[Skip to Content]
[Skip to Content Landing]
March 15, 2000

Time to Reevaluate Risk Stratification Guidelines for Medically Supervised Exercise Training in Patients With Coronary Artery Disease

Author Affiliations

Author Affiliations: Preventive and Rehabilitative Cardiac Center, Division of Cardiology, Department of Medicine, and the Burns and Allen Research Institute, Cedars-Sinai Medical Center; and the Department of Medicine, University of California School of Medicine, Los Angeles, Calif.

JAMA. 2000;283(11):1476-1478. doi:10.1001/jama.283.11.1476

Exercise training is an effective treatment for coronary artery disease. Medically supervised exercise has been demonstrated to halt progression and enhance regression of angiographic stenoses,1 reduce myocardial stress perfusion defects,2 enhance quality of life,3 and reduce mortality.4,5 Compared with other currently available therapies for coronary artery disease, the 25% risk reduction observed with exercise is comparable with the 20% achieved with aspirin,6 the 20% with β blockers,7 and the 15% observed with angiotensin-converting enzyme inhibitors.8 Because the studies demonstrating overall mortality reduction were performed prior to the use of thrombolysis and aggressive revascularization therapy, the net additional benefit of supervised exercise in these lower-risk populations is unknown. Nevertheless, based on these and other randomized controlled trial data, the Agency for Health Care Policy and Research Cardiac Rehabilitation Guidelines9 call for universal application of medically supervised exercise for the more than 12 million patients with established coronary artery disease.

Medically supervised exercise, often in the form of cardiac rehabilitation, typically involves supervision by a program staff member with or without electrocardiographic monitoring. Patients usually initiate programs within 2 to 6 weeks following a recent cardiac event and perform exercise training sessions 3 times weekly with direct supervision and electrocardiographic monitoring for periods of 6 to 12 weeks. Subsequent exercise training is usually performed with decreasing amounts of supervision and electrocardiographic monitoring. These types of programs have been demonstrated to be safe,10,11 cost-effective,12-14 and efficacious for increasing functional capacity and compliance with exercise15 compared with the usual care for patients with coronary artery disease.

Contrary to the Agency for Health Care Policy and Research guidelines,9 there are an increasing number of authoritative guidelines designed to reduce access to and use of cardiac rehabilitation services. Whether these guidelines represent attempts to identify patients most likely to benefit or are simply responses to financial pressure related to the current health care reimbursement climate is unknown. Nevertheless, the American Association of Cardiovascular and Pulmonary Rehabilitation,16 American College of Cardiology,17 American College of Physicians,18 and American Heart Association19 have each proposed exercise-risk stratification criteria as a means of judiciously using medical supervision, electrocardiographic monitoring, or both during exercise training. The American Association of Cardiovascular and Pulmonary Rehabilitation and the American College of Cardiology guidelines suggest that supervision and electrocardiographic monitoring can be limited for intermediate- and low-risk patients, while the American College of Physicians and American Heart Association guidelines suggest that low-risk patients need neither supervision nor monitoring.

Although the impact of these guidelines on exercise program safety and efficacy has not been tested in prospective studies or trials, a retrospective study of 5720 patient-exercise hours demonstrated that the guidelines did not predict cardiac complications during supervised exercise.20 Furthermore, 58% and 75% of cardiac complications occurred in the low-risk groups, as stratified by the American Heart Association's and American College of Physicians' guidelines, respectively.20 These findings suggest that implementation of these guidelines likely will be found to be inadequate for ensuring the safety of patients with coronary artery disease and potentially to be seriously misguided as health care policy.

A current tenet in health care resource management is the concept that therapy will be most efficacious and least costly per amount of benefit among populations at the highest risk. This tenet, however, runs counter to the clinical trial data on supervised exercise-cardiac rehabilitation services. Randomized trials that have demonstrated the utility of supervised exercise in reducing overall mortality have included populations that were heterogeneous for low- and high-risk patients,4,5 and the data were not analyzed according to risk status. Although it could be argued that today's postmyocardial infarction patients have a lower risk than patients enrolled in these previous exercise trials4,5 due to current thrombolysis and revascularization strategies, as well as more aggressive risk-factor modification, such as cholesterol-lowering therapy, today's cardiac rehabilitation populations are also older and more frequently have complex postprocedure disease, likely balancing any risk reduction. In support of this, comparative data on exercise training safety demonstrate stable rates of serious complications from the years 1980 to 1995,11 a period characterized by large increases in interventional therapy use. More recent randomized trials that have demonstrated the efficacy of supervised exercise–cardiac rehabilitation for improving quality of life3 and reducing medical expenditures12 also did not risk stratify the patients enrolled, suggesting that benefit occurs across a range of risk. Thus, exercise–risk stratification used to apply or deny supervised exercise services selectively according to expected benefit is not supported by the clinical trial data.

In the current environment of high technology with the increasing prevalence of the managed care model for covering health care benefits, the role of supervised exercise-cardiac rehabilitation as treatment for patients with coronary artery disease is being increasingly questioned. The rates are low for both major (cardiac arrest, myocardial infarction) complications (1/67 126 patient-exercise hours)11 and serious (sustained chest pain, worsening dyspnea, arrhythmia) complications (1/320 patient-exercise hours)21 during supervised exercise, which demonstrate the relative safety of exercise for patients with coronary artery disease today.

Accordingly, if exercise is relatively safe, it is reasonable to ask why insurance companies, health care organizations, or society should bother to offer and pay for supervised exercise programs? Several issues deserve careful consideration. First, the data documenting the safety of exercise in this population are available only for supervised programs, which predominantly are performed in a standard phase 2 and 3 format.10,11 The comparative safety of unsupervised exercise in patients with coronary artery disease remains untested and unknown, although the mortality rate reported for supervised cardiac rehabilitation (1/784,000 patient-exercise hours)10 is less than half of what has been observed for those in the general public who jog (1/396,000 patient-exercise hours).22 Although this comparison has obvious limitations in population differences, the fact that a diseased population can exercise more safely than a nonovertly diseased population suggests that medical supervision can improve exercise training safety. Second, data suggest that exercise as a therapy addresses a disease manifestation for which there is no other treatment, namely sudden cardiac death. A 43% reduction in out-of-hospital sudden cardiac death was observed in patients randomly assigned to exercise in both a single trial23 and in a meta-analysis.4 While the mechanisms responsible for this effect of exercise are unknown, these data suggest that exercise provides a unique contribution to the treatment of coronary artery disease not currently available via other therapies.

Currently, less than 20% of eligible patients use supervised exercise–cardiac rehabilitation services because of the inadequate number of programs available and because of geographic barriers imposed by travel distance to the program.24 Previous literature, however, demonstrates that medical supervision can be safely and effectively used in home-based rather than center-based programs for selected patients.15 DeBusk and coworkers25 have demonstrated the utility of a home-based case management program, supervised by frequent program staff telephone contact, to be effective for increasing exercise compliance in patients with coronary artery disease.

Changes in health care management have resulted in an increased focus on outcomes and cost, with the overall goal of limiting health care expenditures while maintaining or improving quality of care. Recent data suggest that these changes are resulting in a limitation of supervised exercise cardiac rehabilitation use.26 Indeed, application of some of these authoritative risk-stratification guidelines11,25 would have resulted in 55% of patients being denied services under the American College of Physicians' and American Heart Association's guidelines and in 74% of patients experiencing a services reduction by as much as 2 to 4 weeks under the American College of Cardiology guidelines.

Recent studies suggest that this trend and these exercise–risk stratification guidelines are not optimal. Specifically, several studies have documented that cardiac rehabilitation services are cost-effective by reducing recurrent hospitalization and health care expenditures,12-14 with 1 study reporting a net cost savings of $70 per patient.13 Oldridge and coworkers12 have demonstrated that despite a greater financial expenditure, a comprehensive cardiac rehabilitation program results in an expenditure of $9200 per quality-adjusted life-year saved, an amount comparable to that of hypertension treatment ($16,000 per life-year saved) and coronary artery bypass surgery ($7900 per life-year saved). Two independent meta-analyses of randomized controlled exercise trials performed predominantly in postmyocardial infarction patients suggest that a minimum of 12 weeks of exercise training reduces mortality by 20% to 25%,4,5 an effect that is not apparent in programs of lesser duration.4 Moreover, this survival benefit has been demonstrated to be maintained for up to 10 years,27 and possibly 19 years,28 following the initial program if functional capacity is maintained.

Thus, supervised cardiac rehabilitation exercise programs fulfill both the current health care goals of managing expenses cost-effectively and improving quality of care and patient outcomes. Rather than designing exercise–risk-stratification guidelines that limit access to supervised exercise services, future research should focus on remodeling cardiac rehabilitation programs into on-site and off-site, low-cost, user-friendly, supervised exercise programs24,29-31 with the aim of increasing use from the 20% of eligible patients treated currently to the larger group of 12 million patients with established coronary artery disease in the United States.9

Hambrecht R, Neibauer J, Marburger C.  et al.  Various intensities of leisure time physical activity in patients with coronary artery disease: effects on cardiorespiratory fitness and progression of coronary atherosclerotic lesions.  J Am Coll Cardiol.1993;22:468-477.Google Scholar
Schuler G, Schlierf G, Wirth A.  et al.  Low-fat diet and regular, supervised physical exercise in patients with symptomatic coronary artery disease: reduction of stress-induced myocardial ischemia.  Circulation.1988;77:172-181.Google Scholar
Oldridge NB, Guyatt G, Jones N.  et al.  Effects on quality of life with comprehensive rehabilitation after acute myocardial infarction.  Am J Cardiol.1991;67:1084-1089.Google Scholar
O'Connor GT, Buring JE, Yusuf S.  et al.  An overview of randomized trials of rehabilitation with exercise after myocardial infarction.  Circulation.1989;80:234-244.Google Scholar
Oldridge NB, Guyatt GH, Fischer ME, Rimm AA. Cardiac rehabilitation after myocardial infarction: combined experience of randomized clinical trials.  JAMA.1988;260:945-950.Google Scholar
Antiplatelet Trialists' Collaboration.  Secondary prevention of vascular disease by prolonged antiplatelet treatment.  BMJ.1983;296:320-331.Google Scholar
The Beta-Blocker Pooling Project Research Group.  The Beta-Blockers Pooling Project (BBPP): subgroup findings from randomized trials in post infarction patients.  Eur Heart J.1988;9:8-16.Google Scholar
Pfeffer MA, Braunwald E, Moye LA.  et al.  Effect of captopril on mortality and morbidity in patients with left-ventricular dysfunction after myocardial infarction: results of the survival and ventricular enlargement trial.  N Engl J Med.1992;327:669-677.Google Scholar
Agency for Health Care Policy and Research.  Cardiac Rehabilitation Guidelines.  Silver Spring, Md: Agency for Health Care Policy and Research Publishing Clearinghouse; 1995.
Van Camp SP, Peterson RA. Cardiovascular complications of outpatient cardiac rehabilitation programs.  JAMA.1986;256:1160-1163.Google Scholar
Vongvanich P, Paul-Labrador MJ, Bairey Merz CN. Safety of medically supervised exercise in a cardiac rehabilitation center.  Am J Cardiol.1996;77:1383-1385.Google Scholar
Oldridge N, Furlong W, Feeny D.  et al.  Economic evaluation of cardiac rehabilitation soon after acute myocardial infarction.  Am J Cardiol.1993;72:154-161.Google Scholar
Levin LA, Perk J, Hedback B. Cardiac rehabilitation-a cost analysis.  J Intern Med.1991;230:427-434.Google Scholar
Bondestam E, Breikss A, Hartford M. Effects of early rehabilitation on consumption of medical care during the first year after acute myocardial infarction in patients < or = 65 years of age.  Am J Cardiol.1995;75:767-771.Google Scholar
Miller N, Haskell W, Berra K, DeBusk RF. Home versus group exercise training for increasing functional capacity after myocardial infarction.  Circulation.1984;70:645-649.Google Scholar
American Association of Cardiovascular and Pulmonary Rehabilitation.  Guidelines for Cardiac Rehabilitation ProgramsChampaign, Ill: Human Kinetics: 1995.
American College of Cardiology.  Position report on cardiac rehabilitation.  J Am Coll Cardiol.1986;7:451-453.Google Scholar
Health and Public Policy Committee, American College of Physicians.  Cardiac rehabilitation services.  Ann Intern Med.1988;109:671-673.Google Scholar
Fletcher GF, Balady G, Froelicher VF, Hartley LH, Haskell WL, Pollock ML. Exercise standards: a statement for healthcare professionals from the American Heart Association: medical/scientific statement.  Circulation.1990;82:2286-2322.Google Scholar
Paul-Labrador M, Vongvanich P, Bairey Merz CN. Risk stratification for exercise training in cardiac patients: do the proposed guidelines work?  J Cardiopulm Rehabil.1999;19:118-125.Google Scholar
Vongvanich P, Bairey Merz CN. Supervised exercise and electrocardiographic monitoring during cardiac rehabilitation.  J Cardiopulm Rehabil.1996;16:233-238.Google Scholar
Thompson P, Funk E, Carleton R, Sturner W. Incidence of death during jogging in Rhode Island from 1975 through 1980.  JAMA.1982;247:2535-2538.Google Scholar
Kallio V, Hamalainen H, Hakkila J, Luurila O. Reduction in sudden deaths by a multifactorial intervention programme after acute myocardial infarction.  Lancet.1979;2:1091-1094.Google Scholar
Bairey Merz CN, Rozanski A. Remodeling cardiac rehabilitation into secondary prevention programs.  Am Heart J.1996;132:418-427.Google Scholar
DeBusk RF, Miller NH, Superko HR.  et al.  A case-management system for coronary risk factor modification after acute myocardial infarction.  Ann Intern Med.1994;120:721-729.Google Scholar
Paul-Labrador MJ, Bairey Merz CN. Phase II cardiac rehabilitation exercise conditioning improvement according to program participation.  J Cardiopulm Rehabil.1994;14:328.Google Scholar
Hamalainen H, Luurila OJ, Kallio V.  et al.  Long-term reduction in sudden deaths after a multifactorial intervention programme in patients with myocardial infarction: 10-year results of a controlled investigation.  Eur Heart J.1989;10:55-62.Google Scholar
Dorn J, Naughton J, Imamura D, Trevisan M.for the NEHDP Project Staff.  Results of a multicenter randomized clinical trial of exercise and long-term survival in myocardial infarction patients: the National Exercise and Heart Disease Project (NEHDP).  Circulation.1999;100:1764-1769.Google Scholar
Wenger NK. Supervised versus unsupervised exercise training following myocardial infarction and myocardial revascularization procedures.  Ann Acad Med Singapore.1992;21:141-144.Google Scholar
Pashkow FJ. Issues in contemporary cardiac rehabilitation: a historical perspective.  J Am Coll Cardiol.1993;21:822-834.Google Scholar
Gordon NF, Haskell WL. Comprehensive cardiovascular disease risk reduction in a cardiac rehabilitation setting.  Am J Cardiol.1997;80:69H-73H.Google Scholar