The incidence of early reocclusion is reported to be higher in patients who receive fibrin-specific thrombolytic agents than nonspecific ones. The reason has yet to be clarified. In the present study, we focused on the difference in duration of fibrinolytic activity. The hemostatic parameters of 7 consecutive patients suffering from acute myocardial infarction treated with a fibrin-nonspecific thrombolytic agent (urokinase) were compared with 9 patients who received a fibrin-specific agent (tissue plasminogen activator, t-PA). The plasma concentrations of α2-plasmin inhibitor (α2-PI), plasmin α2-PI complex (PIC), fibrin degradation products E fragment (FDP-E), and D-D dimer (D-dimer) were measured before, soon after, 1 2, 3, 4, and 6 h and 2, 3, 4, and 7 days after thrombolytic therapy to estimate the hemostatic and fibrinolytic state. A significant decrease in α2-PI (less than the lowest measurable level) with a simultaneous increase in FDP-E and D-dimer was induced soon after the administration of urokinase. FDP-E and D-dimer decreased, with a significant increase in α2-PI, more than 6 h after thrombolytic therapy. In contrast, a less significant decrease in α2-PI with a lesser amount and shorter duration of fibrinolysis were observed in patients who received t-PA. The amount of PIC soon after drug administration was not different between the two groups. Our data suggested that fibrinolytic activities induced by fibrin-nonspecific urokinase persisted longer than expected by its plasma half-life. The fibrinolytic activities might be terminated by the production of α2-PI. This prolonged activation of the fibrinolytic system might be effective for preventing early reocclusion.