Physical function by age and sex. Asterisk indicates P<.01 compared with women; dagger, P<.01 compared with older patients.
Ades PA, Maloney A, Savage P, Carhart RL. Determinants of Physical Functioning in Coronary PatientsResponse to Cardiac Rehabilitation. Arch Intern Med. 1999;159(19):2357-2360. doi:10.1001/archinte.159.19.2357
Coronary heart disease is a major cause of disability. There has been little study of the effect of cardiac rehabilitation on measures of self-reported physical functioning.
Physical functioning was measured using the Medical Outcomes Study–Short-Form Questionnaire. Determinants of physical function were analyzed in 303 patients with coronary heart disease. Response of physical function to a 3-month exercise rehabilitation program was then determined.
At baseline, women had lower physical function scores than men, despite similar age and diagnostic distribution. Older patients had lower physical function than younger patients. Aerobic exercise capacity, leg and arm strength, and comorbidity and depression scores were all significant predictors of baseline physical function (r range, 0.46 to −0.22). Physical function score increased significantly after exercise conditioning, with a mean (±SD) overall score increase from 66 ± 23 to 80 ± 20 on a scale of 0 to 100. The best baseline determinant of a favorable physical function improvement after rehabilitation was a low baseline physical function score. The best training-related correlate of improved physical function score was a decrease in mental depression score.
Self-reported physical function in coronary patients is related to age, sex, fitness, and mood state. Physical functioning improves after cardiac rehabilitation in all age, sex, and diagnostic groups, but particularly in patients with low baseline values. These data support the concept that cardiac rehabilitation effectively prevents and treats cardiac disability.
THE EFFECTS of exercise-based cardiac rehabilitation on measurable health outcomes in coronary patients, including morbidity and mortality, exercise capacity, symptoms, and coronary risk factors, have been studied intensively.1- 4 However, despite the fact that coronary heart disease is a major cause of premature disability and of disability in the elderly,5,6 there has been little study of the effect of cardiac rehabilitation on measures of physical functioning.
In our study, self-reported physical function status, a major component of quality of life, was measured using the physical function section from the Medical Outcomes Study–Short-Form Questionnaire.7 This questionnaire describes whether patients are physically limited during a range of physical activities. These range from very strenuous (ie, running or lifting), to moderate (ie, carrying groceries or climbing a single flight of stairs) and low level (ie, bathing or dressing).
We analyzed baseline physical functioning in a population of patients with coronary heart disease entering a cardiac rehabilitation program and determined the subsequent response of physical function score to exercise rehabilitation. From a practical point of view, these data may help reorient cardiac rehabilitation training protocols such that improvements in physical functioning are maximized. They may also prove useful for patients and cardiac rehabilitation professionals to predict which patients are most likely to improve physical functioning with rehabilitation.
The study population at baseline consisted of 303 patients with coronary heart disease referred to the cardiac rehabilitation program at the University of Vermont Fletcher-Allen Health Care, Burlington, from January 1, 1997, through June 30, 1998. All patients had recently been hospitalized with a coronary event a mean (± SD) of 7 ± 4 weeks before entry into the study. All patients were living independently in the community. Index diagnosis was coronary bypass grafting in 121 (39.9%), myocardial infarction in 84 (27.7%), percutaneous angioplasty (without infarction) in 64 (21.1%), unstable angina in 25 (8.2%), and congestive heart failure in 9 (3.0%). Mean (± SD) age of the study population was 60.6 ± 11.3 years (range, 20-87 years). Male patients constituted 74.9% (n = 227) of the study.
Before beginning the exercise rehabilitation program, all patients had a session of baseline data collection that included the following: (1) the Physical Function assessment from the Medical Outcomes Study–Short-Form Questionnaire scaled 0 to 100, where 100 signifies excellent physical function7; (2) the Geriatric Depression Questionnaire8 with measurement of a depression score scaled 0 to 15, where the higher number signifies more depressive symptoms; (3) a symptom-limited treadmill exercise test performed according to a modified Balke protocol, with determination of maximal work capacity (in metabolic equivalents of multiples of resting oxygen uptake [METS]) and collection of expired gas for measurement of peak aerobic capacity (in peak oxygen consumption [V̇O2]), performed with patients taking their usual medications; (4) strength measures by determination of a single-repetition maximal lift for bench press (upper body) and leg extension (lower body); and (5) a comorbidity score by assessment for the presence of diabetes, peripheral vascular disease, cerebrovascular disease, chronic obstructive lung disease, and arthritis. If a comorbid condition was present, it was quantified by severity as follows: 1 indicated present but not exercise limiting; 2, present and had an impact on exercise response; and 3, exercise limiting. A total morbidity score ranging from 0 to 15 was determined for each patient.
The exercise training protocol consisted of three 1-h/wk sessions of combined aerobic and resistance exercise. In general, patients performed 30 to 45 minutes per session of treadmill walking, rowing, and cycling, with exercise heart rates maintained at 70% to 85% of maximal heart rate as determined from results of their baseline exercise test. Patients also participated in a resistance training program on a multisystem Universal Gym apparatus (Universal Inc, Iowa City, Iowa) performing 1 to 2 eight-repetition sets at an intensity of 40% to 50% of a single-repetition maximal lift for bench press, shoulder press, latissimus pulldown, leg extension, and hamstring curl. Adjustments were made for patients with comorbidities such as arthritis. Patients who underwent coronary bypass surgery began upper body resistance training no sooner than 3 months from the date of surgery. All study evaluations were repeated in patients who completed the full 3 months of exercise training.
Data were analyzed using the t test, univariate linear regression, and stepwise multivariate regression analysis. Data are presented as mean ± SD.
The study group ranged from 20 to 87 years of age, with mean ages for men and women of 60.0 ± 11.1 and 62.5 ± 12.0 years, respectively (P = .09) (Table 1). At baseline exercise testing, 21 (6.9%) were characterized by the presence of exertional angina, and 33 (10.9%) were noted to have more than 1-mm ST segment depression on echocardiography.
At baseline, women had lower physical function scores than men (57.4 ± 24.9 vs 68.6 ± 21.4; P = .008), despite no significant difference in age or diagnostic distribution between sexes (Figure 1). Index diagnosis predicted baseline higher physical function score with patients who had coronary angioplasty than with patients who had suffered a myocardial infarction (P<.05), with no other significant differences between diagnostic groups (Table 2). Time since coronary event did not predict physical function score. Patient age was related to baseline physical function score, with older patients having lower function (r = −0.22; P<.001) (Table 3 and Figure 1). Baseline maximal exercise capacity and peak O2 were the best univariate predictors of baseline physical function score (r = 0.46 and 0.40; respectively; P<.001 for both) (Table 3). Other significant predictors of baseline physical function score were strength measures of leg press (r = 0.32) and bench press (r = 0.25), comorbidity score (r = −0.23), and depression score (r = −0.22) (all Ps<.001). Body mass index (calculated as weight in kilograms divided by the square of height in meters) did not predict baseline physical function. The presence of angina or at least 1 mm ST segment depression at the baseline stress test were not predictive of baseline physical function score. By multivariate analysis, baseline maximal exercise capacity and comorbidity score were the best independent determinants of baseline physical function score (cumulative r2 = 0.26).
A total of 218 patients completed the 3-month exercise conditioning protocol. A subset of 40 patients (13.2%) completed an abbreviated conditioning program of 18 sessions and were not included in the present analysis, whereas 45 patients (14.8%) discontinued participation before completion of their planned participation.
Physical function score increased substantially in the 218 patients who completed the prescribed 3-month cardiac rehabilitation protocol, from 65.6 ± 22.8 to 80.2 ± 20.4 (P<.001) (Table 4). Associated with the increase in physical function score were associated increases in peak V̇O2 (+16%), peak exercise capacity (+50%), leg strength (+28%), and upper body strength (+17%) and a decrease in depression score (Table 4). Among the patients who completed cardiac rehabilitation, mean attendance rate (76% ± 8%) did not predict change in physical function score.
We then studied the power of baseline measures to predict change in physical function score in response to cardiac rehabilitation (n = 218). The only baseline measure that predicted change in physical function score after rehabilitation was the baseline physical function score (r = −0.51; P<.001). That is, patients with the lowest baseline physical function score were the most likely to show an improvement in this measure after rehabilitation.
Finally, we analyzed whether exercise-induced changes in variables such as fitness measures, strength measures, and depression score correlated with change in physical function after rehabilitation. Changes in several variables predicted a change in the physical function score after exercise rehabilitation (Table 5). The best predictor of improvement was change in depression score (r = 0.36; P<.001), followed by changes in leg strength (r = 0.29; P = .006), arm strength (r = 0.28; P =
.008), peak V̇O2 (r = 0.22; P = .02), and maximal exercise capacity (r = 0.17; P = .01). By multivariate analysis, the only independent predictors of improvements in physical function score were change in depression score (r2 = 0.08; P = .003) and change in maximal exercise capacity (cumulative r2 = 0.15; P = .007).
Despite the fact that coronary heart disease is a major cause of premature and all-age disability in the United States, the determinants of physical function, the inverse of disability, in this population has received little study. Furthermore, although the effect of cardiac rehabilitation exercise training on treadmill-derived measures of exercise capacity has been studied extensively,3 the relationship between treadmill-derived measures of exercise capacity and physical function in the home setting is actually unclear. In a study by Neill et al9 in middle-aged male coronary patients, there was a surprisingly poor correlation between exercise capacity and questionnaire-derived measures of physical function in the home environment. In many cases, patients consciously limited their actual performance of physical activities owing to fear that an activity was unsafe or to physician or family advice, rather than to an experience of symptoms during the activity.
In our study, we find that baseline characteristics of older age, female sex, low exercise capacity, low strength, presence of medical comorbidities, and higher depression score all predict poor physical function. After cardiac rehabilitation exercise training, patients in all age, sex, and diagnostic categories improved their perceived ability to perform physical activities in the home environment. From a clinical point of view, patients with the lowest baseline physical function scores were the patients who improved to the greatest degree after conditioning. In that the best predictor of cardiac rehabilitation participation is the strength of the referring physician's recommendation, a strong recommendation for cardiac rehabilitation participation is most indicated in patients with the lowest physical function scores, ie, the most disabled.10
The best training-induced correlates of improvement in physical function score were change in depression score, followed by change in exercise capacity. These results highlight the importance of mood state on functioning and well-being in the population tested. Results from the Medical Outcomes Study in more than 11 000 outpatients confirm that patients with depressive symptoms in the absence of depressive disorder had worse physical and social functioning and worse perceived health than patients with hypertension, diabetes, and arthritis.11 Although it is not clear in our study whether depression score improved as a function of the group exercise training program or as a function of the time elapsed since the index coronary event, there is evidence from the literature that participation in cardiac rehabilitation is associated with improvements in measures of depression compared with randomized controls.12,13
Several fitness-related factors were also related to baseline physical function status and to improved function after rehabilitation. These include measures of aerobic fitness and of strength. The recent addition of strength training to cardiac rehabilitation protocols is supported by the results of our study, particularly for patients with the lowest strength measures, ie, women and the elderly.14,15 However, our results also suggest a redirection in the design of rehabilitation programs if the goal is to diminish rates of disability in coronary patients. With a goal of maximizing physical function, attention must be focused not only on physical work capacity, but on mood, and most specifically, depressive symptoms. In addition to group exercise, patients who at baseline have a combination of low physical functioning and a high depression score should be considered for individual or group counseling. It remains to be determined if improvements in depressive symptoms and/or mood alone would lead to an improved physical performance status.
Our study has several limitations. Because patients underwent study on entry into a cardiac rehabilitation program, the study group primarily consisted of patients recently hospitalized for an acute coronary event such as coronary bypass surgery or acute myocardial infarction, with a lesser inclusion of patients with chronic angina and chronic heart failure. Cardiac rehabilitation populations are also characterized by fewer older patients, female patients, and patients in lower socioeconomic subgroups.10 The study was limited also by a lack of a nonexercising control group; thus, some gains in physical work capacity and in mood state may have occurred just with the passage of time since the acute coronary event. Nonetheless, these data focus attention on the importance of measuring the benefits of rehabilitation in terms reported by the patient, such as physical functioning, rather than by measures more familiar to physicians such as exercise testing variables or other diagnostic tests. In summary, cardiac rehabilitation effectively improves physical function in patients with coronary heart disease, thereby preventing and treating coronary disability.
Accepted for publication March 16, 1999.
Supported in part by grant RR-109 from the General Clinical Research Center, University of Vermont College of Medicine, Burlington.
Corresponding author: Philip A. Ades, MD, Cardiac Rehabilitation and Preventive Cardiology, Fletcher-Allen Health Care, McClure 1, Division of Cardiology, Medical Center Hospital of Vermont, Burlington,VT 05401 (e-mail firstname.lastname@example.org).