Introduction: We conducted a systematic review to assess the effect of exercise training in patients with heart failure with preserved ejection fraction (HFPEF). Methods: A number of electronic databases were searched up to November 2011 to identify comparative studies of exercise training in HFPEF. Where possible, outcome data from included studies were pooled using meta-analysis. Results: Three randomised controlled trials, one non-randomised controlled trial and one pre-post study were included, for a total of 228 individuals. The combined duration of exercise programmes and follow-up ranged from 12 to 24 weeks. No deaths, hospital admissions or serious adverse events were observed during or immediately following exercise training. Compared to control, the change in exercise capacity at follow-up was higher with exercise training (between group mean difference: 3.0 ml/kg/min, 95% CI: 2.4 to 2.6). In the four studies, that reported the Minnesota Living with Heart Failure questionnaire, there was evidence of a larger gain in health-related quality of life with exercise training (7.3 units, 3.3 to 11.4). The largest study showed some evidence of improvement in the E/E' ratio with exercise training, but this was not confirmed in the other studies (overall -0.9, -3.8 to 2.0); E/A ratios were not changed. Conclusions: Exercise training for patients with HFPEF confers benefit in terms of enhancements in exercise capacity and health-related quality of life and appears to be safe. The impact on diastolic function remains unclear. Further trials should provide data on long term effects, prognostic relevance and cost-effectiveness.