Low-molecular-weight heparins may simplify the management of deep venous thrombosis. A critical clinical issue is whether this more convenient therapy is as safe and effective as treatment with unfractionated heparin. To compare the safety and efficacy of low-molecular-weight heparins with those of unfractionated heparin for treatment of acute deep venous thrombosis. Reviewers identified studies by searching MEDLINE, reviewing references from retrieved articles, scanning abstracts from conference proceedings, and contacting investigators and pharmaceutical companies. Randomized, controlled trials that compared a low-molecular-weight heparin preparation with unfractionated heparin for treatment of acute deep venous thrombosis. Two reviewers extracted data independently. Reviewers evaluated study quality using a validated four-item instrument. Eleven of 37 studies met inclusion criteria for three major outcomes. Most studies used proper randomization procedures, but only one was double-blinded. Compared with unfractionated heparin, low-molecular-weight heparins reduced mortality rates over 3 to 6 months of patient follow-up (odds ratio, 0.71 [95% CI, 0.53 to 0.94]; P = 0.02). For major bleeding complications, the odds ratio favored low-molecular-weight heparins (0.57 [CI, 0.33 to 0.99]; P = 0.047), but the absolute risk reduction was small and not statistically significant (0.61% [CI, -0.04% to 1.26%]; P = 0.07). For preventing thromboembolic recurrences, low-molecular-weight heparins seemed as effective as unfractionated heparin (odds ratio, 0.85 [CI, 0.63 to 1.14]; P > 0.2). Low-molecular-weight heparin treatment reduces mortality rates after acute deep venous thrombosis. These drugs seem to be as safe as unfractionated heparin with respect to major bleeding complications and appear to be as effective in preventing thromboembolic recurrences.