Current issues of ACP Journal Club are published in Annals of Internal Medicine


Meta-analysis: LMWH is effective in reducing recurrent thromboembolism, bleeding, and death in acute DVT

ACP J Club. 1996 Sept-Oct;125:30. doi:10.7326/ACPJC-1996-125-2-030

Source Citation

Siragusa S, Cosmi B, Piovella F, Hirsh J, Ginsberg JS. Low-molecular-weight heparins and unfractionated heparin in the treatment of patients with acute venous thromboembolism: results of a meta-analysis. Am J Med. 1996 Mar;100:269-77.



To compare the efficacy and safety of low-molecular-weight heparin (LMWH) with unfractionated heparin (UFH) in patients with venous thromboembolism.

Data sources

Studies were identified by searching MEDLINE and EMBASE (1980 to 1994) using the keywords low-molecular-weight heparin, enoxaparin, logiparin, CY 222, fraxiparine, and fragmin, associated with the terms treatment, thromboembolic disease, deep vein thrombosis, and pulmonary embolism. Additional studies were identified by scanning the reference lists and contacting the authors of all retrieved articles.

Study selection

Studies were selected if the patients had a first episode of acute deep venous thrombosis (DVT) objectively confirmed by venography or acute pulmonary embolism confirmed by high-probability lung scan or pulmonary angiography, a randomized comparison was done between LMWH and UFH, and an objective outcome measure was done.

Data extraction

Data were extracted on study design, drug regimen, and the rates of recurrent thromboembolism, bleeding, and mortality. Studies were classified as level 1 if blinded outcome assessment was done, and level 2 if blinded outcome assessment was not assured.

Main results

13 studies were included in the analysis. In the 3 studies with level-1 evidence, the incidence of recurrent venous thromboembolism during the first 15 days of therapy was 0.8% in the LMWH group compared with 3.2% in the UFH group (P = 0.02). {This absolute risk reduction (ARR) of 2.4% means that 41 patients would have to be treated (NNT) with LMWH (rather than UFH) to prevent 1 additional recurrence of venous thromboembolism at 15 days (95% CI 21 to 234); the relative risk reduction was 76%, CI 20% to 94%.}* The incidence of recurrent venous thromboembolism during 90 days of therapy was 2.7% in the LMWH group compared with 6.4% in the UFH group (P = 0.006) {ARR 3.7%; NNT 27, CI 14 to 136; RRR 61%, CI 20% to 70%}*. In the level-1 studies, the incidence of major bleeding was 3.0% in the LMWH group compared with 6.7% in the UFH group (P = 0.01). In level-2 studies, no significant differences existed in the rates of recurrent venous thromboembolism or major bleeding. Mortality in patients with cancer was 14% in the LMWH group and 28% in the UFH group (P = 0.01).


Low-molecular-weight heparin is more effective in reducing recurrent venous thromboembolism, major bleeding, and death than is unfractionated heparin in patients with acute venous thromboembolism.

Sources of funding: IRCCS Policlinico S. Matteo, Pavia, Italy, and Heart and Stroke Foundation of Canada.

For article reprint: Dr. J.S. Ginsberg, McMaster University Medical Centre, 1200 Main Street West, Room 3W15, Hamilton, Ontario, L8N 3Z5, Canada. FAX 905-521-4972.

*Numbers calculated from data in article.


What should we conclude from this meta-analysis of LMWH? LMWH preparations are not interchangeable. The clinical outcomes obtained with one LMWH may not be generalizable to other LMWHs. Siragusa and colleagues note that " ... the current findings should not be considered definitive because they were obtained by pooling the results of studies using different preparations of LMWH. Definitive recommendations await further studies that are large enough to demonstrate true benefits with individual LMWH." The role of meta-analyses that pool studies using different LMWH preparations is to identify hypotheses to be tested by further studies. For example, the observed difference in mortality in favor of LMWH is intriguing. Is it because of a reduction in fatal pulmonary embolism or the antimetastatic potential of the anticoagulant?

How should the clinician apply these data? The decision to use a particular LMWH should be based on the clinical trial data for that preparation. 3 LMWHs have been established to be at least as effective and safe as intravenous UFH by individual clinical trials that measured the outcomes of recurrent thromboembolism, bleeding, and mortality. Trends in favor of LMWH were observed in individual trials (1-5). These LMWH regimens are tinzaparin 175 F Xa units/kg subcutaneously once daily (1), enoxaparin 1 mg/kg subcutaneously twice daily (3), and nadroparin in a fixed dose subcutaneously twice daily according to weight category (4, 5). 2 recent trials (3, 5) established the feasibility, effectiveness, and safety of LMWH treatment at home for patients with DVT. Home treatment with LMWH should markedly reduce the cost of therapy without detracting from the patient's outcome and may potentially improve the outcome. LMWH is a major breakthrough in the clinician's ability to provide effective, safe, and efficient therapy for patients with DVT.

Gary E. Raskob, MSc
University of OklahomaOklahoma City, Oklahoma, USA


1. Hull RD, Raskob GE, Pineo GF, et al. N Engl J Med. 1992;326:975-82.

2. Simonneau G, Charbonnier B, Decousis H, et al. Arch Intern Med. 1993;153:1541-6.

3. Levine M, Gent M, Hirsh J, et al. N Engl J Med. 1996;334:677-81.

4. Prandoni P, Lensing AW, Buller HR, et al. Lancet. 1992;339:441-5.

5. Koopman MM, Prandoni P, Piovella F, et al. N Engl J Med. 1996;3345:682-7.