Abstract
PURPOSE: We hypothesized that anthropometrically predicted visceral adipose tissue (apVAT) accounts for more variance in blood-based biomarkers of glucose homeostasis, inflammation, and lipid metabolism than body mass index (BMI), waist circumference (WC), and the combination of BMI and WC (BMI + WC). METHODS: This was a cross-sectional analysis of 10,624 males and females who participated in the Third National Health and Nutrition Examination Survey (NHANES III; 1988-1994). apVAT was predicted from a validated regression equation that included age, height, weight, waist, and thigh circumferences. Bootstrapped linear regression models were used to compare the proportion of variance (R (2)) in biomarkers explained by apVAT, BMI, WC, and BMI + WC. RESULTS: apVAT accounted for more variance in biomarkers of glucose homeostasis than BMI (∆R (2) = 8.4-11.8 %; P < 0.001), WC (∆R (2) = 5.5-8.4 %; P < 0.001), and BMI + WC (∆R (2) = 5.1-7.7 %; P < 0.001). apVAT accounted for more variance in biomarkers of inflammation than BMI (ΔR (2) = 3.8 %; P < 0.001), WC (ΔR (2) = 3.1 %; P < 0.001), and BMI + WC (ΔR (2) = 2.9 %; P < 0.001). apVAT accounted for more variance in biomarkers of lipid metabolism than BMI (ΔR (2) = 2.9-9.2 %; P < 0.001), WC (ΔR (2) = 2.9-5.2 %; P < 0.001), and BMI + WC (ΔR (2) = 2.4-4.1 %; P ≤ 0.01). CONCLUSIONS: apVAT, estimated with simple and widely used anthropometric measures, accounts for more variance in blood-based biomarkers than BMI, WC, and BMI + WC. Clinicians and researchers may consider utilizing apVAT to characterize cardio-metabolic health, particularly in settings with limited availability of imaging and laboratory data.