Recognition that myocardial infarction is caused by coronary thrombosis has stimulated a search for a safe, rapidly acting, and effective thrombolytic regimen. Tissue plasminogen activator (t-PA) can provide relatively clot-selective thrombolysis, but one quarter of patients fail to achieve reperfusion, lysis speed is not optimal, and higher doses have been associated with an increased incidence of hemorrhagic stroke. We report the results of a multicenter study of pro-urokinase, a second naturally occurring plasminogen activator that has structural similarities to t-PA but has a different mechanism of action. Pro-urokinase was administered 3.9 +/- 1.1 hours after the onset of chest pain to 40 patients with acute myocardial infarction with angiographically confirmed complete coronary occlusion (TIMI grade 0). After a 90-minute intravenous infusion of pro-urokinase (4.7-9 million units, 36-69 mg) 51% (20 of 39) of the patients demonstrated reperfusion (TIMI grade 2 or 3) occurring 64.8 +/- 22.3 minutes after initiation of therapy. Fibrinogen levels fell only 10 +/- 17% from baseline, confirming the fibrin specificity of pro-urokinase. As with t-PA, however, this specificity was only relative. alpha 2-Antiplasmin decreased to 39% and plasminogen decreased to 64% of initial values. Fibrinogen degradation products increased 63% and the fibrin-specific D-dimer increased 8.7-fold. Thus, pro-urokinase produces relatively clot-selective coronary thrombolysis similar to that produced by t-PA, but the use of either pro-urokinase or t-PA alone in higher doses would be likely to produce more nonspecific effects.