The Role of METS-IR in Early Screening for Gestational Diabetes Mellitus in Chinese Women: A Two-Center Prospective Study

Objective: This two-center prospective cohort study aimed to evaluate the predictive value of the Metabolic Score for Insulin Resistance (METS-IR) for gestational diabetes mellitus (GDM) in Chinese women during early pregnancy and compare its performance with conventional insulin resistance (IR) indices. Methods: This prospective investigation evaluated 1450 Chinese gravidas (<12 gestational weeks) without pregestational diabetes from two obstetrical institutions. Baseline clinical-biochemical profiling occurred during the first trimester (6-12 weeks), with GDM confirmation via standardized 75g oral glucose tolerance testing at 24-28 weeks' gestation. Analytical methodologies incorporated multivariable regression modeling and ROC curve optimization to quantify the predictive validity of five metabolic indices (METS-IR, TyG index, TG/HDL-C ratio, HOMA-IR, TyG-BMI) for GDM risk stratification. Results: Among participants, 378 (26.1%) developed GDM. The GDM group (n=378, 26.1%) was older (median age 31.0 vs 30.0 years, p<0.0001) and had higher prepregnancy BMI (22.62 vs 21.32 kg/m², p<0.0001), fasting glucose (4.7 vs 4.5 mmol/L, p<0.0001) compared to NGT. The GDM group exhibited significantly higher METS-IR (31.14 vs 29.03, p <0.001) and other IR indices (p <0.001). In unadjusted models, METS-IR quartile 4 (Q4) was strongly associated with GDM (OR=3.33, 95% CI:2.38-4.70), but this association attenuated after adjusting for age, weight gain, lipids, and insulin (adjusted OR=1.56, 95% CI:1.04-2.35). Comparatively, the TyG index (adjusted OR=3.06, 95% CI:2.03-4.66) and TG/HDL-C ratio (adjusted OR=2.02, 95% CI:1.38-2.99) retained robust predictive power. ROC analysis revealed a moderate discriminative capacity for METS-IR (AUC=0.629 unadjusted; 0.676 fully adjusted), outperformed by HOMA-IR (AUC=0.699) and TyG (AUC=0.699). METS-IR demonstrated high specificity (76.2%) in unadjusted screening but showed dependency on metabolic confounders in adjusted models. Conclusion: METS-IR shows promise for early GDM screening and is outperformed by TyG and HOMA-IR in predictive value. METS-IR in early pregnancy reflects metabolic dysregulation linked to GDM risk, yet its predictive utility is partially mediated by lipid and insulin abnormalities. While METS-IR offers clinical feasibility through routine measurements, TyG and HOMA-IR exhibit superior independent predictive value. These findings highlight the importance of context-specific IR indices for early GDM risk stratification and underscore METS-IR's role as a composite marker of metabolically unhealthy obesity in pregnancy.