Abstracts

Rationale: Removal of regions generating High-Frequency Oscillations (HFOs), namely ripples (R, 80-250 Hz) and fast ripples (FR, >250 Hz) recorded from the intracerebral EEG has been found to be correlated with a good surgical outcome. We evaluated whether the relationship between interictal HFOs and surgical outcome (Engel class) differs according to the type of epilepsy, in particular in temporal lobe epilepsy versus extra-temporal lobe epilepsy.Methods: Twenty-one patients with temporal lobe epilepsy and 9 patients with extra-temporal lobe epilepsy were studied. All underwent depth electrodes placement (median of 6 electrodes per patient) for intracerebral EEG recordings, surgical resection for medically intractable epilepsy, a median 23-month post-surgical follow-up and a post-surgical MRI. The removed electrode contacts were identified within the surgical cavity and the resection volume measured on the postsurgical MRI. HFOs (R, FR) were visually marked on a 5-minute slow-wave sleep interictal EEG sample that was separated by at least 4 hours from any seizure. Spikes (Sp) were marked independently of HFOs. We measured the ratio between HFO rates in removed (Rem) contacts and HFO rates in non-removed (nonRem) contacts: ratio(ev) = [ R(ev)(Rem) - R(ev)(nonRem) ] / [ R(ev)(Rem+nonRem)] , where R is the mean rate of an ev or event (R, FR). The ratio was computed for all patients and separately, for the temporal and extra-temporal lobe epilepsy groups. Similar ratios were computed for Sp rates and for the number of channels within the seizure onset zone (SOZ). We used the Wilcoxon rank sum test to compare ratios in patients with good (Engel 1&2 class) vs those with poor outcome (Engel 3&4 class). A correlation analysis was performed between the surgical outcome and the volume of resection. The level of significance was set at 0.05. Results: In the entire patient sample, the ratio was significantly higher for R in patients with Engel1&2 compared to patients with Engel3&4 (P=0.02). In the temporal lobe epilepsy group, the ratios were significantly higher for FR (P=0.049) and R (P=0.018) in patients with Engel1&2 compared to patients with Engel3&4 (Figure 1). As for analysis of the ratios in the extratemporal lobe epilepsy group, none of the ratios were significant. When analyzing spikes, none of the ratios were significant. When analyzing only the number of channels within the SOZ, the ratios were not significantly different. The volume of resection was not correlated with the outcome.Conclusions: In pharmacoresistant temporal lobe epilepsy, removal of R- and FR-generating areas may lead to improved surgical outcomes. Less consistent findings emerge from pharmacoresistant extratemporal lobe epilepsy, but these may be related to sample size limitations in our study. Future studies are required to elucidate the clinical utility of removing HFO-generating regions in extratemporal lobe epilepsy.