Exploratory Risk Assessment of Cardiovascular Diseases Using Antiphospholipid Antibody-Related Scores in Patients with ANCA-Associated Vasculitis

Background: Antiphospholipid antibodies have been found in patients with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV). This study aimed to evaluate the accuracy of the global antiphospholipid syndrome score (GAPSS) derivatives in predicting cardiovascular diseases in Korean patients with AAV, and to derive a simple candidate index for cardiovascular diseases. Methods: This was a retrospective cohort study including 170 patients with AAV. The adjusted GAPSS (aGAPSS), aGAPSS specific for cardiovascular diseases (aGAPSScvd), and aGAPSS for Korean patients with APS (aGAPSS [KAPS]) were calculated in each patient. The cardiovascular diseases were defined as cerebrovascular accident (CVA) or acute coronary syndrome (ACS). The accuracy of each index was assessed by estimating the area under the curve (AUC). An exploratory multivariable Cox proportional hazards model with a backward elimination method was performed to derive a candidate index. Covariates consisting the aGAPSScvd were included in the model. Results: The AUCs of the aGAPSS, aGAPSScvd, and aGAPSS (KAPS) for CVA were 0.507 (p = 0.926), 0.531 (p = 0.676), and 0.493 (p = 0.923), respectively. The AUCs of the aGAPSS, aGAPSScvd, and aGAPSS (KAPS) for ACS were 0.613 (p = 0.441), 0.752 (p = 0.086), and 0.796 (p = 0.043), respectively. The exploratory multivariable model revealed a candidate index, aGAPSScvd for ACS in Korean patients with AAV (aGAPSScvd [KAAV]) = 2 × lupus anticoagulant (LA) + 2 × obesity. The AUC of the aGAPSScvd (KAAV) in predicting ACS was 0.843 (p = 0.019). Conclusions: The two-item candidate index aGAPSScvd (KAAV), consisting of LA and obesity, showed discriminatory ability for ACS, which warrants further validation in larger studies.