Abstracts

Persistent or recurrent seizures are seen in 20-40% of patients after surgery for intractable epilepsy. Standard non-invasive presurgical evaluation can not always provide sufficient localizing data and a second invasive EEG monitoring is often difficult due to scar formation from previous surgery. Magnetoencephalography (MEG) provides valuable information for localization of the epileptogenic zone in localization-related epilepsies. Magnetic fields are less distorted by the tissues through which they pass than electric fields. Compared to EEG, MEG can provide superior information to localize the residual epileptogenic zone.
We studied the value of MEG in the surgical evaluation of children with recurrent seizures after epilepsy surgery. Twenty three children with persistent or recurrent seizures after epilepsy surgery were evaluated by MEG in addition to standard non-invasive presurgical evaluation including interictal, ictal scalp EEG and MRI. We analyzed location and distribution of MEG spike sources, the concordance of MEG findings with ictal EEG onset and surgical outcome. Initial surgery consisted of lesionectomy in 19 patients and intracranial invasive EEG with subsequent resection in 4 patients. Pathologic diagnoses from previous surgery were: Cortical dysplasia (8), neoplastic (9), mesial temporal sclerosis (2) and other (4). Patients were classified into group 1 with unilateral MEG cluster (17 patients) and group 2 with bilateral MEG clusters or only scattered MEG spikes (6 patients). In group 1 patients, MEG clusters were located adjacent to the margin of previous resection in 16 patients and distant but in the ipsilateral hemisphere in one patient. In group 1 patients, MEG showed regional concordance with ictal EEG onset in 7 patients and hemispheric concordance in 6 patients. Ictal EEG was non lateralizing in two patients and no ictal EEG data was available in the other two. Twelve patients had subsequent surgery (11 in group 1 and 1 in group 2). five patients had invasive monitoring with cortical excision, five patients had resection guided by ECoG and intraoperative neuronavigation and two patients had functional hemispherectomy. Duration of post surgical follow up ranged from 1-40 months. Class I outcome was obtained in nine patients (75%), class II in two patients and class III in one patient. MEG precisely delineated the recurrent or remaining epileptogenic zone in most children with unsuccessful initial epilepsy surgery. Surgical resection including MEG cluster located adjacent to the margins of previous resections can achieve favorable outcome in children with recurrent or residual seizures.