Voxel-Based Morphometric MRI Postprocessing in Nonlesional Pediatric Epilepsy Patients Using Pediatric Normal Databases.

Wei Wang,Yicong Lin,Shan Wang,Stephen Jones,Richard Prayson,Ahsan N V Moosa,Aaron McBride,Jorge Gonzalez-Martinez,William Bingaman,Imad Najm,Andreas Alexopoulos,Irene Wang
DOI: https://doi.org/10.1111/ene.13916
2019-01-01
European Journal of Neurology
Abstract:Background and purpose Pre-surgical evaluation of pediatric patients with drug-resistant focal epilepsy and negative (non-lesional) magnetic resonance imaging (MRI) is particularly challenging. Focal cortical dysplasia (FCD), a frequent pathological substrate in such setting, may be subtle on MRI and evade detection. The aim of this study was to use voxel-based MRI postprocessing to improve the detection of subtle FCD in pediatric surgical candidates. Methods A consecutive cohort of pediatric patients undergoing pre-surgical evaluation with a negative MRI by visual analysis was included. MRI postprocessing was performed using a voxel-based morphometric analysis program (MAP) on T1-weighted volumetric MRI, with comparison to an age-specific normal pediatric database. The pertinence of MAP-positive areas was confirmed by surgical outcome and pathology. Results A total of 78 patients were included. Forty-four patients (56%) had positive MAP regions. Complete resection of the MAP-positive regions was positively associated with seizure-free outcome compared with the no/partial resection group (P < 0.001). Patients with no/partial resection of the MAP-positive regions had worse seizure outcomes than the MAP-negative group (P = 0.002). The MAP-positive rate was 100%, 77%, 63% and 40% in the 3-5, 5-10, 10-15 and 15-21 year age groups, respectively. MAP-positive rates were 45% in patients with temporal resection and 63% in patients with extratemporal resection. Complete resection of the MAP-positive regions was positively associated with seizure-free outcome in the extratemporal group (P = 0.001) but not in the temporal group (P = 0.070). Conclusion Our data suggest the importance of using MRI postprocessing in the pre-surgical evaluation process of pediatric epilepsy patients with apparently normal MRI.
What problem does this paper attempt to address?