Clinical Utility of Automated Electric Source Imaging in Presurgical Evaluation

Overview

Electrical source imaging is part of the presurgical evaluation of patients with drug-resistant focal epilepsy. The software packages that will be used in this study have Declaration of Conformity within the European Economic Area (CE mark) for this specific medical use. In spite of being part of the clinical standard, the evidence for the accuracy and clinical utility of these methods are derived from several smaller-scale and retrospective studies. The PROMAESIS study will provide solid evidence of the accuracy and clinical utility of automated ESI.

One-third of patients with epilepsy have seizures resistant to pharmacotherapy. There are many approaches developed to control these seizures, yet epilepsy surgery is still the most common method. The crucial part of epilepsy surgery is to assess the epileptic zone in order to render patients seizure free. However, accurate localization of the epileptic zone is often challenging due to the multimodal approach. This contains semiology, EEG (obtained during long term video-EEG monitoring), magnetic resonance Imaging (MRI), in addition to certain cases, positron emission computed tomography (PET), and single photon emission computed tomography (SPECT) and magnetoencephalography (MEG). At present, in one of the third patients, seizure remains after epilepsy surgery. Therefore a new preoperative method should be improved to assess the epileptic zone. Automated ESI is a post-processing novel method that estimates the location in the brain of the source of the recorded EEG signals. The objective of this study is: 1. To determine the accuracy of ESI in localizing the epileptic focus. 2. To determine the clinical utility of ESI on clinical decision making Methods: Study design: a prospective diagnostic study in line with the Standards for Reporting Diagnostic Accuracy Studies (STARD). EEG was recorded using the International Federation of Clinical Neurophysiology (IFCN) electrode array of 25 electrodes including six electrodes in the Inferior temporal chain (F9/10, T9/10, and P9/10) in addition to the 19 electrodes of the 10-20 system. Electrode impedance was kept below 5 kilo-ohm. EEG was recorded with a sampling frequency of 256 Hz. The investigators will make a video on how to place the electrodes, to make sure all centers follow a standardized method. Multidisciplinary epilepsy teams classify that in the first step you should keep the multidisciplinary team blinded to the source imaging data, and make the implantation plan without the ESI results. Then show the ESI, adjust the plan and note the changes. The multidisciplinary teams in different centers will classify seizures into "types." A seizure type is a group of seizures that have stereotypical semiology and ictal EEG. Maximum 3 seizures are registered for per type. After Long Term Monitoring, they provide Epilog and Brain Electrical Source Analysis (BESA) teams with datasets including MRI and Long Term Monitoring recording for each patient. For uploading datasets, multicenter teams will use the platform developed by Epilog. Afterward, Epilog and BESA teams (blinded to patient information) will run the automated source imaging-both for interictal epileptiform discharge (EDs) and ictal signs. Both software packages have CE mark for ESI. Finally, the multidisciplinary teams take decisions in two steps: I. Considering all data, except ESI II. Adding ESI to all other data. At each step, the decisions are classified into one of the following categories: 1. Stop (operation not recommended) 2. Implantation of intracranial electrodes 3. Operation In addition, the changes are classified into one of the following categories: 1. No change - but concordant with the decision. 2. No change - but discordant with decision. 3. Change from stop to implantation. 4. Change from implantation to stop. 5. Change in implantation plan: implantation of additional sites (besides the ones planned in step-1). 6. Change from implantation of operation. 7. Change from operation to implantation. 8. Other (specify in free text). If changes are not related to all analyses results, it will be noted, which method(s) triggered the change (SLORETA-interictal / ictal; Equivalent Current Dipole - interictal/ictal; CLARA interictal/ictal). At one-year follow-up, the changes are categorized as useful or not useful. A change is defined useful as follows: (a) change from stop to Intracranial Recording: the Intracranial Recording localized the source; (b) change in implantation strategy: the electrode(s) implanted based on the source imaging identified the source; (c) change from implantation to operation: the patient became seizure-free.