Temporal lobe epilepsy lateralization using retrospective contrast enhanced imaging
Abstract number :
3.436
Submission category :
5. Neuro Imaging / 5B. Functional Imaging
Year :
2017
Submission ID :
392825
Source :
www.aesnet.org
Presentation date :
12/4/2017 12:57:36 PM
Published date :
Nov 20, 2017, 11:02 AM
Authors :
Frank Provenzano, Columbia University; Xinyang Feng, Columbia University; Marla J. Hamberger, Columbia University; and Scott Small, Columbia University
Rationale: Contrast enhanced MRI frequently accompanies Lateralized Temporal Lobe Epilepsy (TLE) for pre-surgical evaluation. The criteria for those scans fortuitously share very similar protocols to specialized cerebral blood volume (CBV) fMRI. CBV offers high resolution interrogation of the hippocampal circuit and CBV measurements correlate to metabolic function. However, there is a wide variance of scanners and scanning protocols. Using an improved method of scan normalization, we were able to generate hippocampal CBV in patients both with and without noted medial temporal sclerosis to determine if it outperforms hippocampal volume measurement in distinguishing ipsi versus contralateral involvement. Methods: Twenty five subjects were recruited with temporal lobe epilepsy who had acquired T1 weighted MRI imaging with and without contrast agent. Those scans without contrast were brain extracted, tissue class segmented, processed through FreeSurfer 6.0 and the post contrast image was scaled to correct for receiver gain using a maximum likelihood estimation (MLE) based on white matter tissue uniformity. Values for mean hippocampal CBV were generated for the ipsilateral and contralateral side, as well as values of the hippocampal volume. Results: We found a statistically significant relative CBV decrease in the hippocampus ipsilateral to seizure-onset (paired t-test, Left-TLE=18, Right-TLE=7, p<0.05) with a mean contralateral ipsilateral difference 0.076 (standard deviation, 0.174). This finding is in line with previous studies showing ipsilateral hypometabolism in mesial temporal lobe using PET, arterial spin labeling (ASL) and DSC-MRI. There was no significant hippocampal volume difference in this group. Conclusions: Using previously acquired scans from an unrelated purpose, we were able to calculate a functional measure that can automatically generate a value useful in establishing laterality. This technique requires only a pair of images that are acquired for the purposes of surgical evaluation. Future applications of this technique to generate CBV may be useful in tracking functional changes in subjects who had previously acquired contrast enhanced imaging, as well as a method that can be acquired at almost any MRI scanner without specialized equipment or training. Funding: American Epilepsy Society Seed Grant (MJH)Taub Institute MRI Platform grant (MJH)
Neuroimaging