Abstract
OBJECTIVE
We evaluated whether early pregnancy oral glucose tolerance testing (OGTT) or continuous glucose monitoring (CGM) improves gestational diabetes mellitus (GDM) or large-for-gestational-age (LGA) birth prediction over clinical factors alone.
RESEARCH DESIGN AND METHODS
In the Glycemic Observation and Metabolic Outcomes in Mothers and Offspring (GO MOMs) study, a prospective observational study at nine U.S. sites (2021–2025), participants with singleton gestation and without preexisting diabetes underwent blinded 75-g OGTT and CGM at 10–14 weeks’ gestation. Predictive models for GDM at 24–28 weeks and LGA were developed using these data and clinical factors (maternal age, BMI, GDM or macrosomia history). New glycemic criteria maximized the area under the receiver operating characteristic curve in training data from four sites and were validated in five. Models incorporating OGTT or CGM criteria versus clinical factors alone compared positive predictive value (PPV) at a prespecified negative predictive value (NPV) at a reported 8.3% national GDM prevalence and 10% for LGA.
RESULTS
Of 2,178 participants, 93.3% who were pregnant at 24–28 weeks completed OGTTs (15.4% GDM), and 98.1% with live births had data to determine LGA (11.2% LGA). At NPV ≥96% and 8.3% GDM prevalence, PPV for GDM was 12.2% (95% CI 11.0–13.3%) for clinical factors alone, 26.5% (23.1–30.3%) for OGTT plus clinical factors, and 19.8% (17.2–22.5%) for CGM plus clinical factors. For LGA, neither glycemic model improved PPV. Validation confirmed findings.
CONCLUSIONS
In a U.S.-representative population, adding 10- to 14-week OGTT or CGM criteria to clinical factors improved GDM but not LGA prediction. Future studies should determine whether predicting GDM in the first trimester facilitates interventions that improve GDM-related pregnancy outcomes.