A Novel Single-Test Approach for GDM Diagnosis: Identification and Prediction of High-Risk Postprandial Hyperglycemia

Background: Early prediction of gestational diabetes mellitus (GDM) remains a major clinical challenge, and the current oral glucose tolerance test (OGTT) is time-consuming and inconvenient for clinical routine. This study aimed to develop a novel predictive model for postprandial hyperglycemia GDM (pp-GDM) and postprandial glucose elevation using fasting serological and metabolic profiles. Method: We used High-Performance Liquid Chromatography-Mass Spectrometry (HPLC-MS) to analyze fasting plasma amino acid profiles at 24-28 weeks of gestation for 60 pp-GDM patients and 120 controls. Binary logistic regression model was constructed to identify potential biomarkers for pp-GDM prediction. Results: By incorporating amino acid indicators such as isoleucine, phenylalanine, threonine, and aspartate into the predictive model alongside traditional predictors (including BMI at sampling, fasting insulin, glycated hemoglobin, and uric acid), the overall predictive performance was significantly improved from 78.2% to 91.1%. A clinically practical nomogram for risk assessment was subsequently developed. Conclusions: This fasting metabolite-based model provides a reliable tool for early prediction of pp-GDM and postprandial hyperglycemia, which may reduce the need for OGTT and facilitate timely clinical decision making.