Exercise Training in Patients with Heart Failure and Preserved Ejection Fraction: Findings Awaiting Discovery

Since the seminal work by Coats et al (1) that investigated the effects of exercise training in patients with heart failure with reduced ejection fraction (HFREF), a large body of clinical research has been reported that includes dozens of single-site randomized trials, numerous metaanalyses and review articles, and one multi-center randomized trial. These studies followed a logical progression from initially investigating the feasibility of exercise training and its impact on exercise tolerance to better understanding how regular exercise helps to mitigate the abnormal pathophysiology of heart failure and its influence on clinical outcomes (2). Clearly, the last 20 years has proven to be fertile and rewarding ground for clinical investigators interested in exercise training, such that regular exercise is now an evidence-based recommendation in the care of patients with stable HFREF (3).

The same can not be said for exercise in patients with heart failure with preserved ejection fraction (HFPEF). In fact, there have been far fewer single-site randomized trials, meta-analyses, and multi-center randomized trials that investigated the effects of exercise training in these patients. This paucity of exercise-related research is unfortunate, given that approximately 50% of the patients with heart failure have HFPEF and these same patients demonstrate a degree of exercise intolerance, impaired quality of life, mortality risk, and a re-hospitalization rate that mirrors or exceeds that of patients with HFREF (4–6).

The trial by Kitzman et al., (7), which investigated the effects of regular exercise on exercise capacity in patients with HFPEF, is in the vanguard of what is hopefully a sustained movement that will progress to a greater understanding of the basic physiology that impacts improving exercise performance in HFPEF. Although there are some similarities in the symptoms and the pathophysiology of HFREF and HFPEF, Kitzman and colleagues illuminate a significant difference relative to the effect of exercise training on peripheral vascular function. Clearly, one can not simply assume that all of the favorable physiologic adaptations now ascribed to exercise training in patients with HFREF will also occur in patients with HFPEF.

Using a controlled design and appropriate eligibility criteria, this study randomized 63 elderly patients with HFPEF (~75% female) to 16 wk of supervised group exercise training versus no exercise attention control. Exercise training duration was progressively increased to 40 minutes per session and included track walking and leg and arm cycle ergometry. Observed findings included 88% attendance to prescribed exercise sessions, and 86% of all patients completed testing at both baseline and 16 week follow-up. Change in brachial artery flow mediated dilation (FMD), which was the primary outcome for the study, was not different between study groups at week 8 or 16. Among exercise trained subjects (versus control) both quality of life scores and exercise capacity (i.e., peak oxygen uptake, six min walk distance, or exercise duration) were improved.

The above observations raise important issues worthy of further discussion. First, the absence of any change in brachial artery FMD differs from what is often reported for exercise training in patients with HFREF (2). This difference between the two forms of heart failure may be due to differences in patient characteristics, differences in the mechanism(s) responsible for the abnormality in cardiac function, the investigator’s decision to exclude patients with known vascular disease, or the possibility that the marked disruption of endothelial function is simply an abnormality that is unique to the pathophysiology of patients with HFREF (8). The latter suggests that endothelial dysfunction is neither a cause or a consequence of HFPEF and therefore, not associated with the disease-related exercise intolerance. This may explain why some therapies shown to be effective in treating patients with HFREF are yet to be proven effective in patients with HFPEF.

Second, Kitzman et al. suggest that their observed 13% increase in peak oxygen uptake among patients with HFPEF, which is similar to the magnitude of increase reported in patients with HFREF (2), was mostly due to exercise training induced improvements in the function (e.g., capillary density, mitochondrial density, oxidative enzyme activity) of the metabolically active skeletal muscles. They base their premise on both the absence of any observed modulation of vascular function in their recent study and prior trials from their laboratory (9) and another group of investigators (10) that showed no exercise training-related change in peak cardiac output. Although their “peripheral premise” is likely correct, such discussion calls for future trials that include evoked measures of skeletal muscle histo-chemistry that are gathered before and after a period of exercise training.

Before closing, I wish to introduce two additional topics into the discussion that stem from this paper. First, like most of the other single site trials conducted to date that involve patients with either HFREF or HFPEF, compliance to exercise in this trial by Kitzman et al., (7) was excellent at 88% of all prescribed sessions. However, this level of adherence is somewhat contrary to what is sometimes observed in large randomized controlled trials that involve maintenance of regular exercise in patients with cardiovascular disease. In the National Exercise and Heart Disease Project that was completed more than 30 years ago, patients suffering a myocardial infarction who were randomized to exercise training (versus non-exercise control) demonstrated an ~47% compliance rate to exercise at one year (11). More recently, the HF ACTION Trial (Heart Failure – A Controlled Trial Investigating Outcomes of Exercise Training) showed a compliance rate to the prescribed amount of exercise to be ~40% at one year (12). Taken in sum, the above begs the larger question of what factors are altered when investigators deploy adherence strategies that are used successfully in single-site trials in larger multi-site trials. Why are higher levels of adherence not maintained? Is it the many co-morbidities that accompany heart failure proper, the less rigorous methods for study over-site and patient screening that sometimes accompanies multi-site trials, or is it a loosening of the eligibility criteria that may occur to help ensure the generalizability of trial findings? Clearly, to better evaluate the impact of exercise training in the clinical care of patients with cardiovascular diseases, more research-derived information is needed that specifically targets improved exercise adherence in multi-site trials.

A final point worth mentioning pertains to the need to formally revisit insurance reimbursement policies for exercise training among Medicare beneficiaries with heart failure. In 2005 the Centers for Medicare and Medicaid Services (CMS) reviewed this issue and subsequently denied reimbursement for cardiac rehabilitation for patients with heart failure, despite a level of evidence at that time that exceeded the level of evidence used by Medicare in its decision to cover patients with percutaneous coronary intervention, cardiac transplant and heart valve replacement/repair. Some 7 years later and with an even greater body of evidence supporting the benefits of exercise training in the care of patients with heart failure, the timing for another formal review by CMS is long over due. Should CMS decide to cover cardiac rehabilitation for patients with heart failure but consider limiting such coverage to only those patients with HFREF, it will miss an opportunity to provide benefit to the larger body of patients with heart failure, both HFREF and HFPEF. Specifically, the evidence supporting the use of exercise training to improve quality of life and exercise capacity in patients with HFPEF is now equal to or exceeds the level of evidence in 2006 that led to the coverage for cardiac transplant, percutaneous coronary intervention and valve replacement/repair.

This paper by Kitzman and colleagues is important because it shows that exercise training improves exercise tolerance in older patients with HFPEF and that such improvement is likely independent of changes in FMD. It forces future research into the study of metabolism within the more active skeletal muscles that are involved with exercise. This paper also stimulates us to gather more knowledge about how exercise adherence can be sustained when trials are scaled up from a single-site to multi-site setting and pushes further the need for a revised policy from CMS that provides coverage for cardiac rehabilitation care for patients with heart failure.