Characterization of Early White Matter Damages in Multiple Sclerosis patients with a Clinically Isolated Syndrome: a Tract Based Spatial Statistics Study

Introduction Diffusion tensor imaging (DTI) has been recognized as a sensitive technique for detecting and quantifying pathological changes in lesions and normal appearing white matter (NAWM) [1]. However, it is usually constrained by an operator-dependant selection of brain regions of interest (ROI). With the emergence of the “tract-based spatial statistics (TBSS)” method [2], whole brain statistical analysis can be performed on DTI data. By creating a mean skeleton of white matter from any diffusivity scalars such as fraction of anisotropy (FA), axial (λa) or radial (λr) diffusivities, TBSS provides voxel-wise statistics across groups of subjects. Therefore, the goal of this study was to characterize the early pathological processes occurring in WM of Multiple Sclerosis (MS) patients with a clinically isolated syndrome (CIS) [3]. Methods This study included 12 CIS (33.0±6.7y) and 12 RR (30.5±6.4y) patients along with 12 control subjects (32.2±7.4y) matched for age and sex. All patients were diagnosed with definite MS according to McDonald’s criteria and their expanded disability status scale ratings (EDSS) measured. MR exams were performed on a 1.5 T Siemens Sonata system. DTI protocol included a spin-echo EPI sequence (TR=3800 ms, TE=96 ms) with 96x96 phases-encoding over a FOV of 240x240 mm and 51 axial slices of 2.5 mm thickness. Processing of DTI data was performed using FSL software [4]. Eddy current correction and brain extraction (factor of 0.35) were applied before generating diffusivity scalar maps (FA, λa, λr) using the FDT module. TBSS registration and tract skeletonisation were then processed. First, statistics based on voxel-wise cross-subject analysis were performed in between groups of patients and controls to produce maps of altered diffusivity scalars (Fig. 1). Second, the altered FA map was extracted as a ROI and applied on each subject to individually quantify changes of FA, λa and λr. One-Way ANOVA test was applied for group comparisons. Non-parametric Spearman’s test was used for analysing correlations between diffusivity scalars, disease duration (DD) and EDSS.