Disrupted topological organization of the default mode network in mild cognitive impairment with subsyndromal

Yang Du1,2, Jing Nie1,2, Jian-Ye Zhang3

  • 1Department of Geriatric Psychiatry, Shanghai Mental Health Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China.

PubMed
Abstract

Insights

Subsyndromal depression in mild cognitive impairment disrupts the default mode network

Area of Science:

  • Neuroscience
  • Cognitive Science
  • Psychiatry

Background:

  • Subsyndromal depression (SSD) is prevalent in mild cognitive impairment (MCI).
  • The neural underpinnings of MCI with SSD (MCID) remain poorly understood.
  • The default mode network (DMN) is implicated in cognition and depression.

Purpose of the Study:

  • To investigate the topological organization of the DMN in patients with MCID.
  • To explore the relationship between DMN topology, depressive symptoms, and cognitive function in MCID.
  • To assess the potential of DMN topological metrics as biomarkers for MCID.

Main Methods:

  • Utilized graph theory to analyze resting-state functional connectivity of the DMN in 42 MCID patients, 34 MCI without SSD (MCIND) patients, and 36 healthy controls (HCs).
  • Performed correlation analyses between DMN network metrics, depressive symptoms, and cognitive scores.
  • Developed and validated Support Vector Machine (SVM) models using DMN topological metrics to differentiate MCID from MCIND.

Main Results:

  • MCID patients exhibited significantly reduced global and nodal efficiency in the left anterior medial prefrontal cortex (aMPFC) compared to MCIND patients.
  • In MCID, DMN small-worldness and global efficiency negatively correlated with depressive symptom severity.
  • Nodal efficiency in the left lateral temporal cortex and left aMPFC positively correlated with cognitive function in MCID.
  • The SVM model achieved high accuracy (0.83) in distinguishing MCID from MCIND.

Conclusions:

  • The co-occurrence of MCI and SSD is associated with significant disruptions in DMN topological organization.
  • DMN topological metrics can effectively differentiate MCID from MCIND.
  • These network metrics show promise as biomarkers for distinct clinical presentations of MCI.