Abstract P665: Thoracic Aorta Calcification and Left Ventricular Structure and Function Relevant to Heart Failure: The Multi-Ethnic Study of Atherosclerosis (MESA)

Background: Aortic stiffness promotes left ventricular (LV) remodeling and dysfunction. Thoracic aorta calcification (TAC) contributes to aortic stiffness, but it remains unclear how TAC presence, volume, and density impact LV structure and function. We tested the hypothesis that TAC presence and higher volume is associated with increased LV mass index, decreased LV end-diastolic volume (LVEDV) index, and impaired mid-wall circumferential strain (Ecc). Methods: We evaluated 1,693 participants from the MESA who underwent chest CT and cardiac MRI from 2010 to 2012. We used linear regression models to determine cross-sectional associations between the presence, volume, and density of TAC with LV mass index, LVEDV index, and Ecc. Each pair of predictor and outcome variables underwent three model adjustments: (1) age, gender, and race/ethnicity; (2) Model 1 plus education level, height, weight, physical activity, sedentary behavior, smoking status, eGFR, SBP, DBP, anti-HTN medication use, total cholesterol/HDL ratio, statin use, and diabetes; (3) Model 2 plus coronary artery calcium (CAC) presence and volume. TAC volume and density were evaluated separately, then jointly in the same models. Results: The mean age was 69 years (SD 9) and 53% were female, while 39% were White, 26% Black, 21% Hispanic/Latino, 15% Chinese. TAC was present in 1,518 (90%) participants. TAC volume and density were highly correlated (r=0.75) with respective median 213 (25%ile, 75%ile: 36, 800) mm 3 and mean 350 (SD 110) Hounsfield Units. LV mass index, LVEDV index, and Ecc had respective means of 65 g/m 2 (SD 14), 120 mL/m 2 (SD 31), and -18.2% (SD 2.8). After full adjustment (Model 3 without density), a 10% increase in TAC volume was associated with a 1.8% multiplicative higher Ecc (i.e., less negative Ecc) (95% CI 0.2-3.4%, p=0.04) reflecting reduced LV systolic and diastolic function. Higher TAC volume was marginally associated with higher LV mass index: a 10% increase in TAC volume was associated with a 5.6% increase (-0.6-11.7%, p=0.08) in Model 2 without density, 13.9% (2.1-25.7%, p=0.03) in Model 2 with density, 3.9% (-2.5-10.4%, p=0.23) in Model 3 without density, and 11.3% (-1.0-23.7%, p=0.08) in Model 3 with density. After adjustments in Models 2 or 3, there were no other significant associations between TAC presence, volume, or density with the outcomes. Conclusions: Higher TAC volume is associated with impaired Ecc independent of CAC and marginally associated with higher LV mass index. These results suggest that the extent of TAC may influence LV remodeling and thus reduced function.