PURPOSE: To determine the efficacy of exercise training and its effects on outcomes in patients with heart failure. METHODS: MEDLINE, Medscape, and the Cochrane Controlled Trials Registry were searched for trials of exercise training in heart failure patients. Data relating to training protocol, exercise capacity, and outcome measures were extracted and reviewed. RESULTS: A total of 81 studies were identified: 30 randomized controlled trials, five nonrandomized controlled trials, nine randomized crossover trials, and 37 longitudinal cohort studies. Exercise training was performed in 2387 patients. The average increment in peak oxygen consumption was 17% in 57 studies that measured oxygen consumption directly, 17% in 40 studies of aerobic training, 9% in three studies that only used strength training, 15% in 13 studies of combined aerobic and strength training, and 16% in the one study on inspiratory training. There were no reports of deaths that were directly related to exercise during more than 60,000 patient-hours of exercise training. During the training and follow-up periods of the randomized controlled trials, there were 56 combined (deaths or adverse events) events in the exercise groups and 75 combined events in the control groups (odds ratio [OR] = 0.98; 95% confidence interval [CI]: 0.61 to 1.32; P = 0.60). During this same period, 26 exercising and 41 nonexercising subjects died (OR = 0.71; 95% CI: 0.37 to 1.02; P = 0.06). CONCLUSION: Exercise training is safe and effective in patients with heart failure. The risk of adverse events may be reduced, but further studies are required to determine whether there is any mortality benefit.