Biomarkers and outcome in children with hypoxic ischemic encephalopathy

Abstract

Approximately 1-6 per 1000 born infants will develop hypoxic-ischemic encephalopathy (HIE) with significant associated mortality and morbidity. For the past 45 years, severity of HIE has been stratified into mild, moderate, and severe. Seizures are common in moderate to severe cases of HIE. There is evidence that neonatal seizures are harmful, especially after perinatal asphyxia, and current clinical consensus is that neonatal seizures should be identified early and treated promptly. Therapeutic hypothermia has been shown to reduce risk of death and disability among newborns with moderate-severe HIE. There is evidence that early cooling improves outcome, which makes timely identification of infants with HIE important.

The overall aim of this thesis was to improve treatment of children with HIE through studies of early identification of infants in need of neuroprotective treatment, and to assess outcome in children with mild HIE. In two separate cohorts, umbilical cord blood miRNA-181b showed potential as a predictor of moderate-severe HIE. Levels of mUCH-L1 were significantly higher in children who developed HIE, but that difference was mainly caused by higher levels in children with severe HIE. In a multicenter randomized controlled trial (RCT) including patients recruited from eight European NICUs, the addition of an automated seizure detection algorithm (ANSeR) did not improve sensitivity in identification of infants with seizures in a clinical setting. We did however observe an increased sensitivity for individual seizure hours, with the largest difference in sensitivity observed during weekends. Using data from four prospective cohorts from Cork, Ireland and Stockholm, Sweden, children with a history of mild HIE at birth were shown to have lower cognitive composite scores measured with BSITD-III at two years of age compared to a healthy control group. The cognitive composite scores of children with mild HIE were not significantly different from that of survivors of moderate HIE treated with therapeutic hypothermia. In a population-based, longitudinal study of all children treated with therapeutic hypothermia at birth between 2007-2009 in Stockholm, Sweden, BSITD-III was shown to detect children at risk of later poor outcome but was insufficient to predict neurocognitive trajectory. We concluded that it is important that studies of outcome in HIE, but also clinical follow-up, continue into school age.