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


Review: Low-molecular-weight heparin and compression stockings are most effective for preventing venous thromboembolism in hip replacement

ACP J Club. 1994 Nov-Dec;121:61. doi:10.7326/ACPJC-1994-121-3-061

Source Citation

Imperiale TF, Speroff T. A meta-analysis of methods to prevent venous thromboembolism following total hip replacement. JAMA. 1994 Jun 8;271:1780-5.



To determine the efficacy and safety of the accepted methods of prophylaxis for venous thromboembolism after total hip replacement using meta-analysis.

Data sources

English-language studies were identified through MEDLINE (January 1966 to December 1993) using the keywords hip replacement or hip prosthesis, thromboembolism, randomized controlled trials, and clinical trials, aspirin, dextran, heparin, low-molecular-weight heparin, compression stockings, and warfarin. Additional articles were identified by scanning the bibliographies of relevant papers, review articles, and textbook chapters.

Study selection

Randomized controlled trials were selected if the patients were having elective total hip replacement; the intervention evaluated was aspirin, warfarin, dextran, subcutaneous heparin, subcutaneous low-molecular-weight heparin, or compression stockings; and the outcomes reported were all deep venous thrombosis, proximal venous thrombosis, pulmonary embolism, or clinically important bleeding.

Data extraction

One observer blinded to quantitative outcome data extracted methodologic and descriptive data from each study in order to assign a quality score. Outcome data were pooled across treatment arms.

Main results

56 trials met the inclusion criteria, with 91 treatment groups and 25 control groups. 4 treatment groups were excluded because of rarely used treatment combinations. The patients studied were a homogeneous group. When compared with the control, all active treatments except aspirin reduced the risk for developing venous thromboembolism with risk differences (the greater the risk difference, the more efficacious was the treatment) that ranged from 0.18 for dextran to 0.31 for low-molecular-weight heparin (Table). All treatments except aspirin reduced the risk for proximal venous thrombosis, with risk differences that ranged from 0.10 for heparin to 0.19 for warfarin (Table). Only low-molecular-weight heparin and compression stockings reduced the risk for pulmonary embolism, both with risk differences of 0.020 and 0.019 , respectively (Table). The crude risks (the percentage of patients who developed the adverse outcome) for clinically important bleeding were 0% for compression stockings, 0.3% for the control, and 1.8% for low-molecular-weight heparin.


Low-molecular-weight heparin and compression stockings are the most effective therapies for preventing venous thromboembolism following total hip replacement. Low-molecular-weight heparin carries a small risk for clinically important bleeding.

Source of funding: Not stated.

For article reprint: Dr. T.F. Imperiale, Department of Medicine, MetroHealth Medical Center, 2500 MetroHealth Drive, Room W603, Cleveland, OH 44109-1998, USA.

Table. Methods to prevent venous thromboembolism after total hip replacement (follow-up not reported)*

Outcomes Type of treatment Treatment Control RRR (95% CI) NNT (CI)
All deep venous thrombosis Aspirin 32% 47% 32% Not significant
Dextran 29% 47% 38% 6 (4 to 11)
Heparin 24% 47% 49% 5 (4 to 6)
LMWH 16% 47% 66% 4 (3 to 4)
Stockings 21% 47% 55% 4 (4 to 6)
Warfarin 24% 47% 49% 5 (3 to 8)
Proximal venous thrombosis Aspirin 15% 23% 35% Not significant
Dextran 9% 23% 61% 8 (5 to 14)
Heparin 13% 23% 43% 11 (7 to 43)
LMWH 7% 23% 70% 7 (5 to 11)
Stockings 13% 23% 43% 10 (6 to 32)
Warfarin 4% 23% 83% 6 (4 to 9)
Pulmonary embolism Aspirin 1.9% 2.4% 21% Not significant
Heparin 2.1% 2.4% 13% Not significant
LMWH 0.4% 2.4% 83% 49 (32 to 111)
Stockings 0.5% 2.4% 79% 51 (30 to 167)
Warfarin 1.6% 2.4% 33% Not significant
Dextran 8.8% 2.4% 267% 16

*LMWH = low-molecular-weight heparin. Other abbreviations defined in Glossary; RRR, RRI, and NNH calculated from data in article.


Venous thromboprophylaxis is recommended to reduce the incidence of venous thromboembolism and pulmonary embolism after total hip replacement. Debate over the most effective and safe method still exists.

Imperiale and Speroff have done a meta-analysis that suggests that low-molecular-weight heparin and warfarin are the most effective methods to prevent proximal venous thrombosis and that low-molecular-weight heparin and compression stockings are the most effective methods to prevent pulmonary embolism.

The results for low-molecular-weight heparin are consistent with the results of individual trials and other meta-analyses showing its efficacy. These trials meet high scientific criteria including a venographic end point for venous thrombosis. Venography was not routinely done in the studies on the other methods of prophylaxis, thus reducing the reliability of their conclusions about compression stockings. The data on pulmonary embolism are even less reliable because no routine surveillance or diagnostic testing was used.

In conclusion, the results of the analysis support the efficacy of low-molecular-weight heparin in prevention of venous thrombosis and pulmonary embolism after total hip arthroplasty. Other recommended forms of prophylaxis in these high-risk patients are warfarin and adjusted-dose heparin (1). Unresolved issues related to the use of primary prophylaxis include the clinical importance of calf venous thrombosis, the need to start prophylaxis before surgery, the total duration of prophylaxis, and the cost-effectiveness of these methods.

Moira Cruickshank, MD, MSc
University HospitalLondon, Ontario, Canada


1. Clagett GP, Anderson FA Jr, Levine MN, Salzman EW, Wheeler HB. Prevention of venous thromboembolism. Chest. 1992;102:391S-407S.