543 - Neonatal seizures: implementation of an EEG grading scale and its correlation with seizures in neonates with and without HIE

Poster Number: 543
Publication Number: 543.139

Julie A. Nelson, Children's Hospital Colorado, Aurora, CO, United States; Laurel Reed, University of Michigan Medical School, Dexter, MI, United States; Lori Silveira, University of Colorado School of Medicine, Aurora, CO, United States; Krista Eschbach, University of Colorado School of Medicine, Aurora, CO, United States

Background: Neonatal seizures are common and estimated to occur in 1-3/1000 live births. EEG is used to diagnose neonatal seizures and can provide important information regarding prognosis.

Objective: We assessed how frequently neonatal EEGs were given a background classification (BC) prior to implementing a specific neonatal background grading scale. We then examined the relationship between BC, seizure occurrence, and time of seizure onset.

Design/Methods: The first cohort analyzed consisted of EEGs performed between 1/1/2020 and 12/31/2020 at Children’s Hospital Colorado on infants with a corrected gestational age between 36 weeks 0 days and 43 weeks 6 days. An intervention followed consisting of epileptologist education regarding a standardized term neonatal BC (Benedetti, 2020) and implementation of a reporting template that incorporated this classification. The post-intervention cohort analyzed EEGs from 5/1/2021 through 10/31/2021 to assess for improvement in documenting BC. Seizure occurrence rates for each BC were then analyzed.

Results: 203 records were analyzed. Pre-intervention, there were 29/111 (26%) EEGs without a BC compared to 4/92 EEGs (4%) post-intervention. 28% of neonates had HIE. In all neonates, the background on the first day of recording was related to the likelihood of seizures. Neonates with an initial moderately abnormal BC were 4.5 times more likely to have a seizure compared to a mildly abnormal background (p=0.0011). In the HIE group, neonates with a moderately abnormal BC were 7.3 times more likely to have a seizure compared to those with normal to mildly abnormal background (p=0.02). There were 3 neonates with HIE with grading of normal or mildly abnormal with seizures; 2 had seizures in the first 24 hours, and one had a seizure at 48-72 hours. 13 neonates without HIE with grading of normal to mildly abnormal had seizures; the majority (84%) had seizures in the first 24 hours, and 2 had seizures beginning at hours 24-48 of recording.Conclusion(s): The frequency of BC improved with the use of a standardized template and education. If the first day reported a moderately abnormal background in both HIE and non-HIE groups, neonates were more likely to have a seizure compared to those with a grading of normal/mildly abnormal. The majority of neonates with normal/mildly abnormal backgrounds in both groups had seizures in the first 24 hours of recording. We recommend the initiation and use of a standardized neonatal background grading system as BC is associated with differing risks of seizure development, and this can help guide management with respect to length of EEG monitoring time.