124 patients undergoing total hip replacement were randomly allocated to receive Kabi 2165, 2,500 anti-Xa units twice a day (group A); Kabi 2165, 2,500 anti-Xa units twice a day during the first 48 h postoperatively and then 5,000 anti-Xa units once a day (group B), or adjusted-dose standard heparin, monitored by activated partial thromboplastin time (group C). The first dose was given 2 h before surgery in the three groups. Deep-vein thrombosis (DVT) was detected by radiolabelled fibrinogen uptake and bilateral venography was performed in patients who had a positive scan. In patients who had a negative scan, bilateral venography was performed routinely the day before discharge from hospital. The frequency of DVT demonstrated by venography was 4.9% in group A, 7.3% in group B and 10% in group C. The difference between the three groups was not statistically significant. The incidence of proximal DVT was 2.4, 2.4 and 7.5%, respectively, for the three groups. There was no significant difference between the three groups with respect to mean estimated blood loss, the number of blood units transfused, wound hematoma formation, or hemoglobin and hematocrit levels.