The true incidence of thromboembolic complications following multiple trauma is unknown, and no method of prophylaxis has been shown to be both safe and effective in managing seriously injured patients. In this prospective study, 113 trauma patients were assigned on admission to receive either low-dose heparin (LDH), (5,000 U subcutaneously every 12 hours) or to wear sequential compression devices (SCDs) as prophylaxis against the development of deep venous thrombosis (DVT). Both groups of patients were serially studied with duplex venous ultrasound imaging to detect thrombus in the veins of the thigh. Ventilation-perfusion lung scans and pulmonary angiograms were performed when pulmonary embolism (PE) was suspected clinically. There were 12 patients who had thromboembolic complications, including 9 of 76 in the SCD group (12%) and 3 of 37 in the LDH group (8%). Five patients had DVT only, four had PE without detectable DVT, and three had both DVT and PE. None of the patients with PE died, and there were no major complications associated with either method of prophylaxis. Compared with the patients who did not develop DVT/PE, those with thromboembolic complications were older (49 +/- 23 vs. 36 +/- 17 years, p less than 0.02), spent more hospital days immobilized (24 +/- 15 vs. 10 +/- 13 days, p less than 0.001), received more transfusions (11 +/- 12 vs. 3 +/- 5 U, p less than 0.001) and had clotting abnormalities on admission, as demonstrated by prolonged PTT values (39 +/- 28 vs. 26 +/- 5 seconds, p less than 0.001). It appears that there is an identifiable subgroup of injured patients at highest risk for PE who warrant both prophylaxis and close surveillance for DVT.