Abstract 4139407: Association of Aortic Valve Calcium with Albuminuria: The Multi-Ethnic Study of Atherosclerosis

Background: Higher aortic valve calcium (AVC) burden is associated with an increased risk of cardiovascular disease (CVD) and non-CVD. Despite shared risk factors between AVC and chronic kidney disease (CKD), the association of AVC with incident albuminuria is unknown. Methods: We examined 5,464 MESA participants who had AVC quantified by cardiac CT at Visit 1 (2000-02), an eGFR of ≥60 mL/min/1.73 m 2 , a normal spot urine albumin to creatinine ratio (ACR) (<30 mg/g at baseline), and at least 1 follow up ACR measurement. AVC was analyzed as a log transformed variable, due to its right skewed distribution, and categorized as 0, 1-99, and ≥100. Incident albuminuria was defined as ACR ≥30 mg/g. We performed multivariable Cox proportional hazard regression along with mixed-effect linear regression models to examine the association of AVC with incidence, and progression of albuminuria per 5 years follow up, respectively. Models adjusted for baseline eGFR, sociodemographic and CVD risk factors, with further adjustment for coronary artery calcium (CAC), lipoprotein (a) (Lp[a]), and the APOE-ε4 genotype. Results: At MESA Visit 1, participants had a mean age 61.6±9.9 years, 2894 (53%) were women, and 616 (11%) had AVC > 0. During a median follow up of 14.2 years, 921 (17%) developed albuminuria. There was a significantly increased rate of incident albuminuria with higher AVC values (p < 0.001) ( Figure ). In multivariable adjusted models, a higher risk of incident albuminuria was observed when AVC was examined as a continuous variable (per log-unit [AVC+1]) HR 1.05; p = 0.03 and for participants with AVC ≥100 HR 1.43; ( p = 0.02) compared to AVC=0, but not for participants with AVC 1-99 (HR 1.13; p = 0.29). A significant progression in log transformed ACR was observed for AVC as a continuous variable (β 0.03; p < 0.001) along with participants who had AVC 1-99 (β 0.13; p < 0.001), and AVC ≥100 (β 0.16; p = 0.001), compared to AVC=0. The associations between continuous AVC and incident albuminuria remained after further adjusting for CAC score ( p = 0.04), Lp(a) ( p = 0.03), and the APOE-ε4 genotype ( p = 0.04). The signal was consistent for ACR progression after further adjusting for CAC score ( p < 0.01), Lp (a) ( p < 0.01), and the APOE-ε4 genotype ( p < 0.01). Conclusions: In a multi-ethnic cohort of participants free of CVD and CKD at baseline, AVC was independently associated with a higher risk of incident albuminuria and progression of ACR.