Abstract
OBJECTIVES: We aim to compare the efficacies of the bioelectrical indices (percentage of body fat, PBF; visceral fat area, VFA) with the conventional anthropometric measures (body mass index, BMI; waist-hip ratio, WHR) for predicting type 2 diabetes (T2D) risk by sex and to determine the sex-specific optimal adiposity indices to predict the T2D risk. DESIGN: Cross-sectional design. SETTING: Tianjin First Central Hospital and Tianjin Union Medical Center, Tianjin, China. PARTICIPANTS: A total of 9,332 adults (41.35% men) undergoing physical examination. PRIMARY AND SECONDARY OUTCOME MEASURES: T2D was defined using the WHO's criteria: fasting blood glucose (FBG) ≥7.0 mmol/L and/or previous diagnosis of T2D. Height, weight, waist, hip, PBF, VFA, and fasting plasma glucose were measured. RESULTS: All studied adiposity indices were associated with T2D among both males and females, and the observed associations differed by sex. The standardized aORs of BMI, WHR, PBF and VFA for T2D were 1.60 (95% CI 1.42-1.81), 1.43 (95% CI 1.25-1.64), 1.42 (95% CI 1.23-1.62) and 1.53 (95% CI 1.35-1.75) for females, and 1.47 (95% CI 1.31-1.66), 1.40 (95% CI 1.25-1.58), 1.54 (95% CI 1.36-1.74) and 1.47 (95% CI 1.31-1.65) for males, respectively. The AUCs of VFA, WHR and BMI were 0.743, 0.742 and 0.717 in women, respectively, whereas none of the indices had AUC larger than 0.70 in men. The AUCs were not significantly different between VFA and WHR, while both demonstrate larger AUCs than BMI and PBF in females (all p < 0.05). The optimal cutoff values of VFA, WHR, and BMI for T2D in women were 103.55 cm(2), 0.905, and 24.15 kg/m(2), respectively. CONCLUSION: Although BMI, WHR, and PBF and VFA as measured by bioelectrical impedance analysis (BIA) were all positively associated with T2D, their efficacy for predicting the risk of T2D differed by sex. VFA, WHR and BMI could be used as biomarkers to predict T2D risk in women, however none of the study indicators demonstrated favorable efficacy of predicting T2D risk in men.