Amplitude-integrated electroencephalography for early diagnosis and severity assessment of neonatal intracranial hemorrhage: a multicenter retrospective study

Neonatal intracranial hemorrhage (ICH) poses substantial neurodevelopmental risks; however, existing neuroimaging modalities (ultrasound, CT, MRI) are limited by accessibility and safety concerns. This study aimed to evaluate the predictive value of amplitude-integrated electroencephalography (aEEG) parameters for the early diagnosis and severity assessment of neonatal ICH. In this multicenter retrospective study, 383 neonates (253 with ICH, 130 without ICH) admitted between 2018 and 2023 were analyzed. aEEG parameters, including cerebral function monitoring (CFM) scores, amplitude limits, and sleep-wake cycling, were compared between ICH and non-ICH groups. ICH severity was classified according to the Papile criteria. Multivariable logistic regression identified independent predictors, and receiver operating characteristic (ROC) curves were used to assess diagnostic performance. CFM scores and lower margin of amplitude (LMA) were significantly reduced in neonates with ICH compared to controls, in both preterm and term subgroups (all P < 0.001). Multivariable logistic regression identified CFM score (OR = 0.38, 95% CI: 0.21–0.68, P = 0.001) and lower limit of amplitude (OR = 0.01, 95% CI: 0.00–0.10, P < 0.001) as protective factors against ICH, while higher LMA was associated with increased risk (OR = 11.56, 95% CI: 3.09–43.20, P < 0.001). The diagnostic model demonstrated good discriminative ability, with an area under the ROC curve (AUC) of 0.849 (95% CI: 0.819–0.879). Furthermore, lower CFM scores and LMA values correlated significantly with increasing ICH severity in term neonates (P < 0.05). aEEG parameters, particularly CFM scores and LMA, serve as reliable indicators for the early diagnosis and severity assessment of neonatal ICH. As a non-invasive bedside tool, aEEG may complement traditional neuroimaging, facilitating earlier intervention and potentially improving neurodevelopmental outcomes.