Voxel-wise assessment of progression of regional brain atrophy in relapsing-remitting multiple sclerosis.

OBJECTIVE: To perform voxel-wise assessments of regional brain atrophy state and rate in subjects with relapsing-remitting (RR) multiple sclerosis (MS). BACKGROUND: Recently, attention has focused on defining the tissue compartments and regions within which brain atrophy occurs. These regional measures of brain volume changes may help to better define the nature of the pathology underlying MS. In this context, specific regional measures of grey matter (GM) volume changes can be obtained by using the voxel-based morphometry (VBM) approach. METHODS: Fifty-nine subjects with RR MS underwent conventional MRI at baseline and after a mean follow-up period of 3 years. Cross-sectionally, two VBM analyses (SPM-VBM, based on the Statistical Parametric Mapping software package, and FSL-VBM, based on the FMRIB Software Library tools) were performed to assess cortical GM volumes in RR MS patients compared to 25 age- and sex-matched normal controls (NC). Longitudinally, FSL-VBM and the regional extension of the SIENA method (SIENAr) were both used to assess regional brain atrophy rate in the RR MS patients and its relationship with increases in T(2)-weighted white matter (WM) lesion volume over the follow-up period. RESULTS: Widespread decrease in cortical GM volume was found in the RR MS patients compared to NC. Both SPM-VBM and FSL-VBM showed similar involvement of cortical regions (frontal, temporal, parietal, occipital lobes and insula), with a close correlation between the numbers of significant voxels obtained with the two different procedures (r=0.73, p<0.001). After 3-year follow-up, both FSL-VBM and SIENAr showed a further significant reduction in GM volume in the lateral frontal and parietal cortices, bilaterally. Regional volume changes also appeared significantly pronounced in correspondence to the increase in T(2)-weighted WM lesion volume over the follow-up period. CONCLUSIONS: By using different methodologies, we showed similar widespread tissue loss in the cerebral cortex of patients with RR MS. This brain tissue loss further progresses over time, particularly in the fronto-parietal cortex and seems to be partially dependent upon the increase of lesion load.