- Email: [email protected]
Postural control of the trunk during unstable sitting in Parkinson's disease
ARTICLE IN PRESS
Parkinsonism and Related Disorders 12 (2006) 492–498 www.elsevier.com/locate/parkreldis
Original article
Postural control of the trunk during unstable sitting in Parkinson’s disease J.C.E. van der Burga,, E.E.H van Wegenb, M.B. Rietbergb, G. Kwakkelb, J.H. van Diee¨na a
Institute for Fundamental and Clinical Human Movement Sciences, Faculty of Human Movement Sciences, Vrije Universiteit Amsterdam, Van der Boechorststraat 9, 1081 BT Amsterdam, The Netherlands b Department of Rehabilitation, Section Physical Therapy, Vrije Universiteit Medical Center, Amsterdam, The Netherlands Received 24 January 2006; received in revised form 12 June 2006; accepted 20 June 2006
Abstract Postural instability and falls, both common in Parkinson’s disease (PD), have been related to altered trunk control. In this study, we investigated dynamic trunk control with subjects balancing on a seat mounted on a hemisphere, for up to 15 s in five trials. We compared eight PD patients with a fall-history, eight without a fall-history, and eight matched healthy subjects. The number of trials completed without balance loss and the time to balance loss were significantly lower in PD patients as compared to healthy controls, whereas the PD patients with a fall-history did not perform significantly less than the patients without a fall-history. Multivariate analysis of variance showed significant effects of group on movements of the center of pressure (CoP) under the seat with the largest amplitudes among the PD fallers and the smallest amplitudes among the healthy controls. Univariate analyses revealed that this effect was mainly based on a significantly larger root mean square CoP displacement in the medio-lateral direction, with significant post hoc differences between all three groups. Trunk angular deviations were significantly smaller among PD patients than controls. Finally, both CoP movements and trunk movements had a significantly lower frequency content and were thus slower in PD patients than in controls, except for anterior–posterior CoP movements. The results show that trunk control is affected in PD and suggest that these changes may be related to postural instability and fall risk. r 2006 Elsevier Ltd. All rights reserved. Keywords: Parkinson’s disease; Trunk; Balance; Sitting; Postural control
1. Introduction Postural instability and falls are common in Parkinson’s disease (PD) [1,2]. In a recent prospective study over 6 months, about 50% of the PD patients fell, compared to about 15% of healthy elderly subjects [3]. A fall can have serious consequences. It can lead to injuries, of which hip fractures are most serious. In addition, corollaries of a fall-event, such as fear of falling, immobility and functional impairments may lead to serious limitations of activities and participation [4].
Corresponding author. Tel.: +31 20 5988501; fax: 31 20 5988529.
E-mail address: [email protected] (J.H. van Diee¨n). 1353-8020/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.parkreldis.2006.06.007
Almost 75% of falls in PD occur due to the inability to control the mass of the body during execution of activities of daily living, such as turning around, standing up and bending forward [3]. It is believed that adequate control of trunk movements is very important for postural stability, as the upper body constitutes twothirds of the total body weight [5]. Due to the heavy mass and its height above the ground, even small, uncoordinated movements of the trunk may increase the risk of balance loss and falls. Mounting evidence suggests that the trunk control is changed in PD patients [6,7]. Recently, Adkin et al. [8] demonstrated that the amplitude of trunk sway measurements during standing can discriminate between healthy controls and PD patients with recent falls.
ARTICLE IN PRESS J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
Only limited insight into the mechanisms underlying the control of the trunk can be gained by studying postural control during quiet stance. In standing, postural adjustments can be accomplished with a wide range of responses at the ankle, knee, hip and trunk joints independently, or combined [9,10]. In quiet standing and after mild perturbations, ankle responses dominate balance control [11,12] In sitting, postural control of the trunk can be studied without the influence of lower extremity responses. Several studies have used an unstable seat paradigm successfully to study postural control of the trunk in healthy and low back pain subjects [13–15]. With this method, the subjects are requested to maintain balance, while sitting on an inherently unstable hemisphere using only trunk movements. In this exploratory study, we investigated whether PD patients show alterations in dynamic trunk control compared to healthy controls during unstable sitting. An additional aim was to study differences in trunk control between PD patients that have a history of falling (fallers), and PD patients that do not have a history of falling (non-fallers). Balance performance was quantified using endurance time (ET) and number of successful trials. In addition, standard measures of center of pressure (CoP) excursions and trunk angular deviations were calculated, as well as their frequency characteristics. We hypothesized that PD patients, and especially fallers would have decreased performance on the unstable seat compared to healthy controls. In addition, it was expected that CoP and trunk angle measures would show evidence of changes in dynamic balance control. Specifically, we anticipated increased CoP excursions related to postural instability, reduced angular deviations related to trunk stiffness, and reduced frequency content of CoP and trunk excursions related to slower trunk movements, between PD patients and controls, as well as between fallers and non-fallers.
493
2. Methods 2.1. Subjects Sixteen PD patients (8 ~, 8 #) and eight age-matched controls (4 ~, 4 #) participated (Table 1). Eight patients had a history of falls, which was defined as at least two reported falls in the last 12 months. Subjects were excluded if they had other neurological, vascular, musculoskeletal and vestibular disorders that could impair balance control, or serious cognitive impairments. Patients were evaluated on the modified Hoehn and Yahr scale and the Unified Parkinson’s Disease Rating scale (UPDRS) by a trained observer. The subjects gave informed consent, in accordance with the ethical standards of the declaration of Helsinki. The local human subjects ethics committee approved the study. 2.2. Experimental setup and procedure Dynamic trunk postural control was measured using an unstable seat (Fig. 1) [13] which was placed on three vertically oriented one-dimensional force transducers (KAP-E, AST, Germany). At the bottom of a seat, an aluminum hemisphere was attached with a diameter of 39 cm, which introduced a rigid, but unstable surface with three rotational degrees of freedom. A leg and foot support was attached to the seat to prevent the influence of lower body movements on balance control. The footplate was adjusted to support the feet with the knees and hips held at a 901 angle. A safety rail was built around the unstable seat to provide security in instances where the subjects lost their balance. Data collection was initiated after the subjects indicated that they had reached a steady state of balance control. Participants were instructed to sit as quietly as possible, holding their hands above the rail as illustrated in Fig. 1. This posture limited compensatory arm movements and allowed them to grab the rail rapidly, when they lost their
Table 1 Mean clinical characteristics of the PD faller, PD non-faller and control group (standard deviation between brackets) Age (year)
Sex
Weight (kg)
Height (cm)
Hoehn and Yahr
UPDRS total score
UPDRS motor score
PD faller
66.0 (78.9)
73 (711)
171.5 (710.8)
2.9 (70.35)
65.6 (718.0)
40.0 (716.8)
PD non-faller
64.3 (79.5)
74 (712)
172.3 (76.1)
2.6 (70.53)
57.4 (717.7)
37.8 (717.4)
Control
63.1 (710.0)
Three females Five males Three females Five males Four females Four males
79 (712)
177.1 (712.6)
—
o10
o5
p-Values for group differences
0.832
0.596
0.500
0.9551
0.752a
0.6741
UPDRS, Unified Parkinson’s Disease Rating Scale. a Non-parametric testing for group differences involved the two groups of PD patients only.
ARTICLE IN PRESS 494
J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
Fig. 1. Subject positioned on the unstable seat.
balance during the trials. The participants wore a bracelet, which emitted a pulse when the rail was touched before the end of the trial. Two minutes of practice were allowed before collecting the data. Five trials of 15 s were performed with 30 s rest between the trials. Subjects were asked to hold on to the safety rail in front of them at all times between the trials to prevent additional learning and fatigue. A goniometer (Penny and Giles type XM, Biometrics Inc.) was used to measure two-dimensional angular deviations between the pelvis and the upper part of the trunk. The goniometer was placed on the back spanning the distance between approximately the T7 and L3 vertebrae and measured the angular deviations in anterior–posterior (AP) and medio-lateral (ML) direction. All data was collected with a sample frequency of 100 Hz and a 20-bit resolution. 2.3. Data analysis Balance performance was quantified using the percentage of trials in which the subjects had to grab the rail for assistance. In addition, the ET was determined, which was defined as the time of start of data collection to either the end of the trial, or the moment of hand contact with the safety rail. In addition, standard characteristics of the CoP excursions were calculated. These were root-mean squared (RMS) excursion (AP and ML), peak excursion (AP and ML) as well sway path and sway area per second. Trunk movements were quantified using the RMS of trunk angular deviations (AP and ML). Finally, the frequency contents of the CoP excursions and trunk
angular deviations were quantified using the mean power frequency (MPF). Only trials in which subjects were able to maintain balance for at least 5 s were included in the analysis. In trials in which subjects lost their balance, the last second before the safety rail was grabbed was excluded from the analysis. 2.4. Statistical analysis First, age, weight, and height of the three subject groups were compared using univariate analyses of variance. Hoehn and Yahr scores and UPDRS scores were compared between the two patient groups using the Mann–Whitney U test. Because of their non-normal distribution, non-parametric (Mann–Whitney U) tests were also performed to compare the primary outcomes: the percentage of trials completed and the ETs, between PD patients and controls and between PD patients with a fall-history and without. Separate MANOVAs were used to analyze the three sets of additional variables: the standard measures of CoP excursion, the angular deviations of the trunk, and the MPF values of CoP and trunk movements. In case significant group differences were found, planned comparisons were made comparing controls to PD patients and PD fallers to PD non-fallers. Effects were considered significant when a p-value o0.05 was found.
3. Results The demographic characteristics were not significantly different between the three groups, nor were the
ARTICLE IN PRESS J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
Hoehn and Yahr and UPDRS scores different between the two patient groups (po0:05, see Table 1), confirming the comparability of the groups. 3.1. ET and successful trials The percentage of successful trials was significantly smaller (p ¼ 0:027) and the ET was significantly shorter (p ¼ 0:045) in the PD patients compared to the healthy controls (Fig. 2). The differences between the PD-fallers
group and the PD non-fallers group were not significant. The percentage successful trials was highly variable within the PD-faller group. Three PD fallers were able to complete all five trials, while in total four successful trials were seen in the other five subjects. One subject could not sit on the unstable seat for more than 4 s. 3.2. CoP excursions A systematic trend was observed for all COP excursion measures: values consistently increased from controls to PD non-fallers to PD fallers (Table 2). In a multivariate analysis including all measures, the CoP excursion was significantly different between the three groups (p ¼ 0:012, see Fig. 3). The ML RMS excursion was significantly larger in the PD fallers than in the nonfallers and in the controls (p ¼ 0:011). In addition, maximal ML CoP excursion, sway area and sway path showed trends towards significance (resp. p ¼ 0:085, 0:070 and 0:069).
2000
Endurance time (ms)
495
PD fallers PD non-fallers controls 1500
1000
500
3.3. Trunk angular deviations 0
80
In a multivariate analysis, the RMS angular deviations of the trunk were significantly smaller in the PD patients compared to controls (po0:001; Table 3). Specifically, RMS angular deviations in AP direction were significantly larger in the healthy group compared to the PD patients (po0:001).
60
3.4. Mean power frequency
40
The expectation regarding slowness of movement in the patient group was confirmed. Overall, a decrease in the mean values of the MPF was observed from controls to PD non-fallers to PD fallers, with a highly significant effect in the MANOVA (Table 3). In univariate analyses, a significant difference between groups was found for all MPF values, except for the MPF of the CoP excursions in AP direction (Table 3).
Percentage successful trials
120 100
PD fallers PD non-fallers controls
20 0
Fig. 2. Bar graphs of the percentage successful trials in each group and the mean endurance times for the three groups (PD fallers, PD nonfallers and the controls). Brackets indicate standard deviations.
Table 2 Mean values of the CoP excursion (mean (SD) in mm) CoP
PD fallers
PD non-fallers
Controls
p-Value
Multivariate RMSml RMSap CoPmax_ml CoPmax_ap Sway area/s Sway path/s
5.4 6.0 20.8 24.6 6.9 692
3.8 5.1 15.9 22.4 5.5 551
3.5 4.8 16.0 20.2 5.2 518
0.012 0.011 0.223 0.085 0.318 0.070 0.069
(4.3) (3.3) (14.7) (12.6) (4.0) (4 0 2)
(2.0) (3.3) (8.1) (13.8) (3.1) (3 0 6)
(2.8) (2.5) (6.9) (8.8) (3.2) (2 3 7)
PDf4PDnf4C
PDf4PDnf4C
PD, Parkinson’s disease; f, fallers; nf, non-fallers; C, controls; RMS, root mean squared; ml, medio-lateral; ap, anterior–posterior; CoP, center of pressure; max, maximal excursion.
ARTICLE IN PRESS J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
496
CoP excursion PD faller
AP excursion (in mm)
20
0
-20
CoPexcursion PD non-faller
AP excursion (in mm)
20
0
-20
CoP excursion control subject
AP excursion (in mm)
20
0
-20
-20
0
20
ML-excursion (in mm) Fig. 3. The center of pressure excursions of a typical Parkinson faller, PD non-faller and a healthy control.
The MPF of the CoP excursions was significantly lower in the PD patients compared to the healthy controls, but only in ML direction (po0:01, Table 3). In both AP and ML direction, the MPF of trunk angular deviations was significantly lower in PD patients compared to controls (Table 3). No significant differences were seen between the MPF of the PD fallers and the PD non-fallers. 4. Discussion As hypothesized, differences were observed in dynamic balance control of unstable sitting between PD
patients and healthy controls. These differences did not only occur in ET and number of successful trials, but also in the amplitude of CoP excursions, trunk angular deviations and their frequency characteristics. The trends of increasing CoP excursions from healthy controls to PD patients non-susceptible for falling to patients who are at risk for falling suggest that increased postural instability is accompanied by decreased task performance. Our results are consistent with a study by Adkin et al. [8], in which larger trunk sway velocity measures in PD patients during stance were interpreted as a sign of poorer balance control. Our results extend these findings in showing that also during unstable sitting, in which influence of lower extremity movement has been eliminated, PD patients show alterations in balance control. In the present study, observed differences were especially evident in the ML direction, perhaps reflecting increased effort to maintain balance in this direction. This is in line with several previous studies that have described ML changes in postural control during stance in PD patients [16–19]. These poor postural responses to a lateral perturbation in stance have been linked to decreased trunk coordination in this population [20]. It has been previously described that PD patients tend to fall mainly in the backward direction after perturbations in stance [20,21], which appears to contradict the observed differences in the ML direction. Changes in ML sway measures may reflect ML instability or poor trunk control in this direction, but how this relates to falls is not clear. During stance the center of mass is controlled by trunk as well as lower extremity movements [22], so movements from the lower extremities also play a role in postural corrections after perturbations. In the current sitting task, contributions of the lower extremities are minimized. In addition, in the sitting task the base of support is very different (i.e. minimal and equal in all directions) compared to standing. Patients may, in an attempt to compensate for instability in AP direction, increase the movements in ML direction [18], although it can be questioned whether this is mechanically effective. Stooped posture and the inability to extend the trunk [23] may limit postural adjustments in AP direction and with that adequate postural corrections. In our experiment, we found indications that the trunk movements in AP direction are limited in PD patients, as the trunk movements in AP direction were decreased, while the trunk movements in ML direction showed no significant changes between PD patients and healthy controls. However, how this can explain the increased ML CoP excursions remains unclear and requires more study. The lower frequency content of trunk and CoP movements in PD patients, reflected in the lower MPF
ARTICLE IN PRESS J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
497
Table 3 Mean values of the trunk angles excursion (mean (SD) in deg.) and the frequency content of the CoP excursions and the trunk angular deviations Trunk angle Multivariate RMS Gonioml RMS Gonioap MPF Multivariate MPF CoPml MPF CoPap MPF Gonioap MPF Gonioml
PD fallers
1.9 (0.7) 2.3 (1.9)
0.30 0.31 2.1 2.3
(0.12) (0.11) (1.8) (1.7)
PD non-fallers
2.2 (1.7) 1.6 (1.1)
0.33 0.34 2.2 2.7
(0.09) (0.12) (1.6) (1.9)
Controls
p-Value
PDf4PDnf4C
2.1 (0.9) 4.0 (2.4)
o0.001 0.70 o0.001
PDf+PDnfoC
0.35 0.32 3.6 3.3
(0.10) (0.12) (1.4) (1.3)
o0.001 o0.01 0.63 o0.01 0.05
PDf+PDnfoC PDf+PDnfoC PDf+PDnfoC
PD, Parkinson’s disease; f, fallers; nf, non-fallers; C, controls. RMS, root mean squared; ml, medio-lateral; ap, anterior–posterior; MPF, mean power frequceny; CoP, center of pressure; Gonio, trunk angular deviations.
values, may be indicative of a reduction in the complexity of the postural control behavior with PD, in line with recent findings by Morales and Kolaczyk [24]. Highfrequency variability in postural patterns has been suggested to reflect deliberate exploratory activity of the postural system, which does not interfere with balance control [25]. The lower frequency content of the CoP excursions and trunk angular deviations in PD patients may therefore indicate that this exploratory behavior is reduced or absent in PD patients. Alternatively, the lower frequency content may also be related to active and/or passive stiffness or rigidity of the trunk. With high trunk stiffness, the trunk and the pelvis will behave like a single inverted pendulum with high inertia above the hemisphere under the unstable seat, which will decrease the movement frequency. The larger CoP excursions in PD patients point also to a higher trunk stiffness in this group, as Reeves et al. [26] showed that high co-activation levels increase the sway path. Visual inspection of video data showed that in the incomplete trials the PD patients fell over as a block, similarly to the results of Martin [27] and later authors [21,28]. The lack of counter-movements of the trunk may be caused by trunk stiffness, or, as Martin [27] stated, may be caused by absence of trunk postural reflexes in PD patients. In more recent literature, it is suggested that PD patients can initiate externally triggered postural synergies, but have problems regulating the gain [28]. However, it is impossible to discriminate in our data between the effects of high trunk stiffness and the effects of lack of, or inappropriately scaled reflexes, as both effects will manifest themselves in the same way. Large differences in balancing performance were observed between the eight individuals in the PD-faller group. Three subjects were able to complete all balance trials. In these patients, other factors than balance deficits, such as reckless behavior and physical activity level in daily life, may have played a role in the occurrence of falls. This reflects a limitation of the
current study as in the determination whether a patient was a faller or not, no distinction was made between socalled external and internal causes of falls. Hence, a fallevent alone may not necessarily distinguish participants with and without balance deficits, in accordance with the results of Laughton et al. [29]. However, trends towards differences between PD fallers and non-fallers may hint at a role that affects postural control of the trunk in the determination of fall risk.
5. Conclusion PD patients show definite alterations in trunk postural control compared to healthy controls. The current results corroborate and extend previous findings that pathological changes in trunk control are related to fall risk and postural instability.
Acknowledgements The authors would like to thank the students Marjolijn Binnekade, Rebecca Baines and Martijn Rhebergen for help with the experiments. This study was supported in part by Grant no. 2004/12 from the research institute MOVE of the VU University Medical Center. References [1] Grimbergen YA, Munneke M, Bloem BR. Falls in Parkinson’s disease. Curr Opin Neurol 2004;17:405–15. [2] Wood BH, Bilclough JA, Bowron A, Walker RW. Incidence and prediction of falls in Parkinson’s disease: a prospective multidisciplinary study. J Neurol Neurosurg Psychiatry 2002;72:721–5. [3] Bloem BR, Grimbergen YA, Cramer M, Willemsen M, Zwinderman AH. Prospective assessment of falls in Parkinson’s disease. J Neurol 2001;248:950–8. [4] Bloem BR, Steijns JA, Smits-Engelsman BC. An update on falls. Curr Opin Neurol 2003;16:15–26.
ARTICLE IN PRESS 498
J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
[5] McConville JT, Churchill TD. Anthropometric relationships of body and body segment moments of inertia. Ohio: Air Force Aerospace Medical Research Laboratory; 1980. [6] Morris ME, Huxham F, McGinley J, Dodd K, Iansek R. The biomechanics and motor control of gait in Parkinson disease. Clin Biomech (Bristol, Avon) 2001;16:459–70. [7] Van Emmerik RE, Wagenaar RC, Winogrodzka A, Wolters EC. Identification of axial rigidity during locomotion in Parkinson disease. Arch Phys Med Rehabil 1999;80:186–91. [8] Adkin AL, Bloem BR, Allum JHJ. Trunk sway measurements during stance and gait tasks in Parkinson’s disease. Gait Posture 2005;22:240–9. [9] Hodges PW, Gurfinkel VS, Brumagne S, Smith TC, Cordo PC. Coexistence of stability and mobility in postural control: evidence from postural compensation for respiration. Exp Brain Res 2002;144:293–302. [10] Krishnamoorthy V, Yang JF, Scholz JP. Joint coordination during quiet stance: effects of vision. Exp Brain Res 2005;164:1–17. [11] Runge CF, Shupert CL, Horak FB, Zajac FE. Ankle and hip postural strategies defined by joint torques. Gait Posture 1999;10: 161–70. [12] Riemann BL, Myers JB, Lephart SM. Comparison of the ankle, knee, hip, and trunk corrective action shown during single-leg stance on firm, foam, and multiaxial surfaces. Arch Phys Med Rehabil 2003;84:90–5. [13] Cholewicki J, Polzhofer GK, Radebold A. Postural control of trunk during unstable sitting. J Biomech 2000;33:1733–7. [14] Radebold A, Cholewicki J, Polzhofer GK, Greene HS. Impaired postural control of the lumbar spine is associated with delayed muscle response times in patients with chronic idiopathic low back pain. Spine 2001;26:724–30. [15] Silfies SP, Cholewicki J, Radebold A. The effects of visual input on postural control of the lumbar spine in unstable sitting. Hum Mov Sci 2003;22:237–52. [16] Morris M, Iansek R, Smithson F, Huxham F. Postural instability in Parkinson’s disease: a comparison with and without a concurrent task. Gait Posture 2000;12:205–16. [17] Hagiwara N, Hashimoto T, Ikeda S. Static balance impairment and its change after pallidotomy in Parkinson’s disease. Mov Disord 2004;19:437–45.
[18] Mitchell SL, Collins JJ, De Luca CJ, Burrows A, Lipsitz LA. Open-loop and closed-loop postural control mechanisms in Parkinson’s disease: increased mediolateral activity during quiet standing. Neurosci Lett 1995;197:133–6. [19] van Wegen EE, van Emmerik RE, Wagenaar RC, Ellis T. Stability boundaries and lateral postural control in Parkinson’s disease. Motor Control 2001;5:254–69. [20] Horak FB, Dimitrova D, Nutt JG. Direction-specific postural instability in subjects with Parkinson’s disease. Exp Neurol 2005; 193:504–21. [21] Carpenter MG, Allum JH, Honegger F, Adkin AL, Bloem BR. Postural abnormalities to multidirectional stance perturbations in Parkinson’s disease. J Neurol Neurosurg Psychiatry 2004;75: 1245–54. [22] Oddsson LIE. Control of voluntary trunk movements in man— mechanisms for postural equilibrium during standing. Acta Physiol Scandinavica 1990;595(140 Suppl):1–60. [23] Bridgewater KJ, Sharpe MH. Trunk muscle performance in early Parkinson’s disease. Phys Ther 1998;78:566–76. [24] Morales CJ, Kolaczyk ED. Wavelet-based multifractal analysis of human balance. Ann Biomed Eng 2002;30:588–97. [25] Riccio GE. Information in movement variability; about the qualitative dynamics of posture and orientation. In: Newell KM, Corcos DM, editors. Variability and motor control. Human Kinetics Publishers: Champaign; 1993. p. 317–58. [26] Reeves NP, Everding V, Cholewicki J, Morrisette DC. The effects of increased active and passive trunk stiffness on spine controllability. Exp Brain Res 2006, doi:10.1007/s00221/-0060516-5. [27] Martin JP. The basal ganglia and posture. London: Pitman; 1967. [28] Dimitrova D, Horak FB, Nutt JG. Postural muscle responses to multidirectional translations in patients with Parkinson’s disease. J Neurophysiol 2004;91:489–501. [29] Laughton CA, Slavin M, Katdare K, Nolan L, Bean JF, Kerrigan DC, et al. Aging, muscle activity, and balance control: physiologic changes associated with balance impairment. Gait Posture 2003;18:101–8.
Parkinsonism and Related Disorders 12 (2006) 492–498 www.elsevier.com/locate/parkreldis
Original article
Postural control of the trunk during unstable sitting in Parkinson’s disease J.C.E. van der Burga,, E.E.H van Wegenb, M.B. Rietbergb, G. Kwakkelb, J.H. van Diee¨na a
Institute for Fundamental and Clinical Human Movement Sciences, Faculty of Human Movement Sciences, Vrije Universiteit Amsterdam, Van der Boechorststraat 9, 1081 BT Amsterdam, The Netherlands b Department of Rehabilitation, Section Physical Therapy, Vrije Universiteit Medical Center, Amsterdam, The Netherlands Received 24 January 2006; received in revised form 12 June 2006; accepted 20 June 2006
Abstract Postural instability and falls, both common in Parkinson’s disease (PD), have been related to altered trunk control. In this study, we investigated dynamic trunk control with subjects balancing on a seat mounted on a hemisphere, for up to 15 s in five trials. We compared eight PD patients with a fall-history, eight without a fall-history, and eight matched healthy subjects. The number of trials completed without balance loss and the time to balance loss were significantly lower in PD patients as compared to healthy controls, whereas the PD patients with a fall-history did not perform significantly less than the patients without a fall-history. Multivariate analysis of variance showed significant effects of group on movements of the center of pressure (CoP) under the seat with the largest amplitudes among the PD fallers and the smallest amplitudes among the healthy controls. Univariate analyses revealed that this effect was mainly based on a significantly larger root mean square CoP displacement in the medio-lateral direction, with significant post hoc differences between all three groups. Trunk angular deviations were significantly smaller among PD patients than controls. Finally, both CoP movements and trunk movements had a significantly lower frequency content and were thus slower in PD patients than in controls, except for anterior–posterior CoP movements. The results show that trunk control is affected in PD and suggest that these changes may be related to postural instability and fall risk. r 2006 Elsevier Ltd. All rights reserved. Keywords: Parkinson’s disease; Trunk; Balance; Sitting; Postural control
1. Introduction Postural instability and falls are common in Parkinson’s disease (PD) [1,2]. In a recent prospective study over 6 months, about 50% of the PD patients fell, compared to about 15% of healthy elderly subjects [3]. A fall can have serious consequences. It can lead to injuries, of which hip fractures are most serious. In addition, corollaries of a fall-event, such as fear of falling, immobility and functional impairments may lead to serious limitations of activities and participation [4].
Corresponding author. Tel.: +31 20 5988501; fax: 31 20 5988529.
E-mail address: [email protected] (J.H. van Diee¨n). 1353-8020/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.parkreldis.2006.06.007
Almost 75% of falls in PD occur due to the inability to control the mass of the body during execution of activities of daily living, such as turning around, standing up and bending forward [3]. It is believed that adequate control of trunk movements is very important for postural stability, as the upper body constitutes twothirds of the total body weight [5]. Due to the heavy mass and its height above the ground, even small, uncoordinated movements of the trunk may increase the risk of balance loss and falls. Mounting evidence suggests that the trunk control is changed in PD patients [6,7]. Recently, Adkin et al. [8] demonstrated that the amplitude of trunk sway measurements during standing can discriminate between healthy controls and PD patients with recent falls.
ARTICLE IN PRESS J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
Only limited insight into the mechanisms underlying the control of the trunk can be gained by studying postural control during quiet stance. In standing, postural adjustments can be accomplished with a wide range of responses at the ankle, knee, hip and trunk joints independently, or combined [9,10]. In quiet standing and after mild perturbations, ankle responses dominate balance control [11,12] In sitting, postural control of the trunk can be studied without the influence of lower extremity responses. Several studies have used an unstable seat paradigm successfully to study postural control of the trunk in healthy and low back pain subjects [13–15]. With this method, the subjects are requested to maintain balance, while sitting on an inherently unstable hemisphere using only trunk movements. In this exploratory study, we investigated whether PD patients show alterations in dynamic trunk control compared to healthy controls during unstable sitting. An additional aim was to study differences in trunk control between PD patients that have a history of falling (fallers), and PD patients that do not have a history of falling (non-fallers). Balance performance was quantified using endurance time (ET) and number of successful trials. In addition, standard measures of center of pressure (CoP) excursions and trunk angular deviations were calculated, as well as their frequency characteristics. We hypothesized that PD patients, and especially fallers would have decreased performance on the unstable seat compared to healthy controls. In addition, it was expected that CoP and trunk angle measures would show evidence of changes in dynamic balance control. Specifically, we anticipated increased CoP excursions related to postural instability, reduced angular deviations related to trunk stiffness, and reduced frequency content of CoP and trunk excursions related to slower trunk movements, between PD patients and controls, as well as between fallers and non-fallers.
493
2. Methods 2.1. Subjects Sixteen PD patients (8 ~, 8 #) and eight age-matched controls (4 ~, 4 #) participated (Table 1). Eight patients had a history of falls, which was defined as at least two reported falls in the last 12 months. Subjects were excluded if they had other neurological, vascular, musculoskeletal and vestibular disorders that could impair balance control, or serious cognitive impairments. Patients were evaluated on the modified Hoehn and Yahr scale and the Unified Parkinson’s Disease Rating scale (UPDRS) by a trained observer. The subjects gave informed consent, in accordance with the ethical standards of the declaration of Helsinki. The local human subjects ethics committee approved the study. 2.2. Experimental setup and procedure Dynamic trunk postural control was measured using an unstable seat (Fig. 1) [13] which was placed on three vertically oriented one-dimensional force transducers (KAP-E, AST, Germany). At the bottom of a seat, an aluminum hemisphere was attached with a diameter of 39 cm, which introduced a rigid, but unstable surface with three rotational degrees of freedom. A leg and foot support was attached to the seat to prevent the influence of lower body movements on balance control. The footplate was adjusted to support the feet with the knees and hips held at a 901 angle. A safety rail was built around the unstable seat to provide security in instances where the subjects lost their balance. Data collection was initiated after the subjects indicated that they had reached a steady state of balance control. Participants were instructed to sit as quietly as possible, holding their hands above the rail as illustrated in Fig. 1. This posture limited compensatory arm movements and allowed them to grab the rail rapidly, when they lost their
Table 1 Mean clinical characteristics of the PD faller, PD non-faller and control group (standard deviation between brackets) Age (year)
Sex
Weight (kg)
Height (cm)
Hoehn and Yahr
UPDRS total score
UPDRS motor score
PD faller
66.0 (78.9)
73 (711)
171.5 (710.8)
2.9 (70.35)
65.6 (718.0)
40.0 (716.8)
PD non-faller
64.3 (79.5)
74 (712)
172.3 (76.1)
2.6 (70.53)
57.4 (717.7)
37.8 (717.4)
Control
63.1 (710.0)
Three females Five males Three females Five males Four females Four males
79 (712)
177.1 (712.6)
—
o10
o5
p-Values for group differences
0.832
0.596
0.500
0.9551
0.752a
0.6741
UPDRS, Unified Parkinson’s Disease Rating Scale. a Non-parametric testing for group differences involved the two groups of PD patients only.
ARTICLE IN PRESS 494
J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
Fig. 1. Subject positioned on the unstable seat.
balance during the trials. The participants wore a bracelet, which emitted a pulse when the rail was touched before the end of the trial. Two minutes of practice were allowed before collecting the data. Five trials of 15 s were performed with 30 s rest between the trials. Subjects were asked to hold on to the safety rail in front of them at all times between the trials to prevent additional learning and fatigue. A goniometer (Penny and Giles type XM, Biometrics Inc.) was used to measure two-dimensional angular deviations between the pelvis and the upper part of the trunk. The goniometer was placed on the back spanning the distance between approximately the T7 and L3 vertebrae and measured the angular deviations in anterior–posterior (AP) and medio-lateral (ML) direction. All data was collected with a sample frequency of 100 Hz and a 20-bit resolution. 2.3. Data analysis Balance performance was quantified using the percentage of trials in which the subjects had to grab the rail for assistance. In addition, the ET was determined, which was defined as the time of start of data collection to either the end of the trial, or the moment of hand contact with the safety rail. In addition, standard characteristics of the CoP excursions were calculated. These were root-mean squared (RMS) excursion (AP and ML), peak excursion (AP and ML) as well sway path and sway area per second. Trunk movements were quantified using the RMS of trunk angular deviations (AP and ML). Finally, the frequency contents of the CoP excursions and trunk
angular deviations were quantified using the mean power frequency (MPF). Only trials in which subjects were able to maintain balance for at least 5 s were included in the analysis. In trials in which subjects lost their balance, the last second before the safety rail was grabbed was excluded from the analysis. 2.4. Statistical analysis First, age, weight, and height of the three subject groups were compared using univariate analyses of variance. Hoehn and Yahr scores and UPDRS scores were compared between the two patient groups using the Mann–Whitney U test. Because of their non-normal distribution, non-parametric (Mann–Whitney U) tests were also performed to compare the primary outcomes: the percentage of trials completed and the ETs, between PD patients and controls and between PD patients with a fall-history and without. Separate MANOVAs were used to analyze the three sets of additional variables: the standard measures of CoP excursion, the angular deviations of the trunk, and the MPF values of CoP and trunk movements. In case significant group differences were found, planned comparisons were made comparing controls to PD patients and PD fallers to PD non-fallers. Effects were considered significant when a p-value o0.05 was found.
3. Results The demographic characteristics were not significantly different between the three groups, nor were the
ARTICLE IN PRESS J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
Hoehn and Yahr and UPDRS scores different between the two patient groups (po0:05, see Table 1), confirming the comparability of the groups. 3.1. ET and successful trials The percentage of successful trials was significantly smaller (p ¼ 0:027) and the ET was significantly shorter (p ¼ 0:045) in the PD patients compared to the healthy controls (Fig. 2). The differences between the PD-fallers
group and the PD non-fallers group were not significant. The percentage successful trials was highly variable within the PD-faller group. Three PD fallers were able to complete all five trials, while in total four successful trials were seen in the other five subjects. One subject could not sit on the unstable seat for more than 4 s. 3.2. CoP excursions A systematic trend was observed for all COP excursion measures: values consistently increased from controls to PD non-fallers to PD fallers (Table 2). In a multivariate analysis including all measures, the CoP excursion was significantly different between the three groups (p ¼ 0:012, see Fig. 3). The ML RMS excursion was significantly larger in the PD fallers than in the nonfallers and in the controls (p ¼ 0:011). In addition, maximal ML CoP excursion, sway area and sway path showed trends towards significance (resp. p ¼ 0:085, 0:070 and 0:069).
2000
Endurance time (ms)
495
PD fallers PD non-fallers controls 1500
1000
500
3.3. Trunk angular deviations 0
80
In a multivariate analysis, the RMS angular deviations of the trunk were significantly smaller in the PD patients compared to controls (po0:001; Table 3). Specifically, RMS angular deviations in AP direction were significantly larger in the healthy group compared to the PD patients (po0:001).
60
3.4. Mean power frequency
40
The expectation regarding slowness of movement in the patient group was confirmed. Overall, a decrease in the mean values of the MPF was observed from controls to PD non-fallers to PD fallers, with a highly significant effect in the MANOVA (Table 3). In univariate analyses, a significant difference between groups was found for all MPF values, except for the MPF of the CoP excursions in AP direction (Table 3).
Percentage successful trials
120 100
PD fallers PD non-fallers controls
20 0
Fig. 2. Bar graphs of the percentage successful trials in each group and the mean endurance times for the three groups (PD fallers, PD nonfallers and the controls). Brackets indicate standard deviations.
Table 2 Mean values of the CoP excursion (mean (SD) in mm) CoP
PD fallers
PD non-fallers
Controls
p-Value
Multivariate RMSml RMSap CoPmax_ml CoPmax_ap Sway area/s Sway path/s
5.4 6.0 20.8 24.6 6.9 692
3.8 5.1 15.9 22.4 5.5 551
3.5 4.8 16.0 20.2 5.2 518
0.012 0.011 0.223 0.085 0.318 0.070 0.069
(4.3) (3.3) (14.7) (12.6) (4.0) (4 0 2)
(2.0) (3.3) (8.1) (13.8) (3.1) (3 0 6)
(2.8) (2.5) (6.9) (8.8) (3.2) (2 3 7)
PDf4PDnf4C
PDf4PDnf4C
PD, Parkinson’s disease; f, fallers; nf, non-fallers; C, controls; RMS, root mean squared; ml, medio-lateral; ap, anterior–posterior; CoP, center of pressure; max, maximal excursion.
ARTICLE IN PRESS J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
496
CoP excursion PD faller
AP excursion (in mm)
20
0
-20
CoPexcursion PD non-faller
AP excursion (in mm)
20
0
-20
CoP excursion control subject
AP excursion (in mm)
20
0
-20
-20
0
20
ML-excursion (in mm) Fig. 3. The center of pressure excursions of a typical Parkinson faller, PD non-faller and a healthy control.
The MPF of the CoP excursions was significantly lower in the PD patients compared to the healthy controls, but only in ML direction (po0:01, Table 3). In both AP and ML direction, the MPF of trunk angular deviations was significantly lower in PD patients compared to controls (Table 3). No significant differences were seen between the MPF of the PD fallers and the PD non-fallers. 4. Discussion As hypothesized, differences were observed in dynamic balance control of unstable sitting between PD
patients and healthy controls. These differences did not only occur in ET and number of successful trials, but also in the amplitude of CoP excursions, trunk angular deviations and their frequency characteristics. The trends of increasing CoP excursions from healthy controls to PD patients non-susceptible for falling to patients who are at risk for falling suggest that increased postural instability is accompanied by decreased task performance. Our results are consistent with a study by Adkin et al. [8], in which larger trunk sway velocity measures in PD patients during stance were interpreted as a sign of poorer balance control. Our results extend these findings in showing that also during unstable sitting, in which influence of lower extremity movement has been eliminated, PD patients show alterations in balance control. In the present study, observed differences were especially evident in the ML direction, perhaps reflecting increased effort to maintain balance in this direction. This is in line with several previous studies that have described ML changes in postural control during stance in PD patients [16–19]. These poor postural responses to a lateral perturbation in stance have been linked to decreased trunk coordination in this population [20]. It has been previously described that PD patients tend to fall mainly in the backward direction after perturbations in stance [20,21], which appears to contradict the observed differences in the ML direction. Changes in ML sway measures may reflect ML instability or poor trunk control in this direction, but how this relates to falls is not clear. During stance the center of mass is controlled by trunk as well as lower extremity movements [22], so movements from the lower extremities also play a role in postural corrections after perturbations. In the current sitting task, contributions of the lower extremities are minimized. In addition, in the sitting task the base of support is very different (i.e. minimal and equal in all directions) compared to standing. Patients may, in an attempt to compensate for instability in AP direction, increase the movements in ML direction [18], although it can be questioned whether this is mechanically effective. Stooped posture and the inability to extend the trunk [23] may limit postural adjustments in AP direction and with that adequate postural corrections. In our experiment, we found indications that the trunk movements in AP direction are limited in PD patients, as the trunk movements in AP direction were decreased, while the trunk movements in ML direction showed no significant changes between PD patients and healthy controls. However, how this can explain the increased ML CoP excursions remains unclear and requires more study. The lower frequency content of trunk and CoP movements in PD patients, reflected in the lower MPF
ARTICLE IN PRESS J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
497
Table 3 Mean values of the trunk angles excursion (mean (SD) in deg.) and the frequency content of the CoP excursions and the trunk angular deviations Trunk angle Multivariate RMS Gonioml RMS Gonioap MPF Multivariate MPF CoPml MPF CoPap MPF Gonioap MPF Gonioml
PD fallers
1.9 (0.7) 2.3 (1.9)
0.30 0.31 2.1 2.3
(0.12) (0.11) (1.8) (1.7)
PD non-fallers
2.2 (1.7) 1.6 (1.1)
0.33 0.34 2.2 2.7
(0.09) (0.12) (1.6) (1.9)
Controls
p-Value
PDf4PDnf4C
2.1 (0.9) 4.0 (2.4)
o0.001 0.70 o0.001
PDf+PDnfoC
0.35 0.32 3.6 3.3
(0.10) (0.12) (1.4) (1.3)
o0.001 o0.01 0.63 o0.01 0.05
PDf+PDnfoC PDf+PDnfoC PDf+PDnfoC
PD, Parkinson’s disease; f, fallers; nf, non-fallers; C, controls. RMS, root mean squared; ml, medio-lateral; ap, anterior–posterior; MPF, mean power frequceny; CoP, center of pressure; Gonio, trunk angular deviations.
values, may be indicative of a reduction in the complexity of the postural control behavior with PD, in line with recent findings by Morales and Kolaczyk [24]. Highfrequency variability in postural patterns has been suggested to reflect deliberate exploratory activity of the postural system, which does not interfere with balance control [25]. The lower frequency content of the CoP excursions and trunk angular deviations in PD patients may therefore indicate that this exploratory behavior is reduced or absent in PD patients. Alternatively, the lower frequency content may also be related to active and/or passive stiffness or rigidity of the trunk. With high trunk stiffness, the trunk and the pelvis will behave like a single inverted pendulum with high inertia above the hemisphere under the unstable seat, which will decrease the movement frequency. The larger CoP excursions in PD patients point also to a higher trunk stiffness in this group, as Reeves et al. [26] showed that high co-activation levels increase the sway path. Visual inspection of video data showed that in the incomplete trials the PD patients fell over as a block, similarly to the results of Martin [27] and later authors [21,28]. The lack of counter-movements of the trunk may be caused by trunk stiffness, or, as Martin [27] stated, may be caused by absence of trunk postural reflexes in PD patients. In more recent literature, it is suggested that PD patients can initiate externally triggered postural synergies, but have problems regulating the gain [28]. However, it is impossible to discriminate in our data between the effects of high trunk stiffness and the effects of lack of, or inappropriately scaled reflexes, as both effects will manifest themselves in the same way. Large differences in balancing performance were observed between the eight individuals in the PD-faller group. Three subjects were able to complete all balance trials. In these patients, other factors than balance deficits, such as reckless behavior and physical activity level in daily life, may have played a role in the occurrence of falls. This reflects a limitation of the
current study as in the determination whether a patient was a faller or not, no distinction was made between socalled external and internal causes of falls. Hence, a fallevent alone may not necessarily distinguish participants with and without balance deficits, in accordance with the results of Laughton et al. [29]. However, trends towards differences between PD fallers and non-fallers may hint at a role that affects postural control of the trunk in the determination of fall risk.
5. Conclusion PD patients show definite alterations in trunk postural control compared to healthy controls. The current results corroborate and extend previous findings that pathological changes in trunk control are related to fall risk and postural instability.
Acknowledgements The authors would like to thank the students Marjolijn Binnekade, Rebecca Baines and Martijn Rhebergen for help with the experiments. This study was supported in part by Grant no. 2004/12 from the research institute MOVE of the VU University Medical Center. References [1] Grimbergen YA, Munneke M, Bloem BR. Falls in Parkinson’s disease. Curr Opin Neurol 2004;17:405–15. [2] Wood BH, Bilclough JA, Bowron A, Walker RW. Incidence and prediction of falls in Parkinson’s disease: a prospective multidisciplinary study. J Neurol Neurosurg Psychiatry 2002;72:721–5. [3] Bloem BR, Grimbergen YA, Cramer M, Willemsen M, Zwinderman AH. Prospective assessment of falls in Parkinson’s disease. J Neurol 2001;248:950–8. [4] Bloem BR, Steijns JA, Smits-Engelsman BC. An update on falls. Curr Opin Neurol 2003;16:15–26.
ARTICLE IN PRESS 498
J.C.E. van der Burg et al. / Parkinsonism and Related Disorders 12 (2006) 492–498
[5] McConville JT, Churchill TD. Anthropometric relationships of body and body segment moments of inertia. Ohio: Air Force Aerospace Medical Research Laboratory; 1980. [6] Morris ME, Huxham F, McGinley J, Dodd K, Iansek R. The biomechanics and motor control of gait in Parkinson disease. Clin Biomech (Bristol, Avon) 2001;16:459–70. [7] Van Emmerik RE, Wagenaar RC, Winogrodzka A, Wolters EC. Identification of axial rigidity during locomotion in Parkinson disease. Arch Phys Med Rehabil 1999;80:186–91. [8] Adkin AL, Bloem BR, Allum JHJ. Trunk sway measurements during stance and gait tasks in Parkinson’s disease. Gait Posture 2005;22:240–9. [9] Hodges PW, Gurfinkel VS, Brumagne S, Smith TC, Cordo PC. Coexistence of stability and mobility in postural control: evidence from postural compensation for respiration. Exp Brain Res 2002;144:293–302. [10] Krishnamoorthy V, Yang JF, Scholz JP. Joint coordination during quiet stance: effects of vision. Exp Brain Res 2005;164:1–17. [11] Runge CF, Shupert CL, Horak FB, Zajac FE. Ankle and hip postural strategies defined by joint torques. Gait Posture 1999;10: 161–70. [12] Riemann BL, Myers JB, Lephart SM. Comparison of the ankle, knee, hip, and trunk corrective action shown during single-leg stance on firm, foam, and multiaxial surfaces. Arch Phys Med Rehabil 2003;84:90–5. [13] Cholewicki J, Polzhofer GK, Radebold A. Postural control of trunk during unstable sitting. J Biomech 2000;33:1733–7. [14] Radebold A, Cholewicki J, Polzhofer GK, Greene HS. Impaired postural control of the lumbar spine is associated with delayed muscle response times in patients with chronic idiopathic low back pain. Spine 2001;26:724–30. [15] Silfies SP, Cholewicki J, Radebold A. The effects of visual input on postural control of the lumbar spine in unstable sitting. Hum Mov Sci 2003;22:237–52. [16] Morris M, Iansek R, Smithson F, Huxham F. Postural instability in Parkinson’s disease: a comparison with and without a concurrent task. Gait Posture 2000;12:205–16. [17] Hagiwara N, Hashimoto T, Ikeda S. Static balance impairment and its change after pallidotomy in Parkinson’s disease. Mov Disord 2004;19:437–45.
[18] Mitchell SL, Collins JJ, De Luca CJ, Burrows A, Lipsitz LA. Open-loop and closed-loop postural control mechanisms in Parkinson’s disease: increased mediolateral activity during quiet standing. Neurosci Lett 1995;197:133–6. [19] van Wegen EE, van Emmerik RE, Wagenaar RC, Ellis T. Stability boundaries and lateral postural control in Parkinson’s disease. Motor Control 2001;5:254–69. [20] Horak FB, Dimitrova D, Nutt JG. Direction-specific postural instability in subjects with Parkinson’s disease. Exp Neurol 2005; 193:504–21. [21] Carpenter MG, Allum JH, Honegger F, Adkin AL, Bloem BR. Postural abnormalities to multidirectional stance perturbations in Parkinson’s disease. J Neurol Neurosurg Psychiatry 2004;75: 1245–54. [22] Oddsson LIE. Control of voluntary trunk movements in man— mechanisms for postural equilibrium during standing. Acta Physiol Scandinavica 1990;595(140 Suppl):1–60. [23] Bridgewater KJ, Sharpe MH. Trunk muscle performance in early Parkinson’s disease. Phys Ther 1998;78:566–76. [24] Morales CJ, Kolaczyk ED. Wavelet-based multifractal analysis of human balance. Ann Biomed Eng 2002;30:588–97. [25] Riccio GE. Information in movement variability; about the qualitative dynamics of posture and orientation. In: Newell KM, Corcos DM, editors. Variability and motor control. Human Kinetics Publishers: Champaign; 1993. p. 317–58. [26] Reeves NP, Everding V, Cholewicki J, Morrisette DC. The effects of increased active and passive trunk stiffness on spine controllability. Exp Brain Res 2006, doi:10.1007/s00221/-0060516-5. [27] Martin JP. The basal ganglia and posture. London: Pitman; 1967. [28] Dimitrova D, Horak FB, Nutt JG. Postural muscle responses to multidirectional translations in patients with Parkinson’s disease. J Neurophysiol 2004;91:489–501. [29] Laughton CA, Slavin M, Katdare K, Nolan L, Bean JF, Kerrigan DC, et al. Aging, muscle activity, and balance control: physiologic changes associated with balance impairment. Gait Posture 2003;18:101–8.