Serial changes in the paroxysmal discharges in rolandic epilepsy may predict seizure recurrence: A retrospective 3-year follow-up study.

OBJECTIVE: The aim of this study was to assess the electrographic criteria related to seizure recurrence and determine age-related seizure recurrence in children with rolandic epilepsy under long-term follow-up.

METHODS: We retrospectively analyzed the data belonging to 109 patients with rolandic epilepsy with sufficient information regarding disease course and follow-up duration longer than 3years. Patients were divided into two categories: Group A (n: 75), comprised of "patients having fewer than four seizures", and Group B (n: 34), the "recurrence group comprised of patients having more than four seizures in the first three months". The number of spikes per minute during both wakefulness and sleep, the localization of spikes other than centrotemporal region, and the duration of spike-wave activity were evaluated longitudinally, with repeated electroencephalogram (EEG) recordings every 6months.

RESULTS: The appearance of rolandic spikes in awake EEGs tended to be more prevalent in Group B than Group A. In Group B, spike rates significantly increased in the 12 and 18months after onset whereas spike rates increased significantly only 6months after onset in Group A. Seizure recurrence is mostly seen at 6-8years, and improvement becomes evident by age 12. The mean number of paroxysmal rolandic discharges during sleep was significantly higher in the younger age groups (3-5, 6-8), and the mean number of spikes per minute significantly decreased at ages 9-11 and over 12.

CONCLUSION: Our study demonstrates that extended periods of high frequency of paroxysmal discharges, initial frontal EEG focus, and persistence of awake interictal abnormalities are highly effective in predicting seizure recurrence in patients with rolandic epilepsy (RE).