Change in general and central adiposity measures in prediction of incident dysglycemia; Tehran Lipid and Glucose Study

ObjectivesTo examine the change in general and central adiposity measures as a predictor of incident dysglycemia during a 6-year follow-up.SubjectsA total of 4029 (2333 women and 1696 men) non-dysglycemic Iranians aged ≥ 20 years, underwent standard fasting and 2-h post-challenge plasma glucose tests at baseline and follow-up.ResultsDuring follow-up, 458 new cases of dysglycemia occurred. In multivariable models including baseline values of each anthropometric measure, odds ratios (ORs) for dysglycemia incidence corresponding to a 1-SD increase in changes of body mass index (BMI), waist circumference (WC), waist-to-hip ratio (WHR), waist-to-height ratio (WHtR) and hip circumference (HC), were 1.32, 1.56, 1.39, 1.51 and 1.39 for men and 1.59, 1.50, 1.37, 1.47, and 1.38, for women, respectively (all P < 0.05). After controlling for weight change and WC change among men and women, respectively, HC change did not remain as a predictor. Using the paired homogeneity test, there was no superiority for changes in central obesity measures compared to changes in BMI to predict dysglycemia.ConclusionThe association between HC changes and incident dysglycemia was dependent upon changes in central and general adiposity measures, where the former did not have higher predictability than the latter in prediction of dysglycemia.