Low molecular weight heparin (LMWH) is effective in the treatment of acute deep vein thrombosis (DVT) in adults. This has not been demonstrated for one LMWH alone. The relationship between venographic changes due to LMWH therapy and clinical outcome in the initial treatment period has not been reported. A pooled analysis of two clinical trials was performed. The trials compared a fixed-dose, body weight-independent, subcutaneous LMWH, certoparin (8000 antifactor Xa [aXa] U twice a day [b.i.d.]), with an adjusted-dose intravenous unfractionated heparin (UFH) with respect to venographic changes expressed as Marder score and occurrence of recurrent venous thromboembolism, major bleeding, and mortality. The Marder score was 23.2 +/- 8.4 in patients randomized to LMWH (n = 299 paired phlebograms) and 23.9 +/- 8.9 in patients allocated to UFH (n = 297 paired phlebograms) at entry (2p = 0.23) and 18.9 +/- 9.7 and 20.5 +/- 9.9 at the end of the initial therapy (2p = 0.04), respectively. The composite outcome of recurrent venous thromboembolism, major bleeding, and mortality occurred less frequently during treatment with LMWH (n = 393) than it did with UFH (n = 404, 1.3% versus 5.0%, risk reduction [RR] 0.26, 95% confidence interval [CI] 0.11 to 0.63, 2p = 0.004). Single events of recurrent thromboembolism (2p = 0.12), major bleeding (2p = 0.03), and mortality (2p = 0.12) were observed less frequently with LMWH. A trend toward a lack of regression of thrombus size was observed in recurrent venous thromboembolism (2p = 0.08). Body weight-independent LMWH significantly reduces thrombus size and the incidence of composite outcome during the initial treatment of acute proximal venous thrombosis compared with adjusted dose intravenous UFH. The data indicate a relation between an unimproved Marder score and a recurrent venous thromboembolism.