A novel clinical risk scoring system for neurodevelopmental outcomes among survivors of neonatal hypoxic-ischemic encephalopathy (HIE)

Abstract

Objective

We aimed to develop a risk scoring system as a predictor of 24-month neurodevelopmental outcomes (cognitive, language, and motor) for neonates treated with therapeutic hypothermia for hypoxic-ischemic encephalopathy (HIE).

Methods

This was a chart review of infants with HIE treated with therapeutic hypothermia who were admitted to the Neonatal Intensive Care Unit (NICU) at the University of Michigan between 2009 and 2019 and followed in the neonatal developmental clinic until 24 months of age. We examined bivariate associations between the neonatal characteristics and Bayley-III scores. We then performed stepwise logistic regression. To create the risk scores, a participant was given one point for each of the factors included in the final model.

Results

Fifty-five infants were included. The final model for Bayley cognitive abnormality included abnormal neonatal neurologic exam (p < 0.0001), white matter/watershed MRI abnormality (p = 0.01), 5-min Apgar score (p = 0.02), and EEG-confirmed seizures (p = 0.04). The model for language abnormality included abnormal neurologic exam (p = 0.0002), seizures (p = 0.007), clinical severity of HIE (p = 0.06), and basal ganglia/thalamus MRI abnormality (p = 0.17). The model for motor abnormality included seizures (p = 0.03), abnormal neurologic exam (p = 0.06) and basal ganglia/thalamus MRI abnormality (p = 0.02). The positive predictive values for the risk scores were 60 %, 85 % and 71 %, respectively, for the Bayley-III cognitive, language and motor domains.

Conclusion

Our study identifies early clinical features that differentially predict domains of neurodevelopmental outcome and associated risk scores that may be of value to both clinicians and families. This novel scoring system should next be validated in a larger, prospective study.