A multicenter double-blind, randomized, placebo-controlled trial of oral enoximone, a phosphodiesterase inhibitor, was conducted in 102 outpatients (50 receiving enoximone and 52 receiving placebo) with moderate to moderately severe congestive heart failure. All were on a long-term regimen of digoxin and diuretics without vasodilators and converting enzyme inhibitors. Symptom score was obtained, and exercise testing was performed monthly for 4 months. There were no differences between groups in symptoms or exercise duration at the end of 4 months. A subgroup undergoing analysis of oxygen consumption with measurement of anaerobic threshold during exercise showed an increase (p less than 0.05) in anaerobic threshold at 1 month with enoximone. (2.7 +/- 0.8 ml O2/kg/min) compared with placebo (-0.8 +/- 1.2 ml O2/kg/min). This improvement was not sustained at 4 months (0.5 +/- 1.7 ml O2/kg/min with enoximone and 0.2 +/- 1.5 ml O2/kg/min with placebo). The dropout rate was significantly higher (p less than 0.02) with enoximone (46%) than with placebo (25%). Adverse effects other than death were slightly, but not significantly, higher with enoximone (32%) than with placebo (22%). During therapy, five deaths occurred in the enoximone group, and none occurred in the placebo group (p less than 0.05). Two deaths were sudden, two were from progressive congestive heart failure, and one was from acute myocardial infarction. With intention-to-treat analysis and inclusion of patients who were removed from therapy because of lack of study drug effect, 10 deaths occurred in the enoximone group, and three occurred in the placebo group (p less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)