The sensitivity and specificity of commercial reagents for the detection of lupus anticoagulant show marked differences in performance between photo-optical and mechanical coagulometers

This study was undertaken to appraise the application of those reagents mos t widely used in the UK for the detection and confirmation of lupus anticoa gulant (LA) on an Amelung KC4A and a Sysmex CA-6000(TM) coagulometer. Five sets of dilute Russell's viper venom time (DRVVT) reagents were assessed as well as the Textarin(R)-PL/Ecarin ratio. Each DRVVT method comprised both LA detection and confirmation reagents provided by the same manufacturer. S amples were obtained from 20 normal healthy subjects, 10 LA-positive patien ts, 10 patients receiving oral anticoagulant therapy (OAT) who had previous ly been documented as LA-positive, a further 10 LA-negative patients receiv ing: OAT and 10 LA-negative patients receiving unfractionated heparin thera py. The sensitivity and specificity of the reagents exhibited considerable vari ation not duly between reagents, but also when the same reagent was used on the two analysers. Sensitivity ranged from 62 to 97% (all reagents both an alysers), specificity went as low as 23% (Gradipore reagent on the CA-6000( TM)) and as high as 100% (American Diagnostica Inc on both KC4A and CA-6000 (TM)). On the KC4A instrument, Unicorn Diagnostics' lupus anticoagulant kit offered the best compromise of sensitivity and specificity (sensitivity 83 % and specificity 81%). On the CA-6000(TM) the reagents supplied by America n Diagnostica Inc exhibited optimal performance (sensitivity 90% and specif icity 100%). The results indicate a need to optimise test reagents for spec ific analyser types, a procedure which can only be undertaken with preparat ions such as the proposed NIBSC reference plasmas for the detection of lupu s anticoagulant.