ReHo is a measure to detect the similarity or synchronization of the time series of nearest neighboring voxels (usually 27 voxels) with the calculation of KCC. This approach was used with fMRI data at rest to investigate the regional homogeneity in CD. Compared with controls, CD patients showed higher ReHo in the right cerebellum crus I but lower ReHo in the right superior MPFC in DMN. Moreover, higher ReHo also existed in the right precentral gyrus, the right insula and the bilateral middle cingulate gyrus. Furthermore, a significant positive correlation was observed between the ReHo value in the right cerebellum crus I and the symptom severity.
By usage, the dystonia that covered the cervical dystonia has been attributed to the dysfunction of the basal ganglia. However, the cerebellum has been recently suggested as a probable originate region. Numerous of Findings from neuroimaging, neurophysiological and pathological studies with CD have shown abnormalities of cerebellar function, which provided strong evidence for the importance of the cerebellum in the pathophysiology of dystonia (Filip, Lungu, & Bares, 2013; Neychev, Gross, Lehericy, Hess, & Jinnah, 2011; Prudente, Hess, & Jinnah, 2014). However, functions of cerebellum in the occurrence of dystonia are still unclear. Traditionally, the cerebellum is identified as a region engaged in motor coordination. It plays a pivotal role in modulation of premotor, sensorimotor and posterior parietal regions for fine-tuning motor control. Pong et al. have demonstrated that the output of the cerebellum and the output of from the basal ganglia work together to participate in movement control of the head and face (Pong, Horn, & Gibson, 2008). Moreover, the cerebellum has been suggested as a processor of sensory information, integrating descending visual input from the parietal cortex and ascending input from the spinocerebellar pathway for promoting a forward model, as well as for predicting sensory consequences of an action (Wolpert, Goodbody, & Husain). Some studies have indicated that lesions on cerebellum can cause secondary cervical dystonia (Filip, Lungu, & Bares, 2013; LeDoux & Brady, 2003). Additionally, other studies have showed an abnormality of motor learning (Filip, Lungu, Shaw, Kasparek, & Bares, 2013) and eyeblink conditioning (Teo, van de Warrenburg, Schneider, Rothwell, & Bhatia, 2009) in cervical dystonia, suggesting that the function of the cerebellum is certainly impaired. Given these findings, we have reason to believe that the abnormality in the right cerebellum crus I with CD could influenced the accuration of the movement control of the head and face impaired because the abnormal modulation. Based on these, the result of a higher ReHo value in the right cerebellum crus I in CD patients here consistently indicated that the dysfunction of this region can be a key factor in the occurrence of motor symptoms in CD, which highlighted the importance of the cerebellum for motor modulation in the pathophysiology of cervical dystonia.
It is noteworthy that a significant positive correlation was observed between ReHo value in the right cerebellum crus I and symptom severity in the present study. Previously, an animal studies observed that pharmacological exciting of the cerebellum leads to dystonia, whereas cerebellar ablation causes ataxia (Pizoli, Jinnah, Billingsley, & Hess, 2002). These findings supported the conjecture that dystonia is caused by aberrant, distorted functional output rather than the loss of it (Jinnah & Hess, 2006). Based on these notions, higher ReHo value in the right cerebellum crus I causes more serious symptoms of dystonia. Overall, the occurrence of CD is likely associated with abnormal regulation of the cerebellum in sensorimotor processing.
Anatomically, MPFC composed of discrete and cytoarchitectonically areas receiving large-scale of sensory information from the external environment and the body (Su et al., 2014). Specifically, the lobule Crus I of the cerebellum closely associated with MPFC (Nitschke, Arp, Stavrou, Erdmann, & Heide, 2005), agrees with a prominent role of both regions in non-motor functions (Krienen & Buckner, 2009). Functionally, MPFC was proposed to be involved in many higher executive functions, involving in emotion, decision-making, goal-directed behavior, working memory and attention(Martin-Cortecero & Nunez, 2016). Therefore, we proposed that the decreased ReHo value of the right superior MPFC possibly influence this region’s function and result in losing top-down regulation, which is suggested as the foundation in the pathophysiology of changes of cognitive, emotional processing, and behavior in CD. In our other study, we have found patients with CD exhibit significantly decreased VMHC in superior MPFC (Jiang et al., 2019). Besides, changes in the cognitive processing of movement have been previously observed in idiopathic dystonia patients (Hebb et al., 2014; Little et al., 2013). In other studies, in addition to organization and execution of movement, aberrant motor cognition consisting of a mental rotation of body parts, temporal processing and alterations of movement, and body representation have been observed in CD (Delorme et al., 2016). These findings strengthen our conjecture.
Additionally, in animal study, MPFC was confirmed receiving afferent projections representing all sensory modalities, which originated from widespread areas of the cortex (and associated thalamic nuclei). The dorsal MPFC presumably integrated and utilized this information for goal directed actions(Hoover & Vertes, 2007). Thus, abnormal neuro-activity in MPFC may influence goal directed actions. On this basis, the “sensory trick”, which we suggested as a goal directed action, may have a correlation with the changes of MPFC. In the present study, the sensory trick phenomenon was observed in 18 patients, which was consistent with previous report (Filip, Sumec, Balaz, & Bares, 2016). These patients usually acquire attenuation of abnormal head movements by slightly touching a particular area of the face or head. It has previously been proposed to use sensory techniques to influence proprioceptive input to balance the inhibition ratio to facilitation (Ramos, Karp, & Hallett, 2014). Recently, the sensory trick was suggested as a modulation of abnormal connections between sensory input and motor output (Filip et al., 2016). The inconsistent standpoint renders the mechanism latent the sensory trick phenomenon still intricate. Our findings probably supported the latter. Consequently, we suggested MPFC involved in the mechanism of the sensory trick.
Furthermore, the superior MPFC partially overlapping with dorsal MPFC is suggested as a critical region in dorsal cognitive system and default mode network (DMN) (Bechara, Damasio, Damasio, & Anderson, 1994; Bechara, Damasio, Tranel, & Damasio, 1997). Meanwhile, The cerebellum was suggested as a region belonging to DMN by Habas et al. (Habas et al., 2009). Moreover, the cerebellum builds reciprocal connection with brain regions belonging to DMN comprising the anterior cingulate and the prefrontal cortex (Vilensky & van Hoesen, 1981). This network exhibits high levels of activity and high degrees of functional connectivity at rest, but deactivated during specific goal-directed behavior is required (Mohan et al., 2016). DMN has been well-detected and showed abnormal neuro-activities in several neurological and neuropsychiatric diseases, for instance Parkinson’s disease, Alzheimer’s disease, epilepsy, schizophrenia (W. Guo, Xiao, et al., 2014; W. Guo, Yao, et al., 2014; Mohan et al., 2016) and affective disorders (e.g., major depressive disorder (W. Guo, Liu, et al., 2014)). Consistent with above results, the present research hence extends the previous findings by offering new evidence that the abnormal regional activity of MPFC possible relevant to the pathophysiology and highlight the importance of DMN for cognitive function in CD. DMN abnormalities may be involved in identifying abnormal posture in CD.
Increasing evidence emphasizes dystonia as a disorder of motor organization, programming, sensorimotor, and execution (Delorme et al., 2016). the primary motor cortex (as the M1), that is the precentral gyrus, received projections from BA2 and BA5 areas that contain the contralateral cutaneous, muscular and articular information, and subsequently corrected movement. Meanwhile, as the target of projections from frontal cortical regions and subcortical regions, precentral gyrus is a probable site converging mechanisms of the selection, initiation and inhibition movement (Stinear, Coxon, & Byblow, 2009). This connectional architecture induced the precentral gyrus to play a crucial role in the mechanism of generated CD. The abnormality in motor system physiology of patients with dystonia was exhibited by reduced surround inhibition resulting in unnecessary contractions of more muscles than what is required for specified motor behavior (Delnooz et al., 2013). Therefore, the increased ReHo in the right precentral gyrus in CD patients could lead to impaired selection, initiation or inhibition of movement though impairing the cortices - basal ganglia – cortices circuit. In other words, it was in line with the notion that the cervical dystonia has been attributed to not only the dysfunction of the basal ganglia but also the cortices. In our other study, patients with CD showed abnormal activities with decreasing GFC in the the M1-SMA motor network, including right supplementary motor area and right precentral gyrus. Moreover, the GFC values in the right precentral gyrus of CD patients was significantly negative correlated to the symptomatic severity (Pan et al., 2020). Hence, abnormal regional homogeneity in this region is important for the pathogenesis of CD.
The right insula and the bilateral middle cingulate gyrus exhibited an increased ReHo in CD patients. The insula participated in processing sensory/motor information, pain, visceral motor/sensory, olfactory, visual, vestibular/auditory, and verbal information and modulating attention, emotion, and inputs related to music and eating (Cauda et al., 2011). The middle cingulate gyrus, a crucial node of the imbic system, has been implicated with environmental monitoring, response selection and skeletomotor body orientation. Meanwhile, a system included the entire insula and the middle cingulate cortex, which possibly participated in these aforementioned functions (Taylor, Seminowicz, & Davis, 2009). Furthermore, the insular and cingulate cortex play crucial roles in integrating multimodal information significant for several functions, such as sensorimotor, allostatic/homeostatic, emotional, and cognitive functions (Taylor et al., 2009). Moreover, both regions have been confirmed to be related to pain perception (Peyron, Laurent, & Garcia-Larrea, 2000). In the present study, a total of 17 patients reported painful neck muscles. The increased ReHo in the right insula and the bilateral middle cingulate gyrus suggested that these brain regions may participate in the processing of CD-related pain and induce an abnormal sensorimotor function in CD patients. The observation of a majority of CD patients with moderate/severe neck pain from a real-world clinical registry (Charles et al., 2016) partially supported our conjecture. In masses of previous findings, CD was linked to multiple brain regions but rarely to the insula and the cingulate gyrus. Thus, our findings extended previous understandings that the these brain regions may also associated with the pathophysiology underlying CD.
We observed asymmetric activity patterns in CD that were strongly involved in the right-hemispheric dystonia-related connectivity pattern. This phenomenon has been previously reported, for instance, during finger movements (de Vries et al., 2008), and at resting-state (Delnooz et al., 2015). It was possibly associated to the larger number of contralateral-side muscles affected patients. However, patients in our study were primary bilateral-side muscles affected. A possible explanation for the laterality of the right-hemispheric may be due to its dominance of position control, that is the right hemisphere determine response modification so that it is dominant for position control (Haaland, Prestopnik, Knight, & Lee, 2004).
Aside from the small sample size, several limitations must be stated in this study. First, all of the patients’ dystonic posturing was minimal in the supine position or absent during scanning. To confirm whether this is a specific sensory trick is difficult. Therefore, influences of sensory trick could not be easily eliminated. Besides, physiological noises including heart rhythm and respiratory cannot be completely eliminated even though a relatively low sampling rate (TR = 2 s) is used. Lastly, with a small sample size, patients were not subdivided further into different groups according to head rotation. Hence, research using a larger sample size is required to expand our results.