135 Individualized High-Density Electroencephalographic Source Imaging Technique in Presurgical Workup: Contribution to Surgical Strategy Making for Intractable Epilepsy Involving Mesial Temporal Lobe Structures. Issue Volume 65:Issue CN(2018)Supplement 1 (16th August 2018)

• Record Type: Journal Article
• Authors: Feng, Rui; Hu, Jie; Wu, Jinsong; Ma, Chengxin; Lang, Liqin; Sun, Bing; Pan, Li
• Is Part Of: Neurosurgery. Volume 65:Issue CN(2018)Supplement 1
• Page Start: 92
• Page End: 92
• Publication Date: 2018-08-16
• Subjects: Nervous system -- Surgery -- Periodicals; 617.48005
• DOI: 10.1093/neuros/nyy303.135
• Languages: English
• ISSNs: 0148-396X

Abstract

Abstract: INTRODUCTION: Localization-related epilepsy frequently involves mesial temporal lobe structures (MTLS), but sometimes presurgical workup is confusing since lack of clear structural lesions or inconsistency among multiple tools. We recently improved accuracy of EEG source imaging technique (ESI) and applied it in presurgical epilepsy workup. This study evaluates its contribution to surgical strategy of epilepsy involving MTLS. METHODS: We included patients who underwent resective surgeries encompassing MTLS in this study. In addition to traditional tools, ESI was available in all cases. ESI was accomplished based on 256-channel high-density EEG and individualized finite difference method head models. Patients accepted either 1-stage or staged resective surgeries (SEEG implantation + stage-2 resective surgeries). Contributions of multiple tools to surgical strategies were evaluated. RESULTS: Twenty-five cases with Engel I + II outcome after surgeries were included. ESI (80%) and MRI (76.0%) showed higher accuracy over ictal EEG (44.0%) and FDG-PET (56.0%) when defining resective scope as epileptogenic zone ( P < .05). In 1-stage cases, ESI showed sources localized within MTLS region in 88.2% (same as MRI positive rate), while PET localized focally only in 64.7% ( P < .05). Staged cases (62.5%) showed complete concordance of ESI sources with SEEG findings, while for PET and MRI the ratio is 25% and 62.5%, respectively. ESI and PET contributed to SEEG plans in all, while MRI and ictal EEG contributed in 87.5% and 50%, respectively. In subtle lesional/MRI-negative cases, 62.5% showed subtle MRI lesions in MTLS firstly diagnosed as "negative." ESI contributed more to detection/confirmation of these lesions (75% showed sources confined within MTLS region) than PET estimates (focally localized in MTLS in 50%) ( P < .05). CONCLUSION: Noninvasive accurate ESI method described here is based on high-density EEG and individualized head model, appearing contributable to surgical planning of epilepsy surgeries involving MTLS, by indicating MTLS region epileptic sources. This feature can help decide strategy of 1-stage resective surgeries and SEEG implantation plans.