Developing a predictive model for early identification of hypoxic-ischemic encephalopathy in high-risk neonates using
Jack Rausch1, Joshua C Euteneuer2, Nick Townley3
1School of Medicine, Creighton University, Omaha, NE, USA.
Journal of Neonatal-Perinatal Medicine
|July 30, 2026
Summary
A new predictive model identifies neonates at risk for hypoxic-ischemic encephalopathy (HIE) using early clinical data. This tool aids timely diagnosis and initiation of therapeutic hypothermia (TH) treatment for better outcomes.
Area of Science:
- Neonatal Medicine
- Neurology
- Biomarker Discovery
Background:
- Neonatal encephalopathy is a major cause of infant death and disability.
- Hypoxic-ischemic encephalopathy (HIE) accounts for most cases and is treated with therapeutic hypothermia (TH).
- Timely TH treatment (within 6 hours) is crucial but challenging due to nonspecific early clinical signs.
Purpose of the Study:
- To develop a predictive model for identifying neonates at risk of HIE requiring TH.
- To improve early diagnosis and timely treatment initiation.
Main Methods:
- Secondary analysis of 362 neonates from the Midwest Neonatal Encephalopathy Registry.
- Logistic regression model developed using early clinical data.
- Included infants with HIE undergoing TH and control infants needing respiratory support.
Main Results:
- HIE cases showed elevated ALT, AST, creatinine, lactate, nucleated red blood cells, and lower pH.
- A predictive model using base excess, fetal hypoxia signs, and 5-min Apgar scores achieved an AUC of 0.831.
- This model effectively discriminated between HIE and control infants.
Conclusions:
- A simple predictive model using early clinical data shows high accuracy for screening HIE.
- The model is a potential bedside tool for identifying infants needing timely therapeutic hypothermia.
- Further validation in larger cohorts is required.
