Introduction: The objective was to examine the association between continuing anti-seizure medications (ASM) until the time of discharge home and length of stay (LOS) among neonates with acute symptomatic seizures enrolled in the Neonatal Seizure Registry.

Methods: Prospective multicenter cohort of term neonates with seizures due to hypoxic-ischemic encephalopathy (HIE), ischemic stroke, or intracranial hemorrhage (ICH) who were discharged home. Propensity for ASM at discharge was determined using logistic regression. The final propensity model included 5-minute Apgar <6, seizure etiology, worst neonatal EEG background, days of EEG seizures, and abnormal discharge examination. LOS was log transformed due to skew and to facilitate percentage change interpretations. Log transformed values were regressed on ASM at discharge and propensity, categorized as quintiles.

Results: ASMs were prescribed at discharge for 122/195(63%). Neonates maintained on ASM were exposed to higher dose of phenobarbital (median 45mg/kg, IQR 30, 63 vs 73mg/kg, IQR 53, 108 p<0.0005) and trended toward longer stay (median 12.5, IQR 9, 22 days vs 11, IQR 7,16 days, p=0.08, Figure).

After propensity adjustment, neonates with stroke discharged home on ASM had a 64% longer hospitalization (95% CI 7%-251% longer, p=0.02) compared with those for whom ASM was discontinued after acute seizures resolved but before neonatal discharge. LOS did not differ among neonates with HIE or ICH.

Conclusions: Discontinuing ASM prior to neonatal discharge may decrease phenobarbital exposure and LOS, particularly for newborns with seizures due to stroke. Long-term studies are needed to determine whether treatment duration is associated with risk of epilepsy and neurodevelopmental disability.