The effect of enoximone was assessed by a randomised double blind trial versus placebo. The clinical status of the patients was evaluated by the NYHA classification and quality of life score. Inotropic state was estimated from the maximum acceleration of aortic and pulmonary blood flow recorded by Doppler echocardiography. Thirty patients with severe cardiac failure, aged 66.4 +/- 14 years, symptomatic despite maximal therapy associating diuretics, digitalis, nitrate derivatives and angiotensin converting enzyme inhibitors, were included. Fifteen patients were given enoximone 100 mg three times a day orally (Group E) and the other 15 were given a placebo (Group P). The NYHA class and quality of life scores were assessed at D0, D4 and D31. Doppler echocardiography and Holter recordings were performed on D0 and D31. The two groups were comparable at D0. Ten patients abandoned the trial, 3 from Group E (including 1 death) and 7 from Group P (including 3 deaths). At D4, 13 patients from Group E and 8 from Group P were clinically improved (p < 0.05). At D31, the clinical state was stable or improved in 10 of the 12 patients in Group E and 6 of the 8 patients in Group P (NS). No secondary effects were severe enough to warrant the withdrawal of treatment: the frequency of ventricular extrasystoles was comparable in the two groups at D0 and D31. At D31 the maximal aortic acceleration had increased by 20% compared with D0 (p < 0.05) and the maximal pulmonary acceleration by 31% (p < 0.05) in Group E. The same parameters showed no significant change in Group P (-6% and +5% respectively).(ABSTRACT TRUNCATED AT 250 WORDS)