Umbilical cord blood markers predict neonatal hypoglycemia in full-term small and large for gestational age infants.

Abstract

Neonatal hypoglycemia in small for gestational age (SGA) and large for gestational age (LGA) infants can lead to neurological complications. Predictive factors including umbilical cord blood analysis, remain largely unexplored. We retrospectively investigated umbilical cord blood parameters as potential predictors of hypoglycemia in full-term SGA (n = 90) and LGA (n = 105) infants. Blood glucose was measured at 11 timepoints over 72 h. Hypoglycemia was defined as blood glucose <40 mg/dL. Hypoglycemia occurred in 13.3% of SGA and 14.3% of LGA infants, with all episodes occurring within the first 2 h of life. In SGA infants, cord blood glucose <69 mg/dL predicted hypoglycemia with 83% sensitivity, 59% specificity, and 96% negative predictive value (NPV); area under the curve (AUC) 0.78, 95% confidence interval (CI): 0.66-0.89. In LGA infants, cord blood bicarbonate >23.1 mmol/L showed 73% sensitivity, 54% specificity, 92% NPV; AUC 0.67, 95% CI: 0.52-0.82, though with marginal statistical significance after multiple comparison correction. Distinct cord blood markers reflect different pathophysiological mechanisms in SGA and LGA infants. While the glucose threshold <69 mg/dL provides a clinically applicable prediction for SGA infants, bicarbonate findings in LGA infants require prospective validation before clinical implementation. This study identifies distinct umbilical cord blood parameters as predictors of hypoglycemia in SGA and LGA infants. The findings reveal different pathophysiological mechanisms of hypoglycemia between SGA and LGA infants. Lower cord blood glucose levels predict hypoglycemia in SGA infants with good accuracy (AUC 0.78), while higher bicarbonate levels show potential association with hypoglycemia risk in LGA infants requiring validation. These group-specific markers could contribute to optimizing the frequency and timing of glucose monitoring strategies tailored to individual risk profiles.