Performance of VAI, BRI and ABSI in the Diagnosis of Metabolic Syndrome Among Adults with Obesity

Background: Metabolic syndrome (MetS) is a group of metabolic abnormalities that include central obesity, hypertension, insulin resistance and atherogenic dyslipidemia. It is considered a public health problem and its early diagnosis is essential. Currently, the potential usefulness of novel composite measures such as body shape index (ABSI), visceral adiposity index (VAI) and body roundness index (BRI) has been reported. Methods: We conducted a cross-sectional study using the database of a private clinic in Lima, Peru. The study population consisted of adults with BMI>30kg/m2 who underwent bariatric surgery between 2017-2020. The outcome was MetS, defined according to the International Diabetes Federation criteria for Latin population. ABSI, VAI and BRI values were calculated according to formulas reported in the literature. To evaluate diagnostic performance, areas under the ROC curve (AUC) were constructed and cut-off points were estimated using Youden's method. Subsequently, we assessed the association between each of the three dichotomized measures (according to the cut-off point) with MetS using generalized linear models with Poisson family, log link function and robust variances. Analyses were presented stratified by sex. Results: Data from 262 participants were analyzed. The prevalence of MetS was 54.2%. Overall, the AUC for ABSI, VAI and BRI were 0.57, 0.91 and 0.57, respectively. The VAI cutoff points for males and females were 1.97 and 2.76, respectively. After adjusting for confounders, elevated VAI, in the whole sample, was associated with an increased likelihood of MetS (PRa: 3.2; 95% CI: 2.51 - 4.08; p