Abstract P165: Associations of Coronary Artery Calcium Volume and Density With Incident Heart Failure in the Multi-Ethnic Study of Atherosclerosis

Abstract

Background: Coronary artery calcium (CAC) predicts incident heart failure (HF) independent of cardiovascular disease (CVD) risk factors. In MESA, Components of CAC, volume and density, have opposite associations with incident CVD, such that for a given volume of CAC, higher CAC density is inversely associated with events. The relationship between CAC volume and density with HF is unknown. Methods: We studied 6814 participants in a multi-ethnic, community-based cohort, free from clinical CVD at recruitment. CAC volume and density were measured by non-contrast cardiac CT at the baseline exam (2000-2002). Adjudicated HF events were assessed through 2014, and analysis limited to those with imaging confirmation and estimated ejection fraction (EF). Cox proportional hazard was used to estimate independent associations of baseline CAC volume and density with incident HF: HF with reduced (< 50%), and preserved EF (HFrEF & HFpEF respectively). Results: The mean age was 62 + 10 years, 47% were men, 38% identified as European-, 28% as African-, 22% as Hispanic-, and 12% as Chinese-ethnicity. Average time to 189 HF events (119 HFrEF & 70 HFpEF) was 6.6 years. In unadjusted models, higher CAC volume (HR 1.27 [1.02-1.59], p=0.03), but not CAC density (HR 0.87 [0.67-1.13], p=0.29) was significantly associated with incident HF, non-significant associations were observed with HFrEF, or HFpEF, and no significant associations were observed for all three outcomes after adjustments for demographics and CVD risk factors (Table). Also, in unadjusted analyses, stratified by sex (p-value for interaction = 0.13), higher CAC volume was associated with increased risk for HF (HR 1.37 [1.03-1.81], p=0.03) and HFpEF (HR 1.76 [0.99-3.16], p=0.06), in males only. No significant associations were observed after adjustments. Conclusion: In a multi-ethnic cohort, CAC volume and density were not independently associated with HF, the trend for volume was positive while density was inverse. Low frequency of incident HF in our cohort was an important limitation.