Response to Endurance Exercise Training in Older Adults with Heart Failure with Preserved or Reduced Ejection Fraction

Editor's Note: Commentary based on Pandey A, Kitzman DW, Brubaker P, et al. Response to endurance exercise training in older adults with heart failure with preserved or reduced ejection fraction. J Am Geriatr Soc 2017;65:1698-1704.

Rationale/Background: The patterns and clinical predictors of peak oxygen uptake (VO2peak) response to exercise training have not been compared between older adults with heart failure with preserved ejection fraction (HFpEF) and HF with reduced ejection fraction (HFrEF). This study characterizes the response in VO2peak to supervised endurance exercise training in older adults with HF in order to determine the predictors of response in HFpEF versus HFrEF.

Study Question: What are the relative magnitudes and predictors of response in peak VO2 to exercise training in older adults with HFpEF and HFrEF?

Funding: This study was supported by the National Institute on Aging of the National Institutes of Health (NIH), the Kermit Glenn Philips II Chair in Cardiovascular Medicine, Wake Forest University, the Moritz Chair in Geriatrics, College of Nursing and Health Innovation, University of Texas at Arlington, the National Institute of Nursing Research of the NIH, the Dedman Family Scholar in Clinical Care endowment at University of Texas Southwestern Medical Center and the American Heart Association Prevention Network.


Study Design: Secondary analysis using individual-level data from a prospective, randomized, blinded, attention-controlled trial of facility-based, supervised endurance exercise training (stratified by EF). Cardiopulmonary exercise testing was performed using upright cycle ergometry with expired gas-exchange analysis at baseline and follow-up according to standardized protocols.

Study Population: Individuals aged 65 and older with chronic, well-compensated HF (n = 24 HFpEF, n = 24 HFrEF) on a stable medication regimen for 6 weeks or more.

Intervention: Training group subjects participated in supervised exercise training three times per week for 16 weeks, each session lasting 1 hour with target walking and cycling durations of 15-20 minutes.

Outcomes: The primary outcome was percent change in VO2peak (mL O2/kg/min) from baseline to 16-week follow-up. Secondary outcomes included exercise time and ventilatory anaerobic threshold.

Statistical Analysis: Outcomes were compared between HFpEF and HFrEF groups using the t-test. Associations between changes in VO2peak and various predictors were assessed using the t-test and trend test, both for the overall study population and separately for HFpEF and HFrEF. Tests for interactions were performed and multivariate linear regression models were constructed.

Results: Endurance training resulted in a 9.2% improvement in VO2peak in both groups combined. Greater improvement was observed in patients with HFpEF compared with HFrEF (18.7 ± 17.6% vs -0.3 ± 15.4%; p < 0.001). Similar patterns were demonstrated with absolute VO2peak, proportion of patients with >5% and >10% improvement in VO2peak, exercise time, and ventilatory anaerobic thresholds. Univariate analysis demonstrated that echocardiographic abnormalities in left ventricular structure and function and lower body mass index (BMI) were associated with blunted response of VO2peak with exercise training. Using multivariate linear regression modelling with stepwise variable selection, higher submaximal systolic blood pressure (P = 0.008) and shorter mitral valve (MV) deceleration time on echo (P = 0.04) were independently associated with blunted response in VO2peak.

Limitations: Only a singular training protocol was tested that included continuous, moderate-intensity exercise, although VO2peak response may differ according to varying types and intensities of training. The absence of data on weight and body composition before and after exercise training does not allow for the determination of how changes in these parameters influenced the outcomes of interest. Additionally, in the statistical analysis, there was no adjustment for multiple comparisons, so findings would need to be confirmed in future studies. Finally, beta-blocker use was very low in the HFrEF group (9%; likely due to older age and comorbid conditions), and the HFrEF group may not have been on optimized therapy (although this may represent real-world conditions, which could additionally strengthen the comparison between HFpEF and HFrEF).

Conclusions: In response to supervised endurance exercise training in older adults with HF, VO2peak improvement was greater in patients with HFpEF than in those with HFrEF.

Perspective: The therapeutic options for patients with HFpEF remain limited. This study is the first of its kind to directly compare peak exercise response to training in older adults with HFpEF and HFrEF and to assess predictors of training response. Since exercise intolerance represents a cardinal symptom of chronic HF (and particularly HFpEF), these results add to the evidence that exercise training improves objective measures in patients with HFpEF. Improvement in VO2peak was demonstrated at clinically meaningful thresholds (>5% and >10% proportional improvement) for HFpEF compared to HFrEF, despite identical training regimens and high adherence rates among both groups, an observation that warrants further study. The mechanisms underlying these observations were partially addressed, demonstrating that higher systolic blood pressure and shorter MV deceleration time were associated with blunted VO2peak response to training. However, as VO2peak is determined by peripheral factors (e.g. vascular and muscle), measures of body composition and vascular impedance could have further elucidated the underlying mechanisms of the differential VO2peak response to training. This study has an important clinical implication—given the persistent failure to demonstrate mortality benefits with pharmacologic interventions in patients with HFpEF, exercise training represents an important practical intervention for such patients with otherwise limited options. The Center for Medicare and Medicaid Services recently expanded cardiac rehabilitation coverage for patients with chronic, stable HFrEF, but not for HFpEF. These findings provide further evidence that cardiac rehabilitation and exercise training is an effective therapeutic strategy for HFpEF, and suggest that reimbursement for such therapy should be reconsidered for this growing cohort of patients.

Clinical Topics: Geriatric Cardiology, Heart Failure and Cardiomyopathies, Noninvasive Imaging, Acute Heart Failure, Echocardiography/Ultrasound

Keywords: Body Mass Index, Cardiac Rehabilitation, Medicaid, Anaerobic Threshold, Stroke Volume, Mitral Valve, Linear Models, Geriatrics, Follow-Up Studies, Prospective Studies, Blood Pressure, Deceleration, Electric Impedance, Exercise Test, Heart Failure, Body Weight, Adrenergic beta-Antagonists, Medicare, Body Composition, Financial Management, Echocardiography, Cohort Studies

< Back to Listings