Antiphospholipid antibodies (aPL) react with negatively charged phospholipids, which may often be complexed with a protein cofactor such as beta2 glycoprotein (beta2GPI) and prothrombin. Cofactor requirements may be assessed by measuring antibodies to beta2GPI or by adding Tween 20 to some reagents in the assays for aPL (anticardiolipin and antiphosphatidyIserine). We have measured anticardiolipin antibodies (aCL), antiphosphatidylserine antibodies (aPS), and anti beta2 glycoprotein antibodies (abeta2GPI) in the serum of 10 normal subjects, 20 patients with systemic autoimmune diseases (SAD) diagnosed as having systemic lupus erythematosus (SLE) or antiphospholipid syndrome (APS), and 12 patients with HIV infection. Adding Tween 20 to aPS, the assay couldn't differentiate protein cofactor dependent from independent antibodies, but this can be done by measuring abeta2GPI (P= 0.0008). There was a significant correlation between aCL and a(beta)2GPI in the control group and in the patients with SAD, but not in the HIV-positive (HIV+) patients. After excluding the HIV+ patients, the best Spearman correlation was obtained between a(beta)2GPI and aCL (0.64, P< 0.0005). In 3 out of 7 patients with positive a(beta)2GPI and in 5 out of 6 patients with moderate or high positive aCL of the group of SAD, there was a history of venous thrombosis. The presence of moderate or high values of aCL either alone or together with a(beta)2GPI was significantly associated with a history of venous thrombosis (P < 0.05). Moderate or high aCL concentrations and their association with a(beta)2GPI seems to be useful for the assessment of the risk of venous thrombosis in unselected patients with SLE or APS.