A total of 1290 patients (Pts) undergoing general surgery were enrolled in a randomized, multicentre double-blind study in order to investigate the efficacy and safety of two different doses of a low molecular weight heparin (LMWH) (Logiparin) for the prevention of deep vein thrombosis. Patients were randomized to either 5,000 IU unfractionated heparin twice daily, 2,500 anti-Xa or 3,500 anti-Xa units of Logiparin once daily. Each treatment was given subcutaneously two hours before surgery and continued for seven to ten days. All coagulation tests were performed blindly in a core laboratory. Blood samples were collected before surgery and then 3 hours after injection on Day 3 and 5 after surgery. Anti-Xa amidolytic activities were significantly higher in the two LMW Heparin groups than in the unfractionated heparin group (mean peak levels +/- s.e.m. on Day 3: 0.097 +/- 0.004; 0.152 +/- 0.004 and 0.034 +/- 0.003 IU respectively). As expected a significant correlation was observed between anti-Xa activity and the dose of LMW Heparin injected. The correlation coefficient was higher when the doses were expressed in anti-Xa units/kg body weight. However, the body weight accounts for only 16% of the interindividual variability of anti-Xa activity. Therefore, there is no clear evidence to suggest that weight-adjusted doses should be recommended when this LMW Heparin is used as prophylactic treatment in general surgery. A weak negative correlation was found between anti-Xa activity and thrombosis as demonstrated by a positive radiolabelled fibrinogen uptake test and confirmed by positive phlebography. No significant correlation was demonstrated between anti-Xa activity and the occurrence of postoperative bleeding.