Abstract
IMPORTANCE: Neonatal hypoglycemia is a common metabolic disorder that may cause long-term neurodevelopmental impairment. Reliable early biomarkers to estimate hypoglycemia risk are lacking. OBJECTIVE: To evaluate whether umbilical cord blood glucose (UCBG) values and derived parameters are associated with transitional neonatal hypoglycemia (TNH) and to establish percentile-based UCBG reference values. DESIGN, SETTING, AND PARTICIPANTS: This prospective cohort study was performed at a tertiary perinatal center in Düsseldorf, Germany. Data were collected from May 27, 2020, to September 1, 2022. The final dataset included neonates (gestational age, ≥35 weeks 0 days) with paired arterial and venous umbilical cord blood samples who were exposed or unexposed to hypoglycemia risk factors. Exclusion criteria consisted of incomplete or invalid cord blood samples, missing glucose measurements within the first 4 hours of life, and transient or persistent hyperinsulinism. Maternal and neonatal clinical data were systematically recorded. Data were analyzed from March 18 to July 15, 2025. EXPOSURES: Paired arterial and venous cord blood samples collected at birth; standardized postnatal glucose screening and management. MAIN OUTCOMES AND MEASURES: The primary end point was hypoglycemia, defined as a glucose level of 45 mg/dL or less during postnatal screening; severe hypoglycemia was defined as a glucose level of less than 30 mg/dL. UCBG parameters included arterial and venous glucose values, venous-arterial difference, and glucose extraction rate. Prospective association of UCBG parameters with postnatal hypoglycemia was assessed using receiver operating characteristic (ROC) analyses. RESULTS: Of 598 neonates included in the analysis, the median gestational age was 39 weeks 0 days (IQR, 37 weeks 6 days to 40 weeks 0 days), and 332 (55.5%) were male. Hypoglycemia occurred in 188 neonates (31.4%) and severe hypoglycemia occurred in 22 neonates (3.7%). The median arterial UCBG level was 67 (IQR, 58-84) mg/dL; venous UCBG level, 85 (IQR, 72-102) mg/dL; venous-arterial level difference, 16 (IQR, 10-23) mg/dL; and extraction rate, 19.0% (IQR, 12.2%-25.5%). Arterial UCBG values of 45 mg/dL or less occurred in 22 neonates (3.7%). Compared with neonates with levels greater than 45 mg/dL, those with levels of 45 mg/dL or less had significantly higher median venous-arterial UCBG level differences (15 [IQR, 10-23] mg/dL vs 24 [IQR, 17-38] mg/dL), UCBG extraction rates (18.7% [IQR, 11.9%-25.0%] vs 35.3% [IQR, 27.5%-56.1%]), and higher incidence of severe TNH (3.3% [19/576] vs 13.6% [3/22]). ROC analyses demonstrated poor discriminative performance for all UCBG parameters (maximum area under the ROC curve, 0.64 [95% CI, 0.45-0.81]), and no percentile-based cutoff achieved clinically useful sensitivity or specificity. Percentiles for UCBG values are provided. CONCLUSIONS AND RELEVANCE: The findings of this prospective cohort study suggest that UCBG parameters were not associated with subsequent TNH. However, this study provides a large, single-center, percentile-based UCBG dataset that can support future research and clinical interpretation. Very low UCBG values may help identify neonates with higher likelihood of pathologic hypoglycemia, but this requires further validation.