Increased thalamic centrality and putamen-thalamic connectivity in patients with parkinsonian resting tremor

Abstract

Introduction: Evidence has indicated a strong association between hyperactivity in the cerebello-thalamo-motor cortical loop and resting tremor in Parkinson's disease (PD). Within this loop, the thalamus serves as a central hub based on its structural centrality in the generation of resting tremor. To study whether this thalamic abnormality leads to an alteration at the whole-brain level, our study investigated the role of the thalamus in patients with parkinsonian resting tremor in a large-scale brain network context.

Methods: Forty-one patients with PD (22 with resting tremor, TP and 19 without resting tremor, NTP) and 45 healthy controls (HC) were included in this resting-state functional MRI study. Graph theory-based network analysis was performed to examine the centrality measures of bilateral thalami across the three groups. To further provide evidence to the central role of the thalamus in parkinsonian resting tremor, the seed-based functional connectivity analysis was then used to quantify the functional interactions between the basal ganglia and the thalamus.

Results: Compared with the HC group, patients with the TP group exhibited increased degree centrality (p < .04), betweenness centrality (p < .01), and participation coefficient (p < .01) in the bilateral thalami. Two of these alterations (degree centrality and participation coefficient) were significantly correlated with tremor severity, especially in the left hemisphere (p < .02). The modular analysis showed that the TP group had more intermodular connections between the thalamus and the regions within the cerebello-thalamo-motor cortical loop. Furthermore, the data revealed significantly enhanced functional connectivity between the putamen and the thalamus in the TP group (p = .027 corrected for family-wise error).

Conclusions: These findings suggest increased thalamic centrality as a potential tremor-specific imaging measure for PD, and provide evidence for the altered putamen-thalamic interaction in patients with resting tremor.

Keywords: Parkinson's disease; functional MRI; functional connectivity; graph theory; resting tremor.

Figure 1

Error plots for centrality metrics of bilateral thalami in patients and controls. The centrality measures were calculated with 31 network densities ranging from 0.10 to 0.40 (x‐axis). For each density, the central bands indicate the mean and the error bars indicate the standard errors. Patients with resting tremor (TP group, in green) showed significantly higher values in degree, betweenness, and participation coefficients compared to healthy controls (HC group, in blue); in contrast, no significant differences were shown for these metrics between patients without resting tremor (NTP, in orange) and healthy controls

Figure 2

Network illustration for centrality alterations of bilateral thalami in patients with resting tremor. The group networks were calculated by averaging the correlation matrices between all subjects in the respective groups, which were thresholded at a density of 0.30. (A) In the healthy control group (HC), the bilateral thalami (in red) were exclusively connected with the nodes in the same module (bilateral pallida, in green). (B) In patients with resting tremor (TP), there were significantly more connections between the bilateral thalami and nodes in other modules (in blue), especially those involved in the cerebello‐thalamo‐motor cortical loop, including the cerebellum, supplementary motor area, paracentral lobule, and middle frontal gyrus. Details of the nodes and links presented in this figure are listed in Table 3

Figure 3

Correlations between centrality measures of the bilateral thalami and UPDRS tremor scores (A: resting tremor scores, B: total tremor scores). In patients with resting tremor (TP, in green), the degree of bilateral thalami and the participation coefficients for the left thalamus were significantly correlated with both tremor scores, and the participation coefficients for the right thalamus showed trend effects. No correlations were shown for patients without resting tremor (NTP, in orange). The solid fitted lines indicate significant effects, and the dashed fitted lines indicate insignificant effects

Figure 4

Group‐level contrast map with the putamen as the seed region Left: the group of patients with resting tremor (TP) showed significantly higher putamen–thalamic connectivity compared to the HC group (small‐volume corrected P_FWE = 0.027). Middle: patients with resting tremor (TP, in green) showed higher putamen–thalamic connectivity compared to healthy controls (HC, in blue). Right: the altered connectivity was significantly correlated with UPDRS Part II scores (motor assessment of daily life) in the TP group. The statistical significance was set at p < .05