Early risk stratification for gestational dysglycemia may help improve maternal and neonatal outcomes. A cross-sectional discovery study with external validation, published in Frontiers in Endocrinology, compared the triglyceride-glucose (TyG) index and the cholesterol-high-density lipoprotein-glucose (CHG) index for associations with gestational dysglycemia and gestational diabetes mellitus (GDM).

The discovery cohort used survey-weighted National Health and Nutrition Examination Survey (NHANES) 2007–2018 data, with self-reported GDM history as the primary outcome. An independent validation cohort included 217 individuals with clinically diagnosed GDM. Multivariable logistic regression, receiver operating characteristic analysis, calibration analysis, and decision curve analysis were used. Additional analyses included adjustment for continuous fasting blood glucose in NHANES and gestational-week-adjusted sensitivity analyses in the validation cohort.

In NHANES, CHG showed a stronger adjusted association with prior GDM than TyG and a numerically higher area under the curve. After adjustment for continuous fasting blood glucose, the TyG association was attenuated, whereas CHG remained statistically significant. In the validation cohort, CHG again showed numerically higher discriminative performance than TyG, with consistent findings after gestational-week adjustment.

Supportive analyses in currently pregnant NHANES participants showed similar directional findings, although estimates were statistically imprecise because of limited sample size. Both indices were simple, low-cost measures associated with gestational dysglycemia or GDM. CHG generally outperformed TyG across cohorts, but overall discrimination remained modest. These indices may serve as potential risk markers rather than standalone screening tools.