HR-EEG Source localization - 01/08/18

Doi : 10.1016/j.neucli.2018.06.027

Martine Gavaret
Service de neurophysiologie clinique, centre hospitalier Sainte-Anne, 1, rue Cabanis, 75014 Paris, France

Résumé

High-resolution EEG (HR-EEG) allows the recording of electric brain activities with excellent temporal resolution. Data have to be recorded with high temporal resolution (sampling rate) and high spatial resolution (number of channels). Data analyses are based on several steps including interictal and/or ictal selections, head model elaboration and use of several algorithms in order to solve the inverse problem. These tools now represent real methods of Electric Source Imaging. HR-EEG constitutes non-invasive and complementary examination with Magnetoencephalography (MEG), characterized by distinct sensitivities according to the location and orientation of intracerebral generators. In the presurgical assessment of drug-resistant partial epilepsies, HR-EEG can characterize and localize interictal activities and thus the irritative zone. HR-EEG often yields significant additional data that are complementary to other presurgical investigations. In some lesional epilepsy cases as epilepsies related to focal cortical dysplasia, localization of the irritative zone using HR-EEG is particularly relevant in presurgical evaluation, epileptogenic and irritative zones being characterized by an overlap. It has been demonstrated that HR-EEG more accurately estimated the epileptogenic zone in subgroups of patients who are often the most difficult cases in epilepsy surgery: frontal lobe epilepsy, negative MRI and the presence of malformation of cortical development [1]. Moreover, source localization results may help in identifying subtle focal cortical dysplasia abnormalities and increasing the diagnostic yield of conventional MRI visual analysis [3]. During presurgical evaluation of pharmacoresistant partial epilepsies, stereoelectroencephalography records interictal and ictal activities directly but is inherently limited in spatial sampling. In contrast, scalp-EEG and MEG are less sensitive but provide a global view on brain activity. Therefore, recording simultaneously these three modalities [2] should provide a better understanding of the underlying brain sources by taking advantage of the different sensitivities of the three recording techniques.