Purpose: Poor seizure outcomes after epilepsy surgery often reflect an incorrect localization of the epileptic sources by standard intracranial EEG interpretation because of limited electrode coverage of the epileptogenic zone. This study investigates whether, in such conditions, source modeling is able to provide more accurate source localization than the standard clinical method that can be used prospectively to improve surgical resection planning.
Methods: Suboptimal epileptogenic zone sampling is simulated by subsets of the electrode configuration used to record intracranial EEG in a patient rendered seizure free after surgery. sLORETA and the clinical method solutions are applied to interictal spikes sampled with these electrode subsets and are compared for colocalization with the resection volume and displacement due to electrode downsampling.
Results: sLORETA provides often congruent and at times more accurate source localization when compared with the standard clinical method. However, with electrode downsampling, individual sLORETA solution locations can vary considerably and shift consistently toward the remaining electrodes.
Conclusions: sLORETA application can improve source localization based on the clinical method but does not reliably compensate for suboptimal electrode placement. Incorporating sLORETA solutions based on intracranial EEG in surgical planning should proceed cautiously in cases where electrode repositioning is planned on clinical grounds.
Departments of *Neurology and
†Neurosurgery, University of Texas Health Science Center, San Antonio, Texas, U.S.A.; and
‡San Antonio VA Epilepsy Center of Excellence, Audie L. Murphy Medical Center, San Antonio, Texas, U.S.A.
Address correspondence and reprint requests to Octavian V. Lie, MD, PhD, 8300 Floyd Curl Drive, MSC: 7883, San Antonio, TX 78229-3900, U.S.A.; e-mail: email@example.com.