HFpEF Exercise: What Training Can—and Cannot—Do

Exercise is one of the few interventions that consistently improves how people with heart failure with preserved ejection fraction (HFpEF) function and feel. It improves exercise capacity and quality of life across randomized trials. It has not yet been shown to reduce deaths or heart-failure hospitalizations.

That distinction matters because HFpEF is a serious disease, not a mildly inconvenient diagnosis with a reassuring ejection fraction. A recent phase 2 trial report cites registry-level five-year survival of 35% to 40%. Drug treatment has improved after decades of disappointing results: SGLT2 inhibitors and finerenone reduce cardiovascular death or heart-failure hospitalization outcomes. Yet residual risk remains high, and no drug class has produced the clear mortality benefit seen with neurohormonal treatment in heart failure with reduced ejection fraction.

For people with HFpEF, structured exercise belongs in the evidence-based plan. It should sit alongside clinician-directed treatment of blood pressure, diabetes, obesity, atrial fibrillation, kidney disease, and other contributors—not in place of it.

Exercise changes functional outcomes, not proven survival

The strongest exercise evidence comes from a 2024 systematic review and meta-analysis by Baral et al. It pooled 14 randomized controlled trials with 629 participants. The average participant was 68.1 years old, and 63.2% were women. Across endurance training, high-intensity interval training (HIIT), inspiratory muscle training, and functional electrical stimulation, exercise improved peak oxygen uptake, six-minute walk distance, and Minnesota Living With Heart Failure quality-of-life scores.

Those are meaningful outcomes. Peak oxygen uptake reflects cardiorespiratory capacity; the six-minute walk test reflects what a person can do outside a physiology lab; quality-of-life scores capture the daily cost of breathlessness and fatigue. But they remain surrogate and patient-reported outcomes, not proof that training changes the disease's mortality trajectory.

The pooled trials were too small to reliably detect differences in death or hospitalization. Their combined sample was 629 people, and their programs varied substantially in age, body size, diabetes prevalence, atrial fibrillation prevalence, exercise method, and duration. A better walk test is worth having. It simply does not answer the same question as whether a program prevents a heart-failure admission.

Two independent reviews point in the same direction. A 2022 meta-analysis by Boulmpou et al. found that exercise programs improved cardiorespiratory reserve in HFpEF. A 2025 systematic review of cardiac rehabilitation also found functional and clinical improvements across the reviewed evidence. Agreement among reviews increases confidence that supervised, structured exercise can improve functional capacity. It does not turn the available trials into mortality trials.

Long enough to adapt, practical enough to continue

In the Baral review, programs lasting at least 12 weeks outperformed shorter programs. The endurance protocols generally used 60% to 70% of heart-rate reserve or peak oxygen uptake for 12 to 20 weeks. HIIT protocols ranged from 4 to 52 weeks and commonly targeted 80% to 90% of peak heart rate or heart-rate reserve. Inspiratory muscle training was typically performed twice daily for 20 minutes per session over 12 to 24 weeks.

These are study protocols, not a universal prescription. HFpEF is heterogeneous: some people have obesity, diabetes, atrial fibrillation, pulmonary disease, frailty, orthopedic limitations, or all of the above. A cardiac-rehabilitation program can tailor intensity, monitor symptoms, and build a plan that a person can actually sustain. Adherence is not a soft endpoint when the intervention is training; it is the intervention.

HIIT has no proven edge over moderate continuous training

HIIT often attracts attention because it sounds efficient and physiologically ambitious. The best head-to-head evidence does not show that it beats moderate continuous training (MCT) in HFpEF.

OptimEx-Clin, the largest and longest individual trial included in the Baral review, randomized 176 participants to 52 weeks of cycling. HIIT targeted 80% to 90% of heart-rate reserve; MCT targeted 35% to 50%. The trial found no HIIT superiority for improvement in peak oxygen uptake.

Smaller studies had suggested an advantage for HIIT, but the larger, year-long comparison did not confirm it. That may reflect differences in sample size, duration, adherence, or the difficulty of delivering very intense training in people with substantial comorbidity. The practical conclusion is straightforward: choose the format a patient can perform safely and maintain. The evidence supports both approaches for functional outcomes; it does not establish a HIIT hierarchy.

What exercise intolerance markers can tell us

Several observational studies describe cardiac and hemodynamic features that track with poor exercise capacity. They are useful for understanding the problem and for designing future trials. They do not prove that exercise reverses those features.

In older adults with obesity and HFpEF, Singleton et al. found that left-atrial stiffness independently predicted exercise intolerance and quality of life. That makes left-atrial stiffness a marker of functional impairment in this population. It does not show that aerobic training lowers left-atrial stiffness.

A 2025 international multicenter cohort study by Rowe et al. included 2,876 people with an ejection fraction of at least 50%, including 347 with HFpEF. Left-ventricular end-diastolic volume independently predicted cardiorespiratory fitness and functional independence. The study was observational and included athletes, healthy non-athletes, people with unexplained dyspnea, and people with HFpEF. It therefore identifies a structural predictor across a broad normal-ejection-fraction population, not an exercise-induced treatment target in HFpEF.

Likewise, Morales-Oyarvide et al. reported that the hemodynamic gain index correlated with exercise capacity in HFpEF. Impaired reserve tracks with worse performance. Whether exercise can improve that index enough to alter clinical outcomes remains unknown.

It is tempting to draw a clean line from training to cardiac remodeling to longer survival. Biology rarely supplies such tidy arrows. These studies make the hypothesis plausible; none tests it.

Inflammation and metabolic disease shape the setting

HFpEF often develops in the setting of hypertension, obesity, diabetes, kidney dysfunction, and systemic inflammation. In an analysis of 374 people with HFpEF, Alogna et al. found that participants with the highest interleukin-6 levels had higher body mass index, more diabetes, worse renal function, and lower exercise tolerance. This supports an inflammatory disease model, but it does not show that exercise lowers interleukin-6 in HFpEF or that lowering the biomarker improves outcomes.

A 2025 observational study of epicardial adipose tissue also linked greater tissue burden with metabolic comorbidities, including obesity and diabetes. The association helps explain why risk-factor treatment matters. It does not establish that exercise reduces epicardial fat in HFpEF.

Drugs have narrowed the gap, but have not closed it

Exercise should not be framed as the only treatment that matters. A 2023 review by Pandey et al. describes how SGLT2 inhibitors changed a field that had resisted effective drug treatment for decades. More recently, the 2026 Nassiri et al. report states that SGLT2 inhibitors and finerenone have reduced cardiovascular death or heart-failure hospitalization outcomes in HFpEF.

That is a real advance, but it does not justify claims that pharmacotherapy has solved HFpEF. The evidence summarized here does not provide individual trial estimates for the major drug trials, so this article cannot rank drugs by effect size or define which patient should receive which therapy. Those decisions require individualized clinical care.

Other signals remain narrower. The INABLE-Training trial tested inorganic nitrite as an adjunct to exercise in a multicenter, double-blind, placebo-controlled design, but the available abstract information does not provide the pooled peak-oxygen-uptake effect estimate needed for a firm conclusion. A spironolactone analysis reported a fibrosis biomarker outcome, procollagen type I carboxy-terminal propeptide, rather than mortality or hospitalization results. Biomarkers can guide research; they cannot substitute for patient outcomes.

Skip the EGCG sales pitch

EGCG, a green-tea polyphenol, has a murine signal: 40 female C57BL/6 mice were divided into four groups of 10. That is preclinical work, not a treatment recommendation. No human randomized trial or cohort study in this evidence base shows that EGCG improves diastolic function in HFpEF, and there are no human HFpEF data to guide dose, duration, bioavailability, or safety. Supplements do not become clinical evidence by wearing a lab coat.

The useful conclusion

For HFpEF, structured exercise training of at least 12 weeks has the most consistent randomized evidence for improving peak oxygen uptake, walking capacity, and quality of life. Moderate continuous training and HIIT both fit that evidence; the 176-person OptimEx-Clin trial found no peak-oxygen-uptake advantage for HIIT after 52 weeks.

The honest boundary is equally clear. Exercise trials have not established fewer deaths or hospitalizations, and observational markers such as left-atrial stiffness, ventricular volume, interleukin-6, and hemodynamic gain are not proven exercise-modifiable mechanisms. Use cardiac rehabilitation when available, build the program around safety and adherence, and treat HFpEF as the high-risk clinical condition it is.