237 patients with a non-traumatic disease of the hip requiring total hip replacement entered a multicentre trial. Treatments were allocated at random in a double-blind manner: 113 patients received 5000 IU of standard heparin t.d.s.; 124 patients received enoxaparin 40 mg once a day. Treatment lasted from 10 to 15 days until bilateral ascending phlebography of the lower limbs (BAPLL) was performed. Its effectiveness was assessed by the results of BAPLL, by evaluation of the risk of bleeding on examination of the surgical wound, by the units of blood required for perfusion and by daily blood count and differential. BAPLL was performed in 228 patients: 15 phlebographies (12.5 p. 100) were positive in the 120 assessed patients of the enoxaparin group, while 27 (25 p. 100) were positive in the 108 assessed patients of the standard heparin group (p = 0.014). Proximal deep vein thrombosis was significantly (p = 0.03) less frequent in the enoxaparin group (7.5 p. 100) than in the standard heparin group (18.5 p. 100). Pulmonary embolism was observed in a patient on standard heparin. The units of blood required for perfusion were significantly (p = 0.03) smaller in the enoxaparin group (3.37 +/- 1.81 U) than in the standard heparin group (3.84 +/- 1.70 U). Thus, one single subcutaneous injection of enoxaparin 40 mg daily was significantly more effective and safer than standard heparin 5000 IU t.d.s. in the prevention of deep vein thrombosis after total hip replacement.