Exercise Training and Heart Failure in Older Adults—Dismal Failure or Not Enough Exercise?

Cardiovascular (CV) disease continues to be a major threat to the health of older patients.1 Despite marked advances in the detection and treatment of most CV diseases, heart failure (HF) continues to produce a heavy burden of CV morbidity and mortality, especially in the elderly population.2, 3 In 2006 in the United States, HF accounted for nearly 750,000 emergency department visits, 1 million hospital admissions, and an estimated cost of nearly $30 billion, with much of this being attributed to the elderly population.4 Cardiopulmonary fitness (CF) and exercise are known to predict prognosis and functional capacity in patients with a variety of CV disorders, including HF.5, 6 We have recently reviewed the marked benefits of CF and exercise training (ET), especially with formal cardiac rehabilitation and ET (CRET) programs in patients with a variety of CV disorders.5 Although many of these benefits have been noted in younger patients with CV disease, the benefits of ET have been noted in elderly patients, including in cohorts well over the age of 70 (Table 1).7-12 Although in most of these studies, the benefits of ET were assessed according to estimated exercise capacity (in metabolic equivalents (METs)), substantial improvements in precisely measured exercise capacity (peak oxygen consumption (peak VO2)), including in 57 adults with a mean age of 78 (Table 2) who had a 13% improvement in peak VO2, were also demonstrated.11 In another large study of 260 older adults with a mean age of 75, although a less impressive, but significant, 7% improvement in peak VO2 after CRET (P<.001) was noted, this was associated with marked improvements in CV risk factors, as well as reductions in the prevalence of depression (−67%; P<.001), hostility (−50%; P<.05), and anxiety (−31%; P<.01), which may affect prognosis. The benefits of ET in patients with advanced HF continues to be controversial and has recently been reviewed.3, 5 A review of 81 studies with 2,383 patients of ET in HF found an average increase in peak VO2 of nearly 17%.13 This meta-analysis demonstrated a trend toward lower mortality with ET (−29%; P=.06). In Exercise Training Meta-Analysis of Trials in Patients with Chronic Heart Failure, which assessed 801 patients from nine studies, ET was associated with 35% lower mortality (P=.02) and 28% fewer hospital admissions (P=.01).14 A major randomized control trial (RCT) of ET in HF, A Randomized Control Trial Investigating Outcomes of Exercise Training (HF-ACTION), is a landmark trial that randomly assigned 2,331 patients with HF to 3 months of supervised ET (3 supervised sessions weekly for 36 ET sessions at 60–70% of the heart rate reserve; patients were instructed to exercise 5 times weekly) followed by unsupervised ET versus usual care.15 Although the primary endpoint was not significantly affected, after preplanned adjustment for confounding factors (baseline peak VO2, ejection fraction, Beck Depression Inventory, and history of atrial fibrillation or flutter), there was a significant 11% reduction in all-cause mortality and a 15% reduction in CV mortality or hospitalizations. Unfortunately, peak VO2 increased significantly, although by less than 5% (P<.001), possibly because the ET group did not perform the volume of exercise that was initially anticipated, thus resulting in much smaller increases in peak VO2 than the greater than 15% improvement anticipated and reported in the meta-analyses.16 Nevertheless, this study confirmed the safety and efficacy of ET in advanced HF in patients receiving vigorous contemporary HF management. If patients had been more adherent to their exercise prescription (e.g., ET for closer to 150 h/wk as opposed to only half this amount), it is likely that greater improvements in peak VO2 would have been noted, along with greater reductions in major CV events and mortality.3 In the study by Brubaker et al. in this issue of the journal,17 the data were “revived” nearly a decade after the study was completed. Presumably, presentation and publication of HF-ACTION has stimulated renewed interest in the study of ET in HF. As the authors point out, the mean age of their population was aged 70 and older, compared with only 59 in HF-ACTION, although their study included only 23 patients with baseline and post-ET data, and they included only three weekly supervised ET sessions, with no comment on other nonsupervised ET performed by their cohort. Unfortunately, there was no improvement in mean peak VO2 in their patients after ET, although the authors noted that, in six of 23 (26%), peak VO2 increased by 10% or more, and in nine of 23 (39%), there was 0% to 9.9% improvement. Therefore, the median change in peak VO2 would be much more meaningful and is clearly in the 0% to 9.9% range in this cohort. It was recently demonstrated that only small improvements in peak VO2 (0–10%) are required to produce improvements in depression and depression-related mortality in a large cohort with coronary heart disease, and greater improvements in peak VO2 (≥10%) produced no additional improvements in prognosis.18 Because it has been reported that depression has considerable effect on morbidity and mortality in patients with advanced HF,19, 20 it is possible that even these small improvements noted in many patients in HF-ACTION and in this small current study may have beneficial effects on psychological distress and its related mortality in these patients with advanced HF. Further analysis of HF-ACTION and other HF studies should assess the effect of ET on depression and other psychological distress that may affect morbidity and mortality. Additionally, other studies, including cohorts of older adults, will need to determine whether a higher volume of exercise (e.g., longer duration per session or more sessions per week) are feasible and produce better improvements in peak VO2 in patients with HF, including older patients. Although all of these points could be ideally assessed in large-scale RCTs that also evaluate morbidity and mortality, unfortunately, it is unlikely that any such study will be forthcoming any time in the near future. Therefore, for the time being, we must depend on the data from the large RCT (HF-ACTION), the large meta-analyses of HF trials, and surrogate endpoints from small-scale studies such as the one in this present issue.17 Although some would conclude that this was a “negative” study, we would conclude that this study had too few subjects and too short a duration and that patients did not perform enough exercise! Nevertheless, modest benefits were still noted in two-thirds of their study cohort, suggesting that ET may still benefit many elderly with HF.