Cognitive impairment in multiple sclerosis is related to reduced functional network state switching

Poster No:


Submission Type:

Abstract Submission 


Tommy Broeders1, Vasco Rauh1, Linda Douw1, Christiaan Vinkers1, Jeroen Geurts1, Menno Schoonheim1


1Amsterdam University Medical Center, Amsterdam, Netherlands

First Author:

Tommy Broeders  
Amsterdam University Medical Center
Amsterdam, Netherlands


Vasco Rauh  
Amsterdam University Medical Center
Amsterdam, Netherlands
Linda Douw  
Amsterdam University Medical Center
Amsterdam, Netherlands
Christiaan Vinkers  
Amsterdam University Medical Center
Amsterdam, Netherlands
Jeroen Geurts  
Amsterdam University Medical Center
Amsterdam, Netherlands
Menno Schoonheim  
Amsterdam University Medical Center
Amsterdam, Netherlands


Multiple sclerosis (MS) is a neurodegenerative and neuroinflammatory disease of the central nervous system. A large proportion of MS patients suffer from cognitive impairment (CI), where disorganization of the functional brain network plays a vital role. While the dynamic behavior of these networks has not been studied extensively in MS, networks appear "stuck" in particular brain states, with a reduction in overall state dynamics. However, especially little is known about the specific brain states MS patients might become "stuck" in. Therefore, this study aimed to investigate whether CI in MS is related to a reduction in the dynamics of specific states.


Resting-state functional MRI (rs-fMRI) was acquired from 332 patients with MS and 96 matched healthy controls (HCs). Seven cognitive domains were assessed with an expanded Brief Repeatable Battery of Neuropsychological tests, from which cognitive groups were defined (Mild CI, ≥2 tests Z<-1.5; severe CI, ≥2 tests Z<-2; preserved, CP). Physical disability was assessed using the expanded disability status scale. A sliding-window approach was used and for each window functional connectivity (FC) was calculated using Pearson correlations. Subsequently, k-means clustering was performed on all windows to identify recurrent patterns of FC, i.e. brain states. Next, meta-states were computed, which represented the distance to each state and, therewith, the position of the brain in state-space. Meta-state analysis computed how the functional network moves through state-space using the number of meta-state changes, number of meta-states visited, state-span, and total distance travelled. For state-specific alterations, fractional occupancy (FO) described the total time spent in each state, dwell-time (DT) represented the number of consecutive windows spent in each state, and the fraction of switches (FS) from other states to a particular state characterised the accessibility of a state. Linear mixed models were used to quantify differences between groups, cognitive relations focused on Bonferroni-corrected pairwise differences between MS patients with CI and CP. A multivariate regression model was performed in MS with backwards selection to identify which measures of state dynamics explain cognition best, beyond conventional measures of structural damage (atrophy, lesion load, and white-matter integrity). All analyses were corrected for age, sex, and education.


Of all patients, 46% showed CI (20% mild CI; 26% severe CI). Five network states were defined (Figure 1). CI patients showed a lower number of meta-states compared to CP patients and HCs (p=0.004; Figure 2). FO (p=0.002) and DT (p=0.005) was higher in CI (compared to CP) in state 2, a state featuring mostly within-network connectivity, but FS was not different between groups. Finally, a longer FO (p=0.010) and DT (p=0.002) in state 2 and a lower number of meta-states (p=0.019) related to worse CI even when controlling for structural damage. For disability, only the number of meta-states was a significant correlate (p=0.011).


These results indicate that CI in MS is related to a broad reduction in state dynamics of the functional brain network. CI patients showed a lower number of meta-states as well as more time spend in a specific state where brain regions mostly communicate to other regions within the same network. These regions communicate less with brain regions from other networks, which could hamper the global integration of information. In addition, time spent in this particular state related to cognition, even after correcting for structural damage, indicating the added clinical value of such network parameters. Although these results show that CI patients are especially "stuck" in a state of within-network communication, future work should elucidate whether these networks reduce their own dynamics or whether dynamics are directly restricted by pathology.

Disorders of the Nervous System:

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 2

Modeling and Analysis Methods:

Connectivity (eg. functional, effective, structural) 1
Task-Independent and Resting-State Analysis


Degenerative Disease
Other - Dynamic Connectivity

1|2Indicates the priority used for review
Supporting Image: Figure1.png
   ·Fig 1. Connectivity matrices of the five states based on all participants.
Supporting Image: Figure2.png
   ·Fig 2. Patients with cognitive impairment(CI) show a higher number of meta-states and fractional occupancy in state 2 than preserved(CP) patients and controls(HCs). *p<0.05,**p<0.01,***p<0.001,#CI≠CP

My abstract is being submitted as a Software Demonstration.


Please indicate below if your study was a "resting state" or "task-activation” study.

Resting state

Healthy subjects only or patients (note that patient studies may also involve healthy subjects):


Was any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.


Was any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.

Not applicable

Please indicate which methods were used in your research:

Functional MRI
Structural MRI
Neuropsychological testing

For human MRI, what field strength scanner do you use?


Which processing packages did you use for your study?


Provide references using author date format

d'Ambrosio, A., Valsasina, P., Gallo, A., De Stefano, N., Pareto, D., Barkhof, F., . . . Rocca, M. A. (2020). Reduced dynamics of functional connectivity and cognitive impairment in multiple sclerosis. Mult Scler, 26(4), 476-488.
Eijlers, A. J. C., Wink, A. M., Meijer, K. A., Douw, L., Geurts, J. J. G., & Schoonheim, M. M. (2019). Reduced Network Dynamics on Functional MRI Signals Cognitive Impairment in Multiple Sclerosis. Radiology, 292(2), 449-457.