Abstracts

Rationale: Epileptic seizures occur in circadian patterns. Only limited information on the circadian patterns, day and nighttime occurrence, and occurrence of ES out of wakefulness and sleep is available. No information of evolution of ES timing peaks in different age groups is available.Methods: 2021 consecutive patients (0-21 years) undergoing V-EEG monitoring over a 10 year period were reviewed for ES. We identified 51 pediatric patients with ES. V-EEG recordings were reviewed for spasm occurrence during the day (6AM-6PM) or night (6PM-6AM), during wakefulness or sleep, and in 3-hour time blocks throughout 24 hours. Epileptic spasms were defined as abrupt movement characterized either by subtle (facial), flexor, extensor or mixed motor movements occurring in clusters. Patients were separately analyzed for patients over and under three years of age. MRIs were reviewed for the presence of lesions. Statistical analysis was conducted using binomial testing. Results: 219 clusters of epileptic spasms were analyzed in 51 patients (15 girls; mean age 2.15 +/- 2.22 yrs). 39 patients (76%) were under three years old and 12 patients (24%) were over three. 37 patients had 157 spasms with generalized EEG seizures and 14 patients had 62 ES with focal EEG seizures. 32 out of 43 patients (74%) had lesional MRIs. ES occurred predominantly during wakefulness (p<0.001) and during daytime (p<0.001), with peaks at 6-9AM, 9AM-12PM, and 3-6PM during wakefulness (p<0.001) and 9PM-12AM during wakefulness (p<0.01). ES associated with generalized EEG seizure patterns occurred primarily during the day (p<0.001) and during wakefulness (p<0.001), especially between the hours of 3-6PM (p<0.001). ES with focal EEG patterns demonstrated no significant day/night or sleep/wakefulness pattern but exhibited diurnal peaks between 9-12PM and 3-6PM (p<0.01). When stratifying patients by age groups (over and under 3 years), only patients over three years had a peak between 6-9AM (p<0.05), which was not seen in the under-three population. In the under-three population, patients with MRI lesions had ES most frequently between 9AM-12PM (p<0.05) and 3-6PM (p<0.01). In the over-three population, patients with MRI lesions had ES most frequently between 3-6PM (p<0.05) but not between 9AM-12PM. Conclusions: ES occur more frequently in the waking state and during the day time, with prominent early morning and afternoon peaks. After stratification by age, early morning peaks shifted to an earlier time and afternoon peaks disappeared. This was most prominent in the population of patients with lesional MRI. Circadian seizure patterns can assist in the identification of ES-sub-types. We speculate that the circadian pattern of epileptic spasms shifts with age, from a two peak pattern under-three to a single early morning peak, similar to previously described circadian patterns in tonic seizures.