The effect of metoprolol on chest pain was compared with that of placebo in all randomized patients. The pain score before and 15 minutes after the injection of trial medication was registered and a reduction in chest pain was observed in the metoprolol group. Increasing chest pain after blind injection was observed in only 16 and 9 patients from the placebo and metoprolol groups, respectively. Comparison with the placebo as well as detailed analysis of clinical data revealed that in these patients the increasing pain could not be explained by coronary spasm evoked by beta-blockade. Similarly, metoprolol did not exhibit any unfavorable effect on the 14 patients who were withdrawn (together with the 28 patients given placebo) from blind treatment due to angina pectoris. Either metoprolol does not induce coronary vasospasm or spasm does not play a role in these patients with definite and suspected acute myocardial infarction as well as unstable angina pectoris. Metoprolol reduced the need for analgesics during the first 4 days and shortened the duration of pain. The effects were similar in patients with early and late treatment, but may depend on initial heart rate, blood pressure and site of infarction.