In order to compare directly the efficacy of two different angiotensin-converting enzyme (ACE) inhibitors in terms of clinical status and exercise capacity, 443 patients with chronic heart failure (New York Heart Association classes II-IV) were randomized into a 24-week double-blind study to receive cilazapril (CLZ) 1-2.5 mg once daily (n = 221), captopril (CPT) 25-50 mg three times daily (n = 108), or placebo (PLA) for 12 weeks followed by CLZ 2.5 mg (n = 114) in addition to their standard heart failure therapy. The majority were New York Heart Association functional class II (56-62%), and the most frequent etiology of chronic heart failure was coronary heart disease (35-42%), followed by dilative cardiomyopathy (≈28%). Both ACE inhibitors prolonged the exercise tolerance test duration at all visits, and the effect at week 12 (CLZ 62.2 ± 7.5 s; CPT 73.1 ± 10.7 s) was significantly greater than after PLA (28.1 ± 12.2 s; p = 0.011 and p = 0.005, respectively). Furthermore, the distance walked during 6 min increased at all visits with ACE inhibitors (NS vs. PLA). CLZ was more effective in patients with the most impaired physical ability at baseline as defined by exercise tolerance test duration < 6 min (p = 0.0036 at week 12 and p = 0.0150 at week 24) and by walking test < 400 m (p = 0.0004 at week 12 and p = 0.0009 at week 24). Similar results were obtained with CPT for the walking test (p = 0.0369 at week 12 and p = 0.0142 at week 24). More patients on PLA tended to worsen their clinical status, as assessed by the New York Heart Association class or increased diuretic use. Furthermore, the mean body weight decreased significantly on CLZ, while it remained unchanged on PLA (p = 0.0335 for CLZ vs. PLA). The ACE inhibitor therapy was well tolerated, with dizziness (CLZ 10%; CPT 10.2%) and coughing (CLZ 9.0%; CPT 9.3%) most frequently reported. CLZ was well tolerated by-young and elderly patients, while CPT tended to cause more adverse events in elderly (63.0%) than in younger (48.4%) patients. Moreover, less patients discontinued treatment for adverse events with CLZ (5.4%) than with CPT (13.0%) or PLA (10.5%). Thus, the two ACE inhibitors CLZ and CPT at the dosages used were equally effective and both significantly better than PLA in prolonging the exercise tolerance test duration. This improvement was achieved with once daily dosing of CLZ as compared with three times daily administration of CPT and with a tendency for less adverse events with CLZ, particularly in older patients. The effect was maintained over 24 weeks and more pronounced in patients with the most impaired baseline physical ability. The similarity in the improvement of exercise tolerance with both ACE inhibitors suggests that this is probably a class effect, while the incidence in adverse events points toward differences in terms of tolerability of these compounds.