Related Experiment Video
Updated: Aug 29, 2025
Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke
Published on: July 14, 2023
Investigating Functional Network Abnormalities and Associations With Disability in Multiple Sclerosis
Antonio Carotenuto1, Paola Valsasina1, Menno M Schoonheim1
1From the Neuroimaging Research Unit (A.C., P.V., M.F., M.A.R.), Division of Neuroscience, IRCCS San Raffaele Scientific Institute, Milan, Italy; Department of Anatomy and Neurosciences (M.M.S., J.J.G.), MS Center Amsterdam, Amsterdam Neuroscience, Amsterdam UMC, Vrije Universiteit Amsterdam, The Netherlands; Radiology and Nuclear Medicine (F.B.), MS Center Amsterdam, Amsterdam Neuroscience, Amsterdam UMC, Vrije Universiteit Amsterdam, The Netherlands; Institutes of Neurology & Healthcare Engineering (F.B.), UCL, London, United Kingdom; National Institute for Health Research (NIHR) (F.B.), University College London Hospitals (UCLH), Biomedical Research Centre, London, United Kingdom; Division of Neurology and 3T MRI Research Center (A.G.,G.T.), Department of Advanced Medical and Surgical Sciences, University of Campania "Luigi Vanvitelli", Naples, Italy; Department of Human Neurosciences (S.T., P.P), Sapienza University, Rome, Italy; IRCCS NEUROMED (P.P.), Pozzilli (IS), Italy; Neurology Unit (M.F., M.A.R.), IRCCS San Raffaele Scientific Institute, Milan, Italy; Neurorehabilitation Unit (M.F.), IRCCS San Raffaele Scientific Institute, Milan, Italy; Neurophysiology Service (M.F.), IRCCS San Raffaele Scientific Institute, Milan, Italy; Vita-Salute San Raffaele University (M.F., M.A.R.), Milan, Italy.
Background And Objectives:
In multiple sclerosis (MS), functional networks undergo continuous reconfiguration and topography changes over the disease course. In this study, we aimed to investigate functional network to pography abnormalities in MS and their association with disease phenotype, clinical and cognitive disability, and structural MRI damage.
Methods:
This is a multicenter cross-sectional study. Enrolled participants performed MRI and neurologic and neuropsychological assessment. Network topography was assessed on resting state fMRI data using degree centrality, which counted the number of functional connections of each gray matter voxel with the rest of the brain. SPM12 age-adjusted, sex-adjusted, scanner-adjusted, framewise displacement, and gray matter-volume adjusted analysis of variance and multivariable regressions were used (p < 0.05, family-wise error [FWE] corrected).
Results:
We enrolled 971 patients with MS (624 female patients; mean age = 43.1 ± 11.8 years; 47 clinically isolated syndrome [CIS], 704 relapsing-remitting MS [RRMS], 145 secondary progressive MS [SPMS], and 75 primary progressive MS [PPMS]) and 330 healthy controls (186 female patients; mean age = 41.2 ± 13.3 years). Patients with MS showed reduced centrality in the salience and sensorimotor networks as well as increased centrality in the default-mode network vs controls (p < 0.05, FWE). Abnormal centrality was already found in CIS vs controls and in RRMS vs CIS (p < 0.001, uncorrected); however, it became more severe in SPMS vs RRMS (p < 0.05, FWE) and in PPMS vs controls (p < 0.001, uncorrected). Cognitively impaired patients (39%) showed reduced centrality in the salience network and increased centrality in the default-mode network vs cognitively preserved patients (p < 0.001, conjunction analysis). More severe disability correlated with increased centrality in the right precuneus (r = 0.18, p < 0.05 FWE). Higher T2 lesion volume and brain/gray matter atrophy were associated with reduced centrality in the bilateral insula and cerebellum (r = range -0.17/-0.15 and 0.26/0.28, respectively; p < 0.05, FWE). Higher brain/gray matter atrophy was also associated with increased centrality in the default-mode network (r = range -0.31/-0.22, p < 0.05, FWE).
Discussion:
Patients with MS presented with reduced centrality in the salience and primary sensorimotor networks and increased centrality in the default-mode network. Centrality abnormalities were specific for different disease phenotypes and associated with clinical and cognitive disability, hence suggesting that voxel-wise centrality analysis may reflect pathologic substrates underpinning disability accrual.
