We have compared, in an open randomized study, the effects of sodium nitroprusside (SNP) and urapidil on haemodynamic state and myocardial function and metabolism in two groups of patients undergoing elective coronary artery surgery. Sixty patients were allocated randomly to one of two groups: group SNP (n = 29) received SNP at an initial rate of 1-2 micrograms kg-1 min-1; group URA (n = 31) received one or more bolus injections of urapidil 25 mg and an i.v. infusion at an initial rate of 11-21 micrograms kg-1 min-1. Baseline measurements were obtained 10 min after introduction of an echotransducer into the oesophagus. Subsequently, vasodilator therapy was started in both groups. Infusion rates were adjusted to maintain systolic arterial pressure at 80-120% of baseline values (or mean arterial pressure < 100 mm Hg). Additional measurements were obtained 10 min after the start of vasodilator therapy and after sternotomy when the pericardium was opened. At each measuring time a complete haemodynamic profile, coronary sinus blood flow (CSBF) curves, transoesophageal echocardiographic images, and arterial and coronary venous blood samples were obtained. Arterial pressure was controlled adequately in both groups. After sternotomy, heart rate and cardiac index increased in both groups. At that time, there was a significant increase in myocardial oxygen consumption and CSBF in group URA (P < 0.05). However, the ratio between myocardial oxygen demand and oxygen supply remained unchanged and there was no difference in the number of ischaemic episodes between the groups.