Abstract
BACKGROUND: Rheumatic atrial fibrillation (AF) patients are at an elevated risk of stroke events, yet the associated risk factors remain unclear. OBJECTIVES: This study aimed to evaluate the effectiveness of the CHA(2)DS(2)-VASc score in predicting stroke events in rheumatic AF patients and explore potential enhancements for increased predicting accuracy in the Asian population, comparing it to nonvalvular AF. METHODS: A retrospective cohort study spanning March 2010 to December 2020 included 29,341 AF patients followed up for up to 10 years, with 1,942 identified as having rheumatic AF. The CHA(2)DS(2)-VASc score was computed for all patients. Using the area under the receiver operating characteristic curve (AUC), we evaluated the performance of the predictive model. The clinical endpoint was ischemic stroke. RESULTS: Rheumatic AF patients exhibited a higher stroke incidence than nonvalvular AF patients (log-rank P = 0.048). While the CHA(2)DS(2)-VASc score predicted stroke events in rheumatic AF, its AUC was lower than in nonvalvular AF (0.693 vs 0.746, P < 0.001). Integrating left atrial (LA) dimension information showed a trend toward an increased AUC for rheumatic AF to 0.707 (P = 0.068), but not for nonvalvular AF (AUC: 0.734 with LA, P = 0.744). The CHA(2)DS(2)-VASc-LA score showed a robust correlation with the annual incidence of ischemic stroke in rheumatic AF patients (P = 0.012 for trend). CONCLUSIONS: Rheumatic AF patients have a higher stroke risk than nonrheumatic AF and CHA(2)DS(2)-VASc-LA score might improve stroke risk assessment in this population.