The Prognostic Weight of Clinical, Biochemical, Electrographic, and Neuroimaging Biomarkers in Perinatal Hypoxic-Ischemic Encephalopathy Spectrum.

Abstract

To assess the prognostic weight of potential biomarkers in infants within the full spectrum of mild, moderate, and severe hypoxic-ischemic encephalopathy. This observational study was conducted as a nested substudy of a prospectively collected perinatal hypoxic-ischemic encephalopathy cohort (RECOVER [Remote Early Intervention for Cerebral Palsy to Improve Outcomes Using Virtual Care Following Perinatal Asphyxia] study, ClinicalTrials.gov ID: NCT04913324) at the University of Toronto, Hospital for Sick Children. Clinical, laboratory, electrographic, and neuroimaging biomarkers were longitudinally collected and objectively evaluated using scoring systems. Neurodevelopmental outcomes were assessed at 18 months corrected age using standardized tests. Of the 200 infants included in the cohort, the severity of neonatal encephalopathy was classified as mild in 40 (20%), moderate in 118 (59%), and severe in 42 (21%). Of these infants, 27 (14%) died. In the multivariable model, brain magnetic resonance imaging deep gray matter (DGM) injury subscore was the only prognostic marker that was associated with adverse outcomes (OR 1.73, 95% CI: 1.20 to 2.49; P = .003) after accounting for Apgar score at 10 minutes, pre- and postrewarming Thompson score, presence of a benign clinical course, absence of electrographic background normalization in 48 hours, absence of sleep-wake cycling in 72 hours, and electrographically confirmed seizures. A DGM injury subscore cutoff of 6 demonstrated a sensitivity of 82% and a specificity of 94% with an area under the curve of 0.91 (95% CI: 0.84 to 0.98; P < .001). The DGM injury subscore emerged as the only independent predictor of adverse outcome at 18 months corrected age, after adjusting for longitudinal clinical, laboratory, electrographic, and neuroimaging biomarkers.