Abstracts

In recent years, converging evidence seems to suggest that Vagus Nerve Stimulation (VNS) can provide effective seizure reduction in patients with intractable epilepsy who have been considered unsuitable for resective surgery. Clinical response has been reported to follow a gradual trend, with best results obtained with longer treatment duration. Less agreement is present on the possible correlation between the interictal EEG changes and seizure reduction. We report on a series of 34 children treated at two centres, in Italy (IT) and United Kingdom (UK), with clinical and EEG follow-up of at least 24 months. We analysed data from 3-day ambulatory (UK) and prolonged (3-hr ) outpatient EEG recording (IT) after implant prior to switching the stimulator on, and after 12 and 24 months. In both groups data was acquired from 19 electrodes placed according the international 10-20 system on digital EEG systems sampled at 256 Hz. Quantitative assessment of interictal epileptiform discharges was performed using a clustering method implemented on a commercial software (Persyst, Prescott, AZ, USA) and manually verified by one of three certified clinical neurophysiologists. Spike density (SD) was computed for each patient. Topography of individual clusters for each subject was analysed using a topographic analysis of variance. Measurement of duration of the spike and slow-wave component of each discharge was performed. Topography of interictal epileptiform discharges remained stable over time in all subjects. A greater then 50% reduction in SD was seen in only 3 patients at 12 months, whereas 15/34 patients showed a significant increase in SD (p[lt].001), with a mean reduction in seizure frequency of 62%. At 24 months only 4 subjects had a [gt]50% reduction in SD, whereas in 12/34 patients an increase in SD with respect to baseline measurement was still present (p[lt].01) with a mean reduction in seizure frequency of 72%. Between 12 and 24 months 12 patients had a significant (p[lt].01) decrease in SD. Quantitative analysis of slow-wave activity showed a significant increase in its duration (p[lt].01), which paralleled the reduction in seizure frequency. Our data suggests that in our subset of patients, seizure reduction and spike density seem only weakly correlated. The significant increase in spike density at 12 months, could be explained with the hypothesis that interictal epileptiform discharges are associated with increased inhibitory phenomena and activity dependent changes in epileptogenic cortex which decreases the likelihood of clinical seizures. In the same line we could interpret the increased duration of slow-wave activity at 12 and 24 months.