To study the haemodynamic and neurohumoral effects of nisoldipine (2 X 10 mg) vs captopril (3 X 25 mg), 24 patients with heart failure (New York Heart Association class II and III) due to coronary artery disease were treated in a randomized double-blind trial over 3 months. Both drugs were well tolerated. Clinical status was similarly improved in both groups, nisoldipine exerted an additional antiischaemic effect. Nisoldipine lowered the mean arterial pressure and capillary wedge pressure acutely and also after long-term treatment. The increase in cardiac index and stroke volume index, however, which was pronounced after acute administration, was no longer present after 3 months of therapy at rest and was abolished during exercise. Norepinephrine plasma concentration increased after the first dose, plasma renin activity did not change, and aldosterone plasma concentration showed a small insignificant decrease. Urine concentrations of norepinephrine and vasopressin were slightly elevated after the 3-month therapy. After captopril, mean arterial pressure and pulmonary capillary wedge pressure decreased acutely and at follow up. Cardiac index and stroke volume index increased significantly only during exercise at follow-up. Plasma renin activity was significantly elevated and aldosterone plasma concentration only slightly lowered. In contrast to what was seen with nisoldipine, plasma norepinephrine concentration and urine catecholamine and vasopressin concentrations remained unchanged. In conclusion, the pronounced haemodynamic effects seen after the first dose of nisoldipine are mostly abolished after long-term treatment, probably due to neurohumoral counterregulation. The haemodynamic response to captopril is complete only after long-term treatment, without evidence of activation of the neurohumoral systems.