OBJECTIVE--To determine the efficacy of antiplatelet therapy as prophylaxis against deep venous thrombosis or pulmonary embolism in surgical and high risk medical patients. DESIGN--Overviews of all randomised trials of antiplatelet therapy that could have been available by March 1990 and in which deep venous thrombosis was assessed systematically. SETTING--53 trials (total 8400 patients) of an average of two weeks of antiplatelet therapy versus control in general or orthopaedic surgery; nine trials (600 patients) of antiplatelet therapy versus control in other types of immobility; 18 trials (1000 patients) of one antiplatelet regimen versus another. RESULTS--Overall, a few weeks of antiplatelet therapy produced a highly significant (2P < 0.00001) reduction in deep venous thrombosis. 25% of patients allocated antiplatelet therapy versus 34% of appropriately adjusted controls had deep venous thrombosis detected by systematic fibrinogen scanning or venography, representing prevention in about 90 patients per 1000 allocated antiplatelet therapy. There was an even greater proportional reduction in pulmonary embolism: such emboli were detected among 47 (1.0%) antiplatelet allocated patients versus an adjusted control total of 129 (2.7%), representing prevention among about 17 patients per 1000 treated (2P < 0.00001). In analyses confined to surgical trials, the proportional reductions were similar and separately significant for nonfatal pulmonary embolism (0.7% antiplatelet therapy v 1.8% control; 2P < 0.00001) and for deaths attributed to pulmonary embolism (0.2% v 0.9%; 2P = 0.0001). There was a slight but non-significant excess of deaths from other causes (1.0% v 0.7%), which made the difference in total mortality nonsignificant, though still favourable (1.2% v 1.5%). Information on adding antiplatelet therapy to heparin was limited but, at least for pulmonary embolism, suggested more protection from the combination than from heparin alone. The proportional reduction in the odds of suffering a deep venous thrombosis was roughly the same in patients having general surgery, traumatic orthopaedic surgery, and elective orthopaedic surgery (and in medical patients who were at increased risk of thromboembolism). For pulmonary embolism the numbers affected were smaller, but again the reductions were highly significant both in general surgery (16 (0.5%) v 58 (1.7%) pulmonary emboli; 2P < 0.0001) and in orthopaedic surgery (28 (2.7%) v 63 (6.1%) pulmonary emboli; 2P < 0.0002). CONCLUSION--It had previously been supposed that antiplatelet therapy did not influence venous thromboembolism, and many surgeons and physicians do not use it routinely for thromboprophylaxis, even for patients who are at substantial risk of deep venous thrombosis or pulmonary embolism. These results indicate that antiplatelet therapy--either alone or, for greater effect, in addition to other proved forms of thromboprophylaxis (such as subcutaneous heparin)--should be considered.