To determine whether prophylaxis for venous thromboembolism (VTE) is effective when it is started post-operatively, and to determine the relative efficacy of pre- and post-operative prophylaxis.

MEDLINE was searched from 1990 for English language publications, using the MeSH 'embolism', 'thrombosis' and 'prophylaxis'. Bibliographies of consensus papers, reviews, book chapters on related topics, and reference files were also searched. Bibliographies of retrieved manuscripts were crosschecked for additional material. Abstracts of recent studies were also included.

Study designs of evaluations included in the review

Studies with randomised allocation of patients to prophylactic regimens and double-blind assessment of outcome. Both placebo and comparison trials were included.

Specific interventions included in the review

Comparisons were made between post-operative prophylaxis and (1) an untreated control group, (2) an alternative method of prophylaxis or (3) the same prophylactic method initiated pre-operatively. The type of VTE prophylaxis were non-pharmacologic or pharmacological, and the type of surgery was nonorthopaedic or orthopaedic.

Participants included in the review

Surgical patients without a substantial risk of having developed deep vein thrombosis (DVT) before operation, including patients undergoing hip or knee replacement and neurosurgery, those undergoing operation for hip fracture, or those who were hospitalised for some time before surgery were excluded.

Outcomes assessed in the review

The primary outcome was post-operative VTE. Rates of bleeding were also examined where possible.

How were decisions on the relevance of primary studies made?

The authors do not state how the papers were selected for the review, or how many of the authors performed the selection. The prophylactic regimen had to be described in sufficient detail such that the trial could be replicated, and with a clear statement of when prophylaxis was initiated relative to surgery.

VTE had to be diagnosed by objective tests that were appropriate for the surgical population studied. Randomised trials, blind assessment of outcome, and appropriate tests were used.

The authors do not state how the data were extracted for the review, or how many of the authors performed the data extraction.

How were the studies combined?

The rates of VTE for individual studies were determined from the binomial distribution, and those for relative risk (RR) were determined by the method of Thomas and Gart (see Other Publications of Related Interest).

How were differences between studies investigated?

The results for non-pharmacologic and pharmacologic prophylaxis were presented separately.

Thirteen studies were included, of which 12 were full accounts and 1 was an abstract. Eight studies compared post-operative prophylaxis with a control group, and the other 5 studies were randomised comparisons of different prophylaxis methods. Two of the studies included a comparison of 2 dosing schedules for the same medication. Five studies evaluated post-operative prophylaxis by mechanical devices, and 9 evaluated pharmacological methods; 1 study evaluated both mechanical devices and pharmacological methods.

Studies comparing the efficacy of post-operative application of mechanical devices with findings in an untreated control group (intermittent pneumatic compression of the calf, or of the calf and thigh combined) led to clinically important reductions of VTE in neurosurgical and orthopaedic patients with RR of 0.18 (95% confidence interval, CI: 0.004, 1.34) and 0.49 (95% CI: 0.35, 0.69).

Of the controlled studies that used pharmacological prophylaxis, 2 evaluating the efficacy of post-operative low-molecular-weight heparin (LMWH) therapy found a clinically important and statistically-significant reduction in rates of VTE after hip and knee replacement, with RR of 0.16 (95% CI: 0.04, 0.48) and 0.30 (95% CI: 0.14, 0.57).

Bleeding rates in the larger comparison studies showed a statistically-significant increase in patients who underwent hip replacement and who received 7500 U of standard heparin twice daily, compared with 30 mg enoxaparin twice daily (p=0.04). Major bleeding (p=0.04) and wound haematomas (p=0.01), but not minor bleeding, were more common in patients who underwent hip and knee replacement and received logiparin, 75 U/kg anti-Xa once a day, compared with warfarin.

The results from this review suggest that methods of VTE prophylaxis, which are effective when started pre-operatively, also achieve substantial risk reductions for VTE when they are started after surgery. Post-operative intermittent pneumatic compression of the calf, or of the calf and thigh in combination, and LMWH lead to statistically-significant and clinically-important reductions of post-operative VTE with RR of 0.16 to 0.49, compared to controls.

Low rates of VTE in randomised trials without an untreated control group that included post-operative prophylactic regimens support this finding. However, the relative efficacy of pre- and post-operative VTE prophylaxis could not be determined definitively, as no direct comparisons of the same regimens have been performed.

Although the search included bibliographies and cross-references, since the search strategy was restricted to MEDLINE other relevant trials may have easily been missed. Another criticism related to the search methodology is that only recent trials, i.e. 1990 onwards, were searched. A comprehensive coverage of the literature should have included a search for relevant trials before 1990. It is not stated explicitly the participants' characteristics, nor the criteria used to assess the validity or quality of the primary studies. No details are given on the way in which the data was extracted from the studies, and whether this included independent extraction by 1, 2 or 3 authors. Although the authors do provide a good qualitative summary of the findings, they do not present a test of heterogeneity.

Thomas DG, Gart JJ. A table of exact confidence limits for differences and ratios of two proportions and their odds ratio. J Am Stat Assoc 1977;72:73-6.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.