Abstracts

Rationale: Previous studies have reported variable rates of epilepsy in both childhood and neonatal arterial ischemic stroke (AIS). We sought to evaluate the probability of developing epilepsy in perinatal and childhood stroke in a combined prospective/retrospective cohort of the children from The Children s Hospital, Denver. Methods: From August 2006 through January 2010, we collected data on clinical presentation, seizure onset, seizure recurrence, and development of refractory epilepsy in 104 consecutively enrolled children with AIS. Patients were classified as childhood AIS (age 29 days - 18 years inclusive), perinatal AIS (PAS- defined as symptomatic AIS presenting at birth - 28 days), or presumed perinatal stroke (PPAIS- defined as chronic presentation of a focal deficit corresponding to a chronic/remote infarct on neuroimaging). Early seizure after stroke was defined as within one week after stroke diagnosis. Epilepsy was defined as two or more seizures separated by 24 hours occurring greater than one week after stroke diagnosis. Kaplan-Meier survival analysis was performed to compare time to onset of epilepsy in those with AIS versus PAS/PPAIS. Results: Of the 104 patients with AIS, 52 met criteria for childhood AIS (50%) and 52 for PAS/PPAIS. Median follow-up time was 16.7 months (IQR: 6.9 mo - 41.6 mo) in childhood AIS and 28.5 months (IQR: 10.0 mo - 52.2 mo) months in PAS/PPAIS. A total of 55 patients (53% of the cohort) had seizures (20 childhood AIS, 35 PAS/PPAIS) and only 18.3% of the entire cohort developed epilepsy (11 childhood AIS and 8 PAS/PPAIS). Two childhood AIS patients had recurrent AIS. Of note, one of these patients had epilepsy, and was only counted once. Two patients (1.9%) became refractory. One was in the childhood AIS category and the other in the PAS/PPAIS group. Survival analysis revealed that the probability of epilepsy-free survival at 12 months did not differ between childhood AIS and PPAIS cases; 84% versus 82%, respectively; log rank P=0.79. Very few patients developed new-onset epilepsy beyond 12 months (1 in the AIS groups at 24 months, and 2 in the PAS/PPAIS group at 21 and 30 months). Conclusions: Risk of epilepsy is high following pediatric stroke, at approximately 20%, and does not appreciably differ between childhood AIS and PAS/PPAIS. However, seizures at diagnosis are more common in PAS/PPAIS. Larger prospective studies are needed to confirm these results and inform seizure management recommendations in perinatal AIS. To our knowledge, this is the first report of the probability of developing epilepsy in a combined perinatal and childhood AIS using survival analysis.