| Objective: In different glucose tolerance subjects, to evaluate thecorrelation, sensitivity and specificity of a simple index (TyG) which wasobtained from the product of fasting serum triglyceride and plasma glucose,TyG was compared with homeostasis model assessment for insulinresistance index (HOMA-IR) and beta cell function indices (HOMA-β, DI).Methods:883different glucose tolerance subjects (269normalglucose tolerance individuals,242newly diagnosed impaired glucoseregulation subjects,372newly diagnosed type2diabetes mellitus) wereincluded. All subjects completed a standard oral glucose tolerance test.Fasting serum insulin, lipid profiles and other biochemical markers weremeasured. TyG index was derived from fasting serum triglyceride andplasma glucose. HOMA-IR, HOMA-β and DI were calculated from fastingserum insulin and plasma glucose.Results: Pearson,s correlation analyses showed a positive correlationbetween TyG and HOMA-IR (r=0.567, p＜0.01) and negative correlationsbetween TyG and HOMA-β, DI (r=-0.526, r=-0.741, respectively; both p＜0.01). When compared with HOMA-IR, the ROC scatter plot revealed that the best cut point of the TyG index for diagnosis of insulin resistancecorresponded to4.92, which showed the highest sensitivity (71.5%) andspecificity (79.7%; AUC+0.803) and the biggest Youden index (0.512).When compared with HOMA-β, the ROC scatter plot revealed that the bestcut point of the TyG index for diagnosis of β cell dysfunction correspondedto4.90, which showed the highest sensitivity (63.6%) and specificity(74.6%; AUC+0.750) and the biggest Youden index (0.383). Whencompared with DI, the ROC scatter plot revealed that the best cut point ofthe TyG index for diagnosis of β cell dysfunction corresponded to4.92either,which showed the highest sensitivity (70.5%) and specificity (86.0%;AUC+0.856) and the biggest Youden index (0.565).Conclusion: The TyG index could be used as surrogate to assessfasting state of insulin resistance and islet beta cell function. |
PDF Full Text Request