Grey matter loss in relapsing–remitting multiple sclerosis: A voxel-based morphometry study
Introduction
Brain tissue volumes have been measured with different techniques in relapsing–remitting multiple sclerosis (RR-MS). A decrease in white matter (WM) (Ge et al., 2001), grey matter (GM) (Quarantelli et al., 2003) or both tissue volumes simultaneously (Chard et al., 2002) has been demonstrated. Moreover, in a recent study on a group of patients with clinically isolated syndromes suggestive of MS (Dalton et al., 2004), GM loss (but not WM loss) was found in the subgroup of patients who developed in 3 years clinically defined MS, thus confirming the presence of GM atrophy already in the very early stages of the disease. None of these studies however addresses the question of whether the observed tissue loss is homogeneously distributed or rather localized in specific cortical or subcortical regions.
In previous studies assessing brain tissue loss over specific predefined brain regions, decreases in both area and axonal density in corpus callosum (Evangelou et al., 2000) and in thalamic GM (Cifelli et al., 2002) have been reported.
Nevertheless, to date, a comprehensive look at the GM regional changes has been done, to the best of our knowledge, only in one study (Sailer et al., 2003) using automatic surface reconstruction to measure the cortical thickness across the entire brain in a group of MS patients, including both RR and secondary progressive subtypes of the disease.
That work reported a highly significant focal atrophy bilaterally in frontal, temporal and motor areas in the MS patients, motor cortex involvement being related to physical disability.
In the present work, we have used voxel-based morphometry (VBM) (Ashburner and Friston, 2000), to compare on a voxel-by-voxel basis GM volumes as measured by segmentation of conventional spin-echo MR images. VBM allows to compare structural features across scans of different subjects in a fully automated manner, thus overcoming problems of intra- and interobserver bias and sensitivity, as well as the need for a priori definition of structures of interest. We applied VBM to a group of RR-MS patients and to an age-matched control group to assess regional GM loss on a voxel-by-voxel basis. VBM was also used to compare brain asymmetries between the two groups and to search the whole brain on a voxel-by-voxel basis for correlations between local GM volumes and disease duration, Expanded Disability Status Scale (EDSS) score and the abnormal WM (aWM) volume (hereinafter referred to as lesion load, LL).
Section snippets
Subjects
Fifty-one patients with clinically defined multiple sclerosis according to Poser criteria (Poser et al., 1983) with a RR course (Lublin and Reingold, 1996) (36 female) and 34 normal volunteer (NV, 15 female) were recruited in the study. The mean age of the patient group was 38.6 ± 7.5 years (age range 23–53), with a mean disease duration of 13.1 ± 6.4 years (range 4–34) and a median EDSS (Kurtzke, 1983) score of 2.6 (range 1.5–4.5). All patients had been previously treated only by brief courses
Results
The preliminary analysis of global GM volumes confirmed previous findings (Chard et al., 2002, Quarantelli et al., 2003) of GM loss in MS patients as compared to NV (607.7 ± 66.1 ml vs. 660.1 ± 49.7 ml, respectively corresponding to 45.7% and 47.0% of ICV), the difference being significant at multiple regression analysis (P < 0.001).
MS patients showed a moderate LL, aWM volume being 22.8 ± 18.8 ml (range 1.8–84.9).
In Fig. 2, the site- and study-specific unsmoothed GM template is displayed in
Discussion
We have applied an optimized VBM method to investigate regional GM loss in RR-MS patients.
Our results demonstrated a preferential left-sided fronto-temporal involvement, with other areas located in the left precuneus and bilaterally in anterior cingulate gyrus and caudate nuclei.
Our results integrate previous findings demonstrating global GM volume reduction selectively in RR-MS (Chard et al., 2002, Ge et al., 2001, Quarantelli et al., 2003).
However, before discussing the
Conclusion
In RR-MS, the loss of GM appears preferentially located in the left fronto-temporal cortex and precuneus, in the anterior cingulate gyrus and bilaterally in caudate nuclei.
The asymmetrical distribution of cortical atrophy is not explained by the volume of aWM or its symmetrical spatial distribution, and thus other mechanisms are probably involved in determining cortical GM loss, beside neuronal degeneration from axonal transection in areas of the brain remote from GM, which probably plays a
References (47)
- et al.
Functional architecture of basal ganglia circuits: neural substrates of parallel processing
Trends Neurosci.
(1990) - et al.
Voxel-based morphometry—The methods
NeuroImage
(2000) - et al.
Incorporating prior knowledge into image registration
NeuroImage
(1997) - et al.
Relating neocortical pathology to disability progression in multiple sclerosis using MRI
NeuroImage
(2004) - et al.
Analysis of fMRI time-series revisited
NeuroImage
(1995) - et al.
Cerebral asymmetry and the effects of sex and handedness on brain structure: a voxel-based morphometric analysis of 465 normal adult human brains
NeuroImage
(2001) - et al.
A voxel-based morphometric study of ageing in 465 normal adult human brains
NeuroImage
(2001) - et al.
A comprehensive study of gray matter loss in patients with Alzheimer's disease using optimized voxel-based morphometry
NeuroImage
(2003) - et al.
A voxel-based approach to gray matter asymmetries
NeuroImage
(2004) - et al.
Relationship between corpus callosum atrophy and cerebral metabolic asymmetries in multiple sclerosis
J. Neurol. Sci.
(1992)