Abstract #7: Voxel Based Morphometry with DARTEL in Multiple Sclerosis

Abstract

Background Voxel based morphometry (VBM) is an automated tool used to assess the topography and magnitude of cerebral atrophy from MRI scans. However, VBM output is highly dependent on the input provided by registration and segmentation algorithms. Diffeomorphic anatomical registration through exponentiated lie algebra (DARTEL) is a new tool which has recently been developed to improve registration in healthy volunteers (HV) and neurodegenerative diseases. This has not yet been applied to multiple sclerosis (MS). Objective To assess regional gray matter (GM) and white matter (WM) atrophy using VBM combined with DARTEL in MS patients. Methods Using a 3T MRI scanner, fluid-attenuated inversion-recovery (FLAIR) and 3D spoiled gradient recalled (SPGR) scans were acquired from 26 relapsing remitting (RR) MS patients [age (mean ± SD) 39.2 ± 9.4 years, disease duration 8.5 ± 5.9 years, Expanded Disability Status Scale (EDSS) score 1.2 ± 0.9, FLAIR lesion volume (FLV) 15.0 ± 14.7 mL] and 28 sex- and age-matched HV. MS lesions on SPGR and FLAIR images were traced using a semi-automated thresholding tool. VBM was performed using DARTEL with statistical parametric mapping (SPM8) software from SPGR images. An analysis of covariance was used to assess differences in GM and WM volume between MS patients and HV. Age, gender, and intracranial volume were included as covariates. Results Compared with HV, MS patients showed significant GM atrophy in the bilateral thalami, bilateral caudate, and right putamen. Furthermore, WM atrophy was found in the brainstem, corpus callosum, fornix, and bilateral temporal and occipital lobes. FLV significantly correlated with GM atrophy in the thalami and with WM atrophy in the brainstem, temporal lobes, and corpus callosum. Conclusions VBM-DARTEL reveals widespread WM and GM cerebral atrophy affecting many areas of the brain, which is associated with overt lesions in MS patients. This pilot study suggests that VBM-DARTEL is a promising tool to derive meaningful disease specific information in MS. Acknowledgements This work was supported by research grants from the National Institutes of Health ( NIH R01 NS055083-01 ) and the National Multiple Sclerosis Society ( RG3705A1 , RG3798A2 ).