Background/Purpose: The clinical scoring systems to quantify the probability for the diagnosis of antiphospholipid syndrome (APS) or future thrombosis, have been proposed as to reflect the diverseness of antiphospholipid antibodies (aPL) profiles into the clinical judgments: They are the Antiphospholipid Score (APLS) and the Global Antiphoshpholipid Syndrome Score (GAPSS). However, the clinical applications of these scores have been obstructed mainly of its burdensome procedure of performing more than 10 aPL tests routinely. The aim of this study is to confirm the function of modified APLS (mAPLS) with reduced number of needed aPL tests.

Methods: This study comprised 261 consecutive patients with autoimmune diseases including APS who visited Hokkaido University Hospital Rheumatology Clinic between 2002 and 2003 and has been followed more than five years afterwards. Five lupus anticoagulant assays (three mixing studies and two confirmatory tests) and six ELISAs (IgG/M of anticardiolipin antibodies, anti-b2-glycoprotein I antibodies and phosphatidylserine dependent antiprothrombin antibodies) were performed in all subjects. All of the aPL were calculated of their hazard ratio for developing thrombosis on definition of the APLS. Using those original data, mAPLS was constituted of the aPL tests that had high odds ratio over 5.0 for thrombotic events. The functions of mAPLS were compared with those of APLS and GAPSS. To evaluate the diagnostic powers, area under the curve (AUC) of receiver operating characteristic (ROC) curves were calculated. Cox proportional hazard regression analyses were performed separately to evaluate the powers of thrombosis prediction. To evaluate and compare the predictive powers of the two scores, Somer’s d coefficient was calculated.

Results: The diagnostic values of the three scores were similar(AUC of the ROC curve: APLS vs mAPLS vs GAPSS : 0.853 vs 0.810 vs 0.831). The Cox multivariate proportional hazard regression analyses revealed that three scores have functions of predicting future thrombotic events. However, Somer’s d coefficient revealed that the APLS and mAPLS have similar power of predicting thrombosis while GAPSS showed lower predictive value(0.50 vs 0.48 vs 0.40, p<0.01). Moreover, when these scores were evaluated in patients with SLE, exactness of accuracy for diagnosis fell rapidly in every scores (APLS(0.91 for non-SLE, 0.80 for SLE), mAPLS( 0.90 for non-SLE, 0.85 for SLE), GAPSS(0.85 for non-SLE, 0.80 for SLE)) and predictive value of future thrombosis was only confirmed by mAPLS; these phenomenon possibly reflecting the pseudo-positive aPL occasionally found in SLE patients. 

Conclusion: Modified APLS with its high performances on APS diagnosis and thrombosis prediction, along with its simplified calculation, may be an available option in general practice.

« Back to ACR/ARHP Annual Meeting 2016

ACR Meeting Abstracts - https://acrabstracts.org/abstract/a-practical-application-of-antiphospholipid-antibodies-profiles-in-the-diagnosis-and-managements-of-antiphospholipid-syndrome-the-modified-antiphospholipid-score/