1035
Association of hypometabolic extension of 18F-FDG PET with DTI in hippocampal sclerosis1UCLA, Los Angeles, CA, United States
Synopsis
To assess associations between hypometabolic extensions of FDG PET and DTI, including MD and FA, 36 unilateral hippocampal sclerosis were stratified into broad hypometabolic (n=26) and localized groups (n=10) by the extension of FDG hypometabolism beyond or within the temporal lobe. Gray matter (GM) ROIs and TBSS were used to compare MD of GM and FA of white matter (WM) between hemispheres ipsilateral and contralateral to epilepsy focus. Associations between hypometabolism and DTI alterations, including more widespread GM abnormalities for broad than localized groups and significantly decreased FA in the temporal and frontal WM for broad group, were identified.
Introduction
Hypometabolic extensions of 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) and diffusion tensor imaging (DTI) may have associations for mesial temporal lobe epilepsy (MTLE) with hippocampal sclerosis (HS). No studies have evaluated the association of DTI abnormalities and hypometabolism depending on the severity of hypometabolism, such as the hypometabolic extension within or beyond the temporal lobe. This study assessed the association between hypometabolic extensions of FDG PET and DTI indices, including mean diffusivity (MD) and fractional anisotropy (FA), for MTLE with unilateral HS.Methods
A total of 36 MTLE subjects with unilateral HS (left HS, 20; right HS, 16) were retrospectively selected between 2014 and 2020, and stratified into broad hypometabolic group (n = 26; FDG hypometabolic regions extended beyond the temporal lobe) and localized hypometabolic group (n = 10; FDG hypometabolic regions remained within the temporal lobe). Forty-one pairs of gray matter (GM) regions of interest (ROIs) were segmented using FreeSurfer software. Asymmetry ratio of hippocampal volume (ratio of HS side to normal side) was compared between groups using Student’s t-test. The GM ROIs were applied to MD maps, and median MD values within each ROI were compared between hemispheres ipsilateral and contralateral to HS using a mixed effect model (fixed slope and random intercept). Tract-Based Spatial Statistics (TBSS; FSL software) was used to evaluate FA of white matter (WM) tracts between hemispheres ipsilateral and contralateral to HS. To combine FA maps for the TBSS analysis, the FA maps of the right MTLE patients were flipped left and right. In both analyses, disease laterality (left or right HS) was controlled because MD and FA may be potentially different between the left and right hemispheres due to the functional asymmetry of the left and right hemisphere. Statistical significance was defined as P < 0.05.Results
The asymmetry ratio of hippocampal volume was significantly lower in the broad hypometabolic group than the localized hypometabolic group (P = 0.042). The median MD values in the thalamus (P = 0.031), caudate (P = 0.043), hippocampus, (P < 0.001) amygdala (P < 0.001), superior frontal gyrus (P = 0.013), middle (P = 0.020) and inferior temporal gyrus (P = 0.001), temporal pole (P < 0.001), and isthmus cingulate gyrus (P = 0.033) were significantly higher in the hemisphere ipsilateral to HS than the contralateral side for HS with broad hypometabolic group. Those in the amygdala (P = 0.010) and banks of the superior temporal sulcus (P = 0.038) were significantly higher in the ipsilateral side than the contralateral side for HS with localized hypometabolic group. The broad hypometabolic group showed more widespread abnormal areas in the GM than the localized hypometabolic group. In the TBSS analysis, the broad hypometabolic group showed significantly decreased FA in the WM tract in the temporal and frontal lobes, while the localized hypometabolic group showed no significantly abnormal tracts.Discussion
The broad hypometabolic group showed more widespread microstructure abnormalities of the GM and WM on MD and FA maps, respectively. A previous study divided MTLE subjects into three groups according to the hypometabolic extension in the anterior, middle, or posterior temporal lobe, and evaluated FA values in the WM tracts.1 WM abnormalities were restricted to the anterior temporal and frontal lobes when the hypometabolism was located in anterior third of the temporal lobe, whereas WM abnormalities were more widespread when metabolic abnormalities extended to the middle or posterior temporal lobes. Our study, which divided subjects depending on the hypometabolic extension beyond or within the temporal lobe, was consistent with this previous study, and newly revealed that GM abnormalities were also associated with the extension of hypometabolism. Although the association between MD values and hypometabolism in GM was contraversial,2,3 imaging features of these two metrics may be partially overlapped. Meanwhile, FDG PET were reported to predict postoperative outcomes based on the extent of hypometabolism in patients with HS with more extensive and extratemporal hypometabolism may lead to a poor prognosis.4,5 Our study, therefore, suggested that DTI alterations have a similar trend with severities of hypometabolic extension, and that DTI indices may also have a potential to become an alternative to FDG PET for predicting post-operative outcomes.Conclusion
The hypometabolic extension for HS was associated with the abnormalities of MD and FA in GM and WM, respectively, as well as hippocampal asymmetry ratio, which may help predict post-operative outcomes.Acknowledgements
This work was supported by the Society of Nuclear Medicine and Molecular Imaging.References
1. Aparicio J, Carreno M, Bargallo N, et al. Combined (18)F-FDG-PET and diffusion tensor imaging in mesial temporal lobe epilepsy with hippocampal sclerosis. Neuroimage Clin 2016;12:976-9892.
2. Kimiwada T, Juhász C, Makki M, et al. Hippocampal and thalamic diffusion abnormalities in children with temporal lobe epilepsy. Epilepsia 2006;47:167-1753.
3. Lippe S, Poupon C, Cachia A, et al. White matter abnormalities revealed by DTI correlate with interictal grey matter FDG-PET metabolism in focal childhood epilepsies. Epileptic Disord 2012;14:404-4134.
4. Willmann O, Wennberg R, May T, et al. The contribution of 18F-FDG PET in preoperative epilepsy surgery evaluation for patients with temporal lobe epilepsy A meta-analysis. Seizure 2007;16:509-5205.
5. Chassoux F, Artiges E, Semah F, et al. (18)F-FDG-PET patterns of surgical success and failure in mesial temporal lobe epilepsy. Neurology 2017;88:1045-1053
Figures
