To better understand the mechanisms contributing to improved exercise capacity with cardiac resynchronization therapy (CRT), we studied the effects of 6 mo of CRT on pulmonary O(2) uptake (Vo(2)) kinetics, exercise left ventricular (LV) function, and peak Vo(2) in 12 subjects (age: 56 ± 15 yr, peak Vo(2): 12.9 ± 3.2 ml·kg(-1)·min(-1), ejection fraction: 18 ± 3%) with heart failure. We hypothesized that CRT would speed Vo(2) kinetics due to an increase in stroke volume secondary to a reduction in LV end-systolic volume (ESV) and that the increase in peak Vo(2) would be related to an increase in cardiac output reserve. We found that Vo(2) kinetics were faster during the transition to moderate-intensity exercise after CRT (pre-CRT: 69 ± 21 s vs. post-CRT: 54 ± 17 s, P < 0.05). During moderate-intensity exercise, LV ESV reserve (exercise - resting) increased 9 ± 7 ml (vs. a 3 ± 9-ml decrease pre-CRT, P < 0.05), and steady-state stroke volume increased (pre-CRT: 42 ± 8 ml vs. post-CRT: 61 ± 12 ml, P < 0.05). LV end-diastolic volume did not change from rest to steady-state exercise post-CRT (P > 0.05). CRT improved heart rate, measured as a lower resting and steady-state exercise heart rate and as faster heart rate kinetics after CRT (pre-CRT: 89 ± 12 s vs. post-CRT: 69 ± 21 s, P < 0.05). For peak exercise, cardiac output reserve increased significantly post-CRT and was 22% higher at peak exercise post-CRT (both P < 0.05). The increase in cardiac output was due to both a significant increase in peak and reserve stroke volume and to a nonsignificant increase in heart rate reserve. Similar patterns in LV volumes as moderate-intensity exercise were observed at peak exercise. Cardiac output reserve was related to peak Vo(2) (r = 0.48, P < 0.05). These findings demonstrate the chronic CRT-mediated cardiac factors that contribute, in part, to the speeding in Vo(2) kinetics and increase in peak Vo(2) in clinically stable heart failure patients.