BACKGROUND AND AIMS: The predictive power of adiposity and body compositions measured from bioelectrical impedance analysis (BIA) for identifying the risk of metabolic syndrome is unknown among ethnic Chinese. METHODS AND RESULTS: We designed a nested case-control study by recruiting 1000 cases of metabolic syndrome and 986 matched controls from a health checkup center. For identifying the metabolic syndrome status, the highest areas under receiver operating characteristic curve (AUCs) were waist-height ratio (WHtR) (0.967, 95% confidence interval [CI], 0.960-0.976). The body fat mass vs. lean body mass and body mass index (BMI) had a similar AUC (0.896 for fat mass vs. lean body mass, 0886 for BMI, P=0.07). WHtR and waist circumference had the highest correctly classified proportions (0.89-0.90) and the highest Youden's index (0.77-0.81). The optimal cut point for WHtR was 52.5, with a sensitivity of 0.92 and specificity of 0.89 for discriminating metabolic syndrome risk. The incremental values of AUC, net reclassification improvement and integrated discrimination improvement values were still highest among WHtR, waist circumference and the percent body fat in the multivariate logistic model. CONCLUSION: Waist circumference and BIA-derived body component measures are suitable for clinical application in identifying the metabolic syndrome status among ethnic Chinese in Taiwan.
BACKGROUND AND AIMS: The predictive power of adiposity and body compositions measured from bioelectrical impedance analysis (BIA) for identifying the risk of metabolic syndrome is unknown among ethnic Chinese. METHODS AND RESULTS: We designed a nested case-control study by recruiting 1000 cases of metabolic syndrome and 986 matched controls from a health checkup center. For identifying the metabolic syndrome status, the highest areas under receiver operating characteristic curve (AUCs) were waist-height ratio (WHtR) (0.967, 95% confidence interval [CI], 0.960-0.976). The body fat mass vs. lean body mass and body mass index (BMI) had a similar AUC (0.896 for fat mass vs. lean body mass, 0886 for BMI, P=0.07). WHtR and waist circumference had the highest correctly classified proportions (0.89-0.90) and the highest Youden's index (0.77-0.81). The optimal cut point for WHtR was 52.5, with a sensitivity of 0.92 and specificity of 0.89 for discriminating metabolic syndrome risk. The incremental values of AUC, net reclassification improvement and integrated discrimination improvement values were still highest among WHtR, waist circumference and the percent body fat in the multivariate logistic model. CONCLUSION: Waist circumference and BIA-derived body component measures are suitable for clinical application in identifying the metabolic syndrome status among ethnic Chinese in Taiwan.