Background: Low-molecular-weight heparins are frequently used to treat venous thromboembolism, but optimal dosing regimens and clinical outcomes need further definition. Methods: In this multicenter, open-label study with blinded adjudication of end points, we randomly assigned patients with acute deep-vein thrombosis to one of three treatment regimens: intravenous administration of unfractionated heparin; subcutaneous administration of a low-molecular-weight heparin, reviparin, twice a day for one week; or subcutaneous administration of reviparin once a day for four weeks. The primary end point was evidence of regression of the thrombus on venography on day 21; secondary end points were recurrent venous thromboembolism, major bleeding within 90 days after enrollment, and death. Results: Of the patients receiving unfractionated heparin, 40.2 percent (129 of 321) had thrombus regression, as compared with 53.4 percent (175 of 328) of the patients receiving reviparin twice daily and 53.5 percent (167 of 312) of the patients receiving reviparin once daily. With regard to thrombus regression, reviparin administered twice daily was significantly more effective than unfractionated heparin (relative likelihood of thrombus regression, 1.28; 97.5 percent confidence interval, 1.08 to 1.52), as was reviparin administered once daily (relative likelihood, 1.29; 97.5 percent confidence interval, 1.08 to 1.53). Mortality and the frequency of episodes of major bleeding were similar in the three groups. Conclusions: In acute deep-vein thrombosis, reviparin regimens are more effective than unfractionated heparin in reducing the size of the thrombus. Reviparin is also more effective than unfractionated heparin for the prevention of recurrent thromboembolism and equally safe. (N Engl J Med 2001;344:626-31.) Copyright (C) 2001 Massachusetts Medical Society.