Identification of insulin-sensitive obese vs. insulin resistant obese postmenopausal women: Evaluation of surrogate indices of insulin sensitivity.
Introduction. Obesity often coexists with insulin resistance, which is related to cardiometabolic risk. However, some obese individuals exhibit comparable insulin sensitivity (IS) to that of normal-weight subjects, a state associated with a reduced cardiometabolic risk. We aimed to determine the efficacy of a panel of surrogate markers of insulin sensitivity (IS) for the identification of insulin sensitive obese (ISO) vs. insulin resistant obese (IRO) with similar total fat mass (FM) and body mass index (BMI). Methods. This is a cross-sectional analysis among 144 overweight and obese post-menopausal women. IS was determined by the hyperinsulinemic-euglycemic clamp (HEC) and by surrogate indices such as Matsuda index, the simple index assessing insulin sensitivity using oral glucose tolerance test (SIisOGTT), Abdul-Ghani liver IS index, HOMA-IR and Abdul-Ghani muscle IR index. Results. When using upper and lower quartiles values or the median as cut-off for IS determined by the reference HEC to define ISO vs. IRO, Matsuda index, SIisOGTT and Abdul-Ghani indices classification identified ISO vs. IRO individuals with similar FM and BMI. With HOMA-IR, the two groups were similar for FM and had borderline significant difference in BMI. Using, receiver operating characteristic curves, Matsuda index AUC was similar to that of SIisOGTT and both indices AUCs were significantly higher than Abdul-Ghani indices AUCs. The best cut-off value for the Matsuda index was 2.5 (83.1% specificity, 54.2% sensitivity) and 0.25 for SIisOGTT (64.8% specificity, 70.8% sensitivity). Conclusion. Whole body IS indices, Matsuda and SIisOGTT indices seem to be reliable indices for the identification of ISO vs. IRO individuals.