Abstracts

Rationale: A certain number of patients exist, who have documented epilepsy without interictal epileptiform discharge (IEDs) on prolonged scalp EEG recordings. This phenomenon has been associated with extratemporal epilepsies in some studies, but its clinical significance remains uncertain. Intracranial EEG (ICEEG) recordings provide the opportunity to uncover previously unrecognized IEDs. In this study we sought to determine the localization of ICEEG-recorded IEDs in relationship to the corresponding seizure focus in patients undergoing invasive surgical evaluations. Methods: Patients who never had documented IEDs on their scalp video-EEG recordings were identified from a total of 425 patients, who underwent intracranial evaluations at our institution from 1997 to 2008. In this subset of patients, all ICEEG-recorded IEDs were co-registered to pre-op MRI and postop CT images and analyzed carefully including location of maximally involved electrodes, and average number of involved electrodes. Subdural grids (platinum electrodes 3.97mm in diameter and 10mm center-to-center interelectrode distance) are used to cover the cerebral convexity, while depth electrodes (platinum 2.5 mm contacts separated by 2.5mm interelectrode distance) are used to target mesial and deeper structures, as dictated by the clinical scenario. Results: One hundred IEDs populations were identified in 36/425 (8.5%) patients, consisting of 18 temporal lobe (TLE), 11 frontal lobe, 4 parieto-occipital, and 1 perirolandic epilepsies (total of 16 extratemporal epilepsies, XTLE), and 2 patients with non-localized seizure onset. In this group of 425 ICEEG patients, scalp IEDs were absent in 18/162 (11%) of TLE and 6.1% (16/263) of XTLE. Most IEDs were recorded from lateral convexity (49% of patients), basal surface (23%), hippocampus (15%), interhemispheric region (12%). The most frequent population arose from the hippocampal region (15%). Average number of involved subdural electrodes was 3.02 (surface area <1 cm2), and depth electrodes 4.3 (volume ~4.6 cm2). Forty percent of identified IED populations originated from an area overlapping the seizure onset zone, especially within the hippocampus and to a lesser degree the lateral convexity. Conclusions: In this study sample of patients with intractable focal epilepsy undergoing ICEEG absence of scalp IEDs was more frequent in TLE (in contrast to previous reports examining this phenomenon during noninvasive video-EEG). Additionally, this study suggests that the irritative zone responsible for “invisible” IEDs is likely to overlap the seizure onset zone in up to 40% of cases, particularly when seizures arise from the hippocampus. ICEEG-recorded IEDs resulted from the synchronized activation of approximate 1 cm2 surface area of cortex, concordant to prior study. The presence of a well-circumscribed generator producing only restricted discharges and/or the inherent inability of scalp EEG to capture discharges from inaccessible/remote areas rendered IEDs invisible prior to ICEEG recordings in 8.5% of patients in this study.