Clinical Characteristics, Predictors for Mortality and Comparison of the Birmingham Vasculitis Activity Score and the Five-Factor Score on Survival in ANCA-Associated Vasculitis in Hong Kong

Abstract

Objective: To describe the clinical profile and predictors of mortality of antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) patients in Hong Kong. To compare the accuracy of the latest Five-Factor Score (FFS-2009) and the Birmingham Vasculitis Activity Score (BVAS) in prediction of survival with this local cohort. Methods: A retrospective observational study on newly diagnosed AAV patients, from January 1, 2011 to March 31, 2022, managed in the Kowloon West Cluster (KWC) hospitals in Hong Kong. Demographic and baseline characteristics, clinical profile, and treatment profile were reviewed. Factors associated with mortality were analyzed with the Cox proportional hazards model. The performances of FFS and BVAS in mortality prediction were analyzed by receiver operating characteristic (ROC) curves. Results: A total of 83 AAV patients were included in the study. The median age was 70.5 years at diagnosis. Microscopic polyangiitis (MPA; 69.9%) was the most common AAV subtype. The median FFS and BVAS were 2 and 20, respectively. The overall mortality was 45.6% across the study period. Multivariate Cox regression identified age at diagnosis (HR 1.043, [Formula: see text]), stabilized peak serum creatinine (HR 1.002, [Formula: see text]), hemoglobin level (HR 0.754, [Formula: see text]), cardiac involvement (HR 3.862, [Formula: see text]), and use of maintenance therapy (HR 0.261, [Formula: see text]) as independent predictors of overall survival. Both FFS and BVAS were significant predictors of overall survival. The areas under the curve (AUC) of ROC curves suggested FFS was a good prediction tool for early mortality in 1 year, with an AUC value of 0.874. Conclusion: Despite the advances in treatment, AAV still carried significant morbidities with high mortality. Clinical predictors and existing scoring systems showed good predictive power on mortality.