Large for Gestational Age and Neonatal Hyperinsulinemia as Predictors of Adverse Outcomes in OGTT-Positive and OGTT-Negative Pregnancies

Introduction: large for Gestational Age (LGA) infants and elevated insulin levels at birth are key indicators of increased risk for adverse perinatal outcomes. Predictive models that integrate maternal risk factors, clinical data, and biomarkers may enable early identification of high-risk pregnancies and support targeted perinatal and postnatal interventions. Objective: to explore the feasibility of assessing associations between maternal and fetal parameters, particularly sequential fetal abdominal circumference (FAC) growth >25%, and the occurrence of large for gestational age (LGA) and/or neonatal hyperinsulinemia in pregnancies with positive and negative OGTT results, in a hypothesis-generating pilot framework Method: this prospective pilot cohort study included 82 singleton pregnancies followed at a tertiary center in Brazil from January to October 2021. Pregnancies with fetal anomalies or maternal comorbidities were excluded. Fetal abdominal circumference (FAC) was measured via ultrasound at 20-22 and 34 weeks. Umbilical venous insulin was assessed at birth using a chemiluminescent assay. We considered LGA, birthweight >90%, and neonatal hyperinsulinemia as insulin >8.0 µU/mL. A multivariable logistic regression model was used to identify association with combined outcome (LGA and/or neonatal hyperinsulinemia). Results: the final association model, based on maternal weight ≥80 kg, estimated fetal weight ≥80th percentile at 34 weeks, and/or >25% increase in FAC during pregnancy, suggested 96.9% sensitivity, 60% specificity, 60.8% positive predictive value (PPV), and 96.8% negative predictive value (NPV). When OGTT+ status was added to the outcome, association model performance remained robust: sensitivity 79%, specifically 73%, PPV 86%, and NPV 61%. 37.5% of OGTT(-). Conclusion: this pilot exploratory study suggests that maternal weight ≥80 kg, estimated fetal weight at 34 weeks, and sequential increases in fetal abdominal circumference >25% are associated with neonatal hyperinsulinemia and/or LGA. These findings do not validate a predictive model, but rather generate hypotheses and demonstrate the feasibility of incorporating sequential fetal biometry into future research protocols. Larger, multicenter, confirmatory studies with external validation are required before any clinical application can be considered. This study should be interpreted strictly as hypothesis-generating, aiming to inform the design, variables, and analytical strategies of future large-scale confirmatory investigations.