Computational protocol: Correlation of neuropsychological and metabolic changes after epilepsy surgery in patients with left mesial temporal lobe epilepsy with hippocampal sclerosis

Similar protocols

Protocol publication

[…] An FDG-PET scan was acquired 30 min after injection of 150 MBq 18F-fluorodeoxyglucose (FDG) under continuous EEG monitoring, using an ECAT EXACT HR+ PET scanner. Prior to the injection, a transmission scan was performed. FDG-PET data were acquired as six frames of 5 min and reconstructed using an iterative reconstruction algorithm including corrections for randoms, dead time, scatter, and attenuation, and taking into account the finite point spread function of the PET system [, ]. Images were reconstructed using ordered subsets expectation-maximization (OSEM) algorithm with the number of iterations and subsets equivalent to 250 MLEM iterations. A correction for small movements was applied using the realignment function in SPM12 (Wellcome Department of Cognitive Neurology, London, UK), and images were summed. These images represent the tracer distribution of FDG in the brain between 30- and 60-min postinjection.All subjects underwent high-resolution MR imaging (T1- and T2-weighted sequences and a magnetization-prepared rapid gradient echo (MPRAGE)), which was performed on a 3T Vision Scanner (Siemens, Erlangen, Germany).For each patient, the preoperative FDG-PET image was co-registered to the preoperative structural MRI (MPRAGE) using SPM12, and the postoperative FDG-PET image was co-registered to the postoperative structural MRI (MPRAGE). Next, we performed a rigid body co-registration of the postoperative to the preoperative MRI scans followed by high-resolution warping of the pre- and postoperative structural MRI scans to account for the shift in anatomical structures in the brain due to the surgery. This image transformation was then applied to the corresponding FDG-PET images so that the deformation between preoperative and postoperative scans was taken into account. The preoperative MRI was segmented into gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) using SPM12. During this segmentation step, we also obtained the warping to the stereotactic Montreal Neurological Institute (MNI) space, and this transformation was applied to all MRI and PET data of that patient. Finally, the warped PET images were smoothed with an isotropic Gaussian 3D kernel with a full width at half maximum (FWHM) of 12 mm. Because we expected changes of metabolism in gray matter, we applied a gray matter mask. This mask was constructed based on all individual postoperative GM masks defined by taking all voxels with GM > 0.5. Since we were interested in the regional changes, we normalized the images by dividing them by the total activity within the final GM mask. Note that this procedure leads to a unit-less value in which differences in administered tracer activity or blood glucose levels between pre- and postoperative conditions or between patients are canceled out. This corrected image is a surrogate measure for glucose metabolism in the brain and reflects the regional distribution of glucose metabolism. Changes in the distribution of regional cerebral glucose metabolism were calculated as a difference image between the corrected postoperative and preoperative FDG-PET images. [...] Neuropsychological data analysis was performed using SPSS statistical software (IBM SPSS Statistics 20; Chicago, IL, USA). A paired t test was used to determine whether there was a significant difference in the neuropsychological test score post- versus preoperatively (p < 0.05). Changes in neuropsychological test scores were calculated as the postoperative value minus the preoperative value.A voxel-based paired t test was used to identify voxels which differ significantly in preoperative versus postoperative regional distribution of FDG metabolism. A voxel-based regression analysis was performed to identify changes in FDG metabolism related to the changes in cognitive functioning. For all voxel-based analyses, we used an uncorrected p < 0.001 at the voxel level combined with a FWE-corrected p < 0.05 at the cluster level. […]

Pipeline specifications

Software tools SPM, SPSS
Applications Miscellaneous, Magnetic resonance imaging