Increased default-mode network centrality in cognitively impaired multiple sclerosis patients

Research output: Contribution to journalArticleAcademicpeer-review

74 Citations (Scopus)


OBJECTIVE: To investigate how changes in functional network hierarchy determine cognitive impairment in multiple sclerosis (MS).

METHODS: A cohort consisting of 332 patients with MS (age 48.1 ± 11.0 years, symptom duration 14.6 ± 8.4 years) and 96 healthy controls (HCs; age 45.9 ± 10.4 years) underwent structural MRI, fMRI, and extensive neuropsychological testing. Patients were divided into 3 groups: cognitively impaired (CI; n = 87), mildly cognitively impaired (MCI; n = 65), and cognitively preserved (CP; n = 180). The functional importance of brain regions was quantified with degree centrality, the average strength of the functional connections of a brain region with the rest of the brain, and eigenvector centrality, which adds to this concept by adding additional weight to connections with brain hubs because these are known to be especially important. Centrality values were calculated for each gray matter voxel based on resting-state fMRI data, registered to standard space. Group differences were assessed with a cluster-wise permutation-based method corrected for age, sex, and education.

RESULTS: CI patients demonstrated widespread centrality increases compared to both HCs and CP patients, mainly in regions making up the default-mode network. Centrality decreases were similar in all patient groups compared to HCs, mainly in occipital and sensorimotor areas. Results were robust across centrality measures.

CONCLUSIONS: Patients with MS with cognitive impairment show hallmark alterations in functional network hierarchy with increased relative importance (centrality) of the default-mode network.

Original languageEnglish
Pages (from-to)952-960
Number of pages9
Issue number10
Publication statusPublished - 7 Mar 2017


  • Adult
  • Aged
  • Brain
  • Cognition Disorders
  • Cohort Studies
  • Female
  • Humans
  • Image Processing, Computer-Assisted
  • Journal Article
  • Magnetic Resonance Imaging
  • Male
  • Middle Aged
  • Multiple Sclerosis
  • Neural Networks (Computer)
  • Neural Pathways
  • Neuropsychological Tests

Cite this