Objective: To determine whether the diagnosis of lupus anticoagulant (LAC) in a large cohort of positive patients was confirmed at a reference laboratory.
Methods: Over a 1-year period, each participating center collected samples from LAC-positive patients. Plasma was filtered and kept deep-frozen until it was sent on dry ice to the reference laboratory by express courier. Centers returned detailed laboratory information and clinical data from each patient. The reference laboratory screened plasma samples by diluted Russell viper venom time (dRVVT) and kaolin clotting time (KCT). When these were prolonged, 1:1 mixing studies were carried out, and confirmatory tests were performed as appropriate. Positive samples were further tested by thrombin time (TT). The presence of heparin was checked by measuring antifactor Xa activity when TT was prolonged. Negative samples were tested by activated partial thromboplastin time using hexagonal phospholipids.
Results: Plasma samples from 302 patients from 29 anticoagulation clinics were analyzed. LAC was excluded in 71 samples (24%), because dRVVT and KCT screening test results were normal (34) or reversed to normal by mixing studies (35). The remaining two samples were considered negative because they contained heparin. LAC-negative patients showed different characteristics from those in whom diagnosis was confirmed. They were significantly older (49.7 vs. 45.0 years, P < 0.03), were more often first diagnosed (66% vs. 41%, P < 0.001), and were more frequently judged as mild in LAC potency (60% vs. 25%, P < 0.0001). Moreover, anticardiolipin and anti-beta(2)-glycoprotein I antibody values were more often normal in LAC-negative (82%) than in LAC-positive (42%) samples (P < 0.0001). LAC-positive samples identified by both dRVVT and KCT (146/231, 63%) showed a LAC potency that was significantly stronger than that in samples in which LAC diagnosis was made by a single test.
Conclusions: A false-positive LAC diagnosis is not uncommon across specialized centers. Patients' characteristics and a complete antiphospholipid antibody profile may help to identify these individuals.