The efficacy of recombinant vampire bat salivary plasminogen activator (bat-PA) as a thrombolytic agent was compared with that of human tissue-type plasminogen activator (t-PA) in a canine model of arterial thrombosis. An occlusive thrombus was formed in the femoral artery by insertion of a thrombogenic copper coil; femoral arterial blood flow was monitored with a Doppler flow meter. Bat-PA and t-PA, when administered by 5-minute intravenous infusion (14 nmol/kg), reperfused seven out of eight and four out of eight dogs, respectively. The median reperfusion times in the bat-PA and t-PA groups were 24 and greater than or equal to 131 minutes, respectively. The mean reperfusion times (+/- SEM) in the recanalized bat-PA- and t-PA-treated dogs were similar (20 +/- 5 and 11 +/- 2 minutes, respectively, p = NS). Maximal blood flow after reperfusion was greater with bat-PA than with t-PA (80 +/- 10% and 41 +/- 15% of control flow, respectively, p less than 0.05). Furthermore, the median reocclusion time was markedly delayed in the bat-PA group relative to the t-PA group (131 versus 34 minutes, respectively, p less than 0.05). Plasma fibrinogen and plasminogen were not significantly depleted by the administration of t-PA or bat-PA. However, plasma alpha 2-antiplasmin activity was moderately depressed in the t-PA group relative to the bat-PA group (p less than 0.05). The clearance profile for t-PA was monoexponential, with a half-life (t1/2) of 2.4 +/- 0.3 minutes and a mean residence time of 3.5 +/- 0.4 minutes. The clearance profile for bat-PA was biexponential, with a t1/2 alpha of 0.9 +/- 0.2 minutes, a t1/2 beta of 20.2 +/- 2.7 minutes, and a mean residence time of 21.3 +/- 4.3 minutes. The steady-state volume of distribution displayed by bat-PA was 16-fold greater than that of t-PA. Zymography of serial plasma samples from the bat-PA-treated dogs failed to demonstrate the apparent generation of a complex between bat-PA and plasminogen activator inhibitor-1; the corresponding complex with t-PA was observed in plasma samples from the t-PA-treated dogs. The sustained recanalization and improved blood flow in the bat-PA group relative to the t-PA group and the avoidance of fibrinogenolysis by bat-PA, despite its prolonged mean residence time, suggest that bat-PA may be superior to t-PA as a thrombolytic agent.