Secondary and Primary Dystonia

Pathophysiological Differences

Maja Kojovic; Isabel Pareés; Panagiotis Kassavetis; Francisco J. Palomar; Pablo Mir; James T. Teo; Carla Cordivari; John C. Rothwell; Kailash P. Bhatia; Mark J. Edwards


Brain. 2013;136(7):2038-2049. 

In This Article

Materials and Methods


We studied 11 patients with secondary dystonia caused by structural brain lesion (five males and six females, mean age 45.8 years, range 28–68, Table 1), 10 patients with primary segmental dystonia (four males and six females, mean age 46.7 years, range 31–67, Table 2) and 10 age-matched healthy participants (five males and five females, mean age 48.7 years, range 27–67). Patients with secondary dystonia were included if they had (i) unilateral distribution of dystonia; (ii) discrete lesion in the basal ganglia and/or thalamus contralateral to the clinically involved side on MRI or CT; and (iii) no significant pyramidal involvement or hemisensory loss, as assessed by the Ashworth Scale and National Institutes of Health Stroke Scale. All patients were clinically examined and videotaped. Three patients with secondary dystonia had resting dystonia with fixed postures (Patients 1 and 2 fixed dystonia at leg, Patient 5 fixed dystonia at arm), the other eight patients had mobile dystonia at rest, worsened by action. All patients with primary dystonia had segmental dystonia with unilateral arm involvement visible at rest or on maintaining outstretched arm posture. Clinical disease severity was assessed with Burke-Fahn-Marsden scale. All patients treated with botulinum toxin were injected at least 15 weeks before participating in the study. One of the patients with secondary dystonia (Patient 5) underwent unilateral thalamotomy 20 years earlier, with only transient improvement of symptoms. At the time of the study, none of the participants were on any medications that could affect the measurements performed. All participants were right-handed. Eye blink classical conditioning testing was performed on patients with secondary dystonia and, for convenience their data on eye blink classical conditioning were compared with the data of patients with primary dystonia [seven males, six females, mean age 63.7 ± 3.4 (SEM) and healthy participants (six males, five females, mean age 61 ± 4.5 (SEM)] obtained using the same experimental protocol (Teo et al., 2009). Written informed consent was obtained from all participants and the study was approved by the local ethics committee and conducted in accordance with the Declaration of Helsinki.

Electromyographic Recordings

EMG recordings were made from the abductor pollicis brevis and adductor digiti minimi muscles on the side contralateral to stimulated cortex with Ag-AgCl surface electrodes using a belly-tendon montage. EMG signals were amplified (×1000) and band-pass filtered (bandwidth 20 Hz to 2 kHz) with a Digitimer D360 amplifier (Digitimer), acquired at a sampling rate of 5 kHz through a 1401 laboratory interface (Cambridge Electronic Design) and stored on a personal computer. The EMG traces were analysed using customized Signal® software version 4.00. The level of background EMG activity was monitored and trials with background EMG activity exceeding 50 μV were rejected online. The background EMG area in at least 200 ms preceding the transcranial magnetic stimulation (TMS) pulse was measured in all trials of each session and EMG root mean square amplitude calculated to ensure comparability of the baseline activity between two sides in patients with secondary dystonia and between patients with secondary and primary dystonia and healthy participants.

Transcranial Magnetic Stimulation

Single and paired pulse TMS of the primary motor cortex was applied using Magstim 2002 magnetic stimulators with a monophasic current waveform (Magstim Company). The magnetic stimulators were connected to a standard figure-of-eight coil with mean loop diameter of 9 cm. The intersection of the coil was held tangentially to the skull with the handle pointing backwards and laterally at an angle of ~45° to the sagittal plane in order to generate a posterior–anterior current in the brain (Kaneko et al., 1996; Di Lazzaro et al., 2004).The 'hot spot' was defined as the optimal scalp position for eliciting motor-evoked potentials (MEPs) of maximal amplitude in the contralateral abductor pollicis brevis muscle.

Corticospinal Excitability

The resting motor threshold and active motor threshold were determined according to standard definitions (Rossini et al., 1994). Single MEPs were recorded using a stimulus intensity adjusted to produce MEP amplitude of ~1 mV in the relaxed abductor pollicis brevis muscle (1 mV MEPs) and this intensity was kept constant for assessment of 1 mV MEPs through the experiment. For assessment of 1 mV MEP, at each time point [before paired associative stimulation (PAS) and 0, 15 and 30 min after PAS] 20 MEPs were collected. Input-output curves were assessed by recording four MEPs at each of the 10 intensities of stimulation, increasing in 10% steps from 80–170% of resting motor threshold.

Intracortical Excitability

Short-latency intracortical inhibition (SICI) was recorded at rest using a standard paired-pulse paradigm (Kujirai et al., 1993). Intensity of conditioning stimulus was set at 80% active motor threshold, whereas the intensity of the test stimulus was adjusted to elicit an MEP of ~1 mV. SICI was probed at an interstimulus interval of 2 ms. Forty MEPs were collected: 20 with conditioning stimulus and 20 with test stimulus alone. For assessment of cortical silent period, 10 single TMS pulses were applied at an intensity of 120% resting motor threshold while participants performed a constant contraction of abductor pollicis brevis at 20% of their maximum voluntary contraction, assisted by visual feedback. Cortical silent period was measured manually from the onset of MEPs to the re-emergence of sustained EMG activity.

Paired Associative Stimulation

PAS consisted of 200 electrical stimuli to the median nerve at the wrist paired with TMS stimuli over the hot spot for the abductor pollicis brevis muscle, given at the rate 0.25 Hz (Ziemann et al., 2004). Each TMS stimulus was preceded by an electrical conditioning stimulus at an interstimulus interval of 25 ms. Electrical stimulation was applied through a bipolar electrode, with the cathode positioned proximally. The electrical stimuli were constant current square wave pulses with a pulse width of 200 μs. Intensity of electrical stimulus was 300% of the perceptual threshold, while TMS intensity was adjusted to 1 mV MEP intensity. Subjects were instructed to look at their stimulated hand and count the peripheral electrical stimuli they perceived in order to ensure comparable attention levels between sessions.

Eye Blink Classical Conditioning

The eye blink classical conditioning protocol was delivered as detailed elsewhere (Teo et al., 2009). In brief, an electrical stimulus was applied through a bipolar electrode, with the cathode positioned proximally. The electrical stimuli were constant current square wave pulses with a pulse width of 200 μs, i.e. unconditioning stimulus, and were delivered to the right supraorbital nerve at an intensity adjusted to obtain stable R2 responses (~4–6 times the sensory threshold). Electrical supraorbital nerve stimulus was preceded by a tone, i.e. the conditioning stimulus, produced by a tone generator and presented bilaterally to the subject through binaural headphones at an intensity 50–70 dB above the individual hearing threshold. Conditioning stimulus intensity was kept constant during the experiment. The conditioning stimulus inconsistently produced an acoustic startle response (alpha blink) occurring within 200 ms after conditioning stimulus onset. Repeated pairs of conditioning stimulus and unconditioning stimulus at 400 ms intervals induced conditioned eye blink response (conditioned response) to appear with onsets within 200 ms before unconditioning stimulus. Eye blink classical conditioning sessions consisted of seven blocks: six acquisition blocks (each block contained 11 trials: nine trials of conditioning stimulus–unconditioning stimulus pairs, the 10th trial was unconditioning stimulus only and trial 11th was conditioning stimulus only) followed by one extinction block (11 trials of conditioning stimulus only). For measurement of eye blink classical conditioning, the conditioned responses were counted manually. EMG bursts were regarded as 'alpha blinks' if their amplitude exceeded 50 μV and if latency was <200 ms after the conditioning stimulus. EMG bursts were regarded as conditioned responses if latency was >200 ms after the conditioning stimulus but before the unconditioning stimulus. For the conditioning stimulus only trials, EMG bursts occurring 200–600 ms after the conditioning stimulus were considered conditioned response.

Experimental Design

Patients with secondary dystonia were tested on both hemispheres, corresponding to the affected and non-affected side in two different TMS sessions, separated by at least 1 week. The order of the tested hemisphere (affected versus unaffected) was balanced between subjects. Patients with primary dystonia were tested on the hemisphere corresponding to the affected side only, since previous studies showed that in primary dystonia abnormalities in TMS measures are present in affected and unaffected parts of the body (Quartarone et al., 2008). Healthy participants were tested on the dominant hemisphere only (Ridding and Flavel, 2006). In each session we began with baseline assessments of resting motor threshold, active motor threshold and 1 mV MEP, input-output curve, SICI and cortical silent period. We then delivered PAS as described above and assessed the effect of this conditioning protocol on corticopinal excitability (resting motor threshold, active motor threshold, and 1 mV MEPs in abductor pollicis brevis and adductor digiti minimi muscles) and cortical silent period at three different time points: 0, 15 min and 30 min after PAS. In addition, patients with secondary dystonia underwent a third session for eye blink classical conditioning testing.

Statistical Analysis

Distribution of data was assessed using Shapiro–Wilk test of normality. Greenhouse-Geisser method was used where necessary to correct for non-sphericity. For parametric tests (ANOVA) post hoc Tukey tests were used to further analyse significant main effects or interactions. The significance was preset at P ≤ 0.05.

To test for age differences between patients with secondary dystonia, patients with primary dystonia and healthy controls we used one-way ANOVA with Group as a between subject factor. The differences in disease duration, Burke-Fahn-Marsden scores and duration of botulinum toxin treatment between patients with secondary and primary dystonia were assessed with Mann-Whitney U-test.

The primary aim of this study was to compare the TMS parameters between affected side in patients with secondary dystonia and patients with primary dystonia and healthy participants. Resting motor threshold, active motor threshold, EMG root mean square amplitude, SICI and cortical silent period were compared between groups using ANOVAs with a factor Group (three levels: secondary dystonia-affected side, primary dystonia and healthy participants) as a between-subject factor. For analysis of SICI, conditioned MEP amplitudes were averaged, normalized to average unconditioned MEP amplitudes and entered into ANOVA with the factor Group as between-subjects factor. Input-output curves were compared between groups in ANOVA with the factor Group and the factor Stimulus intensity (10 levels of stimulator output intensity ranging from 80–170% of resting motor threshold intensity) as a within-subjects factor. For analysis of PAS effect on 1 mV MEP amplitude, MEP amplitudes at each time point were averaged, normalized to baseline MEPs and entered into ANOVA with factor Group as between-subjects factor and factors Muscle (two levels: abductor pollicis brevis and adductor digiti minimi muscle) and Time point (three levels: 0 min, 15 min and 30 min after PAS) as a within-subjects factors. The effect of PAS on resting motor threshold and active motor threshold and cortical silent period was evaluated using separate ANOVAs with factors Group as a between-subject factor and Time point (three levels: normalized resting motor threshold or active motor threshold or cortical silent period at 0 min, 15 min and 30 min after PAS). As a secondary analysis, we assessed how TMS measures compared between the affected and non-affected side in patients with secondary dystonia, using repeated measures ANOVA or paired sample t-test. For eye blink classical conditioning, the percentage of conditioned responses over different blocks did not follow the normal distribution, therefore non-parametric tests were used. We first compared the number of overall conditioned responses (for all blocks) in each group using Kruskal-Wallis ANOVA. The differences in the number of conditioned responses in each block between groups were then assessed by Mann-Whitney U test. Finally, for each group we used Friedman ANOVA to test if there was a conditioning of eye blink responses across blocks.

Possible correlations between clinical and demographic data (disease duration, Burke-Fahn-Marsden score, duration of botulinum toxin injection treatment) and TMS measures (SICI, averaged PAS response) or eye blink classical conditioning measures (the average percentage of conditioned responses per block) were evaluated with the Spearman correlation analysis.

The significance was preset at P ≤ 0.05. Unless otherwise stated, data are given as mean ± standard error of the mean (SEM).