Search tips
Search criteria 


Logo of jathtrainLink to Publisher's site
J Athl Train. 2002 Jul-Sep; 37(3): 269–274.
PMCID: PMC164355

The Spectral Qualities of Postural Control are Unaffected by 4 Days of Ankle-Brace Application


Objective: To determine if the spectral qualities of medial-lateral (ML) and anterior-posterior (AP) center of pressure during a 1-legged stance are affected by 4 days of ankle-brace application.

Design and Setting: The study, which consisted of a pretest-posttest randomized group design, took place in the Sports Injury Research Laboratory at Indiana State University.

Subjects: Twenty-eight Indiana State University students, who had not suffered from any ankle injuries within the past 2 years and were free of any neurologic or vestibular disorders, participated in the study.

Measurements: The effects of 3 independent variables on mean frequency amplitude for both ML and AP center of pressure were examined: (1) treatment (brace, control), (2) frequency bin (0%–20%, 20%–40%, 40%–60%, 60%–80%, and 80%–100% of area), and (3) time (pretest, immediately after brace application, and after 1, 2, 3, or 4 days of brace wear).

Results: We detected no difference (P > .05) for the ML or AP mean frequency when comparing the brace and control groups.

Conclusions: Application of an ankle brace may not require modifications in the postural-control strategies during a 1-legged stance in subjects with healthy ankles.

Keywords: fast Fourier transformation, somatosensation, frequency analysis, balance

In an attempt to prevent recurring ankle sprains, many athletes have resorted to using various external support devices to protect the ankle joint. The rationale is that potentially injurious inversion loads placed on the ankle-foot complex can be resisted with the use of an external ankle support. Such devices include adhesive tape; ready-made, lace-up stabilizers; and semirigid orthoses. The objective of these support systems is to provide mechanical restriction of undesirable ankle joint motion and to relieve ankle joint ligaments of excessive strain while allowing for minimal hindrance to normal joint mechanics.1

Most of the literature regarding ankle bracing has concentrated on understanding bracing's effects on range of motion28 and motor performance.710 More recently, the potential effects of ankle support on the maintenance of postural control have been investigated.1113 The majority of the studies1113 assessing postural control in subjects wearing an external ankle support have been performed after one brace application. Since most athletes who wear ankle braces use them over the course of a season, further investigation is needed to determine if wearing an ankle brace more than once interferes with the proprioceptive system that assists in maintaining postural control.

Two mechanisms by which an ankle brace may act to influence postural control have been proposed: mechanical restriction and increased somatosensation. As previously stated, the main purpose of the ankle brace is to mechanically restrict undesirable motion, primarily plantar flexion and inversion, occurring at the ankle-foot complex. By restricting range of motion at this complex, the postural-control strategy normally used to maintain balance may be altered. Therefore, any differences shown in postural control when wearing external ankle support may reflect deviations in the restrictive properties of such support. In addition to providing mechanical restriction, an ankle brace provides an added tactile stimulus sensed by cutaneous receptors at the ankle-foot complex. It has been suggested12 that the somatosensory cues provided by external support devices may enhance postural control. Therefore, ankle bracing may improve postural control while potentially reducing injury.12

Postural-control testing is often thought to be a direct measure of proprioception; however, this is not the case. The maintenance of upright posture is controlled by the integration of sensory information provided by the visual, vestibular, and somatosensory systems.1416 All 3 of the sensory systems work together to maintain equilibrium. Therefore, performing a balance task challenges the visual, vestibular, and proprioceptive systems collectively. Clinicians and researchers have attempted to focus on the proprioceptive system by introducing somatosensory manipulations (ie, change in support surface).11,17 Although these attempts may challenge the proprioceptive system, the other 2 sensory systems may compensate for the somatosensory change.

Center of pressure (COP) is the center of the distribution of the total force applied to the supporting surface.18 By examining the frequency (spectral) characteristics of COP data, it is possible to “tease out” the 3 sensory systems working to maintain postural equilibrium. It is thought that each of the 3 sensory systems operates within a specific frequency bandwidth.19 Theoretically, when one of these systems is altered, a change in the frequency spectrum is detected within the specified system's operating band. For example, application of an ankle brace introduces an additional somatosensory input, which may influence the proprioceptive system. An alteration in the frequency spectrum after application of the ankle brace would theoretically be due to the added somatosensory input. An increase in the amplitude within the proprioceptive frequency range would represent increased work by the proprioceptive system to maintain posture, whereas a decreased amplitude value would suggest decreased work by the proprioceptive system.

The analysis of the spectral characteristics associated with postural control with the influence of ankle bracing has not been extensively examined. In order for us to gain a thorough understanding of ankle bracing's potential effects on the proprioceptive control of posture, additional examination is warranted. Therefore, our main purpose was to investigate the effects of 4 days of ankle-brace use on the mean frequency amplitude of medial-lateral (ML) and anterior-posterior (AP) COP during a 1-legged stance.



Twenty-eight college students (age = 22 ± 1.8 years, mass = 72.3 ± 12.0 kg, height = 172.6 ± 9.0 cm) volunteered to participate in this study. No subjects had incurred any ankle injuries within the last 2 years, and all were free from any neurologic or vestibular conditions that interfered with their ability to maintain upright stance. Furthermore, subjects had not suffered any head injury resulting in unconsciousness. Any subject reporting prior ankle-brace use was eliminated. Before testing and after the purpose of the study had been explained, all subjects gave written informed consent. The protocol was reviewed and approved by the School of Health and Human Performance Human Subjects Committee at Indiana State University.


We used a strain-gauge force-platform system (Accusway; Advanced Mechanical Technology, Inc, Watertown, MA) to measure postural control. The platform measures 3 translational forces (Fx, Fy, and Fz) and 3 moments of force (Mx, My, and Mz). The data were digitally converted at 50 Hz and interfaced to a controlling laptop computer (model 1250, Compaq Computer Corp, Wilmington, OH). The SWAYWIN software (Advanced Mechanical Technology, Inc) was used to generate the X and Y coordinates representing the COP in the medial-lateral and anterior-posterior directions, respectively. This software was also used to convert the time-domain data into frequency-domain data with the fast Fourier transformation (FFT) technique.


Before testing began, each subject was randomly assigned to either the experimental (brace) or control (no-brace) group. All subjects were asked the preferred leg for kicking a ball in order to determine leg dominance; the leg established as dominant was then used for testing.

A pretest measurement for both groups was recorded without application of an ankle brace. For this and all subsequent measurements, subjects were instructed to step onto the force platform and assume a 1-legged stance. This required the subjects to stand on the test leg with their eyes closed and hands fixed against the body (ie, hands on the iliac crests) (Figure (Figure1).1). Furthermore, during each testing session, all subjects wore opaque goggles to prevent any input from the visual system. The maintenance of upright stance is influenced by the visual system as well as the vestibular and proprioceptive systems. Vision was minimized so that we were able to focus on potential changes due to input from the proprioceptive system. Data sampling was initiated after the subject was properly positioned on the force plate.

Figure 1
One-legged stance.

Additionally, subjects were asked to stand quietly and as motionless as possible in the stance position. They were instructed to keep their hands fixed and not to touch down with the nonstance leg. If subjects began to lose their balance, a quick tap of both elbows by the investigator (R.M.P.) was allowed for each subject to regain control. This touch lasted no longer than 3 seconds.

If this had to be repeated more than twice in a 20-second trial, the trial was considered unsuccessful and was redone. Also, if the subject touched down with the nonstance leg or was unable to sustain the stance position for 20 seconds, the trial was considered incomplete and was redone. Subjects were required to successfully maintain the stance position for 20 seconds per trial for a total of 5 trials, with a 30-second rest period between trials.

After the pretest measurement, all subjects were properly fitted with a lace-up ankle brace (model A101, McDavid Knee Guard Inc, Chicago, IL), according to the manufacturer's specifications, including applying the brace over the sock. Five additional trials were then performed after application of the ankle brace. The instructions given to each subject during pretesting did not change throughout the course of study. However, for the remainder of the study, all subjects wore an ankle brace during testing.

After all 5 trials, the subjects in the control group removed the brace, while the subjects in the brace group continued to wear the ankle brace. After 8 consecutive hours of ankle-brace wear, subjects in the experimental group removed the brace. The next day, subjects in both the brace and control groups again reported to the laboratory for testing. Both groups placed the ankle brace on the test leg and stepped on the force platform for the data to be collected. The procedure used to gather the data on the previous day was followed for each of the subsequent testing sessions. The subjects reported for testing on 5 consecutive days. The time between testing sessions was between 24 and 32 hours. Subjects in the experimental group wore the ankle brace for a total of 32 hours over the 4 days. This duration of time was chosen because we felt it closely represented the usage of this device during a week of preseason training completed by a competitive athlete. Subjects in the control group wore the brace during testing for approximately 1 hour and 15 minutes.

Data Reduction

The trials performed by each subject occurred over a 20-second time period. The FFT was run on the time-domain data in order to convert it to its respective frequency domain. Within a single 20-second trial, ML and AP COP data points were collected at intervals of 0.02 seconds (50 Hz) so that 501 data points each from the ML and the AP COP were recorded for stance.

The FFT algorithm analyzes frequencies up to one half the frequency at which the data were acquired; in this case, the data were collected at 50 Hz, and the highest frequency the FFT detected was 25 Hz. The FFT spectrum was condensed to 256 data points from 501 as described above. The software used to run the FFT on our data omits the first data point. This data point represents the direct current component, and its magnitude reduces the overall resolution of the FFT, which is why it was removed. The FFT spectrum analyzed in this study consisted of a total of 255 points. Once the FFT was performed, the 255 data points were imported into Excel spreadsheets (version 9.0, Microsoft Corp, Redmond, WA) so that the frequency bins could be created. Five frequency bins were developed based on pilot data. Each bin represented 20% of the total area under the FFT curve (Figure (Figure2).2). Frequency bins were created separately for ML and AP COP data.

Figure 2
A, Fast Fourier transformation of medial-lateral center-of-pressure data recorded during a 1-legged stance. This depicts why we chose not to make divisions in equal 5-Hz bins, as shown. More than 80% of the signal would have lain within the first bin, ...

The total area of the frequency spectrum was calculated for each subject and each trial for both ML and AP COP. The area was determined by averaging the 255 amplitude values and then multiplying that value by 25. Twenty-five represents the maximum frequency at which data may have been acquired.

Once the area was calculated, each of the 255 amplitude values was multiplied by 0.02 (the time interval at which data points were collected). This quantity for each value was summed to all prior amplitude values (ie, for amplitude value 232, the sum of all amplitudes from 1 to 232 was taken) and then divided by the total area of the spectrum so that we could determine the percentage of the total area that each data point represents. The data points were reviewed and separated into the frequency bins (each representing 20% of the area under the FFT curve) previously created from the pilot data. The mean amplitude in each bin was used for analyses.


A 2 × 5 × 6 repeated-measures analysis of variance was used to assess differences among conditions, bins, and time and the interaction of brace, time, and bin on ML and AP mean frequency amplitudes. Simple main-effects testing and Tukey multiple comparison procedures were used post hoc to locate specific group differences. The probability was set at P ≤ .05 for all tests.


Mean amplitude values for the ML and AP COP for bins 1 through 5 across each time are displayed in Tables Tables11 and and2,2, respectively. The mean amplitude values for ML and AP COP for each treatment at each time across all bins are displayed in Tables Tables33 and and44.

Table thumbnail
Table 1. Medial-Lateral Mean Amplitude Values for Bins 1 through 5 at Time 0 to Time 5 (cm ± SD)
Table thumbnail
Table 2. Anterior-Posterior Mean Amplitude Values for Bins 1 through 5 at Time 0 to Time 5 (cm ± SD)
Table thumbnail
Table 3. Medial-Lateral Mean Amplitude Values for Brace (B) and Control (C) Conditions at Each Time for Bins 1 through 5
Table thumbnail
Table 4. Anterior-Posterior Mean Amplitude Values for Brace (B) and Control (C) Conditions at Each Time for Bins 1 through 5

The ML (F4,104 = 10.16, P = .081, 1-β = .524, η2 = .075), and AP (F4,104 = 2.26, P = .067, 1-β = .644, η2 = .080) mean amplitudes within the individual bins did not differ between the brace and control groups. This suggests that ankle-brace application did not significantly interfere with the proprioceptive control of posture during a 1-legged stance.

The mean ML (F4,108 = 495.8, P < .0005) and AP (F4,108 = 79.8, P < .0005) amplitudes were different among bins when collapsed over condition (brace and control). Mean ML and AP amplitudes were greater at lower frequency ranges than at higher frequency ranges at all 6 times (P < .05).


Inspection of the components of the postural-control system may help to explain why no changes were seen in the frequency content of ML and AP COP over time with bracing. It is well known that 3 sensory systems (visual, vestibular, and proprioceptive) integrate and synthesize information to control posture. These sensory systems provide afferent input to the central nervous system (CNS). Once the CNS receives these afferent impulses, it organizes the new information and produces an efferent response. This efferent response is what allows the appropriate postural adjustments to be made in order to maintain an upright stance.

Proprioceptive input from the ankle is received through various mechanoreceptors, including Golgi tendon organs, muscle spindles, and cutaneous receptors.20 The McDavid ankle brace used in this investigation may not have provided sufficient input to influence the proprioceptors at the ankle-foot complex and, therefore, would not be expected to alter the efferent signal produced by the CNS. If this is the case, then we would not anticipate changes in the mean frequency amplitude of ML or AP COP.

Previous work supports our results indicating that ankle-brace application does not interfere with the proprioceptive control of posture.11 The effects of 3 selected ankle appliances on postural control under different variations of a modified Romberg test were examined. An increase in COP patterns in the ML and AP directions with the eyes open was noted. When the sensory modalities were challenged, the COP values remained unchanged, which showed that bracing had no effect. These results also suggest that ankle bracing did not interfere with the coherence of the 3 sensory systems that manipulate and control posture.

An argument could also be made that no changes were seen in the ML and AP mean frequency amplitudes because our subject population was uninjured. Placing additional support on an ankle-foot complex that does not need to rely on assistance in order to maintain stability may not alter postural control. However, athletes suffering from chronic ankle instability, which is thought to arise from proprioceptive deficits, may be able to use the additional sensory input provided by the McDavid ankle brace, resulting in alteration of the frequency components of COP.

Before we completed this research, the effect of 4 days of ankle-brace application on one's ability to maintain an upright stance had not been described in the literature. Most athletes use ankle braces over an extended time period in an effort to prevent ankle injury. The finding that ankle-brace application over a 4-day time period did not affect postural control is of great clinical significance. Individuals with decreased postural control are believed to be more susceptible to ankle injury than those with finer postural control.20,21 If the application of an ankle brace decreased one's ability to maintain an upright stance, the purpose of this device would be minimized. The fact that we did not see a decrease in postural control suggests continuous use of a McDavid ankle brace may not adversely affect sensory integration.

The only significant change we found existed within the individual bins at each time. The ML and AP mean frequency amplitudes between 0 and 0.39 Hz were greater than at 0.40 to 25 Hz. The ML and AP mean frequency amplitudes at 0.40 Hz to 0.88 Hz were greater than at 0.89 to 25 Hz and less than the ML and AP mean frequency amplitudes at 0 to 0.39 Hz, etc. This pattern of greater ML and AP mean frequency amplitudes at lower frequencies, decreasing as the frequency increases, is applicable at each time. This finding suggests that performing a 1-legged stance is a low-frequency activity.

A potential criticism of this study is that our statistical analysis revealed relatively low power values (ML 1-β = .524, AP 1-β = .644) when comparing the mean amplitudes of COP contained within each bin between treatment conditions. Therefore, it is possible that with additional subjects, significance would have been achieved, suggesting that the ankle brace did interfere with the proprioceptive control of posture. However, when taking into account the small effect sizes (ML η2 = .075, AP η2 = .080) accompanying the low power values, we question whether a statistically significant result in this case would be clinically significant.

The results of this study may not be applicable to a more dynamic or athletic movement (ie, a cutting maneuver). Data for this study were collected while subjects performed a 1-legged stance, which does not produce the same type of result as completing a more vigorous movement. Perhaps placing an external support on an ankle during athletic competition alters the proprioceptive control of posture.


We do not know if a change did not occur because the application of an ankle brace did not require a modification to be made in the efferent response sent by the central nervous system. The fact that no changes were seen could also indicate that uninjured ankles did not need the added support and, therefore, did not use the added sensory input. A study examining the effects of bracing on the spectral characteristics of postural sway on injured ankles may provide more insight on this issue. A change might have been seen if the study had been carried out over a longer period of time. Future researchers should investigate how ankle-brace application over a longer time period affects the spectral qualities of postural sway in order to more closely mimic actual usage of this device by athletes. Until further investigation is performed, the decision to use an ankle brace based upon its influence on proprioception may not be warranted.


  • Garrick J G, Requa R K. Epidemiology of foot and ankle injuries in sports. Clin Sports Med. 1988;7:29–36. [PubMed]
  • Alves J W, Alday R B, Ketcham D L, Lentell G L. A comparison of the passive support provided by various ankle braces. J Orthop Sports Phys Ther. 1992;15:10–18. [PubMed]
  • De Clercq D L. Ankle bracing in running: the effect of a Push type medium ankle brace upon movement of the foot and ankle during the stance phase. Int J Sports Med. 1997;18:222–228. [PubMed]
  • Gehlsen G M, Pearson D, Bahamonde R. Ankle joint strength, total work, and ROM: comparison between prophylactic devices. J Athl Train. 1991;26:62–65.
  • Cordova M L, Ingersoll C D, LeBlanc M J. Influence of ankle support on joint range of motion before and after exercise: a meta-analysis. J Orthop Sports Phys Ther. 2000;30:170–182. [PubMed]
  • Paris D L, Vassilios V G, Kokkaliaris J. Ankle ranges of motion during extended activity periods while taped and braced. J Athl Train. 1995;30:223–228. [PMC free article] [PubMed]
  • Wiley J P, Nigg B M. The effect of an ankle orthosis on ankle range of motion and performance. J Orthop Sports Phys Ther. 1996;23:362–369. [PubMed]
  • Gross M T, Batten A M, Lamm A L, et al. Comparison of DonJoy ankle ligament protector and subtalar sling ankle taping in restricting foot and ankle motion before and after exercise. J Orthop Sports Phys Ther. 1994;19:33–41. [PubMed]
  • Verbrugge J D. The effects of semirigid Air-Stirrup bracing vs. adhesive ankle taping on motor performance. J Orthop Sports Phys Ther. 1996;23:320–325. [PubMed]
  • Beriau M R, Cox W B, Manning J. Effects of ankle braces upon agility course performance in high school athletes. J Athl Train. 1994;29:224–230. [PMC free article] [PubMed]
  • Kinzey S J, Ingersoll C D, Knight K L. The effects of selected ankle appliances on postural control. J Athl Train. 1997;32:300–303. [PMC free article] [PubMed]
  • Feurbach J W, Grabiner M D. Effect of the Aircast on unilateral postural control: amplitude and frequency variables. J Orthop Sports Phys Ther. 1993;7:149–154. [PubMed]
  • Bennell K L, Goldie P A. The differential effects of external ankle support on postural control. J Orthop Sports Phys Ther. 1994;20:287–295. [PubMed]
  • Diener H C, Dichgans J, Guschlbauer B, Mau H. The significance of proprioception on postural stabilization as assessed by ischemia. Brain Res. 1984;296:103–109. [PubMed]
  • Kelly J P. The Sense of Balance: Principles of Neuroscience. 3rd ed. Appleton & Lange; New York, NY: 1991. pp. 500–511.
  • Soames R W, Atha J, Harding R H. Temporal changes in the pattern of sway as reflected in power spectral density analysis. Agressologie. 1976;17:15–20. [PubMed]
  • Shumway-Cook A, Horak F B. Assessing the influence of sensory interaction on balance: suggestion from the field. Phys Ther. 1986;66:1548–1550. [PubMed]
  • Winter D A, Patla A E, Frank J S. Assessment of balance control in humans. Med Prog Technol. 1990;16:31–51. [PubMed]
  • Soames R W, Atha J. The spectral characteristics of postural sway behaviour. Eur J Appl Physiol Occup Physiol. 1982;49:169–177. [PubMed]
  • Freeman M A, Dean M R, Hanham I W. The etiology and prevention of functional instability of the foot. J Bone Joint Surg Br. 1965;47:678–685. [PubMed]
  • McGuine T A, Greene J J, Best T, Leverson G. Balance as a predictor of ankle injuries in high school basketball players. Clin J Sport Med. 2000;10:239–244. [PubMed]

Articles from Journal of Athletic Training are provided here courtesy of National Athletic Trainers Association