The success of current thrombolytic strategies is undermined by ongoing thrombin activity, but it is uncertain whether prevention of thrombin generation or direct thrombin antagonism is effective in achieving more optimal thrombolysis. To address this question, 24 dogs with electrically induced coronary thrombus undergoing thrombolysis with tissue-type plasminogen activator (1 mg/kg) over 20 min, were given one of the following adjunctive regimens in a random fashion. Twelve dogs received saline, and served as the control group; a direct thrombin antagonist, hirudin, was given at a dose of 20 μg/kg/min for 90 min to six dogs, and a selective factor Xa inhibitor, tick anticoagulant peptide (TAP), was administered to six dogs at a dose of 30 μg/kg/min for 90 min. The time to reperfusion was similar in the saline and hirudin groups (34 ± 4vs37 ± 7 min;P=NS) but shorter in the TAP group (21 ± 4 min;P< 0.05). Coronary blood flow was restored to 100% of its baseline value for 7 ± 2 min in control dogs, and for 20 ± 6 min in the hirudin group (P< 0.05). In the TAP group, coronary blood flow was restored to 100% of its baseline value for more than 120 min in all dogs (P< 0.01vsothers treatments). Reocclusion occurred in 89% and 50% of dogs receiving saline and hirudin, respectively (P= NS), but in none of the TAP-treated dogs (P< 0.01). Plasma fibrinopeptide A (FpA) and thrombin-antithrombin III complex (TAT) levels were determined in all dogs as indicators of thrombin activation. In the saline group, FpA and TAT during reperfusion were 19 ± 2 ng/ml and 104 ± 24 ng/ml respectively (P< 0.02vsbaseline) indicating high thrombin activity. In contrast, during reperfusion in hirudin-treated dogs FpA and TAT remained similar to baseline (10 ± 3 ng/ml and 53 ± 4 ng/ml respectively; bothP< 0.05vssaline). Reperfusion in TAP-treated dogs did not alter FpA and TAT in plasma, which remained similar to baseline (9 ± 1 ng/ml and 39 ± 5 ng/ml respectively; bothP< 0.05 vs saline). Scanning electron microscopy of coronary arteries showed residual thrombi with intense platelet and fibrin deposition adherent to the deendothelialized surface of the vessels following saline and hirudin therapy. In contrast, TAP-treated arteries were characterized by the absence of fibrin and minimal platelet deposition. In conclusion, these hemodynamic, biochemical and morphologic data suggest that adjunctive treatment with a higher tier blockade of the coagulation cascade is superior to direct thrombin inhibition in maintaining coronary artery patency following thrombolysis in the experimental canine electrolytic model. These findings highlight the potential adverse effects of unchecked thrombin generation in the setting of thrombolytic therapy.