Body Composition Indicators Jointly Predict Metabolic Unhealthy Phenotypes in Young and Middle-Aged Obese Individuals: A Cross-Sectional Quantitative Computed Tomography Study.

Abstract

The main aim of this study is to analyze the relationship between body composition indices and metabolic unhealthy phenotypes in young and middle-aged obese patients and to assess their joint predictive ability. A cross-sectional study method was used to select 207 patients who were proposed to undergo weight loss surgery for morbid obesity from March to November 2022. Total adipose tissue (TAT), visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), liver fat content (LFC), cross-sectional area (CSAmuscle), and intermuscular adipose tissue (CSAIMAT) of paraspinal muscles were measured using quantitative computed tomography. Participants were categorized into two groups: metabolically healthy obesity (MHO) and metabolically unhealthy obesity (MUO). The receiver operating characteristic curve comprised body composition variables that correlated with MUO, and the area under the curve (AUC) was calculated to compare their prediction capacity for MUO. There were 71 patients with MHO (34.3%) and 136 patients with MUO (65.7%). The VAT, VAT/TAT ratio, LFC, and CSAmuscle was higher in MUO patients than in MHO (all P < 0.001), and SAT was lower than in MHO (P = 0.008). And all of these metrics were correlated with MUO (all P < 0.05). Inclusion of these body composition metrics in the ROC analysis showed that the AUC values for SAT, VAT, VAT/TAT ratio, LFC and CSAmuscle were 0.615, 0.663, 0.727, 0.694, 0.671, respectively, and the combination of the VAT/TAT ratio and the LFC had the ability to predict MUO best (AUC=0.746, P = 0.025). The combined use of VAT/TAT ratio and LFC is superior to the use of these two metrics alone in terms of their ability to predict the MUO, providing a more accurate approach to the management and prevention of obesity-related metabolic risk.