1135 Classification of Temporal Lobe Epilepsy Anatomo-clinico-electrical Hypotheses: A Stereo EEG Study
Arka Mallela,Jasmine Hect,Michael Pupi,Austin Anthony,David Fogg,Arthur Angonese,Hussam Abou-Al-Shaar,Zachary Roberts,Sharath Kumar Anand,Thandar Aung,Jorge Gonzalez-Martinez
DOI: https://doi.org/10.1227/neu.0000000000002809_1135
IF: 5.315
2024-04-01
Neurosurgery
Abstract:INTRODUCTION: Temporal lobe epilepsy (TLE) has historically been divided into mesial and lateral (neocortical) TLE, but electrophysiological, semiological, and anatomical evidence indicate that the epileptogenic zone (EZ) and areas of propagation involve both temporal and extratemporal structures. Hypothesis-driven, structured interrogation is crucial for a successful stereo EEG (SEEG) exploration and subsequent successful surgical intervention. METHODS: 59 patients (ages 19-77) with drug-resistant TLE underwent SEEG at our institution from 2020-2022. We retrospectively reviewed each patient’s multidisciplinary discussions, history, imaging, and outcomes. We proposed four types of implantation hypotheses: temporal, temporal/basal/occipital, anterior perisylvian, and perisylvian. We quantitatively validated these types against anatomic coverage and semiology using unsupervised clustering and clinically by comparing classification to rates of successful localization of the epileptogenic zone (EZ) and seizure outcomes. RESULTS: Temporal (only) hypotheses were the least frequent (13.6%), followed by temporal/basal/occipital (25.4%), anterior perisylvian (30.5%), and perisylvian (30.5%). The hypotheses were not different in age at SEEG (p = 0.32), sex (p = 0.67), and handedness (p = 0.33). Frequent semiological features by type were: temporal – autonomic features, dÉjÀ vu; temporal/basal/occipital – visual and language disturbance; anterior perisylvian – oromotor; perisylvian – sensory manifestation. Clustering was validated against semiology (RI = 0.63) and lobar coverage (RI = 0.73). Successful localization of the EZ occurred in 79.7% and did not associate with hypothesis type (p = 0.22). Hypothesis type (p = 0.30) or EZ localization (p = 1.00) did not associate with the presence of an MRI lesion. 6-month Engel 1a rate was 71% and did not vary by hypothesis type (p = 0.29). CONCLUSIONS: The anatomo-clinico-electrical hypotheses in TLE must consider both temporal and extratemporal structures and not be limited by a medial vs. lateral TLE mindset. This proposed classification of TLE hypotheses can serve as starting point for multidisciplinary patient discussion as well as a tool to compare SEEG exploration and outcomes across centers.
surgery,clinical neurology