Abstract
The purpose of this study was to clarify the effect of changing footwear conditions on postural response against postural perturbation. Twenty-three healthy subjects participated in this study. Postural response was induced by moving a platform forward, hereafter referred to as forward-perturbation of a platform. The center of pressure (COP) from the force plate and the electromyograms (EMG) of the tibialis anterior (TA) and quadriceps femoris (QUAD), which are both agonists of the response, were measured. The effect of plantar material and shape of footwear on postural response was examined as footwear condition. Changing plantar materials had an effect on integrated EMG of the agonists (IEMG) but not on the response pattern. On the other hand, the shape of footwear had an effect on the response pattern but not on IEMG. It was supposed from this result that changes in somatosensory input, caused by coupling of plantar material and shape of footwear, modifies postural response variously.
Keywords: automatic postural response, footwear types, electromyogram
Falling due to unstable standing balance is considered to be one factor resulting in hip bone fractures, especially in elderly people, so dynamic standing balance of humans is increasingly being given more attention. The study of dynamic standing balance originated in Nashner's scheme in which postural response was examined1) against postural sway induced by a movable platform. Nashner's group found that two comprehensive muscle activation patterns are characteristic to this scheme. When a subject stands on a normal platform and postural disturbance is elicited by horizontal perturbation of the platform, muscle activity begins in the ankle joint muscles and radiates in distal-to-proximal sequence on the same dorsal or ventral aspect. This pattern is termed ankle strategy. When a subject stands on a platform which is short in relation to foot length, the trunk and thigh muscles that are antagonistic to those used in ankle strategy are activated in the opposite proximal-to-distal sequence. This pattern is termed hip strategy. These responses are termed automatic postural response2)3).
Our group has been studying the effect of varying footwear types on dynamic postural response with the practical view of preventing falls among elderly people. In our previous study, we used the EquiTest device. It measures strength, which shows amplitude of the postural response or torque around the ankle joint, as well as latency, which is a period of time from the onset of platform perturbation to the onset of foot pressure. Both strength and latency were examined as parameters. It was confirmed that varying footwear types and materials influenced not only the strength but also the latency of postural response4)5). That is to say, the amplitude of postural response when barefoot was smaller than when wearing slippers with shock-absorbing function. And, the latency when barefoot was shorter than when wearing slippers with shock-absorbing function4)5).
In considering the result of our previous study, the following questions were posed: whether or not the data of a force plate and electromyogram (EMG) data both show the change of latency of postural response.
The purpose of this study was to compare EMG data with force plate data to determine the influence of footwear types on postural response. Our hypothesis was that, as shown in force plate data, EMG latency and amplitude would change according to adaptation and changing footwear types.
Subjects
Twenty-three healthy university students (age: 22.5 ± 3.0) participated in this study.
Method
The subjects stood with their eyes closed and their arms relaxed at their sides on a movable platform (our original portable perturbater; SAKAI Medical Co., Ltd.). A backward body sway was induced by a sudden forward horizontal perturbation of the platform. The velocity and the amplitude of the perturbations were held consistent in all trials (208 cm/sec, 5.0 cm).
Protocol
In order to clarify the effect of changing the footwear types on automatic postural responses, measurements were taken in 18 trials for each subject according to the following protocol.
First of all, three trials were measured for each subject standing barefoot on the platform as practice trials. Second, measurements were conducted under four footwear conditions (Fig. 1): barefoot, plantar material of slippers, slippers with clog thongs, and slippers without clog thongs. Each foot wear was made of same plantar material made of vinyl chloride, which had shock-absorbing function. The height of each footwear was 25 mm. The difference among plantar material of slippers, slippers without clog thongs and slippers with clog thongs was only the existence of cover around the toe and clog thongs. Three trials for each condition were measured at random through 12 trials to avoid the effect of adaptation. A total 12 trials were conducted. Lastly, three barefoot trials were measured again.
Fig. 1.
Footwear conditions.
(1) bare feet, (2) plantar material of slippers, (3) slippers without clog thongs, (4) slippers with clog thongs.
Prior to each perturbation, subjects looked at a fixed mark in front of them and closed their eyes to make their angle of the neck joint consistent. Center of pressure (COP) from the force plate on the platform was also confirmed to be consistent on the monitor and each trial was measured. Subjects were not informed of the onset of perturbation of the platform and the platform began to move suddenly. Each trial was measured following an intermission of one minute throughout the 18 trials.
Automatic postural response caused by forward-perturbation of a platform induces muscle activities of the lower legs symmetrically1). So, surface electromyogram (EMG) recordings from rectus femoris (QUAD) and tibialis anterior (TA) of the left leg, which are agonists of the response, were measured. Each subject was fitted with surface EMG electrodes (blue sensor disposable electrodes; MEDICOTEST Co., Ltd.) spaced 2 cm apart on each muscle. Impedance of the skin was adjusted to be lower than 5 kΩ. Raw electromyograms were rectified with an amplifier (MEG-2100, NIHON KOHDEN CO., Ltd.), and filtered by a low-pass filter at 40 Hz. The excursion of center of pressure (COP) was also measured using the forceplate on the platform without filtration. All EMG and COP data were sampled at 1000 Hz with a MacLab system (MacLab/8s, BIORESEARCH CENTER Co., Ltd.) and recorded on the hard disk of a computer for later analysis.
Data Analysis
Data were analyzed using wave analyzing software (Chart v3.5.6/s ; BIORESEARCH CENTER Co., Ltd.).
COP: From COP data, fore-aft components (COPY) were analyzed. COPY moved backward at first following perturbation of the platform and returned forward. Two parameters of COPY were adopted—the maximum backward COPY excursion as well as COPY latency which refers to the time from the onset of platform perturbation to arrival to the maximum backward COPY excursion.
EMG: From EMG data, EMG latency and integrated EMG (IEMG) were measured. EMG latency was defined as the earliest time when EMG activity exceeded the baseline level by 2 standard deviations for 200 msec. prior to the onset of platform perturbation. The period of IEMG was 100 msec. from the onset of muscle activation of the agonists.
The data were not normally distributed, so Wilcoxon's rank sum test was used to clarify the effects of footwear condition (alpha=0.05).
There are two factors which effect on stability of footwear condition. One factor is the plantar condition and another is the shape of footwear. In order to clarify the effect of the plantar condition on automatic postural responses, the barefoot condition and the condition of plantar material of slippers, hereafter referred to as plantar material were compared. On the other hand, in order to clarify the effect of the shape of footwear on postural responses, the conditions of plantar material, slippers without clog thongs, and slippers with clog thongs were compared. The mean values of EMG and COP parameters of three trials of each condition were measured.
Result
All subjects showed EMG activation of TA following perturbation, as well as signs of QUAD following TA activation. And, as described above, COPY moved backward at first following the perturbation and returned forward (Fig. 2).
Fig. 2.
A typical example of postural response induced by platform perturbation. Amplitude of platform perturbation, muscle response of TA and QUAD, and excursion of COPY.
The effect of changing plantar conditions
COPY: Maximum COPY backward excursion of plantar material was significantly larger than that of bare feet (p<0.05). COPY latency of plantar material was also significantly longer than that of bare feet (p<0.01) (Table1).
Table 1. COP data with plantar material and barefoot.
| plantar material | barefoot | |
|---|---|---|
| COPY latency (sec) | 0.249 ± 0.024 | 0.247 ± 0.020** |
| COPY (cm) | −1.067± 0.234 | −1.024± 0.223* |
p<0.05,
p<0.01.
EMG responses: IEMG of TA with plantar material was significantly larger than that of bare feet (p<0.05). There was no significant difference in IEMG of QUAD between plantar material and bare feet. As for EMG latency of TA and QUAD, there was no significant difference between plantar material and bare feet. The term between onset of TA and QUAD response, which is a parameter for timing postural response, didn't show a significant difference between plantar material and bare feet (Table 2).
Table 2. EMG data with plantar material and barefoot.
| plantar material | barefoot | |
|---|---|---|
| TA latency (sec) | 0.102 ± 0.007 | 0.103 ± 0.008 |
| QUAD latency (sec) | 0.136 ± 0.021 | 0.139 ± 0.025 |
| T-Q period (sec) | 0.031 ± 0.019 | 0.036 ± 0.023 |
| TA IEMG (× 10−3 Vsec) | 0.024 ± 0.010 | 0.022 ± 0.008* |
| QUAD IEMG (× 10−3 Vsec) | 0.004 ± 0.003 | 0.004 ± 0.002 |
p<0.05.
From EMG and COP data, it was shown that the amplitude of postural response was greater in the unstable foot condition of plantar material than the stable barefoot condition. Although it was confirmed that postural adjustment was induced faster in the barefoot condition than in plantar material, in the data of COPY latency, EMG data showed no difference as to latency and timing of postural response.
Effects of changing shape of footwear
COPY: Maximum COPY backward excursion of slippers without clog thongs was significantly larger than that of slippers with clog thongs (p<0.05) and that of plantar material (p<0.01). There was no significant difference in COPY latency among the three footwear conditions (Table 3).
Table 3. COP data with plantar material, slippers without clog thongs and slippers with clog thongs.
 |
p<0.05.
EMG response: There was no significant difference in IEMG of TA and QUAD among the three footwear conditions. Though there was no significant difference in EMG latency of TA among the three conditions, EMG latency of QUAD of slippers with clog thongs was significantly longer than that of slippers without clog thongs (p<0.05) (Table 4).
Table 4. EMG data with plantar material, slippers without clog thongs and slippers with clog thongs.
 |
p<0.05,
p<0.001.
Discussion
Effect of plantar conditions
Inglis et al. reported the importance of somatosensory information from the legs in triggering centrally organized postural synergies6). This led us to the hypothesis that delayed onset, when wearing slippers with shock-absorbing function in our previous studies, was due to a change in the triggering function5)6). We had expected a delayed EMG onset in this study on plantar material.
In COP data, postural sway reduced and latency shortened significantly when the plantar condition was changed from plantar material to bare feet, which was similar to the results of our previous studies. In EMG data, IEMG of TA became significantly smaller when changed to bare feet, which shows postural response was carried out without excessive muscle activities. However, contrary to our hypothesis, EMG latency of both TA and QUAD showed no difference between plantar material and bare feet. The period between onset of TA and QUAD (T-Q period) also showed no difference.
This result means that plantar condition had an effect on amplitude, but not the latency and timing of muscle activity. These results suggest that the changes in postural response caused by plantar conditions were due to mechanical effects of different plantar materials rather than to triggering the function of response.
Shock-absorbing function decreases integral force and delays the conduction of maximum force from one object to the other. When postural response occurs on plantar material, the ankle sinks into the plantar material, disturbing the fixation of plantar foot to the ground. Decreased fixation delays the recovery of the shin yield by TA around the ankle. It was guessed that the mechanical effect of the shock-absorbing function between the plantar foot and the platform caused postural sway to increase, and the conduction of recovery yield by muscle activity of the agonists to be delayed and increased7)8).
Hansen et al. report that the startle-like response was always accompanied by changes in a kind of perturbation of platform9). In our procedure, subjects changed footwear at random, so the characteristics of perturbation were thought to be slightly changed in each trial, which may have caused startle-like responses. Even if the EMG latency of the last three trials in which a startle-like response was inhibited was compared with that of three trials in bare feet in the middle of the 12 trials, there were no significant differences in latencies of either TA or QUAD. So, the probability of the effect of the startle-like response is low.
Horak et al. report that instability of the plantar foot changes the postural synergy2)10). In Horak's report, instability is due to standing on a platform that is short in relation to foot length, whose characteristics are different from the instability of our experiments. In addition, the timing of muscle activity in our study was not affected, so the dynamic change in the postural synergy in this study was thought to be weak. However, we found that the activity of the proximal agonist (QUAD) did not show an increase proportional to the distal agonist (TA) according to the instability of plantar condition. It was guessed that this insufficient activity of the proximal agonist interferes with the center of gravity of proximal body units returning forward, and makes postural response inefficient.
The effect of the shape of footwear types
In three footwear conditions with identical plantar materials, different shapes changed somatosensory input and mechanical support, inducing postural stability. In this study, it was confirmed from the COPY data that, in spite of the advantage of the mechanical support of a cover around the toe, postural disturbance of slippers without clog thongs was larger than that of plantar material. On the other hand, contrary to the result concerning plantar material, a change in IEMG was not confirmed among the three footwear types. This means changes in footwear type don't have an effect on amplitude of muscle activity. This contradictory result suggests muscle activity other than that of the agonists TA and QUAD, participates in regulation of postural response. Ihara describes that there is a high density of mechanoreceptors around the big toe, whose somatosensory input stimulates postural response. Ihara also makes much of the grasping reflex of the toe during postural response11). Certainly, a tight cover around the big toe prevents toes from being flexed. It was supposed that as one factor of increased postural disturbance, changes in somatosensory input caused by the cover around the toe inhibited this grasping reflex.
Though there was no difference in EMG latency and T-Q period when plantar condition was changed, changes in the latency of QUAD and T-Q period occurred when footwear shape was changed. That is to say, both the latency of QUAD and T-Q period without clog thongs was significantly shorter than of that with clog thongs. This result also means that changes in footwear shape effect the triggering function and the regulation of response pattern.
From a clinical point of view, it was confirmed that postural response is facilitated more by barefeet than with plantar material, and by clog thongs rather than a cover around the big toe. These results suggest what kind of footwear should be used to prevent elderly people from falling in facilities for the aged.
In this study, it was suggested that the shock-absorbing function of plantar material has an effect on regulating amplitude of muscle activity of the agonists but not on regulating response pattern. However, shape of footwear has an effect on regulating response pattern but not on regulating amplitude of muscle activity.
References
- 1). Nashner LM, et al. : Fixed patterns of rapid postural responses among leg muscles during stance. Exp Brain Res 30: 13-24, 1977. [DOI] [PubMed] [Google Scholar]
- 2). Horak FB, et al. : Central program of postural movements : Adaptation to alerted support-surface configurations. J Neurophysiology 55: 1369-1381, 1986. [DOI] [PubMed] [Google Scholar]
- 3). Horak FB, et al. : Influence of central set on human postural responses. J Neurophysiology 62: 841, 1989. [DOI] [PubMed] [Google Scholar]
- 4). Hosoda M, et al. : The effect of various footwear types and materials, and of fixing of the ankles by footwear, on upright posture control. J Phys Ther Sci 9: 47-51, 1997. [Google Scholar]
- 5). Hosoda M, et al. : The effects of footwear on standing posture control. J Phys Ther Sci 10: 47-51, 1998. [Google Scholar]
- 6). Inglis JT, et al. : The importance of somatosensory information in triggering and scaling automatic postural responses in humans. Exp Brain Res 101: 159-164, 1994. [DOI] [PubMed] [Google Scholar]
- 7). Shiba N, et al. : Shock-absorbing effect of shoe insert materials commonly used in management of lower extremity disorders. Clin-Orthop 310: 130-136, 1995. [PubMed] [Google Scholar]
- 8). Sanfilippo PB, 2d, et al. : Dynamic plantar pressure analysis. Comparing common insole materials. J-Am-Podiatr-Med-Assoc 82: 507-513, 1992. [DOI] [PubMed] [Google Scholar]
- 9). Hansen PD, et al. : Postural responses to changing task conditions. Exp Brain Res 73: 626-636, 1988. [DOI] [PubMed] [Google Scholar]
- 10). Horak BF, et al. : Cerebella control of postural scaling and central set in stance. J Neurophysiology 72: 479-493, 1994. [DOI] [PubMed] [Google Scholar]
- 11). Ihara H: The Importance of Plantar Function. Joint Trainning, Kyoudoishyo: Tokyo, 1996, pp 89-107. [Google Scholar]