A recent study combining voxel-based morphometry, tractography and resting-state functional MRI suggests that a primary cerebellar defect leads to the emergence of a pathological oscillation which sets the tremor frequency, but the clinical manifestation of tremor is dependent on the cortical output [ 68 ]. Available electrophysiological studies demonstrate that patients with dystonia and tremor had reduced reciprocal inhibition between agonist and antagonist of upper limb muscles, a lack of brainstem interneuronal inhibition BRrc , and abnormal sensory integration TDT , indicating a lack of inhibitory mechanism at multiple levels spinal, brainstem, and cortical [ 31 ].
The neurophysiologic abnormalities in patients with dystonia and tremor resemble those in dystonia but differ from those described in ET, indicating tremor as phenotypic feature of dystonia.
It has further been hypothesized that tremor in dystonia was caused by distorted cerebellar output due to abnormal burst firing pattern in Purkinje cells [ 31 ]. A series of epidemiological studies have yielded variable tremor prevalence among PD, ET and dystonia. These discrepancies may be partly due to sample selection as well as different definitions for rest tremor in parkinsonian tremor, ET and dystonic tremor and their subtypes. A diagnostic challenge comes from patients with mRT.
Besides key clinical phenotypic differences and DAT scan, several transducer-based techniques like accelerometry, gyroscopy, EMG, and digitizing tablet-based meaures may provide extra clues for the distinction [ 70 ]. Compared to rating scales, these transducers are far more sensitive to changes in tremor amplitude and frequency. However, due to the natural variability of tremor, they are not more sensitive in defining the minimal detectable change than rating scales [ 70 ].
Also, their potential diagnostic values sensitivity and specificity still merit further validation in larger cohort studies. More studies on the pathophysiology of the different tremor entities are needed, which may help to develop new diagnostic markers and hence a more tailored therapeutic strategy.
Studies on the natural course and biological basis of tremor are still warranted with standardized terminology, diagnostic criteria, validated evaluation tools and research protocols. Consensus statement of the movement disorder society on tremor.
Google Scholar. Common misdiagnosis of a common neurological disorder: how are we misdiagnosing essential tremor? Arch Neurol. Article PubMed Google Scholar. Tremor entities and their classification: an update. Curr Opin Neurol. Clinical correlates of action tremor in Parkinson disease. Eur J Neurol. The phenomenology of parkinsonian tremor. Parkinsonism Relat Disord. Provocation of Parkinsonian tremor. Rest tremor suppression may separate essential from parkinsonian rest tremor.
J Neurol Neurosurg Psychiatry. J Neural Transm. The Parkinson study group. Resting tremor in Parkinson disease: a negative predictor of levodopa-induced dyskinesia. How common is the most common adult movement disorder? Intention tremor in essential tremor: Prevalence and association with disease duration.
Prevalence and correlates of rest tremor in essential tremor: cross-sectional survey of patients across four distinct cohorts. Testing for alcohol sensitivity of tremor amplitude in a large cohort with essential tremor. Topography of essential tremor. Parkinsonism Relat. Benign essential tremor is not a single entity. In: Yahr MD, editor. Concepts park. Amsterdam: Excerpta Medica; Louis ED.
Early- and late-onset essential tremor patients represent clinically distinct subgroups. Tremor in the elderly: essential and aging-related tremor. Essential tremor and cerebellar dysfunction clinical and kinematic analysis of intention tremor. Rest tremor in patients with essential tremor: prevalence, clinical correlates, and electrophysiologic characteristics.
Essential head tremor is associated with cerebellar vermis atrophy: a volumetric and voxel-based morphometry MR imaging study. Cerebellar atrophy in essential tremor using an automated segmentation method. Torpedoes in the cerebellar vermis in essential tremor cases vs. Is tremor in dystonia a phenotypic feature of dystonia? Pandey S, Sarma N. Tremor in dystonia. Rest and other types of tremor in adult-onset primary dystonia.
Rest tremor in idiopathic adult-onset dystonia. Tremor in primary adult-onset dystonia: prevalence and associated clinical features. Mutations in ANO3 cause dominant craniocervical dystonia: ion channel implicated in pathogenesis. Am J Hum Genet.
Rare variants in ANO3 are not a susceptibility factor in essential tremor. Mutations in GNAL cause primary torsion dystonia. Nat Genet. Benign tremulous parkinsonism. Transl Neurodegener. Heterogeneity of monosymptomatic resting tremor in a prospective study: clinical features, electrophysiological test, and dopamine transporter positron emission tomography.
Chin Med J Engl. Article Google Scholar. Neuropathological findings in benign tremulous parkinsonism. Schwingenschuh P, Deuschl G. Functional tremor. Handb Clin Neurol. McAuley J, Rothwell J. Identification of psychogenic, dystonic, and other organic tremors by a coherence entrainment test.
Blink reflex recovery cycle in patients with dystonic tremor: a cross-sectional study. Temporal discrimination in patients with dystonia and tremor and patients with essential tremor. Arm tremor in cervical dystonia differs from essential tremor and can be classified by onset age and spread of symptoms.
Hallett M. Tremor: pathophysiology. Parkinsonism relat. J Neurol. NMR Biomed. Pallidal dysfunction drives a cerebellothalamic circuit into Parkinson tremor.
Ann Neurol. Current concepts of essential tremor. Rev Neurol. There is no explicit representation of GPi in the model network, so that pallidotomy may be represented in the model by removing the projection from STN to the thalamo-cortical circuits. Even though GPe is silent here presumably due to stronger inhibition from the feedback neuron in the absence of subthalamic inhibitory input , the tonic nature of STN discharge Figure 2C confirms that the system returns in a normal state.
The other kind of lesion reproduced in the model is at the level of cortex or thalamus. In the model that would correspond to a lesion of inputs to the GPe and STN segments or, in other words, removal of the feedback neuron. Note that in the case of this lesion at the level of basal ganglia input the feedback neuron shows somewhat bursty output. However, this bursting activity is at a much higher frequency and, therefore, cannot lead to tremulous movement of limbs.
The characteristic feature of GPe neuron is its tonic activity and high firing rate. The next subsection provides a systematic study of the circuit behavior for varying feedback. To study the effects of dopaminergic modulation we varied dopaminergic parameters s 1 and s 2 as proxy for the presence of dopaminergic modulation see Dopamine-dependent parameters subsection of Methods. The results of the previous subsection suggest that the strength of the feedback is essential for the occurrence of bursting, so we varied the dopaminergic parameters in a broad range to see how bursty the discharge is as quantified by SNR criterion, see Methods.
As an example, we consider the SNR1 as we vary the dopaminergic parameter s 1 in the interval [1] , [2] Figure 3. As the dopaminergic parameter increases, SNR1, which indicates the presence of the tremor-related bursting the presence of oscillations in the tremor frequency band , decreases, first moderately, then sharply to less than 1 the lack of activity in the tremor frequency range.
Thus, Figure 3 illustrates the transition between tremulous and non-tremulous case, as the dopaminergic action changes. Of note is a relatively sharp onset of tremor oscillations in the model and jagged profile of SNR. We think this is most likely due to the simplicity of the model.
This bifurcation cascade is likely to be model-specific. Moreover, if dopamine-dependent parameters are varied in different ways, the SNR profile may be different. While the example above may be illustrative of the role of the dopamine-modulated thalamo-cortical feedback loop, the results of dopamine action on different synapses and cells in the system may be different.
As we explain in the Methods, we study the effect of independent modulation of different properties of the network employing two dopaminergic parameters s 1 and s 2. How exactly dopamine will affect different synaptic and cellular parameters is not known, but the independent variation of two dopaminergic parameters which, in turn, corresponds to variation of several synaptic and cellular parameters, see Methods should give some general knowledge about the effect of the basal ganglia-thalamo-cortical feedback loop on the tremor-like bursting in the basal ganglia circuits.
We varied both s 1 and s 2 in the range from 1 to 1. The presence of tremor-like activity in STN the output node of our simplified basal ganglia network was assessed with SNR criteria. Figure 4 presents the result of this numerical experiment. Lighter shade of grey indicates stronger tremor activity. It is hard to define the exact level of SNR above which the activity can be called tremulous.
However, the present SNR criteria are based on an earlier experimental study of tremor in parkinsonian patients [49] , which uses 3. The four SNR criteria employed here are slightly different one from another and the resulting subplots in Figure 4 are also slightly different. In particular, maximal SNR tends to yield larger values than those of averaged SNR, which may be attributed to the large height and small width of the spectral peaks.
However, overall, tremor-like activity is present in the same regions, regardless of the criteria used. The smallness of the differences between the subplots points to the generic character of the observed phenomena.
The parameters s 1 and s 2 run along vertical and horizontal axis respectively, color codes for the value of SNR. The point 1, 1 corresponds to the bursting mode shown in Figure 2B. Parameters s 1 and s 2 are proxies of dopaminergic status and their higher values correspond to stronger dopamine influence.
Blue color indicates the absence of tremor-band oscillations, red color indicates prominent oscillations. Yellow and green correspond to SNR values termed to be tremulous in [48]. Figure 4 shows that the low values of dopaminergic parameters, i. This indicates that the strength of the coupling in the basal ganglia-thalamo-cortical feedback loop is responsible for the tremor oscillations. However, the dependence of SNR on the dopaminergic parameters is not monotonic.
The relative contribution of s 1 and s 2 is also different. Nevertheless, the general pattern low dopaminergic parameter values — more tremulous activity, high values — less tremulous activity is persistent.
We also study the effect of individual currents on the tremor-like oscillations in the loop. Unlike Figure 4 , dependence of SNR on the dopaminergic parameter s 1 and the AHP current conductance alone is monotonic: SNR abruptly decreases and remains low indicating disappearance of bursting activity in the tremor band as the value of g AHP decreases. Hence, the tremor-like oscillations in the loop substantially depend on the AHP current. The shades of gray code for the value of SNR1 like the color code in Figure 4 so that lighter areas exhibit stronger tremor oscillations.
B Bursting activity with variation of the dopaminergic parameter s 1 and the Ca current conductance g Ca. Parameters are the same as in Figure 2B. Lower values of AHP and Ca currents conductances lead to disappearance of tremor. Interestingly, our study revealed no substantial dependence of oscillations on the T-type current not shown. Hence, these results may indicate that the T current is not strongly involved in generation of tremor-like activity.
Next, we investigated the effect of calcium dynamics on tremor-like bursting activity in the model network. First we slowed down the calcium dynamics as shown in Figure 6A. Around this value, bursting activity in the tremor frequency range changes to tonic firing and this transition is mostly independent from calcium dynamics in the STN cell as can be seen by a nearly vertical transition from tremor oscillations red and yellow to tonic firing blue Figure 6A.
Similar results are obtained when calcium dynamics is accelerated Figure 6B. These results strongly suggest that the tremor-like oscillations in the STN neuron have network origin and are not solely based on the time scale of calcium dynamics in the STN and GPe neurons. Color codes for the value of SNR1 similar to Figure 4 so that yellow and red areas represent tremor oscillations.
The parameters are as in Figure 2B. Tremor oscillations are robust with significant variations of time scale of calcium dynamics in STN and GPe. The model circuit Figure 1B incorporates two delay units, which represent synaptic and conductance delays in polysynaptic pathways from STN to GPi to thalamus to cortico-striatal system.
While these delays are likely to be fixed for each individual subject, we do not know their exact values. Therefore we study the impact of the delays on the tremor-like activity in the model network. This range may include biologically unrealistic delay values but the objective is to ensure that the real delays are in the domain studied. Figure 7 describes how delays affect SNR of tremor frequency oscillations.
The regions of tremulous activity in the plane of delays are in the form of relatively narrow stripes; the slope of these stripes does not vary much and is close to 1. This suggests that the difference between delays may be more important than the values of the delays. Figure 7 also indicates that the oscillations are robust with respect to variation in delay values and tremor-like bursting exists for multiple values of the delays.
Thus even though the exact values of the delays in the loop are not known, there are likely to be some fitting with those at the domains of tremor existence. The parameters are the same as in Figure 2B. The shades of gray code for the value of SNR1 like the color code in Figure 4 so that lighter areas exhibit stronger tremor oscillations grey and white areas are tremulous dynamics. Finally, we substitute the Morris-Lecar-type feedback neuron considered so far in this paper with a more physiologically realistic thalamocortical relay cell model in the form used in [32] , which includes sodium, potassium and leak currents, as well as low-threshold calcium current.
Figure 8 shows how SNR depends on the dopaminergic parameters s 1,2 in the case of this modified model circuit. Similarly to the simple feedback model Figure 4 , STN oscillations in tremor frequency-band exist in the model network in parkinsonian state and cease when the dopaminergic parameter s 1 increases to indicate higher dopamine level and presumably normal state. This suggests that the observed dynamics are robust with respect to different types of the feedback neuron in the model network.
In turn, this suggests that the delayed feedback loop itself is likely to be essential for tremor-like oscillations in the model together with the cellular properties of STN and GPe neurons. Thalamocortical neuron parameters are given in Table 3. The parameters s 1 and s 2 run along vertical and horizontal axis respectively, the shade of grey codes for the value of SNR1. The modeling shows that anatomical and membrane properties of subthalamo-pallidal circuits are prone to generation of tremor-like bursting in the presence of relatively strong basal ganglia-thalamo-cortical feedback.
The destruction of the feedback leads to the suppression of the tremor-like oscillations as one would expect from the outcomes of surgical lesions in parkinsonian patients.
The dependence of the strength of tremor-like oscillations on the strength of dopamine-dependent synaptic projections is not monotonic. However, the model study demonstrates the general pattern of the change: as the basal ganglia-thalamo-cortical feedback loop becomes stronger, oscillations are likely to occur. The phenomenon is robust with respect to different kinds of modulation of the dopamine-dependent parameters.
The phenomenon is also robust with respect to different values of delays in the feedback loop. But the studied phenomenon persists for different values of delays.
Interestingly, recent studies suggest that the dopamine depletion negatively impacts autonomous activity in GPe [52]. Such a decline in GPe pacemaking may be seen at least to some extent similar to the increase in synaptic coupling, since in both cases the degree to which intrinsic dynamics influences the overall activity of a neuron is diminished in comparison with synaptic influence.
There are two important observations regarding the properties and mechanism of tremor oscillations in the model. We simulated the dependence of the tremor-like oscillations on the time scale of calcium dynamics in STN and GPe, and on the strength of calcium and calcium-dependent potassium current.
The calcium and calcium-dependent potassium currents need to be sufficiently strongly expressed to yield tremor-like oscillations. Neither does it rely on the presence of T-type calcium current which is almost inactivated during tremor-like oscillations in numerical experiment. This suggests that disruption of high-threshold calcium current and calcium-dependent potassium current, but not calcium time scale or T-type calcium current, will affect the existence of tremor. Eventually this may be an interesting statement to test experimentally.
Numerical simulation also indicates the importance of delays in the thalamo-cortical feedback. However, the delays in the circuitry are much shorter than the period of oscillations. This again suggests the importance of the network effects and of cortico-subcortical interactions for the genesis of tremor and setting its frequency.
Delays are hard to manipulate with experimentally; however, from a theoretical standpoint, it will be very interesting to study how the delays may promote oscillations of a much longer period. Finally we would like to note that the model effectively utilizes a negative delayed feedback just follow the signs of synaptic connections in the loop for the STN unit, Figure 1B , which is known to be able to give rise to oscillations [53].
A generic model for parkinsonian tremor with delayed negative feedback was studied by [54]. Their model, however, was concerned with delayed proprioceptive feedback which had long been shown not to be significantly involved in the origin of parkinsonian tremor [50] , [51] , [55] , [56]. It also did not represent the cortico-subcortical circuitry and membrane properties of the involved cells.
In that respect it was a more generic study of how the feedback may influence oscillations. We consider the feedback mechanism in Parkinsonian context. Thus this study provided computational rationale to suggest that this is the dopamine-mediated strength of cortico-subcortical loop, which facilitates the birth of tremulous oscillations. The very general nature of the feedback in the model and the robustness of the studied phenomenon indicate that the details of the feedback are unlikely to produce a substantial qualitative change in the modeling results.
The model considered clearly has some limitations. The simplicity of the model basal ganglia-thalamo-cortical feedback is both its advantage as it provides a way to study the generic effects of the feedback and disadvantage as it limits the model in many ways.
Several limitations are discussed below. The model network includes only single STN and GPe neurons following the framework of minimalistic approach to modeling. There are two different ways, in which this may limit the conclusions of the study. First is the very limited representation of the circuitry. The real anatomy of cortico-subcortical loops is complex while we consider simplified representation of striatum, thalamus and cortex and omit the other brain structures related to cortico-subcortical motor circuits.
The minimal circuit considered naturally cannot tell anything about particular effect of this anatomy; however, it suggests that the observed phenomenon is robust, may be generated due to the feedback as a general anatomical feature and may be not very sensitive to the details of the circuitry. Second is the modeling of a whole nucleus with a single neuron. As it was discussed in the modeling subsection this neglects a potential synaptic convergence and associated effects.
So, what is considered here is essentially the case of synchronized oscillations, which appears to be a reasonable case for tremor. In addition, the complexity of the real network and the number of possible and sometimes unknown connection parameters in the loop is huge. The introduction of these elements into the model will substantially increase the number of unknown parameters.
In particular, earlier modeling studies [28] considered oscillations within the basal ganglia network but not cortico-subcortical loops and the effect of the intrapallidal connectivity on these oscillations.
While overall the model neurons exhibit reasonable patterns of neural activity, the case of the model lesions at the level of basal ganglia output may present some problem. Hence, our modeling predicts the reduction in GPe activity after GPi lesion. It is hard to know if this is what happens in parkinsonian patients after lesions in internal pallidum we are not aware of recordings in GPe after lesion in ipsilateral GPi in patients with tremor.
Moreover, some studies indicate that GP intrinsic oscillation capability may increase after STN lesion [58] which would break the feedback to the basal ganglia. Thus we believe the numerical studies still indicate that a stronger basal ganglia-thalamo-cortical feedback promotes tremor oscillations and its destruction suppresses them.
Even though the lack of activity is visible in GPe in the computational results, the output of the model basal ganglia lacks burstiness. Likewise, in the case of the model lesions at the level of basal ganglia input the feedback neuron shows bursty output. Nevertheless, this bursting activity is high-frequency and, therefore, cannot give rise to tremor. Given the minimalistic modeling approach, the model should not be expected to reproduce the results of all known cortical and subcortical lesions with high fidelity.
Similarly, minimalistic modeling approach may not necessarily reproduce the firing rates with high fidelity. Vim is not directly represented in our model circuit. However, the patients in our study were evaluated in OFF time.
Furthermore, as a weighted limb test was not performed, some degree of misdiagnosis cannot be ruled out in cases of physiological action tremor.
The classification of pure action tremor is difficult because rest tremor may be very mild and intermittent, especially when patients are treated with dopaminergic drugs. Finally, another limitation of this cross-sectional study is that we did not study shifts between tremor subtypes; i.
This issue, however, would need to be analyzed in future studies. The data used to support the findings of this study are available from the corresponding author upon request. The authors declare that they have no conflicts of interest with respect to the content of the study. We thank Dr. Ignasi Gich for help with the statistical analysis and Ms. Mar Gironell for the drawings in Figure 1. This is an open access article distributed under the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Article of the Year Award: Outstanding research contributions of , as selected by our Chief Editors. Read the winning articles. Journal overview. Special Issues. Academic Editor: Seyed-Mohammad Fereshtehnejad. Received 12 Apr Revised 07 Aug Accepted 04 Sep Published 30 Sep Abstract Background. Type I. Pure resting tremor Type II. Action tremor without a time lag Type III. Pure action tremor Type IV. Table 1. Figure 1. Tremor types in Parkinson disease graphically explained. Difference in X and Y frequency is above 1.
Association: statistical association between clinical variable and tremor type. Table 2. Characteristics of patients with Parkinson disease by tremor type. Table 3. References A. Extrapyramidal Disorders , P. Vinken, G. Bruyn, and H. Klawans, Eds. View at: Google Scholar M.
Politis, K. Wu, S. Molloy, K. Chaudhuri, and P. Zetusky, J. Jankovic, and F. Teravainen and D. Koller, B. Veter-Overfield, and R. Deuschl, P. Bain, and M. S3, pp. View at: Google Scholar K. Bhatia, P. Bain, N. Bajaj et al. Hughes, S. Daniel, L. Kilford, and A. Zach, M. Dirk, B. Bloem, and R. Hess and S.
0コメント