BACKGROUND Among survivors of acute myocardial infarction, frequent and repetitive ventricular premature depolarizations (VPDs) detected on ambulatory monitoring contribute independently to the risk of all-cause mortality and sudden death. Apart from the beta-blockers, no antiarrhythmic drug has been reliably demonstrated to reduce mortality among patients with VPDs. A pilot study was undertaken to gather data to aid in the design of a multicenter trial of amiodarone for the reduction of mortality from cardiac arrhythmias in such patients. RESULTS Seventy-seven patients with acute myocardial infarction within the previous 6-30 days and 10 or more VPDs/hr or one or more runs of ventricular tachycardia on 24-hour electrocardiographic recording were randomized in a double-blind fashion in a 2:1 amiodarone-to-placebo ratio. The loading dose was 10 mg/kg/day for 3 weeks. The maintenance dose was 300-400 mg/day with reductions at 4-month intervals in response to VPD suppression, excessive plasma levels, or toxicity. VPD suppression at 1 week and 2 weeks was 63% and 85%, respectively, on amiodarone and 17% and 27%, respectively, on placebo. Apart from thyroid-stimulating hormone elevation and skin reactions, no side effects occurred more frequently with amiodarone. The study drug was stopped for side effects or noncompliance in 35% of amiodarone patients and 34% of placebo patients. Patients were followed for a maximum of 2 years (mean, 20 months). Arrhythmic death or resuscitated ventricular fibrillation occurred in two of 48 amiodarone patients (6%) and four of 29 placebo patients (14%), whereas the rates of all-cause mortality were five of 48 (10%) and six of 29 (21%), respectively. CONCLUSIONS Amiodarone, in moderate loading and maintenance dosages with adjustments in response to plasma levels, VPD suppression, and side effects, results in effective VPD suppression and acceptable levels of toxicity.