Early Prognostic Model for Predicting Adverse Outcomes in Neonates with Hypoxic-Ischemic Encephalopathy before Therapeutic Hypothermia

Abstract

Hypoxic-ischemic encephalopathy (HIE) affects 1.3–1.7 per 1000 live births and remains a major cause of neurodevelopmental impairment (NDI). Despite therapeutic hypothermia (TH), nearly half of infants with moderate to severe HIE experience death or NDI. Identifying early prognostic indicators before TH initiation is crucial for improving management and outcomes. We conducted a retrospective case–control study of 144 infants with HIE treated with TH at Kagoshima City Hospital (2000–2022); 100 underwent developmental evaluations at 18 months. Clinical parameters, including amplitude-integrated EEG (aEEG), Thompson scores, and resuscitation details, were analyzed. Logistic regression identified predictors of adverse outcomes: death, cerebral palsy, or developmental quotient <70. Univariate analysis revealed significant predictors, including low Apgar scores, low umbilical artery pH, aEEG abnormalities, high Thompson scores, and resuscitation details. Multivariate regression identified three independent predictors: aEEG abnormalities (adjusted odds ratios [aOR] 7.1, 95% confidence interval [CI]: 1.3–38.2), Thompson score ≥12 (aOR 5.4, 95% CI: 1.5–18.7), and chest compressions (aOR 31.6, 95% CI: 4.3–231.6). We developed and derived early prognostic model from these predictors, assigning +2 points for aEEG abnormalities, +2 points for a Thompson score ≥12, and +3 points for chest compressions. A total score ≥4 achieved high sensitivity (70.4%) and specificity (90.4%), with an area under the curve of 0.87 (95% CI: 0.77–0.94). The early prognostic model may serve as an effective tool for early risk stratification in neonates with HIE before TH initiation, supporting individualized treatment decisions. This score could help identify high-risk neonates who may benefit from additional neuroprotective strategies.