Outcome Prediction in Neonatal Hypoxic-Ischaemic Encephalopathy Using Neurophysiology and Neuroimaging

Abstract

OBJECTIVE: The aim of the study was to determine the predictive power of the combined use of neurophysiological (amplitude-integrated electroencephalography [aEEG], near-infrared spectroscopy [NIRS]) methods and neuroimaging (magnetic resonance imaging [MRI]) for long-term outcome prediction in neonates with hypoxic-ischaemic encephalopathy (HIE). STUDY DESIGN: Prospective cohort study of 56 patients with moderate to severe HIE and hypothermia treatment at the Medical University of Vienna between 2008 and 2020. aEEG and NIRS were recorded continuously over a period of >4 days (102 h) starting at the initiation of hypothermia treatment, MRI was performed at a median age of 8 days. Receiver operating characteristic curves and area under the curve were calculated to evaluate the prognostic ability of aEEG, NIRS, and MRI parameters for outcome assessed via Bayley Scales of Infant Development 3rd edition at 2 years of age. RESULTS: Combined aEEG and MRI parameters showed highest predictive power regarding long-term outcome. The highest area under the curve values (0.96-0.99) were obtained for aEEG (combination of background pattern and sleep-wake cycling) between 66 and 102 h after initiation of hypothermia in combination with MRI findings. NIRS parameters did not differ significantly between infants with favourable and adverse outcome. CONCLUSIONS: Combined aEEG and MRI parameter scores were more predictive than single parameter scores. No further improvement was observed when combining aEEG/MRI with NIRS data.