Direct inhibition of thrombin with agents such as hirudin and argatroban reduces reocclusion rates during experimental coronary thrombolysis. We compared the adjunctive potential of the tripeptide thrombin inhibitor D-methyl-phenylalanyl-prolyl-arginal (LY294468) during thrombolysis with tissue-type plasminogen activator (t-PA) with the less specific tripeptide thrombin inhibitor Boc-D-phenylalanyl-prolyl-arginal (LY178207) and the standard anticoagulant heparin. The left circumflex coronary artery (LCX) was isolated proximal to the first main branch, and coronary blood flow (CBF) was measured in 26 anesthetized dogs. Thrombogenesis was initiated by electrolytic injury of the intimal surface of the artery, producing an occlusive thrombus. Thrombolytic/adjunctive therapy was started 1 h later in the following groups: (a) t-PA alone (0.9 mg/kg, 1-h infusion), (b) t-PA + LY294468 (0.5 or 1 mg/kg/h, 2-h infusion), (c) t-PA + LY178207 (0.5 or 1 mg/kg/h, 2-h infusion), and (d) t-PA + heparin (80 U/kg bolus + 30 U/kg/h, 2-h infusion). LY294468 provided antireocclusive efficacy (time to reocclusion = > 200 min as compared with 65 min for t-PA alone; six of nine patent vessels vs. zero of six, respectively, at the end of the experiment), with no bleeding liability during t-PA-induced thrombolysis. Heparin and LY178207 were ineffective adjunctive agents. Heparin, however, significantly increased template bleeding times. LY294468 was effective as an adjunctive agent during thrombolysis and may represent a safer (less bleeding) and more effective adjunctive agent than heparin.