Internal quality control and external quality assurance in testing for antiphospholipid antibodies: Part II--Lupus anticoagulant

In addition to the presence of appropriate clinical features, the diagnosis of the antiphospholipid antibody syndrome (APS) fundamentally requires the finding of positive antiphospholipid antibody (aPL) test result(s), with these comprising clot-based assays for the identification of lupus anticoagulant (LA) and immunologic ("solid-phase") assays such as anticardiolipin antibodies (aCL) and anti-β2-glycoprotein I antibodies (aβ2GPI). This article is the second of two that review the process for, and provide recommendations to improve, internal quality control (IQC) and external quality assurance (EQA; or proficiency testing) for aPL assays. These processes are critical for ensuring the quality of laboratory test results, and thence the appropriate clinical diagnosis and management of APS. This article covers LA testing. We provide some updated findings from the Royal College of Pathologists of Australasia Haematology Quality Assurance Program, and cover testing results for the past 3 years (2009 to 2011 inclusive). In brief: (1) essentially all laboratories currently perform LA testing using activated partial thromboplastin time (APTT) and dilute Russell viper venom time (dRVVT) methods, and about one-third also employ the kaolin clotting time (KCT); (2) KCT usage has dropped slightly, from around 50% of laboratories in 2009, to around 35% in 2011, presumably reflecting take up of the latest consensus recommendations; (3) other methodologies such as silica clotting time (SCT) and the platelet neutralization procedure (PNP) are only used by <5% of laboratories; (4) interlaboratory coefficients of variation (CVs) are in general moderate, and substantially better than those reported for solid-phase assays such as aCL and aβ2GPI, with median (range) values being 11.6% (9.2 to 25.5%) for APTT ratios, 16.7% (10.1 to 19.2%) for KCT ratios, and 11.7% (5.7 to 17.4%) for dRVVT ratios; (5) CVs increase slightly with increasing LA positivity; (6) most laboratories correctly interpreted test findings for LA, reporting normal samples as normal, and LA-positive samples as positive (albeit with varying gradings of positivity); and (7) however, some laboratories found interpretation to be challenging for some samples, namely a weak LA sample (which was reported as normal by around 50% of laboratories) and a very strong LA sample (which was reported as normal by around 10% of laboratories, primarily those that did not perform mixing studies).