in the elderly and is associated with significant morbidity and mortality. It is a symptom, not a disease, and has a complex differential diagnosis ranging from benign to malignant causes; elderly patients commonly have multiple possible contributing factors. The assignment of the underlying etiology can be greatly facilitated through an understanding of syncope epidemiology and the utility of relevant diagnostic tests. Identifying patients with underlying structural heart disease is important for risk stratification. For

Evaluation of Syncope

671

patients with unexplained syncope, provocative tests (electrophysiological study, tilt-table test, and carotid sinus massage) are more likely to yield a definitive diagnosis than observational testing strategies.

ACKNOWLEDGMENTS Supported by the Society of Geriatric Cardiology and the National Institutes of Health. Dr. Hummel was supported by the Society of Geriatric Cardiology’s Merck Award. Dr. Maurer was supported by the National Institute on Aging, Bethesda, Maryland (K23AG00966A).

REFERENCES 1. 2.

3. 4. 5. 6. 7. 8. 9.

10. 11. 12.

13. 14.

15. 16.

Lipsitz LA, Wei JY, Rowe JW. Syncope in an elderly, institutionalised population: prevalence, incidence, and associated risk. Q J Med 1985; 55(216):45–54. McIntosh S, Da Costa D, Kenny RA. Outcome of an integrated approach to the investigation of dizziness, falls and syncope in elderly patients referred to a ‘‘syncope’’ clinic [comment]. Age Ageing 1993; 22(1):53–58. Soteriades ES, Evans JC, Larson MG, Chen MH, Chen L, et al. Incidence and prognosis of syncope. N Engl J Med 2002; 347(12):878–885. Day SC, Cook EF, Funkenstein H, Goldman L. Evaluation and outcome of emergency room patients with transient loss of consciousness. Am J Med 1982; 73(1):15–23. Kapoor WN, Karpf M, Wieand S, Peterson JR, Levey GS. A prospective evaluation and follow-up of patients with syncope. N Engl J Med 1983; 309(4):197–204. Kapoor W, Snustad D, Peterson J, Wieand HS, Cha R, et al. Syncope in the elderly. Am J Med 1986; 80(3):419–428. Kane R, Ouslander J, Abrass I. Essentials of Clinical Geriatrics. New York: McGraw-Hill, 1989. Murphy SL, Williams CS, Gill TM. Characteristics associated with fear of falling and activity restriction in community-living older persons. J Am Geriatr Soc 2002; 50(3):516–520. Rose MS, Koshman ML, Spreng S, Sheldon R. The relationship between health-related quality of life and frequency of spells in patients with syncope. J Clin Epidemiol 2000; 53(12):1209– 1216. Linzer M, Pontinen M, Gold DT, Divine GW, Felder A, et al. Impairment of physical and psychosocial function in recurrent syncope. J Clin Epidemiol 1991; 44(10):1037–1043. Getchell WS, Larsen GC, Morris CD, McAnulty JH. Epidemiology of syncope in hospitalized patients. J Gen Intern Med 1999; 14(11):677–687. Pires LA, Ganji JR, Jarandila R, Steele R. Diagnostic patterns and temporal trends in the evaluation of adult patients hospitalized with syncope. Arch Intern Med 2001; 161(15):1889– 1895. Sarasin FP, Louis-Simonet M, Carballo D, Slama S, Rajeswaran A, et al. Prospective evaluation of patients with syncope: a population-based study. Am J Med 2001; 111(3):177–184. Ammirati F, Colivicchi F, Santini M. Diagnosing syncope in clinical practice. Implementation of a simplified diagnostic algorithm in a multicentre prospective trial—the OESIL 2 Study (Osservatorio Epidemiologico della Sincope nel Lazio). Eur Heart J 2000; 21:935–940. Savage DD, Corwin L, McGee DL, Kannel WB, Wolf PA. Epidemiologic features of isolated syncope: the Framingham Study. Stroke 1985; 16(4):626–629. Kapoor WN, Hanusa BH. Is Syncope a Risk Factor for Poor Outcomes? Comparison of Patients With and Without Syncope. T Am Journal Med 1996; 100(6):646–655.

672

Hummel and Maurer

17. Linzer M, Yang EH, Estes NA III, Wang P, Vorperian VR, et al. Diagnosis syncope. Part 1: Value of history, physical examination, and electrocardiography. Clinical Efficacy Assessment Project of the American College of Physicians. Ann Intern Med 1997; 126(12):989–996. 18. Hoefnagels WA, Padberg GW, Overweg J, van der Velde EA, Roos RA. Transient loss of consciousness: the value of the history for distinguishing seizure from syncope. J Neurol 1991; 238(1):39–43. 19. Benbadis SR, Wolgamuth BR, Goren H, Brener S, Fouad-Tarazi F. Value of tongue biting in the diagnosis of seizures. Arch Intern Med 1995; 155(21):2346–2349. 20. Sheldon R, Rose S, Ritchie D, Connolly SJ, Koshman ML, et al. Historical criteria that distinguish syncope from seizures. J Am Coll Cardiol 2002; 40(1):142–148. 21. Cummings SR, Nevitt MC, Kidd S. Forgetting falls. The limited accuracy of recall of falls in the elderly. J Am Geriatr Soc 1988; 36(7):613–616. 22. McIntosh S, Lawson J, Kenny R. Clinical characteristics of vasodepressor, cardioinhibitory, and mixed carotid sinus syndrome in the elderly. Am J Med 1993; 95:203–208. 23. Shaw FE, Kenny RA. The overlap between syncope and falls in the elderly. Postgrad Med J 1997; 73(864):635–639. 24. Calkins H, Shyr Y, Frumin H, Schork A, Morady F. The value of the clinical history in the differentiation of syncope due to ventricular tachycardia, atrioventricular block, and neurocardiogenic syncope. Am J Med 1995; 98(4):365–373. 25. Alboni P, Brignole M, Menozzi C, Raviele A, Del Rosso A, et al. Diagnostic value of history in patients with syncope with or without heart disease. J Am Coll Cardiol 2001; 37(7):1921–1928. 26. Martin T, Hanusa B, Kapoor W. Risk stratification of patients with syncope. Anna Emerg Med 1997; 29(4):459–466. 27. Sarasin FP, Junod AF, Carballo D, Slama S, Unger PF, et al. Role of echocardiography in the evaluation of syncope: a prospective study. Heart 2002; 88(4):363–367. 28. Lindroos M, Kupari M, Heikkila J, Tilvis R. Prevalence of aortic valve abnormalities in the elderly: an echocardiographic study of a random population sample. J Am Coll Cardiol 1993; 21(5):1220–1225. 29. Grech ED, Ramsdale DR. Exertional syncope in aortic stenosis: evidence to support inappropriate left ventricular baroreceptor response. Am Heart J 1991; 121(2 Pt 1):603–606. 30. Wilmshurst PT, Willicombe PR, Webb-Peploe MM. Effect of aortic valve replacement on syncope in patients with aortic stenosis. Bri Heart J 1993; 70(6):542–543. 31. Braunwald E, Zipes DP, Libby P. Heart Disease: A Textbook of Cardiovascular Medicine. Philadelphia: W. B. Saunders, 2001:1675–1676. 32. Bayer AJ, Chadha JS, Farag RR, Pathy MS. Changing presentation of myocardial infarction with increasing old age. J Am Geriatr Soc 1986; 34(4):263–266. 33. Link MS, Lauer EP, Homoud MK, Wang PJ, Estes NA III. Low yield of rule-out myocardial infarction protocol in patients presenting with syncope. Am J Cardiol 2001; 88(6):706–707. 34. Gregoratos G, Abrams J, Epstein AE, Freedman RA, Hayes DL, et al. ACC/AHA/NASPE 2002 guideline update for implantation of cardiac pacemakers and antiarrhythmia devices: summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/NASPE Committee to Update the 1998 Pacemaker Guidelines). Circulation 2002; 106(16):2145–2161. 35. Tung RT, Shen WK, Hayes DL, Hammill SC, Bailey KR, et al. Long-term survival after permanent pacemaker implantation for sick sinus syndrome. Am J Cardiol 1994; 74(10):1016– 1020. 36. Shen WK, Hammill SC, Hayes DL, Packer DL, Bailey KR, et al. Long-term survival after pacemaker implantation for heart block in patients > or = 65 years. Am J Cardiol 1994; 74(6): 560–564. 37. DiMarco JP, Philbrick JT. Use of ambulatory electrocardiographic (Holter) monitoring [comment]. Ann Intern Med 1990; 113(1):53–68. 38. Bass EB, Curtiss EI, Arena VC, Hanusa BH, Cecchetti A, et al. The duration of Holter mon-

Evaluation of Syncope

39.

40. 41. 42. 43. 44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

673

itoring in patients with syncope. Is 24 hours enough? Arch Intern Med 1990; 150(5):1073– 1078. Kapoor WN, Cha R, Peterson JR, Wieand HS, Karpf M. Prolonged electrocardiographic monitoring in patients with syncope. Importance of frequent or repetitive ventricular ectopy. Am J Med 1987; 82(1):20–28. Gibson TC, Heitzman MR. Diagnostic efficacy of 24-hour electrocardiographic monitoring for syncope. Am J Cardiol 1984; 53(8):1013–1017. Pires LA, Jangadeesh RG, Jarandila R, SR. Diagnostic patterns and temporal trends in the evaluation of adult patients hospitalized with syncope. Arch Intern Med 2001; 161:1889–1895. Kus T, Nadeau R, Costi P, Molin F, Primeau R. Comparison of the diagnostic yield of Holter versus transtelephonic monitoring. Can J Cardiol 1995; 11(10):891–894. Cumbee SR, Pryor RE, Linzer M. Cardiac loop ECG recording: a new noninvasive diagnostic test in recurrent syncope. Southern Med Assoc J 1990; 83(1):39–43. Krahn AD, Klein GJ, Yee R, Takle-Newhouse T, Norris C. Use of an extended monitoring strategy in patients with problematic syncope. Reveal Investigators. Circulation 1999; 99(3):406– 410. Krahn AD, Klein GJ, Yee R, Norris C. Final results from a pilot study with an implantable loop recorder to determine the etiology of syncope in patients with negative noninvasive and invasive testing. Am J Cardiol 1998; 82(1):117–119. Krahn AD, Klein GJ, Yee R, Manda V. The high cost of syncope: cost implications of a new insertable loop recorder in the investigation of recurrent syncope. Am Heart J 1999; 137(5):870– 877. Krahn AD, Klein GJ, Yee R, Skanes AC. Randomized assessment of syncope trial: conventional diagnostic testing versus a prolonged monitoring strategy [comment]. Circulation 2001; 104(1): 46–51. Sugrue DD, Holmes DR Jr, Gersh BJ, Wood DL, Osborn MJ, et al. Impact of intracardiac electrophysiologic testing on the management of elderly patients with recurrent syncope or near syncope. J Am Geriatr Soc 1987; 35(12):1079–1083. Zipes DP, Wyse DG, Friedman PL, Epstein AE, Hallstrom AP, et al. A comparison of antiarrhythmic-drug therapy with implantable defibrillators in patients resuscitated from nearfatal ventricular arrhythmias. N Engl J Med 1997; 337:1576–1583. Sheldon R, Connolly S, Krahn A, Roberts R, Gent M, et al. Identification of patients most likely to benefit from implantable cardioverter-defibrillator therapy: the Canadian Implantable Defibrillator Study [comment]. Circulation 2000; 101(14):1660–1664. Kuck KH, Cappato R, Siebels J, Ruppel R. Randomized comparison of antiarrhythmic drug therapy with implantable defibrillators in patients resuscitated from cardiac arrest: the Cardiac Arrest Study Hamburg (CASH). Circulation 2000; 102(7):748–754. Pires LA, May LM, Ravi S, Parry JT, Lal VR, et al. Comparison of event rates and survival in patients with unexplained syncope without documented ventricular tachyarrhythmias versus patients with documented sustained ventricular tachyarrhythmias both treated with implantable cardioverter-defibrillators. Am J Cardiol 2000; 85(6):725–728. Garcia-Moran E, Mont L, Cuesta A, Matas M, Brugada J. Low recurrence of syncope in patients with inducible sustained ventricular tachyarrhythmias treated with an implantable cardioverter-defibrillator. Eur Heart J 2002; 23(11):901–907. Doherty JU, Pembrook-Rogers D, Grogan EW, Falcone RA, Buxton AE, et al. Electrophysiologic evaluation and follow-up characteristics of patients with recurrent unexplained syncope and presyncope. Am J Cardiol 1985; 55(6):703–708. Krol RB, Morady F, Flaker GC, DiCarlo LA Jr, Baerman JM, et al. Electrophysiologic testing in patients with unexplained syncope: clinical and noninvasive predictors of outcome. J Am Coll Cardiol 1987; 10(2):358–363. Komajda M, Jais JP, Reeves F, Goldfarb B, Bouhour JB, et al. Factors predicting mortality in idiopathic dilated cardiomyopathy. Eur Heart J 1990; 11(9):824–831.

674

Hummel and Maurer

57. Grimm W, Hoffmann J, Menz V, Luck K, Maisch B. Programmed ventricular stimulation for arrhythmia risk prediction in patients with idiopathic dilated cardiomyopathy and nonsustained ventricular tachycardia. J Am Coll Cardiol 1998; 32(3):739–745. 58. Das SK, Morady F, DiCarlo L Jr, Baerman J, Krol R, et al. Prognostic usefulness of programmed ventricular stimulation in idiopathic dilated cardiomyopathy without symptomatic ventricular arrhythmias. Am J Cardiol 1986; 58(10):998–1000. 59. Raviele A, Proclemer A, Gasparini G, Di Pede F, Delise P, et al. Long-term follow-up of patients with unexplained syncope and negative electrophysiologic study. Eur Heart J 1989; 10(2):127– 132. 60. Kushner JA, Kou WH, Kadish AH, Morady F. Natural history of patients with unexplained syncope and a nondiagnostic electrophysiologic study. J Am Coll Cardiol 1989; 14(2):391–396. 61. Steinberg JS, Prystowsky E, Freedman RA, Moreno F, Katz R, et al. Use of the signal-averaged electrocardiogram for predicting inducible ventricular tachycardia in patients with unexplained syncope: relation to clinical variables in a multivariate analysis. J Am Coll Cardiol 1994; 23(1):99–106. 62. Gold MR, Bloomfield DM, Anderson KP, El-Sherif NE, Wilber DJ, et al. A comparison of T-wave alternans, signal averaged electrocardiography and programmed ventricular stimulation for arrhythmia risk stratification [comment]. J Am Coll Cardiol 2000; 36(7):2247–2253. 63. Atkins D, Hanusa B, Sefcik T, Kapoor W. Syncope and orthostatic hypotension. Am J Med 1991; 91(2):179–185. 64. Lipsitz L. Orthostatic hypotension in the elderly. N Engl J Med 1989; 321:952–957. 65. Thomas JE, Schirger A, Fealey RD, Sheps SG. Orthostatic hypotension. Mayo Clin Proc 1981; 56(2):117–125. 66. Maurer MS, Karmally W, Rivadeneira H, Parides MK, Bloomfield DM. Upright posture and postprandial hypotension in elderly persons [comment]. Ann Intern Med 2000; 133(7):533–536. 67. Kenny RA, Richardson DA, Steen N, Bexton RS, Shaw FE, et al. Carotid sinus syndrome: a modifiable risk factor for nonaccidental falls in older adults (SAFE PACE) [comment]. J Am Coll Cardiol 2001; 38(5):1491–1496. 68. Puggioni E, Guiducci V, Brignole M, Menozzi C, Oddone D, et al. Results and complications of the carotid sinus massage performed according to the ‘‘method of symptoms. Am J Cardiol 2002; 89(5):599–601. 69. Mussi C, Tolve I, Foroni M, Valli A, Ascari S, et al. Specificity and total positive rate of head-up tilt testing potentiated with sublingual nitroglycerin in older patients with unexplained syncope. Aging (Milano) 2001; 13(2):105–111. 70. Kumar NP, Youde JH, Ruse C, Fotherby MD, Masud M. Responses to the prolonged head-up tilt followed by sublingual nitrate provocation in asymptomatic older adults. Age Ageing 2000; 29(5):419–424. 71. Calkins H, Kadish A, Sousa J, Rosenheck S, Morady F. Comparison of responses to isoproterenol and epinephrine during head-up tilt in suspected vasodepressor syncope. Am J Cardiol 1991; 67(2):207–209. 72. Grubb B, Temesy-Armos P, Hahn H, Elliott L. Utility of upright tilt-table testing in the evaluation and management of syncope of unknown origin [comment]. Am J Med 1991; 90(1):6–10. 73. Strasberg B, Rechavia E, Sagie A, Kusniec J, Mager A, et al. The head-up tilt table test in patients with syncope of unknown origin. Am Heart J 1989; 118(5 Pt 1):923–927. 74. Raviele A, Gasparini G, Di Pede F, Delise P, Bonso A, et al. Usefulness of head-up tilt test in evaluating patients with syncope of unknown origin and negative electrophysiologic study. Am J Cardiol 1990; 65(20):1322–1327. 75. Bloomfield D, Maurer M, Bigger J Jr. Effects of age on outcome of tilt-table testing. Am J Cardiol 1999; 83(7):1055–1058. 76. Munro N, McIntosh S, Lawson J, Morley C, Sutton R, et al. Incidence of complications after carotid sinus massage in older patients with syncope. J Am Geriatr Soc 1994; 42(12):1248– 1251.

Evaluation of Syncope

675

77. Morillo CA, Camacho ME, Wood MA, Gilligan DM, Ellenbogen KA. Diagnostic utility of mechanical, pharmacological and orthostatic stimulation of the carotid sinus in patients with unexplained syncope. J Am Coll Cardiol 1999; 34(5):1587–1594. 78. Davis TL, Freemon FR. Electroencephalography should not be routine in the evaluation of syncope in adults [comment]. Arch Intern Med 1990; 150(10):2027–2029. 79. Ausman JI, Shrontz CE, Pearce JE, Diaz FG, Crecelius JL. Vertebrobasilar insufficiency. A review. Arch Neurol 1985; 42(8):803–808. 80. Davidson E, Rotenbeg Z, Fuchs J, Weinberger I, Agmon J. Transient ischemic attack-related syncope. Clin Cardiol 1991; 14(2):141–144. 81. Stetson P, Maurer M, Quint E, Greene R, Hsu NS, et al. Underutilization of cardiovascular testing in elderly patients with syncope (abstract). Am J Geriatr Cardiol 1998; 7(2):58. 82. Brignole M, Alboni P, Benditt D, Bergfeldt L, Blanc JJ, et al. Guidelines on management (diagnosis and treatment) of syncope. Eur Heart J 2001; 22(15):1256–1306.

28 Anticoagulation in the Elderly Jonathan L. Halperin The Zena and Michael Wiener Cardiovascular Institute, Mount Sinai Medical Center, New York, New York, U.S.A.

PHARMACOLOGY OF COUMARIN ANTICOAGULANT DRUGS The coumarin derivative warfarin acts as an anticoagulant by inhibiting formation of the vitamin K–dependent coagulation factors II, VII, IX, and X and interfering with carboxylation of the coagulation-regulating proteins C and S. The anticoagulant and antithrombotic effects of warfarin can be distinguished; reduction of prothrombin and factor X are more important for the antithrombotic effect than reduction of factors VII and IX. Whereas the anticoagulant effect develops in 2 days, the antithrombotic effect requires 6 days of treatment [1]. The relationship between dose and response is influenced by genetics, environment [2,3], drugs, diet, and various diseases. Variability also results from laboratory inaccuracy, patient noncompliance, and miscommunication between patients and physicians. Other drugs may influence the pharmacokinetics of warfarin, reducing gastrointestinal absorption, disrupting metabolic clearance, or inducing hepatic enzymatic activity [4]. Long-term warfarin therapy is sensitive to fluctuating levels of dietary vitamin K [5,6], derived predominantly from the phylloquinone content of plant material [6]. Resistance to warfarin occurs in patients consuming green vegetables or vitamin K–containing supplements, while decreased dietary vitamin K1 intake in malabsorption states and in patients treated with antibiotics and intravenous fluids without vitamin K supplementation potentiates the effect of warfarin. Hepatic dysfunction potentiates the response to warfarin through impaired synthesis of coagulation factors. Hypermetabolic states associated with fever or hyperthyroidism increase responsiveness to warfarin, probably through catabolism of coagulation factors [7,8]. Drugs may inhibit synthesis or increase clearance of coagulation factors or interfere with other pathways of hemostasis. Aspirin [9] and nonsteroidal anti-inflammatory drugs [10] increase the risk of warfarinassociated bleeding by inhibiting platelet aggregation. These drugs can also cause gastric erosions and thereby increase the risk of upper gastrointestinal bleeding. The risk of clinically important bleeding is heightened by high doses of aspirin during high-intensity warfarin therapy (INR 3.0–4.5) [9,11]. In patients with prosthetic heart valves [12] and in asymptomatic individuals at high risk for coronary artery disease [13], low doses of aspirin (75 to 100 mg daily) were also associated with increased rates of minor bleeding when combined with moderate and low-intensity warfarin anticoagulation. 677

678

Halperin

DOSE ADJUSTMENT AND MONITORING OF ANTICOAGULATION INTENSITY The prothrombin time (PT) is the most common test used to monitor oral anticoagulant therapy [14]. The PT responds to reduction of three of the four vitamin K–dependent procoagulant clotting factors (II, VII, and X) at a rate proportionate to their respective halflives. Thus, during the first few days of warfarin therapy, the PT reflects mainly reduction of factor VII, which has a half-life of approximately 6 h, while reduction of factors X and II subsequently contributes. The assay is performed on citrated plasma by adding calcium and thromboplastin (a phospholipid-protein extract containing both the tissue factor and phospholipid necessary to promote activation of factor X by factor VII) [15]. Because of the variability of thromboplastin responsiveness to warfarin, PT monitoring of warfarin treatment is imprecise when expressed as a ratio of the PT value of the patient’s plasma over that of normal control plasma. Adoption of the INR standard increases the safety of monitoring oral anticoagulant therapy. The INR is calculated as follows: INR ¼ ðpatient PT=mean normal PTÞISI or log INR ¼ ISIðlog observed PT ratioÞ; where ISI denotes the International Sensitivity Index of the thromboplastin used to perform the PT measurement. The ISI reflects the responsiveness of a given thromboplastin to reduction of the vitamin K–dependent coagulation factors compared to the primary international standard; the more responsive the reagent, the lower the ISI value [15]. As warfarin therapy is initiated, the INR should be checked daily until it reaches the therapeutic range and is sustained in this range for 2 consecutive days, then 2 or 3 times weekly for 1 to 2 weeks, and then less often, depending upon the stability of the results. Once the INR becomes stable, the frequency of testing can be reduced to intervals as long as 4 weeks. Dose adjustments require more frequent monitoring. The safety and effectiveness of warfarin therapy depends critically on maintaining the INR within the therapeutic range. On-treatment analysis of the primary prevention trials in atrial fibrillation found that a disproportionate number of thromboembolic and bleeding events occurred when the PT ratio was outside the therapeutic range [16], and subgroup analyses of other cohort studies have also shown an increased risk of bleeding when the INR exceeds the therapeutic range [17–20]and increased risk of thromboembolism when the INR falls below 2.0 [21,22]. One of the challenges inherent in warfarin therapy, however, is the difficulty maintaining anticoagulation intensity within such a narrow range. In the primary prevention trials in atrial fibrillation, for example, despite management of a carefully selected patient population by dedicated staff following rigid research protocols, INR values fell within the target range for 42 to 83% of net exposure (Fig. 1). Point-of-care (POC) PT measurements offer the potential to simplify oral anticoagulation management in both the physician’s office and in the patient’s home. Over a 6month period, among 325 newly treated elderly patients randomized to either conventional treatment by personal physicians based on venous sampling or adjustment of dosage by a central investigator based on INR results from patient self-testing at home, the rate of hemorrhage was 12% in the usual care group compared to 6% in the self-testing group [23].

Anticoagulation

679

Figure 1 The challenge of maintaining warfarin anticoagulation intensity within the target therapeutic range was demonstrated in trials involving patients with nonvalvular atrial fibrillation. Despite more favorable conditions than those generally encountered in clinical practice, approximate INR values fell within the targeted range only about half the time. (AFASAK [81], BAATAF [88], SPAF [102], CAFA [103], SPINAF [104], EAFT [77])

Coupled with self-testing, self-management using POC instruments offers independence and freedom of travel to selected patients. When compared with the outcomes of highquality anticoagulation clinic management, there was little difference between groups in time spent within therapeutic range, but patient self-management was associated with less deviation from the target INR. Several computer programs have been developed to guide warfarin dosing based on querying physicians [24], Bayesian forecasting [25], and mathematical equations [26]. In the INR range of 2.0 to 3.0, computerized dosage programs were comparable to dosing by experienced clinic staff, but the computerized approach achieved significantly better control at the higher intensity range of INR 3.0 to 4.5 [27]. Computer-guided warfarin dose adjustment may also be superior to manual dose regulation by inexperienced personnel.

MANAGEMENT OF PATIENTS WITH HIGH INR VALUES Even though the risk of bleeding increases, the daily absolute risk of bleeding is still low when the INR exceeds 4 to 5. The first step when this occurs is to stop warfarin; the second is to administer vitamin K1; the third and most rapidly effective measure is to infuse fresh plasma or prothrombin concentrate. Since no randomized trials have compared these strategies using clinical endpoints, the choice is based on clinical judgment. The INR falls over several days once warfarin is interrupted, reaching the normal range in 4 to 5 days when INR is between 2.0 and 3.0. In contrast, the INR declines substantially within 24 h after oral administration of vitamin K1, 1 to 2.5 mg [28], more rapidy than simply withholding warfarin [29], making oral vitamin K1 the treatment of choice unless more rapid reversal of anticoagulation is required, in which case vitamin K1 can be administered by slow intravenous infusion [30].

680

Halperin

BLEEDING DURING ORAL ANTICOAGULANT THERAPY Factors other than the intensity of anticoagulation that contribute to bleeding are the underlying clinical disorder [31,32] and concomitant administration of aspirin, nonsteroidal anti-inflammatory drugs, or other drugs that impair platelet function, produce gastric erosions, or impair synthesis of clotting factors. Bleeding that occurs at INR 75 years old (1.8% per yr) offset the reduction in ischemic stroke (Fig. 3). The SPAF-II trials demonstrated only small benefits of warfarin over aspirin for unselected AF patients (number needed-to-treat >200 for reduction of ischemic and hemorrhagic stroke) and drew attention to the importance of stroke risk stratification to identify AF patients who benefit substantially from anticoagulation. In the SPAF-III trial, adjusted-dose warfarin (INR 2.0 to 3.0) was much more effective than the combination of fixed, low-dose warfarin (initial INR 1.2 to 1.5) and aspirin (325 mg daily) in patients with AF and additional risk factors for stroke (1.9% versus 7.9%; 95% CI, 3.4 to 8.6; p < 0.0001) without an increase in hemorrhagic complications [22]. The

Figure 3 Rates of disabling stroke in the Stroke Prevention in Atrial Fibrillation (SPAF)-II study. The upper portions of each bar indicate hemorrhagic strokes, while the lower portions represent ischemic strokes. Event rates were higher among elderly patients than younger ones, and the risk of disabling brain hemorrhage on warfarin (mean INR 2.6) outweighed the net gain from reduction of ischemic events, compared to treatment with aspirin, 325 mg daily [82]. Bleeding rates in elderly patients with AF were higher in the SPAF-II study than in other trials, and anticoagulation is currently recommended for most older patients with AF.

686

Halperin

AFASAK-2 study was similarly designed to evaluate the effectiveness of fixed minidose warfarin and aspirin, alone or in combination, for prevention of stroke in patients with AF, but it was stopped prematurely based on the results of SPAF-III [83]. There is a clear reduction in the risk of stroke or systemic embolization associated with the use of adjusted-dose warfarin in patients with AF. A pooled analysis found that adjusted-dose warfarin reduces the risk of stroke by about two-thirds versus control [79,80]. The magnitude of the benefit associated with adjusted-dose warfarin was fairly consistent among the five trials despite variations in design and the intensity of anticoagulation. Current treatment guidelines recommend a target INR of 2.5 (range 2.0 to 3.0) for patients with nonvalvular (nonrheumatic) AF who have risk factors for thromboembolism [79,84]. Compared with aspirin, adjusted-dose warfarin reduces the overall risk of stroke by 36% (CI, 14 to 52%) and ischemic stroke by 46% (CI, 27 to 60%) [80]. The discrepancy is driven largely by the results of the SPAF-II trial (mean INR 2.6) in an patient cohort older than enrolled in many other studies [85]. The risk of intracranial bleeding among patients receiving adjusted-dose warfarin was substantially increased in this study among patients older than 75 years of age. Exclusion of SPAF-II data from the meta-analysis revealed a 49% reduction in the risk of all stroke for adjusted-dose warfarin compared with aspirin (95% CI 26 to 65%) [80]. Despite the well-documented efficacy of warfarin for prevention of stroke, fewer than 40% of patients with AF who are eligible for adjusted-dose warfarin therapy receive warfarin [86]. Concern regarding hemorrhagic complications, especially intracranial bleeding, is thought to influence the use of anticoagulation therapy in prevention of stroke. A recent meta-analysis of randomized trials of anticoagulant/antiplatelet prevention of stroke in patients with AF reported only a 0.2% absolute annual increase in major extracranial bleeding among patients receiving warfarin compared with those receiving aspirin, but a 2.1fold increase in intracranial bleeding [80]. The use of aspirin was not associated with significantly increased rates of major bleeding compared with placebo. Although it is apparent that adjusted-dose warfarin carries the risk of hemorrhagic complications, the stroke-prevention benefit seen in patients with AF justifies the use of this therapy, particularly for secondary prevention. For most patients, therefore, the risk of bleeding does not offset the benefit of therapy. The magnitude of the absolute reduction in stroke achieved through prophylactic therapy in patients with AF varies based upon the inherent risk for stroke [80]. For example, in patients at relatively low risk receiving doseadjusted warfarin as primary prevention, six strokes may be prevented at the cost of three major extracranial bleeds per 1000 patients treated per year, whereas in patients with a high risk of stroke receiving dose-adjusted warfarin, 36 strokes may be prevented per 1000 patients per year treated at the same cost of bleeding. As a result, risk-stratification algorithms have been developed to assist clinicians considering the benefits and bleeding risks of therapy. Deciding which course of antithrombotic therapy is appropriate for a particular patient ultimately depends on the risks of stroke and major hemorrhage, adequacy of monitoring, and patient preference. In balancing the risks of warfarin against its benefits, many physicians conclude that the risk of hemorrhage outweighs the protective effect of prophylaxis with warfarin. Other patient-, physician-, and technology-related impediments to stroke prophylaxis with warfarin have been described [87]. The narrow therapeutic margin of warfarin and interactions with countless other medications and with foods and alcohol contribute to

Anticoagulation

687

considerable underuse of oral anticoagulant therapy in patients with AF who stand to benefit. According to a 1996 national ambulatory care survey, only 33% of patients with AF in whom warfarin was indicated were taking it [86], whereas aspirin was used by 20% of patients not using warfarin. A multiple logistic regression analysis found that patients with AF and a history of stroke were more likely to receive warfarin, and those seeing cardiologists or internists had a higher rate of warfarin use than those under the care of family or general practitioners. Patients older than 80 years were less likely to receive warfarin (odds ratio, 0.60; 95% CI 0.37 to 0.98) than younger patients [86]. Of 312 persons with AF, average age 84 years, who resided in a chronic care facility, rates of stroke over 3 years in those not anticoagulated were 56% in those with no prior history of thromboembolism and 81% in those with prior thromboembolism [78]. Patients 65 years or older with AF and a history of ischemic stroke are at an exceptionally high risk of recurrent thrombotic events (>10% per year), yet a study of such cases among Medicare recipients found warfarin used suboptimally [88,89]. Brass and colleagues found that only 53% of stroke survivors were prescribed warfarin on discharge, and aspirin was prescribed in only 38% of those not receiving warfarin, indicating that a substantial percentage of stroke survivors did not receive any medication at all for secondary prevention [89]. There was a slight, but statistically significant, increase in use of warfarin among patients who had no contraindications to anticoagulation but no increase among patients with additional risk factors for ischemic stroke, which supports the hypothesis that fear of hemorrhage plays a large role in the selection of patients for anticoagulant therapy.

ANTICOAGULATION IN ELDERLY PATIENTS WITH ATRIAL FIBRILLATION Concern over hemorrhagic complications, especially catastrophic intracranial bleeding, influences the use of anticoagulation therapy for prevention of stroke, particularly among patients z80 years old. Older patients with AF are at an increased risk of both ischemic stroke and bleeding. In one study in an academic, hospital-based geriatrics practice, 90% of patients at mean age of 80 years with electrocardiographically documented AF were categorized at high risk for stroke and had no contraindications to warfarin [90]. Only 49% of these patients were receiving adjusted-dose warfarin; 36% were taking aspirin; and the remaining 15% received no prophylaxis. Similar findings were reported in the Cardiovascular Health Study, in which only 37% of these high-risk patients with AF were receiving adjusted-dose warfarin, whereas 47% received aspirin [91]. To assist clinicians in balancing the risks and benefits of anticoagulation in elderly patients based on age and weighted risks of hemorrhage and thromboembolism rather than clinical judgment, Eckman and coworkers proposed a decision-analysis tool [92]. Patient-specific risk ratios are plotted on a decision grid leading to three possible outcomes: anticoagulate; do not anticoagulate; and neither decision will have a significant impact on risk. Currently available antithrombotic regimens for prevention of stroke in patients with AF are saddled with limitations. Warfarin is much more effective than aspirin for prevention of stroke, but is used by less than 40% of patients with AF for whom it is indicated, mainly

688

Halperin

because of the real and perceived risks of hemorrhage as well as the inconvenience and expense imposed by the need for frequent monitoring. An effective, safe oral anticoagulant exhibiting a predictable linear dose response would preclude the routine laboratory monitoring necessary with warfarin. An agent with minimal side effects or food–medication interactions is desirable for patients with AF who are typically more than 65 years old and taking nearly five other medications for various medical conditions. Rapid onset of therapeutic effect and easy reversibility would constitute significant additional advantages over warfarin, which takes several days to become fully effective when started and a similarly delayed offset of anticoagulant effect when discontinued. The oral agent ximelagatran—a prodrug of the direct thrombin inhibitor, melagatran—exhibits rapid absorption and biotransformation to the active form. Melagatran is primarily excreted renally. The linear dose–response relationships and wide therapeutic windows observed with both ximelagatran and melagatran obviate the need for routine anticoagulation monitoring [93–96]. Drug interaction studies involving cytochrome-P450 enzymes and food interaction studies are ongoing with ximelagatran, but no interactions have been reported to date. The drug is currently under investigation for the prevention and treatment of venous thromboembolism and for prevention of ischemic stroke in patients with AF. The efficacy and safety of ximelagatran for prevention of stroke in patients with AF is being evaluated in the Stroke Prevention Using Oral Thrombin Inhibition in Atrial Fibrillation (SPORTIF) clinical trials. SPORTIF-II was a randomized, parallel group, dose-finding, phase 2 trial involving patients with nonvalvular AF and at least one additional risk factor for stroke (median age 70 years). The primary outcome was the number of adverse events (thromboembolism or stroke). A total of 257 patients were randomized to treatment in one of four groups; three groups (n = 187) received ximelagatran 20, 40, or 60 mg twice daily, and the fourth was treated with adjusted-dose warfarin (INR 2.0 to 3.0). Over 3 months, two of the 67 patients in the warfarin group (3%) experienced TIA. In the ximelagatran groups, one of 187 patients had a TIA (0.5%), and one developed ischemic stroke (0.5%); both events occurred in the 60-mg arm. Major bleeding occurred in only one patient who was receiving warfarin. Minor bleeding rates were low and similar among all the groups. Ximelagatran was well tolerated and did not require routine anticoagulation monitoring or dose adjustment. SPORTIF-IV is an open-label extension of SPORTIF-II in which patients originally randomized to ximelagatran continued this agent in a fixed dose of 36 mg twice daily, and the others continued warfarin to prevent stroke and systemic embolism. Two phase 3 trials—SPORTIF-III (in 25 nations, open label) and SPORTIF-V (in North America, double blind)—are each randomizing more than 3000 patients with AF to treatment with adjusted-dose warfarin or ximelagatran, 36 mg twice daily. As an emerging therapeutic modality, oral direct thrombin inhibitors may have a substantial impact on the management of patients with AF in the future.

OTHER INDICATIONS FOR ORAL ANTICOAGULANT THERAPY Other widely accepted indications for oral anticoagulant therapy have not been evaluated in properly designed clinical trials. Among these are atrial fibrillation associated with valvular heart disease and mitral stenosis in the presence of sinus rhythm. Long-term oral

Anticoagulation

689

anticoagulation (INR 2.0 to 3.0) is also indicated in patients who have sustained one or more episodes of systemic thromboembolism. Anticoagulants are not presently indicated in patients with ischemic cerebrovascular disease [20,97]. Reduced left ventricular systolic function is associated with both stroke and mortality even in the absence of documented atrial fibrillation [98]. Warfarin is used frequently in patients with dilated cardiomyopathy, although no randomized trials have confirmed the benefit of anticoagulation [99]. Longterm anticoagulant therapy is also indicated in patients with ischemic stroke of unknown origin who have a combination of a patent foramen ovale and atrial septal aneurysm, since these patients have an increased risk of recurrent stroke despite treatment with aspirin [100,101].

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9.

10. 11.

12.

13.

14. 15. 16. 17. 18.

Wessler S, Gitel SN. Warfarin: From bedside to bench. N Engl J Med 1984; 311:645–652. Breckenridge AM. Oral anticoagulant drugs: pharmacokinetic aspects. Semin Hematol 1978; 15:19–26. O’Reilly RA. Vitamin K and other oral anticoagulant drugs. Annu Rev Med 1976; 27:245–261. Cropp JS, Bussey HI. A review of enzyme induction of warfarin metabolism with recommendations for patient management. Pharmacotherapy 1997; 17:917–928. O’Reilly R, Rytand D. Resistance to warfarin due to unrecognized vitamin K supplementation. N Engl J Med 1980; 303:160–161. Suttie JW, Muhah-Schendel LL, Shah DV, et al. Vitamin K deficiency from dietary vitamin K restriction in humans. Am J Clin Nutr 1988; 47:475–480. Richards RK. Influence of fever upon the action of 3,3-methylene bis-(4-hydroxoycoumarin). Science 1943; 97:313–316. Owens JC, Neely WB, Owen WR. Effect of sodium dextrothyroxine in patients receiving anticoagulants. N Engl J Med 1962; 266:76–79. Dale J, Myhre E, Loew D. Bleeding during acetysalicylic acid and anticoagulant therapy in patients with reduced platelet reactivity after aortic valve replacement. Am Heart J 1980; 99:746–751. Schulman S, Henriksson K. Interaction of ibuprofen and warfarin on primary haemostasis. Br J Rheumatol 1989; 38:46–49. Chesebro JH, 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; 51:1537–1541. Turpie AGG, Gent M, Laupacis A, et al. Aspirin and warfarin after heart-valve replacement. A comparison of aspirin with placebo in patients treated with warfarin after heart-valve replacement. N Engl J Med 1993; 329:524–529. The Medical Research Council General Practice Research Framework. Thrombosis prevention trial: randomised trial of low-intensity oral anticoagulation with warfarin and low-dose aspirin in the primary prevention is ischaemic heart disease in men at increased risk. Lancet 1998; 351: 233–241. Quick AJ. The prothrombin time in haemophilia and in obstructive jaundice. J Biol Chem 1935; 109:73–74. Poller L. Progress in standardisation in anticoagulant control. Hematol Rev 1987; 1:225–241. Hylek EM, Singer DE. Risk factors for intracranial hemorrhage in outpatients taking warfarin. Ann Intern Med 1994; 120:897–902. ASPECT Research Group. Effect of long-term oral anticoagulant treatment on mortality and cardiovascular morbidity after myocardial infarction. Lancet 1994; 343:499–503. Cannegieter SC, Rosendaal FR, Wintzen AR, et al. Optimal oral anticoagulant therapy in

690

19.

20. 21.

22.

23. 24. 25. 26.

27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41.

Halperin patients with mechanical heart valve prostheses: the Leiden artificial valve and anticoagulation study. N Engl J Med 1995; 333:11–17. Stroke Prevention in Reversible Ischemia Trial (SPIRIT) Study Group. A randomized trial of anticoagulants versus aspirin after cerebral ischemia of presumed arterial origin. Ann Neurol 1997; 42:857–865. Mohr JP, Thompson JLP, Lazar RM, et al. A comparison of warfarin and aspirin for the prevention of recurrent ischemic storke. N Engl J Med 2001; 345:1444–1451. Hylek EM, Skates SJ, Sheehan MA, et al. An analysis of the lowest effective intensity of prophylactic anticoagulation for patients with nonrheumatic atrial fibrillation. N Engl J Med 1996; 335:540–546. Stroke Prevention in Atrial Fibrillation Investigators. Adjusted-dose warfarin versus lowintensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation: Stroke Prevention in Atrial Fibrillation III randomised clinical trial. Lancet 1996; 348:633–638. Beyth RJ, Landefeld CS. Prevention of major bleeding in older patients treated with warfarin: results of a randomized trial [abstract]. J Gen Intern Med 1997; 12:66. Wilson R, James AH. Computer assisted management of warfarin treatment. BMJ 1984; 289: 422–424. White RH, Mungall D. Outpatient management of warfarin therapy: Comparison of computerpredicted dosage adjustment to skilled professional care. Ther Drug Monogr 1991; 13:46–50. Poller L, Shiach CR, MacCallum PK, et al. Multicentre randomised study of computerised anticoagulant dosage. European Concerted Action on Anticoagulation. Lancet 1998; 352:1505– 1509. Poller L, Wright D, Rowlands M. Prospective comparative study of computer programs used for management of warfarin. J Clin Pathol 1993; 46:299–303. White RH, McKittrick T, Hutchinson R, et al. Temporary discontinuation of warfarin therapy: Changes in the international normalized ratio. Ann Intern Med 1995; 122:40–42. Park BK, Scott AK, Wilson AC, et al. Plasma disposition of vitamin K1 in relation to anticoagulant poisoning. Br J Clin Pharmacol 1984; 18:655–662. Crowther MA, Donovan D, Harrison L, et al. Low-dose oral vitamin K reliably reverses overanticoagulation due to warfarin. Thromb Haemost 1998; 79:1116–1118. Landefeld CS, Rosenblatt MW. Bleeding in outpatients treated with warfarin: Relation to the prothrombin time and important remediable lesions. Am J Med 1989; 87:153–159. Levine MN, Raskob G, Landefeld S, et al. Hemorrhagic complications of anticoagulant treatment. Chest 1995; 108(suppl):276s–290s. Landefeld CS, Goldman L. Major bleeding in outpatients treated with warfarin: Incidence and prediction by factors known at the start of outpatient therapy. Am J Med 1989; 87:144–152. Beyth RJ, Quinn LM, Landefeld CS. Prospective evaluation of an index for predicting the risk of major bleeding in outpatients treated with warfarin. Am J Med 1998; 105:91–99. 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; 307:1676–1681. Kearon C, Hirsh J. Management of anticoagulation before and after elective surgery. N Engl J Med 1997; 336:1506–1511. Verhagen H. Local hemorrhage and necrosis of the skin and underlying tissues at starting therapy with dicumarol or dicumacyl. Acta Med Scand 1954; 148:455–467. Weinberg AC, Lieskovsky G, McGehee WG, et al. Warfarin necrosis of the skin and subcutaneous tissue of the male genitalia. J Urol 1983; 130:352–354. Broekmans AW, Bertina RM, Loeliger EA, et al. Protein C and the development of skin necrosis during anticoagulant therapy. Thromb Haemost 1983; 49:244–251. Zauber NP, Stark MW. Successful warfarin anticoagulation despite protein C deficiency and a history of warfarin necrosis. Ann Intern Med 1986; 104:659–660. Samama M, Horellou MH, Soria J, et al. Successful progressive anticoagulation in a severe protein C deficiency and previous skin necrosis at the initiation of oral anticoagulation treatment. Thromb Haemost 1984; 51:332–333.

Anticoagulation

691

42. Grimaudo V, Gueissaz F, Hauert J, Sarraj A, Kruithof EK, Bachmann F. Necrosis of skin induced by coumarin in a patient deficient in protein S. BMJ 1989; 298:233–234. 43. Francis CW, Marder VJ, Evarts CM, et al. Two-step warfarin therapy: prevention of postoperative venous thrombosis without excessive bleeding. JAMA 1983; 249:374–378. 44. Powers PJ, Gent M, Jay RM, et al. A randomized trial of less intense postoperative warfarin or aspirin therapy in the prevention of venous thromboembolism after surgery for fractured hip. Arch Intern Med 1989; 149:771–774. 45. Taberner DA, Poller L, Burslem RW, et al. Oral anticoagulants controlled by the British comparative thromboplastin versus low-dose heparin in prophylaxis of deep vein thrombosis. BMJ 1978; 1:272–274. 46. Poller L, McKernan A, Thomson JM, et al. Fixed minidose warfarin: A new approach to prophylaxis against venous thrombosis after major surgery. BMJ 1987; 295:1309–1312. 47. Bern MM, Lokich JJ, Wallach SR, et al. Very low doses of warfarin can prevent thrombosis in central venous catheters. Ann Intern Med 1990; 112:423–428. 48. Poller L, MacCallum PK, Thomson JM, et al. Reduction of factor VII coagulant activity (VIIC), a risk ractor for ischaemic heart disease, by fixed dose warfarin: A double blind crossover study. Br Heart J 1990; 63:231–233. 49. Dale C, Gallus AS, Wycherley A, et al. Prevention of venous thrombosis with minidose warfarin after joint replacement. BMJ 1991; 303:224. 50. Fordyce MJF, Baker AS, Staddon GE. Efficacy of fixed minidose warfarin prophylaxis in total hip replacement. BMJ 1991; 303:219–220. 51. Poller L, Thomson JM, MacCallum PK, et al. Minidose warfarin and failure to prevent deep vein thrombosis after joint replacement surgery despite inhibiting the postoperative riise in plasminogen activator inhibitor activity. Clin Appl Thromb Hemostas 1995; 1:267– 273. 52. Hirsh J. The optimal duration of anticoagulant therapy for venous thrombosis. N Engl J Med 1995; 332:1710–1711. 53. 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; 302:855–858. 54. 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; 306:189–194. 55. Lagerstedt CI, Fagher BO, Albrechtsson U, et al. Need for long-term anticoagulant treatment in symptomatic calf-vein thrombosis. Lancet 1985; 2:515–521. 56. Schulman S, Rhedin A, 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; 332:1661–1665. 57. Schulman S, Granqvist S, Holmstrom M, et al. the Duration of Anticoagulation Trial Study Group. The duration of oral anticoagulant therapy after a second episode of venous thromboembolism. N Engl J Med 1997; 336:393–398. 58. 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; 104:332–338. 59. Simioni P, Prandoni P, Zanon E, et al. Deep venous thrombosis and lupus anticoagulant: A case-control study. Thromb Haemost 1996; 76:187–189. 60. Rance A, Emmerich J, Fiessinger JN. Anticardiolipin antibodies and recurrent thromboem bolism. Thromb Haemost 1997; 77:221–222. 61. Veterans Administration Cooperative Study Group. Anticoagulants in acute myocardial infarction: Results of a cooperative clinical trial. JAMA 1973; 225:724–729. 62. Drapkin A, Merskey C. Anticoagulant therapy after acute myocardial infarction. JAMA 1972; 222:541–548. 63. Medical Research Council Group. Assessment of short-term anticoagulant administration after cardiac infarction: report of the Working Party on Anticoagulant Therapy in Coronary Thrombosis. BMJ 1969; 1:335–342.

692

Halperin

64.

Cairns JA, Hirsh J, Lewis HD Jr, et al. Antithrombotic agents in coronary artery disease. Chest 1992; 102(suppl):456s–481s. Goldberg RJ, Gore JM, Dalen JE, et al. Long term anticoagulant therapy after acute myocardial infarction. Am Heart J 1985; 109:616–622. Leizorovicz A, Boissel JP. Oral anticoagulant in patients surviving myocardial infarction. Eur J Clin Pharmacol 1983; 24:333–336. Sixty-Plus Reinfarction Study Group. A double-blind trial to assess long-term oral anticoagulants therapy in elderly patients after myocardial infarction. Lancet 1980; 2:989–994. Smith P, Arnesen H, Holme I. The effect of warfarin on mortality and reinfarction after myocardial infarction. N Engl J Med 1990; 323:147–151. Gohlke-Barwolf, Acar J, Oakley C, et al., for the Study Group of the Working Group on Valvular Heart Disease of the European Society of Cardiology. Eur Heart J 1995; 16:1320– 1330. Stein PD, Alpert JS, Dalen JE, et al. Antithrombotic therapy in patients with mechanical and biological prosthetic heart valves. Chest 1998; 114(suppl):602–610. The American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients with Valvular Heart Disease). ACC/AHA guidelines for the management of patients with valvular heart disease. J Am Coll Cardiol 1998; 32:1486–1588. Vongpatanasin W, Hillis LD, Lange RA. Prosthetic heart valves. N Engl J Med 1996; 335:407– 416. Feinberg WM, Blackshear JL, Laupacis A, et al. Prevalence, age distribution, and gender of patients with atrial fibrillation: analysis and implications. Arch Intern Med 1995; 155:469–473. Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults. National implications for rhythm management and stroke prevention: the AnTicoagulation and Risk factors In Atrial fibrillation (ATRIA) study. JAMA 2001; 285:2370–2375. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation: a major contributor to stroke in the elderly. The Framingham study. Arch Intern Med 1987; 147:1561–1564. Atrial Fibrillation Investigators. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomized controlled trials. Arch Intern Med 1994; 154:1449–1457. EAFT (European Atrial Fibrillation Trial) study group. Secondary prevention in non-rheumatic atrial fibrillation after transient ischaemic attack or minor stroke. Lancet 1993; 342:1255– 1262. Aronow WS, Ahn C, Kronzon I, Gutstein H. Incidence of new thromboembolic stroke in persons 62 years and older with chronic atrial fibrillation treated with warfarin versus aspirin. J Am Geriatr Soc 1999; 47:366–368. Albers GW, Dalen JE, Laupacis A, et al. Antithrombotic therapy in atrial fibrillation. Chest 2001; 119:194s–206s. Hart RG, Benavente O, McBride R, et al. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis. Ann Intern Med 1999; 131:492–501. Petersen P, Boysen G, Godtfredsen J, et al. Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: the Copenhagen AFASAK study. Lancet 1989; 1:175–179. Stroke Prevention in Atrial Fibrillation Investigators. Warfarin compared to aspirin for prevention of thromboembolism in atrial fibrillation: Results of the Stroke Prevention in Atrial Fibrillation II study. Lancet 1994; 343:687–691. Gullov AL, Koefoed BG, Petersen P. Bleeding during warfarin and aspirin therapy in patients with atrial fibrillation: the AFASAK 2 study. Atrial Fibrillation Aspirin and Anticoagulation. Arch Intern Med 1999; 159:1322–1328. Fuster V, Ryde´n LE, Asinger RW, et al. ACC/AHA/ESC guidelines for the management of patients with atrial fibrillation: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology

65. 66. 67. 68. 69.

70. 71.

72. 73. 74.

75. 76.

77.

78.

79. 80. 81.

82.

83.

84.

Anticoagulation

85.

86. 87. 88.

89. 90.

91. 92. 93.

94.

95.

96. 97. 98. 99. 100. 101. 102. 103. 104.

105.

693

Committee for Practice Guidelines and Policy Conferences (Committee to Develop Guidelines for the Management of Patients With Atrial Fibrillation). J Am Coll Cardiol 2001; 38:1231– 1266. 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; 343:687–691. Stafford RS, Singer DE. Recent national patterns of warfarin use in atrial fibrillation. Circulation 1998; 97:1231–1233. Bungard TJ, Ghali WA, Teo KK, et al. Why do patients with atrial fibrillation not receive warfarin? Arch Intern Med 2000; 160:41–46. 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; 323:1505–1511. Brass LM, Krumholz HM, Scinto JD, et al. Warfarin use following ischemic stroke among Medicare patients with atrial fibrillation. Arch Intern Med 1998; 158:2093–2100. Mendelson G, Aronow WS. Underutilization of warfarin in older persons with chronic nonvalvular atrial fibrillation at high risk for developing stroke. J Am Geriatr Soc 1998; 46:1423– 1424. White RH, McBurnie MA, Manolio T, et al. Oral anticoagulation in patients with atrial fibrillation: adherence with guidelines in an elderly cohort. Am J Med 1999; 106:165–171. Eckman MH, Levine HJ, Salem DN, et al. Making decisions about antithrombotic therapy in heart disease: decision analytic and cost-effectiveness issues. Chest 1998; 114(suppl):699s–714s. Eriksson UG, Johansson L, Frison L, et al. Single and repeated oral dosing of H 376/95, a prodrug of the direct thrombin inhibitor melagatran, to young healthy male subjects. Blood 1999; 94:26. Eriksson H, Eriksson UG, Frison L, et al. Pharmacokinetics and pharmacodynamics of melagatran, a novel synthetic LMW thrombin inhibitor, in patients with acute DVT. Thromb Haemost 1999; 81:358–363. Eriksson UG, Johansson L, Taure K, et al. Pharmacokinetics and pharmacodynamics of melagatran, the active form of the oral, direct thrombin inhibitor H 376/95, in healthy male subjects. Haemostasis 2000; 30(suppl 1):164. Elg M, Gustafsson D, Carlsson S. Antithrombotic effects and bleeding time of thrombin inhibitors and warfarin in the rat. Thromb Res 1999; 94:187–197. Sherman DG, Dyken ML, Gent M, et al. Antithrombotic therapy for cerebrovascular disorders: an update. Chest 1995; 108(suppl):444s–456s. Fuster V, Gersh BJ, Giuliani ER, et al. The natural history of idiopathic dilated cardiomyopathy. Am J Cardiol 1981; 47:525–531. Pullicino PM, Halperin JL, Thompson JLP. Stroke in patients with heart failure and reduced left ventricular ejection fraction. Neurology 2000; 54:288–294. Mas JL, Arquizan C, Lamy C, Zuber M, et al. Recurrent cerebrovascular events associated with patent foramen ovale, atrial septal aneurysm, or both. N Engl J Med 2001; 345:1740. Halperin JL, Fuster V. Patent foramen ovale and recurrent stroke: Another paradoxical twist. Circulation 2002; 105:2580–2582. Stroke Prevention in Atrial Fibrillation Study Investigators. Stroke Prevention in Atrial Fibrillation Study: Final results. Circulation 1991; 84:527–539. Connolly SJ, Laupacis A, Gent M, Roberts RS, Cairns JA, Joyner C. Canadian Atrial Fibrillation Anticoagulation (CAFA) Study. J Am Coll Cardiol 1991; 18:349–355. Ezekowitz MD, Bridgers SL, James KE, et al. Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. Veterans Affairs Stroke Prevention in Nonrheumatic Atrial Fibrillation Investigators. N Engl J Med 1992; 327:1406–1412. Pengo V, Legnani C, Noventa F, Palareti G. Oral anticoagulant therapy in patients with nonrheumatic atrial fibrillation and risk of bleeding. A Multicenter Inception Cohort Study. Thromb Haemost 2001; 85:418–422.

29 Acute Pulmonary Embolism in the Elderly Paul D. Stein and Fadi Kayali St. Joseph Mercy Oakland, Pontiac, Michigan, U.S.A.

The frequency of acute pulmonary embolism (PE) increases sharply with age [1–6]. In a general hospital, PE was diagnosed in 0.39% of hospital admissions of patients 70 years of age or older [7].

PREDISPOSING FACTORS IN ELDERLY PATIENTS Age itself appears to be an important predisposing factor, although the illnesses associated with age, in fact, may be the true predisposing factors. Among patients z70 years old, 67% were immobilized before the pulmonary embolism, and surgery preceded the pulmonary embolism in 44% [8]. Malignancy was more frequent among patients z70 years than among younger patients, occurring in 26% [8].

SYNDROMES OF PULMONARY EMBOLISM IN ELDERLY PATIENTS The usual syndromes of pulmonary embolism are (1) pulmonary hemorrhage or infarction syndrome characterized by pleuritic pain or hemoptysis; (2) isolated dyspnea, unaccompanied by hemoptysis, pleuritic pain, or circulatory collapse; and (3) circulatory collapse. The syndrome of pulmonary hemorrhage or pulmonary infarction is the most common syndrome of acute PE among patients in whom a diagnosis is made antemortem [9,10]. These syndromes were observed with comparable frequency among elderly patients and younger patients [8]. However, 11% of patients z70 years of age, in contrast to younger patients, did not show these syndromes [8]. Pulmonary embolism in these patients was suspected on the basis of unexpected radiographic abnormalities, which may have been accompanied by tachypnea or a history of thrombophlebitis. Unexplained radiographic abnormalities in elderly patients may be an important clue to the diagnosis of pulmonary embolism [8,11]. 695

696

Stein and Kayali

Table 1 Symptoms of Acute Pulmonary Embolism in the Elderly z70 years (n = 72) (%) Dyspnea Pleuritic pain Cough Leg swelling Leg pain Palpitation Wheezing Angina-like pain Hemoptysis

78 51 35 35 31 13 10 10 8

SYMPTOMS OF PE IN ELDERLY PATIENTS In considering all elderly patients, including those who may have had prior cardiopulmonary disease, dyspnea, and pleuritic pain were the most frequent symptoms (Table 1). Dyspnea occurred in 78% of elderly patients with PE and pleuritic pain occurred in 51% [8]. Hemoptysis occurred less frequently among patients z70 years than among younger patients [8]. Other symptoms occurred with comparable frequency among all age groups. Among elderly patients with the pulmonary hemorrhage/infarction syndrome who did not have prior cardiopulmonary disease, pleuritic pain was more frequent than hemoptysis (88% versus 12%) [12].

SIGNS OF PE IN ELDERLY PATIENTS Among all patients z70 years of age, regardless of prior cardiopulmonary disease, tachypnea (respiratory rate z20/min) was the most frequent sign of PE. Tachypnea Table 2 Signs of Acute Pulmonary Embolism According to Age z70 years (n = 72) (%) Tachypnea (z20/min) Rales Tachycardia (>100/min) Increased P2 Deep venous thrombosis Diaphoresis Wheezes Temperature >38.5 jC Third heart sound Pleural friction rub Homans’ sign Cyanosis P2= component of second heart sound.

74 65 29 15 15 8 8 7 7 6 4 3

Acute Pulmonary Embolism Table 3

697

Combinations of Signs and Symptoms in Elderly Patients (%)

Dyspnea or tachypnea Dyspnea or tachypnea Dyspnea or tachypnea Dyspnea or tachypnea Dyspnea or tachypnea Dyspnea or tachypnea Dyspnea or tachypnea abnormality

or or or or or or

hemoptysis pleuritic pain signs of deep venous thrombosis pleuritic pain or signs of deep venous thrombosis radiographic atelectasis or parenchymal abnormality pleuritic pain or radiographic atelectasis or parenchymal

92 92 92 94 94 100 100

The addition of hemoptysis did not improve the sensitivity of the combination for the detection of pulmonary embolism. Tachypnea = respiratory rate z20/min.

occurred in 74% and tachycardia (heart rate > 100/min) occurred in 29% of elderly patients with PE [8] (Table 2). All signs occurred with a comparable frequency among all age groups.

COMBINATIONS OF SYMPTOMS AND SIGNS IN ELDERLY PATIENTS Dyspnea or tachypnea or pleuritic pain occurred in 94% of all elderly patients, including those with prior cardiopulmonary disease. Dyspnea or tachypnea or radiographic evidence of atelectasis or a parenchymal abnormality occurred in 100% [8] (Table 3).

THE ELECTROCARDIOGRAM IN ELDERLY PATIENTS WITH PE Among all elderly patients, some of whom had prior cardiopulmonary disease, nonspecific ST-segment or T-wave changes were the most frequent electrocardiographic abnormalities [8]. Either or both occurred in 56% of patients z70 years old [8] (Table 4). With the

Table 4 Electrocardiographic Findings in Elderly Patients with Acute Pulmonary Embolism (n = 57) z70 Years (%) Normal ST segment or T-wave changes Left axis deviation Left ventricular hypertrophy Acute myocardial infarction pattern Low-voltage QRS Complete right boundle branch block Right ventricular hypertrophy Right axis deviation Right atrial enlargement Incomplete right bundle branch block

21 56 18 12 12 9 7 4 2 2 2

698

Stein and Kayali

exception of left anterior hemiblock (left axis deviation) among patients z70 years of age, other electrocardiographic abnormalities occurred in 12% or fewer elderly patients. No difference in the frequency of any electrocardiographic abnormalities occurred among different age groups [8]. Among elderly patients with the pulmonary hemorrhage/infarction syndrome, and no prior cardiopulmonary disease, the electrocardiogram was normal in 62% [12]. If abnormal, the most frequent abnormalities were nonspecific ST-segment or T-wave changes (38%).

BLOOD GASES IN ELDERLY PATIENTS WITH PE The partial pressure of oxygen in arterial blood (PaO2) was lower among elderly patients with PE than among younger patients [8]. The PaO2 among elderly patients with PE, some of whom had prior cardiopulmonary disease, was 61 F 12 mmHg (mean F standard deviation). The alveolar–arterial oxygen difference (gradient) among elderly patients with PE was 47 F 14 mmHg, which was higher than among younger patients. The alveolar–arterial oxygen difference in normal adults increases with age [13–16]. Elderly patients with the pulmonary hemorrhage/infarction syndrome who had no prior cardiopulmonary disease had a higher pulmonary artery mean pressure (25 F 9 mmHg) and lower PaO2 (64 F 10 mmHg) than patients less than 40 years of age [12].

CHEST RADIOGRAPH AMONG ELDERLY PATIENTS WITH PE The chest radiograph was normal in 4% of all elderly patients with PE, including those with prior cardiopulmonary disease [8]. Atelectasis or pulmonary parenchymal abnormalities were the most frequent radiographic abnormalities (Table 5). All radiographic abnormalities occurred with a comparable frequency among all age groups.

Table 5 Chest Radiograph in Acute Pulmonary Embolism Among Elderly Patients (n = 72) (%) Normal Atelectasis or pulmonary parenchymal abnormality Pleural effusion Pleural-based opacity Prominent central pulmonary artery Elevated diaphragm Cardiomegaly Decreased pulmonary vascularity Pulmonary edema Westermark’s sign

4 71 57 42 29 28 22 19 13 7

Westermark’s sign = prominent central pulmonary artery and decreased pulmonary vascularity.

Acute Pulmonary Embolism

699

Among elderly patients with PE and no prior cardiopulmonary disease who had the pulmonary hemorrhage/infarction syndrome, the chest radiograph showed atelectasis or a pulmonary parenchymal abnormality in 82% [12]. The central pulmonary artery appeared dilated in 29%. A normal chest radiograph was uncommon in elderly patients with the pulmonary hemorrhage/infarction syndrome, occurring in 6% [12].

CLINICAL ASSESSMENT IN ELDERLY PATIENTS When physicians were 80 to 100% confident that pulmonary embolism was present in elderly patients on the basis of clinical judgment and simple laboratory tests, they were correct in 90% in a small sample of patients (9 of 10 patients) [8]. When they believed that there was less than 20% likelihood of pulmonary embolism, they correctly excluded the diagnosis in 81% of patients. In most elderly patients, physicians were uncertain of the diagnosis, believing that there was a 20 to 79% chance of pulmonary embolism. The accuracy of clinical assessment was comparable among patients in all age groups [8].

VENTILATION/PERFUSION LUNG SCANS IN ELDERLY PATIENTS The utility of ventilation/perfusion lung scans among patients z70 years old was comparable with that in younger patients [8]. Among patients z70 years of age with ventilation/ perfusion lung scans indicating a high probability of pulmonary embolism, 94% had pulmonary embolism (Table 6). The positive predictive value of all probabilities of ventilation/perfusion lung scans using original PIOPED criteria [17] were comparable in all age groups [8]. The sensitivity of ventilation/perfusion scans interpreted as a high probability of pulmonary embolism among patients z70 years of age was 47% [8]. The sensitivity did not differ significantly among age groups. The specificity of ventilation/perfusion scans interpreted as a high probability of pulmonary embolism among patients z70 years of age was 99%. The specificity was similar among all age groups [8]. Elderly patients with no prior cardiopulmonary disease who had the pulmonary hemorrhage/infarction syndrome tended to show more mismatched perfusion defects than patients under age 40, regardless of whether the defects were large or moderate in size [12]. In patients with no prior cardiopulmonary disease, a high positive predictive value can be achieved with fewer mismatched perfusion defects than are required for a high-

Table 6 Ventilation-Perfusion Scans in Elderly Patients with Acute Pulmonary Embolism (PE) z70 Years

High Intermediate Low Near normal/normal

Number of patients

%

34/36 27/100 10/71 1/8

94 27 14 13

700

Stein and Kayali

probability interpretation in patients with prior cardiopulmonary disease [18,19]. Stratification according to the presence or absence of prior cardiopulmonary disease was particularly useful for the evaluation of V/Q lung scans in patients z70 years, although only 23% of elderly patients had no prior cardiopulmonary disease. Among patients z70 years who had no prior cardiopulmonary disease, z2 mismatched large- or moderate-size perfusion defects showed a sensitivity of 74%, a specificity of 100%, and a positive predictive value of 100% [20].

COMPLICATIONS OF PULMONARY ANGIOGRAPHY AMONG ELDERLY PATIENTS Major complications of pulmonary angiography occurred in 1.0% of 200 patients z70 years [8]. Renal failure, either major or minor, was the most frequent complication of angiography among elderly patients. It occurred in 3% of patients z70 years of age [8]. ‘‘Minor’’ complications of renal failure were important complications, although dialysis was not required. Patients with these complications showed either an elevation of the serum creatinine from previously normal levels to z2.1 mg/100 mL (range 2.1 to 3.5 mg/100 mL) or an increase in a previously abnormal serum creatinine level z2 mg/100 mL. Minor complications of prior angiography included urticaria, pulmonary edema requiring only diuretics, nausea and vomiting, arrhythmias that were not life threatening, hematomas, interstitial staining with contrast material, and narcotic overdose [8]. Minor complications occurred in 7.0% of patients z70 years old.

CONTRAST-ENHANCED SPIRAL COMPUTED TOMOGRAPHY Contrast-enhanced spiral computed tomography (CT) can be particularly useful in elderly patients because it is noninvasive, although the risk of renal toxicity from contrast material remains [8]. Even though contrast-enhanced spiral CT is used throughout the United States, it has not yet been fully evaluated [21]. Individual small case series showed sensitivities that ranged from 50 to 93% in studies that used 5-mm CT sections [22–27]. In the central pulmonary arteries using CT with 5-mm sections, average sensitivity and specificity were higher, both 94% [22,23,28–30], but in subsegmentaal pulmonary arteries, sensitivity was dismal (13%) [22,23]. With a collimation of 3 mm, better results were reported; the sensitivity was 87% and specificity 95% [31]. More recent collaborative trials failed to clarify the validity of spiral CT. The European Collaborative Trials (ESTIPEP and ANTELOPE) showed sensitivities that ranged from 60 to 95%, depending on whether the images were read centrally or locally (local readers did better) [32–34]. It is speculated that the results will be better with newer multidetector scanners that are faster and reduce motion artifact.

GADOLINIUM-ENHANCED MAGNETIC RESONANCE ANGIOGRAPHY Magnetic resonance angiography (MRA) of the pulmonary vasculature has been hampered by a multitude of factors such as respiratory and cardiac motion artifacts, saturation problems, poor signal-to-noise ratio, long acquisition times, and limited spatial resolution

Acute Pulmonary Embolism

701

[35]. Intravenous gadolinium, a paramagnetic agent, gives additional enhancement of the blood signal. The most appealing approach to the application of gadolinium-enhanced MRA is its potential use for showing intraluminal filling defects. The intraluminal filling defect is considered the characteristic that is diagnostic of PE [17,21]. Data on the use of gadoliniumenhanced MRA for the diagnosis of acute PE are sparse. Gadolinium-enhanced MRA for the diagnosis of acute PE had a sensitivity that ranged from 77% to 100% and specificity that ranged from 95% to 98% [36–38]. With the present state of the art, it is likely that MRA would be most useful in patients with a strong suspicion of PE, in whom results of less expensive tests are equivocal and radiographic contrast material or ionizing radiation are contraindicated.

STRATEGIES OF DIAGNOSIS Serial noninvasive leg tests accompanied by ventilation/perfusion lung scans may permit a diagnosis without the need for pulmonary angiography [39]. Although elderly patients have been shown to tolerate pulmonary angiography, it is sensible to avoid the procedure if possible. The risk of fatal PE has been shown to be small if serial noninvasive leg tests show no deep venous thrombosis [40,41]. Conversely, if deep venous thrombosis is present, the treatment with anticoagulants is the same as with PE [42]. Unfortunately, many insurance programs will not pay for serial noninvasive leg tests, and often patients cannot arrange for serial tests. Alternative diagnostic strategies have been introduced for the exclusion of PE. If the clinical assessment for PE was low probability and the D-dimer was also low, PE was excluded in 99.8% [43]. If the clinical assessment was low, the ventilation/perfusion lung scan was nondiagnostic, and a single leg test was negative, PE was excluded in 98% [43].

ANTICOAGULANT THERAPY Recommendations for prophylaxis and antithrombotic therapy for PE were made in the Sixth American College of Chest Physicians (ACCP) Consensus Conference on Antithrombotic Therapy [42–44]. The recommendations for treatment are shown in Table 7. Several studies found that the frequency of bleeding during warfarin therapy is higher in older patients, although this was not observed by some [45]. In one study, the relative risk for major bleeding was 3.2 for patients aged 65 or older [46]. The risk of major hemorrhagic complications among patients of all ages with thromboembolic disease, treated with ‘‘less intense’’ warfarin, International Normalized Ratio (INR) = 2.0 to 3.0, based on pooled data, was 1.7% [47]. Older patients also have a higher risk of bleeding from heparin than younger patients [48,49]. The frequency of major hemorrhagic complications from heparin among patients of all ages treated for thromboembolic disease, was 4.9%, based on pooled data [47]. The frequency of major bleeding from heparin in high-risk patients (surgery within previous 14 days, history of peptic ulcer disease, gastrointestinal or genitourinary tract bleeding, or platelet count 70% (NASCET) [24–26]. For the patient with significant carotid stenosis without symptoms, selection must be somewhat more restricted. After several other studies failed to conclusively prove the benefits of surgery, ACAS established that, for asymptomatic patients with stenosis >60%, surgery reduces the risk of ipsilateral stroke provided that operative morbidity and mortality be held to 70 years) in itself was correlated with an increased incidence of postoperative MI for a wide variety of major surgical procedure [34,35]. The benefit of surgery for the elderly patient with asymptomatic carotid stenosis having significant concomitant disease imparts a major judgmental challenge to surgical decision making. Perioperative morbidity and mortality should be kept to 70 year old; >75 year old; >80 year old, respectively), there is a statistically significant ( p 75-year-old group (Table 2). No increase in risk for the elderly was found when 70 or 80 years was used as the cutoff. Interestingly, there was a statistically significant increased morbidity and combined morbidity and mortality for younger patients with a cutoff of 70 years. It should be noted that in two of the studies, each progressively older age bracket had a correspondingly higher operative morbidity, mortality, and combined morbidity [46,50]. The failure of individual studies to find a difference in mortality and morbidity rates between younger and older patients can be attributed to various factors. Several studies report results from referral centers with uniform surgical technique and limited patient selection by a small number of surgeons with great familiarity in performing CEA. An additional factor is the operation itself; patients undergoing CEA have relatively little postoperative pain and therefore less respiratory impairment and pulmonary complication [42].

1985 1985 1986 1986 1988 1988 1989

1990 1990 1991

1991 1992 1994 1996

Courbier Plecha Ouriel Rosenthal Schultz Loftus Fisher

Schroe Pinkerton Meyer

Roques Treiman Coyle Perler

75 75 75 80 80 70 65 70 75 80 70 75 70 75 80 75 80 80 75 992 124

483 560

5220 393 1008

Cases (young) 76 782 77 90 116 53 2089 1356 685 212 222 125 749 265 56 81 183 79 63

Cases (old)

26 (2.6) 2 (1.6)

20 (4.1) 2 (0.4)

94 (1.8) 12 (3.1) 20 (2.0)

Morbidity (young)

7 0 23 10 3 2 3 0 3

2 17 3 4 1 1 35

(3.2) (0) (3.1) (3.8) (5.4) (2.5) (1.6) (0) (4.8)

(2.6) (2.2) (3.9) (4.4) (0.9) (1.9) (1.7)

Morbidity (old)

Note: Morbidity indicates postoperative ipsilateral stroke. Percentages are showin in parentheses. Young and Old refer to patients below and above the particular age cutoff for each study.

Year

Author

Age cutoff

16 (1.6) 2 (1.6)

8 (1.7) 5 (0.9)

77 (1.5) 2 (0.5) 6 (0.6)

Mortality (young) 1 (1.3) 18 (2.3) 0 (0) 2 (2.2) 2 (1.7) 0 (0) 52 (2.5) 44 (3.2) 25 (3.6) 10 (4.7) 3 (1.4) 1 (0.8) 10 (1.3) 4 (1.5) 0 (0) 3 (3.7) 3 (1.6) 1 (1.3) 1 (1.6)

Mortality (old)

42 (4.2) 4 (3.2)

28 (5.8) 7 (1.3)

171 (3.3) 14 (3.6) 26 (2.6)

Combined M+M (young)

Table 1 Summary of Carotid Endarterectomy Morbidity and Mortality Rates in Young Versus Old Patients

10 1 33 14 3 5 6 1 4

3 35 3 6 3 1 87

(4.5) (0.8) (4.4) (5.3) (5.4) (6.2) (3.3) (1.3) (6.3)

(3.9) (4.5) (3.9) (6.7) (2.6) (1.9) (4.2)

Combined M+M (old)

No No

No No

No No No

Statistically significant difference?

Peripheral Vascular Disease 715

1 5 7 6

Cutoff

>65 >70 >75 >80 790 6297 2412

Total no. of young patients 2089 1368 1469 543

Total no. of old patients 52 (6.6) 110 (1.7) 82 (3.4)

Morbidity (young) 35 44 37 12

(1.7) (3.2) (2.5) (2.2)

Morbidity (old) 16 (2) 86 (1.4) 34 (1.4)

Mortality (young)

52 25 28 8

(2.5) (1.8) (1.9) (1.5)

Mortality (old)

68 (8.6) 196 (3.1) 116 (4.8)

Combined M+M (young)

87 69 65 20

(4.2) (5) (4.4) (3.6)

Combined M+M (old)

Note: Number of compiled studies totals to >16 because several studies listed results by more than one age cutoff. Pairs of italicized figures indicate statistically significant difference ( p < .05).

No. of studies compiled

Table 2 Carotid Endarterectomy Morbidity and Mortality: Studies Grouped by Age Cutoff

716 Fujitani et al.

Peripheral Vascular Disease

717

Generally, individual reports chronologically show a significant decrease in mortality and morbidity rates over the years in which the studies were conducted. However, the overall morbidity and mortality has changed very little over time (Table 3). Age >65 years was once considered a high-risk threshold for performing CEA. Over the last three decades, the age limit for CEA has been extended to progressively older groups. Literature review suggests that age itself should not be the sole discriminating factor in weighing the value of CEA for clear-cut symptoms or high-grade asymptomatic stenosis. Recently, a statewide analysis of 9918 Maryland patients concluded that CEA, even among the very elderly, is a safe procedure, although the length of stay may be longer due to the concomitant medical illness [53]. There has been recent interest in endovascular stenting of carotid artery lesions in lieu of standard operative CEA. This new development is still unproven and careful prospective studies are currently in progress. Recommendations for carotid stenting will depend upon the outcomes of these randomized trials and are forthcoming. It is clear that elderly patients may benefit from CEA when performed for the proper indications. However, careful assessment of their associated comorbidities is most important to assure optimal long-term benefit from surgical intervention. If these caveats are respected, prevention of ischemic or embolic stroke can greatly enhance the quality of life for the geriatric patient. Vertebrobasilar Insufficiency The most common symptoms of vertebrobasilar insufficiency include nausea, vertigo, ipsilateral facial numbness, ipsilateral Horner’s syndrome, and limb ataxia. While these ischemic conditions may be very mild and often transient, there can be progression to actual posterior fossa infarction, which can be lethal due to extensive edema and midbrain compression. Although microembolization can contribute to posterior cerebral and cerebellar ischemic compromise, occlusive disease of the vertebral arteries or the basilar artery is the most common cause of this process. Thrombosis may occur in the basilar artery proper or the basilar branch vessels that penetrate into the brain stem. Subclavian steal syndrome is a classic vertebrobasilar symptom complex associated with proximal subclavian or innominate artery stenosis. Since the vertebral arteries originate from the subclavian arteries, they can function as collaterals to the upper extremities. With a proximal stenosis, flow is reversed in the vertebral artery following ipsilateral arm exertion, decreasing blood flow and perfusion pressure through the basilar arterial system. This results in posterior cerebral and cerebellar ischemic symptoms. This is exacerbated in the presence of associated carotid occlusive disease. The left subclavian artery is more commonly involved, occurring in 65% of cases. The diagnosis of subclavian steal syndrome

Table 3

Carotid Endarterectomy Morbidity and Mortality by Year of Study Publication

Five-year period

Number of studies compiled

Total cases

Morbidity

Mortality

Morbidity + Mortality

1986–1990 1991–1995 1996–

7 4 1

5216 2084 187

105 (2) 54 (2.6) 5 (2.7)

81 (1.6) 33 (1.6) 3 (1.6)

186 (3.6) 87 (4.2) 8 (4.3)

718

Fujitani et al.

is supported by complaints of intermittent vertigo, light-headedness, nausea, diplopia, and vomiting exacerbated by arm exercise. Findings of a supraclavicular bruit and a significant (>40 mmHg) blood pressure discrepancy between arms on physical examination should prompt further evaluation. The differential diagnosis in geriatric patients includes inner ear disorders and chronic subdural hematomas that may occur even after fairly trivial trauma. Symptomatic patients with vertebrobasilar occlusive disease should be considered for elective operative intervention. Procedures to treat this condition include proximal vertebral artery endarterectomy, carotid-subclavian bypass, or vertebral artery transposition to restore adequate antegrade vertebral flow [54]. Abdominal Aortic Aneurysms The natural history of aortic aneurysms is that of progressive enlargement, usually without accompanying symptoms. Left undetected and untreated, it may lead to eventual rupture, often resulting in death. Early detection with appropriate screening of high-risk patients and elective operative repair of these aneurysms are the key to optimal management of this common vascular disease. Unlike some other atherosclerotic-related disease processes, the incidence of abdominal aortic aneurysms (AAAs) is increasing [55]. This is due, in part, to increased utilization of diagnostic tests that detect smaller asymptomatic aneurysms, but it also reflects the aging of the population. The incidence of aneurysms increases dramatically after the age of 55 years in men and after 70 years in women [56]. As a consequence of this chronologically staggered presentation, aneurysmal disease of the aorta has a substantial male predominance by a ratio of about 3:1. It is estimated that 3 to 7% of men over age 70 have an AAA and approximately 100,000 abdominal aortic aneurysms will be diagnosed in the United States this year [57]. Approximately 15,000 deaths are attributable to AAAs every year in spite of continued efforts at early detection. The risk of rupture is clearly related to size, although the correlation is not perfect. That is, when considering a large group of patients, aneurysms greater than 6 cm in diameter are more likely to rupture (about 30% in 3 years) than smaller aneurysms (about 10% in 3 years). It has been well documented, however, that aneurysms as small as 4 cm do rupture [58]. More importantly, documenting stability in size over time does not necessarily predict a continued benign course [59,60]. The prognosis for a ruptured abdominal aortic aneurysm is very poor. As many as one-half of patients with ruptured AAAs die before operative management can be rendered. Although great progress has been made in surgical technique, anesthesia, and postoperative care, the operative mortality rate for patients with ruptured aneurysms reaching the operating room still remains around 50% in most series [61–63]. Conversely, over the past three decades there has been a progressive improvement in the perioperative mortality rate associated with elective aortic abdominal aneurysmorrhaphy [64]. Mortality rates of 1 or 2% are commonly reported in the literature, although the average national mortality rate is probably closer to 5% [65–66]. It is therefore clear that if the mortality rate attributable to abdominal aortic aneurysms is to be reduced, the disease must be recognized in its asymptomatic state and treated before rupture occurs. Diagnosis depends on careful physical examination and high index of suspicion for the geriatric patient. In nonobese patients, deep palpation of the abdomen between the xiphoid and umbilicus can usually outline the abdominal aorta and provide an estimate of its size. This gross measurement tends to overstate the actual diameter of the aneurysm by 1 to 2 cm. In a very thin patient, even the normal abdominal pulsation may be mistaken for an

Peripheral Vascular Disease

719

aneurysm. Conversely, large aneurysms may be missed in obese patients due to difficulty in manual palpation. During examination, attention should also be directed to the peripheral vasculature including the femoral, popliteal, dorsal pedal, and posterior tibial pulses. Ischemic changes in the foot (‘‘blue-toe syndrome’’) or diminished pulses may suggest distal atheroembolism from the aneurysm. Furthermore, concomitant peripheral aneurysms of the femoral and popliteal arteries may be found in 5 to 8% of patients with AAAs. Ultrasound imaging is an accurate method of diagnosing an abdominal aortic aneurysm. This simple, noninvasive procedure can be tolerated by virtually all patients and is the best screening test for the disease. The biplanar cross-sectional visualization of the aorta allows for determinations of the size and extent of the aneurysm, as well as the detection of intraluminal thrombus and associated iliac aneurysms. Ultrasound determination of AAA diameter is readily reproducible and allows longitudinal assessments of changes in aneurysm size. This is especially useful when following a patient with a ‘‘small’’ (3.5 with the presence of renal artery stenoses of >60% diameter reduction. This study reported an overall sensitivity of 91% and specificity of 95%. Others have confirmed similar results [109]. The B-mode imaging also allows anatomical definition of the stenoses, especially whether the disease is localized to the orifice or the more distal vessel. Duplex scanning is particularly useful in postprocedural follow-up of lesions treated by surgical revascularization or angioplasty. Recurrent stenoses or thromboses can be identified with an accuracy of 93% when compared to arteriography [110]. Arteriography, however, is still the accepted standard for diagnosis of anatomical renovascular disease. Multiplanar views of the renal vessels, including anteroposterior and oblique views, are routinely obtained (Figure 6). Newer digital subtraction techniques allow the utilization of smaller volumes of dye, lessening the chance for contract-mediated renal dysfunction following the necessary pararenal or selective injections. Traditionally, the measurement of renal vein renin levels has been employed as an adjunct to arteriography to help identify functionally significant renal arterial stenoses [111].

Figure 6 Arteriogram demonstrating severe orificial renal artery stenosis.

Peripheral Vascular Disease

729

Such values can be expressed as renal vein renin ratios, which compare one kidney to the other (ratios of >1.5 are considered diagnostic), or renal/systemic renin indices, which quantify the contribution of each kidney to the total systemic renin level. Renal vein ratios are obviously less accurate in bilateral disease. In contrast, renal/systemic indices assess bilateral disease more precisely and have been a more reliable predictor of the response to revascularization [112]. The most favorable situation is unilateral disease in which renin secretion is suppressed in the contralateral kidney. A more aggressive approach in the treatment of renovascular disease has resulted from three recent distinct developments: (1) increased recognition of ischemic nephropathy as an important consequence of renovascular disease separate from hypertension; (2) widespread acceptance of percutaneous renal angioplasty; and (3) improved results of revascularization in elderly patients. In recent reports, substantial improvements in renal function have been noted even in patients over 65 years of age with significant and progressive azotemia (serum creatinine >4 mg%) [113]. Currently accepted indications for renal revascularization include the presence of a hemodynamically significant renal artery stenosis (>60% diameter) in patients with refractory hypertension or hypertension associated with renal functional impairment. Percutaneous renal angioplasty is an attractive option in high-risk elderly patients whose medical risks would otherwise prohibit operative revascularization [114]. However, it is optimally performed in midsegment renal artery stenoses and is not as well suited for treatment of the most prevalent orificial lesions. In appropriately selected stenoses, shortterm patency rates with angioplasty are approximately 90% [115]. The utility of balloonexpandable and self-expanding stents have broadened the number of renal artery lesions amenable to these catheter-based endovascular procedures. Operative options include endarterectomy and aortorenal or hepatorenal bypasses. Endarterectomy through a transaortic route is most useful in bilateral orificial disease but requires suprarenal aortic exposure and clamping with resultant higher cardiac risk. Aortorenal bypass from a relatively disease-free area of the infrarenal aorta has been the most widely used technique in the past, although extra-anatomical hepatorenal and splenorenal bypasses are commonly employed. The latter procedures avoid manipulation of the often-diseased aorta and obviate any need for aortic occlusion. Regardless of the origin of these bypass grafts, long-term patency is excellent, approaching 95% over 5 years if autogenous vein is used. In one recent series of 35 patients over 60 years of age, improvement or cure of hypertension was reported in 91% of patients and improvement in renal function in 37% [116]. Overall morbidity and mortality in this group (5.4% mortality, 23% perioperative morbidity) appeared to be related to generalized atherosclerosis and accompanying risk factors rather than age per se. The argument for aggressive treatment of renovascular disease in the geriatric patient population is further supported by the experience of the physicians at the Cleveland Clinic, who compared medical to surgical management in 50 patients over 3 years [117]. While this was not a randomized study, they noted better control of hypertension (90% improvement in the surgical group versus 66% improvement in the medical group) as well as improvement or stabilization of renal function in those treated surgically (91% surgical versus 50% medical group). Acute Arterial Thrombosis Sudden arterial thrombosis may result from (1) in situ thrombosis of preexisting atheroocclusive disease; (2) arterial–arterial embolization; (3) thrombosis of an aneurysm; and (4)

730

Fujitani et al.

vascular trauma. Obtaining a careful medical history can be very helpful in eliciting the underlying etiology. Patients with in situ thrombosis often describe a history of preexisting claudication or rest pain that has recently worsened. In contrast, arterial embolization, vascular trauma, and aneurysm thromboses usually are not associated with any previous vascular symptoms. When the occlusion is very sudden and is not associated with prior development of collateral arteries, the symptoms tend to be much more severe. Classically, the presentation of acute limb-threatening ischemia includes the six ‘‘Ps.’’ These include pallor, pain, paresthesias, paralysis, pulselessness, and poikilothermia (change in skin temperature). The level of coolness of the extremity is usually limited to the portion of the limb well beyond the occlusion. For example, with an acute femoral artery occlusion, the level of ischemia may not extend above the knee due to recruitment of collateral flow to the thigh. On the other hand, internal organs, especially the intestines, kidneys, and brain are especially prone to early loss of function and infarction with sudden arterial occlusion due to the end-artery nature of their blood supply. With acute occlusions resulting from vascular graft thrombosis or emboli to native vessels, flow through collaterals may sometimes be inadequate to preserve limb viability. The heart is the most common source of emboli, which also originate from plaques and aneurysms in the thoracic and abdominal aorta. Bilateral extremity involvement is common when the origin is proximal; unilateral involvement is the rule with embolism from the iliac or more distal vessels. Atherosclerosis is not the only cause of thombosis. Trauma, collagen vascular disorders, myeloproliferative diseases, and dysproteinemias are other causes for sudden arterial thrombosis. Acute arterial occlusion, regardless of the end organ, requires prompt recognition and treatment. The long nerves of the lower extremity are sensitive to ischemia and weakness, paralysis, or paresthesias, which indicate the presence of acute limb threat requiring urgent intervention correlating well with the degree of neurological impairment. Without timely revascularization, initial numbness may be rapidly followed by ischemic muscle necrosis and skin infarction. Irrevocable tissue loss may occur within 6 h. Often, however, even when rest pain is present, collateral flow is sufficient to preserve the limb while diagnostic studies are obtained prior to operative intervention. Complete paralysis, muscle rigidity, and thrombosis of cutaneous vessels are markers of a high probability of limb loss even with successful revascularization. Delays can result in severe muscle ischemia and amputation. Patients with a sudden onset of acute limb threat often have embolized clots from a cardiac or peripheral source into the infrainguinal arteries. In many instances, the thrombosis of a previously patent bypass graft creates an identical clinical picture. Surgical management entails operative exposure of one or more of the vessels occluded, fashioning an incision (arteriotomy) into the artery, and extraction of the offending thrombus via the arteriotomy. Embolectomy catheters specifically designed for these procedures are passed via the arteriotomy into the proximal and distal arterial segments. A tip-mounted balloon is inflated and traps the thrombus, leading to its extraction as the catheter is gently dragged out of the artery. Repeated passes of the thrombectomy catheter will usually result in satisfactory removal of clot and restoration of blood flow. Adjunctive therapy in patients requiring urgent thrombectomy includes the use of systemic heparin (75 to 100 U/kg bolus, followed by a continuous infusion). Diagnostic angiography before surgery is helpful in planning the operation, and intraoperative contrast angiography can be used to position thrombectomy catheters over guidewires and to confirm the restoration of blood flow. Arteriography is especially helpful in cases of sudden thrombosis of preexisting disease. However, arteriography is not essential in the diagnosis

Peripheral Vascular Disease

731

and treatment of obvious embolic occlusions, especially if thrombolysis is not an option due to advanced ischemia or other contraindications. Thrombolytic drugs may be administered as part of the operation to dissolve clots that are either incompletely removed by embolectomy catheters or lodged in tibial or pedal vessels beyond their reach. The combination of fluoroscopic guidance of embolectomy catheters and the use of intraoperative thrombolytic drugs both improves the overall success of operative interventions and allows distal thrombi to be adequately managed via femoral and popliteal incisions [118,119]. The use of proximal incisions is an advantage because they avoid arteriotomies of more distal vessels with the risk of intimal hyperplasia and late thrombosis. An additional concern is that incisions below the knee may compromise subsequent amputation stump healing if revascularization is unsuccessful. The overall immediate success rate with surgical thrombectomy and embolectomy for limb salvage in the face of acute occlusion is good, on the order of 70 to 90%. Patient mortality after embolic episodes, however, is strikingly high; compared to elective bypass, the reported mortality ranges from 4 to 39%, with an average rate of about 20% in recent series [120]. Some of the factors responsible for the high mortality are the metabolic disturbances that accompany acute arterial occlusion of the lower extremity including hyperkalemia, acidosis, and renal failure secondary to rhabdomyolysis. Some of these disturbances are a consequence of cellular necrosis due to ischemia and may be exacerbated by the reperfusion injury that often attends successful revascularization after severe ischemia. Delay in operative treatment of limb-threatening ischemia correlates with poor outcomes. In one series, treatment instituted within 12 h of symptom onset led to a limb salvage rate of 93% and 12% mortality. After 12 h, limb salvage fell to 78% and mortality increased to 31% [121]. Another factor in the high mortality associated with embolism is the concomitant presence of severe cardiac disease and the frequency of other comorbid conditions. Increased mortality is associated with advanced age; in one series, those younger than 70 years old were found to have 7% mortality compared to 22% for those older than age 70 [120]. Many elderly patients may have already undergone previous vascular reconstructions, which are susceptible to failure. Occluded bypass grafts may be treated by thrombectomy using techniques similar to those employed with primary thromboembolism or pharmacological thrombolysis. Both techniques are used to restore antegrade flow, and the results are good if attendant technical defects restricting flow (most commonly stricture of the distal anastomosis or stenosis in the native artery just beyond) are also corrected. After initial treatment of embolic occlusions, long-term postoperative anticoagulation must be strongly considered. Without anticoagulation, emboli may recur in about one-third of patients within 30 days. With the use of heparin and warfarin postoperatively, the recurrence rate decreases to less than 10%. Unfortunately, even with the best treatment, the mortality rate for acute arterial occlusion remains relatively high. This can be attributed to the often advanced age of these patients suffering from either in situ thrombosis or embolization together with the severity of associated myocardial disease. The most common cause of death in these patients is myocardial infarction and pulmonary embolization.

LOWER EXTREMITY ATHERO-OCCLUSIVE DISEASE Moderate peripheral arterial disease (PAD) may induce pain or weakness with walking, which resolves rapidly with rest. This symptom complex is known as intermittent claudi-

732

Fujitani et al.

cation. Lower extremity blood flow at rest is approximately 500 mL/min but can increase tenfold to 5 L/min with maximum exercise stress. Resting skeletal muscle perfusion is 2 mL/ 100 g/min but rises to 70 mL/100 g/min with maximum exercise—a 35-fold increase. As muscle tends to be more sensitive to the imbalance between blood flow requirements and what the diseased arterial system can deliver, most patients experience symptoms in muscle groups, with the calf being the most common site involved, but buttock, hip, foot, thigh, or even the inferior back muscles can be affected. As the imbalance between the demand and actual perfusion is transient, claudication is considered a functional impairment. When flow cannot meet the needs of resting tissue metabolism, evidenced by pain at rest or tissue loss, the lower extremity ischemia is considered critical. Rest pain from critical ischemia is usually experienced in the toes or foot. Patients may dangle their legs from bed or stand to relieve symptoms. Without intervention, rest pain may progress to ulceration or gangrene. Chronic arterial insufficiency ulcers tend to be bland and lack intense inflammation or infection. They commonly develop at the ankle, heel, or leg. Mummified, dry, black toes or devitalized soft tissue covered by a crust represents gangrene from ischemic infarction. With time, suppuration often develops, and ‘‘dry’’ changes to ‘‘wet’’ gangrene. Recent epidemiology studies indicate that only one-half of patients with documented PAD (ABI 120 mmHg and clinical evidence of impending or actual end-organ damage [22], poses a definite risk of MI and cerebrovascular accident. Several precipitants of hypertensive crises have been identified [22], including pheochromocytoma, abrupt clonidine withdrawal prior to surgery, and the use of chronic monoamine oxidase inhibitors with or without sympathomimetic drugs in combination [23]. Chronic HTN may indirectly predispose patients to perioperative myocardial ischemia since CAD is more prevalent in these patients. Even in the absence of CAD, patients with chronic HTN may have episodes of myocardial ischemia [24], perhaps due to impaired coronary vasodilator reserve and autoregulation [25], such that higher arterial pressures are required to maintain adequate perfusion of vital organs. The Study of Perioperative Ischemia Research Group showed that a history of HTN was an independent predictor of postoperative ischemia and increased postoperative mortality [16,26]. Patients with a history of HTN had almost twice the risk of developing

768

Falcone et al.

postoperative myocardial ischemia [16] and almost four times the risk of postoperative death [26] than did patients without HTN in the first 48 h postoperatively. The link between systemic HTN and perioperative cardiac complications may relate to an increased risk of silent myocardial ischemia, which has been shown to be a major predictor of postoperative cardiac morbidity [27,28]. In patients who underwent ambulatory ECG ST-segment monitoring to determine the incidence of silent myocardial ischemia before elective noncardiac surgery, at least one episode of ST-segment depression consistent with silent ischemia occurred in 20% of patients [29]. Patients with HTN despite antihypertensive therapy were at particularly high risk, with more than a 50% incidence of silent myocardial ischemia. However, continuous ECG ST-segment monitoring may overestimate the incidence of myocardial ischemia in the setting of left ventricular hypertrophy [29,30]. In patients with significant diastolic HTN (diastolic blood pressure >110 mmHg,) the potential benefits of delaying surgery in order to optimize antihypertensive medications should be weighed against the risk of delaying the surgical procedure [31,32]. Isolated systolic HTN (systolic blood pressure greater than 160 mmHg and diastolic blood pressure less than 90 mmHg) has been identified as a risk factor for cardiovascular complications in the general population, and treatment has been shown to reduce the future risk of stroke [33]. However, there has been only one study to directly assess the relationship between cardiovascular disease and preoperative isolated systolic HTN. In a multicenter study of patients undergoing coronary artery bypass grafting (CABG), the presence of isolated systolic HTN was associated with a 30% increased incidence of cardiovascular complications [34]. In summary, noncardiac surgery should not be postponed or canceled in the otherwise uncomplicated patient with mild-to-moderate HTN [9]. Antihypertensive medications should be continued perioperatively and blood pressure should be maintained near preoperative levels to reduce the risk of myocardial ischemia [9,25]. More severe or poorly controlled HTN (systolic blood pressure z180 mmHg or diastolic blood pressure z110 mmHg), should be controlled before any elective noncardiac surgery [10]. Continuation of antihypertensive treatment throughout the perioperative period is indicated.

Physical Examination A complete preoperative physical examination is necessary for every patient undergoing noncardiac surgery. Blood pressure and heart rate should be determined in both the supine and standing positions to assess intravascular volume status or autonomic dysfunction. Careful cardiac auscultation should be performed to detect clinically important cardiac findings, including the presence of an S3 gallop suggestive of CHF and murmurs suggestive of significant valular disease, particularly aortic stenosis. The pulmonary exam and evaluation for jugular venous distention and lower extremity edema can also help determine intravascular volume status and the presence of CHF. Peripheral arterial pulses and bruits may suggest cardiac valvular disease or the presence of occult atherosclerotic disease.

ESTIMATION OF PERIOPERATIVE CARDIAC RISK Identifying Surgery-Specific Risk The type of surgical procedure itself has a significant impact on perioperative cardiac risk. Knowledge of the urgency, type, and anticipated duration of the surgical procedure and the

Perioperative Cardiovascular Evaluation

769

expected blood loss and intravascular volume shifts is necessary to make perioperative management plans. Three classifications of surgical risk have been suggested by the ACC/AHA Task Force (Table 1) [10]. Surgical procedures which are not emergent and are not associated with significant blood loss are not associated with a high risk of cardiac ischemia. For example, cataract surgery is associated with minimal cardiac stress and exceedingly low cardiac morbidity and mortality rates, even after a recent MI [35]. Similarly, outpatient procedures have also been shown to be associated with a low incidence of cardiac morbidity and mortality [36]. In such patients, changes in perioperative management are rarely needed unless the patient demonstrates unstable angina or signs or symptoms of CHF. On the other hand, intra-abdominal, orthopedic, and intrathoracic procedures are associated with an intermediate cardiac risk. In this group, the duration of surgery and extent of fluid shifts may modify the need for further cardiac evaluation. In contrast, the patient undergoing a noncardiac surgical procedure associated with a high risk of cardiac morbidity and mortality may benefit from a more extensive preoperative cardiac evaluation. For example, peripheral vascular surgery is associated with a high risk of cardiac morbidity [37]. Since further determination of cardiac status may alter perioperative care, the benefit of further evaluation and treatment would be expected to be greater than the associated costs or risks of such testing. Identifying Clinical Risk Predictors Several approaches can be used to estimate perioperative cardiac risk in patients before noncardiac surgery. In early studies, clinical parameters associated with CAD were proven predictors of perioperative morbidity and mortality. In a landmark study in 1977 by Goldman et al. [3], there were nine independent clinical correlates of postoperative cardiac complications in patients undergoing noncardiac surgery. A point system was assigned to each Table 1 Cardiac Riska Stratification for Noncardiac Surgical Procedures High

Intermediate

Lowb

a

Reported cardiac risk often >5% Emergent major operations, particularly in the elderly Aortic and other major vascular Peripheral vascular Anticipated prolonged surgical procedures associated with large fluid shifts and/or blood loss Reported cardiac risk generally 10 METs

Abbreviations: MET-metabolic equivalent. Source: Adapted from the Duke Activity Status Index and AHA Exercise Standards. Reproduced with permission from Ref. 10.

Figure 1 The American Heart Association/American College of Cardiology Task Force on Perioperative Evaluation of Cardiac Patients undergoing Noncardiac Surgery has proposed an algorithm for decisions regarding the need for further evaluation. This represents one of multiple algorithms proposed in the literature. It is based upon expert opinion and incorporates six steps. First, the clinician must evaluate the urgency of the surgery and the appropriateness of a formal preoperative assessment. Next, he or she must determine whether the patient has had a previous revascularization procedure or coronary evaluation. Those patients with unstable coronary syndromes should be identified, and appropriate treatment should be instituted. The decision to have further testing depends on the interaction of the clinical risk factors, surgery-specific risk, and functional capacity. (Reproduced with permission from Ref. 10.)

Perioperative Cardiovascular Evaluation

773

inversely related to the number of blocks that could be walked or flights of stairs that could be climbed.

NONINVASIVE TESTING BEFORE NONCARDIAC SURGERY General Comments The approach to the preoperative evaluation of an elderly patient for noncardiac surgery includes consideration of noninvasive cardiac stress testing. The sensitivity, specificity, and accuracy of the stress-testing modality and its cost must be considered along with the prevalence of CAD in the population. A positive stress test for myocardial ischemia may not have the same meaning in a population with a low prevalence of CAD as in a population with a high prevalence of CAD. The optimal approach for cardiac risk stratification for patients undergoing noncardiac surgery remains controversial. Several algorithms exist for approaching cardiac risk assessment. Some studies raise questions about the usefulness of preoperative noninvasive stress testing before noncardiac surgery, while other studies consider it useful in specific populations [13,41–43]. In general, the negative predictive values of the specific stress testing modalities studied in these patient populations are high. Therefore, in a patient with a negative stress test for myocardial ischemia, the risk of a perioperative cardiac event is relatively low. On the other hand, the positive predictive values of available stress tests are consistently low so that a patient with evidence of stress-induced myocardial ischemia often still has a good perioperative outcome. This may be explained by the fact that an acute MI is often caused by the rupture of a non-flow-limiting, unstable coronary artery plaque [44]. Stress testing detects flow-limiting coronary artery stenoses (>50 to 70% arterial lumen narrowing), but cannot detect stenoses that are non–flow limiting. Approach to the Patient Importantly, no preoperative cardiovascular testing should be performed if the results will not change perioperative management. The algorithm to determine the need for cardiac stress testing proposed by The ACC/AHA Task Force is based upon observational or retrospective studies and expert opinion before the decision to proceed to further cardiac testing is made [9,10] (Fig. 1). An alternative approach was developed by the American College of Physicians and attempts to apply an evidence-based approach for the determination of preoperative cardiac risk [45]. The initial decision point is the assessment of risk using the Detsky modification of the Cardiac Risk Index [4] and other clinical factors [13,46].

NONINVASIVE PREOPERATIVE CARDIAC TESTING Although cardiac risk for noncardiac surgery is increased with preexisting CAD, the indications for cardiac testing to evaluate the presence of CAD are not based on adequately contolled or randomized clinical trials. Thus, the optimal strategy for testing is based on a conservative approach where the use of testing and treatment is best done only if the results will change preoperative management of the patient. The indications for resting and ambulatory ECG, evaluation of systolic and diastolic left ventricular function, and, finally, preoperative cardiac stress testing are discussed below.

774

Falcone et al.

Resting Electrocardiography All patients undergoing preoperative evaluation for noncardiac surgery should have a resting ECG in order to identify preexisting cardiac disease. It has been estimated that up to 30% of MIs are silent and only detected on routine ECG, especially in diabetics and patients with HTN. The baseline ECG has limited applicability in the assessment of myocardial ischemia but may be useful to detect significant conduction disturbances and arrhythmias. Electrocardiographic abnormalities may not always lead to delay of a noncardiac surgical procedure but may lead to increased vigilance by the anesthesiologist intraoperatively. High-grade atrioventricular block, symptomatic ventricular arrhythmias in the presence of underlying heart disease, and supraventricular arrhythmias with uncontrolled ventricular rate are considered high-risk clinical predictors by the ACC/AHA Guidelines [9]. Pathological Q-waves are considered intermediate-risk clinical predictors. The presence of left ventricular hypertrophy, left bundle branch block, ST-segment abnormalities, and rhythm other than sinus are considered minor risk predictors. Assessment of Left Ventricular Function In general, an assessment of baseline left ventricular function should be reserved for those with unexplained cardiopulmonary symptoms, using echocardiography, radionuclide angiography, and/or contrast ventriculography. Left ventricular ejection fraction has been correlated with short- and long-term prognosis in multiple studies in patients undergoing noncardiac surgery; in general, the lower the ejection fraction, the greater the perioperative risk [9]. Some recent studies have found that left ventricular systolic dysfunction does not predict cardiac complications after vascular surgery [9,47–49]. Importantly, Halm and colleagues were unable to demonstrate that preoperative echocardiographic information added to clinical risk factors for risk stratification [50]. The ACC/AHA Task Force Guidelines recommend that preoperative assessment of left ventricular systolic function prior to noncardiac surgery should be limited to patients with current or poorly controlled CHF, but not as a routine test in patients without prior CHF [9,10]. Among patients with a history of CHF and with dyspneas of unknown etiology, the indications are less clear. Continuous ECG ST-Segment Monitoring Multiple studies have demonstrated the association between major cardiac events and perioperative ST-segment changes detected by continuous ECG monitoring. The duration of perioperative ST-segment changes has been shown to be the strongest predictor of poor outcome [51,52]. Continuous ST-segment monitoring is typically used during the intraoperative period for high-risk patients, although ST-segment changes may not always indicate myocardial ischemia [53,54]. The value of continuous ECG ST-segment monitoring for all patients is unclear. In one study, preoperative myocardial ischemia detected by this method was found to be the most significant independent predictor of adverse postoperative cardiac events [55]. The absence of ischemia was associated with a very high negative predictive value. Cardiac events occurred more often in patients with higher postoperative heart rates or if patients had a history of diabetes, clinical manifestations of CAD, or were older than 70 years of age.

Perioperative Cardiovascular Evaluation

775

Intraoperative ECG ST-segment changes consistent with ischemia, on the other hand, were a significant, but relatively weak, predictor of postoperative events, especially in patients with a low prevalence of CAD. Since it is unclear whether early detection of STsegment changes will lead to improved treatment and patient outcome, the utility of continuous ECG ST-segment monitoring is unclear, and it should not be used routinely in the perioperative period. In fact, overly aggressive treatment of ST-segment changes of nonischemic origin could theoretically increase morbidity, costs, or both.

PREOPERATIVE NONINVASIVE CARDIAC STRESS TESTING Early studies assessed only clinical parameters from a population with a relatively low risk of CAD [3,4]. Applying these clinical indexes to elderly patients undergoing vascular surgery in whom the prevalence of CAD is high and exercise capacity is limited may not be as reliable [49]. The ultimate goal of a preoperative cardiac stress test is to identify patients with myocardial ischemia in whom further cardiac interventions would significantly lower perioperative cardiac risks. In patients undergoing major vascular surgery, which carries the highest perioperative cardiac risk, noninvasive cardiac testing has been found to be helpful for cardiac risk stratification [13,41,43,56–61]. Unfortunately, there are few prospective, randomized studies that establish the value of preoperative testing and that clearly demonstrate that therapy based on these test results affects perioperative outcomes. An important aspect of cardiac stress testing relies on Bayes’ theorem, which states that the correct interpretation of the results of noninvasive stress testing requires estimating the pretest probability of CAD in the patient being studied [62,63]. False-positive stress test results are more common in those in whom CAD is unlikely to be present and false-negative results are more common in those with a high likelihood of CAD. Therefore, determining the probability of CAD in a specific patient will often guide decision making even before a stress test is considered. A limited number of prospective studies have investigated the predictive value of noninvasive cardiac stress tests in determining the risk of postoperative cardiac events. The positive predictive values of all stress testing modalities are poor (10 to 20%), so that a patient with evidence of myocardial ischemia often will not have a postoperative cardiac event even though the estimated cardiac risk is high. On the other hand, negative predictive values are high (95% to 100%), so that patients without evidence of ischemia are at the lowest risk for an adverse perioperative outcome. The likelihood of an adverse cardiac event after noncardiac surgery, even in patients with evidence of CAD, is less than 10% [64,65]. Stress testing should generally be limited to patients in whom the risk of significant CAD is high and to those undergoing the highest risk surgical procedures. The ACC/AHA Task Force Guidelines 2002 update states that the choice of specialized testing in the preoperative evaluation for noncardiac surgery in ambulatory patients is the exercise treadmill ECG to determine functional capacity and to detect myocardial ischemia [9,10]. In patients with baseline ECG abnormalities or in those who are unable to ambulate, nuclear scintigraphic or echocardiographic imaging should be used, depending on the expertise of the interpreters. Dobutamine echocardiography, stress treadmill echocardiography, and dipyridamole thallium have been shown to have similar positive and negative predictive values, sensitivities, and specificities when performed by clinicians with expertise [66–69].

776

Falcone et al.

Exercise ECG Cardiac Stress Testing In patients with a normal baseline ECG without a prior history of CAD, the exercise ECG response is abnormal in up to 25% of cases. In those with a prior history of MI or an abnormal resting ECG, the exercise ECG response can be abnormal in up to 50% of cases [70]. If an ischemic response occurs at a low cardiac workload, the positive predictive value of the test for determining a high cardiac risk is further increased [38,71,72]. Reduced exercise duration and exercise-induced ST-segment depression, in most [43,71] but not all [73], studies have been shown to correlate with an increased likelihood of postoperative cardiac events [43,71]. In the general population, the usefulness of treadmill ECG test without imaging is somewhat limited. The mean sensitivity and specificity are 68% and 77% for detection of single-vessel disease, 81% and 66% for detection of multivessel disease, and 86% and 53% for detection of three-vessel or left main CAD [9]. The older age of patients undergoing noncardiac and vascular surgery reduces the sensitivity and prognostic utility of exercise stress testing in this group [70,74]. Often these patients will have a submaximal treadmill exercise study, not being able to achieve their maximum predicted heart rate due to medical therapy, such as beta-blocker use, or to comorbid states, which can limit the results. Patients with intermediate- to high-risk clinical profiles who reach a cardiac workload >5 METs or a heart rate greater than 75 to 85% of maximum age-predicted with a nonischemic ECG response are at low-risk for postoperative cardiac events [38,74]. A patient’s performance on a treadmill or bicycle ergometer may also be predictive of postoperative cardiac outcomes [75]. The level at which ischemia is evident on the exercise ECG can be used to estimate an ‘‘ischemic threshold’’ for a patient to guide perioperative medical management. This may support further intensification of perioperative medical therapy in highrisk patients, which may impact on perioperative cardiovascular events [76]. Pharmacological Cardiac Stress Testing Pharmacological stress testing has been advocated for preoperative cardiac risk assessment for patients in whom exercise tolerance is limited. Often, these patients may not exercise sufficiently during daily life to provoke symptoms of myocardial ischemia or CHF. Pharmacological stress tests with echocardiographic or nuclear scintigraphic imaging have been studied extensively in preoperative cardiac risk assessment for noncardiac, and especially vascular, surgery and will be briefly reviewed here [9]. Nuclear Scintigraphy Myocardial Perfusion Imaging An abnormal preoperative dipyridamole-thallium scintigraphy scan has been shown to be a sensitive predictor of postoperative cardiac events [41,43,56,58,61,63,77–80]. Pooled data, though, show that the positive predictive value for adverse cardiac outcomes is low, ranging from 36 to 45%. The negative predictive value, on the other hand, is high (up to 97%). In several studies, the presence of a fixed defect was shown to have no predictive value for adverse postoperative cardiac outcomes [41,43,46,56,58,61,77,81] although, in two studies, there was a higher risk compared to patients with no ischemic defect [62,65]. Preoperative dipyridamole thallium scintigraphy was found to be superior to the clinical assessment alone for the determination of cardiac risk in early studies [56]. More recent studies support the use of preoperative dipyridamole thallium scintigraphy, in combination with clinical parameters, to identify patients at high risk for adverse cardiac outcomes after noncardiac surgery. In the initial study by Eagle et al. of patients

Perioperative Cardiovascular Evaluation

777

undergoing vascular surgery, an abnormal preoperative dipyridamole thallium scan was the most significant predictor of postoperative ischemic events. The likelihood of an ischemic event was 7% if the scan was negative and 45% if it revealed ischemia [41]. Clinical predictors of postoperative events were pathological Q-waves on ECG and evidence of redistribution defects. Patients with no clinical predictors were at low perioperative cardiac risk, and their perioperative outcomes were not affected by the results of preoperative stress testing. In a second study by Eagle [13] in a similar patient population, five clinical variables were predictive of postoperative cardiac events: age >70 years, Q-waves on baseline ECG, history of angina, ventricular ectopic activity requiring therapy, and DM. Dipyridamole thallium scintigraphy was found to be most useful in further stratifying patients considered at intermediate clinical risk (one or two clinical variables). In this group, the presence of a redistribution defect was associated with a 30% event rate compared to a 3% event rate in those without a thallium redistribution defect. In more than 50% of cases, the dipyridamole thallium stress test did not add incremental information to the preoperative assessment after clinical variables were evaluated. One study by Lette [74] has challenged the findings of Eagle [13]. Using 18 clinical parameters and several scoring systems, clinical variables were not found to predict postoperative events. Dipyridamole thallium scintigraphy was the only predictor of adverse postoperative events. Patients without evidence of ischemia or with only a fixed defect had no adverse cardiac outcomes. Some studies have also found that dipyridamole thallium scintigraphy scans with a large number and size of redistribution defects, presence of left ventricular dilatation after stress, or pulmonary radiotracer uptake are predictive of a higher postoperative cardiac risk [61,74,77,81]. The accuracy of dipyridamole thallium scintigraphy in the preoperative evaluation of patients before noncardiac surgery has been challenged by more recent trials [82–85]. In one study, intraoperative myocardial ischemia was assessed using continuous ECG monitoring and transesophageal echocardiography in patients undergoing vascular surgery [82]. All had dipyridamole thallium scintigraphy preoperatively, and all treating physicians were blinded to the results. There was a 5% incidence of adverse postoperative cardiac outcomes, with no association between redistribution defects and adverse cardiac outcomes. The sensitivity and specificity of thallium scintigraphy for all adverse outcomes were low (40 to 54% and 65 to 71%, respectively). The positive predictive value was low (27 to 47%) and the negative predictive value relatively high (61 to 82%). It was proposed that routine use of dipyridamole thallium scans for preoperative screening of patients before vascular surgery may not be warranted. A study by Baron [83] using SPECT, which has been found to have an increased sensitivity for the detection of CAD, confirmed these findings. Initial data using technetium 99m sestamibi, a newer myocardial perfusion tracer, indicate that it has the same diagnostic accuracy as thallium 201 for the detection of myocardial ischemia [86]. There are few studies that evaluate the long-term postoperative outcomes of patients with abnormal dipyridamole thallium scans. In one study [60], an abnormal dipyridamole thallium scan was associated with a significantly increased risk of cardiac death in the perioperative period and in late follow-up in comparison to those with a normal scan. A reversible defect was the only predictor of death or MI during late follow-up and was associated with a twofold greater risk of a cardiac event than if the defect was fixed. The number of perfusion defects, a history of angina, and the presence of chest pain during the study were independent predictors of perioperative cardiac events. Fleisher et al. utilized criteria from the TIMI-IIIB trials for quantification of dipyridamole thallium results [39].

778

Falcone et al.

They reported a significantly increased long-term risk only in the subset of patients with high-risk thallium markers, including increased lung uptake and multiple segments with reversible defects. Recently, the ACC/AHA Task Force algorithm was applied to high-risk patients before major vascular surgery [87]. An evaluation of functional status was also incorporated, and patients were stratified according to predicted cardiac risk. Overall mortality was 3.5% and cardiac mortality was 1%. There was no significant difference in cardiac morbidity between groups and no single independent predictor of morbidity. This algorithm for highrisk patients proved to be a safe and economical strategy. One study of patients who underwent elective aortic abdominal surgery over an 8-year period [88] compared data of two consecutive 4-year periods (1993–1996 (control period) versus 1997–2000 [intervention period] based upon the ACC/AHA Guidelines including perioperative medical therapy). Implementation of ACC/AHA Guidelines in the second time period resulted in more preoperative myocardial scanning and coronary revascularization, fewer cardiac complications (11.3% versus 4.5%), and higher event-free survival at 1 year after surgery (98.2% versus 91.3%). In a small, prospective, randomized trial, 99 patients undergoing elective vascular surgery were randomized to preoperative cardiac stress testing or to no stress testing and followed for up to 12 months postoperatively for adverse cardiac outcomes [89]. There were three perioperative adverse cardiac events, no perioperative deaths, and two additional adverse cardiac events during 12-month follow-up. There was no difference in postoperative adverse cardiac outcomes between these two groups, suggesting that preoperative cardiac stress was unnecessary in this patient group, who had a high functional capacity at baseline. Larger randomized clinical trials are needed to confirm these findings. Dobutamine Stress Echocardiography Dobutamine stress echocardiography (DSE) involves the identification of new or worsening myocardial wall motion abnormalities using two-dimensional echocardiography during infusion of intravenous dobutamine. This technique has been shown to have the same accuracy as dipyridamole thallium scintigraphy for the detection of CAD [69,90–92]. More recently, DSE has been found to be useful for the assessment of preoperative cardiac risk among patients undergoing noncardiac surgery [9,67,93,94]. The estimated low positive predictive values (17 to 43%) and high negative predictive values (93 to 100%) are similar to those for dipyridamole thallium [41,56,58,67,69,82,95]. There are several advantages to DSE compared to dipyridamole thallium scintigraphy; DSE can also assesses left ventricular function and valvular abnormalities. The cost of DSE is significantly lower [67,69], there is no radiation exposure, the duration of the study is significantly shorter, and results are immediately available. Several studies have assessed the value of DSE in preoperative risk assessment. In one of the first studies, postoperative cardiac events were noted in 21% of patients undergoing noncardiac surgery who had evidence of preoperative dobutamine stress-induced myocardial ischemia, while no patient had a cardiac even if ischemia was not induced [59]. Other authors have reported similar findings [68,69,96]. The results of DSE add significantly to the clinical risk assessment of patients undergoing major vascular surgery [69], particularly in patients with intermediate clinical risk [96]. As with other forms of preoperative risk assessment, however, DSE has a relatively low positive predictive value [68] and many patients with abnormal test results do not have adverse postoperative cardiac events.

Perioperative Cardiovascular Evaluation

779

Presently, there are few studies that report long-term cardiac outcomes after noncardiac surgery among patients with abnormal preoperative DSE. In one study, patients were followed for up to 2 years after major vascular surgery [67]. There were two cardiac events (3%) among patients with negative DSE. Among patients with positive DSE, 68% subsequently underwent coronary revascularization before the noncardiac surgery was performed. There were no perioperative events in the group of patients with positive DSE who underwent coronary revascularization before noncardiac surgery. By contrast, 40% of those with positive DSE who did not undergo coronary revascularization had perioperative adverse cardiac outcomes. In this study, DSE predicted perioperative and long-term outcome among patients undergoing major vascular surgery, with a high negative predictive value. In a second study, patients undergoing major vascular surgery were evaluated by clinical parameters and results of DSE and followed for an average of 19 months postoperatively [97]. The presence of extensive dobutamine-induced wall motion abnormalities and a previous history of MI independently predicted late cardiac events, increasing risk up to 6-fold. A meta-analysis of 15 studies to compare dipyridamole-thallium and DSE for risk stratification before vascular surgery in intermediate-risk patients found that the prognostic value of both techniques is comparable, but that the accuracy varies with CAD prevalence [95]. In summary, the ACC/AHA guidelines recommend that exercise or pharmacological cardiac stress testing be performed before noncardiac surgery with the same indications as for those who are not undergoing noncardiac surgery [10].

CORONARY ANGIOGRAPHY AND REVASCULARIZATION BEFORE NONCARDIAC SURGERY Coronary Angiography An abnormal noninvasive study in the preoperative period may lead to coronary angiography and revascularization, to changes in anesthetic technique, to utilization of expensive resources such as intensive care units, or to aggressive treatment of hemodynamic fluctuations [98]. Historically, the use of coronary angiography as a screening procedure before elective peripheral vascular surgical procedures was advocated [11] due to the high perioperative mortality associated with the high prevalence of CAD in this patient group [60]. Many argue that using coronary angiography as a preoperative screening tool is too costly and places patients at further risk due to the cumulative risks of these procedures, especially in the elderly with significant comorbid diseases. This has led to increased interest in preoperative noninvasive cardiac stress testing to assist in risk-stratifying patients before elective noncardiac surgery [99]. It is generally assumed that coronary angiography should be performed preoperatively in patients with evidence of ischemia on stress testing [41,56,58,60], although there are no published randomized trials to support this approach. The present indications for coronary angiography in the preoperative evaluation before noncardiac surgery are adapted from the ACC/AHA Guidelines for coronary angiography in the general population [100]. Coronary angiography should be considered in patients with unstable symptoms of angina pectoris, with high-risk or equivocal preoperative noninvasive stress test results, or in patients with multiple clinical risk parameters who are undergoing a high-risk noncardiac surgical procedure. It is generally agreed

780

Falcone et al.

that performance of coronary angiography is not indicated in patients with low-risk results on preoperative noninvasive cardiac stress testing, in those who are asymptomatic up to 5 years after prior coronary revascularization, or for CAD screening without a prior noninvasive cardiac stress test.

Percutaneous Coronary Intervention The role of percutaneous coronary intervention (PCI) before noncardiac surgery to reduce the incidence of perioperative cardiac events has not been studied in a prospective randomized fashion. In a small, retrospective study of patients who underwent PCI prior to noncardiac surgery, 10% required urgent CABG [101]. Successful preoperative PCI in highrisk patients was shown to be associated with a low perioperative cardiac risk during noncardiac surgery. In a large 10-year study of patients who underwent PCI and/or CABG prior to elective abdominal aortic aneurysm surgery from 1980 to 1990, there were no perioperative deaths in patients with prior coronary revascularization compared to a 2.9% perioperative mortality for the group as a whole [102]. Three-year survival was similar with either revascularization procedure (92% PCI versus 83% CABG), although patientswith PCI had a significantly higher number of late events at 3 years (56.5% versus 27.3%). This study suggests that coronary revascularization before high-risk vascular surgery may improve perioperative outcomes in a select group of patients. Posner et al. [103] retrospectively analyzed hospital discharge abstracts, matching patients with CAD undergoing noncardiac surgery with and without prior PCI. In this nonrandomized study, they noted a significantly lower rate of 30-day cardiac complications in patients who underwent PCI at least 90 days before the noncardiac surgery. PCI performed within 90 days of noncardiac surgery was not associated with improvement in outcome. These results suggest that PCI performed ‘‘to get the patient through surgery’’ may not improve perioperative outcome, although this study did not control for CAD severity, medical management, or comorbidities. One study evaluated the value of multivessel coronary angioplasty on subsequent noncardiac surgery a median of 29 months after the most recent coronary revascularization procedure [104]. Mortality or non-fatal MI occurred in 4 of the 250 surgery-assigned patients and 4 of the 251 angioplasty-assigned patients. Therefore, there does not appear to be an optimal choice of revascularization procedure for multivessel disease, although a larger scale study will be necessary to confirm these findings. Evidence regarding the value of coronary stent placement is currently lacking. Of importance, the outcome in 40 patients who underwent coronary stent placement less than 6 weeks before noncardiac surgery requiring a general anesthesia was recently reported [105]. There were 7 myocardial infarctions, 11 major bleeding episodes, and 8 deaths. All deaths and MIs, as well as the majority of bleeding episodes, occurred in patients subjected to surgery fewer than 14 days from the stent procedure. Four patients expired after undergoing surgery 1 day after stenting. The time between stenting and surgery appeared to be the main determinant of outcome, and the authors recommend that a minimum of 2 weeks, and preferably 4 weeks, elapse before elective surgery. The risk of significant complications during PCI limits its generalized use, especially in elderly patients [101,102,106]. The indications for the use of PCI before noncardiac surgery should therefore be considered similar to the indications for PCI in other settings [107].

Perioperative Cardiovascular Evaluation

781

Coronary Artery Bypass Graft Surgery Studies have shown that the preoperative presence of CAD significantly increases the risk for a perioperative cardiac event after noncardiac surgery [108–111], although there are no randomized trials to assess whether preoperative CABG lowers this risk [112]. Some evidence exists that preoperative revascularization may decrease postoperative cardiac risk two- to four-fold in patients undergoing elective vascular surgery [113–115]. In one study, CABG surgery was performed in 216 patients before noncardiac surgery with a 5.2% mortality [11]. The overall surgical mortality for both the cardiac and noncardiac procedures combined in the total cohort of 1000 patients was 2.6%. This indicates that a significant proportion of patients had inoperable or severe CAD. The risk of adverse cardiac outcomes after noncardiac surgery in these patients may be reduced with CABG surgery, but with the added risk of the CABG procedure itself. Many patients referred for noncardiac surgery, and especially for vascular surgery, are elderly with comorbid conditions, and it is in these patients that the risks of CABG itself are significant [11,116]. Several studies suggest that preoperative CABG can reduce postoperative risk if the patient is felt to be at low risk for the CABG itself [11,117–122]. In a subset of patients in the Coronary Artery Surgery Study (CASS) Registry enrolled from 1978 to 1981, the influence of CABG on perioperative cardiac risks prior to noncardiac surgery was studied [122]. The operative mortality for patients with CABG prior to noncardiac surgery was 0.9% but was significantly higher at 2.4% among patients without prior CABG. However, there was a 1.4% mortality rate associated with the CABG procedure itself [112]. Eagle and colleagues reported a long-term analysis of patients in the CASS registry who received medical or surgical therapy for CAD and who subsequently underwent 3368 noncardiac operations over the next 10 years [123]. The rate of perioperative MI and death was stratified by the type of surgical procedure. Specifically, low-risk surgeries such as skin, breast, urological, and minor orthopedic procedures were associated with a total morbidity and mortality of less than 1%, and prior revascularization did not affect outcome. Intermediate-risk surgery, such as abdominal or thoracic procedures or carotid endarterectomy, was associated with a combined morbidity and mortality of 1 to 5%, with a small, but significant, improvement in outcome among patients who underwent prior coronary revascularization. The most significant improvement in outcome was noted among patients undergoing major vascular surgery who underwent prior coronary revascularization. Therefore, in this nonrandomized study, CABG was beneficial in selected subgroups if they survived the initial coronary revascularization operation. Few data support the use of CABG solely to improve perioperative outcomes among patients undergoing noncardiac surgery. The indications for CABG are based on ACC/ AHA Task Force Guidelines and indications for CABG for the general population [100]. Timing of CABG prior to noncardiac surgery is an individualized decision, and there are no prospective, randomized data to suggest the optimal strategy. An alternative approach to determine the optimal strategy for medical care is construction of decision analyses. Decision analyses have been published on cardiovascular testing before major vascular surgery [115,124,125]. All analyses assumed that patients with significant CAD would undergo CABG surgery prior to noncardiac surgery. The models indicated that the optimal decision was sensitive to local morbidity and mortality rates within the clinically observed range. These models suggest that preoperative testing for the purpose of coronary revascularization is not the optimal strategy if the anticipated perioperative morbidity and mortality are low.

782

Falcone et al.

First and foremost, the primary cost of preoperative testing and revascularization is the revascularization procedure itself. Therefore, the indications for revascularization, and thus the frequency of its use, has a significant impact on the model. Second, potential long-term benefits of coronary revascularization in this population were not included in the analysis, potentially biasing against the revascularization arm. If long-term survival is included in the models, then coronary revascularization may lead to improved overall outcome and be a cost-effective intervention. However, a patient’s age should be included in the equation. For example, an 80-year-old diabetic patient with significant comorbid diseases may gain little additional life-years by undergoing coronary revascularization. By contrast, a 55-year-old man with an abdominal aortic aneurysm who is found to have occult left main coronary disease would have a substantial increase in both the length and quality of his life from preoperative cardiovascular testing and coronary revascularization.

PERIOPERATIVE INTERVENTIONS TO REDUCE RISK Table 4 indicates strategies to reduce risk of noncardiac surgery. Intra-Aortic Baloon Counterpulsation Among patients with unstable angina or severe CAD, placement of an intra-aortic balloon counterpulsation device has been used before induction of anesthesia in patients considered high-risk for noncardiac surgical procedures. In several small case series, perioperative morbidity and mortality was low [126]. Perioperative CHF It is critical to optimize the medical management of CHF, both systolic and diastolic, before an elective noncardiac surgical procedure. For patients with left ventricular systolic dysfunction, this often involves the use of diuretics, angiotensin-converting enzyme inhibitors, and beta-blockers. These medications can alter hemodynamic and metabolic parameters that may impact on perioperative cardiac risk. Patients at risk for CHF postoperatively are those with a history of arrhythmia and DM and those undergoing prolonged surgical procedures [3,122]. Patients who develop isolated CHF postoperatively have a better long-term prognosis than those in whom a postoperative myocardial ischemic event occurs [127]. Pharmacological Agents Beta-Blockers Several studies in the late 1970s indicated that beta-blocker therapy could be safely continued preoperatively, often resulting in a reduction in the incidence of myocardial ischemia [128]. These studies, and the concern for beta-blocker withdrawal-induced tachycardia, HTN, and myocardial ischemia [129,130], led to recommendations to continue beta-blocker therapy before surgery [131]. By decreasing adrenergic stimulation of the heart, beta-blockers may reduce the incidence of perioperative arrhythmias and myocardial ischemia. In a study of patients with

Perioperative Cardiovascular Evaluation

783

mild-to-moderate HTN undergoing general anesthesia, the use of a single low-dose oral beta-blocker produced a 13-fold reduction in the incidence of intraoperative myocardial ischemia by ECG [132]. Among patients not receiving beta-blocker therapy, ischemia was detected in 28%. All episodes were either during tracheal intubation or emergence from anesthesia. In those treated with beta-blocker therapy, only 2.2% had evidence of myocardial ischemia. Those who received beta-blockers more frequently had bradycardia, had a greater fall in mean arterial pressure during premedication, and had less of a pressor response during tracheal intubation and emergence from anesthesia. The prophylactic use of beta-blocker therapy in patients at high risk of postoperative cardiac complications appears to be safe and effective [133]. The first well-designed randomized, placebo-controlled study in high-risk noncardiac surgical patients involved the perioperative use of atenolol, which was administered beginning 2 days preoperatively and continued for 7 days postoperatively [76,134]. A significantly lower incidence of perioperative ischemia and improved event-free 6-month survival was observed in the atenolol group. No difference in perioperative MI or cardiac death was noted between groups. Of note, several risk factors and medications were not equally distributed in the two groups, with the placebo group having a higher risk profile. Further research is required to confirm these findings. In a study of the perioperative use of bisoprolol in elective major vascular surgery, this medication was administered at least 7 days perioperatively, titrated to achieve a resting heart rate V60 beats per minute, and continued postoperatively for 30 days [135]. Of note, the study was confined to patients with one or more clinical markers of cardiac risk (prior MI, DM, angina pectoris, CHF, age >70 years or poor functional status), and with myocardial ischemia on DSE. Patients with large zones of myocardial ischemia were excluded from the trial. Bisoprolol led to an approximate 80% reduction in perioperative MI or cardiac death. The efficacy of bisoprolol in the highest risk group, those who would be considered for coronary revascularization or modification or cancellation of the surgical procedure, is unknown. However, the event rate in the placebo group (nearly 40%) suggests that all but the highest risk patients were enrolled in the trial. The same data were then reevaluated with respect to clinical factors, DSE results and beta-blocker usage [136]. A clinical risk score was calculated by assigning one point for each of the following characteristics: age z 70 years, current angina, MI, CHF, prior cerebrovascular, DM, and renal failure. Importantly, DSE was performed only in patients with a significant number of risk factors. Those patients who demonstrated new wall motion abnormalities had higher event rates compared to those without new wall motion abnormalities, for the same clinical risk score. When the risk of death of MI was stratified by perioperative beta-blocker usage, there was no significant improvement in those without any of the prior risk factors (Fig. 2). In those with fewer than three clinical risk factors, the use of beta-blockers was associated with a lower rate of cardiac events (0.8% versus 2.3%), and beta-blocker therapy was very effective in reducing cardiac events in those with limited stress-induced ischemia on DSE (33% versus 2.8%). By contrast, beta-blocker therapy had no effect in those patients with more extensive stress-induced ischemia on DSE. Several recent reviews also highlight the importance of perioperative beta-blocker therapy, although there is lack of strong evidence regarding the optimal agents or timing [137]. A scientific review provides recommendations regarding the use of perioperative betablocker therapy [138]. If patients are taking beta-blockers preoperatively, then it is important to continue them in the perioperative period. If patients are undergoing vascular surgery and demonstrate areas of myocardial ischemia on preoperative cardiac stress

Patients requiring calcium channel blockers to control angina and/or provocable ischemia Patients previously requiring nitrates to control angina and/or provocable ischemia; patients developing perioperative ischemia without hypovolemia

Calcium channel blockers

Nitrates

Patients with existing CAD undergoing vascular surgery

Patients already on; evidence of existing CAD (prior MI, angina or its equivalent, and/or provocable myocardial ischemia in presence of risk factors for CAD

Patient selection

Alpha-2 agonists

Medical Therapy Beta-blockers

Category

Table 4 Strategies to Reduce Cardiac Risk of Noncardiac Surgery

Continue prior regimen; institution of intravenous delivery

Continue prior regimen

Continue prior regimen; ideally begin treatment days to weeks preoperative (e.g., atenolol 25–50 mg q.d. or metoprolol 25–50 mg b.i.d. titrating to achieve heart rate V60 beats per minute) 2 Ag/kg of clonidine orally as premedication; mizaverol (not commercially available) given to achieve a target concentration of 20 ng/mL

Specific therapy

Data on clonidine supports reduction in myocardial ischemia, not MI or death; mivazerol only demonstrates efficacy in subset of large clinical trial Not well studied; no convincing data to support efficacy Limited trial data; no convincing prophylactic value

Most convincing data exist for patients undergoing vascular surgery with known CAD and provocable ischemia on stress test

Quality of supporting data

784 Falcone et al.

Coronary artery bypass surgery

Percutaneous coronary intervention

Patient with poorly controlled angina despite maximal medical therapy; patients with large zones of provocable myocardial ischemia on preoperative stress test, particularly when already on maximal medical therapy and undergoing high stress noncardiac surgery Patients with poorly controlled angina despite maximal medical therapy; Patientswith significant (>50%) left main coronary stenosis; Patients with severe 2- or 3- vessel coronary artery disease with involvement of the proximal left anterior descending artery and easily provocable myo cardial ischemia on stress testing and/or impaired left ventricular systolic function at rest

All data are observational, based on small sample sizes or utilize administrative databases

All data are observational; much is old (e.g. >10 years ago)

PCI is justified independent of noncardiac surgery; dilation and stenting of severe lesions—allow 4 weeks of post-stent antiplatelet therapy and coronary endothelial healing before elective noncardiac surgery CABG is justified independent of noncardiac surgery; CABG should generally be performed before elective, high stress noncardiac surgery to reduce risk of perioperative cardiac events

Perioperative Cardiovascular Evaluation 785

786

Falcone et al.

Figure 2 Perioperative cardiac risk and death in dierent populations of patients enrolled in the DECREASE trial. Risk is defined according to clinical risk and use of beta-blockers both as part of the randomized and nonrandomized cohort of individuals. Patients with risk factors and a positive stress test demonstrating 1 to 4 areas of regional wall motion abnormality on dobutamine stress echocardiography were randomized to perioperative bisoprolol titrated preoperatively to a heart rate