In our study, we reported that CMB, particularly located in the frontal and temporal lobes, basal ganglia and infratentorial, were associated with gait dysfunction and balance problems, regardless of the presence of other CSVD categories. A novel finding was that CMBs located in the lobes (frontal, parietal and temporal) and basal ganglia correlated with motor disorder of the upper extremity. The results indicate that CMB can impair upper and lower extremity functions, and the analysis of different regions reveals the possible pathogenesis of CMBs. This type of CSVD needs to be identified and managed appropriately in the treatment of movement disorders.
In the previous studies, the relationships between individual CSVD imaging markers and gait performance remained controversial[3, 8, 12, 19]. The association between WMH and lacunes and gait has consistently been reported[12, 20-22]. However, most studies found no correlation between CMBs and motor performance[3, 12, 13]. As we known, just a few research studies have shown that the presence of CMB magnifies the effects of the volume of WMH on gait but not on postural stability[23]. In studies focusing specifically on CMBs and movement disorders, it was observed that the presence of CMBs were associated with a slower walking speed in the Leiden Longevity study [24] and with a longer stride time and a longer stance phase in the Taizhou study[25]. According to another study, a higher number of CMB, measured by T2*-weighted gradient-echo sequences, was associated with shorter stride length and worse performance on Tinetti and the Timed-Up-and-Go tests[11]. In our research, we discovered that the number of CMBs had correlations with both gait and balance disorders. In order to examine the specific reasons for this, we reviewed the previous studies and identified several differences which need to be addressed. First, the incidence of CMBs in previous studies was lower than in our study, which may affect the research about the location of CMBs. Second, some studies detected CMBs using GRE, which was not as sensitive in terms of number and distribution as SWI. Third, CMBs in most patients are not severe enough to have a functional impact, so just looking at general gait parameters are not sufficient to detect early changes in gait performance of people with CMBs. During our study, we measured gait function with both quantitative and semi-quantitative methods which were more comprehensive to detect gait disorders. The SPPB test measured gait and balance, providing more validity, reliability and responsiveness to assess physical function, increasing the accuracy between CMBs and motor function[26].
Comparatively to lower extremity, the association between CMBs and upper extremity has never been investigated. In our study, we demonstrated that CMBs in the frontal, parietal, temporal and basal ganglia, were associated with the function of pronation-supination. As the population ages, executive function plays an increasing role in the ability of individuals to function independently. Upper extremity function, as an important part of it, has been shown to decrease the ability of manual tasks to be completed[27, 28]. Many studies have shown that people with mild cognitive impairment have a greater decline in finger dexterity[29]. And a further study from the RISE study indicated participants with mild cognition impairment had significantly effects on reactions times and pronation-supination of the hands[30]. As one of the important causes of cognitive impairment, CSVD may also have an important impact on hand function. From now on, investigations between the upper extremity and CSVD are still rare, only two of them had a positive outcome. One of the studies involving 30 individuals from the Austrian Stroke Prevention Study, Purdue's Pegboard Test was used to assess upper extremity function. The results showed a positive correlation between finger movement and increasing WMH in the frontal lobe, but the correlation was not significant[2]. Another study found that WMH was associated with motor deficits in pronation–supination, as well as brain atrophy was associated with both pronation–supination and finger-tapping, but the other kinds of CSVD were not adjusted[8, 13]. In this study, we examined the relationship between CMB and upper extremity function independently and expanded the sample size appropriately.
CMBs seem to play an important role in motor function, but the mechanisms are still not well understood. Multisensory information such as somatosensory, visual and vestibular sensation acts on various areas of the brain so that adaptable posture-gait control can be achieved. To walk in unfamiliar circumstances requires cognitive control of posture that depends on self-body knowledge such as body schema and body motion in space[31]. The CMBs were found in many parts of the brain, and they were able to damage the functional areas in the brain. As with our results, CMBs in specific areas (frontal, temporal, basal ganglia, and infratentorial) had a greater effect on gait. According to De Laat's study, CMBs especially located in the frontal lobe and basal ganglia (and thalamus) interfered with gait, an area that controls movement in the lower extremities[11]. The results of another study, looking at cerebral function patterns in CSVD patients, indicated that lesions in the sensorimotor network and frontoparietal network, such as the left supplementary motor area, left superior parietal gyrus, and right inferior frontal gyrus, were associated with gait disorder[32]. Our study also found that lesions in the temporal lobe and infratentorial had a correlation with gait and balance disturbances. The temporal lobe has many neuronal networks which can connect the visual and vestibular signals and control the gait performance and keep balance[33]. At the same time, the cerebellum and brainstem are also important regions in the control of balance in the body.
The blood-brain barrier (BBB) dysfunction and endothelial injury play an important role in the pathogenesis of CMBs, and the pathological changes appeared not only in CMBs, but also in other types of CSVD [34], and quantity and distribution of CMBs were closely related to lacunar infarction and WMH have been noted[35]. BBB dysfunction is considered one of the leading causes of WMH, and extravasation of intravascular substances may result in diffuse brain tissue damage and nerve cell death, thereby contributing to motor dysfunction. This could partially explain the origin of motor dysfunction caused by CMB.
Takakusaki pointed out that upper extremities had sophisticated abilities depending on integrated visual, somatosensory, and action systems of motor cortex, as well as its subcortical connections with brainstem and cerebellum[36]. The process of pronation–supination involves hand movements with visuospatial and coordination components, strength, and speed, may require the overall function of the intact nervous system. In agreement to our findings, CMBs involved multiple brain regions that controlled motor, sensory, and visual functions associated with hand movements.
The strengths of our study include the use of high-resolution MRI, evaluating both of number and location of CMBs, improved accuracy of imaging markers of CMBs. Another strength was the exploring of relationships between CMBs and motor performances in lower and upper extremities which were rarely discussed. There are limitations in our study. First, its cross-sectional design precluded assumptions of causality. Longitudinal studies are needed to further explore this and to finally get insight into the optimal timing with respect to prevention and disease progression. Second, although we used the more accurate MRI, visual ratings of CSVD imaging might have introduced some potential errors, for example, misclassification, which could not be overruled. Third, the measuring equipment was not accurate enough which might induce the error of movement data and more advanced equipment should be used. Finally, the inclusion of patients only come from one hospital which limited the generalizability of our results.