We randomly assigned 56 patients who presented within 12 hours of their first symptoms of acute myocardial infarction to treatment with either intracoronary streptokinase or coronary angioplasty. The mean (+/- SD) duration of symptoms (3.0 +/- 1.2 hours in the group treated with angioplasty vs. 3.6 +/- 1.8 in the group treated with streptokinase; P not significant) and time to recanalization (4.1 +/- 1.4 hours vs. 4.8 +/- 1.7 hours; P not significant) were similar in both groups. Coronary recanalization was achieved in 83 percent of the patients treated with angioplasty and in 85 percent of those treated with streptokinase (P not significant). Residual luminal stenosis in the coronary artery was significantly decreased after angioplasty, as compared with streptokinase therapy (43 +/- 31 percent of patients vs. 83 +/- 17; P less than 0.001). Residual stenosis of 70 percent or more was present in 4 percent of the angioplasty-treated patients and in 83 percent of the streptokinase-treated patients (P less than 0.01). Ventricular function after therapy was assessed by serial contrast ventriculograms. Increases in both global ejection fraction (8 +/- 7 percent vs. 1 +/- 6; P less than 0.001) and regional wall motion (+1.32 +/- 1.32 SD vs. +0.59 +/- 0.79 SD; P less than 0.05) were greater for the angioplasty group. We conclude that angioplasty and streptokinase produce similar rates of early coronary reperfusion during evolving transmural myocardial infarction. However, angioplasty is significantly more effective in alleviating the underlying coronary stenoses, and this may result in more effective preservation of ventricular function after therapy.