Abstracts

Rationale: Evidence based management of seizures in neonates has not progressed for more than a decade, even though existing treatments are incompletely effective and may be harmful when used long term. The Neonatal Seizure Registry (NSR) aims to define contemporary practice of neonatal seizure management using a prospective observational cohort study of all neonates treated for seizures at 8 American pediatric epilepsy centers that follow the American Clinical Neurophysiology Society’s guideline for neonatal EEG monitoring.Methods: Every newborn with seizures cared for at the NSR sites was enrolled. Details regarding medical management and treatment response were recorded. Descriptive statistics are presented.Results: From 01/2013 to 05/2015, 434 newborns with seizures were enrolled. Phenobarbital (PB) was the most commonly prescribed first line seizure medication (89%) and was often administered before EEG monitoring began (N=186; 44%). The mean initial PB loading dose was 19±5mg/kg, but varied by institution (p=0.014). Electrographic seizures persisted after the first PB load in 266 subjects (64.7%). Use of levetiracetam (LEV) varied across study sites (p=0.04): 138 infants received LEV during their admission and 102 (24%) were discharged home on LEV. Twenty-one (5%) received LEV as their initial seizure treatment. The mean LEV loading dose was 48±33mg/kg and varied by institution (p=0.03). Phenytoin was used for 123 infants (28%) and the mean loading dose (23±7mg/kg) did not vary by study site (p=0.5). Seizure medications were discontinued prior to hospital discharge for 91 of the 370 survivors who were discharged to home (25%) and this varied by study site (p=0.04; range 3% to 73% for individual centers). Among the 39 infants with neonatal-onset epilepsy who survived and were discharged to home, 100% continued on medication (vs. 74% of 267 with acute symptomatic seizures, p=0.0003). Infants discharged home without medication had lower maximum PB levels than those whose anticonvulsants were continued (35±14ug/ml vs. 44±15; p=0.0003). Those with the lowest electrographic seizure burden were also most likely to have medication discontinued prior to discharge home (p<0.0001).Conclusions: Despite concerns about the neurodevelopmental consequences of long term early life exposure, PB remains the first-line treatment for neonatal seizures. There is widespread practice variability with regard to medication selection, doses prescribed, and duration of treatment. This may be due to lack of evidence, especially for newer medications like LEV. Careful study of the impact of treatment on long term outcomes after neonatal seizures is urgently needed. Financial disclosure: This work was supported by the Pediatric Epilepsy Research Foundation.