GAPSS: the Global Anti-Phospholipid Syndrome Score

• Published in: Rheumatology (Oxford, England) England), 2013-08, Vol.52 (8), p.1397-1403
• Main Authors: Sciascia, Savino, Sanna, Giovanni, Murru, Veronica, Roccatello, Dario, Khamashta, Munther A, Bertolaccini, Maria Laura
• Format: Article
• Language: English
• Subjects: Abortion, Spontaneous - diagnosis; Abortion, Spontaneous - immunology; Adult; Analysis of Variance; Antibodies, Antiphospholipid - immunology; Antiphospholipid Syndrome - diagnosis; Cardiovascular Diseases - diagnosis; Cardiovascular Diseases - immunology; Cross-Sectional Studies; Female; Humans; Logistic Models; Lupus Erythematosus, Systemic - diagnosis; Middle Aged; Multivariate Analysis; Pregnancy; Prothrombin - metabolism; Risk Assessment; Sensitivity and Specificity; Severity of Illness Index; Thrombosis - diagnosis; Thrombosis - immunology; United Kingdom
• ISSN: 1462-0324; 1462-0332
• DOI: 10.1093/rheumatology/kes388

To develop and validate a risk score [global APS score (GAPSS)] derived from the combination of independent risk for thrombosis and pregnancy loss (PL), taking into account the aPL profile, conventional cardiovascular risk factors and the autoimmune antibody profile. This cross-sectional study included 211 consecutive SLE patients. Data on clinical manifestations, conventional cardiovascular risk factors, aPL profile, ANAs, ENA and anti-dsDNA were collected. Long-term low-dose aspirin, oral anticoagulant and HCQ treatment were also included in the analysis. Patients were randomly divided into two sets by a computer-generated randomized list. We developed GAPSS in the first set of patients (n = 106), assigning the risk factors identified by multivariate analysis weighted points proportional to the β-regression coefficient values. GAPSS was validated in the second set of patients (n = 105). The relationship between GAPPS and thrombosis and/or PL was analysed. In the first set, higher values of GAPSS were seen in patients who experienced thrombosis and/or PL compared with those without clinical events [GAPSS 9.3 (4.8) (range 1-19) and 5.3 (4) (range 0-16), P < 0.001]. Also taken separately, patients who experienced thrombosis or PL showed higher GAPSS compared with those without clinical events [GAPSS 9.6 (4.8) (range 1-19) vs 4.9 (5) (range 0-14), P = 0.027 for thrombosis; 7.3 (5) vs 3.9 (5.1) (range 0-16), P = 0.024 for PL, respectively]. In the second set, the results were similar, with statistically higher values of GAPSS in patients with a clinical history of thrombosis and/or PL compared with those without events [GAPSS 9.5 (5.6) (range 0-20) and 3.9 (4.1) (range 0-17), P < 0.001). Higher values were also seen when subclassifying the patients according to the clinical manifestation, thrombosis or PL [GAPSS 9.5 (5.6) (range 0-20) vs 4.8 (5.4) (range 0-17), P = 0.036 for thrombosis; 7.9 (3.3) vs 3.8 (5.4) (range 0-16), P = 0.037 for PL, respectively). These data propose a substantial improvement in risk prediction of thrombosis or PL in SLE based on assessment of the GAPSS, a quantitative scoring system.