A prospective, randomized trial was done to evaluate the prevalence of deep venous thrombosis following primary unilateral or bilateral total hip arthroplasty with use of hypotensive epidural anesthesia, external pneumatic-compression boots, and aspirin (Group I) and with use of hypotensive epidural anesthesia and aspirin (Group II). All operations were performed by two of us (E. A. S. and T. P. S.) through a posterolateral approach. Two hundred and thirty-one patients who were more than thirty-nine years old and who had a total of 250 primary total hip arthroplasties were included in the study. There were 113 patients (124 hips) in Group I and 118 patients (126 hips) in Group II. All patients had venography on the sixth, seventh, or eighth postoperative day. Group I had no proximal thrombi, seven distal thrombi (6 per cent), and one late pulmonary embolus (1 per cent). Group II had one proximal thrombus (popliteal) (1 per cent), eight distal thrombi (6 per cent), and one late pulmonary embolus (1 per cent). The difference was not significant (p = 0.65). However, a significant difference may have been noted if the study population had been larger. The combination of hypotensive epidural anesthesia and aspirin is effective prophylaxis against deep venous thrombosis in patients who have a primary total hip arthroplasty. The extremely low rate of deep venous thrombosis in the present study may be attributed to the use of hypotensive epidural anesthesia and the associated decrease in blood loss and transfusion requirements.(ABSTRACT TRUNCATED AT 250 WORDS)