Easy-to-use nomogram to predict neonatal hyperbilirubinemia.

BACKGROUND

Neonatal hyperbilirubinemia is a common condition and a leading cause of hospitalization in newborns in their first week of life. Thus early identification of infants at risk is particularly important. In this study, we explored risk factors for its development of neonatal hyperbilirubinemia, and then constructed and validated an easy-to-use nomogram for the early prediction.

METHODS

This study was conducted retrospectively and non-interventionally, involving 646 neonates born at the Second Hospital of Dalian Medical University between January 2021 and January 2024. The study population was systematically partitioned through cluster sampling into a training set comprising of 454 neonates and a validation set of 192 neonates, adhering to a 7:3 ratio, utilizing the R-4.4.0 program. Independent predictors of neonatal hyperbilirubinemia were identified using least absolute shrinkage and selection operator (LASSO) regression from the training set, and a nomogram was constructed based on these predictors. The performance of the nomogram was assessed using receiver operating characteristic (ROC) curves, calibration curves and decision curve analysis (DCA).

RESULTS

Among 646 newborns, there were 350 males and 296 females, with a mean gestational age (GA) of 38.4 ± 1.4 weeks and birth weight (BW) of 3,264.1 ± 490.7 g. Six independent factors associated with hyperbilirubinemia were identified: GA, BW, premature rupture of membranes (PROM) ≥ 18 hours or concurrent maternal fever, maternal-infant blood type incompatibility with positive direct Coombs test, supplementation with probiotics, and weight loss > 9% within 3 days. Calibration curves indicated that the nomogram closely matched the actual observed values in both the training and validation sets. The areas under the ROC curves for predicting hyperbilirubinemia were 0.825 (95% confidence interval (CI) [0.777-0.874]) in the training set and 0.829 (95% CI [0.757-0.901]) in the validation set. DCA showed that the nomogram has clinical applicability.

CONCLUSION

The nomogram constructed in this study has good differentiation, calibration and clinical applicability, and has the potential to be used for predicting neonatal hyperbilirubinemia.