These authors contributed equally to this work.
Impaired force control in writer's cramp showing a bilateral deficit in sensorimotor integration
Article first published online: 10 OCT 2013
Copyright © 2013 Movement Disorder Society
Volume 29, Issue 1, pages 130–134, January 2014
How to Cite
Bleton, J.-P., Teremetz, M., Vidailhet, M., Mesure, S., Maier, M. A. and Lindberg, P. G. (2014), Impaired force control in writer's cramp showing a bilateral deficit in sensorimotor integration. Mov. Disord., 29: 130–134. doi: 10.1002/mds.25690
Relevant conflicts of interest/financial disclosures: Nothing to report.
Full financial disclosures and author roles may be found in the online version of this article.
- Issue published online: 23 JAN 2014
- Article first published online: 10 OCT 2013
- Manuscript Accepted: 10 SEP 2013
- Manuscript Revised: 11 JUL 2013
- Manuscript Received: 29 APR 2013
- writer's cramp;
- force control;
- basal ganglia;
- sensorimotor feedback
Abnormal cortical processing of sensory inputs has been found bilaterally in writer's cramp (WC). This study tested the hypothesis that patients with WC have an impaired ability to adjust grip forces according to visual and somatosensory cues in both hands.
A unimanual visuomotor force-tracking task and a bimanual sense of effort force-matching task were performed by WC patients and healthy controls.
In visuomotor tracking, WC patients showed increased error, greater variability, and longer release duration than controls. In the force-matching task, patients underestimated, whereas controls overestimated, the force applied in the other hand. Visuomotor tracking and force matching were equally impaired in both the symptomatic and nonsymptomatic hand in WC patients.
This study provides evidence of bilaterally impaired grip-force control in WC, when using visual or sense of effort cues. This suggests a generalized subclinical deficit in sensorimotor integration in WC. © 2013 International Parkinson and Movement Disorder Society
Writer's cramp (WC) is a task-specific form of dystonia with abnormal movements and postures of the hand during writing. Grip-force control is important for hand writing, and one hypothesis in WC is that there is a disruption in the underlying cortical sensorimotor network, including the premotor and parietal areas.[2, 3] In WC, grip-force control is affected in the symptomatic hand.[4, 5] We tested the hypothesis that grip-force adjustments according to visual and somatosensory (sense of effort) information would be affected in WC in both the symptomatic and asymptomatic hand, suggesting a general deficit in sensorimotor integration.
Patients and Methods
Fifteen patients (9 females and 6 males; mean age: 47.4 ± 14.3 years; mean disease duration: 10.9 ± 7.2 years) affected by WC in the dominant hand (14 right-handed and one left-handed) were included. Six patients were affected by simple WC, the remaining by dystonic cramp (DC; clinical details: see Table 1). Dystonic patterns during writing were primarily found in the index finger and thumb. No involuntary movements were present during the force-tracking task or at rest in either hand. None of the patients received botulinum toxin injections for at least 6 months before the study. Patients were compared to a group of 15 healthy control subjects (mean age: 44.9 ± 19.5 years). The procedures of the study complied with the Declaration of Helsinki, and subjects provided informed consent.
|Patient||Gender||Age (Years)||Dom Hand||Diagnosis||Symptom Duration (Years)||Pattern of Dystonic Postures During Writing||Mirror Movements||BFM||WST (Letters/min)||Grip MVC (N)||Sensibility Test (g)||Pick-up Test (sec)|
|4||F||19||R||SC||4||Shoulder, forearm, F1, 2, 3, 4, 5||Yes||1||81||34||34||0.07||0.07||13||12|
|5||M||32||R||DC||8||Forearm, wrist, F1||No||2||68||46||40||0.07||0.07||15||17|
|7||F||34||R||DC||3||Wrist, F1, 2, 3, 4, 5||Yes||1||107||26||32||0.07||0.07||10||10|
|8||M||51||R||DC||21||Wrist, F1, 2||No||3||60||50||54||0.07||0.07||35||18|
|11||F||41||R||DC||19||Wrist, F1, 2, 3, 4, 5||Yes||1||104||30||28||0.07||0.07||13||13|
|15||F||52||R||SC||1||Wrist, F1, 2, 3, 4, 5||Yes||2||99||26||22||0.07||0.07||11||12|
Visuomotor and Matching (Sense of Effort) Tasks
Grip force was recorded using strain gauge-force sensors, and a visuomotor force-tracking task previously described was used (Fig. 1A). A series of ramp hold-and-release target-force trajectories with two target-force levels were used: a low absolute level at 5 N and a higher relative level at 10% maximal voluntary contraction (MVC). Force tracking was performed in two configurations: (1) unimanual tracking: one hand performed the task with visual feedback (a cursor that moved vertically as a linear function of grip force) and (2) bimanual force matching: the task was performed with the two hands simultaneously, but visual feedback was provided for only one hand. Thus, the tracking hand (with visual feedback) performed as in the unimanual task, whereas the other (matching) hand was supposed to simultaneously produce the same force, but without visual feedback. Each configuration was performed twice, with feedback from the right and alternatively from the left hand. Task conditions (with 32 trails each) were randomized among subjects.
Performance measures for each trial were arranged as follows:
- Relative error (Ns): the total error (the absolute sum of positive error above and negative error below target) between the applied force and the target force trajectory, normalized to the target force level.
- Variability: the coefficient of variation (CV) of force (i.e., standard deviation/mean). The CV expresses variability relative to the mean force level and is often used in studies of force variability.
- Release duration (ms): Subjects were instructed to release force completely and immediately at the end of the hold phase. The time taken to abruptly reduce the grip force (from 75% to 25% of the target force) was calculated. The release onset (i.e., time of initial force reduction) was also quantified as the time when the slope (dF/dt) first crossed a negative threshold.
- Force-matching difference (N): the accuracy of force matching during the hold phase (DIFF). DIFF = the mean force of the tracking hand (with visual feedback) minus the mean force exerted in the matching hand (without visual feedback).
Data Analysis and Statistics
Force data were analyzed using Matlab v7 (The MathWorks, Inc., Natick, MA). Relative error was analyzed using a general linear model repeated measures Analysis of variance (ANOVA) with one GROUP factor and three within-subject factors: PHASE (ramp or hold); HAND (dominant or nondominant);and FORCE (5 N or 10%). CV, release duration, release onset, and DIFF were analyzed similarly without PHASE factor. Post-hoc tests were applied to significant differences. Relative error was visibly reduced on the second trial, compared with the first, and this effect was compared between groups (ANOVA with additional TRIAL effect). Statistical analysis was performed using Statistica 10 (StatSoft, Inc., Tulsa, OK), and the level of significance was set to P ≤ 0.05.
No difference was found between groups in sensory function, dexterity, MVC, age, laterality, and gender (P > 0.2; Table 1).
Visuomotor Grip-Force Control
All subjects completed the task successfully (examples, Fig. 1). Grand average error (across ramp and hold phases and across both force levels) was higher in patients than in control subjects (F(1,28) = 12.48; P = 0.002;, Fig. 1B). Error was greatest during the ramp (F(1,28) = 40.6; P < 0.001), and error was force-level dependent: 33% greater error was found at low target forces (5 N), compared with the higher 10% level (F(1,28) = 38.33; P < 0.001). This effect was stronger in patients, compared to controls (GROUP*FORCE: F(1,28) = 5.2; P = 0.03). Patients thus showed greatest increase in error at the low force level. No difference in error was found between the dominant and nondominant hand in control subjects. Surprisingly, this was also the case in WC patients (F(1,28) = 1.2; P = 0.28). All subjects showed greater accuracy on the second trial, compared with the first, with an average error decrease of 10% (F(1,17) = 15.7; P = 0.001), with no difference between groups (GROUP*TRIAL: F(1,17) = 0.3; P = 0.58).
Grand average CV was approximately 50% higher in patients than in control subjects (F(1,28) = 7.66; P = 0.01). CV was higher in the dominant than in the nondominant hand (F(1,28) = 15.80; P < 0.001). This asymmetry was similar in patients and control subjects. CV was four times higher at the low force level (F(1,28) = 85.78; P < 0.001), and this effect was stronger in patients (GROUP*FORCE: F(1,28) = 6.05; P = 0.02).
Time taken to release the grip force was longer in patients (grand average: 109 ms) than in control subjects (84 ms; F(1,28) = 3.96; P = 0.05). This effect was magnified in patients at the 5-N level (GROUP*FORCE: F(1,28) = 7.33; P = 0.001). No difference in release duration was found between the dominant and nondominant hand in both groups (F(1,28) = 0.2; P = 0.67). No group difference was found in time of onset of grip release (P > 0.15).
Matching (Sense of Effort) Task
Figure 1C shows examples of the bimanual matching task. In control subjects, the grand average force difference (DIFF) between the tracking hand with visual feedback and the matching hand without feedback was negative (−1.45 ± 1.29 N). Thus, control subjects produced more force in the matching hand, compared with the tracking hand. In patients, the reverse was observed with a positive grand average DIFF (1.32 ± 2.62 N). This group difference was highly significant (F(1,28) = 13.55; P < 0.001). No difference was found between dominant and nondominant hands (F(1,28) = 2.1; P = 0.16), and no interaction was found between GROUP*HAND (F(1,28) = 0.06; P = 0.8). A GROUP*FORCE interaction was found (F(1,28) = 14.9; P < 0.001) showing that patients differed from control subjects at the 10% force level only (Fig. 1D). Thus, a force of 38 N (mean 10% level in patients) was perceived by a WC patient as 36.5 N, compared to 44 N in controls. Patients with the highest DIFF 10% also showed greatest error during unimanual force tracking (R = 0.64; P = 0.01).
All force-control parameters were similarly affected in patients with simple or dystonic WC and with or without mirror dystonia (P > 0.2). No significant correlations were found between force-control measures and clinical scales.
We found deficient grip-force control in both the symptomatic and nonsymptomatic hand in WC patients. Grip-force parameters changed as a function of sensory feedback, suggesting that inaccurate grip-force scaling is a manifestation of impaired sensorimotor integration. During unimanual force tracking, relying on visual input, all three performance measures (error, CV, and release duration) were similarly affected in both hands. Visuomotor deficits were greater at the low force level, compatible with reduced cortical inhibition at low force levels. During the bimanual force-matching task, relying on tactile and proprioceptive inputs, we also found a bilateral deficiency in WC. WC patients systematically underestimated their force at 10% grip MVC, contrary to the overestimation observed in control subjects. A likely mechanism is disturbed integration of muscle afferents (important for sense of effort perception) in the corticostriatal loop. This is coherent with an underestimation of force matching in healthy subjects when the muscle tendon is vibrated and preliminary findings of impaired integration of force feedback in other types of dystonia. Improving force matching in WC may be beneficial, as suggested by a study on proprioceptive training in musician's dystonia. The correlation between accuracy during unimanual force tracking and bimanual force matching suggests a multimodal deficiency of sensorimotor integration. We cannot rule out abnormal transcallosal communication as a cause of impaired force matching, although we found no motor overflow to the resting hand during unimanual tracking (P = 0.5, results not shown). Both groups improved equally across trials, suggesting that WC patients could predict force output on repeated trials. Thus, we did not detect deficits in movement prediction that may be specific to writing tasks. The lack of correlation of our findings to the clinical characteristics of writer's cramp may be the result of our nonwriting task. The subclinical and generalized nature of the deficit in sensorimotor integration may indicate a vulnerability trait for the development of WC.[15, 16]
Our findings are coherent with bilateral changes in sensory pathways[17-19] and in premotor-parietal activation patterns in WC.[3, 20] Indeed, frontoparietal activation plays an important role in grip tasks, and altered grip-force scaling in either hand can be induced by “virtual lesions” of the posterior parietal cortex using transcranial magnetic stimulation. Therefore, we hypothesize that altered activity in frontoparietal sensorimotor areas may be involved in abnormal grip-force control.
In conclusion, this study provides evidence of a similar bilateral sensorimotor impairment in grip-force control in WC patients. The tasks used may prove useful in exploring new endophenotypes in dystonia and for the development of targeted training approaches.
1. Research Project: A. Conception, B. Organization, C. Execution; 2. Statistical Analysis: A. Design, B. Execution, C. Review and Critique; 3. Manuscript Preparation: A. Writing of the First Draft, B. Review and Critique.J.P.B..: 1A, 1B, 1C, 2C, 3A, 3BM.T.: 1B, 1C, 2A, 2B, 2C, 3BM.V.:1B, 2C, 3BS.M.: 2C, 3BM.A.M.: 1B, 2C, 3BP.G.L. 1B, 1C, 2A, 2B, 2C, 3A, 3B
Nothing to report.