Usefulness of the CHA2DS2-VASc Score to Predict Mortality in Defibrillator Recipients

Abstract

The CHA2DS2-VASC score is a well-validated stratification tool that predicts the risk of thromboembolism and stroke in patients with nonvalvular atrial fibrillation. Several studies have examined its application as a predictor of mortality in clinical applications other than atrial fibrillation. However, there are current no studies examining its use as an outcome prediction tool in a population of patients with implantable cardiac defibrillators (ICDs). In this study, we examined data from 2,258 patients who underwent ICD device implantation at the hospitals of the University of Pittsburgh Medical Center from February 2002 to April 2014 (median follow-up 5.1years) and examined the impact of their CHA2DS2-VASC score at the time of device implantation on all-cause mortality. Survival curves based on CHA2DS2-VASC scores were generated using the Kaplan-Meier method and were adjusted for unbalanced covariates using the Cox proportional hazards model. The mean CHA2DS2-VASC score was 3.15 ± 1.52 (range 1 to 8, mode 3). The CHA2DS2-VASC score predicted all-cause mortality in a significant and dose-dependent fashion. Analyzing the population by quartiles revealed increasing all-cause mortality from Q1 to Q4 (p <0.001). Using a Cox multivariate model adjusting for ejection fraction, BMI, serum creatinine, hemoglobin level, and QRS width, the CHA2DS2-VASC score remained a strong predictor of all-cause mortality (hazard ratio 1.26 per 1-point increase, 95% confidence interval 1.20 to 1.32). In conclusion, the CHA2DS2-VASC score is a simple tool that highly predicts all-cause mortality in patients with ICD.