Cardiac rehabilitation (CR) refers to the provision of a wide range of secondary prevention services to patients with cardiovascular disease. Although exercise training is a core component of any CR program, modern comprehensive CR programs provide a comprehensive approach to disease modification including risk factor modification, nutritional counseling, weight management, and psychosocial management.1 The largest amount of evidence regarding the benefits of CR exists for individuals with coronary artery disease, in which clinical trial data and more recent observational data indicate that CR reduces morbidity and mortality.2,3 As a result, CR is considered a class I indication in numerous national guidelines following myocardial infarction, coronary revascularization, or in the setting of chronic, stable angina.4–8
As a result of the benefits from CR, the American Association of Cardiovascular and Pulmonary Rehabilitation (AACVPR), the American College of Cardiology (ACC), and the American Heart Association (AHA) first published, in 2007, performance measures for the referral to, and delivery of, CR services,9 followed by an update in 2010.10 These performance measures state that all hospitalized patients with a qualifying cardiovascular disease event should be referred to outpatient CR prior to hospital discharge. Qualifying cardiovascular disease events include myocardial infarction, percutaneous coronary intervention, coronary artery bypass graft surgery, stable angina, heart transplantation, and heart valve surgery.10 However, a simple diagnosis of heart failure is not listed as a qualifying diagnosis for CR referral in these performance measures.10
Although heart failure is not currently listed as a qualifying diagnosis for referral to, and enrollment in, CR in the previously cited CR performance measures, significant data exist regarding the potential benefits of exercise training in heart failure patients. In fact, current ACC–AHA guidelines list exercise training as a class I indication for individuals with heart failure.11 The objective of this article is to review the available data supporting the benefits of exercise training and formal CR programs in heart failure patients.
Mechanisms by which Exercise Training May Be Beneficial in Heart Failure
Decreased exercise tolerance, including exertional dyspnea and fatigue, is a significant cause of morbidity in patients with heart failure12 and serves as an important clinical indicator, as the New York Heart Association (NYHA) classification, which is based on patient-reported symptoms, is a strong predictor of prognosis.13 Formerly, patients with heart failure were restricted from exercise for fear of exacerbating their symptoms and hastening their disease process. Although early studies of exercise training in heart failure patients suggested that cardiac function worsened,14,15 several subsequent small, single-center studies and meta-analyses suggested that exercise training is not only safe, but may also benefit heart failure patients by improving hemodynamic, pathophysiologic, and clinical parameters.12,16–23
Several hemodynamic changes take place as left ventricular (LV) function declines. Initially, cardiac output is preserved through elevated filling pressures. As filling pressures increase and persist, LV function declines further through a series of complex mechanical and neurohormonal mechanisms. Cardiac output begins to decline and total peripheral resistance increases to maintain blood pressure, increasing LV work. Elevated filling pressures and depressed cardiac output result in dyspnea and fatigue, especially in the decompensated state. Exercise training has been shown not only to improve LV ejection fraction, end-systolic volumes, and end-diastolic volumes in chronic, compensated heart failure patients, but it has also been shown to improve LV relaxation, reverse LV remodeling, and reduce total peripheral resistance.18–21
In addition to cardiac muscle changes, skeletal muscle dysfunction contributes to decreased exercise tolerance in heart failure.12,16,24,25 The onset of heart failure symptoms frequently leads to decreased physical activity. As activity level decreases, changes in skeletal muscle occur, including muscle atrophy, loss of strength, and early anaerobic metabolism.12,25–28 Chronic hypoperfusion and deconditioning lead to oxidative stress, decreased capillary density, decreased proportion of type I muscle fibers in relation to type II muscle fibers, and reduction in anaerobic enzyme content.26,28–30 Exercise training has been shown to modify skeletal muscle changes, including decreasing lactic acid levels, increasing mitochondrial content and oxidative enzymes, and delaying anaerobic threshold.12,26 Functional benefits include increased muscle strength and endurance.31,32
Maladaptive changes in the regulation of vascular tone and intravascular volume are also seen. Elevated levels of the neurohormones norepinephrine, endothelin, vasopressin, and natriuretic peptides and increased activation of the renin–angiotensin–aldosterone system cause an increase in fluid retention and vasoconstriction.33 Higher resting sympathetic tone, decreased nitric oxide production, and abnormal endothelial function contribute to the progression of heart failure.33–35 Exercise therapy results in reduced levels of vasoconstrictive hormones, increased levels of vasodilators, higher vagal tone, and decreased sympathetic tone.36,37
Heart failure has been recognized as a pro-inflammatory state. Markers of inflammation, such as interleukin-6, interleukin-1β, tumor necrosis factor-α, and C-reactive protein, are increased in patients with heart failure.38 Improvement in the inflammatory state of chronic heart failure patients after undergoing exercise training, as measured by these markers, has been observed.39
Two Systematic Reviews of the Benefit of Exercise Training in Heart Failure
Two systematic reviews were conducted and published in 2004 looking at several single-center randomized controlled trials to determine if reduced mortality in heart failure patients who underwent exercise training could be demonstrated.
Exercise training meta-analysis of trials in patients with chronic heart failure (ExTraMATCH) was a collaborative meta-analysis that looked at nine randomized controlled trials which had assessed 801 individuals with heart failure due to LV systolic dysfunction.40 Trial participants had been on stable heart failure therapy for at least three months and had an LV ejection fraction <50 %. The treatment group consisted of a total of 395 individuals and received at least eight weeks of exercise training. The control group comprised a total of 406 participants. Compared with controls, mortality was significantly lower in the exercise group (hazard ratio [HR] 0.65, 95 % confidence interval [CI] 0.46–0.92).40
The results of a systematic review by Smart and Marwick were less promising.41 Their review comprised 30 randomized, five non-randomized, and nine randomized cross-over trials, as well as 37 longitudinal cohort studies. In total, 2,387 participants with an LV ejection fraction of <40 % received more than 60,000 patient-hours of exercise training. There were no deaths reported as directly related to exercise training. Overall, there was a non-significant trend toward lower mortality in those undergoing exercise training (odds ratio [OR] 0.71, 95 % CI 0.37–1.02).41
Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION)
The paucity of data on the morbidity and mortality benefits of exercise training in chronic stable heart failure patients led to the design of Heart failure: a controlled trial investigating outcomes of exercise training (HF-ACTION).42 The trial was designed to provide a definitive assessment of the effect of exercise training on the clinically relevant endpoints of mortality, hospitalization, and quality of life in patients with heart failure. Its inclusion and exclusion criteria are summarized in Table 1. Prior to randomization, individuals underwent cardiopulmonary exercise (CPX) testing to exclude significant arrhythmias, early ischemic changes, and abnormal blood pressure responses to exercise.42
Qualifying individuals were randomized in a 1:1 fashion between usual-care and exercise-training groups. Reasonable efforts were made to ensure that all individuals were on appropriate medical therapy. Regardless of randomization, all participants were encouraged to engage in up to 30 minutes of moderate-intensity activity, if tolerated, on most days of the week, although no formal exercise program was designed for those in the usual-care group. Individuals in the exercise-training group received formal exercise training based on AHA and American College of Sports Medicine recommendations. These individuals were prescribed a total of 36 supervised training sessions to be completed in up to six months. After 36 sessions of supervised exercise training, they transitioned to home exercise and were encouraged to continue exercising five times per week. Both groups received an equal number of follow-up visits and phone calls, unless poor adherence was noted. CPX testing was repeated at three, 12, and 24 months, with measurement of gas exchange. In addition, the Kansas City Cardiomyopathy Questionnaire (KCCQ) was used to assess health status. The primary outcome was a composite of all-cause mortality or all-cause hospitalization. Secondary outcomes were all-cause mortality, a composite of cardiovascular mortality and cardiovascular hospitalization, and a composite of cardiovascular mortality and heart failure hospitalization.42
A total of 2,331 individuals from 82 centers in the US, Canada, and France were enrolled. Of these, 1,172 were randomized to usual care and 1,159 to the exercise-training intervention. The baseline characteristics in each group were similar. The median age of the participants was 59 years; 28 % were women and 40 % were from ethnic or racial minorities. The median LV ejection fraction was 25 %. The etiology of heart failure was ischemic in 51 % of cases and the median duration of follow-up was 30.1 months.42
In the exercise-training group, only 736 (64 %) individuals completed the 36 supervised training sessions. During the first three months, these individuals exercised for a median of 76 minutes per week, although the protocol-defined goal was 90 minutes. At months 4–6, their median exercise time increased to 95 minutes per week, but it subsequently decreased to 74 minutes per week at months 10–12. From month 4 onward, the protocol-defined goals were 120 minutes per week. By the third year of follow-up, the median exercise time had dropped to 50 minutes per week. At any given time during the duration of the study, only 30 % of participants were exercising at or above their target time.42
The major findings of HF-ACTION are summarized in Table 2. Exercise training resulted in a non-significant reduction in the primary outcome of all-cause mortality or all-cause hospitalization after adjustment for heart failure etiology. Additionally, there were also non-significant reductions in all three of the secondary outcomes of all-cause mortality, cardiovascular mortality or cardiovascular hospitalization, and cardiovascular mortality or heart failure hospitalization (see Table 2). The authors pre-specified that analyses of the primary and secondary outcomes would be adjusted for key predictors of the primary outcome. Additional analyses were therefore conducted with further statistical adjustment. Independent predictors of all-cause mortality and hospitalization were baseline CPX duration, LV ejection fraction, history of atrial fibrillation or flutter, and Beck Depression Inventory II score. The results after this additional statistical adjustment are also shown in Table 2. The primary outcome of all-cause mortality or all-cause hospitalization, as well as the secondary outcomes of cardiovascular mortality or heart failure hospitalization, were significantly reduced after multivariable adjustment (see Table 2).42
Although exercise training did achieve statistically significant improvements in six-minute walk test distance at three months (but not at 12 months), cardiopulmonary exercise time at three and 12 months, and peak oxygen consumption (VO2) at three and 12 months, the median percentage improvement in peak VO2 was only 4 % in the exercise-training group. As the HF-ACTION authors point out, this fell short of the protocol-specified goal of a 10 % improvement, which is usually used to define a clinically meaningful improvement.42 Exercise training was also associated with improvements in health status. After adjustment for heart failure etiology, the overall KCCQ score increased by 5.21 points in the exercise arm as compared with 3.28 points in the usual-care group at three months (p<0.001).43
In addition to these clinical outcomes, a subsequent economic evaluation of the HF-ACTION trial was conducted and found no difference in costs between the two groups. Total direct medical costs per participant were $50,857 ± $81,488 in the exercise-training group and $56,177 ± $92,749 in the control group (p=0.10). Of the total direct costs in the exercise arm, an estimated $1,006 ± $337 was for exercise training.44
HF-ACTION was of adequate size and duration to evaluate the effect of exercise training in heart failure patients with reduced ejection fraction on all-cause mortality or all-cause hospitalization. It is the largest multi-center randomized controlled trial of exercise training and heart failure ever conducted. Although the primary analysis did not demonstrate a benefit of exercise training, the safety of regular exercise training in patients with systolic heart failure was demonstrated. In addition, self-reported health status in the exercise group was statistically improved compared with the usual-care arm. Poor participant adherence in the treatment arm and cross-over from the usual-care arm may have contributed to the lack of differentiated findings between the two groups. As previously stated, at any given time in the study, only 30 % of participants in the exercise group were meeting their weekly goals for exercise time, and 22–28 % of participants in the usual-care group reported some kind of regular exercise (although their amount of reported exercise was only half that of participants in the exercise group). The HF-ACTION investigators have acknowledged the difficulty of designing a study that ensures long-term adherence to regular exercise in a population with multiple comorbidities.
The Role of Cardiac Rehabilitation in Patients with Diastolic Dysfunction
Heart failure is a clinical diagnosis that includes patients with reduced and preserved ejection fraction. It is estimated that over half of patients with a diagnosis of heart failure have a preserved ejection fraction, and the prevalence of heart failure with a preserved ejection fraction will continue to increase as the population ages.45 Until recently, the role of exercise training in patients with heart failure and a preserved ejection fraction had not been well studied.
In 2010, Kitzman et al. published a single-center randomized controlled trial evaluating the effect of exercise training in older individuals with heart failure and a preserved ejection fraction.46 Individuals were included if they had an LV ejection fraction ≥50 % and signs or symptoms of heart failure as determined by a board-certified cardiologist. The primary and secondary outcomes studied were improvement in peak exercise VO2 and disease-specific measures of quality of life, respectively. Participants were randomized to receive either exercise training for one hour, three times a week for 16 weeks, or telephone calls every two weeks for the duration of the study, with conversations focused on new medical events since the prior contact and retention in the study, including reminders and encouragement to keep upcoming appointments. No discussion about exercise took place. The exercise group received an exercise prescription in accordance with standard methods. Perceived exertion and heart rate were monitored. For the initial two weeks, participants exercised to 40–50 % of their heart rate reserve while the duration of exercise was gradually increased. Over the following weeks, participants exercised to 60–70 % of their heart rate reserve with increasing duration of exercise. Each participant in the exercise group completed a minimum of 40 sessions.46
A total of 53 participants were enrolled, 26 of whom were assigned to receive exercise training and 27 to form the control group. Of the 53 participants, 46 completed follow-up—24 in the exercise-training group and 22 in the control group. Significantly larger increases in peak exercise oxygen uptake were observed in the exercise-training group (13.8 ± 2.5 to 16.1 ± 2.6 ml/kg/minute) as compared with the control group (12.8 ± 2.6 to 12.5 ± 3.4 ml/kg/min, p=0.0002). Additionally, an improvement in the physical score on the Minnesota Living with Heart Failure Questionnaire (MLHFQ) was observed in the exercise-training group, although the overall MLHFQ score was unchanged. Likewise, no changes in the scores on the Short Form 36-item Health Survey or the Center for Epidemiologic Studies Depression Survey were observed in either group.46
In 2011, the results of the Exercise training in diastolic heart failure (Ex-DHF) pilot study were published.47 This was the first multi-center randomized controlled trial to investigate exercise training in individuals with heart failure and a preserved ejection fraction. Individuals were included if they were >45 years of age, had an LV ejection fraction of ≥50 %, echocardiographic evidence of diastolic dysfunction, sinus rhythm, NYHA class II to III symptoms, and at least one cardiovascular disease risk factor. Eligible participants were randomized in a 2:1 fashion to exercise training or usual care.
Individuals randomized to exercise training underwent 32 supervised sessions of resistance and endurance training. From week 1 to week 4, the program comprised aerobic endurance training (cycling) twice per week with increasing intensity and duration. Intensity was tailored to reach a target heart rate of 50–60 % of peak VO2. From week 5, the frequency of the training was increased to three times weekly with a target heart rate of 70 % of baseline peak VO2 and resistance training was added at 15 repetitions per exercise session. The control group was instructed to continue with its current level of activity. The primary endpoint was improvement in peak VO2 after three months. Secondary endpoints included effects on cardiac structure, diastolic function, and quality of life.47
A total of 64 individuals were enrolled from three centers, with 44 individuals randomized to the exercise arm and 20 to the usual-care arm. Peak VO2 increased significantly (16.1 ± 4.9 to 18.7 ± 5.4 ml/min/kg, p< 0.001) in the exercise group but remained unchanged in the control group (16.7 ± 4.7 to 16.0 ± 6.0 ml/min/kg, p=not stated). Similar improvements were seen in cardiac structure, diastolic function, and quality of life in the exercice group.47
These randomized controlled trials of exercise training in individuals with heart failure and preserved ejection fractions have demonstrated clinical benefit, with improved exercise tolerance, improved indices of diastolic function, reversed remodeling, and improvements in quality of life. A larger trial is needed to further investigate the clinical benefit, safety, and effects on morbidity and mortality of exercise training in individuals with heart failure and preserved ejection fractions.
Summary and Conclusions
The totality of the evidence suggests that exercise training and formal CR programs improve outcomes in individuals with heart failure. Exercise training results in improvements in cardiac and skeletal muscle function and modifies the biochemical, neurohormonal, and inflammatory responses to heart failure. Meta-analyses of clinical trials conducted in patients with heart failure have suggested that mortality is lower in heart failure patients who undergo formal exercise training. However, the results from the recently published HF-ACTION trial failed to definitively prove that exercise training reduces mortality in patients with heart failure and reduced systolic function, largely as a result of non-adherence to the exercise prescription in a large percentage of study participants. However, HF-ACTION has demonstrated that formal exercise training is safe in this patient population and does not result in significantly increased costs to the healthcare system. Recent studies of exercise training in heart failure patients with preserved ejection fractions, in which higher levels of adherence to the exercise prescription were achieved than in HF-ACTION, have demonstrated improvements in exercise tolerance and indices of diastolic function, as well as in symptom burden and quality of life. Further large-scale, multi-center clinical trials are needed to definitively establish the role of exercise training and formal CR programs in this patient population.