Abstracts

Rationale: Acute symptomatic seizures have been well characterized for neonatal cohorts experiencing the most common acute etiologies of brain injury (hypoxic ischemic encephalopathy (HIE), ischemic stroke, intracranial hemorrhage, intracranial infection), but they have been less defined in other high risk cohorts, such as those with severe cardiopulmonary disease (CPD) including congenital heart disease (CHD), congenital diaphragmatic hernia (CDH), or exposure to extracorporeal membrane oxygenation (ECMO). We compared the clinical presentation, EEG monitoring strategy, seizure characteristics, management, and outcomes between neonates with and without CPD to determine if neonates with CPD are at less risk for seizure and poor outcomes. Methods: The Neonatal Seizure Registry (NSR-I) is a multi-center, prospective, consecutive cohort of neonates with clinical or EEG-confirmed seizures. We compared neonates with acute symptomatic etiologies of seizures with and without CPD. Frequencies and percentages were used to summarize categorical variables, and median and interquartile ranges (IQR) were used for continuous variables due to the skewness in distribution. To compare our cohorts, the Pearson's chi-squared test were used to assess binary and nominal categorical variables, ordered logistic regression for ordinal categorical variables, and Wilcoxon rank sum test for continuous variables. Results: 83 neonates with CPD (73 CHD, 5 CDH, 26 ECMO; 20 with multiple diagnoses) and 271 neonates with acute symptomatic etiologies (144 HIE, 55 ischemic stroke, 44 hemorrhage, 28 infection) without CPD were included (Table). Neonates with CPD were less likely to present with clinical seizures (57% vs 69%, p=0.03) and more likely to have EEG-only seizures (40% vs 21%, p<0.001). Neonates with CPD started EEG monitoring later (141 vs 29 hours after birth, p<0.001), had their first EEG seizure later (174 vs 33 hours after birth, p<0.001), and had higher seizure burden (p=0.05). The initial antiseizure medication (ASM) loaded was phenobarbital (PHB) in both cohorts, but the PHB total loading dose was higher in the CPD cohort (40 [20, 50] vs 30 [20, 40] mg/kg, p=0.02) with highest PHB levels for both cohorts of median 45 (p=0.75). Neonates with CPD were less likely to have an incomplete response to initial ASM loading dose (64% vs 68%, p=0.05) and more likely to be discharged home on ASM (82% vs 68%, p=0.01). In-hospital mortality was higher in the CPD cohort (30% vs 14%, p=0.01), and death occurred at a later age (23 vs 5 days, p<0.001). Conclusions: While seizures occur later in the clinical course of neonates with CPD, these children were less likely to present with clinical seizures, more likely to have EEG-only seizures, and had a higher seizure burden than neonates with other acute symptomatic etiologies. These data support guidelines that recommend EEG monitoring in neonates with CPD. While PHB remains the mainstay of treatment for neonates with CPD, seizure refractoriness is high. Future studies should focus on ASM efficacy and tolerability in neonates with CPD with the goal of optimizing seizure management. Funding: PCORI (1507-31187)Pediatric Epilepsy Research Foundation