NOVEL ANTHROPOMETRIC INDICES WERE BETTER ASSOCIATED WITH HYPERTENSIVE TARGET ORGAN DAMAGES IN THE COMMUNITY-DWELLING ELDERLY CHINESE: THE NORTHERN SHANGHAI STUDY

Abstract

Objective: Body mass index (BMI), the most commonly used body composition assessing indice, has recently been challenged by two novel anthropometric indices, a body shape index (ABSI) and body roundness index (BRI), in arterial stiffness predicting capacity[1,2]. The purpose of this study was to detect and compare the different predictive capacities of the four anthropometric indices (ABSI, BRI, waist-hip ratio (WHR) and BMI) in predicting hypertensive target organ damages (HTODs). Design and method: Overall, 3077 community-dwelling elderly Chinese (age > 65 years, 1329 men and 1748 women) in northern Shanghai were enrolled prospectively from June 2014 to August 2019. ABSI and BRI were calculated with validated formula. HTODs were defined as three levels: (1) left ventricular hypertrophy (LVH), (2) arterial stiffness (AS) and lower limb atherosclerosis (LLA), (3) micro-albuminuria (MAU). Multivariate linear and logistic regression were performed to detect the relations between anthropometric indices and HTODs. Results: Spearman correlation analysis revealed that all anthropometric indices were correlated with HTODs indices (all P < 0.001). In multivariate logistic regression models, compared to the lowest quartiles, participants with the highest quartiles of ABSI, BRI and WHR all exhibited a significantly higher risk of LVH, AS and MAU independent of confounders and BMI (all P for trend < 0.01). Results of receiver operator characteristic (ROC) analysis performed within the female subgroup demonstrated that BRI and WHR were the better anthropometric indices in predicting HTODs compared to BMI. However, similar trend disappeared in the male subgroup. Further comparison analysis about the predictive priorities in the female subgroup were tested with stepwise logistic regression, after all anthropometric indices were put into models, only BRI stayed and was significantly associated higher prevalence of LVH (OR = 1.42, 95%CI 1.30∼1.55) and LLA (OR = 1.48, 95%CI 1.23∼1.77) together with WHR stayed to predict higher risks of AS (OR = 1.30, 95%CI 1.18∼1.43) and MAU (OR = 1.48, 95%CI 1.28∼1.72, all P < 0.01). Conclusions: Novel anthropometric indices, especially BRI and WHR, compared to the conventional indice BMI, have the superior capacities in predicting HTODs in the elderly female Chinese population.