Abstract 11328: CHADS2 Score Predicted New-Onset Dementia in Patients With Atrial Fibrillation: A Nationwide Cohort Study

Introduction: Atrial fibrillation (AF) is associated with increased risk of dementia. CHADS2 scores were reported to be important risk factors for the development of AF. The goal of this study sought to evaluate whether the CHADS2 scores could predict the incidence of new-onset dementia in patients with AF.Methods: A total of 5224 patients with AF who had no history of dementia and were above 20 years old were enrolled from National Health Insurance Research Database of Taiwan from January 1, 1997 to January 1, 1998. The CHADS2 scores were calculated for every patient. Finally, 500 (score 0), 2761 (score 1 and 2), and 1963 (score > 2) patients were studied and followed for the development of dementia.Results: During a mean follow-up of 10.9±0.7 years, there were 163 (3.1%) patients occurring new-onset dementia. The incidence of new-onset dementia was 2.86 per 1000 patient-year. The incidence increased from 1.09 per 1000 patient-years for patients with a CHADS2 score of 0 to 4.04 per 1000 patient-years for those with a CHADS2 score > 2. The Kaplan-Meier curve also demonstrated that CHADS2 scores were associated with the incidence of new-onset dementia in patients with AF (log rank P < 0.001, Figure 1). After an adjustment for the gender and comorbidities, cox-regression analysis showed that the hazard ratio (95% confidence interval) of CHADS2 scores were 0.268 (score 0 vs > 2; P = 0.002) and 0.585 (score 1, 2 vs > 2; P = 0.001).Conclusion: In conclusion, the CHADS2 scores predicted new-onset dementia and were useful in risk stratification of new-onset dementia in patients with AF.