Treadmill Test - Early Exercise Program After Myocardial Infarction

Transcription

1 Treadmill Test Responses to an Early Exercise Program After Myocardial Infarction: A Randomized Study ERIKA S. SIVARAJAN, R.N., M.A., ROBERT A. BRUCE, M.D., BRONWEN D. LINDSKOG, R.P.T., M.S., MARY JANE ALMES, M.S., LAURENT BELANGER, M.D., AND BERT GREEN, M.D. SUMMARY The effects of an exercise program started early after myocardial infarction and the added effects of an outpatient teaching-counseling program were studied. At random, 84 patients were allocated to a control group (A), 88 patients to an exercise group (Bi) and 86 patients to an exercise and teaching-counseling group (B2). The same exercise program was prescribed for patients in groups Bi and B2 and was started about 4.5 days after myocardial infarction and continued for 3 months. The outpatient teaching-counseling program consisted of eight group sessions pertaining to risk factor reduction and psychosocial adjustment to myocardial infarction. A low-level treadmill test and an exercise test were performed at 3 months and the exercise test was repeated at 6 months. The clinical, hemodynamic and electrocardiographic responses to these tests were not different among the three groups. However, by the end of 3 months, patients in group B1 and B2 reported walking greater distances than patients in group A. The incidence of morbidity and mortality was not different between the groups. No deleterious or beneficial physiologic effects of an exercise program either by itself or combined with a teaching-counseling program were demonstrated. Routine medical care and our interventions were equally effective in permitting the spontaneous hemodynamic improvements after myocardial infarction. More than 3 months after myocardial infarction, the group as a whole manifested spontaneous recovery in the form of a significant decrease in resting heart rate (p < 0.001) and a significant increase in systolic and diastolic blood pressure at rest and with submaximal exercise (p < 0.001). No further improvements were observed between 3 and 6 months. IN RECENT YEARS there has been a trend toward early mobilization and discharge of patients after acute myocardial infarction.1 Evidence for the safety of this practice has been presented from controlled and uncontrolled studies.24 One study demonstrated that regular exercises started 2 months after myocardial infarction and maintained for 3 months improves hemodynamic function." Accordingly, recommendations have been made' 7 for initiation of in-hospital exercises after myocardial infarction. Although there is evidence that cardiac rehabilitation achieved by a multidisciplinary team approach reduces the incidence and severity of adverse psychosocial reactions to the acute illness,8 neither the benefits and hazards of early exercises nor their maintenance through the convalescent phase has been studied. We designed a controlled, randomized study to determine the effects of early exercises and also of teaching and counseling after myocardial infarction. Early in-hospital exercise was found not to improve low-level treadmill performance by the time of discharge.9 In this study, we sought to determine whether an exercise program started early after myocardial infarction and continued for 3 months would improve subsequent exercise performance and outcome. We also hypothesized that teaching and counseling in From the Department of Physiological Nursing, the Division of Cardiology, and the Department of Biostatistics, University of Washington, Seattle, Washington. Supported by research grant 5 ROI NU , Bureau of Health Professions, Division of Nursing, Department of Health and Human Services. Address for correspondence: Erika S. Sivarajan, R.N., Center for Health Enhancement, Education and Research, A3-307 Center for the Health Sciences, University of California, Los Angeles, Los Angeles, California Received June 15, 1981; revision accepted October 8, Circulation 65, No. 7, addition to exercise might result in greater adherence to the exercise program and further improvement in exercise performance. Methods For this prospective randomized trial, 6561 patients admitted to coronary care units in seven Seattle hospitals were screened for eligibility for study from September 1, 1977, to December 2, Of these, 1418 had confirmed myocardial infarctions by two of three criteria: a history of chest pain, electrocardiographic changes and enzymatic changes. Two hundred fifty-eight patients met all of the screening criteria and were enrolled in the study. Table 1 is a list of reasons for excluding patients from the study. A 6-month follow-up was completed May The Human Subjects Review Committee requirements for informed consent were observed. Randomization Randomization was designed so that every patient in every hospital had an equal chance to be assigned to one of three groups: A, a control group; B1, a group of patients who were provided with an exercise program in the hospital and a continued exercise program during weekly clinic appointments for the first 3 months after discharge; and B2, a group of patients who were provided with the same exercise program as the Bl group and were also given a teaching-counseling program about risk factors and emotional adjustment after discharge from the hospital. Control Group Patients in group A received conventional medical and nursing management throughout all phases of hospitalization and convalescence at home. The inhospital exercises for patients in the exercise groups,

2 EARLY EXERCISE PROGRAM AFTER MI/Sivarajan et al TABLE 1. Study Sample and Categories tember 1, 1977 to December 2, 1979 of Exclusion, Sep- Admissions 6561 Confirmed myocardial infarctions 1418; study exclusions 912 (64%)* Reason for exclusion n % Age over 70 years Prolonged complications Physical limitations Noncardiac diseases Other cardiac diseases Distance over 50 miles Communications Other reasons (e.g., massive obesity, psychological problems, etc.) More than one reason Eligible for study 36% 506 Physician refusals (16%) 80 Physician permission granted 426 Patient refusals 84% (39%) 168 Patients enrolled 61% 258 *Percent taken from total at previous step. B 1 and B2, consisted of passive and active arm and leg exercises performed in bed with a system of pulleys and weights, gradually progressing to an out-of-bed calisthenic and walking program requiring an energy expenditure of 2-3 metabolic equivalents (mets*).lo 12 On the day before discharge from the hospital, a lowlevel treadmill test was performed to evaluate the effects of in-hospital exercise and to form the basis for further exercise prescription after discharge. Exercise Groups The outpatient exercise program was identical for the patients in groups Bl and B2. It consisted of a gradually progressive calisthenic and walking program prescribed at weekly 30-minute clinic visits and performed by the patient at home. Patients were instructed to exercise twice a day until they returned to work and once a day thereafter. They were taught to monitor themselves during the exercises for adverse symptoms such as dizziness, shortness of breath and chest discomfort, and signs such as excessive heart rate increase, as taught during hospitalization. The specific calisthenics for upper-body musculature and the walking program that were given to the patient have been described.10 If the patient was symptomfree, the prescription was gradually increased over the succeeding weeks to add calisthenics of increasing intensity. Patients practiced new calisthenics in the clinic to familiarize themselves and to allow the staff *A met, or metabolic equivalent, is the energy expended while sitting quietly in a chair, and corresponds to ml of oxygen per kilogram of body weight per minute. to observe their cardiovascular responses. The distance and time (or rate) of walking were gradually advanced. The exercise program progressed on an individual basis, varying with the age, physical status, current clinical status, low-level treadmill test responses, current responses to exercises and the patient's motivation and preference for the type of exercise. Patients were given tables showing energy expenditures in mets for various activities and exercises to help them choose activities within the energy expenditure prescribed for them. Exercise and Teaching-Counseling Because the quantity, quality and structure of patient teaching efforts at the different hospitals participating in the study varied widely, and because there is evidence that patients have limited recall of teaching done during hospitalization, the teaching-counseling program was offered only after discharge from hospital. Patients in group B2, in addition to receiving exercise prescriptions as described above, attended a series of eight 1-hour group sessions during weekly clinic visits. The sessions emphasized the practical aspects of anatomy and physiology of the heart, coronary artery disease, myocardial infarction and medications; risk factors, including smoking, hypercholesterolemia, hypertension, stress and sedentary living; nutritional aspects of fats, cholesterol, salt and alcohol; activities and exercises; emotional reactions to myocardial infarction in patients and their families; resumption of sexual activity; and issues concerning return to work or, if retired, to an alternative, meaningful lifestyle. Follow-up Evaluations The maximum level of activity before myocardial infarction was assessed for each patient by individually questioning the patients about the most strenuous activities at home, at work and during recreation. This constituted the baseline activity data. Progression of activity was documented for all patients at weekly intervals using an activity summary questionnaire13 that consisted of 12 activities representing aspects of daily living, recreation and exercises. Patients in the group A mailed their responses to the questionnaire each week. This activity summary was collected from patients in groups B 1 and B2 at their weekly clinic visits. The activity level was converted into mets. Follow-up evaluation of all patients was performed 3 and 6 months after discharge from the hospital at a centrally located clinic. It included a low-level treadmill test at the 3-month visit and a symptom-limited treadmill test at the 3- and 6-month visits. All subsequent admissions to the hospital, except those required solely for performing cardiac catheterization, were considered morbidity events. For the follow-up evaluation, mortality was defined as a death that occurred within 6 months after enrollment into the study. The low-level treadmill test protocol consisted of

3 1422 CI RCULATION VOL 65, No 7, JUNE 1982 four stages, each lasting 3 minutes. The level of each treadmill stage was about 2.0, 2.5, 3.0, and 4.0 mets. This protocol has been described in detail.9' 10, 14 At least 30 minutes elapsed between completion of the low-level treadmill test and the symptom-limited maximal treadmill test using the Bruce protocol"' at the 3-month follow-up visit. Symptoms and signs were continuously observed and recorded. Blood pressure was measured by the arm cuff method using an aneroid sphygmomanometer and a stethoscope for auscultation."' Heart rate, arrhythmias, ST-segment displacement, and QRS morphology were monitored with a single CB, ECG lead system. Arrhythmias were recorded at rest, exercise and recovery and are reported as present or absent. ST-segment displacement was considered significant when displacement was at least 1 mm below or above the PR segment (at second after the J point), which was chosen as the isoelectric line. Blood pressure and ECG were recorded at 1-minute intervals throughout the test and for 6 minutes into recovery. Functional aerobic impairment from the symptom-limited treadmill tests at 3 and 6 months was calculated from age and sex using the Bruce protocol nomogram (sedentary activity level).'5 The hemodynamic responses to the low-level treadmill test at the 3-month visit were compared with those to the low-level treadmill test at the time of discharge. Similarly, hemodynamic responses to the exercise test at 6 months were compared with those to the exercise test at 3 months. Analysis of variance and an unpaired t test were used to detect differences between groups cross-sectionally and the paired t test was used to compare differences within each group over time. Differences in proportions were tested by chi-square analysis. Differences were considered significant if p < Results Seven hospitals provided the study sample. Fortyseven percent of the patients came from health maintenance organizations, 40% from private hospitals and 13% from teaching hospitals. The mean age (± SD) of the patients was 57.1 ± 7.3, 55.6 ± 9.3 and 56.3 ± 8.3 years in groups A, Bl and B2, respectively. More than 80% of the patients in each group were male, Caucasian and married. The median years of education and income, work status, and Hollingshead socioeconomic index17 were similar in the three groups. Twenty-three percent of the patients in group A and 12% in groups Bl and B2 reported a history of myocardial infarction (NS). Patients in group A reported a higher prevalence of a family history of coronary artery disease (60%) than patients in group B1 (48%) or group B2 (46%) (NS). Risk factors for coronary artery disease and events in the medical history of significance were similar in all three groups. Electrocardiographic location of infarction was equally distributed among the three groups. Killip classifications for severity of illness'8 were also similar, as were complications during initial hospitalization.9 Patients in the A, Bl and B2 groups were discharged an average 10.3 ± 2.7, 10.7 ± 3.7, and 10.1 i 2.7 days after admission to the hospital. Attrition Attrition due to withdrawal, cardiac surgery or death at different stages of the study is shown by group in table 2. Total attrition at the end of 3 months was 15.8% and at the end of 6 months, 25%. The rates of attrition in the three groups at the end of 6 months were comparable. None of the patients in group A, two patients in group B1 and four in group B2 had cardiac surgery during initial hospitalization. After discharge, eight patients in group A, two in group Bl and three in group B2 had cardiac surgery. Exercise Program Results Patients in group A reported walking greater distances once a day at least three times a week immediately upon discharge from hospital compared with patients in groups Bl and B2 (fig. 1). However, by 3 months, patients in groups Bl and B2, who received weekly exercise prescriptions, walked significantly (p < 0.001) greater distances at least three times a week (2.4 and 2.2 miles, respectively) than patients in group A (1.5 miles). Also by the end of 3 months, patients in groups B1 and B2 had exceeded their preinfarction level of activity, whereas patients in group A, who had had a higher met level than group B1 and B2 patients before their myocardial infarction, were below their own preinfarction levels of activity (table 3). To compare the responses to the low-level treadmill test at discharge and at 3 months and to the exercise test at 3 months and 6 months, only data from patients who performed the tests at both times were pooled, so that the differences within each group over time could be assessed. Because propranolol modified hemodynamic response to exercise,19 responses of patients in each group who were not receiving propranolol were analyzed separately and were found to be not significantly different from the responses of patients who were not receiving propranolol. Data for patients in each group and across time are presented in tables 4-9. Response to Low-level Treadmill Test Clinical and electrocardiographic responses to the low-level treadmill tests are shown in table 4. There were no significant differences among the groups in the number of patients who finished the test or in the prevalence of chest pain, fatigue, excessive heart rate increase, arrhythmias or ST-segment displacement either at the time of hospital discharge or at 3 months. The hemodynamic responses to the low-level treadmill test at hospital discharge and at 3 months are shown in table 5. There were no differences among the groups in the hemodynamic responses to either lowlevel treadmill test. The differences in hemodynamic responses between the low-level treadmill test at hospital discharge and low-level treadmill test at 3 months were examined for the total group, since there were no differences

4 EARLY EXERCISE PROGRAM AFTER MI/Sivarajan et al TABLE 2. Attrition Due to Withdrawal, Cardiac Surgery or Death During 6 Months of Follow-up Exercise/ Control Exercise teaching group A group BI group B2 Total n % n % n % n.% Enrollment Withdrawal Surgery Death I Discharge Withdrawal Surgery Death Three-month follow-up Withdrawal Surgery Death I Six-month follow-up Cumulative attrition Withdrawal Surgery Death All reasons between these tests in each group (table 6). Systolic and diastolic blood pressures at rest and at the start of recovery were significantly higher (p < 0.01) during the low-level treadmill test at 3 months than the test at hospital discharge. Heart rate at rest was significantly lower (p < 0.001) at the time of the low-level treadmill test at 3 months than during the test at hospital discharge; heart rate at the start of recovery in the two tests was not significantly different. Of the patients who had a change in clinical responses between the hospital and the 3 months low-level treadmill test, the proportion of patients who completed the test was significantly greater (p < ) at 3 months; the a) y~.e a) Y a) c E. a) U) I-.e -a)c (nq). m E >, E -0 x X.c Control (A) n = 58 Exercise (B1) n = 75 3 Exercise/Teaching (B2) n = L 1-. A, e!-. *_ - *, Q Weeks after discharge from hospital at 12 weeks at 24 weeks A vs Bl px.001 pt.04 A vs B2 p- 01 N.S. Bl vs B2 N.S. N.S. FIGURE 1. Progression of walking over the first 3 moni ths after hospital discharge and at 6 months. - proportion who showed no ST-segment elevation was significantly lower (p < 0.003) (table 7). Conversely, the proportion of patients who had arrhythmias at 3 months was significantly greater (p < 0.02) (table 7). Responses to Exercise Tests Clinical and ECG responses of all groups to the exercise tests at 3 and 6 months are shown in table 8. There were no differences among the groups. ST-segment elevation disappeared at 6 months in the group A patients, but there were no significant differences in group Bl and B2 patients. The hemodynamic responses during the exercise test at 3 months and at 6 months are shown in table 9. There were no differences among the groups or within each group between the tests. There were no significant differences in functional aerobic impairment among the groups or within each group between the tests (table 10). Morbidity and Mortality Data Total events (including mortality) were similar in all three groups. Cardiac events were also evenly distributed across the three groups. One patient in group A TABLE 3. Activity Level Before and After Myocardia Infarction Exercise/ 24 Control Exercise teaching group A Met level group BI Met level group B2 Met level n Median n Median n Median Before MI months months

5 1424 CI RCULATION VOL 65, No 7, JUNE 1982 TABLE 4. Clinical and Electrocardiographic Responses to Low-level Treadmill Test by Groups, In-hospital and at 3 Months Control Exercise Exercise/ teaching group A group BI group B2 (n=61) (n=72) (n=63) H 3M H 3M H 3M Completed test 51%o 74% 60%/o 75% 67% 81% Reasons for stopping Chest pain Fatigue Heart rate increase > 120 beats/ min ECG changes Arrhythmias ST-segment elevation > 1 mm ST-segment depression. 1 mm Abbreviations: H = in hospital; 3M = at 3 months. died from metastatic cancer. All other deaths were due to cardiac causes and occurred within the first 3 months. The cardiac mortality rate in group A was 2% (one of 65); in group B 1, 4% (three of 71); and in group B2, 5% (three of 65). These rates were not significantly different. Discussion Our sample was drawn from seven Seattle hospitals that represented a wide range of cardiology practices. Nevertheless, it was a select sample; only 18% (258 patients) of all patients who had confirmed myocardial infarctions were admitted to the study. The TABLE 5. Hemodynamic Responses to Low-level Treadmill Test by Groups, In-hospital and at 3 Months Exercise! teaching Control group A Exercise group BI group B2 (n=61) (n=72) (n=63) H 3M H 3M H 3M Systolic blood pressure (mm Hg) Rest ± ± ± 18 Stage ±20 139± ± 19 0 recovery ± ± ± 19 3 recovery ± ± ± ± 20 6 recovery ± ± ± ± 18 Diastolic blood pressure (mm Hg) Rest ±9 79 ± ± 9 83 ± 10 Stage 1 73 ± ± ± ± ± ± 10 O recovery 71 ± ± ± ± ± ± 10 3recovery 73±9 81 ±11 76±9 84±10 74±9 85±10 6recovery 73±9 81 ±11 76±9 83±11 75±8 83±10 Heart rate (beats/ min) Rest 75 ± ± ± ± ± ± 11 Stage 1 92 ± ± ± ± ± ± 11 0 recovery 99 ± ± ± ± ± ± 13 3 recovery 81 ± ± ± ± ± ± 12 6 recovery 78 ± ± ± ± ± ± 12 Rate pressure product (SBP X HR 100) Rest 78 ± ± ± ± 19 Stage I 111±30 116±23 113±27 122±29 108±22 115±23 0 recovery 125 ± ± ± ± ± 30 3 recovery 92 ± ± ± ± ±23 6 recovery 86 ± ±20 86±20 92±21 85 ± ±21 0 recovery = at the end of exercise.

6 EARLY EXERCISE PROGRAM AFTER MI/Sivarajan et al TABLE 6. Statistical Comparison of Differences in Hemodynamic Response to the Low-level Treadmill Test from Hospital to 3 Months for the Total Group Response of total group -n Mean ± SD P Systolic blood pressure at rest <0.001 Systolic blood pressure at0 recovery ±20.9 <0.001 Diastolic blood pressure at rest <0.001 Diastolic blood pressure ato recovery ± 13.4 <0.001 Heart rate at rest ± 12.4 <0.001 Heart rate ato recovery NS 0 recovery = end of exercise. selectivity was due to the screening criteria, which excluded patients who were older than 70 years of age, physically limited or who had severe systemic diseases, complications of myocardial infarction longer than 7 days, and who did not choose to be enrolled in the study. These criteria specifically excluded patients who were perhaps more in need of rehabilitative assistance than were those selected. Group assignments were made using a random number table. Nevertheless, small differences emerged among the groups. Patients' in group A had almost twice the pr'evalence of'prior myocardial infarction than patients in the other two groups. However, this apparently had no influence on the Killip prognostic index, duration of stay in the coronary care unit or hospital, or complications during hospitalization. Thus, the three groups were clinically similar and comparable. Attrition' was also similar among the three groups. This rate of compliance is better than that reported in other studies.2>22 The exercises prescribed for patients in groups B1 and B2 were performed at'home, and hence-could not be directly observed by the research staff. Compliance with exercise prescription was assessed from' the patients' answers to the activity summary questionnaire. Wilhelmsen et al.21 and others22 23 have suggested that'exercise programs that require patients to travel to the exercise facility, 'such as community centers, have greater noncompliance. The high compliance in our study tends to confirm this suggestion. Performance during the low-level treadmill test at TABLE 7. Statistical Comparison of Changes in Clinical Responses to Low-level Treadmill Test Between Hospital and 3 MAonths Observed Observed at hospital at 3 but not months at 3 but not month at hospital Total group LLTT LLTT- p Completion of test < Arrhythmias < 0.02 ST-segment elevation. 1 mm 21 5 <0.003 *Hypothesis test of difference in proportions in columns 1 and 2, meaning proportion I = proportion 2 = 0.50; i.e., if a patient changed status, the change was as likely to be a regression as an improvement. Abbreviation: LLTT = low-level treadmill test. the time of hospital discharge showed the three groups to be similar. Results of the low-level treadmill test and exercise test at 3 months showed that the exercise program did not have a conditioning effect on patients in either group Bl and B2. The influence of propranolol on the conditioning effect of the exercise program could not be discerned, because even the subset of patients in groups B 1 and B2 not receiving propranolol failed to show improvement in performance. Improvement in clinical and hemodynamic response to exercise stress during the first 3 months was evident in all'three groups when performance on the low-level treadmill test was compared with performance at discharge. This is attributable to spontaneous recovery from bedrest and left ventricular dysfunction and confirms recovery trends reported by Wohl and associates.24 The decreased prevalence of ST-segment elevation during the low-level treadmill test at 3 months is similar (table 7) to the observations made by Atterhog et al.25 using a symptomlimited test 6 months after myocardial in-farction and can also be attributed to recovery. The exercise program was initiated on about the fourth day after myocardial 'infarction. To our knowledge, no other controlled randomized study of exercise intervention has been initiated this early. Considering the unprecedented institution of exercise this early, the intensity was kept low, perhaps too low to produce a conditioning effect. DeBusk- et al.26 reported the effects of 8 weeks of exercise training on a TABLE 8. Months Clinical and Electrocardiographic Responses to Treadmill Test by Groups, at 3 Months and at 6 Control Exercise Exercise/ teaching group A group Bl group B2 (n=51) (n=61) (n53) 3M 6M 3M 6M 3M 6M Chest pain 27% 20% 23% 19% 26% 19% Arrhythmias ST-segment elevation > 1 mm ST-segment depression. 1 mm Abbreviations: 3M = at 3 months; 6M = at 6 months.

7 1426 CI RCULATION VOL 65, No 7, JUNE 1982 TABLE 9. Hemodynamic Responses ( Mean ± SD) to Low-level Treadmill Test by Groups, at 3 Months and 6 Months Exercise/ teaching Control group A Exercise group BI group B2 (n = 51) (n = 61) (n = 53) 3M 6M 3M 6M 3M 6M Systolic blood pressure (mm Hg) Rest Stage I Recovery 0 minute I minute 3 minute 5 minute Diastolic blood pressure (mm Hg) Rest Stage I Recovery 0 minute I minute 3 minute 5 minute Heart rate (beatsl min) Rest Stage I Recovery 0 minute I minute 3 minute 5 minute Rate-pressure product/ 100 Rest Stage I Recovery 0 minute 1 minute 3 minute 5 minute 0 recovery = at the end of exercise. Abbreviations: 3M = at 3 months; 6M = at 6 months. 121 ± ± ± ± ± ± ± ± ± ± ± ± 18 81±10 82±11 82 ± ±9 81± ± ± ± ± ± 9 84 ± 9 81 ± 8 71±14 67± ± ± ± ± ± ±27 87 ± ± ± ± 15 86±20 83± ± ± ± ± ± ±62 132±43 136± ± ±30 very select group of patients who were stratified by presence or absence of ischemic responses to a symptom-limited treadmill test 3 weeks after myocardial infarction and were then randomized into a gymnasium-trai'ned group, a' home-trained group (only those without ischemic response), and a control group that received no training. The exercise intensity produced heart rate responses 70-85% of peak values attained during the symptom-limited treadmill test. The highly select group studied by DeBusk et al.,2 performing higher intensity exercises than did patients in our study, demonstrated no statistically significant improvement in functional capacity over that of the control group that recei'ved no training. Only the subgroup of patients without exercise-induced 124 ± ± ± ± ± ± ± ± ± ± ± ± ± 8 83 ± 8 80 ±9 83 ±9 80± 10 80± ± ± 10 82±8 83±9 83 ±8 83 ±8 69± 10 68± ± ± ±22 144± ± ±22 83 ± ± 18 82±12 85±16 85 ± ± ± ± ± ± ±51 196± ± ± ±24 116± ± ± ± ± ± ± ± ± ± ± ± ± 16 86±10 83±9 84±12 82±12 80 ± ± ± ± 12 84±9 82±10 85 ± ± 9 70 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±38 159± ± ± ± ± ± ± ± ± 21 ischemia (ST-segment depression or angina pectoris) showed significant improvement in functional capacity over the control group. The authors concluded that formal training may not be necessary to restore functional capacity after uncomplicated myocardial infarction. Other randomized studies in which higher intensity exercises were introduced at a later stage in the convalescent phase for a longer duration than our study also failed to show significant improvement in functional performance In contrast to these studies, Cunningham' et al.81 reported that h'ighintensity endurance training for '2 years after myocardial infarction, compared with low-intensity training, resulted in a significant decrease in heart rate response to exercise.

8 EARLY EXERCISE PROGRAM AFTER MI/Sivarajan et al TABLE 10. Functional Aerobic Impairment by Groups, at 3 Months and at 6 Months After Hospital Discharge Control Exercise Exercise/ teaching group A group BI group B2 (n = 51) (n = 61) (n = 53) 3M 6M 3M 6M 3M 6M FAI(%) 18.1 ± ± ± ± ± ±21 Change in FAI from 3 to 6 months -2.6 ± ± ± 15 Abbreviations: FAI = functional aerobic impairment; 3M = 3 months; 6M = 6 months. Our activity summary questionnaires revealed that patients in group A performed exercises on their own initiative, so much so that exercise levels reported by patients in groups B 1 and B2 did not exceed those of patients in group A until about 6 weeks after hospital discharge (fig. 1). This might have been due to the high level of awareness in this community, where cardiac rehabilitation programs have had wide public support for the past 10 years. Under such circumstances, a structured, low-intensity exercise program might not distinguish the control group from the experimental groups. Exercise prescriptions were stopped at 3 months. Patients in the control group had a smaller decrease in the maximum distance walked between 3 and 6 months, the period during which intervention had stopped in groups Bl and B2. Patients who initiated walking without formalized prescription may have been more motivated to maintain their exercise than patients who received a weekly exercise prescription. This might explain why patients in the control group showed the most improvement in functional capacity after intervention had ceased. The exercise program did have some influence, as evidenced by significantly greater exercise levels reported by patients in groups B 1 and B2 than by patients in group A at the end of 3 months (fig. 1). Patients in group A reported higher levels of activity immediately upon discharge from hospital than did patients in groups Bi and B2. This raises an interesting question of whether the general awareness of the risks of inactivity and benefits of exercise induces patients who do not receive specific exercise prescriptions to do more on their own. If so, exercise prescriptions might serve the useful, cautionary purpose of preventing overexertion and its complications. One tenet of rehabilitation is to return patients to preinfarction levels of activity or, if this is not possible, to return them to a healthier lifestyle. By this measure, the exercise program prescribed for the patients in groups Bi and B2 achieved the purpose within the duration of the follow-up. By the end of 3 months, patients in groups B 1 and B2 reported higher (though not significant) activity levels than before their myocardial infarction, whereas patients in group A were still below their preinfarction levels of activity. Although no beneficial physiologic effects could be discerned, there were no complications directly attributable to the exercise program, since total events were similar in all three groups. Patients who required cardiac surgery during initial hospitalization were in groups Bl and B2, and a greater proportion of patients requiring cardiac surgery after discharge from the hospital were in group A. This suggests that the inhospital exercise program may have helped to identify patients in the B 1 and B2 groups who required cardiac surgery. Such patients in group A were identified only after discharge from hospital, when an increase in activity was imposed on them by the demands of daily living. Although the incidence of cardiac mortality among patients in B1 and B2 groups was twice that of patients in group A, this difference was not significant and was quite low. Moreover, significance of mortality rates cannot be accurately assessed when the sample is small, the duration of follow up is short (6 months), and the incidence of mortality itself is low.20 The National Exercise and Heart Disease Project32 and the interim report by Rechnitzer33 have not shown convincing evidence that a program of supervised exercises significantly decreases mortality and morbidity. This study was also designed to measure the effects of teaching-counseling alone on rehabilitation outcomes. When activity levels and hemodynamic responses to exercise stress in patients in group B2 were compared with those of patients in group B 1, no differences could be discerned, leading us to conclude that teaching-counseling offers no physiologic benefits. The teaching-counseling program did significantly improve patients' knowledge about their medications. In conclusion, routine medical care and our interventions were equally effective in permitting the spontaneous hemodynamic improvement after myocardial infarction. No clinical or physiologic benefit or deleterious effect of structured, formal, early, lowintensity exercises could be demonstrated. Thus, it is difficult to justify the cost associated with structured, formalized exercise programs. Acknowledgment We thank the following research staff and consultants for their efforts: Marjorie Anderson, R.N., M.A., Janet Carriveau, B.S.N., Theresa Kempf, B.S.N., Kathleen Kominski, R.N., M.N., Martha Livingston, B.S.N., Katherine M. Newton, R.N., M.A., Martha Shively, R.N., M.A., and Sandra Solack, R.N., M.S.N., cardiac nurse specialists; Maryann Savina, O.P.T., M.S., occupational therapist; L. John Clarke, M.D., and Kenneth Hossack, M.B.B.S., cardiology fellows; Peggy Goforth, program assistant; Rosaria Fontanilla, data control technician; William Owen M.S., statistician; and Timothy DeRouen Ph.D., senior biostatistician. We are greatly indebted to the many nurses and physicians at the participating hospitals for their superb cooperation, especially the physicians who helped to coordinate these efforts: John R. Blackmon, M.D., John G. Doces, M.D., Phillip L. Hall, M.D., Theodore Hegg, M.D., J. Ward Kennedy, M.D., Robert M. Levinson, M.D., John A. Murray, M.D., Simeon A. Rubenstein, M.D., and Nathaniel Wagner, Ph.D. (deceased). We thank Alison Ross

9 1428 CIRCULATION VOL 65, No 7, JUNE 1982 for her editorial assistance. We are indebted to Louise W. Mansfield, R.N., M.A., for her pioneering work in cardiac rehabilitation and for her continued guidance and support. References 1. U.S. Department of Health, Education and Welfare: Workshop on early hospital discharge of patients with uncomplicated acute myocardial infarction. NHLBI, NIH publ. no , Tucker H, Carson PHM, Bass NM, Sharratt GP, Stock JPP: Results of mobilization and discharge early after myocardial infarction. Br Med J 1: 10, Bloch A, Maeder JP, Maissly J: Early mobilization after myocardial infarction: a controlled study. Am J Cardiol 34: 152, Abraham AS, Sever Y, Weinstein M, Dollberg M, Manczel J: Value of early ambulation in patients with and without complications after acute myocardial infarction. N Engl J Med 292: 719, Detry JMR, Rousseau M, Brasseur LA: Early hemodynamic adaptations to physical training in patients with healed myocardial infarction. Eur J Cardiol 2: 307, Haskell WL: Physical activity after myocardial infarction. Am J Cardiol 33: 776, American Heart Association Coronary Care Committee of the Council on Clinical Cardiology and the Committee on Medical Education: Coronary Care: Rehabilitation After Myocardial Infarction. Dallas, American Heart Association, Pozen MW, Stechmiller JA, Harris W, Smith S, Fried DD, Voigt GC: A nurse rehabilitator's impact on patients with myocardial infarction. Med Care 15: 830, Sivarajan ES, Bruce RA, Almes MJ, Green B, Belanger L, Newton KM, Mansfield LW: In-hospital exercise after myocardial infarction does not improve treadmill performance. N Engl J Med 305: 357, Sivarajan ES, Snydsman A, Smith B, Irving JB, Mansfield LW, Bruce RA: Low-level treadmill testing of 41 patients with acute myocardial infarction prior to discharge from the hospital. Heart Lung 6: 975, Sivarajan ES, Lindskog B, Savina M, Mansfield LW, Bruce RA: Exercise protocol for early cardiac rehabilitation. Early mobilization and testing, significance of arrhythmias after myocardial infarction. In Council on Rehabilitation of the International Society and Federation of Cardiology, Proceedings from Symposium, edited by Konig K. Waldkirch, Freiburg, 1979, pp Kempf TM, Halpenny CJ: Oxygen uptake and hemodynamic responses during in bed arm and leg exercises. (abstr) Circulation 62 (suppl III): III-219, Lindskog BD, Sivarajan ES: A method of evaluation of activity and exercise in a controlled study of early cardiac rehabilitation. J Cardiac Rehab 2: 156, Mansfield LW, Sivarajan ES, Bruce RA: Exercise testing of myocardial infarction patients prior to hospital discharge: a quantitative basis for exercise prescription. Cardiac Rehab 8: 17, Bruce RA, Kusumi F, Hosmer D: Maximal oxygen intake and nomographic assessment of functional aerobic impairment. Am Heart J 84: 546, American Heart Association. Recommendations for human blood pressure determination by sphygmomanometers. Dallas, American Heart Association, Miller DC: Handbook of Research Design and Social Measurement, 3rd ed. New York, Longman, 1977, pp Killip T, Kimball JT: Treatment of myocardial infarction in a coronary care unit: a two year experience with 250 patients. Am J Cardiol 20: 457, Hossack KF, Bruce RA, Clarke LJ: Influence of propranolol on exercise prescription of training heart rates. Cardiology 65: 47, Sanne H, Elmfeldt D, Grimby G, Rydin C, Wilhelmsen L: Exercise tolerance and physical training of non-selected patients after myocardial infarction. Acta Med Scand (suppl 551): 1, Wilhelmsen L, Sanne H, Elmfeldt D, Grimby G, Tibblin G, Wedel H: A controlled trial of physical training after myocardial infarction. Prev Med 4: 491, Oldridge NB: Compliance of post myocardial infarction patients to exercise programs. Med Sci Sports 11: 373, Andrew GM, Parker JO: Factors related to dropout of post myocardial infarction patients from exercise programs. Med Sci Sports 11: 376, Wohl AJ, Lewis HR, Campbell W, Karlsson E, Willerson JT, Mullins CB, Blomqvist CG: Cardiovascular function during early recovery from acute myocardial infarction. Circulation 56: 931, Atterhog JH, Ekelund LG, Kaijser L: Electrocardiographic abnormalities during exercise 3 weeks to 18 months after anterior myocardial infarction. Br Heart J 33: 871, DeBusk RF, Houston N, Haskell W, Fry G, Parker M: Exercise training soon after myocardial infarction. Am J Cardiol 44: 1225, Kallio V, Hamalainen H, Hakkila J, Luurila OJ: Reduction in sudden deaths by a multifactorial intervention program after acute myocardial infarction. Lancet 2: 1091, Kentala E: Physical fitness and feasibility of physical rehabilitation after myocardial infarction in men of working age. Ann Clin Res 4 (suppl IX): 1, Palatsi I: Feasibility of physical training after myocardial infarction and its effect on return to work, morbidity and mortality. Acta Med Scand (suppl 599): 1, Saunamaki KI: Feasibility and effect of physical training with maximum intensity in men after acute myocardial infarction. Scand J Rehab Med 10: 155, Cunningham DA, Ingram KJ, Rechnitzer PA: The effect of training: physiological responses. Med Sci Sports 11: 379, Shaw LW: Effects of a prescribed supervised exercise program on mortality and cardiovascular morbidity in patients after a myocardial infarction. The National Exercise and Heart Disease Project. Am J Cardiol 48: 39, Rechnitzer PA: The effects of training: reinfarction and death - an interim report. Med Sci Sports 11: 382, 1979