Association between triglyceride-glucose index and stages of T2DM: potential indirect effects of inflammatory indicators.

Abstract

Background: The triglyceride-glucose (TyG) index is a reliable surrogate marker of insulin resistance and has been closely linked to type 2 diabetes mellitus (T2DM) risk. However, it remains unclear whether the TyG index is nonlinearly associated with cross-sectional glycemic status categories, from normal glucose to T2DM, and whether inflammatory indicators may partly account for this association. We therefore aimed to examine this nonlinear association and to estimate potential indirect associations through inflammatory indicators.

Methods: A retrospective analysis of cross-sectional data was conducted, including 1,117 participants. Generalized additive models (GAM) combined with restricted cubic spline (RCS) regression were used to examine the nonlinear association between the TyG index and stages of glucose metabolism. Stage specificity analyses were performed across normal glucose, impaired fasting glucose (IFG), impaired glucose tolerance (IGT), combined IFG and IGT, and T2DM. In exploratory cross-sectional indirect association analyses, inflammatory markers accounted for part of the cross-sectional association between the TyG index and glycemic outcomes.

Results: A significant nonlinear relationship was identified between the TyG index and T2DM (P < 0.05), with an inflection point at TyG = 9.14. The TyG index showed a progressively increasing trend across advancing disease stages. In exploratory cross-sectional analyses of potential indirect associations, only MLDLR and NLDLR exhibited significant statistical indirect associations linking the TyG index to 2-hour postprandial plasma glucose (2hPG). However, because all variables were measured concurrently, causal mediation cannot be inferred. These exploratory findings may warrant further investigation in longitudinal studies, particularly across early stages of dysglycemia.

Conclusions: The TyG index was nonlinearly associated with cross-sectional glycemic categories. Exploratory cross-sectional mediation analyses further revealed statistically significant indirect associations through MLDLR and NLDLR in the TyG index-2hPG relationship. These observations suggest that inflammatory markers may be related to glycemic deterioration primarily via postprandial-related pathways in early dysglycemia, although causal inference cannot be drawn from cross-sectional data. These findings are hypothesis-generating and support the prospective evaluation of the TyG index for risk stratification in primary-care settings.