Discriminatory capacity of obesity indicators as predictors of high liver fat in US adolescents.

Abstract

The aims of this study were to assess the potential of 16 anthropometric, body composition and endocrine indexes as predictors of high liver fat and determine the most appropriate cut-off points in US adolescents. A cross-sectional study was conducted on a population of 816 adolescents aged 12-17years. The FibroScan® 502V2 device was used to estimate the controlled attenuation parameter (CAP). Body fat percentage, fat mass, trunk fat percentage and trunk fat mass were measured by dual-energy X-ray absorptiometry. Anthropometric data and metabolic parameters were determined. Receiver operating characteristic curves were analysed to estimate the optimal cut-off points that best identify adolescents with high liver fat (CAP ≥90th percentile). In boys, triponderal mass index (TMI) had the highest area under curve (AUC) value (0.865) and the optimal cut-off score for TMI was 17.47kg/m3 , which had 81.32 sensitivity and 82.99 specificity. In girls, trunk fat index (TFI) had the highest AUC value (0.826) and its optimal cut-off score in screening for high liver fat was 3.76kg/m2 , which had 74.04 sensitivity and 88.03 specificity. Fat mass index (FMI) index had the second highest AUC values (0.863 in boys 0.812 in girls) in both sex; the cut-off point for the detection of high liver fat was <8.66kg/m2 for girls and <7.45kg/m2 for boys. Assessment of TMI in boys, TFI in girls, and FMI in both sexes are low-cost and easy-to-use parameters that may be useful as early screening tools for possible high liver fat in adolescents.