Abstracts

Rationale: Focal cortical dysplasia (FCD) has been recognized as frequent cause of pharmacoresistant focal epilepsies, for which epilepsy surgery is an effective treatment option. FCDs are increasingly detected using optimized MR imaging protocols, however, subtle lesions may escape visual analysis. Advanced MRI postprocessing techniques, such as voxel-based morphometry (VBM) provides additional information. As a functional complement, magnetencephalography (MEG) enables localization of epileptic activity, providing evidence for pathophysiological involvement of detected structural alterations. A combined VBM/MEG approach is suggested for detection of FCD. Methods: MEG and VBM data of 16 consecutive patients with suspected FCD type 2 were analyzed retrospectively. MEG had been acquired using a whole-head 248 channel magnetometer system (Magnes 3600WH, 4D-Neuroimaging, San Diego, CA, USA) and a two sensor gradiometer system with 2x37 channels (Magnes II, 4D-Neuroimaging). Analysis of interictal activity consisted of single moving dipole analysis in a spherical volume conductor (Curry 7, Compumedics Neuroscan, Singen, Germany). Voxel-based morphometry was calculated based on 3D-MPRAGE T1 images using the MAP software (morphometric analysis program, Huppertz et al., 2005) and a scanner specific database of controls. MEG and VBM were coregistered and compared. Based on all clinical findings, including reinspection of MRI and histology when available, the location of the "true" FCD was defined and compared to VBM and MEG. Results: Spikes in MEG were detected and successfully localized to single clusters in all patients. VBM showed significant alterations in all patients, which were always regionally concordant with MEG. However, in some cases, VBM showed several significant areas next to the maximum. Distance of closest VBM alteration and MEG localizations was on average 12.6mm (+-14.1). Average distance to the real FCD was 10.8mm (+-7.0). VBM maximum was concordant with the real FCD in 6 cases, whereas this was the case for the MEG-marked VBM alterations in 13 of the 16 cases (p = 0.03, Fisher's exact test). Conclusions: The combination of MEG and VBM represents a valuable tool for detection of FCD in pharmacoresistant epilepsies. It may enable patients to proceed to surgery who were previously considered not eligible. The improved delineation of subtle lesions is likely to improve success of epilepsy surgery. Funding: No funding was received in support of this abstract.