Mri-negative Temporal Lobe Epilepsy and Presurgical FDG-PET
Abstract number :
2.185
Submission category :
5. Neuro Imaging / 5A. Structural Imaging
Year :
2022
Submission ID :
2204411
Source :
www.aesnet.org
Presentation date :
12/4/2022 12:00:00 PM
Published date :
Nov 22, 2022, 05:24 AM
Authors :
Julia Simon, MS – Keck School of Medicine of University of Southern California; Yaqiong Chai, PhD – Keck School of Medicine of University of Southern California; Yan Li, PhD – Neurology and UCSF Weill Institute for Neurosciences; Wolfgang Muhlhofer, MD – Epilepsy Center at UW; Robert Knowlton, MD – Neurology and UCSF Weill Institute for Neurosciences; Hosung Kim, PhD – Keck School of Medicine of University of Southern California
Rationale: Temporal lobe epilepsy (TLE) is the most common medically refractory form of focal epilepsy, with resection of the affected mesial temporal tissue being the most effective treatment for this disease (Epilepsy Res. 2010;89(2-3):310-318). Localization and confirming lateralization of epileptogenic tissue remains paramount for success. However, at least 30% of TLE patients have normal appearing mesial temporal structures on MRI (MRI-) and no detectable hippocampal sclerosis (HS). Presumably due to this problem, MRI- patients more often do not achieve seizure freedom after surgery (Epilepsia. 2017;58(5):727-742). Visual interpretation of 2-[18F]fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET) has been identified to be particularly valuable for MRI- patients when relative focal hypometabolism is demonstrated in anterior temporal regions of the electroclinically suspected temporal lobe (Epilepsia. 2017;58(5):727-742). Good surgical outcome correlates with such ipsilateral PET hypometabolism (Neuroradiology. 2013;55(5):541-550). Here we investigated the utility of objectively measured cortical PET features to identify hypometabolic regions in MRI- TLE.
Methods: We studied 77 MRI- patients (age yrs=39.6±12.4) including 40 left lateralized TLE (LMRI-;18 Male) and 37 right TLE (RMRI-;13M). To compare, we also studied 34 HS+ patients (43.1±13.2), 21 left (LHS+;9M) and 13 right (RHS+;4M). Presurgical MRI- TLE cases were prospectively identified at two sites: UCSF (n=70) and UAB (n=41). We then chose 37 age- and sex-matched healthy controls (38.1±11.4; 17M) with FDG-PET and T1 MRI from the CERMEP-IDB-MRXFDG database (EJNMMI Res. 2021;11(1): 1-10). We coregistered each subject’s MRI and PET images, reconstructed cortical surface, sampled PET on the mid cortical surface, and normalized PET relative to the mean PET value in lateral ventricles, as previously described (Neuroimage. 2018;166:10-18). LTLE and RTLE subjects were pooled by flipping hemispheres. PET values were compared between hemispheres for HS+ and MRI- using paired t-tests. To measure lateralization we used the PET asymmetry index: 2x(L-R)/(L+R). General linear models (GLM) were used to compare cortical PET features between each TLE group and controls, correcting for age and sex.
Results: PET asymmetry analysis showed significant hypometabolism in the ipsilateral hemisphere compared to contralateral, mainly found in the entire temporal lobe (corrected p< 0.05). Our PET individual analysis shows that 34/40 LMRI-, 32/37 RMRI-, 19/21 LHS+, and 13/13 RHS+ were correctly lateralized. GLM showed a similar pattern of hypometabolism in both HS+ and MRI- patients compared to controls, including ipsilateral temporal lobe as well as bilateral frontal, insular, and cingulate cortices.
Neuro Imaging