In the Göteborg Metoprolol Trial, 1395 patients with suspected acute myocardial infarction were, on admission, randomly allocated to double-blind treatment, 697 to placebo and 698 to metoprolol (15 mg i.v. + 200 mg/day) for 90 days. During this period, there were 62 deaths in the placebo group (8.9%) and 40 in the metoprolol group (5.7%), a mortality reduction of 36% (p less than 0.03). This effect persisted regardless of age, previous infarction or previous chronic beta blockade. All deaths were classified as cardiovascular. After 3 months, all patients were recommended open treatment with metoprolol, and the difference in mortality between the two groups was maintained after 1 year. Early institution of metoprolol (within 12 hours) influenced infarct development during the first 3 days (infarct diagnosis and indirect measures of infarct size). Metoprolol also reduced the incidence on fatal and nonfatal infarction during the next 4-90 days by 35%. Furthermore, fewer episodes of ventricular fibrillation were recorded in the metoprolol than in the placebo group (six vs 17 patients). The tolerance was judged to be very good. The same percentage of patients (19%) was withdrawn from the blind treatment in the two groups. Fewer patients in the metoprolol group used lidocaine, furosemide and analgesics. We conclude that metoprolol therapy instituted on admission in patients with suspected acute myocardial infarction reduced 3-month mortality and exerted beneficial clinical effects.