Predictors of adverse outcome in infants with hypoxic-ischemic encephalopathy treated with therapeutic hypothermia: a real-world cohort study

INTRODUCTION: To identify robust predictors of adverse outcomes in infants with hypoxic-ischemic encephalopathy (HIE) from a real-world referral center cohort, using readily available clinical characteristics, biochemical markers, electroencephalography (EEG), and cranial magnetic resonance imaging (MRI), and to develop a practical prediction approach applicable to diverse settings, particularly in low- and middle-income countries (LMICs). METHODS: We conducted a retrospective cohort study of 116 infants with HIE who underwent therapeutic hypothermia (TH). Clinical features, biochemical parameters, EEG data, and MRI findings were analyzed. MRI lesions were assessed using a modified MRI scoring system, and neurodevelopmental outcomes were evaluated at 2 years of age. Variables with P < 0.05 in the univariate analysis, selected based on established clinical relevance and previously published evidence, were entered into the multivariate logistic regression model to identify independent factors associated with adverse outcomes. Receiver operating characteristic (ROC) curve analysis was performed to assess predictive performance. RESULTS: Among the 116 infants, 32 (27.6%) experienced adverse outcomes, including 20 deaths (17.2%). In multivariate analysis, the gray matter (GM) subscore on MRI emerged as the sole independent predictor of adverse outcomes (OR 1.6, 95% CI 1.3-2.1). A GM subscore ≥ 9 demonstrated excellent predictive accuracy, with an area under the ROC curve of 0.96, sensitivity of 84.4%, and specificity of 100%. CONCLUSION: The GM subscore is a practical and reliable independent predictor of adverse neurodevelopmental outcomes at 2 years of age in infants with HIE treated with TH, making it a valuable prognostic tool with significant clinical implications, particularly in LMICs.