G Model
GAIPOS-4257; No. of Pages 6 Gait & Posture xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals Yannick Mullie a,b, Cyril Duclos a,b,* a Pathokinesiology Laboratory, Centre for Interdisciplinary Research in Rehabilitation (CRIR), Institut de re´adaptation Gingras-Lindsay-de-Montre´al (IRGLM), 6300 avenue Darlington, Montreal, QC H3S 2J4, Canada b School of Rehabilitation, Universite´ de Montreal, 7077 avenue du Parc, Montreal, QC H3N 1X7, Canada
A R T I C L E I N F O
A B S T R A C T
Article history: Received 16 January 2014 Received in revised form 29 May 2014 Accepted 11 July 2014
Proprioceptive information is important for balance control yet little is known about how it is used during gait or how a stroke affects its use. The aim of this study was to evaluate the role of proprioception in controlling balance during gait in healthy participants and after stroke. Twelve healthy and 9 hemiparetic participants walked on an instrumented treadmill in a fully lit room, while whole-body, three-dimensional kinematics were quantified. Vibration was applied continuously or during the stance phase only, on the posterior neck muscles and triceps surae tendon on the non-dominant/paretic side. Difficulty in maintaining dynamic and postural balance was evaluated using stabilizing and destabilizing forces, respectively. Continuous and stance phase vibration of the triceps surae reduced the difficulty in maintaining both dynamic and postural balance in healthy participants (p < .05), with a greater distance between the center of pressure and the limit of the potential base of support, a more backward body position, and no change in spatio-temporal gait parameters. No effect of neck muscle vibration was observed on balance (p = .63 and above). None of the vibration conditions affected balance or gait parameters among stroke participants. The results confirmed that proprioceptive information was not used to control balance during gait in stroke participants. The importance of proprioceptive information may depend on other factors such as walking and visual conditions. Changes in sensory integration ability likely explain the results after stroke. Further study is needed to understand the integration of proprioceptive and visual information to control balance during gait after stroke. ß 2014 Elsevier B.V. All rights reserved.
Keywords: Kinesthesis Equilibrium Locomotion Cerebrovascular accident Human
1. Introduction Control of body posture and dynamics, i.e., segment alignment and velocity of the center of mass (CoM), depends on the integration and weighting of somatosensory, vestibular and visual sensory information [1,2]. Among them, muscle proprioception continuously informs the central nervous system (CNS) on the position of each segment of the body [3]. Muscle vibration is commonly used to study the role of proprioceptive information in motor control as it generates a strong proprioceptive stimulation when applied at a frequency of 80–100 Hz and with amplitude between 0.5 and 1 mm [4–6]. It also induces a contraction of the
* Corresponding author at: Pathokinesiology Laboratory, Centre for Interdisciplinary Research in Rehabilitation (CRIR), Institut de re´adaptation Gingras-Lindsay-de-Montre´al (IRGLM), 6300 avenue Darlington, Montreal, QC H3S 2J4, Canada. E-mail address: [email protected] (C. Duclos).
vibrated muscle or its antagonist depending on the conditions of application [7]. When applied during quiet standing, muscle vibration induces oriented postural reactions [8,9]: vibration of the triceps surae resulted in a backward postural response [8] while vibration to the back neck muscle resulted in a forward postural reaction [8,10]. During gait, the role of proprioception in controlling balance has scarcely been studied. Vibration has been shown to affect kinematics, speed, and muscle activity during gait [9,11,12]. The results most closely related to balance control during gait showed changes in CoM displacements and accelerations during gait when vibration was applied at the ankle [11]. Contrary to quiet standing, sensory activity during gait is cyclical (due to repetitive movements of the limbs) and random in the case of gait perturbations. As such, sensory activity plays different roles in the motor control of gait. For example, the transition between stance and swing phase, and the organization of alternated flexor and extensor activity in the lower limbs is mediated by phasic sensory activity [13]. Moreover, sensory
http://dx.doi.org/10.1016/j.gaitpost.2014.07.008 0966-6362/ß 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
G Model
GAIPOS-4257; No. of Pages 6 Y. Mullie, C. Duclos / Gait & Posture xxx (2014) xxx–xxx
2
background affects responses to new sensory information: during standing, the position of the joints alters gains in cutaneous reflexes [14], and instability of the support surface inhibits postural reactions to vibration of ankle muscles [15]. Thus, sensory afferents may have different effects on balance control during gait depending on the gait phase examined. The use of proprioceptive information to control balance is also affected by neurological lesions such as stroke. Persons with hemiparesis generally have impaired proprioception [16], associated with increased postural sway [17]. Alterations in the center of pressure (CoP) excursion due to sensory manipulation during quiet standing [17] or gait [18] also indicate that stroke-related sensorimotor impairments affect the neuromuscular activity necessary to control balance. However, gait speed has been increased via proprioceptive stimulation applied at the ankle during gait in participants with hemiparesis [19]. Ankle vibration only affected temporal (vs. spatial) gait parameters in stroke patients, regardless of any ankle joint position sense impairment [20], suggesting that the use of proprioception information to control gait might be reduced at the sensorimotor integration level rather than the sensory perception level after stroke [20–22]. To our knowledge, no study has evaluated how proprioceptive information is used to control balance during gait after stroke. The first objective of the study was to evaluate how continuous and phasic proprioceptive information from the neck and ankle alters postural and dynamic balance during gait in healthy participants. A secondary objective was to determine how hemiparesis due to stroke affects the use of proprioceptive information in postural and dynamic balance control during gait.
2. Methods 2.1. Participants A convenience sample of 21 volunteers (12 healthy, 9 with hemiparesis due to stroke) was recruited for this study. The inclusion criteria were to be able to walk on a treadmill without any assistance, have no orthopedic or neurological problems affecting gait, or cognitive deficits prior to the experiment (for healthy participants) or prior to stroke (for hemiparetic participants) and be able to sustain 90 min of activity with rest periods as
required. The main characteristics of the participants are presented in Table 1. All participants gave written consent to participate in the study after having been informed of the details of the experiment according to local ethics board recommendations. 2.2. Data collection and proprioceptive stimulation Three-dimensional whole-body kinematics were recorded at 60 Hz with an Optotrak Certus system (NDI, Waterloo, Canada), using three to six non-collinear infrared markers placed on each main segment of the body (15 segments for a total of 75 markers). A digitizing probe was used to locate the contour of the shoe soles, with respect to the infrared markers on the respective foot segments, and anatomical landmarks to complete the definition of the rigid bodies representing each body segment along with anthropometric measurements and to define a 3-D link-segment model for each participant [23]. Ground reaction forces and moments were measured under each foot using an instrumented split-belt treadmill (Bertec Fit1). Kinetic data were collected at a frequency of 600 Hz, filtered with a fourth-order Butterworth zero-lag filter with a cut-off frequency of 10 Hz and re-sampled at 60 Hz to match the kinematic data. Belt speed was set to the participants’ comfortable gait speed using progressive speed increases and decreases. Proprioceptive stimulation was applied using an electromechanical vibrator (VB115, Technoconcept, France) at 80 Hz (amplitude between 0.5 and 1 mm) on the tendon of the nondominant or paretic triceps surae and on the bilateral posterior neck muscles during gait. Continuous (with the vibrator on throughout the entire trial) and phasic (with the vibrator turned on only when the heel was in contact with the ground as detected by a foot switch placed on the heel of the shoe) modes of vibration were used. Preliminary tests showed delays between vibrator activations and stance phase of about less than 80 ms. 2.3. Experimental protocol The following conditions were tested among all participants: control condition without vibration, posterior neck muscle vibration in the continuous and phasic mode, and triceps surae tendon vibration in the continuous and phasic mode. The control
Table 1 Clinical characteristics of stroke participants. Age (years)
S1 S2 S3 S4 S5 S6 S7 S8 S9 Mean SD or range Mean SD
60 32 64 53 55 54 39 37 36 47.8 11.8 47.6 14.5
BMI (kg/m2)
31.2 21.6 27.5 22.8 27.7 25.7 25.7 37.4 20.6 26.7 5.2 26.4 6.0
OG velocity (m/s)
0.85 0.82 0.80 0.72 0.73 0.55 1.08 1.31 1.22 0.90 0.25 1.55 0.15
Treadmill velocity (m/s)
0.35 0.75 0.60 0.70 0.70 0.55 0.65 0.95 0.60 0.65 0.16 1.02 0.17
Sensory perception
CMMSA Leg score
Foot score
Cut. (/4)
Mov. (/10)
Pall. (s)
5 6 5 4 4 6 5 6 5 5 [4:6]
4 4 3 2 4 4 4 5 3 4 [2:4]
3/3 4/4 3/3 3/3 3/3 2/2 4/3 2/2 4/4 3/3 [2:4/2:4]
10/10 10/10 10/10 10/10 10/10 10/10 10/10 10/10 10/10 10/10 [10/10]
13.2/13.2 14.7/15.4 9.6/10.0 10.7/11.6 17.1/14.0 13.1/12.4 8.0/9.2 14.1/14.8 15.5/16.2 12.9/13.0 2.9/2.4
BMI, body mass index; OG, overground; CMMSA, Chedoke-McMaster Stroke Assessment; SD, standard deviation. Pallesthesia: duration perception with a 128 Hz tuning fork on the malleolus lateralis. Mov.: Number of passive movements at the big toe perceived correctly out of 10. Cut.: cutaneous perception of 2 out of 3 Semmes–Weinstein filament contacts on the malleolus lateralis; 1: anesthesia (6.65 filament perceived); 2: severe deficit (5.18 filament perceived); 3: hypoesthesia (4.31 filament perceived); 4: normal sensitivity (4.17 filament perceived). Paretic/Non-paretic are indicated for these three tests. The last two lines in italics are the mean results for the group of healthy participants.
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
G Model
GAIPOS-4257; No. of Pages 6 Y. Mullie, C. Duclos / Gait & Posture xxx (2014) xxx–xxx
condition was always recorded first. The first site of stimulation (triceps or neck) was alternated between participants. The two modes of vibration (continuous or phasic) were applied in a pseudo-random order for each muscle. One-minute trials were recorded for each condition at a comfortable gait speed on the treadmill. A 90-s familiarization period was allowed with vibration before data collection. Control trials were also performed in quiet standing position, eyes closed. Vibration was tested at each site in one trial, before the corresponding experimental gait trials. The participants were standing on the non-moving treadmill during 30-s trials. After 10 s of quiet standing, the vibrator was turned on for 20 s until the end of the trial. Only the ground reaction forces were recorded. 2.4. Data analysis For the gait trials, kinetic and kinematic data were translated according to the treadmill belt speed to yield forward displacements of the CoM and CoP, equivalent to overground walking [24]. The positions of the CoM, CoP and limits of the potential base of support (pBoS, defined as the vertical ground projection of the contour of both feet for the whole gait cycle) were obtained from the translated kinetic and kinematic data and used to calculate the stabilizing and destabilizing forces [25,26] for each step in the trials. The stabilizing force is the theoretical force needed to stop the motion of the body’s CoM and CoP at the limit of the pBoS in the direction of the displacement of the body (Eq. (1)). Its maximal value indicates the greatest difficulty in dynamic balance, and was calculated. The destabilizing force is the theoretical force necessary to move the body into an unstable position (Eq. (2)). Its minimal value represents the greatest postural difficulty during the task. The distance between the CoP and the limit of the pBoS, the position of the CoM relative to the anteroposterior length of the pBoS, and velocity of the CoM were also calculated to further understand any significant effect of vibration on balance difficulty. These values were averaged over a minimum of 20 steps for each foot. The stabilizing force mglobal :~ vCM :~ vCM ~ ~ F ST ¼ DCP 2D2CP
(1)
vCM : linear velocity of the body center of mglobal: body mass, ~ ~CP : distance between the center of pressure and the limit of mass, D
3
the potential base of support. The destabilizing force
~ FD ¼
! ~ n ~ Fr ~ DCP hCM
(2)
~ n: normal unitary vector to ground F r : ground reaction forces, ~ ~CP : distance between surface, hCM: height of the center of mass, D the center of pressure and the limit of the potential base of support. Spatiotemporal gait parameters were calculated based on the variations of ground reaction forces and kinematics of malleolus markers: step time (i.e., time between two consecutive heel contacts), swing time (i.e., time between toe-off and heel strike of the same leg), step length (i.e., anteroposterior distance between the two malleolus at the time of respective heel contact). For the standing trials, the mean anteroposterior position of the CoP was measured during the first 10 s of the trials, when the vibration was off, and during the last 18 s of the trials, when the vibration was on. For each group, Friedman ANOVAs were used for each variable during gait trials to compare the experimental conditions in each group. Planned contrasts (Wilcoxon ranked test) were applied depending on the results of the ANOVAs to test which vibration condition differed from the control condition without vibration. Wilcoxon ranked tests were also used to compare the position of the CoP before and during vibration in the standing trials, for each vibration site, and each group. For all statistical tests, significance was set at a = .05. The results of the stroke participants during gait trials were also analyzed individually. The percentage of change in stabilizing/ destabilizing force values between the control and experimental conditions was calculated for both groups. For each condition, individual changes among stroke participants were compared to the range of changes measured in the healthy group. Response to vibration was considered similar to the healthy participants when the change observed for each stroke participant was within the healthy participants’ range. 3. Results Data of the gait trials were averaged between dominant and non-dominant steps as Wilcoxon tests did not demonstrate any differences among the healthy or stroke participants (p = .21 and above). In both groups, vibration stimulation made no difference in any of the spatiotemporal characteristics (step length, step and swing time (p = .20 and above) (Table 2).
Table 2 Spatial gait parameters for each experimental condition.
Healthy participants CoM position (%) CoP distance (cm) CoM velocity (m/s) Step time (s) Swing time (s) Step length Stroke participants Paretic step time (s) Non-paretic step time (s) Paretic swing time (s) Non-paretic swing time Paretic step length Non-paretic step length
Control
Triceps continuous
62.19 10.37 1.02 1.12 0.35 0.51
(3.86) (0.03) (0.21) (0.09) (0.03) (0.07)
61.22 11.22 1.01 1.12 0.35 0.51
1.34 1.34 0.42 0.35 0.38 0.42
(0.13) (0.13) (0.06) (0.06) (0.09) (0.09)
1.34 1.34 0.39 0.35 0.38 0.42
(3.57)** (0.03)*** (0.21) (0.12) (0.04) (0.07) (0.14) (0.14) (0.05) (0.09) (0.10) (0.10)
Triceps stance
55.59 11.53 1.02 1.13 0.36 0.51 1.35 1.35 0.40 0.34 0.39 0.42
Neck continuous
Neck stance
(2.48)*** (0.03)*** (0.21) (0.12) (0.05) (0.07)
1.12 (0.10) 0.35 (0.04) 0.51 (0.07)
1.11 (0.09) 0.35 (0.04) 0.51 (0.07)
(0.14) (0.14) (0.06) (0.08) (0.09) (0.09)
1.28 1.28 0.39 0.32 0.38 0.42
1.33 1.32 0.40 0.34 0.38 0.41
(0.08) (0.08) (0.04) (0.09) (0.11) (0.10)
(0.12) (0.11) (0.05) (0.09) (0.10) (0.09)
CoM position is the position of the center mass (CoM) relative to the anteroposterior length of the potential base of support. CoP distance is the distance between the center of pressure (CoP) and the limit of the potential base of support in the direction of the CoM displacement. Mean (standard deviation) are presented. Empty cells indicates that the values were not evaluated in the absence of changes in balance variables. ** p < .01. *** p < .005.
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
G Model
GAIPOS-4257; No. of Pages 6 4
Y. Mullie, C. Duclos / Gait & Posture xxx (2014) xxx–xxx
3.1. Healthy participants
A. Triceps surae vibraon (standing)
Only ankle muscle vibration altered the difficulty in maintaining both postural and dynamic balance, as measured by destabilizing and stabilizing forces, respectively (Friedman ANOVA, x2(4) = 12.2, p < .05, Fig. 1A). The vibration of triceps surae in continuous and phasic modes significantly decreased the stabilizing force (Wilcoxon ranked test, p < .05) and increased the destabilizing force (p < .01). However, continuous and stance phase neck muscle vibration did not alter the stabilizing (p = .63 and p = .81, respectively) or destabilizing force values (p = .38 and p = .27, respectively). In the conditions where balance was altered, vibration application changed the distance between the CoP and the limit of the pBoS (x2(4) = 15.9, p < .005) and the relative position of the CoM (x2(4) = 27.8, p < .001) depending on its location (Table 2). During ankle muscle vibration, the distance between the CoP and the anterior limit of the pBoS increased during both continuous and stance phase vibration (p < .005). In terms of body position over the pBoS, the relative position of the CoM was more backward for stance phase (p < .005) and continuous vibration (p < .01). No significant alteration of these variables was found during neck muscle vibration (p = .07). Neither the location nor the mode of vibration significantly altered the velocity of the CoM (x2(4) = 1.3, p = .23).
0.01
(m)
0.00
10
20
30 (s)
-0.01 -0.02 -0.03 -0.04 -0.05
B. Neck muscle vibraon (standing) 0.12
(m)
0.10 0.08 0.06 0.04
3.2. Stroke participants Ankle and neck muscle vibration did not alter the difficulty in maintaining postural and dynamic balance (x2(4) = 1.8, p = .77 and x2(4) = 4.6, p = .33, respectively, Fig. 1). During both continuous and stance phase ankle vibration conditions, the individual analysis indicated that the changes in stabilizing force for each stroke participant were lower than the range of changes measured among the healthy participants ([2.2; – 19.8%]). Similarly, the changes in destabilizing force during continuous and stance phase ankle vibration were lower in each stroke participant than the range measured in the healthy participants ([3.1; 18.4%]). During neck muscle vibration, only two and three stroke participants showed changes within the range of the non-significant changes observed in stabilizing and destabilizing forces measured for the healthy participants.
0.02 0.00
10
20
30 (s)
-0.02
Fig. 2. Group mean (thick line) standard deviation (thin line) of the position of the center of pressure (CoP) during standing in healthy (black) and stroke participants (gray) during 30-s trials where 80 Hz vibration was applied on the triceps surae (A) or back neck muscles (B) during the last 20 s of the trial (gray bar at the bottom of each line graph).
3.3. Standing trials
400
A. Healthy parcipants (N=12)
120 (N)
(N)
*
350
The mean position of the CoP was more forward during neck muscle vibration than before in both groups (stroke participants: +1.1 (1.0) cm, p < .05; healthy group: +3.5 (3.2) cm, p < .005) (Fig. 2A). The mean position of the CoP was more backward during triceps surae vibration than before in both groups (stroke participants: 1.1 (0.9) cm, p < .01; healthy group: +1.4 (1.0) cm, p < .005) (Fig. 2B).
110
*
100
300
4. Discussion
90 250
80
200
**
***
150 100
70
Stabilizing force (le axis) Destabilizing force (right axis)
Control Gait
Triceps Connuous
Triceps Stance
Neck Connuous
Neck Stance
B. Stroke parcipants (N=9)
300
60 50
150
(N)
(N)
140
250
130 200
120 110
150
100
100
90 50 0
Stabilizing force (le axis) Destabilizing force (right axis)
Control Gait
Triceps Connuous
Triceps Stance
Neck Connuous
Neck Stance
80 70
Fig. 1. Mean and standard deviation of maximal stabilizing force values (gray square, left axis) and minimal destabilizing force values (black square, right axis) during control gait condition (left) and vibration conditions (4 right columns) in healthy (A) and stroke (B) participants. Values are expressed in Newtons (N). * For p < .05, ** for p .01, and *** for p .005 (planned contrasts, comparison with the control gait condition).
Our results showed that ankle proprioceptive stimulation altered both postural and dynamic components of balance in healthy participants and that stroke affects the use of proprioceptive information to maintain balance during gait. The increase in destabilizing force indicated that the difficulty to maintain postural balance was reduced during the vibration of the triceps surae tendon during gait in healthy participants. This was associated with a position of the CoM 1 to 6.5% more posterior within the pBoS. Although of small amplitude, i.e., mean change between 0.5 cm and 3 cm between control and ankle vibration conditions, similar changes have been reported between comfortable (1.38 m/s) and slow (1.0 m/s) gait speed in young healthy subjects [27]. These results obtained during walking are consistent with the backward tilt during vibration of the triceps surae in quiet standing [8,28], also reproduced in the present study. This postural effect of vibration is generally interpreted through the kinesthetic illusion of lengthening of the vibrated muscle and the backward compensatory reaction generated to restore body posture [8]. However, a forward body position was also reported during triceps surae vibration during gait [9]. At least two experimental conditions may explain these effects in opposite directions. Firstly, gait pattern adaptation was less constrained in [9]. Participants walked either on a self-paced treadmill which adapted to any change in the participant’s gait speed, or at set speed but with the vibration turned off when the participant reached the front end of the treadmill. Participants were thus able to adapt their behavior in the direction of the illusion, i.e., forward. In the present study,
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
G Model
GAIPOS-4257; No. of Pages 6 Y. Mullie, C. Duclos / Gait & Posture xxx (2014) xxx–xxx
forward progression was limited to the length of the treadmill, which may have caused participants to adapt their behavior to counter the illusion, i.e., lean backward during gait. Secondly, this experiment was carried out in a fully lit room, rather than with dim light as in [9], and may have altered sensory integration (see below). These differences also likely explain why no change was measured in spatio-temporal gait parameters, contrary to previous results [9,12]. In terms of dynamic balance, the stabilizing force decreased when vibration was applied on the triceps surae. This was mainly due to a greater distance between the CoP and the anterior limit of the pBoS, given that no change was observed in the velocity of the CoM. Greater distance between the CoP and the anterior BoS is considered a condition where dynamic balance is easier to maintain as it provides more opportunity to elicit a postural response to stop any forward displacement of the body. The amplitude of change of this distance (roughly 1 cm) has already been observed when comparing comfortable and fast gait speed [18], or during obstacle negotiation [11]. The results suggest that the integration of proprioceptive input depends on the visual environment during gait. When visual information is reliable, it seems that weighting of proprioceptive input is decreased, particularly when conflict appears between visual and proprioceptive information. Neck muscle vibration did not alter the dynamic and postural component of balance or temporal characteristics of gait in healthy participants. These findings are opposite to previously reported increase in gait speed and step frequency, and forward body leaning during neck muscle vibration during gait in a dimly lit room [10]. It is thus possible that when visual information is sufficient (e.g., in a fully lit room) to determine head orientation, neck muscle information is downweighted and its stimulation does not trigger balance adaptation. Reduced integration of proprioceptive information may also explain the absence of effect of ankle muscle vibration in the stroke group. Several studies have suggested that stroke is responsible for reorganization of the sensory integration process [20,29], resulting in increased use of visual compared to proprioceptive input during quiet standing [30]. Our results during walking also support this hypothesis since proprioceptive stimulation on ankle muscles did not alter balance during gait in most stroke participants, contrary to healthy participants. This interpretation is also strengthened by the results obtained during standing with the eyes closed. In these conditions, the postural reactions in stroke participants indicated that they did use the proprioceptive information from the ankle or neck when the eyes were closed. During gait with normal visual information, these postural reactions were not seen, supporting the reduction of the weight of proprioceptive information in balance control during gait after stroke. The main limitation of this study is its small group of stroke participants with fairly good locomotor function. However, it is likely that stroke participants with lower gait performance and motor control would show similar or worse balance performance. We believe the individual analysis strengthened our conclusion, but a larger group of stroke participants would be necessary to confirm the results. Secondly, small changes in balance due to vibration stimulation may not have been detected by our evaluation methods. Nevertheless, changes observed in the stabilizing and destabilizing forces were associated with changes in the positions of the CoM and CoP relative to the limit of the pBoS, showing the sensitivity of the stabilizing and destabilizing forces. Finally, random proprioceptive stimulations (which would cause unpredictable gait perturbations) have not been analyzed, preventing any conclusion to be drawn on the use of proprioceptive information to detect unexpected events during gait. To conclude, proprioception information is used to control balance during gait. Its importance likely depends on the walking
5
and visual conditions. After stroke, the use of proprioception for balance control was affected during gait, but not during standing, pointing toward alteration of sensory integration during gait. Further study is needed to fully understand the integration of proprioceptive and visual information to control balance during gait. Clinicians should be more aware of the deficits in sensorimotor integration in stroke patients and specific evaluation interventions should be developed in this field. Funding This work was supported by the Lindsay Rehabilitation Hospital Foundation. Y. Mullie was supported by a scholarship from Universite´ de Montre´al. Acknowledgements The authors wish to thank the subjects for their participation as well as P. Gourdou and F.B. Loiselle for their technical assistance. Conflict of interest statement The authors have no conflicts of interest to disclose. References [1] Massion J, Alexandrov A, Frolov A. Why and how are posture and movement coordinated? Prog Brain Res 2004;143:13–27. [2] Horak FB. Postural orientation and equilibrium: what do we need to know about neural control of balance to prevent falls? Age Ageing 2006;35:ii7–11. [3] Gurfinkel VS, Ivanenko YP, Levik YS, Babakova IA. Kinesthetic reference for human orthograde posture. Neuroscience 1995;68:229–43. [4] Roll JP, Vedel JP, Ribot E. Alteration of proprioceptive messages induced by tendon vibration in man: a microneurographic study. Exp Brain Res 1989;76: 213–22. [5] Fallon JB, Macefield VG. Vibration sensitivity of human muscle spindles and Golgi tendon organs. Muscle Nerve 2007;36:21–9. [6] Roll JP, Vedel JP. Kinaesthetic role of muscle afferents in man, studied by tendon vibration and microneurography. Exp Brain Res 1982;47:177–90. [7] Forner-Cordero A, Steyvers M, Levin O, Alaerts K, Swinnen SP. Changes in corticomotor excitability following prolonged muscle tendon vibration. Behav Brain Res 2008;190:41–9. [8] Kavounoudias A, Gilhodes JC, Roll R, Roll JP. From balance regulation to body orientation: two goals for muscle proprioceptive information processing? Exp Brain Res 1999;124:80–8. [9] Ivanenko YP, Grasso R, Lacquaniti F. Influence of leg muscle vibration on human walking. J Neurophysiol 2000;84:1737–47. [10] Ivanenko YP, Grasso R, Lacquaniti F. Neck muscle vibration makes walking humans accelerate in the direction of gaze. J Physiol 2000;525:803–14. [11] Sorensen K, Hollands M, Patla A. The effects of human ankle muscle vibration on posture and balance during adaptive locomotion. Exp Brain Res 2002;143:24–34. [12] Courtine GPT, Lucas B, Schieppati M. Continuous, bilateral Achilles’ tendon vibration is not detrimental to human walk. Brain Res Bull 2001;55:107–15. [13] Pearson KG. Generating the walking gait: role of sensory feedback. Prog Brain Res 2004;143:123–9. [14] Knikou M, Kay E, Schmit BD. Parallel facilitatory reflex pathways from the foot and hip to flexors and extensors in the injured human spinal cord. Exp Neurol 2007;206:146–58. [15] Ivanenko YP, Talis VL, Kazennikov OV. Support stability influences postural responses to muscle vibration in humans. Eur J Neurosci 1999;11:647–54. [16] Connell LA, Lincoln NB, Radford KA. Somatosensory impairment after stroke: frequency of different deficits and their recovery. Clin Rehabil 2008;22: 758–67. [17] Niam S, Cheung W, Sullivan PE, Kent S, Gu X. Balance and physical impairments after stroke. Arch Phys Med Rehabil 1999;80:1227–33. [18] Chisholm AE, Perry SD, McIlroy WE. Inter-limb centre of pressure symmetry during gait among stroke survivors. Gait Posture 2011;33:238–43. [19] Kawahira K, Higashihara K, Matsumoto S, Shimodozono M, Etoh S, Tanaka N, et al. New functional vibratory stimulation device for extremities in patients with stroke. Int J Rehabil Res 2004;27:335–7. [20] Lin SI, Hsu LJ, Wang HC. Effects of ankle proprioceptive interference on locomotion after stroke. Arch Phys Med Rehabil 2012;93:1027–33. [21] Marigold DS, Eng JJ, Tokuno CD, Donnelly CA. Contribution of muscle strength and integration of afferent input to postural instability in persons with stroke. Neurorehabil Neural Repair 2004;18:222–9. [22] Oliveira CB, Medeiros IRT, Greters MG, Frota NAF, Lucato LT, Scaff M, et al. Abnormal sensory integration affects balance control in hemiparetic patients within the first year after stroke. Clinics 2011;66:2043–8.
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
G Model
GAIPOS-4257; No. of Pages 6 6
Y. Mullie, C. Duclos / Gait & Posture xxx (2014) xxx–xxx
[23] de Leva P. Adjustments to Zatsiorsky–Seluyanov’s segment inertia parameters. J Biomech 1996;29:1223–30. [24] van Ingen Schenau GJ. Some fundamental aspects of the biomechanics of overground versus treadmill locomotion. Med Sci Sports Exerc 1980;12:257–61. [25] Duclos C, Desjardins P, Nadeau S, Delisle A, Gravel D, Brouwer B, et al. Destabilizing and stabilizing forces to assess equilibrium during everyday activities. J Biomech 2009;42:379–82. [26] Duclos C, Mieville C, Gagnon D, Leclerc C. Dynamic stability requirements during gait and standing exergames on the wii fit system in the elderly. J Neuroeng Rehabil 2012;9:28.
[27] Lugade V, Lin V, Chou LS. Center of mass and base of support interaction during gait. Gait Posture 2011;33:406–11. [28] Ivanenko YP, Solopova IA, Levik YS. The direction of postural instability affects postural reactions to ankle muscle vibration in humans. Neurosci Lett 2000;292:103–6. [29] de Oliveira CB, de Medeiros IRT, Frota NAF, Greters ME, Conforto AB. Balance control in hemiparetic stroke patients: main tools for evaluation. J Rehabil Res Dev 2008;45:1215–26. [30] Bonan I, Derighetti F, Gellez-Leman MC, Bradai N, Yelnik A. Visual dependence after recent stroke. Ann Readapt Med Phys 2006;49:166–71.
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
GAIPOS-4257; No. of Pages 6 Gait & Posture xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals Yannick Mullie a,b, Cyril Duclos a,b,* a Pathokinesiology Laboratory, Centre for Interdisciplinary Research in Rehabilitation (CRIR), Institut de re´adaptation Gingras-Lindsay-de-Montre´al (IRGLM), 6300 avenue Darlington, Montreal, QC H3S 2J4, Canada b School of Rehabilitation, Universite´ de Montreal, 7077 avenue du Parc, Montreal, QC H3N 1X7, Canada
A R T I C L E I N F O
A B S T R A C T
Article history: Received 16 January 2014 Received in revised form 29 May 2014 Accepted 11 July 2014
Proprioceptive information is important for balance control yet little is known about how it is used during gait or how a stroke affects its use. The aim of this study was to evaluate the role of proprioception in controlling balance during gait in healthy participants and after stroke. Twelve healthy and 9 hemiparetic participants walked on an instrumented treadmill in a fully lit room, while whole-body, three-dimensional kinematics were quantified. Vibration was applied continuously or during the stance phase only, on the posterior neck muscles and triceps surae tendon on the non-dominant/paretic side. Difficulty in maintaining dynamic and postural balance was evaluated using stabilizing and destabilizing forces, respectively. Continuous and stance phase vibration of the triceps surae reduced the difficulty in maintaining both dynamic and postural balance in healthy participants (p < .05), with a greater distance between the center of pressure and the limit of the potential base of support, a more backward body position, and no change in spatio-temporal gait parameters. No effect of neck muscle vibration was observed on balance (p = .63 and above). None of the vibration conditions affected balance or gait parameters among stroke participants. The results confirmed that proprioceptive information was not used to control balance during gait in stroke participants. The importance of proprioceptive information may depend on other factors such as walking and visual conditions. Changes in sensory integration ability likely explain the results after stroke. Further study is needed to understand the integration of proprioceptive and visual information to control balance during gait after stroke. ß 2014 Elsevier B.V. All rights reserved.
Keywords: Kinesthesis Equilibrium Locomotion Cerebrovascular accident Human
1. Introduction Control of body posture and dynamics, i.e., segment alignment and velocity of the center of mass (CoM), depends on the integration and weighting of somatosensory, vestibular and visual sensory information [1,2]. Among them, muscle proprioception continuously informs the central nervous system (CNS) on the position of each segment of the body [3]. Muscle vibration is commonly used to study the role of proprioceptive information in motor control as it generates a strong proprioceptive stimulation when applied at a frequency of 80–100 Hz and with amplitude between 0.5 and 1 mm [4–6]. It also induces a contraction of the
* Corresponding author at: Pathokinesiology Laboratory, Centre for Interdisciplinary Research in Rehabilitation (CRIR), Institut de re´adaptation Gingras-Lindsay-de-Montre´al (IRGLM), 6300 avenue Darlington, Montreal, QC H3S 2J4, Canada. E-mail address: [email protected] (C. Duclos).
vibrated muscle or its antagonist depending on the conditions of application [7]. When applied during quiet standing, muscle vibration induces oriented postural reactions [8,9]: vibration of the triceps surae resulted in a backward postural response [8] while vibration to the back neck muscle resulted in a forward postural reaction [8,10]. During gait, the role of proprioception in controlling balance has scarcely been studied. Vibration has been shown to affect kinematics, speed, and muscle activity during gait [9,11,12]. The results most closely related to balance control during gait showed changes in CoM displacements and accelerations during gait when vibration was applied at the ankle [11]. Contrary to quiet standing, sensory activity during gait is cyclical (due to repetitive movements of the limbs) and random in the case of gait perturbations. As such, sensory activity plays different roles in the motor control of gait. For example, the transition between stance and swing phase, and the organization of alternated flexor and extensor activity in the lower limbs is mediated by phasic sensory activity [13]. Moreover, sensory
http://dx.doi.org/10.1016/j.gaitpost.2014.07.008 0966-6362/ß 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
G Model
GAIPOS-4257; No. of Pages 6 Y. Mullie, C. Duclos / Gait & Posture xxx (2014) xxx–xxx
2
background affects responses to new sensory information: during standing, the position of the joints alters gains in cutaneous reflexes [14], and instability of the support surface inhibits postural reactions to vibration of ankle muscles [15]. Thus, sensory afferents may have different effects on balance control during gait depending on the gait phase examined. The use of proprioceptive information to control balance is also affected by neurological lesions such as stroke. Persons with hemiparesis generally have impaired proprioception [16], associated with increased postural sway [17]. Alterations in the center of pressure (CoP) excursion due to sensory manipulation during quiet standing [17] or gait [18] also indicate that stroke-related sensorimotor impairments affect the neuromuscular activity necessary to control balance. However, gait speed has been increased via proprioceptive stimulation applied at the ankle during gait in participants with hemiparesis [19]. Ankle vibration only affected temporal (vs. spatial) gait parameters in stroke patients, regardless of any ankle joint position sense impairment [20], suggesting that the use of proprioception information to control gait might be reduced at the sensorimotor integration level rather than the sensory perception level after stroke [20–22]. To our knowledge, no study has evaluated how proprioceptive information is used to control balance during gait after stroke. The first objective of the study was to evaluate how continuous and phasic proprioceptive information from the neck and ankle alters postural and dynamic balance during gait in healthy participants. A secondary objective was to determine how hemiparesis due to stroke affects the use of proprioceptive information in postural and dynamic balance control during gait.
2. Methods 2.1. Participants A convenience sample of 21 volunteers (12 healthy, 9 with hemiparesis due to stroke) was recruited for this study. The inclusion criteria were to be able to walk on a treadmill without any assistance, have no orthopedic or neurological problems affecting gait, or cognitive deficits prior to the experiment (for healthy participants) or prior to stroke (for hemiparetic participants) and be able to sustain 90 min of activity with rest periods as
required. The main characteristics of the participants are presented in Table 1. All participants gave written consent to participate in the study after having been informed of the details of the experiment according to local ethics board recommendations. 2.2. Data collection and proprioceptive stimulation Three-dimensional whole-body kinematics were recorded at 60 Hz with an Optotrak Certus system (NDI, Waterloo, Canada), using three to six non-collinear infrared markers placed on each main segment of the body (15 segments for a total of 75 markers). A digitizing probe was used to locate the contour of the shoe soles, with respect to the infrared markers on the respective foot segments, and anatomical landmarks to complete the definition of the rigid bodies representing each body segment along with anthropometric measurements and to define a 3-D link-segment model for each participant [23]. Ground reaction forces and moments were measured under each foot using an instrumented split-belt treadmill (Bertec Fit1). Kinetic data were collected at a frequency of 600 Hz, filtered with a fourth-order Butterworth zero-lag filter with a cut-off frequency of 10 Hz and re-sampled at 60 Hz to match the kinematic data. Belt speed was set to the participants’ comfortable gait speed using progressive speed increases and decreases. Proprioceptive stimulation was applied using an electromechanical vibrator (VB115, Technoconcept, France) at 80 Hz (amplitude between 0.5 and 1 mm) on the tendon of the nondominant or paretic triceps surae and on the bilateral posterior neck muscles during gait. Continuous (with the vibrator on throughout the entire trial) and phasic (with the vibrator turned on only when the heel was in contact with the ground as detected by a foot switch placed on the heel of the shoe) modes of vibration were used. Preliminary tests showed delays between vibrator activations and stance phase of about less than 80 ms. 2.3. Experimental protocol The following conditions were tested among all participants: control condition without vibration, posterior neck muscle vibration in the continuous and phasic mode, and triceps surae tendon vibration in the continuous and phasic mode. The control
Table 1 Clinical characteristics of stroke participants. Age (years)
S1 S2 S3 S4 S5 S6 S7 S8 S9 Mean SD or range Mean SD
60 32 64 53 55 54 39 37 36 47.8 11.8 47.6 14.5
BMI (kg/m2)
31.2 21.6 27.5 22.8 27.7 25.7 25.7 37.4 20.6 26.7 5.2 26.4 6.0
OG velocity (m/s)
0.85 0.82 0.80 0.72 0.73 0.55 1.08 1.31 1.22 0.90 0.25 1.55 0.15
Treadmill velocity (m/s)
0.35 0.75 0.60 0.70 0.70 0.55 0.65 0.95 0.60 0.65 0.16 1.02 0.17
Sensory perception
CMMSA Leg score
Foot score
Cut. (/4)
Mov. (/10)
Pall. (s)
5 6 5 4 4 6 5 6 5 5 [4:6]
4 4 3 2 4 4 4 5 3 4 [2:4]
3/3 4/4 3/3 3/3 3/3 2/2 4/3 2/2 4/4 3/3 [2:4/2:4]
10/10 10/10 10/10 10/10 10/10 10/10 10/10 10/10 10/10 10/10 [10/10]
13.2/13.2 14.7/15.4 9.6/10.0 10.7/11.6 17.1/14.0 13.1/12.4 8.0/9.2 14.1/14.8 15.5/16.2 12.9/13.0 2.9/2.4
BMI, body mass index; OG, overground; CMMSA, Chedoke-McMaster Stroke Assessment; SD, standard deviation. Pallesthesia: duration perception with a 128 Hz tuning fork on the malleolus lateralis. Mov.: Number of passive movements at the big toe perceived correctly out of 10. Cut.: cutaneous perception of 2 out of 3 Semmes–Weinstein filament contacts on the malleolus lateralis; 1: anesthesia (6.65 filament perceived); 2: severe deficit (5.18 filament perceived); 3: hypoesthesia (4.31 filament perceived); 4: normal sensitivity (4.17 filament perceived). Paretic/Non-paretic are indicated for these three tests. The last two lines in italics are the mean results for the group of healthy participants.
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
G Model
GAIPOS-4257; No. of Pages 6 Y. Mullie, C. Duclos / Gait & Posture xxx (2014) xxx–xxx
condition was always recorded first. The first site of stimulation (triceps or neck) was alternated between participants. The two modes of vibration (continuous or phasic) were applied in a pseudo-random order for each muscle. One-minute trials were recorded for each condition at a comfortable gait speed on the treadmill. A 90-s familiarization period was allowed with vibration before data collection. Control trials were also performed in quiet standing position, eyes closed. Vibration was tested at each site in one trial, before the corresponding experimental gait trials. The participants were standing on the non-moving treadmill during 30-s trials. After 10 s of quiet standing, the vibrator was turned on for 20 s until the end of the trial. Only the ground reaction forces were recorded. 2.4. Data analysis For the gait trials, kinetic and kinematic data were translated according to the treadmill belt speed to yield forward displacements of the CoM and CoP, equivalent to overground walking [24]. The positions of the CoM, CoP and limits of the potential base of support (pBoS, defined as the vertical ground projection of the contour of both feet for the whole gait cycle) were obtained from the translated kinetic and kinematic data and used to calculate the stabilizing and destabilizing forces [25,26] for each step in the trials. The stabilizing force is the theoretical force needed to stop the motion of the body’s CoM and CoP at the limit of the pBoS in the direction of the displacement of the body (Eq. (1)). Its maximal value indicates the greatest difficulty in dynamic balance, and was calculated. The destabilizing force is the theoretical force necessary to move the body into an unstable position (Eq. (2)). Its minimal value represents the greatest postural difficulty during the task. The distance between the CoP and the limit of the pBoS, the position of the CoM relative to the anteroposterior length of the pBoS, and velocity of the CoM were also calculated to further understand any significant effect of vibration on balance difficulty. These values were averaged over a minimum of 20 steps for each foot. The stabilizing force mglobal :~ vCM :~ vCM ~ ~ F ST ¼ DCP 2D2CP
(1)
vCM : linear velocity of the body center of mglobal: body mass, ~ ~CP : distance between the center of pressure and the limit of mass, D
3
the potential base of support. The destabilizing force
~ FD ¼
! ~ n ~ Fr ~ DCP hCM
(2)
~ n: normal unitary vector to ground F r : ground reaction forces, ~ ~CP : distance between surface, hCM: height of the center of mass, D the center of pressure and the limit of the potential base of support. Spatiotemporal gait parameters were calculated based on the variations of ground reaction forces and kinematics of malleolus markers: step time (i.e., time between two consecutive heel contacts), swing time (i.e., time between toe-off and heel strike of the same leg), step length (i.e., anteroposterior distance between the two malleolus at the time of respective heel contact). For the standing trials, the mean anteroposterior position of the CoP was measured during the first 10 s of the trials, when the vibration was off, and during the last 18 s of the trials, when the vibration was on. For each group, Friedman ANOVAs were used for each variable during gait trials to compare the experimental conditions in each group. Planned contrasts (Wilcoxon ranked test) were applied depending on the results of the ANOVAs to test which vibration condition differed from the control condition without vibration. Wilcoxon ranked tests were also used to compare the position of the CoP before and during vibration in the standing trials, for each vibration site, and each group. For all statistical tests, significance was set at a = .05. The results of the stroke participants during gait trials were also analyzed individually. The percentage of change in stabilizing/ destabilizing force values between the control and experimental conditions was calculated for both groups. For each condition, individual changes among stroke participants were compared to the range of changes measured in the healthy group. Response to vibration was considered similar to the healthy participants when the change observed for each stroke participant was within the healthy participants’ range. 3. Results Data of the gait trials were averaged between dominant and non-dominant steps as Wilcoxon tests did not demonstrate any differences among the healthy or stroke participants (p = .21 and above). In both groups, vibration stimulation made no difference in any of the spatiotemporal characteristics (step length, step and swing time (p = .20 and above) (Table 2).
Table 2 Spatial gait parameters for each experimental condition.
Healthy participants CoM position (%) CoP distance (cm) CoM velocity (m/s) Step time (s) Swing time (s) Step length Stroke participants Paretic step time (s) Non-paretic step time (s) Paretic swing time (s) Non-paretic swing time Paretic step length Non-paretic step length
Control
Triceps continuous
62.19 10.37 1.02 1.12 0.35 0.51
(3.86) (0.03) (0.21) (0.09) (0.03) (0.07)
61.22 11.22 1.01 1.12 0.35 0.51
1.34 1.34 0.42 0.35 0.38 0.42
(0.13) (0.13) (0.06) (0.06) (0.09) (0.09)
1.34 1.34 0.39 0.35 0.38 0.42
(3.57)** (0.03)*** (0.21) (0.12) (0.04) (0.07) (0.14) (0.14) (0.05) (0.09) (0.10) (0.10)
Triceps stance
55.59 11.53 1.02 1.13 0.36 0.51 1.35 1.35 0.40 0.34 0.39 0.42
Neck continuous
Neck stance
(2.48)*** (0.03)*** (0.21) (0.12) (0.05) (0.07)
1.12 (0.10) 0.35 (0.04) 0.51 (0.07)
1.11 (0.09) 0.35 (0.04) 0.51 (0.07)
(0.14) (0.14) (0.06) (0.08) (0.09) (0.09)
1.28 1.28 0.39 0.32 0.38 0.42
1.33 1.32 0.40 0.34 0.38 0.41
(0.08) (0.08) (0.04) (0.09) (0.11) (0.10)
(0.12) (0.11) (0.05) (0.09) (0.10) (0.09)
CoM position is the position of the center mass (CoM) relative to the anteroposterior length of the potential base of support. CoP distance is the distance between the center of pressure (CoP) and the limit of the potential base of support in the direction of the CoM displacement. Mean (standard deviation) are presented. Empty cells indicates that the values were not evaluated in the absence of changes in balance variables. ** p < .01. *** p < .005.
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
G Model
GAIPOS-4257; No. of Pages 6 4
Y. Mullie, C. Duclos / Gait & Posture xxx (2014) xxx–xxx
3.1. Healthy participants
A. Triceps surae vibraon (standing)
Only ankle muscle vibration altered the difficulty in maintaining both postural and dynamic balance, as measured by destabilizing and stabilizing forces, respectively (Friedman ANOVA, x2(4) = 12.2, p < .05, Fig. 1A). The vibration of triceps surae in continuous and phasic modes significantly decreased the stabilizing force (Wilcoxon ranked test, p < .05) and increased the destabilizing force (p < .01). However, continuous and stance phase neck muscle vibration did not alter the stabilizing (p = .63 and p = .81, respectively) or destabilizing force values (p = .38 and p = .27, respectively). In the conditions where balance was altered, vibration application changed the distance between the CoP and the limit of the pBoS (x2(4) = 15.9, p < .005) and the relative position of the CoM (x2(4) = 27.8, p < .001) depending on its location (Table 2). During ankle muscle vibration, the distance between the CoP and the anterior limit of the pBoS increased during both continuous and stance phase vibration (p < .005). In terms of body position over the pBoS, the relative position of the CoM was more backward for stance phase (p < .005) and continuous vibration (p < .01). No significant alteration of these variables was found during neck muscle vibration (p = .07). Neither the location nor the mode of vibration significantly altered the velocity of the CoM (x2(4) = 1.3, p = .23).
0.01
(m)
0.00
10
20
30 (s)
-0.01 -0.02 -0.03 -0.04 -0.05
B. Neck muscle vibraon (standing) 0.12
(m)
0.10 0.08 0.06 0.04
3.2. Stroke participants Ankle and neck muscle vibration did not alter the difficulty in maintaining postural and dynamic balance (x2(4) = 1.8, p = .77 and x2(4) = 4.6, p = .33, respectively, Fig. 1). During both continuous and stance phase ankle vibration conditions, the individual analysis indicated that the changes in stabilizing force for each stroke participant were lower than the range of changes measured among the healthy participants ([2.2; – 19.8%]). Similarly, the changes in destabilizing force during continuous and stance phase ankle vibration were lower in each stroke participant than the range measured in the healthy participants ([3.1; 18.4%]). During neck muscle vibration, only two and three stroke participants showed changes within the range of the non-significant changes observed in stabilizing and destabilizing forces measured for the healthy participants.
0.02 0.00
10
20
30 (s)
-0.02
Fig. 2. Group mean (thick line) standard deviation (thin line) of the position of the center of pressure (CoP) during standing in healthy (black) and stroke participants (gray) during 30-s trials where 80 Hz vibration was applied on the triceps surae (A) or back neck muscles (B) during the last 20 s of the trial (gray bar at the bottom of each line graph).
3.3. Standing trials
400
A. Healthy parcipants (N=12)
120 (N)
(N)
*
350
The mean position of the CoP was more forward during neck muscle vibration than before in both groups (stroke participants: +1.1 (1.0) cm, p < .05; healthy group: +3.5 (3.2) cm, p < .005) (Fig. 2A). The mean position of the CoP was more backward during triceps surae vibration than before in both groups (stroke participants: 1.1 (0.9) cm, p < .01; healthy group: +1.4 (1.0) cm, p < .005) (Fig. 2B).
110
*
100
300
4. Discussion
90 250
80
200
**
***
150 100
70
Stabilizing force (le axis) Destabilizing force (right axis)
Control Gait
Triceps Connuous
Triceps Stance
Neck Connuous
Neck Stance
B. Stroke parcipants (N=9)
300
60 50
150
(N)
(N)
140
250
130 200
120 110
150
100
100
90 50 0
Stabilizing force (le axis) Destabilizing force (right axis)
Control Gait
Triceps Connuous
Triceps Stance
Neck Connuous
Neck Stance
80 70
Fig. 1. Mean and standard deviation of maximal stabilizing force values (gray square, left axis) and minimal destabilizing force values (black square, right axis) during control gait condition (left) and vibration conditions (4 right columns) in healthy (A) and stroke (B) participants. Values are expressed in Newtons (N). * For p < .05, ** for p .01, and *** for p .005 (planned contrasts, comparison with the control gait condition).
Our results showed that ankle proprioceptive stimulation altered both postural and dynamic components of balance in healthy participants and that stroke affects the use of proprioceptive information to maintain balance during gait. The increase in destabilizing force indicated that the difficulty to maintain postural balance was reduced during the vibration of the triceps surae tendon during gait in healthy participants. This was associated with a position of the CoM 1 to 6.5% more posterior within the pBoS. Although of small amplitude, i.e., mean change between 0.5 cm and 3 cm between control and ankle vibration conditions, similar changes have been reported between comfortable (1.38 m/s) and slow (1.0 m/s) gait speed in young healthy subjects [27]. These results obtained during walking are consistent with the backward tilt during vibration of the triceps surae in quiet standing [8,28], also reproduced in the present study. This postural effect of vibration is generally interpreted through the kinesthetic illusion of lengthening of the vibrated muscle and the backward compensatory reaction generated to restore body posture [8]. However, a forward body position was also reported during triceps surae vibration during gait [9]. At least two experimental conditions may explain these effects in opposite directions. Firstly, gait pattern adaptation was less constrained in [9]. Participants walked either on a self-paced treadmill which adapted to any change in the participant’s gait speed, or at set speed but with the vibration turned off when the participant reached the front end of the treadmill. Participants were thus able to adapt their behavior in the direction of the illusion, i.e., forward. In the present study,
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
G Model
GAIPOS-4257; No. of Pages 6 Y. Mullie, C. Duclos / Gait & Posture xxx (2014) xxx–xxx
forward progression was limited to the length of the treadmill, which may have caused participants to adapt their behavior to counter the illusion, i.e., lean backward during gait. Secondly, this experiment was carried out in a fully lit room, rather than with dim light as in [9], and may have altered sensory integration (see below). These differences also likely explain why no change was measured in spatio-temporal gait parameters, contrary to previous results [9,12]. In terms of dynamic balance, the stabilizing force decreased when vibration was applied on the triceps surae. This was mainly due to a greater distance between the CoP and the anterior limit of the pBoS, given that no change was observed in the velocity of the CoM. Greater distance between the CoP and the anterior BoS is considered a condition where dynamic balance is easier to maintain as it provides more opportunity to elicit a postural response to stop any forward displacement of the body. The amplitude of change of this distance (roughly 1 cm) has already been observed when comparing comfortable and fast gait speed [18], or during obstacle negotiation [11]. The results suggest that the integration of proprioceptive input depends on the visual environment during gait. When visual information is reliable, it seems that weighting of proprioceptive input is decreased, particularly when conflict appears between visual and proprioceptive information. Neck muscle vibration did not alter the dynamic and postural component of balance or temporal characteristics of gait in healthy participants. These findings are opposite to previously reported increase in gait speed and step frequency, and forward body leaning during neck muscle vibration during gait in a dimly lit room [10]. It is thus possible that when visual information is sufficient (e.g., in a fully lit room) to determine head orientation, neck muscle information is downweighted and its stimulation does not trigger balance adaptation. Reduced integration of proprioceptive information may also explain the absence of effect of ankle muscle vibration in the stroke group. Several studies have suggested that stroke is responsible for reorganization of the sensory integration process [20,29], resulting in increased use of visual compared to proprioceptive input during quiet standing [30]. Our results during walking also support this hypothesis since proprioceptive stimulation on ankle muscles did not alter balance during gait in most stroke participants, contrary to healthy participants. This interpretation is also strengthened by the results obtained during standing with the eyes closed. In these conditions, the postural reactions in stroke participants indicated that they did use the proprioceptive information from the ankle or neck when the eyes were closed. During gait with normal visual information, these postural reactions were not seen, supporting the reduction of the weight of proprioceptive information in balance control during gait after stroke. The main limitation of this study is its small group of stroke participants with fairly good locomotor function. However, it is likely that stroke participants with lower gait performance and motor control would show similar or worse balance performance. We believe the individual analysis strengthened our conclusion, but a larger group of stroke participants would be necessary to confirm the results. Secondly, small changes in balance due to vibration stimulation may not have been detected by our evaluation methods. Nevertheless, changes observed in the stabilizing and destabilizing forces were associated with changes in the positions of the CoM and CoP relative to the limit of the pBoS, showing the sensitivity of the stabilizing and destabilizing forces. Finally, random proprioceptive stimulations (which would cause unpredictable gait perturbations) have not been analyzed, preventing any conclusion to be drawn on the use of proprioceptive information to detect unexpected events during gait. To conclude, proprioception information is used to control balance during gait. Its importance likely depends on the walking
5
and visual conditions. After stroke, the use of proprioception for balance control was affected during gait, but not during standing, pointing toward alteration of sensory integration during gait. Further study is needed to fully understand the integration of proprioceptive and visual information to control balance during gait. Clinicians should be more aware of the deficits in sensorimotor integration in stroke patients and specific evaluation interventions should be developed in this field. Funding This work was supported by the Lindsay Rehabilitation Hospital Foundation. Y. Mullie was supported by a scholarship from Universite´ de Montre´al. Acknowledgements The authors wish to thank the subjects for their participation as well as P. Gourdou and F.B. Loiselle for their technical assistance. Conflict of interest statement The authors have no conflicts of interest to disclose. References [1] Massion J, Alexandrov A, Frolov A. Why and how are posture and movement coordinated? Prog Brain Res 2004;143:13–27. [2] Horak FB. Postural orientation and equilibrium: what do we need to know about neural control of balance to prevent falls? Age Ageing 2006;35:ii7–11. [3] Gurfinkel VS, Ivanenko YP, Levik YS, Babakova IA. Kinesthetic reference for human orthograde posture. Neuroscience 1995;68:229–43. [4] Roll JP, Vedel JP, Ribot E. Alteration of proprioceptive messages induced by tendon vibration in man: a microneurographic study. Exp Brain Res 1989;76: 213–22. [5] Fallon JB, Macefield VG. Vibration sensitivity of human muscle spindles and Golgi tendon organs. Muscle Nerve 2007;36:21–9. [6] Roll JP, Vedel JP. Kinaesthetic role of muscle afferents in man, studied by tendon vibration and microneurography. Exp Brain Res 1982;47:177–90. [7] Forner-Cordero A, Steyvers M, Levin O, Alaerts K, Swinnen SP. Changes in corticomotor excitability following prolonged muscle tendon vibration. Behav Brain Res 2008;190:41–9. [8] Kavounoudias A, Gilhodes JC, Roll R, Roll JP. From balance regulation to body orientation: two goals for muscle proprioceptive information processing? Exp Brain Res 1999;124:80–8. [9] Ivanenko YP, Grasso R, Lacquaniti F. Influence of leg muscle vibration on human walking. J Neurophysiol 2000;84:1737–47. [10] Ivanenko YP, Grasso R, Lacquaniti F. Neck muscle vibration makes walking humans accelerate in the direction of gaze. J Physiol 2000;525:803–14. [11] Sorensen K, Hollands M, Patla A. The effects of human ankle muscle vibration on posture and balance during adaptive locomotion. Exp Brain Res 2002;143:24–34. [12] Courtine GPT, Lucas B, Schieppati M. Continuous, bilateral Achilles’ tendon vibration is not detrimental to human walk. Brain Res Bull 2001;55:107–15. [13] Pearson KG. Generating the walking gait: role of sensory feedback. Prog Brain Res 2004;143:123–9. [14] Knikou M, Kay E, Schmit BD. Parallel facilitatory reflex pathways from the foot and hip to flexors and extensors in the injured human spinal cord. Exp Neurol 2007;206:146–58. [15] Ivanenko YP, Talis VL, Kazennikov OV. Support stability influences postural responses to muscle vibration in humans. Eur J Neurosci 1999;11:647–54. [16] Connell LA, Lincoln NB, Radford KA. Somatosensory impairment after stroke: frequency of different deficits and their recovery. Clin Rehabil 2008;22: 758–67. [17] Niam S, Cheung W, Sullivan PE, Kent S, Gu X. Balance and physical impairments after stroke. Arch Phys Med Rehabil 1999;80:1227–33. [18] Chisholm AE, Perry SD, McIlroy WE. Inter-limb centre of pressure symmetry during gait among stroke survivors. Gait Posture 2011;33:238–43. [19] Kawahira K, Higashihara K, Matsumoto S, Shimodozono M, Etoh S, Tanaka N, et al. New functional vibratory stimulation device for extremities in patients with stroke. Int J Rehabil Res 2004;27:335–7. [20] Lin SI, Hsu LJ, Wang HC. Effects of ankle proprioceptive interference on locomotion after stroke. Arch Phys Med Rehabil 2012;93:1027–33. [21] Marigold DS, Eng JJ, Tokuno CD, Donnelly CA. Contribution of muscle strength and integration of afferent input to postural instability in persons with stroke. Neurorehabil Neural Repair 2004;18:222–9. [22] Oliveira CB, Medeiros IRT, Greters MG, Frota NAF, Lucato LT, Scaff M, et al. Abnormal sensory integration affects balance control in hemiparetic patients within the first year after stroke. Clinics 2011;66:2043–8.
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
G Model
GAIPOS-4257; No. of Pages 6 6
Y. Mullie, C. Duclos / Gait & Posture xxx (2014) xxx–xxx
[23] de Leva P. Adjustments to Zatsiorsky–Seluyanov’s segment inertia parameters. J Biomech 1996;29:1223–30. [24] van Ingen Schenau GJ. Some fundamental aspects of the biomechanics of overground versus treadmill locomotion. Med Sci Sports Exerc 1980;12:257–61. [25] Duclos C, Desjardins P, Nadeau S, Delisle A, Gravel D, Brouwer B, et al. Destabilizing and stabilizing forces to assess equilibrium during everyday activities. J Biomech 2009;42:379–82. [26] Duclos C, Mieville C, Gagnon D, Leclerc C. Dynamic stability requirements during gait and standing exergames on the wii fit system in the elderly. J Neuroeng Rehabil 2012;9:28.
[27] Lugade V, Lin V, Chou LS. Center of mass and base of support interaction during gait. Gait Posture 2011;33:406–11. [28] Ivanenko YP, Solopova IA, Levik YS. The direction of postural instability affects postural reactions to ankle muscle vibration in humans. Neurosci Lett 2000;292:103–6. [29] de Oliveira CB, de Medeiros IRT, Frota NAF, Greters ME, Conforto AB. Balance control in hemiparetic stroke patients: main tools for evaluation. J Rehabil Res Dev 2008;45:1215–26. [30] Bonan I, Derighetti F, Gellez-Leman MC, Bradai N, Yelnik A. Visual dependence after recent stroke. Ann Readapt Med Phys 2006;49:166–71.
Please cite this article in press as: Mullie Y, Duclos C. Role of proprioceptive information to control balance during gait in healthy and hemiparetic individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.07.008
