A simple venous thrombosis model in rabbits was used for the quantitative evaluation of the thrombolytic effect of human extrinsic (tissue-type) plasminogen activator as compared with urokinase.A thrombus was formed in an isolated segment of the jugular vein from a mixture of (125)I-labeled fibrinogen, whole rabbit blood, and thrombin. In order to immobilize the thrombus during lysis, it was formed around a woolen thread introduced longitudinally in the lumen of the vein. Thrombotic extension of the clot was prevented by subcutaneous injection of heparin. The extent of thrombolysis was measured as the difference between the radioactivity introduced in the clot and that recovered in the vein segment at the end of the experiment. In control animals the extent of thrombolysis was 5.6+/-1.4% (n = 5) after 6 h, 14.5+/-1.7% (n = 10) after 30 h, 16.0+/-1.5% (n = 11) after 78 h, and 48.1+/-2.7% (n = 10) after 174 h (mean+/-SEM). Extrinsic (tissue-type) plasminogen activator, highly purified from the culture fluid of a human melanoma cell line, was administered systemically or locally over a time period of 4 h and the percent thrombolysis measured 2 h after the end of the infusion. One- and two-chain extrinsic plasminogen activator had very similar thrombolytic potency. Systemic infusion resulted in a dose-dependent degree of thrombolysis. The activator-induced thrombolysis, after infusion of 100,000 IU ( congruent with1 mg protein), was approximately 75% for fresh clots, 35% for 1-d-old clots, 30% for 3-d-old clots, and 50% for 7-d-old clots. The thrombolytic activity of urokinase was more than five times lower than that of extrinsic plasminogen activator: Infusion of 500,000 IU resulted in approximately 40% lysis of fresh clots and 25% of 1-3-d-old clots, while 7-d-old clots appeared to have become resistent to urokinase. Local infusion resulted in a 5-10 times higher thrombolytic effect of both extrinsic plasminogen activator and urokinase. Thrombolysis with extrinsic plasminogen activator was not associated with systemic activation of the fibrinolytic system as evidenced by unaltered plasma levels of fibrinogen, plasminogen, and alpha(2)-antiplasmin. Systemic infusion of urokinase resulted in significant thrombolysis only at doses that were associated with disseminated plasminogen activation. Local infusion of urokinase required a 5-10-fold higher dose than extrinsic plasminogen activator to obtain a similar degree of thrombolysis, which also occurred in the absence of systemic activation of the fibrinolytic system. It is concluded that the extent of thrombolysis by extrinsic plasminogen activator is mainly determined by the dose of activator and its delivery in the vicinity of the thrombus and much less by the age of the thrombus or the molecular form of the activator. Extrinsic plasminogen activator appears to be superior to urokinase because of its higher (5-10-fold) specific thrombolytic activity and the absence of systemic activation of the fibrinolytic system, which results in defibrinogenation and a bleeding tendency.