- Email: [email protected]
Postural reactions to soleus muscle vibration in Parkinson's disease: Scaling deteriorates as disease progresses
Neuroscience Letters 401 (2006) 92–96
Postural reactions to soleus muscle vibration in Parkinson’s disease: Scaling deteriorates as disease progresses Peter Valkoviˇc a,b , Siegbert Krafczyk a , Kai B¨otzel a,∗ a
b
Department of Neurology, Ludwig-Maximilians University, Marchioninistraße 15, 81366 Munich, Germany 2nd Department of Neurology, School of Medicine, Comenius University, Limbov´a 5, 85104 Bratislava, Slovakia Received 30 December 2005; received in revised form 13 February 2006; accepted 24 February 2006
Abstract Previous research has shown that Parkinson’s disease (PD) patients, especially those with postural instability, respond hyperactively to visual, vestibular, and neck proprioceptive sensory manipulation. To determine if this impairment of the sensory information scaling holds true for the lower leg proprioceptive system, we studied postural responses to mechanical vibration (which affects the muscle spindle Ia afferents) applied to the soleus muscles of PD subjects and healthy controls. Early-stage and advanced-stage PD patients as well as age-matched control subjects participated. Each group comprised 11 subjects. Nine pulses of 3-s long vibration were applied randomly to both soleus muscles while subjects kept their eyes closed. Postural responses to these stimuli were measured by static posturography. The effect of dopaminergic medication was established by testing patients in both ON and OFF treatment phases. There was no intergroup difference in the pattern or latencies of responses. However, the amplitudes were significantly larger in advanced PD patients; controls did not differ from early-stage PD patients. Dopaminergic medication had no significant effect on any of the measures. The scaling of postural reactions triggered by lower leg proprioception is disturbed in advanced PD. Neither afferent proprioceptive deficits nor inaccurate timing is involved. This study gives further evidence for the generalized impairment of the scaling of postural responses evoked whenever there is a sudden change of sensory conditions, as occurs with the progression of PD. Such impairment could play a significant role in the pathophysiology of postural instability and falls in PD patients. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Parkinson’s disease; Posture; Vibration; Proprioception; Dopaminergic medication
Postural instability is a major clinical problem of patients with Parkinson’s disease (PD) [6]. It is known that multiple factors are involved, but details about the pathophysiological mechanism underlying this phenomenon are still lacking [12]. While a proprioceptive disturbance is a likely candidate, the evidence is mostly indirect and based of studies on deficits in limb kinesthesia [19,20,39]. We recently tried to directly assess the proprioceptive influence on posture using vibration applied to the dorsal neck muscles [37]. This stimulus mainly affects muscle spindle Ia afferents [33] and can induce postural reactions and postural kinesthetic illusions in standing subjects [21]. We found that patients with clinically significant postural instability demonstrated larger vibration-evoked postural responses than early-stage PD and control subjects. This study indicated that impaired scaling of postural reactions triggered by (neck) pro-
∗
Corresponding author. Tel.: +49 89 7095 3673; fax: +49 89 7095 3677. E-mail address: [email protected] (K. B¨otzel).
0304-3940/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2006.02.073
prioceptive stimulus could contribute to postural instability in advanced PD [37]. Several questions, however, about the contribution of such impaired processing of proprioceptive information to balance problems in PD subjects still remain open. For example, neck proprioception is thought to be interpreted in the context of vestibular signals of head movement, indicating its assistance role in postural control [22,23,25,32]. Postural responses to direct vestibular (galvanic) stimulation were found to also be enhanced in severely affected Parkinson subjects [29]. Thus, larger responses to neck stimulation could be an epiphenomenon related to the larger responses to vestibular stimulation. To elucidate this issue, we stimulated lower leg proprioception, which is interpreted independently of vestibular information [22,25], by means of vibration applied to both soleus muscles (Sol). We hypothesized that postural reactions to Sol vibration would be enhanced only in severely affected PD patients as it was in the case of neck vibration. This would indicate that impairment of the scaling of proprioception-triggered
P. Valkoviˇc et al. / Neuroscience Letters 401 (2006) 92–96
postural reactions is generalized and related to the progression of PD. Twenty-two patients with idiopathic PD diagnosed according to the standard criteria [16] and 11 age-matched healthy subjects (control group, CONT; 8 men; mean age ± S.D. = 66.1 ± 4.8) participated. The informed consent of the subjects was obtained in agreement with the Declaration of Helsinki, and the local institutional review board approved the study. Patients were subdivided into two groups on the basis of the modified Hoehn and Yahr score (H&Y) [17]. Numerical data in this section are given as arithmetic mean ± standard deviation. Group PDA consisted of 11 moderately affected patients with H&Y scores ≤ 2.5 (early-stage PD; 9 men; age = 65.1 ± 4.9; duration of disease since diagnosis = 3.1 ± 1.5; H&Y score ON/OFF treatment phase = 1.7 ± 0.5/2.2 ± 0.5; motor component of the Unified Parkinson’s Disease Rating Scale (UPDRS) [11] in ON/OFF = 16.7 ± 7.1/28.8 ± 9.1). Group PD-B was composed of 11 severely affected patients in an advanced-stage of PD with H&Y score ≥ 3 (8 men; age = 66.6 ± 7.7; duration of disease = 11.1 ± 5.3; H&Y score ON/OFF = 3.1 ± 0.3/3.3 ± 0.5; UPDRS score ON/OFF = 26.2 ± 8.2/40.2 ± 9.7). Individuals with significant cognitive, musculoskeletal, and otoneurologic problems were not included. The possibility of polyneuropathy was excluded by clinical neurological assessments, including deep tendon reflexes and sensory testing. All patients were examined during their ON phase, about 45–60 min after they had taken their usual morning dose of antiparkinson medication, which had been recalculated on the basis of levodopa equivalents [31] and administered as a soluble levodopa preparation; all PD patients also underwent testing after withdrawal of medication (OFF) for at least 12 h; cabergoline washout lasted at least 36 h. Clinical measures – the modified H&Y stages and motor component of the UPDRS – were assessed (both ON and OFF) immediately before the balance tests. All patients were free of dyskinesia and significant body tremor that might have affected posturographic measures when tested. Center of foot pressure (COP) displacements were measured in the anterior-posterior (AP) direction on a posturographic platform (Kistler, type 9281B). Subjects stood with their heels touching each other and the feet splayed out at an angle of 30◦ . The subjects were instructed to remain upright and to refrain from any voluntary movements during the recording. Signals from the platform were amplified and digitized at a sampling rate of 40 Hz for 90 s. AP displacements of the COP were calculated from the raw data in mm. Muscle vibration was applied bilaterally with two identical vibrators, constructed as cylinders (8.8 cm × 3.2 cm; 180 g) and each containing a small DC motor (Buehler, type 1.13.055.221, Germany). The motors had the eccentric weights at both ends of the axis. A moderate vibration was applied (50 Hz, 1 mm). The vibrators were attached to the lower-third of the leg on the Sol muscles by elastic bands. Electromechanical delay between the computer command and the physical vibration stimulus (socalled trigger delay) was assessed by placing the accelerometer on the vibrator according to the method of Popov et al. [30]. The very first mechanical event occurred 23 ms after trigger impulse; the first peak (25% of the first rotation) was reached at 31 ms
93
when it had 20% of the steady-state peak-to-peak amplitude. This moment was arbitrarily taken as the actual onset of vibration. The first revolution of the motors was completed 44 ms after the trigger impulse. All values referring to latencies are given after subtraction of the trigger delay in the following. To allow for a clear identification of latencies and amplitudes of postural responses, the protocol was designed to gather data for averaging. Therefore, nine pulses of vibration each lasting 3 s were applied randomly to both Sol muscles while the subjects kept their eyes closed. The vibrators were switched on manually after the experimenter was sure that the subjects had recovered the baseline position of the COP after a preceding vibration (duty cycle was 5–15 s). This was checked on-line on the screen of the recording computer. The vibration was terminated automatically. Clinical variables (UPDRS scores) were compared with the Mann–Whitney U-test and Wilcoxon test for non-parametric data (p < 0.05). Postural responses to vibration pulses (path of the COP in AP direction) of each subject (nine responses) were averaged in the epoch from 1000 ms before to 3000 ms after the trigger impulse. From these curves, the mean latencies and amplitudes of peaks and troughs (see schematic drawing, Fig. 1A) were compared by ANOVA and Tukey’s HSD post hoc procedure (p < 0.05). ON-OFF comparison of amplitudes was made by paired t-tests. Correlation between amplitudes (A1 and A2; see below) was done using a two-tailed Pearson’s test. The functional involvement differed significantly in both patient groups. This was documented by the values of H&Y (selection criterion) and UPDRS scores; PD-A < PDB (Mann–Whitney U-test: UPDRSON , p < 0.005; UPDRSOFF p = 0.011; individual values are not shown). There was also a significant difference between ON and OFF scores in both PD groups; ON < OFF (Wilcoxon test: p < 0.003 both groups). After the onset of the Sol vibration, all subjects demonstrated a similar pattern of postural responses (Fig. 1). The COP was first shifted transiently forward and then was displaced backward for a longer period. The earlier, forward response started about 87 ms (latency L1) and peaked 250–300 ms (L2) after the beginning of vibration; the later response began after 500 ms (L3) and peaked after 2000 ms (L4; Table 1). There was no significant group difference for any of latencies (L1 (F = 0.002) p = 1; L2 (F = 0.927) p = 0.456; L3 (F = 1.430) p = 0.238; L4 (F = 0.548) p = 0.701). In contrast to the similar pattern and latencies in the three groups, the size of the responses, i.e., the amplitudes, were clearly different for the severely affected patients (Fig. 1B and C, Fig. 2). There was a significant difference between the groups Table 1 Latencies of postural responses L1 CONT PD-A-ON PD-A-OFF PD-B-ON PD-B-OFF
86.8 87.5 87.2 86.8 86.8
L2 ± ± ± ± ±
22.4 20.6 28.2 22.5 22.5
271.1 257.2 301.7 284.8 284.8
L3 ± ± ± ± ±
73.4 46.6 60.3 51.8 51.8
511.9 471.0 593.9 523.5 548.0
L4 ± ± ± ± ±
165.6 153.8 94.6 88.3 107.6
2102.1 2041.6 1900.2 2030.0 2199.4
± ± ± ± ±
542.8 456.4 439.9 456.3 544.1
Values are given in milliseconds as mean ± standard deviation. Note that there was no significant difference in any of latencies between the groups.
94
P. Valkoviˇc et al. / Neuroscience Letters 401 (2006) 92–96
Fig. 1. (A) Schematic drawing of the COP movement in AP direction around the onset of vibration with regards to the points at which latencies and amplitudes were measured. (B) Original traces of center of foot pressure of one representative subject from each group. (C) Averaged COP paths in AP from each group. Traces of the patients in their OFF medication phase are not shown. COP: centre of foot pressure, AP: anterior-posterior.
for amplitude A1 ((F = 5.646) p < 0.001) as well as for amplitude A2 ((F = 9.685) p < 0.001). Post hoc comparisons revealed that PD-B patients demonstrated significantly larger responses (both A1 and A2) than the other two groups, either ON or OFF (amplitude A1: PD-B-ON versus both PD-A-ON and CONT, p < 0.004; PD-B-OFF versus PD-A-OFF, p = 0.032; PD-B-OFF versus CONT, p < 0.007; amplitude A2: PD-B-ON versus both PD-A-ON and CONT, p < 0.002; PD-B-OFF versus both PD-AOFF and CONT, p < 0.001). There was no significant difference in the amplitudes between ON and OFF medication phases neither in PD-A nor in PD-B group (A1PD-A , p = 0.313; A1PD-B , p = 0.675;A2PD-A , p = 0.073; A2PD-B , p = 0.592). Finally, in order to explore the relationship between A1 and A2 we correlated both amplitudes, separately for all groups. The results showed that both amplitudes mutually significantly correlated for the CONT group (r = 0.73, p = 0.01). But, in both patient groups, regardless ON or OFF medication, these corre-
Fig. 2. Amplitudes of postural responses. Asterisks denote significant group differences between severely affected patients and both other groups in ONmedication treatment phase; ** p < 0.01, and squares indicate significant differences in OFF-medication treatment phase; p < 0.05; p < 0.01; p < 0.001. Note that the group comparison does not reveal any significant difference in the amplitudes of postural responses between ON- and OFF-medication trials in the two PD groups.
lations were not significant (PD-A-ON group: r = 0.42, p = 0.2; PD-A-OFF: r = −0.08, p = 0.81; PD-B-ON: r = 0.49, p = 0.12; PD-B-OFF: r = 0.29, p = 0.37). This study shows that severely affected PD patients swayed significantly more during artificial modification of the proprioceptive signal from the lower legs than did mildly affected patients and healthy controls. The patterns and latencies of these reactions did not, however, differ between the three groups. Dopaminergic medication did not have any significant effect on any of the measures. By averaging multiple COP traces, we identified a biphasic configuration of the COP displacement, which was obvious in all tested subjects. Although, one might expect that both components represent separate reactions with distinct functions, a correlation analysis revealed a high interdependence at least in the healthy state and thus we assume a common origin and function. The COP started to move ∼87 ms after the real onset of vibration. This is caused by a plantarflection of the feet which is induced by activation of the corresponding muscles. Our latencies are similar to those in tests with toe-up rotation of the supporting platform, which causes stretching and following medium-latency (ML) activation of the triceps surae muscles [2,9,27]. The following second phase (backward COP shift) begins about 500 ms after vibration onset. This reflects the alignment of the body caused by processing the false proprioceptive signal from the soleus muscles, which seem to be stretched, a condition naturally present during forward leaning. Thus, to maintain their balance, subjects counteract this illusory forward leaning by actually leaning backward [13,18,21]. This complex reaction is a result of a continuous updating of sensory information from several sources, including the cerebral cortex. Moreover, our onset latencies for this later response coincide with published data on the latency of vibration-induced illusory motion perception [24] and the latencies of the later portions of the fronto-central vibration-evoked potentials (400–800 ms), which are specific for kinesthetic illusions [26]. The finding that postural responses are enhanced only in severely affected Parkinson patients confirms our preliminary
P. Valkoviˇc et al. / Neuroscience Letters 401 (2006) 92–96
hypothesis that the scaling of postural reactions evoked by lower leg proprioceptive stimulation deteriorates as the disease progresses. In a previous study with neck vibration, we drew a similar conclusion [37]. Therefore, the present results provide more evidence that the generalized impairment of the scaling of proprioception-triggered postural reactions is related to the progression of PD. Moreover, the correlations of the amplitudes of both components of the postural synergy (A1 and A2) in both PD patients groups (regardless of dopaminergic therapy) were not significant. We interpret this finding as a de-coupling or dyscoordination of a postural synergy in PD. A couple of mechanisms seem to be involved in the enhancement of these reactions. The first response (forward) possibly reflects ML activation of the triceps surae muscles; its exaggeration could correspond to the advanced PD subject’s tendency to respond to external perturbation with abnormally large automatic postural responses, particularly the ML stretch responses. This was shown for lower leg muscles [2,15,35], as well as for hip, trunk, and arm muscles, suggesting a global impairment of ML gain control [8]. The attenuation and/or inflexibility of correcting long-latency automatic postural responses [1,3,4,34] probably also influences this postural impairment. The enlargement of the later (backward) response could simply reflect an enlargement of the fast response in the increasingly stiff and severely affected Parkinson subjects [8,10,14]. However, there is some evidence that these patients also react hyperactively to misleading stimuli from other sensory modalities, which depend on higher-level sensory re-weighting. This holds true for visually induced postural tilts [7], as well as for postural responses evoked by galvanic stimulation of the vestibular system [29]. In this context, our findings of enhanced responses in severely affected PD patients suggest that the progression of the disease affects the ability of the postural system to generate accurately calibrated motor commands, regardless of which sensory modality is manipulated. Parkinson patients with clinical postural instability (H&Y stage ≥ 3) seem to respond to a sudden change of sensory conditions with hyperactive postural reactions. They are not reduced as might be expected on the basis of earlier experiments with another type of proprioception, i.e., kinesthesia [19,20,39]. In contrast to the different amplitudes, the similarity of pattern and latencies in the three groups generally excludes any role of afferent proprioceptive deficits in postural dyscontrol in persons with PD. The central integrative organization and timing of postural responses are not involved in PD pathology. This is congruent with previous work on sensory manipulation [7,29,36,38] and movable platform experiments [2,5,8,14,15,34]. Dopaminergic therapy has been reported to cause an acute decrease in proprioception in the upper extremities of PD patients, and it was speculated that drug-induced dyskinesias could be a compensation for this impairment [28]. Thus, one can logically hypothesize that our pivotal finding might also account for subtle dyskinesias. Although medication substantially improved the clinical disability in our PD population, we failed to detect any significant difference in the calculated postural measures between “ON” and “OFF” phases. Therefore, subtle dyskinesias do not explain the enlargement of postural
95
responses in advanced PD. The failure of medication supports clinical experience and corroborates previous experimental studies that suggested that other non-dopaminergic structures are involved in the pathophysiology of balance impairment in PD [5,8,14]. In conclusion, our study showed that only those PD patients who exhibit clinical postural instability have problems scaling their postural reactions to a sudden change of proprioceptive signals from the lower legs. This corroborates our previous finding made with proprioceptive manipulation of the neck and indicates that impairment of the scaling of proprioception-triggered postural reactions is generalized and related to the progression of PD. We did not detect any impairment in the timing or in the sensory integration mechanisms. Likewise, we did not find any effects of dopaminergic medication, which suggests that nondopaminergic structures are involved in the genesis of balance deterioration in PD. In the light of previous studies using sensory manipulation of other modalities (e.g., vision, vestibular system), we hypothesize that whichever sensory input is manipulated, standing patients in the advanced-stage of PD (H&Y ≥ 3) tend to respond hyperactively (despite correct timing), approach their limits of stability, and therefore are more prone to lose their balance and fall. This mechanism could significantly explain postural instability and falls in PD. Acknowledgments We wish to thank Judy Benson for copyediting the manuscript. Peter Valkoviˇc was supported by an ENS Fellowship for 2005. References [1] D.J. Beckley, B.R. Bloem, M.P. Remler, Impaired scaling of long latency postural reflexes in patients with Parkinson’s disease, Electroencephalogr. Clin. Neurophysiol. 89 (1993) 22–28. [2] D.J. Beckley, B.R. Bloem, J.G. van Dijk, R.A. Roos, M.P. Remler, Electrophysiological correlates of postural instability in Parkinson’s disease, Electroencephalogr. Clin. Neurophysiol. 81 (1991) 263–268. [3] B.R. Bloem, D.J. Beckley, M.P. Remler, R.A. Roos, J.G. van Dijk, Postural reflexes in Parkinson’s disease during ‘resist’ and ‘yield’ tasks, J. Neurol. Sci. 129 (1995) 109–119. [4] B.R. Bloem, D.J. Beckley, J.G. van Dijk, A.H. Zwinderman, M.P. Remler, R.A. Roos, Are medium and long latency reflexes a screening tool for early Parkinson’s disease? J. Neurol. Sci. 113 (1992) 38–42. [5] B.R. Bloem, D.J. Beckley, J.G. van Dijk, A.H. Zwinderman, M.P. Remler, R.A. Roos, Influence of dopaminergic medication on automatic postural responses and balance impairment in Parkinson’s disease, Mov. Disord. 11 (1996) 509–521. [6] B.R. Bloem, J.P. van Vugt, D.J. Beckley, Postural instability and falls in Parkinson’s disease, Adv. Neurol. 87 (2001) 209–223. [7] A.M. Bronstein, J.D. Hood, M.A. Gresty, C. Panagi, Visual control of balance in cerebellar and parkinsonian syndromes, Brain 113 (Pt 3) (1990) 767–779. [8] M.G. Carpenter, J.H. Allum, F. Honegger, A.L. Adkin, B.R. Bloem, Postural abnormalities to multidirectional stance perturbations in Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 75 (2004) 1245–1254. [9] H.C. Diener, J. Dichgans, F. Bootz, M. Bacher, Early stabilization of human posture after a sudden disturbance: influence of rate and amplitude of displacement, Exp. Brain Res. 56 (1984) 126–134.
96
P. Valkoviˇc et al. / Neuroscience Letters 401 (2006) 92–96
[10] D. Dimitrova, F.B. Horak, J.G. Nutt, Postural muscle responses to multidirectional translations in patients with Parkinson’s disease, J. Neurophysiol. 91 (2004) 489–501. [11] S. Fahn, R.L. Elton, and the UPDRS Development Committee, Unified Parkinson’s disease rating scale. In: S. Fahn, C.D. Marsden, M. Goldstein, D.B. Calne (Eds.), Recent Developments in Parkinson’s Disease, vol. 2, Macmillan Healthcare Information, Florham Park, NJ, 1987, pp. 153–163. [12] Y.A. Grimbergen, M. Munneke, B.R. Bloem, Falls in Parkinson’s disease, Curr. Opin. Neurol. 17 (2004) 405–415. [13] F. Hlavaˇcka, M. Kriˇzkov´a, F.B. Horak, Modification of human postural response to leg muscle vibration by electrical vestibular stimulation, Neurosci. Lett. 189 (1995) 9–12. [14] F.B. Horak, J. Frank, J. Nutt, Effects of dopamine on postural control in parkinsonian subjects: scaling, set, and tone, J. Neurophysiol. 75 (1996) 2380–2396. [15] F.B. Horak, J.G. Nutt, L.M. Nashner, Postural inflexibility in parkinsonian subjects, J. Neurol. Sci. 111 (1992) 46–58. [16] A.J. Hughes, S.E. Daniel, S. Blankson, A.J. Lees, A clinicopathologic study of 100 cases of Parkinson’s disease, Arch. Neurol. 50 (1993) 140–148. [17] J. Jankovic, M. McDermott, J. Carter, S. Gauthier, C. Goetz, L. Golbe, S. Huber, W. Koller, C. Olanow, I. Shoulson, The Parkinson Study Group, Variable expression of Parkinson’s disease: a base-line analysis of the DATATOP cohort, Neurology 40 (1990) 1529–1534. [18] A. Kavounoudias, J.C. Gilhodes, R. Roll, J.P. Roll, From balance regulation to body orientation: two goals for muscle proprioceptive information processing? Exp. Brain Res. 124 (1999) 80–88. [19] E. Khudados, F.W. Cody, D.J. O’Boyle, Proprioceptive regulation of voluntary ankle movements, demonstrated using muscle vibration, is impaired by Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 67 (1999) 504–510. [20] T. Klockgether, M. Borutta, H. Rapp, S. Spieker, J. Dichgans, A defect of kinesthesia in Parkinson’s disease, Mov. Disord. 10 (1995) 460–465. [21] J.R. Lackner, M.S. Levine, Changes in apparent body orientation and sensory localization induced by vibration of postural muscles: vibratory myesthetic illusions, Aviat. Space Environ. Med. 50 (1979) 346–354. [22] H. Lekhel, K. Popov, A. Bronstein, M. Gresty, Postural responses to vibration of neck muscles in patients with uni- and bilateral vestibular loss, Gait Posture 7 (1998) 228–236. [23] S. Lund, Postural effects of neck muscle vibration in man, Experientia 36 (1980) 1398. [24] P.B.C. Matthews, Where does Sherrington’s ‘muscular sense’ originate? Muscles, joints, corrolary discharges, Annu. Rev. Neurosci. 5 (1962) 189–218.
[25] T. Mergner, T. Rosemeier, Interaction of vestibular, somatosensory and visual signals for postural control and motion perception under terrestrial and microgravity conditions—a conceptual model, Brain Res. Rev. 28 (1998) 118–135. [26] T.F. M¨unte, E.M. J¨obges, B.M. Wieringa, S. Klein, M. Schubert, S. Johannes, R. Dengler, Human evoked potentials to long duration vibratory stimuli: role of muscle afferents, Neurosci. Lett. 216 (1996) 163–166. [27] L.M. Nashner, Fixed patterns of rapid postural responses among leg muscles during stance, Exp. Brain Res. 30 (1977) 13–24. [28] P. O’Suilleabhain, J. Bullard, R.B. Dewey, Proprioception in Parkinson’s disease is acutely depressed by dopaminergic medications, J. Neurol. Neurosurg. Psychiatry 71 (2001) 607–610. [29] M.A. Pastor, B.L. Day, C.D. Marsden, Vestibular induced postural responses in Parkinson’s disease, Brain 116 (1993) 1177–1190. [30] K.E. Popov, H. Lekhel, M. Faldon, A.M. Bronstein, M.A. Gresty, Visual and oculomotor responses induced by neck vibration in normal subjects and labyrinthine-defective patients, Exp. Brain Res. 128 (1999) 343–352. [31] J. Reimer, M. Grabowski, O. Lindvall, P. Hagell, Use and interpretation of on/off diaries in Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 75 (2004) 396–400. [32] J.P. Roll, J.P. Vedel, E. Ribot, Alteration of proprioceptive messages induced by tendon vibration in man: a microneurographic study, Exp. Brain Res. 76 (1989) 213–222. [33] J.P. Roll, J.P. Vedel, R. Roll, Eye, head and skeletal muscle spindle feedback in the elaboration of body references, Prog. Brain Res. 80 (1989) 113–123. [34] M. Schieppati, A. Nardone, Free and supported stance in Parkinson’s disease. The effect of posture and ‘postural set’ on leg muscle responses to perturbation, and its relation to the severity of the disease, Brain 114 (1991) 1227–1244. [35] E. Scholz, H.C. Diener, J. Noth, H. Friedemann, J. Dichgans, M. Bacher, Medium and long latency EMG responses in leg muscles: Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 50 (1987) 66–70. [36] A.L. Smiley-Oyen, H.Y. Cheng, L.D. Latt, M.S. Redfern, Adaptation of vibration-induced postural sway in individuals with Parkinson’s disease, Gait Posture 16 (2002) 188–197. ˇ [37] P. Valkoviˇc, S. Krafczyk, M. Saling, J. Benetin, K. B¨otzel, Postural reactions to neck vibration in Parkinson’s disease, Mov. Disord. 21 (2006) 59–65. [38] J.A. Waterston, M.B. Hawken, S. Tanyeri, P. Jantti, C. Kennard, Influence of sensory manipulation on postural control in Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 56 (1993) 1276–1281. [39] S. Zia, F. Cody, D. O’Boyle, Joint position sense is impaired by Parkinson’s disease, Ann. Neurol. 47 (2000) 218–228.
Postural reactions to soleus muscle vibration in Parkinson’s disease: Scaling deteriorates as disease progresses Peter Valkoviˇc a,b , Siegbert Krafczyk a , Kai B¨otzel a,∗ a
b
Department of Neurology, Ludwig-Maximilians University, Marchioninistraße 15, 81366 Munich, Germany 2nd Department of Neurology, School of Medicine, Comenius University, Limbov´a 5, 85104 Bratislava, Slovakia Received 30 December 2005; received in revised form 13 February 2006; accepted 24 February 2006
Abstract Previous research has shown that Parkinson’s disease (PD) patients, especially those with postural instability, respond hyperactively to visual, vestibular, and neck proprioceptive sensory manipulation. To determine if this impairment of the sensory information scaling holds true for the lower leg proprioceptive system, we studied postural responses to mechanical vibration (which affects the muscle spindle Ia afferents) applied to the soleus muscles of PD subjects and healthy controls. Early-stage and advanced-stage PD patients as well as age-matched control subjects participated. Each group comprised 11 subjects. Nine pulses of 3-s long vibration were applied randomly to both soleus muscles while subjects kept their eyes closed. Postural responses to these stimuli were measured by static posturography. The effect of dopaminergic medication was established by testing patients in both ON and OFF treatment phases. There was no intergroup difference in the pattern or latencies of responses. However, the amplitudes were significantly larger in advanced PD patients; controls did not differ from early-stage PD patients. Dopaminergic medication had no significant effect on any of the measures. The scaling of postural reactions triggered by lower leg proprioception is disturbed in advanced PD. Neither afferent proprioceptive deficits nor inaccurate timing is involved. This study gives further evidence for the generalized impairment of the scaling of postural responses evoked whenever there is a sudden change of sensory conditions, as occurs with the progression of PD. Such impairment could play a significant role in the pathophysiology of postural instability and falls in PD patients. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Parkinson’s disease; Posture; Vibration; Proprioception; Dopaminergic medication
Postural instability is a major clinical problem of patients with Parkinson’s disease (PD) [6]. It is known that multiple factors are involved, but details about the pathophysiological mechanism underlying this phenomenon are still lacking [12]. While a proprioceptive disturbance is a likely candidate, the evidence is mostly indirect and based of studies on deficits in limb kinesthesia [19,20,39]. We recently tried to directly assess the proprioceptive influence on posture using vibration applied to the dorsal neck muscles [37]. This stimulus mainly affects muscle spindle Ia afferents [33] and can induce postural reactions and postural kinesthetic illusions in standing subjects [21]. We found that patients with clinically significant postural instability demonstrated larger vibration-evoked postural responses than early-stage PD and control subjects. This study indicated that impaired scaling of postural reactions triggered by (neck) pro-
∗
Corresponding author. Tel.: +49 89 7095 3673; fax: +49 89 7095 3677. E-mail address: [email protected] (K. B¨otzel).
0304-3940/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2006.02.073
prioceptive stimulus could contribute to postural instability in advanced PD [37]. Several questions, however, about the contribution of such impaired processing of proprioceptive information to balance problems in PD subjects still remain open. For example, neck proprioception is thought to be interpreted in the context of vestibular signals of head movement, indicating its assistance role in postural control [22,23,25,32]. Postural responses to direct vestibular (galvanic) stimulation were found to also be enhanced in severely affected Parkinson subjects [29]. Thus, larger responses to neck stimulation could be an epiphenomenon related to the larger responses to vestibular stimulation. To elucidate this issue, we stimulated lower leg proprioception, which is interpreted independently of vestibular information [22,25], by means of vibration applied to both soleus muscles (Sol). We hypothesized that postural reactions to Sol vibration would be enhanced only in severely affected PD patients as it was in the case of neck vibration. This would indicate that impairment of the scaling of proprioception-triggered
P. Valkoviˇc et al. / Neuroscience Letters 401 (2006) 92–96
postural reactions is generalized and related to the progression of PD. Twenty-two patients with idiopathic PD diagnosed according to the standard criteria [16] and 11 age-matched healthy subjects (control group, CONT; 8 men; mean age ± S.D. = 66.1 ± 4.8) participated. The informed consent of the subjects was obtained in agreement with the Declaration of Helsinki, and the local institutional review board approved the study. Patients were subdivided into two groups on the basis of the modified Hoehn and Yahr score (H&Y) [17]. Numerical data in this section are given as arithmetic mean ± standard deviation. Group PDA consisted of 11 moderately affected patients with H&Y scores ≤ 2.5 (early-stage PD; 9 men; age = 65.1 ± 4.9; duration of disease since diagnosis = 3.1 ± 1.5; H&Y score ON/OFF treatment phase = 1.7 ± 0.5/2.2 ± 0.5; motor component of the Unified Parkinson’s Disease Rating Scale (UPDRS) [11] in ON/OFF = 16.7 ± 7.1/28.8 ± 9.1). Group PD-B was composed of 11 severely affected patients in an advanced-stage of PD with H&Y score ≥ 3 (8 men; age = 66.6 ± 7.7; duration of disease = 11.1 ± 5.3; H&Y score ON/OFF = 3.1 ± 0.3/3.3 ± 0.5; UPDRS score ON/OFF = 26.2 ± 8.2/40.2 ± 9.7). Individuals with significant cognitive, musculoskeletal, and otoneurologic problems were not included. The possibility of polyneuropathy was excluded by clinical neurological assessments, including deep tendon reflexes and sensory testing. All patients were examined during their ON phase, about 45–60 min after they had taken their usual morning dose of antiparkinson medication, which had been recalculated on the basis of levodopa equivalents [31] and administered as a soluble levodopa preparation; all PD patients also underwent testing after withdrawal of medication (OFF) for at least 12 h; cabergoline washout lasted at least 36 h. Clinical measures – the modified H&Y stages and motor component of the UPDRS – were assessed (both ON and OFF) immediately before the balance tests. All patients were free of dyskinesia and significant body tremor that might have affected posturographic measures when tested. Center of foot pressure (COP) displacements were measured in the anterior-posterior (AP) direction on a posturographic platform (Kistler, type 9281B). Subjects stood with their heels touching each other and the feet splayed out at an angle of 30◦ . The subjects were instructed to remain upright and to refrain from any voluntary movements during the recording. Signals from the platform were amplified and digitized at a sampling rate of 40 Hz for 90 s. AP displacements of the COP were calculated from the raw data in mm. Muscle vibration was applied bilaterally with two identical vibrators, constructed as cylinders (8.8 cm × 3.2 cm; 180 g) and each containing a small DC motor (Buehler, type 1.13.055.221, Germany). The motors had the eccentric weights at both ends of the axis. A moderate vibration was applied (50 Hz, 1 mm). The vibrators were attached to the lower-third of the leg on the Sol muscles by elastic bands. Electromechanical delay between the computer command and the physical vibration stimulus (socalled trigger delay) was assessed by placing the accelerometer on the vibrator according to the method of Popov et al. [30]. The very first mechanical event occurred 23 ms after trigger impulse; the first peak (25% of the first rotation) was reached at 31 ms
93
when it had 20% of the steady-state peak-to-peak amplitude. This moment was arbitrarily taken as the actual onset of vibration. The first revolution of the motors was completed 44 ms after the trigger impulse. All values referring to latencies are given after subtraction of the trigger delay in the following. To allow for a clear identification of latencies and amplitudes of postural responses, the protocol was designed to gather data for averaging. Therefore, nine pulses of vibration each lasting 3 s were applied randomly to both Sol muscles while the subjects kept their eyes closed. The vibrators were switched on manually after the experimenter was sure that the subjects had recovered the baseline position of the COP after a preceding vibration (duty cycle was 5–15 s). This was checked on-line on the screen of the recording computer. The vibration was terminated automatically. Clinical variables (UPDRS scores) were compared with the Mann–Whitney U-test and Wilcoxon test for non-parametric data (p < 0.05). Postural responses to vibration pulses (path of the COP in AP direction) of each subject (nine responses) were averaged in the epoch from 1000 ms before to 3000 ms after the trigger impulse. From these curves, the mean latencies and amplitudes of peaks and troughs (see schematic drawing, Fig. 1A) were compared by ANOVA and Tukey’s HSD post hoc procedure (p < 0.05). ON-OFF comparison of amplitudes was made by paired t-tests. Correlation between amplitudes (A1 and A2; see below) was done using a two-tailed Pearson’s test. The functional involvement differed significantly in both patient groups. This was documented by the values of H&Y (selection criterion) and UPDRS scores; PD-A < PDB (Mann–Whitney U-test: UPDRSON , p < 0.005; UPDRSOFF p = 0.011; individual values are not shown). There was also a significant difference between ON and OFF scores in both PD groups; ON < OFF (Wilcoxon test: p < 0.003 both groups). After the onset of the Sol vibration, all subjects demonstrated a similar pattern of postural responses (Fig. 1). The COP was first shifted transiently forward and then was displaced backward for a longer period. The earlier, forward response started about 87 ms (latency L1) and peaked 250–300 ms (L2) after the beginning of vibration; the later response began after 500 ms (L3) and peaked after 2000 ms (L4; Table 1). There was no significant group difference for any of latencies (L1 (F = 0.002) p = 1; L2 (F = 0.927) p = 0.456; L3 (F = 1.430) p = 0.238; L4 (F = 0.548) p = 0.701). In contrast to the similar pattern and latencies in the three groups, the size of the responses, i.e., the amplitudes, were clearly different for the severely affected patients (Fig. 1B and C, Fig. 2). There was a significant difference between the groups Table 1 Latencies of postural responses L1 CONT PD-A-ON PD-A-OFF PD-B-ON PD-B-OFF
86.8 87.5 87.2 86.8 86.8
L2 ± ± ± ± ±
22.4 20.6 28.2 22.5 22.5
271.1 257.2 301.7 284.8 284.8
L3 ± ± ± ± ±
73.4 46.6 60.3 51.8 51.8
511.9 471.0 593.9 523.5 548.0
L4 ± ± ± ± ±
165.6 153.8 94.6 88.3 107.6
2102.1 2041.6 1900.2 2030.0 2199.4
± ± ± ± ±
542.8 456.4 439.9 456.3 544.1
Values are given in milliseconds as mean ± standard deviation. Note that there was no significant difference in any of latencies between the groups.
94
P. Valkoviˇc et al. / Neuroscience Letters 401 (2006) 92–96
Fig. 1. (A) Schematic drawing of the COP movement in AP direction around the onset of vibration with regards to the points at which latencies and amplitudes were measured. (B) Original traces of center of foot pressure of one representative subject from each group. (C) Averaged COP paths in AP from each group. Traces of the patients in their OFF medication phase are not shown. COP: centre of foot pressure, AP: anterior-posterior.
for amplitude A1 ((F = 5.646) p < 0.001) as well as for amplitude A2 ((F = 9.685) p < 0.001). Post hoc comparisons revealed that PD-B patients demonstrated significantly larger responses (both A1 and A2) than the other two groups, either ON or OFF (amplitude A1: PD-B-ON versus both PD-A-ON and CONT, p < 0.004; PD-B-OFF versus PD-A-OFF, p = 0.032; PD-B-OFF versus CONT, p < 0.007; amplitude A2: PD-B-ON versus both PD-A-ON and CONT, p < 0.002; PD-B-OFF versus both PD-AOFF and CONT, p < 0.001). There was no significant difference in the amplitudes between ON and OFF medication phases neither in PD-A nor in PD-B group (A1PD-A , p = 0.313; A1PD-B , p = 0.675;A2PD-A , p = 0.073; A2PD-B , p = 0.592). Finally, in order to explore the relationship between A1 and A2 we correlated both amplitudes, separately for all groups. The results showed that both amplitudes mutually significantly correlated for the CONT group (r = 0.73, p = 0.01). But, in both patient groups, regardless ON or OFF medication, these corre-
Fig. 2. Amplitudes of postural responses. Asterisks denote significant group differences between severely affected patients and both other groups in ONmedication treatment phase; ** p < 0.01, and squares indicate significant differences in OFF-medication treatment phase; p < 0.05; p < 0.01; p < 0.001. Note that the group comparison does not reveal any significant difference in the amplitudes of postural responses between ON- and OFF-medication trials in the two PD groups.
lations were not significant (PD-A-ON group: r = 0.42, p = 0.2; PD-A-OFF: r = −0.08, p = 0.81; PD-B-ON: r = 0.49, p = 0.12; PD-B-OFF: r = 0.29, p = 0.37). This study shows that severely affected PD patients swayed significantly more during artificial modification of the proprioceptive signal from the lower legs than did mildly affected patients and healthy controls. The patterns and latencies of these reactions did not, however, differ between the three groups. Dopaminergic medication did not have any significant effect on any of the measures. By averaging multiple COP traces, we identified a biphasic configuration of the COP displacement, which was obvious in all tested subjects. Although, one might expect that both components represent separate reactions with distinct functions, a correlation analysis revealed a high interdependence at least in the healthy state and thus we assume a common origin and function. The COP started to move ∼87 ms after the real onset of vibration. This is caused by a plantarflection of the feet which is induced by activation of the corresponding muscles. Our latencies are similar to those in tests with toe-up rotation of the supporting platform, which causes stretching and following medium-latency (ML) activation of the triceps surae muscles [2,9,27]. The following second phase (backward COP shift) begins about 500 ms after vibration onset. This reflects the alignment of the body caused by processing the false proprioceptive signal from the soleus muscles, which seem to be stretched, a condition naturally present during forward leaning. Thus, to maintain their balance, subjects counteract this illusory forward leaning by actually leaning backward [13,18,21]. This complex reaction is a result of a continuous updating of sensory information from several sources, including the cerebral cortex. Moreover, our onset latencies for this later response coincide with published data on the latency of vibration-induced illusory motion perception [24] and the latencies of the later portions of the fronto-central vibration-evoked potentials (400–800 ms), which are specific for kinesthetic illusions [26]. The finding that postural responses are enhanced only in severely affected Parkinson patients confirms our preliminary
P. Valkoviˇc et al. / Neuroscience Letters 401 (2006) 92–96
hypothesis that the scaling of postural reactions evoked by lower leg proprioceptive stimulation deteriorates as the disease progresses. In a previous study with neck vibration, we drew a similar conclusion [37]. Therefore, the present results provide more evidence that the generalized impairment of the scaling of proprioception-triggered postural reactions is related to the progression of PD. Moreover, the correlations of the amplitudes of both components of the postural synergy (A1 and A2) in both PD patients groups (regardless of dopaminergic therapy) were not significant. We interpret this finding as a de-coupling or dyscoordination of a postural synergy in PD. A couple of mechanisms seem to be involved in the enhancement of these reactions. The first response (forward) possibly reflects ML activation of the triceps surae muscles; its exaggeration could correspond to the advanced PD subject’s tendency to respond to external perturbation with abnormally large automatic postural responses, particularly the ML stretch responses. This was shown for lower leg muscles [2,15,35], as well as for hip, trunk, and arm muscles, suggesting a global impairment of ML gain control [8]. The attenuation and/or inflexibility of correcting long-latency automatic postural responses [1,3,4,34] probably also influences this postural impairment. The enlargement of the later (backward) response could simply reflect an enlargement of the fast response in the increasingly stiff and severely affected Parkinson subjects [8,10,14]. However, there is some evidence that these patients also react hyperactively to misleading stimuli from other sensory modalities, which depend on higher-level sensory re-weighting. This holds true for visually induced postural tilts [7], as well as for postural responses evoked by galvanic stimulation of the vestibular system [29]. In this context, our findings of enhanced responses in severely affected PD patients suggest that the progression of the disease affects the ability of the postural system to generate accurately calibrated motor commands, regardless of which sensory modality is manipulated. Parkinson patients with clinical postural instability (H&Y stage ≥ 3) seem to respond to a sudden change of sensory conditions with hyperactive postural reactions. They are not reduced as might be expected on the basis of earlier experiments with another type of proprioception, i.e., kinesthesia [19,20,39]. In contrast to the different amplitudes, the similarity of pattern and latencies in the three groups generally excludes any role of afferent proprioceptive deficits in postural dyscontrol in persons with PD. The central integrative organization and timing of postural responses are not involved in PD pathology. This is congruent with previous work on sensory manipulation [7,29,36,38] and movable platform experiments [2,5,8,14,15,34]. Dopaminergic therapy has been reported to cause an acute decrease in proprioception in the upper extremities of PD patients, and it was speculated that drug-induced dyskinesias could be a compensation for this impairment [28]. Thus, one can logically hypothesize that our pivotal finding might also account for subtle dyskinesias. Although medication substantially improved the clinical disability in our PD population, we failed to detect any significant difference in the calculated postural measures between “ON” and “OFF” phases. Therefore, subtle dyskinesias do not explain the enlargement of postural
95
responses in advanced PD. The failure of medication supports clinical experience and corroborates previous experimental studies that suggested that other non-dopaminergic structures are involved in the pathophysiology of balance impairment in PD [5,8,14]. In conclusion, our study showed that only those PD patients who exhibit clinical postural instability have problems scaling their postural reactions to a sudden change of proprioceptive signals from the lower legs. This corroborates our previous finding made with proprioceptive manipulation of the neck and indicates that impairment of the scaling of proprioception-triggered postural reactions is generalized and related to the progression of PD. We did not detect any impairment in the timing or in the sensory integration mechanisms. Likewise, we did not find any effects of dopaminergic medication, which suggests that nondopaminergic structures are involved in the genesis of balance deterioration in PD. In the light of previous studies using sensory manipulation of other modalities (e.g., vision, vestibular system), we hypothesize that whichever sensory input is manipulated, standing patients in the advanced-stage of PD (H&Y ≥ 3) tend to respond hyperactively (despite correct timing), approach their limits of stability, and therefore are more prone to lose their balance and fall. This mechanism could significantly explain postural instability and falls in PD. Acknowledgments We wish to thank Judy Benson for copyediting the manuscript. Peter Valkoviˇc was supported by an ENS Fellowship for 2005. References [1] D.J. Beckley, B.R. Bloem, M.P. Remler, Impaired scaling of long latency postural reflexes in patients with Parkinson’s disease, Electroencephalogr. Clin. Neurophysiol. 89 (1993) 22–28. [2] D.J. Beckley, B.R. Bloem, J.G. van Dijk, R.A. Roos, M.P. Remler, Electrophysiological correlates of postural instability in Parkinson’s disease, Electroencephalogr. Clin. Neurophysiol. 81 (1991) 263–268. [3] B.R. Bloem, D.J. Beckley, M.P. Remler, R.A. Roos, J.G. van Dijk, Postural reflexes in Parkinson’s disease during ‘resist’ and ‘yield’ tasks, J. Neurol. Sci. 129 (1995) 109–119. [4] B.R. Bloem, D.J. Beckley, J.G. van Dijk, A.H. Zwinderman, M.P. Remler, R.A. Roos, Are medium and long latency reflexes a screening tool for early Parkinson’s disease? J. Neurol. Sci. 113 (1992) 38–42. [5] B.R. Bloem, D.J. Beckley, J.G. van Dijk, A.H. Zwinderman, M.P. Remler, R.A. Roos, Influence of dopaminergic medication on automatic postural responses and balance impairment in Parkinson’s disease, Mov. Disord. 11 (1996) 509–521. [6] B.R. Bloem, J.P. van Vugt, D.J. Beckley, Postural instability and falls in Parkinson’s disease, Adv. Neurol. 87 (2001) 209–223. [7] A.M. Bronstein, J.D. Hood, M.A. Gresty, C. Panagi, Visual control of balance in cerebellar and parkinsonian syndromes, Brain 113 (Pt 3) (1990) 767–779. [8] M.G. Carpenter, J.H. Allum, F. Honegger, A.L. Adkin, B.R. Bloem, Postural abnormalities to multidirectional stance perturbations in Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 75 (2004) 1245–1254. [9] H.C. Diener, J. Dichgans, F. Bootz, M. Bacher, Early stabilization of human posture after a sudden disturbance: influence of rate and amplitude of displacement, Exp. Brain Res. 56 (1984) 126–134.
96
P. Valkoviˇc et al. / Neuroscience Letters 401 (2006) 92–96
[10] D. Dimitrova, F.B. Horak, J.G. Nutt, Postural muscle responses to multidirectional translations in patients with Parkinson’s disease, J. Neurophysiol. 91 (2004) 489–501. [11] S. Fahn, R.L. Elton, and the UPDRS Development Committee, Unified Parkinson’s disease rating scale. In: S. Fahn, C.D. Marsden, M. Goldstein, D.B. Calne (Eds.), Recent Developments in Parkinson’s Disease, vol. 2, Macmillan Healthcare Information, Florham Park, NJ, 1987, pp. 153–163. [12] Y.A. Grimbergen, M. Munneke, B.R. Bloem, Falls in Parkinson’s disease, Curr. Opin. Neurol. 17 (2004) 405–415. [13] F. Hlavaˇcka, M. Kriˇzkov´a, F.B. Horak, Modification of human postural response to leg muscle vibration by electrical vestibular stimulation, Neurosci. Lett. 189 (1995) 9–12. [14] F.B. Horak, J. Frank, J. Nutt, Effects of dopamine on postural control in parkinsonian subjects: scaling, set, and tone, J. Neurophysiol. 75 (1996) 2380–2396. [15] F.B. Horak, J.G. Nutt, L.M. Nashner, Postural inflexibility in parkinsonian subjects, J. Neurol. Sci. 111 (1992) 46–58. [16] A.J. Hughes, S.E. Daniel, S. Blankson, A.J. Lees, A clinicopathologic study of 100 cases of Parkinson’s disease, Arch. Neurol. 50 (1993) 140–148. [17] J. Jankovic, M. McDermott, J. Carter, S. Gauthier, C. Goetz, L. Golbe, S. Huber, W. Koller, C. Olanow, I. Shoulson, The Parkinson Study Group, Variable expression of Parkinson’s disease: a base-line analysis of the DATATOP cohort, Neurology 40 (1990) 1529–1534. [18] A. Kavounoudias, J.C. Gilhodes, R. Roll, J.P. Roll, From balance regulation to body orientation: two goals for muscle proprioceptive information processing? Exp. Brain Res. 124 (1999) 80–88. [19] E. Khudados, F.W. Cody, D.J. O’Boyle, Proprioceptive regulation of voluntary ankle movements, demonstrated using muscle vibration, is impaired by Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 67 (1999) 504–510. [20] T. Klockgether, M. Borutta, H. Rapp, S. Spieker, J. Dichgans, A defect of kinesthesia in Parkinson’s disease, Mov. Disord. 10 (1995) 460–465. [21] J.R. Lackner, M.S. Levine, Changes in apparent body orientation and sensory localization induced by vibration of postural muscles: vibratory myesthetic illusions, Aviat. Space Environ. Med. 50 (1979) 346–354. [22] H. Lekhel, K. Popov, A. Bronstein, M. Gresty, Postural responses to vibration of neck muscles in patients with uni- and bilateral vestibular loss, Gait Posture 7 (1998) 228–236. [23] S. Lund, Postural effects of neck muscle vibration in man, Experientia 36 (1980) 1398. [24] P.B.C. Matthews, Where does Sherrington’s ‘muscular sense’ originate? Muscles, joints, corrolary discharges, Annu. Rev. Neurosci. 5 (1962) 189–218.
[25] T. Mergner, T. Rosemeier, Interaction of vestibular, somatosensory and visual signals for postural control and motion perception under terrestrial and microgravity conditions—a conceptual model, Brain Res. Rev. 28 (1998) 118–135. [26] T.F. M¨unte, E.M. J¨obges, B.M. Wieringa, S. Klein, M. Schubert, S. Johannes, R. Dengler, Human evoked potentials to long duration vibratory stimuli: role of muscle afferents, Neurosci. Lett. 216 (1996) 163–166. [27] L.M. Nashner, Fixed patterns of rapid postural responses among leg muscles during stance, Exp. Brain Res. 30 (1977) 13–24. [28] P. O’Suilleabhain, J. Bullard, R.B. Dewey, Proprioception in Parkinson’s disease is acutely depressed by dopaminergic medications, J. Neurol. Neurosurg. Psychiatry 71 (2001) 607–610. [29] M.A. Pastor, B.L. Day, C.D. Marsden, Vestibular induced postural responses in Parkinson’s disease, Brain 116 (1993) 1177–1190. [30] K.E. Popov, H. Lekhel, M. Faldon, A.M. Bronstein, M.A. Gresty, Visual and oculomotor responses induced by neck vibration in normal subjects and labyrinthine-defective patients, Exp. Brain Res. 128 (1999) 343–352. [31] J. Reimer, M. Grabowski, O. Lindvall, P. Hagell, Use and interpretation of on/off diaries in Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 75 (2004) 396–400. [32] J.P. Roll, J.P. Vedel, E. Ribot, Alteration of proprioceptive messages induced by tendon vibration in man: a microneurographic study, Exp. Brain Res. 76 (1989) 213–222. [33] J.P. Roll, J.P. Vedel, R. Roll, Eye, head and skeletal muscle spindle feedback in the elaboration of body references, Prog. Brain Res. 80 (1989) 113–123. [34] M. Schieppati, A. Nardone, Free and supported stance in Parkinson’s disease. The effect of posture and ‘postural set’ on leg muscle responses to perturbation, and its relation to the severity of the disease, Brain 114 (1991) 1227–1244. [35] E. Scholz, H.C. Diener, J. Noth, H. Friedemann, J. Dichgans, M. Bacher, Medium and long latency EMG responses in leg muscles: Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 50 (1987) 66–70. [36] A.L. Smiley-Oyen, H.Y. Cheng, L.D. Latt, M.S. Redfern, Adaptation of vibration-induced postural sway in individuals with Parkinson’s disease, Gait Posture 16 (2002) 188–197. ˇ [37] P. Valkoviˇc, S. Krafczyk, M. Saling, J. Benetin, K. B¨otzel, Postural reactions to neck vibration in Parkinson’s disease, Mov. Disord. 21 (2006) 59–65. [38] J.A. Waterston, M.B. Hawken, S. Tanyeri, P. Jantti, C. Kennard, Influence of sensory manipulation on postural control in Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 56 (1993) 1276–1281. [39] S. Zia, F. Cody, D. O’Boyle, Joint position sense is impaired by Parkinson’s disease, Ann. Neurol. 47 (2000) 218–228.