Abstracts

Rationale: Consolidation of declarative memories occurs during sleep in both healthy adults and children. Focal idiopathic epileptic syndromes of childhood are characterized by frequent focal interical epileptiform discharges (IED) more abundant during non-rapid eye movement (NREM) sleep, thus potentially disrupting sleep-related memory consolidation processes. Sleep-dependent impairment in the consolidation of verbal declarative memory was shown by our group in a pilot study including 4 children aged 8-10 years (Urbain et al. 2011, Epilepsy Behav 2:380-4). Aims of the study are (1) to confirm our preliminary data on a larger set of patients, (2) to investigate a non-verbal memory task, allowing to include younger children, and (3) to correlate behavioural data with sleep EEG parameters.Methods: We investigated 15 patients and 8 control subjects aged 6 to 12 years. All subjects had performance IQ above 80. Epilepsy was diagnosed in 12 patients who presented at least one seizure. The 3 other patients had IED without seizures. Seven patients were taking anti-epileptic drugs (AED). IED were centro-temporal in 12 patients, and occipital in 3 patients. Two declarative memory tasks for which sleep-dependent maintenance of performance has been demonstrated in healthy children (Wilhelm et al. 2008) were administered: a verbal (word-pairs association) task and a non-verbal (2D object location) task. Children aged under 8 years or non-native French speakers completed the non-verbal task only. In the evening, participants learned the location of 12 pairs of objects presented on a grid or the association of 32 (or 22) word-pairs until they reached a 60% learning criterion. The next morning after a night of sleep, memory for previously learned material was tested. Memory performance was measured as the difference between the percentage of items recalled at retrieval minus the percentage of items memorized to reach the learning criteria. All the procedure occurred at the pediatric sleep unit under continuous EEG monitoring.Results: Analyses were conducted separately for each task. Analyses on sleep EEG parameters did not show significant differences between populations, apart from the presence of IED in patients. Analyses on memory performance disclosed lower overnight recall in patients than controls for both verbal and non-verbal tasks (ps <0.05), see figure. Memory performance was similar in patients taking vs. free of AED. Regression analyses showed that IED frequency during NREM sleep stage 2 and a history of epileptic seizures predicted poor performance in the non-verbal task (ps<0.05), whereas history of epileptic seizures only was related to performance in the verbal task (p<0.05).Conclusions: This study suggests impaired verbal and non-verbal memory consolidation processes in childhood focal idiopathic epilepsies. Identification of a history of seizures as a predictor of poor performance suggests that the impairment of sleep-related memory consolidation in this type of epileptic disorder is not merely related to IED during sleep.