Background: The mechanism by which angiotensin-converting enzyme inhibitors reduce mortality rates and disease progression in patients with heart failure is likely mediated in part through prevention of adverse ventricular remodeling. This study examined the effects of the angiotensin- converting enzyme inhibitor captopril and the angiotensin II type 1 receptor antagonist losartan on ventricular volumes and function in elderly patients with heart failure and reduced left ventricular ejection fraction (≤40%). Methods: Patients underwent radionuclide ventriculograms (RVG) at baseline and were randomized to either captopril (n = 16) or losartan (n = 13). After 48 weeks, another RVG was obtained. Therapy was then withdrawn for at least 5 days, and the RVG was repeated while the patient was not receiving the drug. Results: At 48 weeks both captopril and losartan significantly reduced left ventricular (LV) end-diastolic volume index (13.5 ± 26 to 128 ± 23 mL/m2 for losartan, P < .05 vs baseline; 142 ± 25 to 131 ± 20 mL/m2 for captopril, P < .01. mean (SD). Captopril also reduced LV end-systolic volume index (98 ± 24 to 89 ± 21 mL/m2, P < .01 vs. baseline), whereas a nonsignificant trend was observed for the losartan group (97 ± 23 to 90 ± 16 mL/m2, P = not significant). The between-group differences in the changes in LV volumes were not statistically significant. After drug withdrawal, LV end-diastolic volume index remained significantly lower than baseline in the captopril group (P < .01). Conclusions: Both captopril and losartan prevent LV dilation, representing adverse ventricular remodeling, previously seen with placebo treatment. Reverse remodeling was observed in the captopril group. On the basis of these results, the relative effects on LV remodeling do not provide a rationale for a survival benefit of losartan over captopril.