Abstract
Research is conflicting whether kidney function should be incorporated in thromboembolism risk prediction. Our published data showed that the CHA2DS2-VASc score predicts thromboembolism and mortality in those without atrial fibrillation. We used the Rochester Epidemiology Project medical records system to retrospectively evaluate whether adding renal impairment (1 point) to the CHA2DS2-VASc score (-R) enhances the score's prediction of mortality, thromboembolism, and atrial fibrillation in patients without atrial fibrillation. We identified patients that had an implantable cardiac monitoring device placed from January 1, 2004 to December 31, 2013, which was defined as the start date. Follow-up was through March 7, 2016. An implantable device was required to discern the absence of atrial fibrillation. Renal impairment was defined as chronic kidney disease stage 3 or greater. The population (n = 1,606) had a mean age of 69.8 years and median follow-up of 4.8 years. Baseline renal impairment was predictive of mortality (hazard ratio [HR] 2.06, 95% confidence interval [CI] 1.64 to 2.60, p <0.001), thromboembolism (HR 1.34, 95% CI 0.96 to 1.87, p = 0.09), and atrial fibrillation (HR 1.31, 95% CI 0.98 to 1.74, p = 0.07). Lower glomerular filtration rate correlated significantly with mortality. Increasing CHA2DS2-VASc-R score correlated significantly with mortality, thromboembolism, and incident atrial fibrillation. The addition of renal impairment to the CHA2DS2-VASc score improved the C-statistics for thromboembolism and survival from 0.72 to 0.73 (p = 0.01) and 0.70 to 0.72 (p <0.001). Adding renal impairment to the CHA2DS2-VASc score improves the score's prediction of thromboembolism and mortality in a population without atrial fibrillation, although the incremental benefit appears mild. © 2018 Elsevier Inc.