Given the close relationship between lipids and gestational diabetes mellitus(GDM), the present study sought to examine the association between the triglyceride high-density cholesterol-glucose body index(TyHGB) and the risk of developing GDM.
In order to investigate the relationship between TyHGB and GDM, a retrospective case-control study and a prospective cohort study were conducted involving 1862 individuals. A multifactorial logistic regression model was employed to assess the relationship between TyHGB and GDM, and to reduce potential bias and confounding variables, the association between TyHGB and GDM was evaluated using 1:1 propensity score further matching (PSM) multivariate logistic regression of post-PSM data. The restricted cubic splines(RCS) was used to observe the dose-response relationship between TyHGB and GDM. Receiver operating characteristic (ROC) curves were employed to evaluate the predictive accuracy of GDM. Subsequently, likelihood ratio tests were utilized to investigate the correlation between TyHGB and GDM in specific subgroups.
After adjustment for all confounders, multivariate logistic regression analysis showed that TyHGB was significantly associated with the risk of developing GDM.The ROC curve shows the excellent predictive performance of the TyHGB index.TyHGB was better than other indices in predicting GDM in the first and second trimester.
This study demonstrates that TyHGB is independently associated with GDM risk. TyHGB may be used as a screening and monitoring tool for pregnant women at high risk of GDM rather than other lipid indicators.

© 2025. The Author(s).