Recent reviews have suggested a higher frequency of the lupus anticoagulant or related antiphospholipid antibodies in patients with systemic lupus erythematosus (21% to 65%) than was found in earlier studies (6% to 18%). In our study of 60 consecutive patients, we found the frequency of the lupus anticoagulant by Russell viper venom time was 6.7% (95% confidence interval, 16.2 to 1.8) and by anticardiolipin antibody assay was 25% (95% Cl, 37.0 to 15.7), compared with 0% (p = not significant) and 2.5% (p = 0.002), respectively, in the normal control population. The Russell viper venom time (p = 0.0001 by t-test) and anticardiolipin antibody levels (p = 0.01) were significantly associated with presumed thrombotic events (stroke, deep venous thrombosis, and digital gangrene). No association with miscarriage or pulmonary hypertension was detected. The Russell viper venom time was more specific than the anticardiolipin antibody level in the prediction of past presumed thrombotic events, miscarriage, or pulmonary hypertension (100% compared with 84%, p = 0.01).