Abstracts

Rationale: There is no consensus on optimal management of newborns with seizures, a population at high risk for subsequent epilepsy and neurodevelopmental disorders. The underlying etiology of seizures is known to be associated with outcomes following neonatal seizures and one of factors determining treatment duration. This study aimed to evaluate the association between seizures etiology, discharge status, and postdischarge anti-epileptic drug (AED) treatment in newborns with seizures. Methods: We performed a retrospective observational study of newborns using Truven Health MarketScan^® commercial claims data collected throughout the United States. We built a newborn cohort born and hospitalized in 2013. Newborns with seizures were identified if they had ICD-9-CM codes of birth (V3X.XX) and convulsions in newborn (779.0). We examined mortality during hospitalization, disposition status (i.e., discharged to home and transferred to other care facilities), and AEDs prescribed at or after discharge. We also measured differences in disposition status and AED prescriptions by gestational age and etiology of seizures including perinatal morbidity (i.e., intrauterine hypoxia and birth asphyxia, or perinatal intracranial hemorrhage), metabolic disturbances, brain malformation, hydrocephalus, stroke or cardiovascular disease (CVD), central nervous system (CNS) infection, and congenital infection.  Results: Out of 219231 newborns (male = 113961, 52.0%; urban residence = 191083, 87.2%), 250 neonatal seizure patients were identified (male = 146, 58.4%; urban residence = 218, 87.2%). Fifty-seven (22.8%) were preterm babies. Most patients (n = 208, 83.2%) had at least one seizure etiology. Perinatal morbidity (n = 185, 74.0%) was most frequent, followed by metabolic disturbances (n = 60, 24.0%), stroke or CVD (n = 33, 13.2%), brain malformation (n = 17, 6.8%), hydrocephalus (n = 5, 2.0%), congenital infection (n = 5, 2.0%), and CNS infection (n = 3, 1.2%). A median length of hospital stay was 7 days (interquartile range, IQR = 3.0-14.3 days). One hundred forty-seven (58.8%) were discharged to home. Four were deceased during hospitalization. Postdischarge AEDs were prescribed to 61 patients (24.4%, 61/246 survivors). A median treatment duration was 80 days (IQR = 30-197 days). Forty-three patients (70.5%) had one AED. The AEDs included Phenobarbital (n = 37, 60.7%), Levetiracetam (n = 24, 39.3%), Benzodiazepines (n = 10, 16.4%), Prednisolone (n = 10, 16.4%), Phenytoin (n = 3, 4.9%), Topiramate (n = 3, 4.9), Vigabatrin (n = 1, 1.6%), and Oxcarbazepine (n = 1, 1.6%). Disposition status and postdischarge AED treatment were not associated with gestational age or seizures etiology (P > 0.05). No difference was found in AED treatment according to disposition status (P > 0.05). Conclusions: More than 40% of newborns with seizures were discharged to other facilities than home for further care, whereas less than one-fourth of this cohort received postdischarge AEDs. Seizures etiology was not associated with postdischarge AED treatment in this cohort. This finding showed that AEDs were used for short-term period, only during hospitalization, for neonatal seizures. Funding: Children’s Healthcare of Atlanta research grant funded by the Goizueta Foundation.