Assessing the correlation between grey and white matter damage with motor and cognitive impairment in multiple sclerosis patients

Abstract

Background: Multiple sclerosis (MS) is characterized by demyelinating and degenerative processes within the central nervous system. Unlike conventional MRI,new advanced imaging techniques improve pathological specificity and better highlight the relationship between anatomical damage and clinical impairment.

Objective: To investigate the relationship between clinical disability and both grey (GM) and white matter (WM) regional damage in MS patients.

Methods: Thirty-six relapsing remitting-MS patients and 25 sex- and age-matched controls were enrolled. All patients were clinically evaluated by the Expanded Disability Status Scale and the Multiple Sclerosis Functional Composite (MSFC) scale, which includes the 9-hole peg test (9HPT), the timed 25-feet walking test (T25FW) and the paced auditory serial addition test (PASAT). All subjects were imaged by a 3.0 T scanner: dual-echo fast spin-echo, 3DT1-weighted and diffusion-tensor imaging (DTI) sequences were acquired. Voxel-based morphometry (VBM) and tract-based spatial statistics (TBSS) analyses were run for regional GM and WM assessment, respectively. T2 lesion volumes were also calculated, by using a semi-automated technique.

Results: Brain volumetric assessment of GM and DTI measures revealed significant differences between patients and controls. In patients, different measures of WM damage correlated each-other (p<0.0001), whereas none of them correlated with GM volume. In patients, focal GM atrophy and widespread WM damage significantly correlated with clinical measures. In particular, VBM analysis revealed a significant correlation (p<0.05) between GM volume and 9HPT in cerebellum and between GM volume and PASAT in orbito-frontal cortex. TBSS showed significant correlations between DTI metrics with 9HPT and PASAT scores in many WM bundles (p<0.05), including corpus callosum, internal capsule, posterior thalamic radiations, cerebral peduncles.

Conclusions: Selective GM atrophy and widespread WM tracts damage are associated with functional impairment of upper-limb motion and cognition. The combined analysis of volumetric and DTI data may help to better understand structural alterations underlying physical and cognitive dysfunction in MS.

Figure 1. Differences in GM volume between patients and controls.

GM volume is decreased in patients group in comparison to controls. Differences are evident in the cerebellum, thalamus, subgenual gyrus, middle cingulate cortex, superior frontal gyrus, occipital and temporal cortices, bilaterally. Clusters of significant differences are superimposed on sagittal, coronal and axial slices of the single-subject T1 template provided with SPM8. (VBM analysis, SPM 8, two-sample, p<0.05, FWE corrected; t value are shown).

Figure 2. Correlation between GM volume in patients and MSFC subscales.

On the top of the figure, GM volume negatively correlates with 9HPTD score at cerebellum level. On the bottom of the figure, GM volume positively correlates with PASAT 2 s at level of orbitofrontal cortex. Clusters of significant correlations are superimposed on sagittal, coronal and axial slices of the single-subject T1 template provided with SPM8. (VBM analysis, SPM 8, one-sample, p<0.05, FWE corrected; t values are shown).

Figure 3. Differences in DTI measures between patients and controls.

Statistical maps showing voxels which exhibit differences in DTI parameters in patients versus controls (red and yellow colours, according to the lower and higher significance, respectively ). FA is significantly decreased and MD, AD, RD are increased in patients group in comparison to controls. Differences are widespread and evident in the corona radiata, corpus callosum, internal and external capsule, superior cerebellar peduncles, superior and inferior longitudinal fasciculum, posterior thalamic radiations and cerebral peduncle bilaterally. All WM tracts are overlaid on MNI152 1 mm standard image.(TBSS analysis, two-sample, p<0.05, tfce corrected).

Figure 4. Correlation between FA in patients and MSFC subscales.

Statistical map showing voxels which correlate with clinical scores. a) FA negatively correlates with 9HPT score obtained with dominant and non-dominant hand in corpus callosum, internal and external capsule, posterior thalamic radiations, cerebral peduncles. FA is positively correlated with the PASAT score at 2 seconds in the internal and external capsule, corpus callosum, posterior thalamic radiations, cerebral peduncles (red and yellow colours according to the lower and higher significance, respectively). b) WM tracts with reduced FA that are common to the WM skeletons resulted by the correlation between FA values and 9HPTD, 9HPTND and PASAT 2 s. All WM tracts are overlaid on MNI152 1 mm standard image. (TBSS analysis, one-sample, p<0.05, tfce corrected).