Twelve patients with frequent ventricular premature depolarizations (VPDs) received amiodarone, 600 mg/day, for up to 8 weeks. On days 0, 1, 4, 8, 15, 22, 36, and 57 of treatment, 24-hour ambulatory ECGs were obtained, and multiple blood samples were taken for determination of amiodarone and desethylamiodarone plasma concentrations. All patients had at least 75% suppression of VPDs. The mean duration of therapy before the onset of antiarrhythmic effect was 13.2 days (range 1 to 36 days). Trough amiodarone and desethylamiodarone plasma concentrations at the time of onset of antiarrhythmic effect were 0.86 +/- 0.48 mg/L and 0.23 +/- 0.15 mg/L, respectively. Sixty-seven percent of patients responded at amiodarone concentrations below 1.0 mg/L. For each patient there was a progressive decrease in frequency of VPDs as both amiodarone and desethylamiodarone concentrations increased. Regression modeling indicated that both amiodarone and desethylamiodarone plasma concentrations explained significant variability in the frequency of VPDs, and amiodarone and desethylamiodarone plasma concentrations were highly correlated with each other. There was a trend for desethylamiodarone to explain more variability in frequency of VPDs than amiodarone.