A large proportion of hospitalized patients who are at high risk for venous thrombo-embolism (VTE) do not receive prophylaxis. Reluctance to use VTE prophylaxis in surgical patients may be due to fear of perioperative bleeding when anticoagulants are given preoperatively. We preformed a literature review to determine (1) whether prophylaxis for VTE is effective when it is started postoperatively and (2) the relative efficacy of preoperatively and postoperatively initiated prophylaxis. Studies were included in the review (1) if they were randomized trials with "blind" assessment of appropriate VTE outcomes, and (2) if prophylaxis was started postoperatively. Randomized, controlled trials establish that pharmacologic and nonpharmacologic methods of prophylaxis that are effective when started preoperatively are also effective when they are started postoperatively, with relative risks for VTE of 0.16 to 0.49. Low rates of VTE in noncontrolled randomized trials that included postoperatively intitiated prophylactic regimens support this finding. The relative efficacy of preoperatively and postoperatively initiated VTE prophylaxis could not be determined definitively, as direct comparisons of the same regimens have not been performed. Indirect comparisons suggest that any loss of efficacy resulting from deferring VTE prophylaxis until after surgery is unlikely to be marked. Randomized trials are required to resolve this question. This comparison may be of greatest clinical importance when twice-daily, low-molecular-weight heparin is used to prevent VTE after major orthopedic surgery.
(Arch Intern Med. 1995;155:366-372)