A controlled double-blind trial was carried out to assess the efficacy and safety of a continuous intravenous infusion of diltiazem in preventing perioperative myocardial ischaemia in patients with coronary artery disease. Sixty-six patients undergoing non cardiac surgical procedures (vascular surgery, n = 37; other, n = 29) were randomly chosen to receive either diltiazem (group D, n = 32); or placebo (group P, n = 34); there was no difference between these groups in the number of patients in each NYHA class (I: 13/16; II: 14/14; III: 5/4) or having had a previous myocardial infarct (20/22). ECG leads CM5 and CL5 were recorded continuously with an ICR 7200 Holter monitor. After starting recording, either placebo or a loading dose (0.5 mg.kg-1) of diltiazem was given, followed by an infusion of 5 micrograms.kg-1.min-1. Anaesthesia was induced by thiopentone and suxamethonium, and maintained with nitrous oxide (50%), fentanyl and either halothane or droperidol. The number of myocardial ischaemic episodes was significantly (p less than 0.05) lower in group D (2 ST depressions in two patients) than in group P (8 ST depressions in six patients, 2 myocardial infarcts and 1 pulmonary oedema). No conduction disturbance was observed; the lowest cardiac frequency was found in group P (32 b.min-1). Systolic and diastolic arterial blood pressures were lower in group D than in group P, but no difference was found in heart rate and rate-pressure product.(ABSTRACT TRUNCATED AT 250 WORDS)