Altered motor network dynamics in myoclonus-dystonia

The exact mechanisms underlying myoclonus-dystonia (M-D) remain unknown, although the basal ganglia-thalamo-cortical (BGTC) and cerebello-thalamo-cortical (CTC) networks are hypothesized to be involved. We aimed to investigate the static and dynamic features of networks related to motor control during rest in M-D patients using functional magnetic resonance imaging (fMRI). Resting-state fMRI data from 19 M-D patients and 19 healthy volunteers were analysed. Symptom severity was measured using the Clinical Global Impression-Severity Scale, anxiety and depression with the Hospital Anxiety and Depression Scale and cognitive impairment with the Montreal Cognitive Assessment. Independent component analysis was used to identify brain components corresponding to the BGTC and CTC networks. Static within-network (i.e. spatial contribution of voxels to the average network signal time course) and between-network (i.e. correlation between network time courses) functional connectivity were examined. We additionally performed a dynamic functional connectivity analysis focused on recurrent connectivity patterns (i.e. brain states) using k-means clustering of windowed functional connectivity correlation matrices, whereby we identified three prototypical dynamic connectivity states in the BGTC and CTC circuits. The following brain state summary measures were computed: fraction of time spent in a state, dwell time, number of state transitions and number of state visits. Static analysis revealed that patients with M-D showed increased functional connectivity (FC) of the left supramarginal gyrus within a cognitive control network (corresponding to the cortical part of BGTC and CTC circuits), suggesting stronger integration of this area in this network (within-network P _fdr < 0.05). There were no significant group differences in between-network functional connectivity. Dynamic analysis revealed three dynamic connectivity states in BGTC and CTC circuits. Patients with M-D engaged less in a state characterized by high segregation of sensorimotor from basal ganglia and cerebellar domains, with high connectivity between networks within these domains. This finding may reflect reduced basal ganglia and cerebellar contributions during motor preparation in M-D. We found no relationships between fMRI findings and motor symptom severity in M-D patients. Our findings provide further evidence for disrupted brain network dynamics in BGTC and CTC circuits in M-D patients, supporting the hypothesis of compromised basal ganglia and cerebellar functioning. Particularly, the association of cerebellar and anterior parietal alterations is proposed to reflect impaired sensory prediction in feed-forward motor planning, potentially underlying irregular 'non-prepared movements'.