Abstract 13124: Better Prognostic Value of Single Time Coronary Artery Calcium Score Over the 30-y Averaged Pooled Cohort Equation in Older Adults: The Atherosclerosis Risk in Communities (ARIC) Study

Abstract

Background: Cardiovascular disease (CVD) risk prediction in the 75-and-older population is challenging in part because traditional risk factors at older age may not reflect the risk factor profile over the life course. In this regard, coronary artery calcium (CAC) can be considered an integrated biomarker of exposure to CVD risk factors. However, the predictive value of a single time point CAC at older age vs. the cumulative exposure of traditional risk factors over three decades is unknown. Methods: In 1,875 ARIC participants (age 75-94 years) without a history of coronary heart disease (CHD), stroke, and heart failure (HF) at Visit 7 (2018-2019), we assessed the association of Visit 7 CAC vs. a weighted average of Pooled Cohort Equation (PCE) calculated 10-year atherosclerotic CVD (ASCVD) risk from Visit 1 (1987-89) to Visit 7 with the risk of ASCVD (CHD or stroke), HF, and all-cause mortality using adjusted Cox models. Risk prediction improvement was assessed using Harrel’s C-statistics. Results: The median CAC was 221 (IQI 35-743), and the median averaged PCE score was 15.0% (IQI 10.4%-21.2%). Over a median follow-up of 2.1 (IQI 1.7-2.5) years, there were 48 ASCVD events, 36 HF events, and 85 deaths. A higher CAC score was associated with a higher risk of ASCVD and HF, even after adjusting for averaged PCE (adjusted hazard ratio 3.33 [95% CI 1.53, 7.26] and 5.57 [1.95, 15.89] for CAC 1000+ vs. 0-99, respectively), but not all-cause mortality ( Table ). In contrast, averaged PCE was not associated with ASCVD or HF independent of CAC. However, averaged PCE showed robust associations with all-cause mortality. Notably, the addition of CAC to averaged PCE significantly improved risk prediction of HF, whereas adding averaged PCE to CAC did not significantly improve the risk prediction for any adverse outcome. Conclusions: The present study shows that a single time point CAC outperforms 30-y cumulative exposure of PCE for ASCVD and HF prediction in the 75-and-older population.