Abstracts

Rationale: A pseudo-temporal ictal pattern refers to an EEG ictal pattern that is mis-localized to the temporal region when the epileptogenic zone is actually outside of the temporal lobe. They are misleading and can lead to faulty pre-surgical localization. Our aim is to analyze the EEG features of patients with pseudo temporal ictal patterns and differentiate them from patients with true temporal lobe epilepsy if at all possible. Methods: We retrospectively identified 10 patients (37 szs) with pseudo-temporal (PT) ictal patterns chosen based on their presurgical EEG reports of temporal ictal patterns, and preliminary visual confirmation by one of us (SE). They included 5 frontal, 3 parietal, 1 insular and 1 occipital lobe epilepsy patients who had extratemporal surgery. They were compared with 12 patients (45 szs) with mesial temporal epilepsy due to pathologically proven hippocampal sclerosis (HS), and 11 (41 szs) with neocortical temporal epilepsy (NT) who had lateral temporal resections with preservation of mesial structures. All patients had an Engel class I outcome for a minimum of one year after surgery. Interictal discharge populations were compared amongst the 3 groups. Structured visual analysis of up to 5 seizures in each patient was performed by two blinded investigators (NS, RE) using the full capabilities of digital EEG in bipolar and referential montages for ictal onset and later significant ictal patterns with regards to frequency, location, timing, duration, and spread. Source analysis using FOCUS in BESA 5.1 was performed by one blinded investigator (RE) on the first 5-10 seconds of EEG ictal onset from the best artifact free seizure of each patient and classified as temporal or extratemporal. Results: The distribution of scalp interictal discharges overlapped amongst the 3 groups. The distribution of ictal patterns in PT group in terms of location and duration were not separable from those in true temporal groups. In PT group 9/37 Szs (6 pts) started with regional temporal patterns (4 rhythmic delta, 4 rhythmic theta, 1 rhythmic alpha), and rhythmic temporal theta was seen in 8 szs of 5 pts as a later significant pattern. All seizures in PT group spread to the contralateral side while 13 szs (6HS, 7 NT) of 4 pts in true temporal groups did not show contralateral spread (P <0.05). Source analysis did not improve the separation amongst groups. Conclusions: Pseudo-temporal ictal patterns are morphologically indistinguishable from true temporal ictal patterns by visual or source analysis. Rhythmic temporal theta at ictal onset or as later significant pattern can be a spread pattern in extra-temporal epilepsy. One possible differentiating point is that while all PT seizures showed bilateral spread, temporal ictal patterns that remained strictly unilateral without contralateral spread were only found in true temporal lobe epilepsy. The findings support the notion that once activated, whether from an intrinsic generator or by extrinsic propagation, the temporal lobe structures can generate a more or less identical electrical seizure discharge.