Objective: To study functional connectivity changes using graph theory and their association with cognitive evolution in a sample of non-demented Parkinson’s disease (PD) and matched healthy controls (HC) after a four-year follow-up.

Background: Previous studies demonstrated that cognitive deficits in PD are associated with alterations in different parameters derived from resting-state functional connectivity. Evidence regarding the relationship between connectivity changes and cognitive evolution is nonetheless very scarce.

Methods: A final sample of 27 HC and 45 matched PD patients, assessed twice at an interval of approximately four years, was included. Resting-state fMRI was used for functional connectivity analysis. Connectivity matrices, obtained after parcellating the brain into 246 regions using the Brainnetome atlas, were used to compute graph-theory parameters that describe network organization (path length, clustering coefficient, modularity, small-world coefficient). For cognitive assessment, a thorough neuropsychological battery was used, and tests were subsequently grouped into four cognitive domains (attention/working memory, executive function, memory, and visuospatial/perceptual function). A repeated-measures general linear model was used to test group and time effects, as well as group-by-time interactions.

Results: Analysis of cognitive data in patients showed a significant effect of time for executive function (p=.004) and visuospatial/perceptual scores (p=.015). Significant effects of group (p=.034, patients performing worse than controls), time (p=.006), and group-by-time interaction (p=0.027) were found. Connectivity parameters showed significant effects of group for clustering coefficients and modularity (p=.023 and p=.045, respectively, with patients showing higher values). For path lengths, a significant effect of time was found (p=.022, both groups showing increasing values). When performing correlations between longitudinal cognitive and connectivity changes, we found that differences in executive function scores correlated significantly and positively with differences in clustering coefficients (r=0.404, p=0.007), modularity (r=0.384, p=0.011), and small-world coefficients (r=0.335, p=0.028).

Conclusions: Our findings indicate that functional connectivity changes explain executive function in PD patient’s evolution.

« Back to 2018 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/resting-state-connectivity-and-cognitive-changes-in-parkinsons-disease-a-four-year-follow-up-study/