Rechnitzer, Sample size: 761 (390 | Intervention: High Cunningham, high-intensity exercise, intensity (65-85% of estimated maximal oxygen uptake) or lowintensity (50% or less estimated maximal oxygen uptake). Supervised 2 times/ week for mean of 3.3 years. Followup: Mean 3.3 years Intervention: Supervised monitored aerobic Fatal reinfarction: 15 high-intensity group 13 low-intensity group No significant difference between groups Total mortality: 6 exercise No significant difference between groups exercise 1 hour 4 times/5 control week at 80-90% maximal baseline exercise test heart rate for 8-9 weeks. Controls, home exercise cycling, walking, and calisthenics. Followup: 4.5 years Table 8. (continued) Reference, Country Patients Intervention, Followup Outcome Total mortality: Hamalainen, Sample size: 375 (188 | Intervention: Luurila, Kallio, intervention, 187 control) Multifactorial. Dietary, 44% intervention 52% control No significant difference between groups Sudden death: 24 (13%) intervention 43 (23%) control (p = .01) Total coronary mortality: 35% intervention 47% control Total mortality: RP vs. CP+ UC (p = .03) Total mortality: 10 high-intensity exercise 9 light exercise No significant difference between groups daily at home. Followup: 4 and 12 months Table 8. (continued) Reference, Country DeBusk, Patients Intervention, Followup Outcome Cardiac mortality: 3.8% intervention 3.1% usual care Sample size: 585 (293 | Intervention: NurseHouston Miller, intervention, 292 usual | directed case Total mortality: 4.1% intervention 3.4% usual care Neither difference statistically significant between groups Total cardiac events Cardiac or sudden 2 intervention Note: bpm excluded. beats per minute; CABG = coronary artery bypass graft surgery; CHD = coronary heart disease; LDL = low-density lipoprotein; MI = myocardial infarction; PTCA = percutaneous transluminal coronary angioplasty. of cardiac death is warranted because the data represent an aggregate analysis of information received from cardiac rehabilitation programs that were sent survey questionnaires and provided data in a retrospective manner. Furthermore, the survey represents data from the 1980s and may not reflect contemporary management of coronary patients. Conclusions A survival benefit among patients participating in exercise training as a component of multifactorial cardiac rehabilitation is suggested from review of the scientific data, but this benefit cannot be attributed solely to exercise training. The relatively low mortality rates reported in questionnaire surveys of supervised exercise rehabilitation and the inability of single studies to enroll adequate numbers of patients in a randomized controlled trial could underestimate the potential benefit of cardiac rehabilitation in reducing mortality rates. For this reason, the panel drew on meta-analyses to gain additional information about mortality outcomes. Two meta-analyses of 21 randomized controlled trials of cardiac rehabilitation established significant mortality reduction in rehabilitation patients compared with control patients53,54; the meta-analyses established an approximately 25-percent relative reduction in mortality at the 3-year followup in rehabilitation patients compared with control patients. This mortality reduction is similar to that with other interventions for patients with coronary disease (e.g., trials of beta-blocking drug therapy following myocardial infarction; ACE inhibitors therapy for left ventricular systolic dysfunction and heart failure). The meta-analyses of randomized controlled trials included more than 4,000 patients with coronary disease; the beneficial mortality outcome was greater in the 15 trials that used multifactorial cardiac rehabilitation compared with the seven trials that used exercise training as the sole intervention. Specifically, the odds ratio (proportion of events in intervention group divided by proportion of events in control group) for cardiovascular mortality comparing intervention with control patients at the 3-year followup was 0.74 in the multifactorial intervention compared with 0.85 in the exercise-only trials.53 Most studies involved principally male patients younger than 65 years old following myocardial infarction and excluded high-risk complex patients, thereby limiting the generalizability of the data. The percentage of females, when enrolled, was 20 percent or less. Furthermore, subsequent to the research studies cited as scientific evidence, the reduction in mortality was increased by nonrehabilitation interventions such as myocardial revascularization procedures and newer pharmacologic agents that have far more powerful effects on survival. The randomized controlled trials that reported mortality rates included a total of 7,063 patients. In no trial was the rate of fatal events greater in the intervention group than in the control group. Information obtained from two large surveys provided retrospective data from cardiac rehabilitation program responses to questionnaires.59,65 The proportion of fatal cardiac events occurring during or immediately following exercise training was low: 1/116,400 patient-hours of participation in supervised exercise training in the 1978 report59 and 1/784,000 patient-hours in the 1986 report.65 The data from both survey reports antedate the use of contemporary risk stratification procedures and contemporary medical and surgical therapies for coronary disease and heart failure. Further analysis of the more recent rehabilitation program survey data documented that 16 of 20 cardiac arrest patients were categorized as high risk by standard risk stratification methods and that three of the four non-high-risk patients were in programs with intermittent ECG monitoring. Only one patient was reported to exercise above the recommended exercise training level.65 However, in one randomized controlled trial patients who experienced cardiac arrest had been documented to exceed their prescribed training heart rate.37 No mortality data were reported by gender or patient age, nor was definitive information available regarding the effect of levels of supervision and ECG monitoring of exercise training. Pathophysiologic Measures Extent of Coronary Atherosclerosis RECOMMENDATION Cardiac rehabilitation exercise training as a sole intervention does not result in regression or limitation of progression of angiographically documented coronary atherosclerosis. When combined with intensive dietary intervention-with and without lipid-lowering drugs-exercise training may result in regression or limitation of progression of angiographically documented coronary atherosclerosis. Scientific Evidence (Strength of Evidence for Lack of Efficacy of Exercise Training Only A; Strength of Evidence for Efficacy of Multifactorial Intervention = = B) The scientific data that address the effect of exercise training on coronary atherosclerosis are based on five randomized controlled clinical trials,22,37,39,41,155 one nonrandomized controlled study,251 and three observational studies,83,186,193 All of these studies involved patients with cardiac catheterizationdocumented coronary disease. Ascertainment of regression or progression was based on serial coronary angiography. In the earlier studies, assessment was by visual inspection of the coronary angiogram. The more recent studies reported digitally processed coronary angiographic findings.22,37,39,41,251 One of the five randomized controlled trials, 155 which used exercise training as the sole intervention, showed that atherosclerosis had a "tendency to progress," but there was no significant difference between the exercise and control populations. Two randomized controlled trials of multifactorial rehabili |