A multicentre, double-blind, randomised trial was conducted to compare the efficacy of a low-molecular-weight (LMW) heparin, Logiparin, with that of an unfractionated (UF) heparin in the prophylactic treatment of thrombosis in patients undergoing general surgery. A total of 1,290 patients were randomised to receive a single daily dose of Logiparin (2,500 IU: 431 patients; 3,500 IU: 430 patients) or UF heparin (2 × 5,000 IU: 429 patients). The incidence of the main end point, deep venous thrombosis, was found to be significantly different between the groups (p = 0.03), whereas the incidence of severe haemorrhage was not (p = 0.5). The plasma anti-Xa activity was found to be correlated with body weight, but correlated only very weakly with antithrombotic activity (p = 0.045) after adjustment in a stepwise multivariate analysis, and did not significantly correlate with the incidence of haemorrhage. Logiparin at 3,500 IU and UF heparin showed similar efficacy. Although a correlation between plasma anti-Xa activity and body weight was observed, there is not sufficient evidence to recommend the adjustment of the Logiparin dose on patient’s weight for prophylaxis in general surgery patients.