Abstract
AIM: Rapid and growing rise in obesity and diabetes mellitus, as serious human health-threatening issues, is alarming. The aim of the present study was assessing the accuracy of several obesity indices to predict hyperglycemia in overweight and obese Iranian populations and determining the value of such indices in comparison to the conventional parameters. We also evaluated new latent combined scores in this matter. PATIENTS AND METHODS: Overall, there were 2088 patients recruited from the weight loss clinic of Sina Hospital, an educational hospital of Tehran University of Medical Sciences for this cross-sectional study. Demographic information, anthropometric indices and biochemical measurements were collected and calculated. The multivariable regression modeling as well as area under the receiver-operating characteristic (ROC) analysis was used. To detect the existence of new combined scores, we used SEM (structural equation modeling) analysis through SmartPLS. RESULTS: Combined latent scores and WHtR (waist-to-height ratio) gave us a higher area under the curve in predicting hyperglycemia associated with WC (waist circumference) in women, whereas FFMI (fat-free mass index) gave low values. Additionally, BRI (body roundness index) and latent scores had slightly higher AUC values in predicting hyperglycemia in men. According to the age-adjusted odds ratio (OR) in the presence of hyperglycemia, OR was the highest for WHR (waist to hip ratio) in women (OR, 7.74; 95% confidence interval [CI], 1.71-15.13). The association of WHR and hyperglycemia remained significant by adjusting for BMI (body mass index), WC and menopausal status. CONCLUSION: WHR had the strongest association with hyperglycemia in women with only sufficient discrimination ability. However, neither BSI (body shape index) and BAI (body adiposity index) nor FMI (fat mass index) and FFMI were superior to BMI (body mass index), WC or WHtR in predicting hyperglycemia. It was revealed that BRI and combined scores had a more predictive power compared to the BSI, BAI, FMI and FFMI, simplifying hyperglycemia evaluation.