PURPOSE: Aerobic exercise training has been used in patients with stable heart failure (HF) to reduce the risk of clinical events. However, due to patient heterogeneity, some patients may experience a decrease in functional capacity due to such training. The purpose of this study was to estimate the proportion of HF patients participating in a training program who had negative responses to such therapy and to compare them with a concurrent control group. METHODS: Baseline and 3-month peak V˙O2 measurements were obtained on 1870 HF subjects who were randomized to receive either an exercise training program or a control program of usual care without exercise training. The exercise program consisted of supervised walking or stationary cycling 3 d·wk(-1) for 12 wk as well as a 2-d·wk(-1) home exercise program after completing 18 supervised sessions. A negative response was defined as a baseline-to-3-month decrease in peak V˙O2 of at least 5 mL·kg(-1) min(-1), which was two times the SD of the control group's change in peak V˙O2. RESULTS: The mean ± SD change in peak V˙O2 in the exercise group and control group was 0.8 ± 2.5 mL·kg(-1)min(-1) and 0.2 ± 2.5 mL·kg(-1)min(-1), respectively (P < 0.001). The percentage of negative responders in the exercise and control groups was 0.9% and 2.3% (P = 0.02). CONCLUSIONS: The low negative response rate in the exercise group combined with the slightly higher rate in the control group and equal variability in the exercise and control groups suggests that few if any subjects had training-related negative peak V˙O2 responses. These findings support current exercise recommendations for HF patients.