Jiayue Feng1, Sen He1, Xiaoping Chen2. 1. Department of Cardiology, West China Hospital, Sichuan University, Chengdu, China. 2. Department of Cardiology, West China Hospital, Sichuan University, Chengdu, China. Electronic address: feng_jiayue@yeah.net.
Abstract
BACKGROUND: Body adiposity index (BAI) and body roundness index (BRI), initially developed to assess obesity, were evaluated here to detect insulin resistance in comparison with traditional anthropometric indices of body mass index (BMI), waist circumference (WC), weight-to-height ratio (WHtR), visceral adiposity index (VAI) and abdominal volume index (AVI). METHODS: In this cross-sectional study, 570 Chinese individuals without diabetes were evaluated. RESULTS: The Spearman rank test showed that insulin resistance correlated most strongly with WC and AVI in men and BMI in women, and most weakly with BAI in men and VAI in women. The prevalence of insulin resistance increased per quartile for all 7 anthropometric indices. Multivariate logistic regression identified BAI as the weakest predictor of insulin resistance in both genders (men, odds ratio [OR] 3.34, 95% confidence interval [CI] 1.09-10.18; women, OR 4.90, 95% CI 1.89-12.69), AVI as the strongest predictor in men (OR 19.73, 95% CI 2.51-155.04) and BMI as the strongest predictor in women (OR 15.55, 95% CI 4.71-51.28). The area under the receiver operating characteristic curve (AUC) showed that BAI exhibited the lowest AUCs for men (0.653, 95% CI 0.574-0.731) and women (0.701, 95% CI 0.627-0.774). BRI showed significantly higher AUCs for men (0.769, 95% CI 0.699-0.838) and women (0.763, 95% CI 0.699-0.827), and WHtR showed equal AUCs to BRI. CONCLUSIONS: Neither BAI nor BRI were superior to BMI, WC, WHtR, VAI or AVI for predicting insulin resistance. BAI showed the weakest predictive ability, while BRI showed reasonable potential to serve as an alternative anthropometric index to detect insulin resistance.
BACKGROUND: Body adiposity index (BAI) and body roundness index (BRI), initially developed to assess obesity, were evaluated here to detect insulin resistance in comparison with traditional anthropometric indices of body mass index (BMI), waist circumference (WC), weight-to-height ratio (WHtR), visceral adiposity index (VAI) and abdominal volume index (AVI). METHODS: In this cross-sectional study, 570 Chinese individuals without diabetes were evaluated. RESULTS: The Spearman rank test showed that insulin resistance correlated most strongly with WC and AVI in men and BMI in women, and most weakly with BAI in men and VAI in women. The prevalence of insulin resistance increased per quartile for all 7 anthropometric indices. Multivariate logistic regression identified BAI as the weakest predictor of insulin resistance in both genders (men, odds ratio [OR] 3.34, 95% confidence interval [CI] 1.09-10.18; women, OR 4.90, 95% CI 1.89-12.69), AVI as the strongest predictor in men (OR 19.73, 95% CI 2.51-155.04) and BMI as the strongest predictor in women (OR 15.55, 95% CI 4.71-51.28). The area under the receiver operating characteristic curve (AUC) showed that BAI exhibited the lowest AUCs for men (0.653, 95% CI 0.574-0.731) and women (0.701, 95% CI 0.627-0.774). BRI showed significantly higher AUCs for men (0.769, 95% CI 0.699-0.838) and women (0.763, 95% CI 0.699-0.827), and WHtR showed equal AUCs to BRI. CONCLUSIONS: Neither BAI nor BRI were superior to BMI, WC, WHtR, VAI or AVI for predicting insulin resistance. BAI showed the weakest predictive ability, while BRI showed reasonable potential to serve as an alternative anthropometric index to detect insulin resistance.