Site-specific decrease in cortical reactivity during sensory trick in cervical dystonia patients

Study participants

Thirteen CD patients (mean age: 59.4 ± 13.3 (SD) years; 7 females) and 13 age-and gender-matched healthy controls participated in the study (Table 1). All participants were right handed as assessed by the Edinburgh Handedness Inventory of Manual Preference (Oldfield, 1971). All participants were screened by a physician or nurse practitioner for eligibility to undergo TMS and magnetic resonance imaging (MRI). All patients had a diagnosis of CD confirmed by a movement disorders specialist and did not have any other neurological illness. They all demonstrated sensory tricks (i.e., gently touching their face improved their symptoms). Their last injection of Botulinum toxin was at least 11 weeks prior to the study. Patients were considered ineligible for participation if they had any abnormal neurological signs other than dystonia, any history of brain tumor, stroke or head trauma. Other exclusion criteria included the use of any medications, including benzodiazepines, anticholinergics and antidepressants, in the two weeks prior to the study. Healthy volunteers (HV) underwent general physical and neurological examination as part of the screening for participation. They were excluded if they had any current or history of neurological/psychiatric illness or had used any medication acting on the central nervous system in the last two weeks. All participants were asked to abstain from alcohol and caffeine for 48 hours prior to the study. The study was approved by the CNS Institutional Review Board of the National Institutes of Health and conformed to the ethical guidelines of the Declaration of Helsinki. Written informed consent was obtained from all participants prior to the study.

Experimental procedure

Participants were seated comfortably on a chair with their forearms resting on a side table. An EEG cap (Braincap MR, Easycap, Munich, Germany) with 32 electrodes was positioned on their heads. Subjects also wore glasses with a subject tracker for neuronavigation whose position was secured throughout the experiment and was calibrated using the infrared camera of the optical neuronavigation system. Two Magstim200² stimulators connected through a Bistim unit (Magstim Company, Whitland, Dyfed, UK) and to a figure-of-eight coil (70mm external loop diameter) were used to deliver TMS pulses. Stimulation was delivered either to M1 or to SMA contralateral to the resting hand at 110% of resting motor threshold intensity. Further details of the EMG, neuronavigation, TMS and EEG methods are provided in the supplementary information.

The experiment comprised of 8 blocks in total for each subject. We had two blocks (50 pulses/block) for each of the following conditions – M1_rest, M1_trick, SMA_rest, and SMA_trick, that is, 4 blocks were designated for each stimulation site. Altogether, 100 pulses were delivered over a stimulation site during either rest or SeT condition. Patients were asked to either stay in their resting dystonic position or perform the SeT that relieved the dystonia during the entirety of the respective block. Seven patients used their left hand for the SeT, while the others used their right hand. Controls were asked to mimic the rest and SeT head positions of their age- and gender-matched patient during the respective blocks. The investigator demonstrated the dystonic head position of the patient matched to every healthy volunteer. The order of the blocks was randomized across participants. One hundred single, monophasic TMS pulses were delivered over M1 or SMA contralateral to the resting hand (i.e., the hand that did not perform the SeT). Supplementary figure 1B shows an illustration of the experimental protocol. EEG data analysis was performed using the EEGLAB [18] and Fieldtrip [19] open source MATLAB toolboxes. A detailed description of the data analysis methods is provided in the supplementary information. We performed Spearman’s correlation analysis of the statistically significant EEG parameters with clinical measures - duration of symptoms, disease severity and disability as assessed by Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS).

• Download figure
• Open in new tab

Figure 1.

(A) TMS-evoked potential (TEP) in healthy controls at rest (red) and during trick (blue). (B) TEP in cervical dystonia patients at rest (red) and during trick (blue). TEPs were evoked by stimulation over M1. Time window showing statistically significant difference between rest and trick conditions is in grey. (C) Topographic map of the scalp localizing the difference in cortical activity between rest and trick conditions in patients. (D) Bar graph showing the mean area under the curve of the TEP within the significant time window in both conditions (black bars: rest; shaded bars: trick) and subject groups (HV: healthy controls; CD: cervical dystonia patients). Error bars are standard error of mean. Asterisk indicates p<0.05.

Post hoc analysis: EMG data from the right SCM (RSCM) and the left SCM (LSCM) muscles were band-pass filtered (0.5Hz – 100Hz) then a notch filter (60Hz) was applied to remove power-line noise. Temporal profile of the muscle activation was obtained by applying a low-pass filter (5Hz) to the rectified EMG signals. The relative change in the muscle activation during SeT was computed from the muscle activation during the baseline and during SeT; here, the log change in percent (L%; Δ_L) was used to estimate the change in the SCM muscle activity (Δ_LSCM), which resolves the problem of asymmetry and non-additivity of the relative measure [20].

M1 stimulation

Time-frequency analysis

No significant differences in oscillatory power were observed between conditions in either of the subject groups (Supplementary figure 2). A between-groups comparison at rest also did not reveal any significant difference.

• Download figure
• Open in new tab

Figure 2.

(A) Topographic map of the scalp showing the significant positive cluster of electrodes (Fz, FCz, Cz, FC1, FC2, FC5 and FC6), that is, electrodes where there was significant increase in alpha power during trick in patients with cervical dystonia for SMA stimulation. (B) Bar graph showing the mean alpha power during the significant time window in both conditions (black bars: rest; shaded bars: trick) and subject groups (HV: healthy controls; CD: cervical dystonia patients) for SMA stimulation.

SMA stimulation

TEP analysis

No significant differences in TEPs were observed between conditions or subject groups (Supplementary figure 3).

• Download figure
• Open in new tab

Figure 3.

Correlation analysis (A) between symptom duration and trick-induced change in area under the curve of TMS-evoked potential within the significant time window (B) between TWSTRS severity score and trick-induced change in alpha power within the significant time window

Impact of SeT on TMS-evoked response

In the current study, we have shown that SeT in CD patients significantly decreased the amplitude of the long-latency positive component of the TEP – the P180. The neural basis of the different TEP components is still remains unclear [13]. Many studies suggest that the N100-P180 complex is contaminated by the auditory and somatosensory evoked responses, particularly bone conducted sound [23, 24]. Nevertheless, there is evidence to suggest that P180 also contains significant cortical activity evoked by the TMS pulse [15, 25, 26] and hence it cannot be ignored as artefactual. It is thought that P180 may represent a late excitatory component [27, 28]. Our results are most consistent with a change in cortical activity induced by SeT modulating the P180 amplitude. It is unlikely that this effect is solely due to bone conduction because we compared the TEPs elicited at the same stimulation intensity in all conditions, and the P180 amplitude was affected only following M1 stimulation in the patients. The site specificity of the effect does not support the purely auditory origin of P180. Moreover, post hoc analysis revealed that the P180 reduction was seen exclusively in those patients who exhibited an effective SeT and not in those who had a forcible trick. This reinforces that the reduction in P180 amplitude could be primarily due to change in cortical activity induced by SeT. TMS studies have demonstrated that CD patients have high intracortical facilitation, which is reduced by SeT [6]. Considering that P180 represents an excitatory component of TEP, our results also suggest that reduction in cortical facilitation could underlie an effective SeT.

Impact of SeT on TMS-induced oscillations

Contrary to our expectations, we did not find any significant effect of SeT on TMS-induced oscillations. Although we observed an increase in alpha power that was significant in CD patients, post hoc analysis revealed that this might not be relevant to the neurophysiology of SeT since this effect was observed irrespective of the effectiveness of SeT. Further, HVs who mimicked the head posture of the patients also showed similar trend. Figure 3 shows that both HVs and patients had low alpha power in the dystonic posture (alpha desynchronization). Changing the head position to neutral relaxed the neck muscles and thereby could have abolished this desynchronization. This seems likely because the SeT-induced change in alpha power correlated negatively with disease severity. That is, those who had severe disease, had stronger contraction of the neck muscles even at neutral head position achieved by SeT and hence less change in alpha power. Based on past studies, we expected that SeT would cause desynchronization of low frequency bands especially in the theta range [7]. There has been recent evidence showing that pallidal theta may be significant in the pathophysiology of CD [9]. CD patients showed high pallidal theta activity and SeT caused theta desynchronization in them. Further, Tang et al. [3] also demonstrated that pallidal and thereby cortical theta desynchronization could underlie an effective SeT. However, we did not observe any changes in the cortical theta activity. We think that it could be because we recorded EEG during tonic phase of SeT. The trick was maintained for about 5 minutes until 50 TMS pulses were delivered at 5-7s intervals within a block. Tang and colleagues recorded theta desynchronization about 4-6s following the onset of SeT. Due to our experiment design, we could have missed out on this observation.