Comparison of test methods for the lupus anticoagulant: international survey on lupus anticoagulants-I (ISLA-1).

Six lyophilized plasma samples were sent to 20 "expert" laboratories for assessment of lupus anticoagulant (LA). Four samples contained pooled LA of graded potency mixed with aged normal plasma. One contained LA plus cephalin phospholipid and one contained a nonspecific venom anticoagulant. Sixteen methods were used overall with some participants using up to 8 methods. Results were scored in regard to the known potencies of LA in the samples and other known induced defects. Activated partial thromboplastin time (APTT) tests used by most participants for preliminary screening were relatively sensitive, but non-specific. Platelet or phospholipid neutralization procedures (PNP) appeared to be sensitive and specific but showed a non-linear response to increased LA content. Kaolin clotting time (KCT) tests showed the most sensitive response to increased LA content but the weaker LA were not scored as abnormal by most laboratories as the samples may have contained platelet fragments. Other commonly used tests such as the tissue thromboplastin inhibition (TTI) test and the dilute Russell's viper venom test (DRVVT) were carried out somewhat inconsistently. The variability in performance of tests in different laboratories indicates that standardization of methodology is urgently required. Generally it seemed that most clotting tests were "bypassed" by the addition of phospholipid to a known LA-positive sample in apparently direct proportion to their sensitivity. Sample preparation, especially prevention of contamination with activated platelets is a vital preliminary part in the assay of LA.