Umbilical Artery Cord Blood Glucose Predicted hypoglycemia in Gestational Diabetes Mellitus and Other at-risk Newborns

Abstract

Abstract Backgrounds: To explore the value of umbilical artery cord blood glucose (UACBG) in predicting hypoglycemia in gestational diabetes mellitus (GDM) and other at-risk newborns, and to provide a cut-off UACBG value for predicting hypoglycemia occurrence. Methods: In this prospective study, we enrolled at-risk infants delivered vaginally, including neonates born to mothers with GDM, premature, macrosomia, and low birth weight. We separated the infants into GDM group and other at-risk group. All subjects underwent UACBG measurement during delivery. Neonatal peripheral blood glucose measurement was performed at 0.5 hand 2 h after birth. The predictive performance of UACBGfor neonatal hypoglycemia was assessed using receiver operating characteristic curve (ROC), area under curve (AUC), sensitivity and specificity. Results: 916 newborns were included, 538 in GDM group and 378 in other at-risk group. For hypoglycemia prediction within 2 h, the best cut-off of UACBG was 4.150 mmol/L, yielding an AUC of 0.688 (95% CI 0.625-0.751). UACBG had a better performance at 0.5 h (AUC 0.757; 95% CI 0.690-0.824). In detail, the AUC was 0.703 in GDM group (95% CI 0.594-0.811), with the optimal cut-off of 4.550 mmol/L. In other at-risk group, the AUC was 0.797（95% CI 0.716-0.877), and the best threshold was 4.350 mmol/L. No significant differences were observed between GDM group and other at-risk group in AUC at 0.5 h, 2 h and within 2 h. Conclusions: UACBG may be an efficient, non-invasive approach to predicting neonatal hypoglycemia within 2 h after birth.