During a double-blind trial in which patients with suspected myocardial infarction received metoprolol or placebo, we analyzed the occurrence of ventricular tachyarrhythmias. Metoprolol (15 mg intravenously) was given as soon as possible after admission, and thereafter 200 mg was given daily for three months. Antiarrhythmic drugs were given only for ventricular fibrillation and sustained ventricular tachycardia (greater than 60 beats per second). Definite acute myocardial infarction developed in 809 of the 1395 participants, and probable infarction in 162. Metoprolol did not influence the occurrence of premature ventricular contractions or short bursts of ventricular tachycardia. However, there were 17 cases of ventricular fibrillation in the placebo group (697 patients) and only 6 in the metoprolol group (698 patients, P less than 0.05). During the hospital stay significantly fewer patients receiving metoprolol (16) than placebo (38) (P less than 0.01) required lidocaine. In a separate analysis of 145 patients, metoprolol did not influence the occurrence of premature ventricular contractions or short bursts of ventricular tachycardia during the first 24 hours of treatment. Despite a lack of effect on less serious ventricular tachyarrhythmias, metoprolol had a prophylactic effect against ventricular fibrillation in acute myocardial infarction.