Default Mode Network Structural Integrity and Cerebellar Connectivity Predict Information Processing Speed Deficit in Multiple Sclerosis

Abstract

Cognitive impairment affects about 50% of multiple sclerosis (MS) patients, but the mechanisms underlying this remain unclear. The default mode network (DMN) has been linked with cognition, but in MS its role is still poorly understood. Moreover, within an extended DMN network including the cerebellum (CBL-DMN), the contribution of cortico-cerebellar connectivity to MS cognitive performance remains unexplored. The present study investigated associations of DMN and CBL-DMN structural connectivity with cognitive processing speed in MS, in both cognitively impaired (CIMS) and cognitively preserved (CPMS) MS patients. 68 MS patients and 22 healthy controls (HCs) completed a symbol digit modalities test (SDMT) and had 3T brain magnetic resonance imaging (MRI) scans that included a diffusion weighted imaging protocol. DMN and CBL-DMN tracts were reconstructed with probabilistic tractography. These networks (DMN and CBL-DMN) and the cortico-cerebellar tracts alone were modeled using a graph theoretical approach with fractional anisotropy (FA) as the weighting factor. Brain parenchymal fraction (BPF) was also calculated. In CIMS SDMT scores strongly correlated with the FA-weighted global efficiency (GE) of the network [GE(CBL-DMN): ρ = 0.87, R ² = 0.76, p < 0.001; GE(DMN): ρ = 0.82, R ² = 0.67, p < 0.001; GE(CBL): ρ = 0.80, R ² = 0.64, p < 0.001]. In CPMS the correlation between these measures was significantly lower [GE(CBL-DMN): ρ = 0.51, R ² = 0.26, p < 0.001; GE(DMN): ρ = 0.48, R ² = 0.23, p = 0.001; GE(CBL): ρ = 0.52, R ² = 0.27, p < 0.001] and SDMT scores correlated most with BPF (ρ = 0.57, R ² = 0.33, p < 0.001). In a multivariable regression model where SDMT was the independent variable, FA-weighted GE was the only significant explanatory variable in CIMS, while in CPMS BPF and expanded disability status scale were significant. No significant correlation was found in HC between SDMT scores, MRI or network measures. DMN structural GE is related to cognitive performance in MS, and results of CBL-DMN suggest that the cerebellum structural connectivity to the DMN plays an important role in information processing speed decline.

Keywords: cerebellum; connectomics; default mode network (DMN); diffusion weighted imaging (DWI); magnetic resonance imaging (MRI); multiple sclerosis (MS); symbol digit modalities test (SDMT); tractography.

FIGURE 1

Default mode network (DMN) nodes. Medial frontal cortex (orange), angular gyri (yellow), precuneus/posterior cingulate cortex (green), middle temporal gyri (blue), cerebellum (red).

FIGURE 2

Connectivity matrix of the DMN (A), CBL (B), and CBL-DMN network (C). Symmetric entries of the matrix are shown in yellow and they represent bi-directional bundles of fibers linking network nodes like those tracts connecting DMN regions in the brain cortex. Asymmetric entries are shown in orange and they represent bundles of fibers where the signal is mono-directional like those tracts projecting from the cerebellum to the cerebral cortex. In gray are shown anatomically non-existent connections between nodes and existent connections which are neglected in a specific model, like connections between cortical regions in the model of the cortico-cerebellar loops (B); all of these are numerically represented by zero-elements. The elements along the principal diagonal that would represent self-connections are displayed in white. L/R = left/right. MFC, medial frontal cortex; AG, angular gyrus; PCC, precuneous/posterior cingulate cortex; MTG, middle temporal gyrus; CBL, cerebellum.

FIGURE 3

Tractography results from HC were registered to MNI152 standard space at 1 mm resolution to create a population map for each tract. Here are shown the superimposed population maps of the tracts connecting the left hemisphere of the cerebellum to the DMN regions of the right cerebral cortex. On the left are shown cerebello-thlamo-cortical tracts, while on the right are shown cortico-ponto-cerebellar tracts. The voxel color (from dark red to yellow) represents the frequency of occurrence in the specific tract. Light blue regions identify the most consistent part of the tracts, which are common to at least 50% of HC. In the unthresholded population maps, it can be noted that these tracts originate from (cerebello-thalamo-cortical tracts) and project to (cortico-ponto-cerebellar tracts) the posterior lobe of the cerebellum that is associated to cognitive functions.

FIGURE 4

The population map of each tract is thresholded at 50% to select its most consistent part across subjects and to eliminate spurious streamlines. Here are shown all 32 resulting tract masks in MNI 1 mm standard space. These tract masks were subsequently registered to MS patients to compute tract-averaged diffusion FA. Tracts connecting cerebral nodes are displayed in blue, while cortico-cerebellar connections are displayed in red.

FIGURE 5

Boxplots representing the summary of clinical scores in the different groups of subjects. Statistically significant differences between groups are indicated with ^∗p < 0.05 and ^∗∗p < 0.01 according to results obtained with ANOVA post hoc tests or T-tests according to the specific case (see section “Statistical Analysis” and Table 2). It can be observed that SDMT and depression (HADS-D) are significantly different in MS patients and HC. EDSS and, to a lesser extent, disease duration and NART can discriminate between CIMS and CPMS.

FIGURE 6

Boxplots representing the summary of MRI metrics in each group of subjects. Statistically significant differences between groups are indicated with ^∗p < 0.05 and ^∗∗p < 0.01 according to results obtained with ANOVA post hoc tests or T-tests according to the specific case (see section “Statistical Analysis” and Table 2). It can be observed that all measures can distinguish between MS patients and HC. BPF is significantly reduced in CIMS with respect to CPMS. Despite the absence of statistical significance, a similar trend can be observed also for GE and WM-FA.

FIGURE 7

Scatter plots representing SDMT vs. GE (CBL-DMN) for HC (left), CPMS (center), and CIMS (right). Solid black lines represent the linear regression model fit, while dashed red lines represent 95% confidence intervals. It is noteworthy that GE (CBL-DMN) predicts SDMT performance progressively better going from HC (no significant correlation) to CPMS (ρ = 0.51, p < 0.001) to CIMS (ρ = 0.87, p < 0.001). The difference between correlation coefficients for CPMS and CIMS is statistically significant (p < 0.01), as verified by applying Fisher z-transformation.