Abstracts

MICROSTRUCTURAL ABNORMALITIES IN THE UNCINATE FASCICULUS AS DETERMINED BY TBSS AND PROBABILISTIC TRACTOGRAPHY IN TEMPORAL LOBE EPILEPSY

Abstract number : 3.140
Submission category : 5. Human Imaging
Year : 2008
Submission ID : 9325
Source : www.aesnet.org
Presentation date : 12/5/2008 12:00:00 AM
Published date : Dec 4, 2008, 06:00 AM

Authors :
Jeff Riley, V. Choi, S. Cramer and Jack Lin

Rationale: We have previously used tract-based spatial statistics (TBSS) to show subnormal white matter integrity, as evidenced by decreased fractional anisotropy (FA) in the anterior temporal lobe ipsilateral to a temporal lobe epilepsy (TLE) focus. However, the white matter tract associated with this area, as well as the underlying pathophysiology of this white matter derangement, has not been elucidated. To address these questions, we performed probabilistic tractography to construct connectivity probabilities with the area of altered white matter integrity. We also performed TBSS using components of the diffusion tensor in order to look for changes in mean diffusivity (MD), parallel diffusivity (PD), and radial diffusivity (RD) in individuals with TLE, when compared to controls. Methods: Twelve subjects with unilateral TLE (10 left/2 right; mean ± SD age, 37.9 ± 10.7 yrs; seizure onset, 14.3 ± 12.4 yrs; duration, 23.6 ± 12.9 yrs) and 10 controls (age 42.1 ± 10.1) were included in this study. DTI data were obtained using a 3-Tesla MR scanner with gradients applied in 12 non-collinear directions with b value=1000 s/mm2, repeated two times to increase signal-to-noise ratio. The two right TLE subject data were flipped across the midline to maintain consistent side of seizure onset. Voxelwise statistical analysis of FA, MD, PD, and RD was carried out using TBSS [Smith 2006], part of FSL [Smith 2004]. TBSS projects all subjects' data onto a mean FA tract skeleton, before applying voxelwise cross-subject statistics (cluster-forming threshold t=2, corrected cluster size significance level p<0.05) with subject age entered as a nuisance variable. A cluster of significant FA difference found in TBSS was then used as a seed mask for probabilistic tractography using FMRIB’s Diffusion Toolbox. Results: Using TBSS, a cluster of reduced FA was again found in the ipsilateral temporal lobe of individuals with TLE, as compared to controls. Probabilistic tractography, using this cluster as a seed mask, showed a pattern of connectivity consistent with the uncinate fasciculus (UF). TBSS of the diffusion tensor components showed increased MD and increased RD, with no change in PD, coinciding with the cluster of decreased FA. Conclusions: White matter integrity is reduced in the anterior temporal lobe ipsilateral to seizure onset in individuals with TLE. Probabilistic tractography demonstrates that this area is a portion of the UF, an important frontotemporal connection implicated in seizure propagation in TLE. The profile of the diffusion tensor components suggest the reduced white matter integrity is primarily due to derangement of myelin (increased RD) with relative preservation of axonal integrity (unchanged PD). Funding supported by: NIH T32 NS45540, PI: Baram, T. Z. and GCRC at UC Irvine.
Neuroimaging