Abstracts

Rationale: According to the existing literature, absence of a lesion in the magnetic resonance imaging (MRI) is associated with poorer prognosis after resective epilepsy surgery. Delineation of the epileptic zone by invasive intracranial recording is required in such patients. Invasive monitoring has potential complications and is perceived to be relatively more difficult due to lack of a clear radiographic target to implant the recording electrodes in patients with non-lesional epilepsy. Electroencephalogram (EEG) dipole source analysis is based on noninvasive scalp EEG data and allows estimation of the underlying generator of a given EEG potential. Its clinical utility has not been well studied in non lesional epilepsy. Here we compare the ability of dipole source analysis using scalp EEG with intracranial recording to identify the epileptic focus in these patients. Methods: All patients with intractable epilepsy and no detectable lesion on MRI, who underwent resection of the epileptogenic zone after intracranial recording, were included in the study. Data regarding patient characteristics, scalp EEG findings, brain MRI, intracranial electroencephalogram recordings, surgical details, pathology findings and the results of dipole source analysis were collected and analyzed. Non-invasive scalp EEG's were analyzed using the Brain Electrical Source Analysis (BESA) software, developed by MEGIS Software. We performed the dipole source analysis on averaged ictal waveforms and co-registered the EEG source dipole on the patient's brain MRI. We compared the effectiveness of EEG dipole source analysis in localizing the seizure focus with intracranial recordings. Results: Twelve patients fulfilling the inclusion criteria were identified. Three patients, who failed the resective surgery and one with incomplete follow up were excluded. Eight patients (66%) who underwent resection of the epileptogenic zone after intracranial recording and became completely seizure free were included in the final analysis. We correctly lateralized the seizures in 7 (87%) patients using the dipole source analysis. The dipole source localization results matched the intracranial recordings in exact localization of the epileptogenic zone in 4(50%) patients. In remaining three patients, the dipole source localization using BESA was able to correctly localize the seizure onset zone to different lobar regions of the brain. Conclusions: Dipole source analysis using scalp EEG was able to correctly lateralize the seizure onset zone in 87% of the patients with non-lesional epilepsy. Furthermore, it precisely localized the epileptogenic zone in at least half of these patients. Estimating the EEG dipole source before invasive electrode evaluation will potentially help in guiding the clinicians in placement of intracranial electrodes and also will limit the extensiveness of the coverage, thereby potentially decreasing the complication rate, as well as financial burden.