Abstracts

Rationale: The clinical recognition of seizures in the neonate remains a challenge due to their subtle clinical presentation. The American Clinical Neurophysiology Society had published continuous EEG (cEEG) monitoring guidelines for neonates who have abnormal paroxysmal events recommending prompt cEEG monitoring when clinically indicated¹. However, initiation of cEEG monitoring is often delayed due to technical and staffing challenges. As a result, the decision to start treatment is based solely on clinical identification of paroxysmal or autonomic events._x000D_ _x000D_ Purpose: The aim of this study is to compare rates of abnormal EEG findings among neonates having autonomic, paroxysmal, and mixed events.

Methods: This is a single-center, retrospective study of neonates born ≥36 weeks of gestation age (GA) from January 2018 to April 2022 who had initial clinical suspicion for seizures and were monitored with cEEG.  Infants were classified into 3 groups based on their clinical presentation. Group 1 (autonomic signs, n=45), Group 2 (paroxysmal events without autonomic signs, n=38), and Group 3 (paroxysmal events with autonomic signs, n=31). Underlying etiologies for clinical events were identified and classified in each group. An adverse outcome was defined as the presence of seizures and/or markedly abnormal EEG background²(score 3-4) based on the cEEG. The associations between the groups, perinatal characteristics, various etiologies and presence of adverse outcome were assessed using logistic regression analysis._x000D_
Results: A total of 114 infants were included. Infants in group 1 had lower GA and fewer identifiable causes for their symptoms than infants in the other groups (table). Perinatal stroke and neonatal asphyxia accounted for 73% of abnormal EEGs and were more likely to be associated with paroxysmal events with or without autonomic symptoms (groups 2 and 3). Congenital infection and brain anomalies accounted for 24 % of abnormal EEGs, and their clinical characteristics were similar among the groups. The presence of paroxysmal events (groups 2 and 3), stroke, asphyxia, congenital infection and brain anomalies were significantly associated with higher rates of abnormal EEG. On further multivariate analysis, significantly higher rates of abnormal EEG was exclusively associated with infants in group 3 (OR 9.27 with 95%CI (1.36-63.3), p=0.023). The area under the curve of the ROC curve of this model is 0.81. _x000D_
Conclusions: Neonates who present with mixed paroxysmal spells and autonomic signs are likely to have markedly abnormal EEG background and/or seizures. Early treatment is recommended prior to onset of EEG monitoring in these infants. _x000D_ _x000D_ References:_x000D_ _x000D_ 1.    Shellhaas RA, Chang T, Tsuchida T, et al. The American Clinical Neurophysiology Society's Guideline on Continuous Electroencephalography Monitoring in Neonates. J Clin Neurophysiol. 2011;28(6):611-617. _x000D_ _x000D_ 2.   Murray DM, Boylan GB, et al. Early EEG findings in hypoxic-ischemic encephalopathy predict outcomes at 2 years. Pediatrics. 2009 Sep;124(3):e459-67._x000D_

Funding: Not applicable