A Pilot Study on Dynamic Multimodal Neuromonitoring for Predicting Neurodevelopmental Outcomes in Neonatal Hypoxic-Ischemic Encephalopathy

Abstract

Listen

Translate article

Introduction: Multimodal neuromonitoring at the bedside is essential for understanding the pathophysiological mechanisms of brain injury and neurodevelopmental outcomes associated with neonatal hypoxic-ischemic encephalopathy (HIE). While previous research has focused on single modality neuromonitoring biomarkers to predict neurodevelopmental impairment (NDI) at 2 years of age, there remains significant gap in exploring the potential of multimodal physiological signal biomarkers to improve predictive accuracy. This study aimed to evaluate multimodal quantitative neuromonitoring biomarkers within the first day of life to improve prediction of NDI. Methods: This prospective cohort study enrolled newborns (≥36 weeks) diagnosed with HIE at Parkland Hospital in Dallas, TX. A Sarnat examination was performed to determine the severity of HIE within the first 6 h of life, and the Total Sarnat Score (TSS) was calculated. Newborns with moderate and severe HIE received therapeutic hypothermia (TH). Neuronal noninvasive biomarkers including electroencephalogram (EEG) delta power (DP, 0.5–4 Hz) and neurovascular coupling (NVC), calculated as wavelet coherence between amplitude-integrated EEG and cerebral tissue oxygen saturation (SctO2), were measured. NDI was defined as death or a cognitive score <85 on the Bayley Scales of Infant and Toddler Development. The predictive ability of individual biomarkers (TSS, DP, and NVC) and their combination for NDI was evaluated using receiver operating characteristic (ROC) curves, with the area under the ROC curve (AUC) indicating prediction accuracy. Additionally, a Net Reclassification Index (NRI) analysis was conducted to assess the predictive performance of the three baseline models (TSS, DP, and NVC). Results: Forty-six newborns with mild to severe HIE were enrolled and neuromonitoring was initiated at 12 ± 6 h of life. Death or NDI was diagnosed in 18 (6 mild, 10 moderate, 2 severe) infants. Eight out of 46 infants did not complete the 18–24 months follow-up but had a normal examination prior. The combination of all three biomarkers (TSS, DP, and NVC) yielded the highest AUC of 0.755 (95% CI: 0.569–0.941), with sensitivity of 0.750, specificity of 0.769, positive predictive value of 0.800, and negative predictive value of 0.714, outperforming individual biomarkers or two-marker combinations. Furthermore, the NRI analysis demonstrated that the combined model achieved the highest NRI value (0.5577), indicating the strongest improvement in risk classification. Conclusion: This study emphasizes the importance of implementing multimodal neuromonitoring and integrating quantitative biomarkers at the bedside during the first day of life to provide objective metrics on brain health in addition to neurological exam. These approaches demonstrate potential for enhancing the prediction of encephalopathy severity, brain injury, and NDI in the early hours of life, aiding clinicians in timely and effective decision-making for neuroprotective interventions. However, validation through multicenter studies with large cohorts is needed for clinical implementation.