Five-year outcomes after cooling for hypoxic-ischemic encephalopathy and predictive value of aEEG

Therapeutic hypothermia (TH) is standard practice for neonatal hypoxic-ischemic encephalopathy (HIE). Persistently abnormal amplitude-integrated electroencephalography (aEEG) background patterns are associated with adverse outcomes. We aimed to determine neurodevelopmental outcomes at age five and the predictive value of aEEG.

We studied infants with HIE treated with TH between 2013 and 2018. At age five, we assessed motor (neurological examination, movement-ABC) and cognitive outcomes. We classified children as developing typically (IQ ≥ 85 and M-ABC > p16), mildly abnormal (IQ 70-84 and/or cerebral palsy with GMFCS 1-2 and/or M-ABC ≤ p16), and severely abnormal (IQ < 70 and/or cerebral palsy GMFCS >2). During the first 96 h after birth, we assessed aEEG background patterns and presence of seizures.

Out of 95 infants 14 (15 %) died and we lost 18 (22 %) survivors to follow-up. At age five, 33 (52 %) were developing typically, 21 (33 %) mildly abnormal, and 9 (14 %) severely abnormal. A severely abnormal aEEG at 48 h was the best predictor for death/severely abnormal outcome with a PPV of 88 %, and a PPV of 71 % for severe disability in survivors. Children with death/severely abnormal outcomes had electrographic seizures more frequently and a higher seizure burden than children developing typically/mildly abnormal (61 % vs 11 %, P < .001, median 40 vs 0 min, P < .001. Absence of electrographic seizures was highly predictive (NPV 92 %) of typical/mildly abnormal outcome in survivors.

Among survivors, five-year outcomes after TH were favorable. aEEG background patterns at 48 h and seizure burden were reliable predictors of five-year outcomes.