Abstracts

Rationale: Patients with mesial temporal lobe epilepsy (MTLE) resistant to AEDs may be candidates for surgery. While scalp video-EEG often identifies the side and site of seizure onset, some patients need more investigation to distinguish the epileptogenic zone from targets of propagated activity starting in mesial areas. We have developed experience using foramen ovale electrodes (FOs) as a minimally invasive diagnostic approach in these patients. FOs provide the opportunity to record close to the mesial temporal structures, avoiding the filtering and dispersing effects of the skull. We analyzed features, prevalence and significance of patterns of ictal propagation seen with FOs. Methods: We studied 147 seizures from 17 MTLE patients that had evaluation with FOs from 2005-2010. For each seizure we established time of 1st EEG change on FOs and on scalp, time of 1st clinical manifestation, pattern of electrographic propagation and EEG features. To analyze propagation delay (FO-scalp or FO-FO), individual seizure values for each patient were averaged to calculate a per-patient mean delay. Results: Patients had a mean of 8.6 seizures(Range2-29).In all seizures the 1st EEG change occurred in a unilateral FO. 3 major propagation patterns were identified: Type I-Early propagation to contralateral side (FO-contralateral FO-scalp)(13.6% of seizures); Type II-Late propagation to contralateral side (FO-ipsi /contralateral scalp-contralateral FO)(33.3% of seizures); and Type III- No propagation to contralateral side (FO-ipsilateral scalp or FO only)(53.1% of seizures). 27.2 % of seizures had no clinical correlate. All seizures without clinical correlate were Type III (51.3% of the events of this type). Most seizures had a scalp signature characterized by a recognized ictal pattern or slowing, however, 4.8% of the seizures were only seen on FOs. When there was scalp signature the mean delay of initial FO-scalp propagation per-patient was 13.4s (Range2.8-24.9). 4.1% of seizures with scalp signature originated from the contralateral FO. When there was propagation from FO-scalp-FO (Type II), the mean delay per-patient was 22.8s (Range4-40.5). When direct FO-FO propagation occurred (Type I), the mean delay per patient was 8.3s (Range3-28). 5 patients had predominantly Type I seizures, 8 had >50% Type II, and 4 had >50% Type III. Ictal EEG patterns on FOs consisted of a buzz of beta activity in 54% of seizures, and rhythmic spike-waves in the remainder. Conclusions: Three distinct patterns of ictal propagation can be identified using combined scalp-FOs. Most patients manifest predominance of one pattern, although many have a minority of seizures that have other routes of propagation. FO-FO propagation occurs rapidly. FO recordings sometimes clarify the lateralization, and frequently define the localization of ictal onset. In some cases FO recordings reveal that seizures arise from mesial areas contralateral to the scalp signature, a phenomenon we have termed ping-pong seizures . Seizures arising from FOs frequently have no obvious behavioral correlate, however subtle effects on cognition cannot be ruled out.