Background The study objective was to determine whether Hirulog, a direct thrombin inhibitor, has potential efficacy and safety in the prevention of deep vein thrombosis (DVT) in orthopedic patients.A phase 2 open-label, dose-escalating design was used to study 222 unselected patients undergoing major hip or knee surgery in tertiary-care, university-affiliated hospitals.Methods and Results Subcutaneous Hirulog was initiated postoperatively.Patients were evaluated for bleeding and symptomatic pulmonary embolism, and mandatory bilateral venography was performed before discharge. Dose escalations were made on the basis of observed rates of bleeding and venous thrombosis. There were five dosage regimens used: 0.3 mg/kg every 12 hours, 0.6 mg/kg every 12 hours, 1.0 mg/kg every 12 hours for 3 days followed by 0.6 mg/kg every 12 hours for up to 11 days, 1.0 mg/kg every 12 hours, and 1.0 mg/kg every 8 hours. One hundred seventy-seven patients who had technically adequate bilateral venography or objectively documented pulmonary embolism were included in the primary analysis of efficacy. The highest dosage regimen (1.0 mg/kg every 8 hours) provided the lowest rates of total DVT (17%) and proximal DVT (2%), both of which were significantly lower (P=.010 and P=.023, respectively) than the pooled rates of total (43%) and proximal (20%) DVT seen with the first four regimens. Bleeding rates were low (<5%) with all regimens.Conclusions This study demonstrates that 1.0 mg/kg Hirulog every 8 hours started postoperatively is potentially efficacious and safe for the prevention of DVT after major hip or knee surgery. (Circulation. 1994;90:2385-2389.)