Abstracts

Subtraction techniques using ictal and interictal SPECT improve identification of ictal hyperperfusion compared to direct visual comparison of ictal and interictal SPECT images. The importance of perfusion changes in the interictal SPECT image to this subtraction image is unclear. Our aim was to determine the extent to which interictal SPECT perfusion changes contribute to the detection of ictal hyperperfusion in subtraction ictal SPECT coregistered to MRI (SISCOM) in temporal lobe epilepsy (TLE) patients. Sixteen consecutive patients who met the study criteria and who underwent epilepsy surgery at Mayo Clinic, Rochester, MN for intractable TLE were included. For all patients: MR imaging was normal, or demonstrated mesial temporal lobe sclerosis only; a standard anterior temporal lobectomy with amygdala-hippocampectomy was performed; and postsurgical seizure control was excellent at one year (Engel class I, no seizures or auras only). Patients were divided into right and left TLE on the basis of excellent surgical outcome. A reference dataset of SPECT images of 19 normal volunteers was used for comparison. Statistical parametric mapping (SPM2) was used to identify areas of increased or decreased perfusion in the individual interictal SPECT images compared to the normal SPECT images. The study was approved by the Mayo Foundation Institutional Review Board. On group analysis, multiple areas of hyperperfusion were identified ipsilateral to the epileptogenic zone involving the inferior frontal, superior temporal, middle temporal, and inferior temporal gyri, as well as the parahippocampus and hippocampus. No ipsilateral hypoperfusion of the epileptogenic TL was seen on group comparison. A separate analysis was performed to model the elapsed time from ictal to interictal scan as a covariate of interest. Using this model, no statistically significant correlation was seen for the elapsed time from ictal to interictal scan, and hyperperfusion ipsilateral to the epileptogenic zone.
On individual interictal SPECT image analysis, using the a priori hypothesis of ipsilateral TL hypoperfusion and directly comparing the individual interictal SPECT image to the group of controls, we did not localize the epileptogenic TL in any of the 16 patients. Interictal SPECT images performed for SISCOM display interictal increases TL perfusion in patients with TLE. These findings may represent an uncoupling of cerebral blood flow (CBF) and metabolism in the context of frequent seizures, or conversely, an increase in TL perfusion in response to increased interictal activity. Statistical comparison of ictal SPECT images to normals using SPM to screen for asymmetry in ictal TL perfusion is a potential alternative to subtraction techniques for SPECT image analysis. (Supported by Mayo Clinic Foundation for Education and Research.)