To investigate the mechanisms of ventricular arrhythmia suppression by propranolol, we determined the antiarrhythmic efficacy of d-propranolol in 10 patients with frequent ventricular ectopic depolarizations (VEDs) and nonsustained ventricular tachycardia. After an initial placebo phase, 40 mg d-propranolol was administered orally every 6 h with dosage increased every 2 d until arrhythmia suppression (greater than or equal to 80% VED reduction), intolerable side effects, or a maximal dosage (1,280 mg/d) was reached. Response was verified by documenting return of arrhythmia during a final placebo phase. Arrhythmia suppression occurred in six patients while two more had partial responses. Effective dosages were 320-1,280 mg/d (mean 920 +/- 360, SD) of d-propranolol with corresponding plasma concentrations of 60-2,280 ng/ml (mean 858 +/- 681). For the entire group, the QTc interval shortened by 4 +/- 4% (P = 0.03). Arrhythmia suppression was accompanied by a reduction in peak heart rate during exercise of 0-29%. To determine whether arrhythmia suppression could be attributed to beta-blockade, racemic propranolol was then administered in dosages producing the same or greater depression of exercise heart rate. In 3/8 patients, arrhythmias were not suppressed by racemic propranolol indicating that d-propranolol was effective via a non-beta-mediated action. By contrast, in 5/8 patients racemic propranolol also suppressed VEDs. We conclude that propranolol suppresses ventricular arrhythmias by both beta- and non-beta-adrenergic receptor-mediated effects.