INTERHEMISPHERIC FUNCTIONAL CONNECTIVITY CORRELATES OF UPPER-LIMB MOTOR PERFORMANCE IN MULTIPLE SCLEROSIS PATIENTS WITH DIFFERENT CLINICAL PHENOTYPES

C. Cordani¹, P. Valsasina¹, F. Esposito², M. Filippi^1,2,3,4, M.A. Rocca^1,2

Background: Interhemispheric communication plays an important role in human motor control. Voxel-mirrored homotopic connectivity (VMHC) analysis, based of resting state (RS) functional connectivity (FC) magnetic resonance imaging (MRI) metrics, could provide new insights interhemispheric homotopic connections and their contribute explain motor impairment in multiple sclerosis (MS) patients.

Purpose: Using VMHC, we investigated in a large cohort of multiple sclerosis (MS) patients:
i) abnormalities of inter-hemispheric RS FC according to MS clinical phenotype;
ii) correlations between altered VMHC metrics and measures of manual dexterity, as well as global clinical disability.

Methods: One-hundred and twenty-one right-handed healthy controls (HC) and 340 MS patients (234 relapsing-remitting [RR] and 106 progressive [P] MS) underwent a manual-dexterity assessment, using Nine-Hole-Peg Test (9HPT) and finger-tapping (FT). In MS, clinical disability was rated with Expanded Disability Status Scale (EDSS) score. 3T brain 3D T1-weighted and RS fMRI scans were acquired and used to perform a VMHC analysis to produce maps of homotopic RS FC in all study subjects. SPM and voxel-wise general linear models were used to investigate between-group differences of VMHC and correlations with clinical measures.

Results: Compared to HC, RRMS patients showed significantly increased VMHC in occipital, thalamic, hippocampal, superior temporal and insular regions. Conversely, decreased VMHC was observed in cerebellar, middle temporal, superior and middle frontal regions, as well as in the precunei, cingulate cortices and supplementary-motor areas (SMAs). A similar VMHC increase was observed in PMS vs HC while VMHC reductions were detected in frontal, sensorimotor, parietal and cerebellar regions. These reductions were significant also between PMS and RRMS. In RRMS patients, worse motor scores correlated with increased VMHC in thalamic and insular regions, and with decreased VMHC in SMAs, cingulate and frontal cortices, and small clusters in cerebellar lobules
(r-range=-0.35/0.25, p-range <0.001/0.05). EDSS was associated with increased thalamic (r=0.30, p<0.001) and decreased cerebellar VMHC (r=-0.15, p=0.02). In PMS, worse left-9HPT and right-FT scores correlated with a widespread decrease of VMHC in parietal and cerebellar regions (r-range=0.19/0.22, p-range=<0.01/0.05). Conversely, a higher EDSS score was correlated with increased VMHC in thalamic (r=0.19, p=0.05) and inferior frontal (r=0.21, p=0.03) regions.

Conclusion(s): Distinct VMHC abnormalities characterized different MS clinical phenotype and contributed to explain upper-limb motor deficits and global clinical disability. A closer relationship between altered VMHC and upper-limb motor impairment was found in RRMS than in PMS.

Implications: A better comprehension of the brain MRI substrates of upper limb motor performance could provide new pieces of information about MS motor deficits and contribute to develop new and more specific rehabilitative intervention in patients with MS.

Funding, acknowledgements: None

Keywords: Multiple sclerosis, upper limb impairment, magnetic resonance imaging

Topic: Neurology: multiple sclerosis

Did this work require ethics approval? Yes

Institution: IRCCS San Raffaele Scientific Institute, Milan, Italy

Committee: Ethical Committee of IRCCS San Raffaele Hospital, Milan, Italy

Ethics number: Prot. 336/11