LMWH better than unfractionated heparin for preventing DVT in medical patients

Clinical Question

Which pharmacological agents most effectively prevent venous thromboembolism in hospitalized medical patients?

Bottom Line

Low-molecular-weight heparin (LMWH) was more effective than unfractionated heparin (UFH) for reducing deep venous thrombosis (DVT) and injection site hematoma. Bleeding rates were comparable with the 2 agents. (LOE = 1a)


Wein L, Wein S, Haas SJ, Shaw J, Krum H. Pharmacological venous thromboembolism prophylaxis in hospitalized medical patients: a meta-analysis of randomized controlled trials. Arch Intern Med 2007;167(14):1476-1486.  [PMID:17646601]

Study Design

Meta-analysis (randomized controlled trials)




Inpatient (any location)


This meta-analysis included randomized controlled trials that compared either UFH, LMWH, or a selective factor Xa inhibitor with a control, or LMWH with UFH in hospitalized medical patients for the reduction of DVT, PE, or mortality after administration of therapy. Trials of central venous catheters in cancer patients, those involving patients in intensive care units, and studies with fewer than 30 patients were excluded. The authors performed a complete search of the English literature through June 2006 and 2 independent reviewers handled selection, validity assessment, and data abstraction. Meta-analysis was only done if at least 3 trials were identified for a particular outcome. Both UFH (14 trials) and LMWH (11 trials) were more effective than control for reducing the risk of DVT and PE, but not mortality. Both increased the risk of bleeding. Studies that compared UFH at a dose of 5000 units 3 times per day with placebo showed a greater reduction in DVT than those comparing a twice-daily regimen with placebo. Pooled results from the 10 trials directly comparing UFH with LMWH showed that LMWH reduced the risk of DVT [number needed to treat (NNT) = 59; 95% CI, 34 - 212] and injection site hematoma (NNT= 13; 95% CI, 0.7 - 20.6) more than UFH. There was no difference between agents in the rate of PE, mortality, total or major bleeding, or thrombocytopenia. Only 1 trial addressed an Xa inhibitor -- fondaparinux -- and found it effective in the prevention of VTE without an increase in bleeding over placebo. Publication bias could not be excluded.