PGDM

This study aimed to explore predictors of gestational diabetes mellitus (GDM) by integrating both simple maternal measures and novel biomarkers to determine how effectively GDM can be predicted in the first trimester. The research involved 124 women who developed GDM and 248 control subjects, with data collected on factors such as age, BMI, parity, race, smoking, prior GDM, family history of diabetes, and blood pressure. Blood samples were analyzed for both routine (lipids, high-sensitivity C-reactive protein, and γ-glutamyltransferase) and novel biomarkers (adiponectin, E-selectin, and tissue plasminogen activator [t-PA]). Stepwise regression identified elevated t-PA and low HDL cholesterol as independent predictors of GDM beyond basic maternal factors. Incorporating these biomarkers improved the area under the receiver-operating characteristic curve (AUC-ROC) from 0.824 to 0.861 and the integrated discrimination improvement (IDI) by 0.052, suggesting that GDM prediction can be enhanced by including specific blood measures like lipids and t-PA alongside simple clinical data.