Abstracts

INTRACRANIAL EEG MONITORING IN PATIENTS WITH MEDICALLY INTRACTABLE EPILEPSY

Abstract number : 2.463
Submission category :
Year : 2003
Submission ID : 2256
Source : www.aesnet.org
Presentation date : 12/6/2003 12:00:00 AM
Published date : Dec 1, 2003, 06:00 AM

Authors :
Margarita Pondal, David C. Diosy, Samuel Wiebe Department of Neurology, Hospital Severo Ochoa, Madrid, Madrid, Spain; Clinical Neurological Sciences, London Health Science Centre, London, ON, Canada; Clinical Neurological Sciences, London Health Science C

To analyse the contribution of intracranial EEG monitoring (IEM) to epilepsy surgery decision-making in patients with intractable localisation-related epilepsy.
Review of medical records for all patients with chronic IEM at the London Health Sciences Centre between 1993 and 1999. Data included demographics, preoperative seizure frequency, antiepileptic drug (AED) use, etiology, MRI findings, scalp EEG data, type and location of intracranial electrodes, duration of monitoring, locations of seizures and spikes captured during intracranial monitoring, offer of surgery, and complications. Patients were grouped according to increasing uncertainty of seizure localisation: 1 [ndash] focal MRI lesion, focal scalp ictal and interictal EEG; 2 [ndash] focal MRI lesion, focal ictal and multifocal interictal scalp EEG; 3 [ndash] focal MRI lesion, non-localising scalp EEG; 4 [ndash] normal or multifocal MRI, focal ictal EEG; 5 [ndash] normal or multifocal MRI, EEG non-localising.
171 patients identified, 101 male (59%), 70 female (41%), mean age 28.5 years (range 3 [ndash] 61). Etiologies included MTS (34), tumour (25), migration disorder (17), stroke (3), trauma (13), infection (12), cerebral palsy (5), cavernoma (2), AVM (3), meningioangiomatosis (4), Rasmussen encephalitis (2), and unknown (51). Patients were taking an average of 2 AEDs (range 1-5). Monthly seizure frequencies (median): simple-partial 20, complex-partial 10, generalised tonic-clonic 4. An average of 7.3 electrodes (range 1-14) were inserted per patient. Most were strips, 25 were grids, 4 depth. MRI revealed a focal lesion in 115 patients, multifocal in 12, hemispheric/diffuse lesions in 3, and no lesion in 40. Nine patients were not analysed due to multiple surgeries (7), hypothalamic hamartoma (1), and technical problems with subdural electrodes(1). Group 1 (n=14) [ndash] in 12 patients IEM was congruent with scalp EEG and MRI, IEM added information in 2 cases; Group 2 (n=13) [ndash] in 9 IEM was congruent for seizures, in 4 patients IEM was incongruent with scalp EEG or MRI; Group 3 (n=71) [ndash] IEM was congruent with scalp EEG in 27, supplemented MRI data in 11, and was incongruent with either in 11; Group 4 (n=10) [ndash] IEM was congruent with scalp in only 4, congruent with MRI in 4, and congruent with neither in 4; Group 5 (n= 54) [ndash] IEM was congruent with scalp EEG in 18, all offered surgery. 8 of 10 patients with incongruent saclp EEG and IEM and multifocal MRI abnormalities were offered surgery, and half of patients with normal MRI and incongruent scalp EEG and IEM and were offered surgery.
EIM provided a gradation in contribution of information. In simpler cases EIM data were more congruous with scalp EEG and MRI data and provided little addional information, whereas in more complex cases EIM data were more often incongruous and were more instrumental in surgical decision-making.