AbstractThe paradoxical correlation between thrombosis and the lupus anticoagulant (LAC) effect is an enigmatic feature of the antiphospholipid (aPL) syndrome. The Dutch authors previously reported that thrombosis-related anti–β2-glycoprotein I (β2GPI) antibodies recognize domain I and cause LAC. The American authors reported that aPLs disrupt an anticoagulant annexin A5 (AnxA5) crystal shield. We investigated whether antidomain I antibodies correlate with disruption of AnxA5-anticoagulant activity. We studied a well-characterized group of 33 patients including subgroups with β2GPI-dependent LAC that recognize domain I (n = 11), with β2GPI-independent LAC (n = 12), and lacking LAC (n = 10). The effects on AnxA5-anticoagulant activity were determined with an AnxA5 resistance assay that measures coagulation times with and without AnxA5. Patients with β2GPI-dependent LAC (group A, all with thrombosis) had significantly lower AnxA5-anticoagulant ratios than those with β2GPI-independent LAC (group B, thrombosis n = 4; 157.8% versus 235.6%, P < .001) and those without LAC (group C, thrombosis n = 2; 157.8% versus 232.5%, P < .001). There was no difference in the ratios between groups B and C (P = .92). Plasmas with β2GPI-dependent LAC that recognize domain I displayed significantly increased AnxA5 resistance, suggesting that specifically anti-β2GPI antibodies compete with AnxA5 for anionic phospholipids. These results are consistent with a model in which aPL antibodies may promote thrombosis by interfering with the anticoagulant activity of AnxA5.