To evaluate the effectiveness of intravenous and intrathrombic injection of the thrombin inhibitor argatroban during pulse-spray pharmacomechanical thrombolysis (PSPMT) in experimental venous thrombosis. Clots were produced in the inferior vena cava in 52 rabbits by placement of steel coils and balloon injury to the vessel wall. Two days later, clots were treated with PSPMT. Several treatment methods were used: intrathrombic saline, intrathrombic tissue plasminogen activator (t-PA), intrathrombic t-PA with intrathrombic and intravenous heparin, intrathrombic t-PA with intravenous argatroban, and intrathrombic t-PA with intrathrombic and intravenous argatroban at two different doses. After treatment, the rabbits were killed and residual clot was weighed. Pretreatment clot weight was estimated and clot lysis was assessed. PSPMT with t-PA and adjunctive intrathrombic heparin resulted in greater lysis than PSPMT with only t-PA (percentage of residual clot, 59% +/- 14 vs 81% +/- 28; P = .02). Addition of intravenous argatroban did not increase lysis, but adjunctive intrathrombic argatroban significantly increased lysis at low doses (37% +/- 16; P = .02) and high doses (34% +/- 6; P = .006) compared with t-PA and intrathrombic heparin. In a rabbit model of venous thrombosis, the use of intrathrombic argatroban during PSPMT with t-PA significantly improved clot lysis.