A LASSO-based nomogram for predicting acute bilirubin encephalopathy in newborns with severe hyperbilirubinemia.

Authors

Qifa Hu, Zhenzhu Yao, Haixia Zhu, Xue Feng, Hongxing Li

Published in

Frontiers in pediatrics. Volume 14. Pages 1818387. Epub May 15, 2026.

Abstract

To develop and validate a nomogram for predicting acute bilirubin encephalopathy (ABE) in newborns with severe hyperbilirubinemia.
A retrospective analysis was conducted on 287 newborns with severe hyperbilirubinemia who visited the neonatal department of Shenzhen Children's Hospital from January 2015 to December 2022. A simple random sampling method was used to divide the subjects into a training group (200 cases) and a validation group (87 cases) at a ratio of 7:3, collecting general information and biochemical indicators of the neonates. LASSO regression and cross-validation were performed using RStudio (4.2.3) to select optimal predictors. A multivariate logistic regression model was then constructed and visualized as a nomogram. Model performance was evaluated using receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA).
LASSO regression combined with multivariate logistic analysis identified six potential predictors selected by LASSO. Among these, four were independently associated with ABE in the multivariate model: delivery method (OR = 3.563, 95%CI: 1.391-9.145), birth trauma-related hemorrhage (OR = 3.024, 95%CI: 1.156-7.816), total bilirubin (OR = 1.012, 95%CI: 1.006-1.019), and reticulocyte percentage (OR = 1.185, 95%CI: 1.019-1.478) (all P < 0.05). Breastfeeding (OR = 0.454, 95%CI: 0.084-1.628, P = 0.279) and abnormal hemoglobin (OR = 1.821, 95%CI: 0.654-4.811, P = 0.235) were retained by LASSO but did not reach statistical significance in the multivariate analysis. The area under the curve of the nomogram model for the training and validation sets was 0.792 and 0.822, respectively, with Hosmer-Lemeshow goodness-of-fit values of 4.894 and 3.032, and P-values of 0.558 and 0.805, indicating that the model has good predictive ability and consistency. Decision curve analysis (DCA) for both the training and validation sets showed that the model has good efficacy in predicting the risk of ABE in newborns with severe hyperbilirubinemia.
The nomogram developed in this study demonstrates good accuracy and predictive value, providing a reference for clinical individualized prediction of ABE risk in newborns with severe hyperbilirubinemia.

PMID:
42220995
Bibliographic data and abstract were imported from PubMed on 01 Jun 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.