THU0167 Evaluation of two assays for antiphospholipid antibodies in 712 SLE patients; clinical associations depend on isotypes and cut-off levels

Abstract

Background There are several problems with the current methods for detection of antiphospholipid antibodies (aPL) and there is an on-going discussion about the interpretation of different aPL-tests in everyday practice. (1) Objectives I. To evaluate agreement and performance of two tests for aPL, including a new automated assay. II. To study the importance of isotypes and titers with respect to thrombotic events in patients with Systemic Lupus Erythematosus (SLE). Methods We investigated 712 SLE-patients and 280 controls. Antibodies against cardiolipin and b2 glycoprotein-I were analysed by a routine ELISA method (Orgentec, Mainz, Germany) and a new, automated method (Elia Cardiolipin IgG/IgM, Elia –β2 –Glycoprotein I IgG/IgM performed on Phadia 250, Phadia AB, now Thermo Fisher Scientific, Germany). We used three cut-offs for positivity and calculated specificity, sensitivity and odds ratio (OR) for objectively verified previous thrombotic events. Results were compared with outcomes for Lupus anticoagulant (LAC) performed in 380 of the patients. Results Agreement between and performance of both aPL-methods were modest with no strong advantage in performance for either test. Using a cut-off corresponding to established criteria (2) sensitivity for the individual aPL-tests for any previous thrombosis ranged from 3.7 to 24.8% while specificity ranged from 84.7% to 97.7%. Regardless of assay, antibodies of IgG-isotype were associated with venous thrombosis and ischemic cerebrovascular disease while IgM-antibodies were associated with ischemic heart disease. Associations were highly affected by the positivity cut-off-level. ORs for LAC were generally higher than for the specific aPL-tests. For any previous thrombosis, OR (95% Confidence Interval) for LAC was 5.4 (3.1-9.4) while OR for the individual aPL-tests using a cut-off corresponding to established criteria ranged from non-significant to 1.9 (1.03-3.4) with the highest OR for routine ELISA cardiolipin IgG. Conclusions The most interesting finding in this study is that aPL of IgG and IgM isotypes were associated with different types of thrombotic events. Only moderate agreement between, but no clear advantage in relation to thrombotic manifestations, was observed when comparing a routine ELISA for aPL with a new automated method. The LAC test performed better than both aPL-assays. References Galli M. Clinical utility of laboratory tests used to identify antiphospholipid antibodies and to diagnose the antiphospholipid syndrome. Semin Thromb Hemost. 2008 Jun;34(4):329-34. Miyakis S, Lockshin MD, Atsumi T, Branch DW, Brey RL, Cervera R, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost. 2006 Feb;4(2):295-306. Disclosure of Interest None Declared