Affiliations: INSERM U1105, GRAMFC, University of Picardie Jules Verne, Faculty of Medicine, SFR CAP Santé, Amiens, France | INSERM U1105, EFSN Pediatrics, CHU Amiens, Amiens, France
Note: [] Corresponding author: Emilie Bourel-Ponchel, Inserm U1105, GRAMFC, University of Picardie Jules Verne, CHU Amiens, SFR CAP Santé, 3 rue des Louvels, 80036 Amiens, France. Tel.: +33 322 66 8741; Fax: +33 322 66 8622; E-mail: [email protected].
Abstract: Surgery is increasingly proposed as a treatment for children with medically refractory partial epilepsy. The outcome after surgery depends on complete resection of the epileptogenic zone correlates. Despite multimodal preoperative assessment, 40–60% of children require intracranial electroencephalography (EEG) recording. Standard EEG can be easily performed with 20 to 30 electrodes in children of all ages. Increasing the number of electrodes (high-density [HD] EEG) significantly improves the spatial resolution of EEG and allows source localization of the epileptogenic focus. In adults, this technique has been validated for source localization of the epileptic focus and contributes to a better understanding of the neural networks involved in epilepsy. In this review, we discuss the pediatric specificities of HD EEG and source localization in order to define the value of this technique in the preoperative assessment of children with refractory partial epilepsy. Source localization using HD EEG requires several steps. The first step, elaboration of the head model, must take into account the heterogeneity of skull bone thickness, persistence of the fontanelles in younger infants, and the unknown conductivities of anatomical structures in children. The second step consists of defining the number of electrodes, which depends on the age and cortical gyration of the brain. Finally, after extraction of meaningful information (interictal or ictal features), the third step consists of defining the specific modalities required in pediatric subjects to resolve the inverse problem to determine source localization.
