Accepted Manuscript

Muscle activity of leg muscles during unipedal stance on therapy devices with
different stability properties

Thomas Wolburg, M.D., Walter Rapp, Ph.D., Dr., Jochen Rieger, Thomas Horstmann,
M.D.

PII: S1466-853X(15)00037-1
DOI: 10.1016/j.ptsp.2015.05.001
Reference: YPTSP 665

To appear in: Physical Therapy in Sport

Received Date: 1 December 2014

Revised Date: 15 April 2015
Accepted Date: 6 May 2015

Please cite this article as: Wolburg, T., Rapp, W., Rieger, J., Horstmann, T., Muscle activity of leg
muscles during unipedal stance on therapy devices with different stability properties, Physical Therapy in
Sports (2015), doi: 10.1016/j.ptsp.2015.05.001.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Leg muscle activity during unipedal stance on therapy devices

with different stability properties

PT
Thomas Wolburg, M.D.1
Walter Rapp, Ph.D.2,3

RI
Jochen Rieger3
Thomas Horstmann, M.D.4,5

SC
1
Kantonsspital Baselland, Bruderholz, Switzerland
2
Institute for Sport and Sport Sciences, Albert-Ludwigs University Freiburg, Germany
3
Department of Sports Medicine, University Hospital Tübingen, Germany

U
4
Medical Park Bad Wiessee St. Hubertus, Bad Wiessee, Germany
5
Faculty for Sport and Health Sciences, Technische Universität München, Munich,
AN
Germany
M

January 2015
D

Original Research
TE

Revised manuscript for: Physical Therapy in Sport

EP

Ms. Ref. No.: PTIS-14179

C

Corresponding author: Dr. Walter Rapp

AC

Department of Sport and Sport Science

Albert-Ludwigs University Freiburg
Schwarzwaldstrasse 175
79117 Freiburg
Germany
Tel.: +49 761 / 2034551
E-Mail: walter.rapp@sport.uni-freiburg.de

Word count: 2,729 words

1
 ACCEPTED MANUSCRIPT

Muscle activity of leg muscles during unipedal stance on

therapy devices with different stability properties

PT
RI
U SC
AN
April 2015
M

Original Research
D

Revised manuscript for: Physical Therapy in Sport

TE

Ms. Ref. No.: PTIS-14179

C EP
AC

Word count: 2,773 words

1
 ACCEPTED MANUSCRIPT
ABSTRACT

Objectives. To test the hypotheses that less stable therapy devices require greater

muscle activity and that lower leg muscles will have greater increases in muscle activity

with less stable therapy devices than upper leg muscles.

Design. Cross-sectional laboratory study.

PT
Setting. Laboratory setting.

RI
Participants. Twenty-five healthy subjects.

Main Outcome Measures. Electromyographic activity of four lower (gastrocnemius

SC
medialis, soleus, tibialis anterior, peroneus longus) and four upper leg muscles (vastus

medialis and lateralis, biceps femoris, semitendinosus) during unipedal quiet barefoot

U
stance on the dominant leg on a flat rigid surface and on five therapy devices with
AN
varying stability properties.
M

Results. Muscle activity during unipedal stance differed significantly between therapy

devices (P<0.001). The order from lowest to highest relative muscle activity matched
D

the order from most to least stable therapy device. There was no significant interaction
TE

between muscle location (lower versus upper leg) and therapy device (P=0.985).

Magnitudes of additional relative muscle activity for the respective therapy devices
EP

differed substantially among lower extremity muscles.

Conclusions. The therapy devices offer a progressive increase in training intensity, and
C
AC

thus may be useful for incremental training programs in physiotherapeutic practice and

sports training programs.

2
 ACCEPTED MANUSCRIPT
KEYWORDS

Balance training, muscle activity, lower extremity, stability properties

PT
RI
U SC
AN
M
D
TE
C EP
AC

3
 ACCEPTED MANUSCRIPT
INTRODUCTION

Proprioceptive and sensorimotor training therapies are successful in treating sport-

related disorders and injuries that are linked to deficits in proprioceptive regulation and

muscular imbalances (Ergen & Ulkar, 2008; Jerosch, Pfaff, Thorwesten, & Schoppe,

1998; van Ochten, van Middelkoop, Meuffels, & Bierma-Zeinstra, 2014). Frequently,

PT
proprioceptive training programs utilize therapy devices with varying stability

RI
properties: less stable therapy devices pose a greater challenge on the neuromuscular

system (Hupperets, Verhagen, & van Mechelen, 2009). Therapy devices include

SC
unstable mats, spinning tops, balls, tilt platforms, and oscillating devices. It is generally

accepted that the training stimulus in training sessions should be gradually increased

U
(Holm, et al., 2004; Hupperets, Verhagen, & van Mechelen, 2008). In therapeutic
AN
practice, this is usually achieved by combining different motor tasks such as balancing
M

while catching a ball. This method requires greater cognitive attention and consequently

higher neuromuscular activation. Frequently, the tasks are not performed

D

simultaneously but rather successively, and it remains unclear if this method will
TE

impose a greater stimulus to specific regions such as the ankle joint complex. Therefore,

increasing a controlled stimulus to the affected anatomical structure is desirable and

EP

might be achieved by increasing the difficulty of the balancing task by using successive

therapy devices with decreasing stability properties.

C
AC

Two main mechanisms contribute to stabilizing the joints of the lower extremities and

thus protect against injury. First, anatomical congruity of the joint facilitates passive

mechanical stability and is especially effective at high axial loads. This congruent

system is supported and guided by the capsule-ligament system (Hintermann, 1996;

Scheuffelen, Rapp, Gollhofer, & Lohrer, 1993). Second, neuromuscular activation and

4
 ACCEPTED MANUSCRIPT
the mechanical stiffness of associated tendons enable active joint stability. Specifically,

coordinated neuromuscular activation of antagonistic and agonistic muscles stabilizes

the ankle joint both voluntarily and involuntarily through reflex mechanisms

(Solomonow & Krogsgaard, 2001). In addition, postural control strategies are

characterized primarily by muscle activation patterns and body kinematics, and early

PT
activation of dorsal ankle muscles followed by activation of dorsal thigh and trunk

RI
muscles during perturbations are indicators for an ankle strategy (Horak & Nashner,

1986; Wang, Molenaar, Challis, Jordan, & Newell, 2014).

SC
Despite of the common use of therapy devices, to date information on their effects on

U
muscular coordination is scarce. Van Ooteghem et al. (Van Ooteghem, et al., 2008)
AN
described that young participants improved their balance control in response to
M

continuous, variable amplitude motion of a translating platform by shifting from an

ankle strategy toward a multi-segmental control strategy. A similar change in balance

D

control strategy as been reported for task with versus without visual information (Wang,
TE

et al., 2014). While those results imply that ankle muscles may be involved in

maintaining balance while standing on an unstable surface, these conclusions were

EP

based on kinematic data only. Greater insight into neuromuscular effects of balance

training with therapy devices is necessary for improving the efficacy of training
C
AC

interventions aimed at treating and preventing sport-related disorders and ankle injuries.

The objective of this study was to quantify the effects of therapy devices with different

stability properties on muscle activity of lower extremity muscles during unipedal

stance. We hypothesized that less stable therapy devices require greater muscle activity

5
 ACCEPTED MANUSCRIPT
and that lower leg muscles will have greater increases in muscle activity with less stable

therapy devices than upper leg muscles.

METHODS

A convenience sample of 25 healthy subjects (22 men, 3 women; mean ± 1SD; age:

PT
25.2 ± 4.5 years; body mass: 75.3 ± 9.9 kg; height: 178.4 ± 6.9 cm) participated in this

RI
study after providing informed consent. All subjects were active in different sport

activities between 1 and 4 hours per week. Exclusion criteria of this study were:

SC
cerebral or neurological conditions or balance difficulties, joint related functional

limitations, muscle related functional limitations of the dominant leg, physiotherapy or

U
surgery of the dominant leg in the preceding 12 months and post-exercise fatigue on the
AN
day of testing. This study was approved by the University’s ethics review board and
M

conducted in accordance with the Declaration of Helsinki.

D

Experimental procedure
TE

Subjects were asked to perform unipedal quiet stance trials on their dominant leg. For

all trials, subjects were barefoot with their eyes open and arms folded behind their back.
EP

Subjects balanced for 15 seconds on each of six surfaces: a flat rigid surface and five

surfaces with varying stability properties, respectively (Figure 1). The therapy devices
C
AC

with different stability properties were a Therapy Top (TT, Thieme Sport, Grasleben,

Germany), three different deformable balance pads (Thera Band stability trainer™,

Hygenic Corporation, Akron, OH) and an Airex mat (Airex AG, Sins, Switzerland). The

different stability properties were achieved by different material properties and

structural design (Table 1).

6
 ACCEPTED MANUSCRIPT

PT
Figure 1. The five therapy devices from most stable to least stable.

RI
Table 1. Characteristics of the balancing surfaces. Different stability properties of the

therapy devices were achieved by different material properties and surface structures.

SC
Dimensions
Therapy Material Stability
(length×width×height; Surface structure
device property property

U
diameter)
Control 50.0cm×41.0cm×3.0cm hard smooth very stable
AN
Therapy Top Ø=40cm hard concentric grooves slightly unstable
Thera Band stability trainera
green 36.5cm×20.0cm×4.5cm hard horizontal grooves slightly unstable
blue 40.5cm×23.0cm×5.0cm soft horizontal grooves unstable
M

black 44.0cm×25.5cm×6.3cmb very soft horizontal grooves very unstable

Airex mat 50.0cm×41.0cm×6.0cm soft smooth unstable
a
elliptic shape; bmeasured at the edge of the device.
D

All subjects first balanced on the flat rigid surface. This condition served as baseline
TE

measurement (control condition). Data for three trials with 10-second breaks between
EP

trials were collected for the control condition. To minimize potential systematic fatigue

effects, the therapy devices with different stability characteristics were tested in
C

randomized order immediately after the control condition by choosing consecutive

AC

orders from a randomization table. Data for three trials with 10-second breaks between

trials were collected for each of the device conditions.

Data recording

Muscle activity was recorded using surface electromyography (EMG). Bipolar surface

electrodes were placed according to the proposals of the SENIAM project group

7
 ACCEPTED MANUSCRIPT
(Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles;

www.seniam.org) on the vastus medialis, vastus lateralis, biceps femoris,

semitendinosus, gastrocnemius, soleus, tibialis anterior and peroneus longus muscles.

The recording area was shaved, and fine-grained sand paper was used to remove the top

layer of skin to reduce skin impedance. The electrodes (Ambu Blue Sensor, Ambu A/S,

PT
Ballerup, Denmark Denmark) were fixed over the muscle body along the fiber direction

RI
with a 2-cm inter-electrode distance. EMG signals were recorded using the MyoSystem

2000™ (Noraxon U.S.A. Inc., Scottsdale, AZ) with 1000 Hz recording frequency. We

SC
confirmed that there was no crosstalk between muscles by computing cross-correlations

between raw signals. Signals were preprocessed with a 10 Hz low pass and a 500 Hz

U
high pass filter. Further signal processing was performed in MyoResearch XP™
AN
(Version 1.06.05; Noraxon U.S.A. Inc., Scottsdale, AZ). All EMG signals were
M

rectified, and the mean EMG amplitude over the 15-sec interval was calculated. Values

for each trial were normalized to the average value of the control condition and
D

expressed as percent control condition. Ensemble averages of the three trials for each
TE

condition were used for further analysis.

EP

Statistical analysis

All statistical tests were performed in SPSS Statistics version 19.0.0 (IBM Corporation,
C
AC

Somers, NY). Data are presented as means and 95% confidence intervals. Multivariate

analysis of variance (MANOVA) was used to detect an overall difference in muscle

activation between the test surfaces. One-way analyses of variance (ANOVAs) were

used to detect differences in muscle activity between the test surfaces for each muscle.

A mixed factor ANOVA was used to detect differences in lower versus upper leg

muscle activity between the test surfaces. The significance level was set a priori to

8
 ACCEPTED MANUSCRIPT
α=0.05. Paired t-tests were used for posthoc analyses with Bonferroni adjustment

(α=0.005).

RESULTS

Muscle activity for all muscles was higher for all therapy devices than for the control

PT
condition, except that of the peroneus longus muscle for the Therapy top and the green

RI
Thera Band stability trainer (Figure 2). However, the increase in muscle activity during

unipedal balancing differed significantly between therapy devices (Table 2). All

SC
muscles required the highest muscle activity for the black Thera Band stability trainer

(P<0.001 compared to all other devices; Figure 2). The greatest amounts of additional

U
relative muscle activity were observed for the tibialis anterior muscle (Figure 2). The
AN
amount of additional relative muscle activity required corresponded mostly to the
M

stability properties of the therapy devices: for most muscles, the order from smallest to

highest additional relative muscle activity matched the order from least to most stable
D

training device (Figure 2). However, for upper leg muscles, the smallest increases in
TE

relative muscle activity were observed for the green Thera Band stability trainer.
EP

Table 2. Results of the one-way analysis of variance (ANOVA) testing for statistically

significant differences in muscle activity among therapy devices for each muscle.
C
AC

Location Muscle Degrees of F-value P-value

freedom
Gastrocnemius medialis 5 4.575 <0.001
Lower

Soleus 5 4.585 <0.001

leg

Tibialis anterior 5 9.291 <0.001

Peroneus longus 5 15.654 <0.001
Vastus medialis 5 7.270 <0.001
Upper

Vastus lateralis 5 7.979 <0.001

leg

Biceps femoris 5 3.550 <0.001

Semitendinosus 5 13.859 <0.001

9
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP
AC

Figure 2. Mean (± 95% confidence interval, 95% CI) difference in muscle activity

between the five therapy devices (in the order from most to least stable) and the control

condition for lower (top) and upper (bottom) leg muscles. Horizontal bars indicate

statistically significant differences (paired t-tests, P<0.05).

10
 ACCEPTED MANUSCRIPT
The overall pattern of additional relative muscle activity for the five therapy devices

was similar for all measured lower extremity muscles. There was no significant

interaction between muscle location (lower versus upper leg) and therapy device

(P=0.99; Table 3). However, while the magnitude of additional relative muscle activity

for the respective therapy devices were similar among upper leg muscles, these

PT
magnitudes differed substantially among lower extremity muscles with the smallest

RI
increases for the soleus muscle and the largest increases for the antagonistic tibialis

anterior muscle (Figure 2).

SC
Table 3. Results of the two-way analysis of variance (ANOVA) with interaction (factors

U
muscle location and therapy device) testing for statistically significant differences in
AN
muscle activity between muscle location and therapy device.
M

Factor Degrees of F-value P-value

freedom
Muscle location 2 11.247 0.91
D

Therapy device 5 0.012 <0.001

Muscle location × therapy device 30 0.088 0.99
TE
EP

DISCUSSION

We found that the less stable therapeutic therapy devices required more additional
C

muscle activity than the more stable therapy devices confirming our first hypothesis. To
AC

date, there are no other reports of differences in muscle activity while balancing on

different therapy devices and hence the observed differences in muscle activity cannot

be related to the literature. Contrary to our second hypothesis, the overall magnitude of

increases in relative muscle activity with the therapy devices were similar for upper and

lower leg muscles. However, the increases in relative muscle activity with increasing

instability were similar among upper leg muscles but differed between lower leg

11
 ACCEPTED MANUSCRIPT
muscles. Additional relative muscle activity of antagonistic muscles suggest that

subjects have greater joint stiffness in the hip and knee joint when balancing on

progressively unstable therapy devices. Moreover, additional relative muscle activity

when balancing on unstable therapy devices represents a greater demand on these

muscles, and hence this type of training will likely improve upper leg muscle strength.

PT
Heitkamp et al. (Heitkamp, Horstmann, Mayer, Weller, & Dickhuth, 2001) reported that

RI
balance training alone can increase knee flexor and extensor muscle strength. While the

effects of balancing on therapy devices on muscle strength were not measured in this

SC
study, we would expect these effects to be strongest with the least stable device because

of the correlation between muscle activation with force output (Alkner, Tesch, & Berg,

2000).
U
AN
M

Comparing the amount of additional relative muscle activity for the more stable therapy

devices (Therapy Top and green Thera Band stability trainer) between upper and lower
D

leg muscles suggests that the task of balance control seems to move from lower
TE

extremity muscles to the more proximal muscles. The increase in relative muscle

activity in the upper leg muscles for the Therapy Top was comparable to that for the
EP

Airex mat and the blue Thera Band stability trainer. Hence, the Therapy Top seems to

provide an adequate stimulus for the thigh muscles, which may result from the fact that
C
AC

a hard surface requires a relatively rigid tarsal and metatarsal joint. Gruber et al.

(Gruber, Bruhn, & Gollhofer, 2006) found that with fixed ankles the control of stability

tasks moves from the distal shank to the proximal thigh muscle groups: in their study,

activity in upper leg muscles was significantly higher when balancing on the

Posturomed with a fixed ankle compared with a normal freely moveable ankle joint.

12
 ACCEPTED MANUSCRIPT
This shift in control strategy is further supported by the differences in muscle activity

between lower and upper leg muscles for the different therapy devices suggesting that

subjects may use an ankle strategy to maintain their balance when standing on less

stable therapy devices. Electrodes were not removed between conditions and hence

methodological factors such as motor point, the length of muscle fibers or connective

PT
tissue remained constant and did not contribute to the observed differences. Relative

RI
muscle activity in the gastrocnemius medialis, soleus and peroneus muscles were

increased by less than 15% with the Therapy Top and the green Thera Band stability

SC
trainer compared to the control surface suggesting that balancing on these two surfaces

does not pose a great challenge on muscles controlling the ankle joint. These results

U
may explain why Dias et al. (Dias, Pezarat-Correia, Esteves, & Fernandes, 2011) did
AN
not find differences in peroneus longus activity between a training (balance training on
M

a Therapy Top) and a control group in their four-week training study. Hence, only the

unstable therapy devices (blue Thera Band stability trainer, Airex mat and black Thera
D

band stability trainer) achieve an adequate stimulus to the neuromuscular system

TE

controlling the ankle joint.

EP

Interestingly, the tibialis anterior muscle showed the greatest increases in relative

muscle activity when balancing on the therapy devices. Poldzielny (Podzielny, 2000)
C
AC

reported that the two antagonistic muscles tibialis anterior and peroneus longus tightly

interact during balance control tasks on therapy devices including the Airex mat.

Specifically, they stated that peroneus muscle strength can be improved by training on a

balance device which is particularly relevant in person with peroneus muscle weakness.

However, our results suggest that only training on less stable therapy devices require

additional peroneus activity and hence may be effective. Moreover, according to Horak

13
 ACCEPTED MANUSCRIPT
and Nashner (Horak & Nashner, 1986) early activation of dorsal ankle muscles

followed by activation of dorsal thigh and trunk muscles during perturbations are

indicators for an ankle strategy. Hence, the large increase in tibialis anterior muscle

activity and no or small increases in peroneus longus muscle activity suggest that

subjects indeed adopted an ankle strategy without stiffening the ankle joint. Further,

PT
these results showed that unstable elastic materials have a significant influence on the

RI
activation of individual muscles needed to achieve posture control. Unstable therapy

devices could therefore be used for rehabilitation training of the leg and in particular for

SC
stabilizing the subtalar joint.

U
In this cross-sectional study, healthy subjects were asked to balance on five different
AN
therapy devices. While we found distinct differences in additional relative muscle
M

activity for lower extremity muscles between therapy devices, it is unknown if the

differences in muscle activity between therapy devices persist over a prolonged period
D

of time and the benefits of training interventions using these therapy devices in patients
TE

remain unclear. In addition, kinematic data would be necessary to confirm which

control strategies subjects employ when balancing on more or less stable therapy
EP

devices or surfaces. Because we measured muscle activity for the entire balancing task

and did not resolve the electromyographic signals in the time domain or into individual
C
AC

bursts, we were unable to determine if the additional muscle activity can be attributed to

reflex mechanisms (Solomonow & Krogsgaard, 2001). Previous research (Clark &

Burden, 2005) has shown that timing of muscle activity is a critical aspect of balance

control that can be affected by training on less unstable therapy devices such as the

Therapy Top in persons with unstable ankles. Nevertheless, we found that the largest

increases in relative muscle activity occurred when balancing on the black Thera Band

14
 ACCEPTED MANUSCRIPT
stability trainer. Hence, this training device appears to provide adequate stimuli during

sensorimotor training, which is a common component of rehabilitation programs after

lower leg injuries. All tested therapy devices can be used for home training, additional

training sessions supplementing guided programs, or injury prevention.

PT
Conclusion

RI
The tested therapy devices offer a gradual increase in training intensity reflected by

proportional increases in neuromuscular activation, and thus may be useful for

SC
proprioceptive training programs not only in physiotherapeutic practice but also in

sports training programs. These results for healthy subjects should be confirmed for

U
patients with musculoskeletal or neurological conditions and diseases.
AN
M
D
TE
C EP
AC

15
 ACCEPTED MANUSCRIPT
REFERENCES

Alkner, B. A., Tesch, P. A., & Berg, H. E. (2000). Quadriceps EMG/force relationship

in knee extension and leg press. Medicine and science in sports and exercise, 32,

459-463.

Clark, V. M., & Burden, A. M. (2005). A 4-week wobble board exercise programme

PT
improved muscle onset latency and perceived stability in individuals with a

RI
functionally unstable ankle. Phys Ther Sport, 6, 181-187.

Dias, A., Pezarat-Correia, P., Esteves, J., & Fernandes, O. (2011). The influence of a

SC
balance training program on the electromyographic latency of the ankle

musculature in subjects with no history of ankle injury. Phys Ther Sport, 12, 87-

92.
U
AN
Ergen, E., & Ulkar, B. (2008). Proprioception and ankle injuries in soccer. Clinics in
M

sports medicine, 27, 195-217.

Gruber, M., Bruhn, S., & Gollhofer, A. (2006). Specific adaptations of neuromuscular
D

control and knee joint stiffness following sensorimotor training. International

TE

journal of sports medicine, 27, 636-641.

Heitkamp, H. C., Horstmann, T., Mayer, F., Weller, J., & Dickhuth, H. H. (2001). Gain
EP

in strength and muscular balance after balance training. International journal of

sports medicine, 22, 285-290.

C
AC

Hintermann, B. (1996). [Biomechanics of the ligaments of the unstable ankle joint].

Sportverletzung Sportschaden : Organ der Gesellschaft fur Orthopadisch-

Traumatologische Sportmedizin, 10, 48-54.

16
 ACCEPTED MANUSCRIPT
Holm, I., Fosdahl, M. A., Friis, A., Risberg, M. A., Myklebust, G., & Steen, H. (2004).

Effect of neuromuscular training on proprioception, balance, muscle strength,

and lower limb function in female team handball players. Clinical journal of

sport medicine : official journal of the Canadian Academy of Sport Medicine,

14, 88-94.

PT
Horak, F. B., & Nashner, L. M. (1986). Central programming of postural movements:

RI
adaptation to altered support-surface configurations. Journal of

neurophysiology, 55, 1369-1381.

SC
Hupperets, M. D., Verhagen, E. A., & van Mechelen, W. (2008). The 2BFit study: is an

unsupervised proprioceptive balance board training programme, given in

U
addition to usual care, effective in preventing ankle sprain recurrences? Design
AN
of a randomized controlled trial. BMC musculoskeletal disorders, 9, 71.
M

Hupperets, M. D., Verhagen, E. A., & van Mechelen, W. (2009). Effect of unsupervised

home based proprioceptive training on recurrences of ankle sprain: randomised

D

controlled trial. BMJ, 339, b2684.

TE

Jerosch, J., Pfaff, G., Thorwesten, L., & Schoppe, R. (1998). [Effects of a

proprioceptive training program on sensorimotor capacities of the lower

EP

extremity in patients with anterior cruciate ligament instability]. Sportverletzung

Sportschaden : Organ der Gesellschaft fur Orthopadisch-Traumatologische

C
AC

Sportmedizin, 12, 121-130.

Podzielny, S. (2000). [Biomechanical and neuromuscular mechanisms of training

programs for preventing ankle injuries]. Unpublished PhD Dissertation,

University of Stuttgart, Stuttgart.

17
 ACCEPTED MANUSCRIPT
Scheuffelen, C., Rapp, W., Gollhofer, A., & Lohrer, H. (1993). Orthotic devices in

functional treatment of ankle sprain. Stabilizing effects during real movements.

International journal of sports medicine, 14, 140-149.

Solomonow, M., & Krogsgaard, M. (2001). Sensorimotor control of knee stability. A

review. Scandinavian journal of medicine & science in sports, 11, 64-80.

PT
van Ochten, J. M., van Middelkoop, M., Meuffels, D., & Bierma-Zeinstra, S. M. (2014).

RI
Chronic complaints after ankle sprains: a systematic review on effectiveness of

treatments. The Journal of orthopaedic and sports physical therapy, 44, 862-

SC
C823.

Van Ooteghem, K., Frank, J. S., Allard, F., Buchanan, J. J., Oates, A. R., & Horak, F. B.

U
(2008). Compensatory postural adaptations during continuous, variable
AN
amplitude perturbations reveal generalized rather than sequence-specific
M

learning. Experimental brain research. Experimentelle Hirnforschung.

Experimentation cerebrale, 187, 603-611.

D

Wang, Z., Molenaar, P. C., Challis, J. H., Jordan, K., & Newell, K. M. (2014). Visual
TE

information and multi-joint coordination patterns in one-leg stance. Gait &

posture, 39, 909-914.

C EP
AC

18
 ACCEPTED MANUSCRIPT
FIGURE LEGENDS

Figure 1. The five therapy devices from most stable to least stable.

Figure 2. Mean (± 95% confidence interval, 95% CI) difference in muscle activity

between the five therapy devices (in the order from most to least stable) and the control

PT
condition for lower (top) and upper (bottom) leg muscles. Horizontal bars indicate

RI
statistically significant differences (paired t-tests, P<0.05).

U SC
AN
M
D
TE
C EP
AC

19
 ACCEPTED MANUSCRIPT
ACKNOWLEDGEMENTS

The authors thank PD Dr. Annegret Mündermann for her writing assistance on

behalf of the authors.

PT
RI
U SC
AN
M
D
TE
C EP
AC

1
 ACCEPTED MANUSCRIPT
HIGHLIGHTS

• We measured upper and lower leg muscle activity during unipedal balancing

tasks.

• We compare balance training devices with different stability and surface

PT
structure.

RI
Relative muscle activity increases with decreasing training device stability.

• Lower extremity muscles are affected differently by training device stability.

SC
• Training intensity can be gradually increased with different training devices.

U
AN
M
D
TE
C EP
AC