- Email: [email protected]
Plasticity of static graviceptive function in patients with cervical dystonia
Journal of the Neurological Sciences 373 (2017) 230–235
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Plasticity of static graviceptive function in patients with cervical dystonia Kirsten Platho-Elwischger a,⁎,1, Gottfried Kranz a, Thomas Sycha a, Daniela Dunkler b, Paulus Rommer a, Christian Mueller a, Eduard Auff a, Gerald Wiest a a b
Department of Neurology, Medical University of Vienna, Waehringer Guertel 18-20, 1090 Vienna, Austria Center for Medical Statistics, Informatics, and Intelligent Systems, Section for Clinical Biometrics, Medical University of Vienna; Spitalgasse 23, BT88/E 03, 1090 Vienna, Austria
a r t i c l e
i n f o
Article history: Received 11 March 2016 Received in revised form 29 December 2016 Accepted 3 January 2017 Available online 4 January 2017 Keywords: Dystonia Botulinum toxin Plasticity Subjective visual vertical Graviceptive function
a b s t r a c t Objective: Botulinum toxin (BoNT) is effective in improving abnormal head posture in cervical dystonia (CD) within a period of several weeks to an upright position. These dynamic alterations over time represent a unique model to study the plasticity of static graviceptive function in CD by assessing the subjective visual vertical (SVV). Methods: SVV was assessed in 30 CD patients and 13 controls, in their habitual head posture and with fixed head positions at different angles of head tilt. The patients were tested before and 3 weeks after BoNT administration. Results: At baseline, the patient's estimates in their habitual head posture and at forced head tilt angle of 30° differed significantly from those of controls. This effect was no longer visible after BoNT injection, or when the patient's head was fixed in an upright, and at a 15° head tilt position, respectively. A moderate positive correlation between disease severity and SVV aberrations was found. When comparing within-group changes, in all participants significant aberrations occurred when the head was tilted at 30°, which may be explained by the physiological E-effect. Conclusion: Our findings demonstrate that patients with CD exhibit altered static graviceptive perception when tested in their habitual head posture and an overshooting E-effect at the major forced head tilt of 30°. This graviceptive perceptual deficit can be reversed by modification of the somatosensory input (i.e. head fixation) as a short-term effect, and by changes in the proprioceptive input (i.e. BoNT injections) as a long-term effect. © 2017 Elsevier B.V. All rights reserved.
1. Introduction The evaluation of verticality perception is usually performed by the assessment of the SVV, in healthy individuals with a maximum aberration of b2° from the ‘true’ gravitational vertical [1,2]. Major aberrations of the SVV, up to 10–30° in the direction of tilt, can be produced by a body tilt of 80–90° to one side, which is also referred to as the “A-effect” [1]. Lateral head and body tilt up to 50° leads to perception of the SVV overshooting to the opposite side of the tilt, which is called the “E-effect” [3–6]. In healthy persons, small head roll angles (under 30°) revealed more accurate estimations, close to the gravitational vertical [2]. Verticality perception is mediated by visual, otolithic and Abbreviations: AIC, Akaike information criterion; BoNT, botulinum toxin; CCW, counter clockwise; CD, cervical dystonia; CI, confidence interval; CNS, central nerve system; CW, clockwise; P, p-value; SD, standard deviation; SVV, subjective visual vertical. ⁎ Corresponding author: Kirsten Platho-Elwischger, MD, Koeglergasse 2a, 1120 Vienna, Austria. E-mail addresses: [email protected] (K. Platho-Elwischger), [email protected] (G. Kranz), [email protected] (T. Sycha), [email protected] (D. Dunkler), [email protected] (P. Rommer), [email protected] (C. Mueller), [email protected] (E. Auff), [email protected] (G. Wiest). 1 Present address: Rehabilitation Clinic Meidling, Koeglergasse 2a, 1120 Vienna, Austria.
http://dx.doi.org/10.1016/j.jns.2017.01.007 0022-510X/© 2017 Elsevier B.V. All rights reserved.
somatosensory input [1]. Thereby, the somatosensory system and proprioceptive input from neck muscles, respectively play a major role in the estimation of the SVV during lateral body tilt [1,7–10]. The pathophysiological mechanism of CD still remains unclear. Animal models suggest involvement of multiple cerebral integration centers. The “sensory trick”, supports the assumption of altered sensorimotor processing in CD. In addition, vestibular abnormalities have been suggested [11,12]. Previous studies on the perception of verticality in CD yielded conflicting results [7,13,14]. When treated with BoNT by an experienced clinician, head posture in people with CD usually improves within a period of 3 weeks to an upright position and diminishes subsequently after approximately 9 weeks. These dynamic alterations of head posture over time represent a unique model to study the effects of altered somatosensory and otolithic input on the perception of verticality. In this study we tested the SVV of CD before and after BoNT, in order to evaluate the plasticity of static graviceptive function in these patients. The purpose of the present study was to 1) assess static graviceptive function at habitual head posture and at different head tilts (i.e. short-term modulation of the somatosensory input) in normal subjects and in patients with CD and 2) study the effects of BoNT injections (i.e. long-term modulation of the proprioceptive input) on static graviceptive function in patients with CD.
K. Platho-Elwischger et al. / Journal of the Neurological Sciences 373 (2017) 230–235
2. Materials and methods 2.1. Subjects 32 consecutive patients with idiopathic CD undergoing regular treatment with BoNT at the BoNT Outpatient Clinic of the Department of Neurology, Medical University Vienna, were included in the study. All patients participated at baseline evaluation, however, two patients refused to complete the follow-up examination (3 weeks after baseline evaluation), so that a total of 30 patients completed the whole study (mean age 59 [SD 12] years; females 19, 63.3%). The patient's clinical details are listed in Table 1. At baseline, 17 (56.7%) patients presented with habitual head deviation (chin-nasion line rotated in the roll plane) to the right, and 13 (43.3%) to the left, respectively. Patients with a history of vestibular disorders, as well as patients with head rotation in the horizontal plane exceeding 15° or secondary CD were excluded. 2.2. Standard protocol approval and patients consents The study was approved by the ethical committee of the Medical University of Vienna. All patients and controls signed an informed consent prior to their inclusion in the study. The study has been performed in accordance with the ethical standards laid down in the Declaration of Helsinki. This pilot study has been registered at http://clinicaltrials.gov (NCT01180270).
231
Assessment of SVV, Tsui score and head deviation was performed at the baseline visit directly before routine BoNT treatment (i.e. at time of minimum treatment effect). Three weeks after injection (median 21 days, range 18–28), i.e. at time of maximum treatment response, the initial assessment was repeated. As controls, 13 healthy subjects (age 52.8 [11.6], nine females, 69.2%) without a history of vestibular disorders, with normal vestibular- and ocular motor function tests (System 2000, Micromedical Technologies, Illinois, USA) and no history of CNSactive medications participated in the study. 2.4. Assessment of SVV SVV was assessed in an upright sitting position in a completely dark room. In front of the participants, at a distance of 100 cm, there was a dim light bar, 2 mm wide and 10 cm long, which could be rotated about its midpoint by means of an electronic motor and a remote control device (System 2000, Micromedical Technologies, Illinois, USA). All participants adjusted the bar six times from randomized starting positions for parallel alignment with the perceived gravitational vertical. The six estimates were averaged for further analysis. SVV alignments were performed both by patients and healthy controls at five different head positions (head fixed upright 0°, CW and CCW: fixed head deviation at 15° and fixed head deviation at 30°). A plastic head ring and a hook-and-loop tape that was mounted on an adjustable neck rest (which covered the occiput and the posterior neck) fixed the head. All patients and controls also performed one trial with habitual head posture. The interval between each test/head position was approximately 3 min.
2.3. Procedures 2.5. Statistical analysis Patients were treated with four different preparations of BoNT (Table 1). The median interval between first SVV assessment at baseline and previous BoNT injection was 105 days (range 83–371). The severity of CD was assessed by the Tsui score, head deviation was measured by referencing the chin-nasion line to earth vertical with a finger goniometer (KaWe Germany) [15]. In order to rule out vestibular disorders, which might interfere with SVV perception, such as vestibulopathy or neuritis vestibularis, all patients underwent vestibular- and ocular motor function tests by means of digital videooculography and a rotational chair system (System 2000, Micromedical Technologies, Illinois, USA) at their baseline visit. Table 1 Clinical details of 32 patients at baseline assessment. N (%)
Median (IQR)
Male Female Age in years (a) Duration of disease (a) BoNT treatment duration (a) Severity of disease measured by Tsui score
11 (36.7) 19 (63.3) 30 (100) 30 (100) 30 (100) 30 (100)
59.0 (12.0) 15.6 (9.4) 9.4 (7.0) 6.2 (3.9)
Predominant head deviation in degrees (°) Clockwise head deviation Counterclockwise head deviation
17 (56.7) 13 (43.3)
10.0 (4.8) 8.5 (7.7)
Amount of BoNT preparation applied in MU Abo-BoNT A, Dysport®, Ipsen Inco-BoNT A, Xeomin®, Merz Ona-BoNT A, Botox®, Allergan Rima-BoNT B, NeuroBloc®, USWorldMed
14 (46.7) 7 (23.3) 8 (26.7) 1 (3.3)
579.8 (233.0) 149.6 (59.6) 212.8 (93.4) 15,500
Patients with CNS-active co-medication Total Antidepressant Anticonvulsants Anticholinergics Musclerelaxants Neuroleptics
12 (40.0) 9 (30.0) 4 (13.3) 2 (6.7) 1 (3.3) 0
BoNT: botulinum toxin; MU: mouse units; N: number of patients; SD: standard deviation.
Categorical variables are given as absolute numbers and percentages; continuous variables are given as median and interquartilerange (IQR). SVV was measured in degrees of deviation from gravitational vertical. CCW deviations were calculated as negative values, CW deviations as positive. For the evaluation of differences in the perception of verticality in CD patients and healthy controls the direction of SVV deviations has not been taken into account, i.e. that absolute deviations from earth vertical - irrespective of their direction - were calculated. Controls were assessed once, as no injections were performed, and the median aberrance of the SVV at 0° was very small at baseline measures. For comparison of differences between patients and controls, fixed head positions and visits, repeated measures ANOVA was applied: In the model only for patients deviations, the visits and the interaction of the fixed head positions were included. Two separate models for each visit compared patients with controls (including fixed head positions, deviations of patients and controls, and the interaction). The last model compared the deviations of controls. Covariance structure was selected minimizing AIC. In addition, t-tests or Wilcoxon-tests, depending on the distribution of the respective continuous variable, were used. Pearson correlation was used to measure the association between Tsui score and SVV aberrations. Boxplots and scatter-plots were generated to visualize distributions. A two-sided p b 0.05 was considered statistically significant. The statistical analysis was performed with the SAS system version 9.2 (2008; SAS Institute Inc., Cary, NC). 3. Results 3.1. Absolute SVV deviations at habitual head position Absolute SVV-deviations from true gravitational vertical were calculated irrespective of CW- or CCW direction. At baseline, all CD patients presented with marked abnormal habitual head posture, were reflected by elevated Tsui scores (Table 1). SVV judgments of patients tested without head fixation, i.e. at their “habitual” head position, yielded major aberrations of SVV from earth vertical. The median SVV
232
K. Platho-Elwischger et al. / Journal of the Neurological Sciences 373 (2017) 230–235
judgement was 2.65° (95% CI 0.17 to 7.83). In contrast, healthy controls, tested at neutral head position (approximately 0° head tilt), aligned their SVV responses close to gravitational vertical, with a median of 1.33°([0.97], 95% CI 0 to 3). When compared with the estimates of controls, the median difference of SVV aberrations of patients was −1.34° with a 95% CI from − 2.5 to − 0.33. This difference was significant (p = 0.017; Fig. 1). Three weeks after BoNT injection, the patient's clinical symptoms had significantly improved, reflected by a significant decrease of Tsui score (mean reduction −2.4°[2.12]; p b 0.0001, Paired t-test) and the degree of head deviation in roll plane (mean reduction − 6.23°[4.85]; p b 0.001). The absolute SVV judgements of patients had also improved. The median SVV judgement was 1.42° ([2.05], 95% CI 0 to 6.5). When compared with controls, the SVV difference was not significant anymore (p = 0.2893; Fig. 1). 3.2. Absolute SVV deviations at forced head positions (for descriptive data please see Table 2) 3.2.1. Deviations at baseline and 3 weeks after injection, tilt-angle of 0° At baseline, when the patient's head was forced to upright position (i.e. 0° position), the median SVV response was 2° (95% CI 0.33 to 9.3). There was no significant difference of SVV judgements of patients from those of controls (p = 0.1134; mean difference −−1.2°; 95% CI −2.7° to 3°; repeated measure ANOVA). Three weeks after BoNT injection the median SVV judgement had not changed, the median response was 2.42° (95% CI 0.33 to 8.33). However, when compared with those of controls, a statistically significant difference was found (p = 0.019; mean difference −1.7°, 95% CI −3.11° to −0.3°; Fig. 2a). 3.2.2. Deviations at baseline and 3 weeks after injection, tilt-angle of 15° The median estimates of patients (irrespective of CW or CCW rotation) were 2.25° (95% CI 0.34 to 7.83) at baseline and 2.17° (95% CI 0.16 to 9.5) at week 3. Controls tested at head tilt-angle of 15° estimated SVV with a median of 1.66° (95% CI 0.34 to 5.34). When the patients judgements at baseline, or week 3 were compared with those of controls, no significant difference was detected (Baseline p = 0.2207,
Fig. 1. Habitual head position. Horizontal axis, left to right: SVV estimates in controls (“Control”), patient's estimates at habitual head posture at baseline (“CD baseline”) and 3 weeks after BoNT injection (“CD week 3”). Vertical axis: Absolute SVV deviations from gravitational vertical in degree. Baseline SVV aberrations of patients were significantly deviated when compared with controls (p = 0.017). CD: cervical dystonia; SVV: subjective visual vertical.
mean difference −− 1.01°, 95% CI − 2.65° to 0.63°; 3 weeks after BoNT injection: p = 0.2187; mean difference − 1.23°, 95% CI − 3.22° to 0.76°; repeated measure ANOVA; Fig. 2b). 3.2.3. Deviations at baseline and 3 weeks after injection, tilt-angle of 30° At tilt angle of 30° (irrespective of CW or CCW rotation), the patients' median estimates at baseline were 6.42° (95% CI 2.5 to 15). Median SVV judgements of controls were 5.33° (95% CI 0.17 to 9.84). A significant difference between patients and controls was found (p = 0.026, mean difference −3.20, 95% CI − 6 to −0.4; repeated measure ANOVA; Fig. 2c). Three weeks after BoNT injections the median estimate was 4.75°(0.34 to 18.3) and did not differ significantly from those of controls (p = 0.1913, mean difference −2.31, 95% CI −5.83 to 1.20). 3.2.4. Within-group differences of SVV estimates at different forced head tilt positions (mean differences and p-values are listed in Table 2) Within the patient and the control group we noticed an increase of SVV aberrations with increased forced tilt angles. In both groups, the increase from 0° to 15° was b1° and not significant. Within the patient group (patients baseline and week 3), as well as within the control group, there was a mean difference of SVV aberrations between 0° and 30° tilt angle of N 3° and between 15° and 30° tilt angle N 2°, which was significant (Fig. 3). 3.3. Influence of symptom severity, disease duration and duration of BoNT therapy We observed a moderate positive correlation between severity of symptoms (reflected by increasing Tsui score) and SVV aberrations at baseline (Pearson correlation 0.61; p = 0.0002; Fig. 4). No correlation between disease duration and duration of BoNT therapy was found. 4. Discussion The main aim of this study was to assess static graviceptive perception in patients with CD and to evaluate whether this perceptual function may be influenced by means of BoNT therapy. We have demonstrated in our study that, when tested with their abnormal, habitual head posture, the patient's perception of their SVV differed significantly from that of healthy subjects. In addition, a moderate positive correlation between the severity of symptoms (reflected by increasing Tsui score) and SVV aberration was found. Our data also revealed that modulation of the somatosensory input (in particular the head fixation in upright position) may overcome this effect, but might not fully explain that phenomenon, which will be discussed below. Furthermore, an overshooting E-effect could be demonstrated at head tilts of 30°. Our findings confirmed the plasticity of static graviceptive function in CD, as their SVV estimates in habitual head position and at 30° fixed head position did not differ anymore from those of controls three weeks after BoNT injection, i.e. at time of maximal treatment response. The data provided evidence that BoNT therapy in CD not only improves dystonic head posture but also static graviceptive function and the E-effect (i.e. an overshooting of the SVV with a lateral head tilt), likely caused by modulation of the proprioception. In our study, controls were able to adjust their SVV estimates (in upright head position) close to the objective vertical, with a range from 0° to 3°, corresponding with normative data from age-matched healthy subjects [16]. At 15° head tilts no significant differences in the estimates were found compared to the head-upright position. Tilt angles of 30°, however, yielded significant aberrations of the adjustments in controls, which may be explained by the physiological E-effect. This effect has been shown in previous experiments in healthy subjects testing the SVV at 30° head tilt [2,17,18]. The comparison of our findings in the patient group with previous results is hampered, not least because of the different experimental setups and the techniques applied for the assessment of verticality
K. Platho-Elwischger et al. / Journal of the Neurological Sciences 373 (2017) 230–235
233
Table 2 Descriptive data of absolute SVV aberration and within group differences. Patients
N
Baseline
32
Week 3
30
Controls
13
Head tilt
0° 15° 30° 0° 15° 30° 0° 15° 30°
SVV median (95% CI)
2.00 (0.33 to 9.30) 2.25 (0.34 to 7.83) 6.42 (2.50 to 15.0) 2.42 (0.33 to 8.33) 2.17 (0.16 to 9.50) 4.75 (0.34 to 18.3) 1.33 (0 to 3.00) 1.66 (0.34 to 5.34) 5.33 (0.17 to 9.84)
Within group differences Mean difference (95% CI)
p-Values
0° vs. 15°: −0.67° (−2.15 to 0.81) 15° vs. 30°: −4.51° (−5.99 to −3.02) 0° vs. 30°: −5.18° (−6.66 to −3.69) 0° vs. 15°: −0.42° (−2.20 to 1.36) 15 vs. 30°: −3.38° (−5.16 to −1.59) 0° vs. 30°: −3.80° (−5.58 to −2.02) 0° vs. 15°: −0.85° (−2.1 to 0.36) 15 vs. 30°: −2.31° (−3.72 to −0.90) 0° vs. 30°: −3.17° (−5.39 to −0.94)
0.3729 b0.001⁎ b0.001⁎ 0.6363 0.0003⁎
b0.0001⁎ 0.1525 0.0039⁎ 0.0091⁎
⁎ Statistically significant (p b 0.05). For comparison of differences repeated measures ANOVA was performed. SVV is measured in degree (°). CI: confidence interval; N: number of patients; SVV: subjective visual vertical.
perception in former studies. Previous reports did not find any abnormality of static graviceptive function in CD [13]. In the study by Straube et al., the SVV has been assessed in a head-fixed condition without head tilt. These findings correspond with our results in the patient group when the head was fixed at 0° head tilt at both time points, at baseline examination, and 3 weeks after BoNT injection. Here, 14 of 19 patients had SVV aberrations of b2.5°, the rest showed SVV responses with a maximum of 5.9°. However, when our patients were tested with their habitual head posture, their SVV estimates differed significantly from those of controls, which implies that CD patients exhibit altered static graviceptive processing. Also, a moderate positive correlation between symptom severity and SVV aberrations was found. It might be argued in this regard, that the upright head position of healthy subjects may not be compared with the habitual head position of CD patients. However, the argument that the patient's significant SVV aberrations in the habitual head position may be explained by the E-effect can be refuted by the fact that the E-effect could only be elicited at 30° head tilt in the standardized (i.e. head-fixed) condition in patients and in controls in our study, which is also in line with previous data in healthy subjects [2,17,18]. In addition, the E-effect has been overshooting in the patient-group. Furthermore, the assessment of the habitual head tilt angle at baseline examination confirmed that the mean head tilt of our patients was 10°(4.8) (tilted to the left) and 8.5°(7.7) (tilted to the right) and thus definitely below 30°. To our knowledge no significant E-effect has been reported at these small tilt angles in previous studies. Another study, which applied a “Rod and Frame Test” for the assessment of the SVV in CD, found no statistical differences in the absolute errors between patients and controls [14]. Furthermore, these authors
found that whole body postural control is not affected in CD. Although the SVV was also assessed at the patient's habitual head posture in this study, the results should not be directly compared with our data due to the different experimental setup, in particular different environmental or orientational cues in the rod and frame test may have biased the patient's judgements. To our knowledge, there is only one study that investigated the SVV in patients with CD also in lateral tilt [7]. However, Anastasopoulos et al. applied whole body tilts at 80–90° for the evaluation of static graviceptive function. Also, judgements were made against a wholefield random-dot background that was either stationary or rotating around the line of sight. The authors found that CD patients set the SVV normally, unaffected by their chronically maintained head deviation and that they exhibit the A-effect during whole body tilts at 80–90°. In contrast, our findings implied that CD patients exhibit altered graviceptive perception at their habitual head posture which may be overcome when the head is being fixed at upright (0°) head position. The latter may not only be explained by a “correction” of otolithic input in the upright head position. As the perception of verticality is mediated by multisensory inputs, it is conceivable, that other sensory mechanisms are also contributing to this phenomenon. It is well established in this regard, that CD changes proprioceptive input of the neck by abnormal muscle activity [1]. It has been suggested that abnormal co-contraction of antagonists results in abnormal processing of muscle spindle input [19]. In the present experimental setting, an additional somatosensory input was provided by the patient's head fixation. This might have produced the so-called “sensory trick” [20]. However, sensory cues have been shown to induce a significant reduction of
Fig. 2. Forced head positions – allocated to head tilts at 0°,15° and 30°. Horizontal axis, left to right: SVV estimates of patients at baseline and 3 weeks after BoNT injection in forced head tilt positions, compared with controls. Vertical axis: Absolute SVV deviations from gravitational vertical in degrees. The head was tilted (a) 0°, (b) 15°, and (c) 30° in CW and CCW directions in the roll plane. (b) and (c) show the summation of SVV responses irrespective of direction of forced head tilt. At baseline, the patient's SVV estimates showed a significant difference at 30° head tilt, when compared with controls (Fig. 2c). CCW: counter clockwise; CD: cervical dystonia; CW: clockwise; SVV: subjective visual vertical.
234
K. Platho-Elwischger et al. / Journal of the Neurological Sciences 373 (2017) 230–235
Fig. 3. Forced head positions – allocated to controls, pre- and posttreatment-group. Horizontal axis, left to right: SVV estimates of (a) controls at 0°, 15° and 30°, (b) patients at baseline (“CD baseline”) at 0°, 15° and 30° and (c) three weeks after injection (“CD week 3”) at 0°, 15° and 30°. Vertical axis: Absolute SVV deviations from gravitational vertical in degrees. CD: cervical dystonia; SVV: subjective visual vertical.
electromyography (EMG) activity in neck muscle's agonists and antagonists [21]. Furthermore, even the voluntary movement of the head in midline position causes significant reduction of EMG activity [22]. Consequently, altered muscle activity and reduction of proprioceptive input might be plausible mechanisms of these short-term adaptations of verticality perception. Interestingly, three weeks after BoNT treatment, SVV response at 0° fixation differed significantly from those of controls. The current study cannot explain this finding and further research is required. At fixed head-tilt angles of 15° the patient's estimates did not differ significantly from those of controls, both at baseline and after BoNT application. This shows, that graviceptive perception in CD patients is normal at small head tilts, if a head-fixed condition with additional sensory cues is applied. When comparing within-group changes at different head-tilt angles, both, in the patient and control groups, significant aberrations of estimates occurred only when the head was tilted at least 30° (with even overshooting aberrations in the patient group at baseline). This demonstrates, that the E-effect can be seen both, in controls
and CD patients starting at head tilt angles of 30° in a head-fixed condition. The effect of BoNT therapy on static graviceptive function has to our knowledge not yet been examined in CD patients. In the present study, symptomatic BoNT therapy improved the patient's verticality perception in habitual head posture and the overshooting E-effect with longterm effect, as their SVV estimates did not differ from healthy subjects 3 weeks after BoNT injection. Symptomatic therapy with BoNT denervates extra- and intrafusal muscle fibres, i.e. structures, that are involved in proprioception [19,23]. Altered propriception of the neck muscles might change central sensorimotor integration [19]. BoNT might in this regard act similar to the “sensory trick” and may thus account for the plasticity of sensorimotor function. Plasticity on the level of the cortex, more precisely the premotor-cortex area, has been shown recently. After symptomatic treatment of CD with BoNT, a reversible reorganisation of areas involved in motor planning and spatial cognition has been demonstrated [24]. Furthermore, when comparing SVV judgements at different head-tilt angles, patients treated with BoNT exhibited the same SVV judgement patterns as at baseline examination and as controls, i.e. significant SVV aberrations occurred in each group only at 30° head-tilt. This demonstrates, in our opinion, that BoNT cannot reverse errors in the perception of verticality at certain head-tilt angles, which can be explained physiologically by the E-effect both in CD patients and in healthy subjects.
5. Conclusion In conclusion, altered verticality perception in CD patients appears to be a secondary phenomenon, caused by abnormal increase of proprioceptive input of neck muscles. This is supported by a positive correlation between symptom severity and impairment of verticality perception. This graviceptive perceptual deficit can be reversed by modification of the somatosensory input (i.e. head fixation) as a short-term effect, and by changes in the proprioceptive input (i.e. BoNT injections) as a long-term effect.
Individual contributions to the manuscript
Fig. 4. Correlation of symptom severity and SVV aberration. Horizontal axis: Patient's TSUI Score at baseline assessment. Vertical axis: Absolute SVV deviations from gravitational vertical in degrees (°). Positive Pearson correlation 0.61. SVV: subjective visual vertical.
Platho-Elwischger K: design and conception of the study, analysis and interpretation of data, drafting or revising the manuscript for intellectual content. Kranz G: design and conception of the study, drafting or revising the manuscript for intellectual content. Sycha T: drafting or revising the manuscript for intellectual content.
K. Platho-Elwischger et al. / Journal of the Neurological Sciences 373 (2017) 230–235
Dunkler D: design and conception of the study, analysis and interpretation of data, drafting or revising the manuscript for intellectual content. Rommer P: drafting or revising the manuscript for intellectual content. Mueller C: drafting or revising the manuscript for intellectual content. Auff E: drafting or revising the manuscript for intellectual content. Wiest G: design and conception of the study, analysis and interpretation of data, drafting or revising the manuscript for intellectual content. Disclosures Dr. Platho-Elwischger received funding for travel from Ipsen, Merz Pharmaceuticals, Allergan and the Germain Society for Neurology and speaker honoria from Ipsen. Dr. Kranz received funding for travel from the NIH/NINDS and Ipsen, speaker honoraria from Allergan, Inc., Ipsen, and Merz Pharmaceuticals, LLC, and research support from the Max Kade Foundation and the NIH/ NINDS (Intramural Grant). Dr. Sycha has nothing to disclose. Dr. Dunkler has nothing to disclose. Dr. Rommer has nothing to disclose. Dr. Mueller has nothing to disclose. Dr. Auff has nothing to disclose. Dr. Wiest has nothing to disclose. Funding Trial center and sponsor: Medical University of Vienna, Department of Neurology. The study has not been funded by external funds or grants. References [1] A.M. Bronstein, The interaction of otolith and proprioceptive information in the perception of verticality. The effects of labyrinthine and CNS disease, Ann. N. Y. Acad. Sci. 871 (1999) 324–333. [2] A.A. Tarnutzer, C.J. Bockisch, D. Straumann, Roll-dependent modulation of the subjective visual vertical: contributions of head- and trunk-based signals, J. Neurophysiol. 103 (2010) 934–941, http://dx.doi.org/10.1152/jn.00407.2009. [3] R. Held, H. Leibowitz, H.-L. Teubel, Handbook of sensory physiology, Perception, Volume 8, Springer Verlag, Berlin Heidelberg New York, 1978.
235
[4] R.T. Dyde, M.R. Jenkin, L.R. Harris, The subjective visual vertical and the perceptual upright, Exp. Brain Res. 173 (2006) 612–622. [5] A.D. Van Beuzekom, W.P. Medendorp, J.A. Van Gisbergen, The subjective vertical and the sense of self orientation during active body tilt, Vis. Res. 41 (2001) 3229–3242. [6] U. Schönfeld, A.H. Clarke, A clinical study of the subjective visual vertical during unilateral centrifugation and static tilt, Acta Otolaryngol. 131 (2011) 1040–1050, http://dx.doi.org/10.3109/00016489.2011.584902. [7] D. Anastasopoulos, K. Bhatia, A.M. Bronstein, C.D. Marsden, M.A. Gresty, Perception of spatial orientation in spasmodic torticollis. Part 2: the visual vertical, Mov. Disord. 12 (1997) 709–714. [8] L. Yardley, Contribution of somatosensory information to perception of the visual vertical with body tilt and rotating visual field, Percept. Psychophys. 48 (1990) 131–134. [9] G.J. McKenna, G.C. Peng, D.S. Zee, Neck muscle vibration alters visually perceived roll in normals, J. Assoc. Res. Otolaryngol. 5 (2004) 25–31. [10] T. Kawase, A. Maki, Y. Takata, H. Miyazaki, T. Kobayashi, Effects of neck muscle vibration on subjective visual vertical: comparative analysis with effects on nystagmus, Eur. Arch. Otorhinolaryngol. 268 (2011) 823–827, http://dx.doi.org/10.1007/ s00405-010-1467-9. [11] R. Stell, A.M. Bronstein, C.D. Marsden, Vestibulo-ocular abnormalities in spasmodic torticollis before and after botulinum toxin injections, J. Neurol. Neurosurg. Psychiatry 52 (1989) 57–62. [12] A. Münchau, A.M. Bronstein, Role of the vestibular system in the pathophysiology of spasmodic torticollis, J. Neurol. Neurosurg. Psychiatry 71 (2001) 285–288. [13] A. Straube, M. Dieterich, Neuro-ophthalmologic and posturographic studies of patients with idiopathic torticollis, Nervenarzt 64 (1993) 787–792. [14] F. Vacherot, M. Vaugoyeau, S. Mallau, S. Soulayrol, C. Assaiante, J.P. Azulay, Postural control and sensory integration in cervical dystonia, Clin. Neurophysiol. 118 (2007) 1019–1027. [15] J.K. Tsui, A. Eisen, A.J. Stoessl, S. Calne, D.B. Calne, Double-blind study of botulinum toxin in spasmodic torticollis, Lancet 2 (1986) 245–247. [16] H. Kobayashi, Y. Hayashi, K. Higashino, et al., Dynamic and static subjective visual vertical with aging, Auris Nasus Larynx 29 (2002) 325–328. [17] M. Hoppenbrouwers, F.L. Wuyts, P.H. Van de Heyning, Suppression of the E-effect during the subjective visual vertical test, Neuroreport 15 (2004) 325–327. [18] N.J. Wade, R.H. Day, Development and dissipation of a visual spatial aftereffect from prolonged head tilt, J. Exp. Psychol. 76 (1968) 439–443. [19] R.L. Rosales, D. Dressler, On muscle spindles, dystonia and botulinum toxin, Eur. J. Neurol. 17 (2010) 71–80, http://dx.doi.org/10.1111/j.1468-1331.2010.03056.x. [20] A. Poisson, P. Krack, S. Thobois, et al., History of the ‘geste antagoniste’ sign in cervical dystonia, J. Neurol. 259 (2012) 1580–1584, http://dx.doi.org/10.1007/s00415011-6380-7. [21] A. Schramm, K. Reiners, M. Naumann, Complex mechanisms of sensory tricks in cervical dystonia, Mov. Disord. 19 (2004) 452–458. [22] A.S. Buchman, C.L. Comella, S. Leurgans, G.T. Stebbins, C.G. Goetz, The effect of changes in head posture on the patterns of muscle activity in cervical dystonia (CD), Mov. Disord. 13 (1998) 490–496. [23] C. Trompetto, A. Currà, A. Buccolieri, A. Suppa, G. Abbruzzese, A. Berardelli, Botulinum toxin changes intrafusal feedback in dystonia: a study with the tonic vibration reflex, Mov. Disord. 21 (2006) 777–782. [24] C.C. Delnooz, J.W. Pasman, B.P. van de Warrenburg, Dynamic cortical gray matter volume changes after botulinum toxin in cervical dystonia, Neurobiol. Dis. 73 (2014) 327–333, http://dx.doi.org/10.1016/j.nbd.2014.10.013.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Plasticity of static graviceptive function in patients with cervical dystonia Kirsten Platho-Elwischger a,⁎,1, Gottfried Kranz a, Thomas Sycha a, Daniela Dunkler b, Paulus Rommer a, Christian Mueller a, Eduard Auff a, Gerald Wiest a a b
Department of Neurology, Medical University of Vienna, Waehringer Guertel 18-20, 1090 Vienna, Austria Center for Medical Statistics, Informatics, and Intelligent Systems, Section for Clinical Biometrics, Medical University of Vienna; Spitalgasse 23, BT88/E 03, 1090 Vienna, Austria
a r t i c l e
i n f o
Article history: Received 11 March 2016 Received in revised form 29 December 2016 Accepted 3 January 2017 Available online 4 January 2017 Keywords: Dystonia Botulinum toxin Plasticity Subjective visual vertical Graviceptive function
a b s t r a c t Objective: Botulinum toxin (BoNT) is effective in improving abnormal head posture in cervical dystonia (CD) within a period of several weeks to an upright position. These dynamic alterations over time represent a unique model to study the plasticity of static graviceptive function in CD by assessing the subjective visual vertical (SVV). Methods: SVV was assessed in 30 CD patients and 13 controls, in their habitual head posture and with fixed head positions at different angles of head tilt. The patients were tested before and 3 weeks after BoNT administration. Results: At baseline, the patient's estimates in their habitual head posture and at forced head tilt angle of 30° differed significantly from those of controls. This effect was no longer visible after BoNT injection, or when the patient's head was fixed in an upright, and at a 15° head tilt position, respectively. A moderate positive correlation between disease severity and SVV aberrations was found. When comparing within-group changes, in all participants significant aberrations occurred when the head was tilted at 30°, which may be explained by the physiological E-effect. Conclusion: Our findings demonstrate that patients with CD exhibit altered static graviceptive perception when tested in their habitual head posture and an overshooting E-effect at the major forced head tilt of 30°. This graviceptive perceptual deficit can be reversed by modification of the somatosensory input (i.e. head fixation) as a short-term effect, and by changes in the proprioceptive input (i.e. BoNT injections) as a long-term effect. © 2017 Elsevier B.V. All rights reserved.
1. Introduction The evaluation of verticality perception is usually performed by the assessment of the SVV, in healthy individuals with a maximum aberration of b2° from the ‘true’ gravitational vertical [1,2]. Major aberrations of the SVV, up to 10–30° in the direction of tilt, can be produced by a body tilt of 80–90° to one side, which is also referred to as the “A-effect” [1]. Lateral head and body tilt up to 50° leads to perception of the SVV overshooting to the opposite side of the tilt, which is called the “E-effect” [3–6]. In healthy persons, small head roll angles (under 30°) revealed more accurate estimations, close to the gravitational vertical [2]. Verticality perception is mediated by visual, otolithic and Abbreviations: AIC, Akaike information criterion; BoNT, botulinum toxin; CCW, counter clockwise; CD, cervical dystonia; CI, confidence interval; CNS, central nerve system; CW, clockwise; P, p-value; SD, standard deviation; SVV, subjective visual vertical. ⁎ Corresponding author: Kirsten Platho-Elwischger, MD, Koeglergasse 2a, 1120 Vienna, Austria. E-mail addresses: [email protected] (K. Platho-Elwischger), [email protected] (G. Kranz), [email protected] (T. Sycha), [email protected] (D. Dunkler), [email protected] (P. Rommer), [email protected] (C. Mueller), [email protected] (E. Auff), [email protected] (G. Wiest). 1 Present address: Rehabilitation Clinic Meidling, Koeglergasse 2a, 1120 Vienna, Austria.
http://dx.doi.org/10.1016/j.jns.2017.01.007 0022-510X/© 2017 Elsevier B.V. All rights reserved.
somatosensory input [1]. Thereby, the somatosensory system and proprioceptive input from neck muscles, respectively play a major role in the estimation of the SVV during lateral body tilt [1,7–10]. The pathophysiological mechanism of CD still remains unclear. Animal models suggest involvement of multiple cerebral integration centers. The “sensory trick”, supports the assumption of altered sensorimotor processing in CD. In addition, vestibular abnormalities have been suggested [11,12]. Previous studies on the perception of verticality in CD yielded conflicting results [7,13,14]. When treated with BoNT by an experienced clinician, head posture in people with CD usually improves within a period of 3 weeks to an upright position and diminishes subsequently after approximately 9 weeks. These dynamic alterations of head posture over time represent a unique model to study the effects of altered somatosensory and otolithic input on the perception of verticality. In this study we tested the SVV of CD before and after BoNT, in order to evaluate the plasticity of static graviceptive function in these patients. The purpose of the present study was to 1) assess static graviceptive function at habitual head posture and at different head tilts (i.e. short-term modulation of the somatosensory input) in normal subjects and in patients with CD and 2) study the effects of BoNT injections (i.e. long-term modulation of the proprioceptive input) on static graviceptive function in patients with CD.
K. Platho-Elwischger et al. / Journal of the Neurological Sciences 373 (2017) 230–235
2. Materials and methods 2.1. Subjects 32 consecutive patients with idiopathic CD undergoing regular treatment with BoNT at the BoNT Outpatient Clinic of the Department of Neurology, Medical University Vienna, were included in the study. All patients participated at baseline evaluation, however, two patients refused to complete the follow-up examination (3 weeks after baseline evaluation), so that a total of 30 patients completed the whole study (mean age 59 [SD 12] years; females 19, 63.3%). The patient's clinical details are listed in Table 1. At baseline, 17 (56.7%) patients presented with habitual head deviation (chin-nasion line rotated in the roll plane) to the right, and 13 (43.3%) to the left, respectively. Patients with a history of vestibular disorders, as well as patients with head rotation in the horizontal plane exceeding 15° or secondary CD were excluded. 2.2. Standard protocol approval and patients consents The study was approved by the ethical committee of the Medical University of Vienna. All patients and controls signed an informed consent prior to their inclusion in the study. The study has been performed in accordance with the ethical standards laid down in the Declaration of Helsinki. This pilot study has been registered at http://clinicaltrials.gov (NCT01180270).
231
Assessment of SVV, Tsui score and head deviation was performed at the baseline visit directly before routine BoNT treatment (i.e. at time of minimum treatment effect). Three weeks after injection (median 21 days, range 18–28), i.e. at time of maximum treatment response, the initial assessment was repeated. As controls, 13 healthy subjects (age 52.8 [11.6], nine females, 69.2%) without a history of vestibular disorders, with normal vestibular- and ocular motor function tests (System 2000, Micromedical Technologies, Illinois, USA) and no history of CNSactive medications participated in the study. 2.4. Assessment of SVV SVV was assessed in an upright sitting position in a completely dark room. In front of the participants, at a distance of 100 cm, there was a dim light bar, 2 mm wide and 10 cm long, which could be rotated about its midpoint by means of an electronic motor and a remote control device (System 2000, Micromedical Technologies, Illinois, USA). All participants adjusted the bar six times from randomized starting positions for parallel alignment with the perceived gravitational vertical. The six estimates were averaged for further analysis. SVV alignments were performed both by patients and healthy controls at five different head positions (head fixed upright 0°, CW and CCW: fixed head deviation at 15° and fixed head deviation at 30°). A plastic head ring and a hook-and-loop tape that was mounted on an adjustable neck rest (which covered the occiput and the posterior neck) fixed the head. All patients and controls also performed one trial with habitual head posture. The interval between each test/head position was approximately 3 min.
2.3. Procedures 2.5. Statistical analysis Patients were treated with four different preparations of BoNT (Table 1). The median interval between first SVV assessment at baseline and previous BoNT injection was 105 days (range 83–371). The severity of CD was assessed by the Tsui score, head deviation was measured by referencing the chin-nasion line to earth vertical with a finger goniometer (KaWe Germany) [15]. In order to rule out vestibular disorders, which might interfere with SVV perception, such as vestibulopathy or neuritis vestibularis, all patients underwent vestibular- and ocular motor function tests by means of digital videooculography and a rotational chair system (System 2000, Micromedical Technologies, Illinois, USA) at their baseline visit. Table 1 Clinical details of 32 patients at baseline assessment. N (%)
Median (IQR)
Male Female Age in years (a) Duration of disease (a) BoNT treatment duration (a) Severity of disease measured by Tsui score
11 (36.7) 19 (63.3) 30 (100) 30 (100) 30 (100) 30 (100)
59.0 (12.0) 15.6 (9.4) 9.4 (7.0) 6.2 (3.9)
Predominant head deviation in degrees (°) Clockwise head deviation Counterclockwise head deviation
17 (56.7) 13 (43.3)
10.0 (4.8) 8.5 (7.7)
Amount of BoNT preparation applied in MU Abo-BoNT A, Dysport®, Ipsen Inco-BoNT A, Xeomin®, Merz Ona-BoNT A, Botox®, Allergan Rima-BoNT B, NeuroBloc®, USWorldMed
14 (46.7) 7 (23.3) 8 (26.7) 1 (3.3)
579.8 (233.0) 149.6 (59.6) 212.8 (93.4) 15,500
Patients with CNS-active co-medication Total Antidepressant Anticonvulsants Anticholinergics Musclerelaxants Neuroleptics
12 (40.0) 9 (30.0) 4 (13.3) 2 (6.7) 1 (3.3) 0
BoNT: botulinum toxin; MU: mouse units; N: number of patients; SD: standard deviation.
Categorical variables are given as absolute numbers and percentages; continuous variables are given as median and interquartilerange (IQR). SVV was measured in degrees of deviation from gravitational vertical. CCW deviations were calculated as negative values, CW deviations as positive. For the evaluation of differences in the perception of verticality in CD patients and healthy controls the direction of SVV deviations has not been taken into account, i.e. that absolute deviations from earth vertical - irrespective of their direction - were calculated. Controls were assessed once, as no injections were performed, and the median aberrance of the SVV at 0° was very small at baseline measures. For comparison of differences between patients and controls, fixed head positions and visits, repeated measures ANOVA was applied: In the model only for patients deviations, the visits and the interaction of the fixed head positions were included. Two separate models for each visit compared patients with controls (including fixed head positions, deviations of patients and controls, and the interaction). The last model compared the deviations of controls. Covariance structure was selected minimizing AIC. In addition, t-tests or Wilcoxon-tests, depending on the distribution of the respective continuous variable, were used. Pearson correlation was used to measure the association between Tsui score and SVV aberrations. Boxplots and scatter-plots were generated to visualize distributions. A two-sided p b 0.05 was considered statistically significant. The statistical analysis was performed with the SAS system version 9.2 (2008; SAS Institute Inc., Cary, NC). 3. Results 3.1. Absolute SVV deviations at habitual head position Absolute SVV-deviations from true gravitational vertical were calculated irrespective of CW- or CCW direction. At baseline, all CD patients presented with marked abnormal habitual head posture, were reflected by elevated Tsui scores (Table 1). SVV judgments of patients tested without head fixation, i.e. at their “habitual” head position, yielded major aberrations of SVV from earth vertical. The median SVV
232
K. Platho-Elwischger et al. / Journal of the Neurological Sciences 373 (2017) 230–235
judgement was 2.65° (95% CI 0.17 to 7.83). In contrast, healthy controls, tested at neutral head position (approximately 0° head tilt), aligned their SVV responses close to gravitational vertical, with a median of 1.33°([0.97], 95% CI 0 to 3). When compared with the estimates of controls, the median difference of SVV aberrations of patients was −1.34° with a 95% CI from − 2.5 to − 0.33. This difference was significant (p = 0.017; Fig. 1). Three weeks after BoNT injection, the patient's clinical symptoms had significantly improved, reflected by a significant decrease of Tsui score (mean reduction −2.4°[2.12]; p b 0.0001, Paired t-test) and the degree of head deviation in roll plane (mean reduction − 6.23°[4.85]; p b 0.001). The absolute SVV judgements of patients had also improved. The median SVV judgement was 1.42° ([2.05], 95% CI 0 to 6.5). When compared with controls, the SVV difference was not significant anymore (p = 0.2893; Fig. 1). 3.2. Absolute SVV deviations at forced head positions (for descriptive data please see Table 2) 3.2.1. Deviations at baseline and 3 weeks after injection, tilt-angle of 0° At baseline, when the patient's head was forced to upright position (i.e. 0° position), the median SVV response was 2° (95% CI 0.33 to 9.3). There was no significant difference of SVV judgements of patients from those of controls (p = 0.1134; mean difference −−1.2°; 95% CI −2.7° to 3°; repeated measure ANOVA). Three weeks after BoNT injection the median SVV judgement had not changed, the median response was 2.42° (95% CI 0.33 to 8.33). However, when compared with those of controls, a statistically significant difference was found (p = 0.019; mean difference −1.7°, 95% CI −3.11° to −0.3°; Fig. 2a). 3.2.2. Deviations at baseline and 3 weeks after injection, tilt-angle of 15° The median estimates of patients (irrespective of CW or CCW rotation) were 2.25° (95% CI 0.34 to 7.83) at baseline and 2.17° (95% CI 0.16 to 9.5) at week 3. Controls tested at head tilt-angle of 15° estimated SVV with a median of 1.66° (95% CI 0.34 to 5.34). When the patients judgements at baseline, or week 3 were compared with those of controls, no significant difference was detected (Baseline p = 0.2207,
Fig. 1. Habitual head position. Horizontal axis, left to right: SVV estimates in controls (“Control”), patient's estimates at habitual head posture at baseline (“CD baseline”) and 3 weeks after BoNT injection (“CD week 3”). Vertical axis: Absolute SVV deviations from gravitational vertical in degree. Baseline SVV aberrations of patients were significantly deviated when compared with controls (p = 0.017). CD: cervical dystonia; SVV: subjective visual vertical.
mean difference −− 1.01°, 95% CI − 2.65° to 0.63°; 3 weeks after BoNT injection: p = 0.2187; mean difference − 1.23°, 95% CI − 3.22° to 0.76°; repeated measure ANOVA; Fig. 2b). 3.2.3. Deviations at baseline and 3 weeks after injection, tilt-angle of 30° At tilt angle of 30° (irrespective of CW or CCW rotation), the patients' median estimates at baseline were 6.42° (95% CI 2.5 to 15). Median SVV judgements of controls were 5.33° (95% CI 0.17 to 9.84). A significant difference between patients and controls was found (p = 0.026, mean difference −3.20, 95% CI − 6 to −0.4; repeated measure ANOVA; Fig. 2c). Three weeks after BoNT injections the median estimate was 4.75°(0.34 to 18.3) and did not differ significantly from those of controls (p = 0.1913, mean difference −2.31, 95% CI −5.83 to 1.20). 3.2.4. Within-group differences of SVV estimates at different forced head tilt positions (mean differences and p-values are listed in Table 2) Within the patient and the control group we noticed an increase of SVV aberrations with increased forced tilt angles. In both groups, the increase from 0° to 15° was b1° and not significant. Within the patient group (patients baseline and week 3), as well as within the control group, there was a mean difference of SVV aberrations between 0° and 30° tilt angle of N 3° and between 15° and 30° tilt angle N 2°, which was significant (Fig. 3). 3.3. Influence of symptom severity, disease duration and duration of BoNT therapy We observed a moderate positive correlation between severity of symptoms (reflected by increasing Tsui score) and SVV aberrations at baseline (Pearson correlation 0.61; p = 0.0002; Fig. 4). No correlation between disease duration and duration of BoNT therapy was found. 4. Discussion The main aim of this study was to assess static graviceptive perception in patients with CD and to evaluate whether this perceptual function may be influenced by means of BoNT therapy. We have demonstrated in our study that, when tested with their abnormal, habitual head posture, the patient's perception of their SVV differed significantly from that of healthy subjects. In addition, a moderate positive correlation between the severity of symptoms (reflected by increasing Tsui score) and SVV aberration was found. Our data also revealed that modulation of the somatosensory input (in particular the head fixation in upright position) may overcome this effect, but might not fully explain that phenomenon, which will be discussed below. Furthermore, an overshooting E-effect could be demonstrated at head tilts of 30°. Our findings confirmed the plasticity of static graviceptive function in CD, as their SVV estimates in habitual head position and at 30° fixed head position did not differ anymore from those of controls three weeks after BoNT injection, i.e. at time of maximal treatment response. The data provided evidence that BoNT therapy in CD not only improves dystonic head posture but also static graviceptive function and the E-effect (i.e. an overshooting of the SVV with a lateral head tilt), likely caused by modulation of the proprioception. In our study, controls were able to adjust their SVV estimates (in upright head position) close to the objective vertical, with a range from 0° to 3°, corresponding with normative data from age-matched healthy subjects [16]. At 15° head tilts no significant differences in the estimates were found compared to the head-upright position. Tilt angles of 30°, however, yielded significant aberrations of the adjustments in controls, which may be explained by the physiological E-effect. This effect has been shown in previous experiments in healthy subjects testing the SVV at 30° head tilt [2,17,18]. The comparison of our findings in the patient group with previous results is hampered, not least because of the different experimental setups and the techniques applied for the assessment of verticality
K. Platho-Elwischger et al. / Journal of the Neurological Sciences 373 (2017) 230–235
233
Table 2 Descriptive data of absolute SVV aberration and within group differences. Patients
N
Baseline
32
Week 3
30
Controls
13
Head tilt
0° 15° 30° 0° 15° 30° 0° 15° 30°
SVV median (95% CI)
2.00 (0.33 to 9.30) 2.25 (0.34 to 7.83) 6.42 (2.50 to 15.0) 2.42 (0.33 to 8.33) 2.17 (0.16 to 9.50) 4.75 (0.34 to 18.3) 1.33 (0 to 3.00) 1.66 (0.34 to 5.34) 5.33 (0.17 to 9.84)
Within group differences Mean difference (95% CI)
p-Values
0° vs. 15°: −0.67° (−2.15 to 0.81) 15° vs. 30°: −4.51° (−5.99 to −3.02) 0° vs. 30°: −5.18° (−6.66 to −3.69) 0° vs. 15°: −0.42° (−2.20 to 1.36) 15 vs. 30°: −3.38° (−5.16 to −1.59) 0° vs. 30°: −3.80° (−5.58 to −2.02) 0° vs. 15°: −0.85° (−2.1 to 0.36) 15 vs. 30°: −2.31° (−3.72 to −0.90) 0° vs. 30°: −3.17° (−5.39 to −0.94)
0.3729 b0.001⁎ b0.001⁎ 0.6363 0.0003⁎
b0.0001⁎ 0.1525 0.0039⁎ 0.0091⁎
⁎ Statistically significant (p b 0.05). For comparison of differences repeated measures ANOVA was performed. SVV is measured in degree (°). CI: confidence interval; N: number of patients; SVV: subjective visual vertical.
perception in former studies. Previous reports did not find any abnormality of static graviceptive function in CD [13]. In the study by Straube et al., the SVV has been assessed in a head-fixed condition without head tilt. These findings correspond with our results in the patient group when the head was fixed at 0° head tilt at both time points, at baseline examination, and 3 weeks after BoNT injection. Here, 14 of 19 patients had SVV aberrations of b2.5°, the rest showed SVV responses with a maximum of 5.9°. However, when our patients were tested with their habitual head posture, their SVV estimates differed significantly from those of controls, which implies that CD patients exhibit altered static graviceptive processing. Also, a moderate positive correlation between symptom severity and SVV aberrations was found. It might be argued in this regard, that the upright head position of healthy subjects may not be compared with the habitual head position of CD patients. However, the argument that the patient's significant SVV aberrations in the habitual head position may be explained by the E-effect can be refuted by the fact that the E-effect could only be elicited at 30° head tilt in the standardized (i.e. head-fixed) condition in patients and in controls in our study, which is also in line with previous data in healthy subjects [2,17,18]. In addition, the E-effect has been overshooting in the patient-group. Furthermore, the assessment of the habitual head tilt angle at baseline examination confirmed that the mean head tilt of our patients was 10°(4.8) (tilted to the left) and 8.5°(7.7) (tilted to the right) and thus definitely below 30°. To our knowledge no significant E-effect has been reported at these small tilt angles in previous studies. Another study, which applied a “Rod and Frame Test” for the assessment of the SVV in CD, found no statistical differences in the absolute errors between patients and controls [14]. Furthermore, these authors
found that whole body postural control is not affected in CD. Although the SVV was also assessed at the patient's habitual head posture in this study, the results should not be directly compared with our data due to the different experimental setup, in particular different environmental or orientational cues in the rod and frame test may have biased the patient's judgements. To our knowledge, there is only one study that investigated the SVV in patients with CD also in lateral tilt [7]. However, Anastasopoulos et al. applied whole body tilts at 80–90° for the evaluation of static graviceptive function. Also, judgements were made against a wholefield random-dot background that was either stationary or rotating around the line of sight. The authors found that CD patients set the SVV normally, unaffected by their chronically maintained head deviation and that they exhibit the A-effect during whole body tilts at 80–90°. In contrast, our findings implied that CD patients exhibit altered graviceptive perception at their habitual head posture which may be overcome when the head is being fixed at upright (0°) head position. The latter may not only be explained by a “correction” of otolithic input in the upright head position. As the perception of verticality is mediated by multisensory inputs, it is conceivable, that other sensory mechanisms are also contributing to this phenomenon. It is well established in this regard, that CD changes proprioceptive input of the neck by abnormal muscle activity [1]. It has been suggested that abnormal co-contraction of antagonists results in abnormal processing of muscle spindle input [19]. In the present experimental setting, an additional somatosensory input was provided by the patient's head fixation. This might have produced the so-called “sensory trick” [20]. However, sensory cues have been shown to induce a significant reduction of
Fig. 2. Forced head positions – allocated to head tilts at 0°,15° and 30°. Horizontal axis, left to right: SVV estimates of patients at baseline and 3 weeks after BoNT injection in forced head tilt positions, compared with controls. Vertical axis: Absolute SVV deviations from gravitational vertical in degrees. The head was tilted (a) 0°, (b) 15°, and (c) 30° in CW and CCW directions in the roll plane. (b) and (c) show the summation of SVV responses irrespective of direction of forced head tilt. At baseline, the patient's SVV estimates showed a significant difference at 30° head tilt, when compared with controls (Fig. 2c). CCW: counter clockwise; CD: cervical dystonia; CW: clockwise; SVV: subjective visual vertical.
234
K. Platho-Elwischger et al. / Journal of the Neurological Sciences 373 (2017) 230–235
Fig. 3. Forced head positions – allocated to controls, pre- and posttreatment-group. Horizontal axis, left to right: SVV estimates of (a) controls at 0°, 15° and 30°, (b) patients at baseline (“CD baseline”) at 0°, 15° and 30° and (c) three weeks after injection (“CD week 3”) at 0°, 15° and 30°. Vertical axis: Absolute SVV deviations from gravitational vertical in degrees. CD: cervical dystonia; SVV: subjective visual vertical.
electromyography (EMG) activity in neck muscle's agonists and antagonists [21]. Furthermore, even the voluntary movement of the head in midline position causes significant reduction of EMG activity [22]. Consequently, altered muscle activity and reduction of proprioceptive input might be plausible mechanisms of these short-term adaptations of verticality perception. Interestingly, three weeks after BoNT treatment, SVV response at 0° fixation differed significantly from those of controls. The current study cannot explain this finding and further research is required. At fixed head-tilt angles of 15° the patient's estimates did not differ significantly from those of controls, both at baseline and after BoNT application. This shows, that graviceptive perception in CD patients is normal at small head tilts, if a head-fixed condition with additional sensory cues is applied. When comparing within-group changes at different head-tilt angles, both, in the patient and control groups, significant aberrations of estimates occurred only when the head was tilted at least 30° (with even overshooting aberrations in the patient group at baseline). This demonstrates, that the E-effect can be seen both, in controls
and CD patients starting at head tilt angles of 30° in a head-fixed condition. The effect of BoNT therapy on static graviceptive function has to our knowledge not yet been examined in CD patients. In the present study, symptomatic BoNT therapy improved the patient's verticality perception in habitual head posture and the overshooting E-effect with longterm effect, as their SVV estimates did not differ from healthy subjects 3 weeks after BoNT injection. Symptomatic therapy with BoNT denervates extra- and intrafusal muscle fibres, i.e. structures, that are involved in proprioception [19,23]. Altered propriception of the neck muscles might change central sensorimotor integration [19]. BoNT might in this regard act similar to the “sensory trick” and may thus account for the plasticity of sensorimotor function. Plasticity on the level of the cortex, more precisely the premotor-cortex area, has been shown recently. After symptomatic treatment of CD with BoNT, a reversible reorganisation of areas involved in motor planning and spatial cognition has been demonstrated [24]. Furthermore, when comparing SVV judgements at different head-tilt angles, patients treated with BoNT exhibited the same SVV judgement patterns as at baseline examination and as controls, i.e. significant SVV aberrations occurred in each group only at 30° head-tilt. This demonstrates, in our opinion, that BoNT cannot reverse errors in the perception of verticality at certain head-tilt angles, which can be explained physiologically by the E-effect both in CD patients and in healthy subjects.
5. Conclusion In conclusion, altered verticality perception in CD patients appears to be a secondary phenomenon, caused by abnormal increase of proprioceptive input of neck muscles. This is supported by a positive correlation between symptom severity and impairment of verticality perception. This graviceptive perceptual deficit can be reversed by modification of the somatosensory input (i.e. head fixation) as a short-term effect, and by changes in the proprioceptive input (i.e. BoNT injections) as a long-term effect.
Individual contributions to the manuscript
Fig. 4. Correlation of symptom severity and SVV aberration. Horizontal axis: Patient's TSUI Score at baseline assessment. Vertical axis: Absolute SVV deviations from gravitational vertical in degrees (°). Positive Pearson correlation 0.61. SVV: subjective visual vertical.
Platho-Elwischger K: design and conception of the study, analysis and interpretation of data, drafting or revising the manuscript for intellectual content. Kranz G: design and conception of the study, drafting or revising the manuscript for intellectual content. Sycha T: drafting or revising the manuscript for intellectual content.
K. Platho-Elwischger et al. / Journal of the Neurological Sciences 373 (2017) 230–235
Dunkler D: design and conception of the study, analysis and interpretation of data, drafting or revising the manuscript for intellectual content. Rommer P: drafting or revising the manuscript for intellectual content. Mueller C: drafting or revising the manuscript for intellectual content. Auff E: drafting or revising the manuscript for intellectual content. Wiest G: design and conception of the study, analysis and interpretation of data, drafting or revising the manuscript for intellectual content. Disclosures Dr. Platho-Elwischger received funding for travel from Ipsen, Merz Pharmaceuticals, Allergan and the Germain Society for Neurology and speaker honoria from Ipsen. Dr. Kranz received funding for travel from the NIH/NINDS and Ipsen, speaker honoraria from Allergan, Inc., Ipsen, and Merz Pharmaceuticals, LLC, and research support from the Max Kade Foundation and the NIH/ NINDS (Intramural Grant). Dr. Sycha has nothing to disclose. Dr. Dunkler has nothing to disclose. Dr. Rommer has nothing to disclose. Dr. Mueller has nothing to disclose. Dr. Auff has nothing to disclose. Dr. Wiest has nothing to disclose. Funding Trial center and sponsor: Medical University of Vienna, Department of Neurology. The study has not been funded by external funds or grants. References [1] A.M. Bronstein, The interaction of otolith and proprioceptive information in the perception of verticality. The effects of labyrinthine and CNS disease, Ann. N. Y. Acad. Sci. 871 (1999) 324–333. [2] A.A. Tarnutzer, C.J. Bockisch, D. Straumann, Roll-dependent modulation of the subjective visual vertical: contributions of head- and trunk-based signals, J. Neurophysiol. 103 (2010) 934–941, http://dx.doi.org/10.1152/jn.00407.2009. [3] R. Held, H. Leibowitz, H.-L. Teubel, Handbook of sensory physiology, Perception, Volume 8, Springer Verlag, Berlin Heidelberg New York, 1978.
235
[4] R.T. Dyde, M.R. Jenkin, L.R. Harris, The subjective visual vertical and the perceptual upright, Exp. Brain Res. 173 (2006) 612–622. [5] A.D. Van Beuzekom, W.P. Medendorp, J.A. Van Gisbergen, The subjective vertical and the sense of self orientation during active body tilt, Vis. Res. 41 (2001) 3229–3242. [6] U. Schönfeld, A.H. Clarke, A clinical study of the subjective visual vertical during unilateral centrifugation and static tilt, Acta Otolaryngol. 131 (2011) 1040–1050, http://dx.doi.org/10.3109/00016489.2011.584902. [7] D. Anastasopoulos, K. Bhatia, A.M. Bronstein, C.D. Marsden, M.A. Gresty, Perception of spatial orientation in spasmodic torticollis. Part 2: the visual vertical, Mov. Disord. 12 (1997) 709–714. [8] L. Yardley, Contribution of somatosensory information to perception of the visual vertical with body tilt and rotating visual field, Percept. Psychophys. 48 (1990) 131–134. [9] G.J. McKenna, G.C. Peng, D.S. Zee, Neck muscle vibration alters visually perceived roll in normals, J. Assoc. Res. Otolaryngol. 5 (2004) 25–31. [10] T. Kawase, A. Maki, Y. Takata, H. Miyazaki, T. Kobayashi, Effects of neck muscle vibration on subjective visual vertical: comparative analysis with effects on nystagmus, Eur. Arch. Otorhinolaryngol. 268 (2011) 823–827, http://dx.doi.org/10.1007/ s00405-010-1467-9. [11] R. Stell, A.M. Bronstein, C.D. Marsden, Vestibulo-ocular abnormalities in spasmodic torticollis before and after botulinum toxin injections, J. Neurol. Neurosurg. Psychiatry 52 (1989) 57–62. [12] A. Münchau, A.M. Bronstein, Role of the vestibular system in the pathophysiology of spasmodic torticollis, J. Neurol. Neurosurg. Psychiatry 71 (2001) 285–288. [13] A. Straube, M. Dieterich, Neuro-ophthalmologic and posturographic studies of patients with idiopathic torticollis, Nervenarzt 64 (1993) 787–792. [14] F. Vacherot, M. Vaugoyeau, S. Mallau, S. Soulayrol, C. Assaiante, J.P. Azulay, Postural control and sensory integration in cervical dystonia, Clin. Neurophysiol. 118 (2007) 1019–1027. [15] J.K. Tsui, A. Eisen, A.J. Stoessl, S. Calne, D.B. Calne, Double-blind study of botulinum toxin in spasmodic torticollis, Lancet 2 (1986) 245–247. [16] H. Kobayashi, Y. Hayashi, K. Higashino, et al., Dynamic and static subjective visual vertical with aging, Auris Nasus Larynx 29 (2002) 325–328. [17] M. Hoppenbrouwers, F.L. Wuyts, P.H. Van de Heyning, Suppression of the E-effect during the subjective visual vertical test, Neuroreport 15 (2004) 325–327. [18] N.J. Wade, R.H. Day, Development and dissipation of a visual spatial aftereffect from prolonged head tilt, J. Exp. Psychol. 76 (1968) 439–443. [19] R.L. Rosales, D. Dressler, On muscle spindles, dystonia and botulinum toxin, Eur. J. Neurol. 17 (2010) 71–80, http://dx.doi.org/10.1111/j.1468-1331.2010.03056.x. [20] A. Poisson, P. Krack, S. Thobois, et al., History of the ‘geste antagoniste’ sign in cervical dystonia, J. Neurol. 259 (2012) 1580–1584, http://dx.doi.org/10.1007/s00415011-6380-7. [21] A. Schramm, K. Reiners, M. Naumann, Complex mechanisms of sensory tricks in cervical dystonia, Mov. Disord. 19 (2004) 452–458. [22] A.S. Buchman, C.L. Comella, S. Leurgans, G.T. Stebbins, C.G. Goetz, The effect of changes in head posture on the patterns of muscle activity in cervical dystonia (CD), Mov. Disord. 13 (1998) 490–496. [23] C. Trompetto, A. Currà, A. Buccolieri, A. Suppa, G. Abbruzzese, A. Berardelli, Botulinum toxin changes intrafusal feedback in dystonia: a study with the tonic vibration reflex, Mov. Disord. 21 (2006) 777–782. [24] C.C. Delnooz, J.W. Pasman, B.P. van de Warrenburg, Dynamic cortical gray matter volume changes after botulinum toxin in cervical dystonia, Neurobiol. Dis. 73 (2014) 327–333, http://dx.doi.org/10.1016/j.nbd.2014.10.013.