Abstract W MP69: CHADS2 Score Predicts Stroke Related In-hospital Mortality In Patients With Atrial Fibrillation: An Analysis Of The California State Inpatient Database
Background: The CHADS2 score predicts stroke risk in patients with atrial fibrillation. Although strokes caused by atrial fibrillation carry the highest mortality when compared to other etiologies, it is unclear whether the CHADS2 score has an impact on stroke related mortality in patients with atrial fibrillation. We hypothesize that higher CHADS2 scores are associated with higher stroke related in-hospital mortality.
Methods: Data were obtained from the California State Inpatient Database between 2008 and 2011, using appropriate ICD-9 codes. Frequencies and descriptive analysis adjusting for influence of comorbidities and confounders were utilized. Age and ICD-9 codes for hypertension, diabetes, congestive heart failure, and prior stroke were used to calculate the CHADS2 score of patients with atrial fibrillation. A multivariate Cox regression model adjusted for age, gender and race was used to further explore the relationship. The primary outcome was in-hospital stroke mortality. We hypothesized that higher CHADS2 scores increase the risk of stroke related mortality.
Results: Between 2008-2011, 18,089 patients with atrial fibrillation had stroke as one of the admitting diagnoses; 70% were Caucasians, 5% African Americans, 15% Hispanic, and 10% Asian; 57% were females and 66% of patients were ≥ 75 years. The in-hospital mortality rates of the CHADS2 score were as follows: 0 (18.7%), 1 (18.6%), 2 (21.3%), 3 (26.3%), 4 (23.8%), 5 (22.8%), and 6 (22.2%). After adjusting for baseline demographics, the odds of in-hospital mortality was significantly higher with a CHADS2 score ≥ 2 vs. < 2 (OR 1.15 95% CI 1.08-1.23). Among the individual CHADS2 score items, predictors of increased in-hospital mortality were congestive heart failure (OR 1.61 95% 1.53-1.70), age ≥ 75 years (OR 1.27 95% 1.19-1.35), and diabetes (OR 1.24 95% CI 1.14-1.35).
Conclusion: CHADS2 ≥ 2 not only increases the risk of stroke but is a predictor of stroke related mortality. Factors driving this association appear to be age, congestive heart failure, and diabetes. This may be useful when deciding on anticoagulation use for stroke prevention especially in patients with elevated bleeding risk.
Author Disclosures: S. Yaghi: None. M. Pilot: None. D. Sherzai: None. A. Sherzai: None.
- © 2015 by American Heart Association, Inc.