Abstracts

Voxel-based Morphometry in Nonlesional Epilepsies and Focal Cortical Dysplasia: Correlation with Intraoperative Ecog and Surgical Results

Abstract number : 3.471
Submission category : 9. Surgery / 9C. All Ages
Year : 2024
Submission ID : 20
Source : www.aesnet.org
Presentation date : 12/9/2024 12:00:00 AM
Published date :

Authors :
Presenting Author: William Martins, MD, PhD – Hospital São Lucas da PUCRS

Eliseu Paglioli, MD, PhD – Hospital São Lucas da PUCRS
Ricardo Paganin, MD, PhD – Hospital São Lucas da PUCRS
Rafael Paglioli, MD – Hospital São Lucas da PUCRS
Ricardo Soder, MD, PhD – Hospital São Lucas da PUCRS
Thomas Frigeri, MD, PhD – Hospital São Lucas da PUCRS
Natalie da Silveira Donida, MD – Hospital São Lucas da PUCRS
Bibiana Liberman Thomé, MD – Hospital São Lucas da PUCRS
Tais Michele Werle, MD – Hospital São Lucas da PUCRS
Fernanda Schuh Martins, MD – Hospital São Lucas da PUCRS
Andre Palmini, MD, PhD – Hospital São Lucas da PUCRS

Rationale: Epilepsy surgery is still the only hope of achieving seizure freedom for a significant number of patients with epilepsy. However, a visible MRI epileptogenic lesion, whose complete resection would be the best predictor of surgical success, is only found in 60-70 % of patients undergoing epilepsy surgery evaluation. Therefore, there is urgent need for noninvasive methods that could unveil small lesions that passes unnoticed on structural MRI. Voxel based morphometry (VBM) is an emerging method that may help identify and determine the extent of cortical abnormalities unseen in conventional imaging. Morphometric analysis program (MAP) is an automated VBM-based tool for evaluating nonlesional MRI that may aid in diagnosis of subtle or occult lesions.

Methods: : In this single center retrospective study, all patients submitted to epilepsy surgery between 2006 and 2020 at the Porto Alegre Epilepsy Surgery Program were screened and 141 entered the study. They had between 2 and 70 years of age and MRI was either normal or positive for FCD. Concordance of the extent of VBM-MAP+ lesions with conventional MRI, FDG- PET and electrocorticography (ECoG) was evaluated, and completeness of resection of such lesions was correlated with surgical outcome using Engel classification.


Results: Of the 141 patients included, 62 had a nonlesional MRI and 79 had a lesional MRI. MAP was positive in 93 (65.9%) patients, 66 lesional and 27 nonlesional. Resection of MAP+ regions was associated with seizure freedom in both in the whole cohort (76.6% x 26.1%; P=0.0001) and in nonlesional patients (72.7% x 25%; p=0.02). In addition, resection of ECoG/MAP concordant regions led to a higher rate of seizure-freedom (75.8% x 16.6%, p=0.001). For non-lesional patients, MAP showed a sensitivity of 40% (CI 95% 19-64) and a specificity of 93% (CI 95% 80-98), with accuracy of 76% (CI 95% 63-86) for correctly localizing the epileptogenic zone. A novel pattern of MAP positivity in the bottom of sulcus was found (Boomerang sign) in 31% of MAP+ patients, corresponding to FCD type II in 72.7% of cases.


Conclusions: VBM-MAP provides useful information in planning epilepsy surgery even in the absence of invasive extra-operative investigations. Combining a strong electroclinical hypothesis with ECoG guided resection of MAP+ regions may be a feasible strategy in low resource settings. Novel methods of non-invasive evaluation are warranted for expanding accessibility to epilepsy surgery in low resource countries.

Funding: “This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nivel Superior - Brasil (CAPES) – Finance Code 001”


Surgery