Gait and balance impairments in ET have been described in numerous studies. These problems are not simply sub-clinical phenomena, observed only in the laboratory [7, 10, 12]. In some patients, the problems have the potential to increase falls and fear of falls [11, 12]. The presence of specific gait impairments (including decreased speed and impaired dynamic balance) suggests that these impairments may have a cerebellar origin [7, 8, 24–27].
Cerebellar dysfunction is suggested to result in impairments in the timing control of movements [18]. In ET, performance of motor timing tasks, such as repetitive finger tapping, is impaired [21, 22, 28]. In particular, finger-tapping movements are slower [21, 22] and more variable compared with controls [22]. However, these motor impairments (slow and variable movements) were not correlated with clinical markers of tremor severity [21, 22]. Target interception using a key press movement was also impaired in ET patients with head tremor, suggesting that the cerebellum may be implicated in predictive timing of discrete movements [23].
In the present study, we examined whether motor timing was impaired during gait in a large sample of 215 subjects, which included 155 ET subjects. The results of our study show that timing control during gait is not impaired in ET in comparison with matched controls. Cadence (step frequency) was lower in ET patients compared with controls while producing and reproducing criterion cadences (step frequencies). Lower cadence in patients with ET may be a function of slower gait speed, which has been previously reported [7, 29]. However, cadence error was similar across the two groups, indicating that both groups were comparably accurate at producing and reproducing criterion cadence during gait. Similarly, cadence standard deviation (a measure of precision) was also similar in ET patients and controls. Analysis of step time and step time variability yielded similar results: patients with ET performed similar to controls. Our results, in the context of previous studies on motor timing deficits in ET, suggest that timing control impairment in ET may be specific to some activities but not others.
Why is timing control in ET impaired during finger tapping but not during gait? One explanation is that discrete motor tasks such as finger tapping may have distinct control mechanisms in comparison with continuous motor tasks such as circle drawing and gait [30–32]. Support for this hypothesis comes from results showing that timing variability in discrete movements, such as finger tapping, is not correlated with timing variability during circle drawing movements [30–32] or with variability during gait [33]. Additional support for this hypothesis comes from imaging studies [34, 35]. Functional imaging of healthy subjects while performing discrete and continuous movements indicates that the cerebellar vermis [35], the dentate nucleus and cerebellar homunculus [34] are selectively active during discrete rather than continuous movements. Control mechanisms for discrete movements may include an explicit timing representation. In contrast, timing in continuous movements is thought to be an emergent property of movement control [32].
Patients with cerebellar lesions have deficits in timing of discrete (discontinuous) movements (such as finger tapping) but not continuous movements (such as circle drawing and gait) [31, 36]. These results suggest that the cerebellum may be involved in timing control of discrete movements that require an explicit timing representation. In contrast, timing deficits in continuous movements are pronounced in patients with basal ganglia disease such as Parkinson’s disease (PD) and Huntington’s disease (HD) [37–40].
The results of our regression analysis suggest that cranial tremor (tremor in midline structures) was associated with gait measures (cadence and cadence error). Postural instability (measured by the retropulsion test) was also associated with gait measures. In contrast, kinetic tremor and intention tremor (tremor in the extremities) were not associated with gait measures. These results are similar to our prior work and suggest that gait impairments may be related to pathology in midline cerebellar structures which may be implicated in head/neck control and postural stability [9]. In our results, presence of neck tremor was not associated with gait. The results of this study, along with our prior work [7, 9–12, 41] indicate that gait impairments in ET primarily include deficits in speed and balance control, whereas timing control during gait is spared. Future work should compare timing control in discontinuous tasks (such as finger tapping) with continuous tasks (such as gait or circle drawing) in order to further understand the role of the cerebellum in timing control of movements. In addition, imaging of the cerebellum during observation or imagery of gait may be important in further elucidating the role of the cerebellum during gait. Recent imaging studies of motor imagery or action observation of gait have shown that similar cortical and sub-cortical regions are active during performance of gait and observation or imagery of gait [42, 43]. However, results of imaging of observation or imagery of gait must be interpreted with caution because imaging or observation of gait do not involve balance and postural control, which is an inherent aspect in the control of functional gait.
One potential limitation is that our study tested production and reproduction of two criterion cadences. Future work should test production and reproduction at additional criterion cadences in order to generalize our findings. Second, we did not administer a test of timing control in a discrete task such as finger tapping. Given the extensive battery of clinical assessments (~ 3Â h), we limited the quantitative motor assessment in order to minimize subject fatigue. The strengths of our study include a large sample size and uniform administration of assessments across participants.