Standardization of lupus anticoagulant. Feasibility study of a calibration model to minimize between-method variability

Results for lupus anticoagulant (LA) are currently expressed as ratio of patient-to-normal clotting times (LA-ratio). Yet, numerical results do vary according to the method used for testing, thus making difficult the between-method comparison of results. We hypothesized that the standardization model currently used for the INR for patients on oral-anticoagulants (OAT) would be of value also for LA standardization. PATIENTS AND METHODS: To test this hypothesis we determined a sensitivity index valid for LA (called LASI) for six LA-detection methods against a common-standard using two sets of calibration-plasmas: (i)normal-plasmas spiked with IgG derived from patients strongly-positive for LA or (ii)plasmas from LA-positive patients. The LASI was then used to convert the LA-ratio into the standardized-LA-ratio (SLA-ratio) according to the equation: SLA-ratio = (LA-ratio)(LASI). RESULTS: We demonstrate that (i)the model is feasible because calibration plots of log-transformed clotting times obtained for the LA-detection methods-vs.-the common-standard gave acceptable LASI values; (ii)the model is effective because between-method variability expressed as coefficient of variation, which was 42.8% with results expressed as LA-ratio, decreased to 7.8% with results expressed as SLA-ratio; (iii)the LASI value calculated with the LA-positive plasmas is more effective in minimizing between-method variability than the LASI value calculated with IgG-spiked plasmas. CONCLUSIONS: A model of LA calibration similar to the INR for patients on OAT is feasible by using plasmas from LA-positive patients instead of patients on OAT. Potential application of the model are:(i)to compare the relative responsiveness of different LA-detection methods,(ii)to minimize differences between their results and (iii)to quantify LA potency.