Skip to main content
PMC home page
Journal of Athletic Training logoLink to Journal of Athletic Training
. 2007 Jan-Mar;42(1):60–65.

Test-Retest Reliability of Sudden Ankle Inversion Measurements in Subjects With Healthy Ankle Joints

Christophe Eechaute 1, Peter Vaes 1, William Duquet 1, Bart Van Gheluwe 1
PMCID: PMC1896082  PMID: 17597945

Abstract

Context: Sudden ankle inversion tests have been used to investigate whether the onset of peroneal muscle activity is delayed in patients with chronically unstable ankle joints. Before interpreting test results of latency times in patients with chronic ankle instability and healthy subjects, the reliability of these measures must be first demonstrated.

Objective: To investigate the test-retest reliability of variables measured during a sudden ankle inversion movement in standing subjects with healthy ankle joints.

Design: Validation study.

Setting: Research laboratory.

Patients or Other Participants: 15 subjects with healthy ankle joints (30 ankles).

Intervention(s): Subjects stood on an ankle inversion platform with both feet tightly fixed to independently moveable trapdoors. An unexpected sudden ankle inversion of 50° was imposed.

Main Outcome Measure(s): We measured latency and motor response times and electromechanical delay of the peroneus longus muscle, along with the time and angular position of the first and second decelerating moments, the mean and maximum inversion speed, and the total inversion time. Correlation coefficients and standard error of measurements were calculated.

Results: Intraclass correlation coefficients ranged from 0.17 for the electromechanical delay of the peroneus longus muscle (standard error of measurement = 2.7 milliseconds) to 0.89 for the maximum inversion speed (standard error of measurement = 34.8 milliseconds).

Conclusions: The reliability of the latency and motor response times of the peroneus longus muscle, the time of the first and second decelerating moments, and the mean and maximum inversion speed was acceptable in subjects with healthy ankle joints and supports the investigation of the reliability of these measures in subjects with chronic ankle instability. The lower reliability of the electromechanical delay of the peroneus longus muscle and the angular positions of both decelerating moments calls the use of these variables into question.

Keywords: joint stability, accelerometry, electromyography, ankle joint

Key Points

  • The validity of a measure primarily depends on the degree of its consistency and its magnitude of measurement error.

  • Studying inversion angles, time, and speed during a sudden ankle perturbation may teach us more about the mechanisms of ankle joint control than simply studying latency times.

  • The latency time of the peroneus longus muscle, the total inversion time, and the mean and maximal angular inversion speeds are reliable measures in subjects with healthy ankle joints.

  • Not knowing how reliable peroneal latency times are in subjects with chronic ankle instability prevents us from determining if such instability is related to delayed peroneal muscle reaction time.

Sudden ankle inversion tests in the standing position have been frequently applied to investigate whether onset of muscle activity is delayed in the peroneal muscles of subjects with chronically unstable ankle joints.1–12 Furthermore, Eils and Rosenbaum13 used such a test procedure to assess the effect of a 6-week wobble board training program on the latency time of the peroneal and tibialis anterior muscles. Before a test procedure is implemented as a valid and responsive measure, acceptable test-retest reliability (ie, the degree of consistency and agreement of test results among repeated measurements) must be demonstrated.14,15 The degree of “acceptable” clinical reliability depends on both the magnitude of the correlation coefficients and the standard error of measurement (SEM).16 Correlation coefficients reflect the degree of consistency among different sets of scores, and the SEM indicates the precision of the measurement.

Several authors3,12,17–19 have measured latency times of the peroneal muscles in subjects with healthy ankles. However, whether peroneal latency time is a reliable measure is unknown. Specifically, Benesch et al18 reported little to no relationship among peroneus longus muscle latency measurements (Spearman ρ = .00). Moreover, SEMs were not reported.

As ligament injuries are related to the strain rate,20,21 especially at high rates, it is useful to study the inversion speed during a sudden ankle inversion. In chronically sprained ankles, the biomechanical properties of ligaments and joint capsule may have been affected.22–24 Differences in ankle joint control between healthy ankle joints and chronically sprained ankles (ie, shorter inversion time and higher inversion speed represent worse control) during a sudden standing ankle inversion may be reflected in the magnitude and course of the ongoing inversion speed. Vaes et al11,12 documented acceleration and deceleration of inversion speed during sudden ankle inversions in standing subjects with healthy ankle joints and in patients with unstable ankles. During this test procedure, the ankle moves into plantar flexion, adduction, and supination, thereby stressing the anterolateral capsule and the anterior talofibular ligament of the ankle joint in particular. This experimental set-up was chosen because most lateral ankle sprains involve injuries to these tissues. Also, but to a lesser extent, the calcaneofibular ligament, posterior talofibular ligament, and capsuloligament tissue of the subtalar joint are specifically stressed during the test procedure.

Using accelerometry, Vaes et al12 observed 2 distinct decelerating moments during a sudden ankle inversion: a first and early deceleration at the moment of the highest inversion speed (Figure) and a second decelerating moment in the late phase of the inversion. In addition to the timing and angular position of these 2 decelerating moments, variables such as the total inversion time and mean and maximum inversion speed can be measured with this test procedure. These variables can give us supplementary information about how stabilizing mechanisms in unstable ankle joints may differ from those in healthy ankle joints, as most authors have only measured latency time of the peroneal muscles.

Variables of the sudden inversion in the standing position. A downward course of the velocity graph indicates an increase of inversion speed; an upward course indicates a decrease of inversion speed. 1 = latency time of the peroneus longus muscle (milliseconds), 2 = total inversion time (milliseconds), 3 = maximum inversion speed (m/s), 4 = time of the first deceleration (milliseconds), 5 = time of the second deceleration (milliseconds), 6 = angular position of the first deceleration (°), and 7 = angular position of the second deceleration (°).

Variables of the sudden inversion in the standing position. A downward course of the velocity graph indicates an increase of inversion speed; an upward course indicates a decrease of inversion speed. 1 = latency time of the peroneus longus muscle (milliseconds), 2 = total inversion time (milliseconds), 3 = maximum inversion speed (m/s), 4 = time of the first deceleration (milliseconds), 5 = time of the second deceleration (milliseconds), 6 = angular position of the first deceleration (°), and 7 = angular position of the second deceleration (°).

Open in a new tab

A longer inversion time, a higher inversion speed, and decelerating moments occurring later in time and at larger inversion angles indicate less control of the ankle joint during a sudden ankle inversion. Also, possible effects of external supports or rehabilitation programs on the deceleration of inversion speed and peroneal reaction time in patients with chronic ankle instability can be assessed with this test procedure. However, the test-retest reliability of these variables is unknown, both in subjects with healthy ankle joints and in patients with chronic ankle instability. Our purpose, therefore, was to investigate the test-retest reliability of these variables with a sudden ankle inversion in standing subjects with healthy ankle joints.

METHODS

Subjects

A total of 15 subjects (12 men, 3 women: age = 22.1 ± 2.4 years, height = 177.7 ± 9.2 cm, mass = 96.7 ± 8.2 kg) with healthy ankle joints participated in this reliability study. Inclusion criteria were age between 17 and 30 years and activity in recreational or competitive sports. Exclusion criteria were a history of injury to the lower extremities during the 3 months before testing or a traumatic lateral ankle sprain, ankle surgery, or ankle fracture; signs of recurrent ankle sprains or feelings of the “ankle giving way”; vestibular deficits; presence of muscle fatigue or pain in the lower extremity at the time of testing; or having been ill (ie, any health problem that required a medical consultation and prevented the subject from performing regular professional, recreational, or sports activities) between the 2 test sessions. Conformity with these criteria was assessed by means of a questionnaire or clinical examination of the lower extremity. This study was approved by the Ethical Committee of the Vrije Universiteit Brussel, and subjects signed written informed consents.

Test Procedures

We conducted the sudden ankle inversion in the same manner as described by Vaes et al.11,12 Subjects stood on a custom-designed ankle inversion platform, with both feet tightly fixed on independently movable trapdoors12,25 (constructed by the Physical Therapy Department and the Human Biometry and Biomechanics Department, Vrije Universiteit Brussel, Brussels, Belgium). Each footplate was positioned in 40° of plantar flexion, with the shoe in 15° of adduction. The tight fixation of the foot in the shoe ensured ankle positions with similar plantar flexion and inversion. The subject was asked to put full body weight on the tested ankle. The knee of the tested leg was in extension, while the opposite knee was flexed and totally unloaded. The operator visually controlled the exact positioning of the subject, who was listening to music via headphones and, therefore, was oblivious to external noise. The operator then imposed a sudden ankle inversion of 50°. During the sudden ankle inversion, the acceleration and deceleration of the tilting trap were measured with a Brüel and Kjör accelerometer (model 4393; Nacrum, Denmark) mounted on the platform close to the lateral side of the heel. The angular position of the falling platform was registered using electrogoniometers (custom design; Becker Meditec, Karlsruhe, Germany) permanently fixed on the back of the inversion platforms.

Electromyography (EMG) of the peroneus longus muscle was recorded using disposable, self-adhesive, and ready-to-use surface electrodes (22 mm by 14 mm; Medicotest, Lystkke, Denmark). These surface electrodes were fixed on the only palpable part of the peroneus longus muscle (distal to the caput fibula of the tibia at the superior lateral side of the lower leg) and parallel with its longitudinal axis, 5 mm to 10 mm apart, side to side. As the EMG amplifiers measured the incoming signal differentially, a third reference electrode, placed on the frontal part of the patella, was necessary. These surface electrodes were fixed close to the motor point of the peroneus longus muscle and parallel to its longitudinal axis. In order to enhance electric conductance, the skin was shaved and rubbed with alcohol before the electrodes were attached. Lower and upper cut-off frequencies were 25 Hz and 700 Hz, respectively. The EMG signal did not undergo any further processing and was always analyzed in its raw form.

The criterion for the onset of EMG of the peroneus longus muscle during the sudden ankle inversion was an increase in the signal to more than twice the noise level (Figure). The accelerometric, electrogoniometric, and EMG signals were sampled simultaneously at 1000 Hz and recorded using a Varioport data logger (Becker Meditec). The sampling rate was set at 1000 Hz, implying a time resolution of 1 millisecond. This means that time intervals or durations can theoretically be measured with an accuracy of 1.4 milliseconds. However, as the latter error figure may be doubled due to additional reading errors, it is safer to estimate the final time accuracy at 3 milliseconds. The recorded signals were represented on the computer screen by 3 graphs (Figure).

During the sudden ankle inversion, 2 distinct decelerating moments can be observed: a first and early deceleration at the moment of the highest inversion speed (see 4 in Figure) and a second and late deceleration (see 5) just before the end of the sudden ankle inversion.

The same examiner tested subjects twice, with an interval of 1 week between the 2 test sessions. Both ankles of each subject were tested. The order of ankle testing was randomly chosen, and the order of testing was the same for both test occasions. To habituate the subjects to the test procedure (ie, subjects were able to stand with a fully relaxed peroneus longus muscle), several practice trials were performed. Then 3 to 6 inversions of each ankle were recorded in each session.

Trials were discarded in 2 situations: when the muscle activity of the peroneus longus muscle exceeded twice the level of the noise in the 100-millisecond time period before the start of the perturbation and when the end of the perturbation was not characterized by a steep decline of the inversion speed, indicating that the subject did not put full body weight on the tested leg.

In order to avoid observer bias during the processing of the data, the observer was blinded for the assessment of the variables. We randomly converted the identity of the subjects and the recording date of the data files and graphs into numeric values so that the observer did not know whether the data originated from the test or retest session.

The measured variables were assessed manually from the curves representing the time histories for inversion speed (hardware integrated from the accelerometer signal), for EMG activity of the peroneus longus muscle, and for goniometric inversion displacement of the platform, as recorded by the Varioport data logger. Special measuring tools of the accompanying Varioport software allowed onscreen extraction of amplitude and time of any data point on a displayed curve. The following 10 variables were assessed in this way:

  • The latency time of the peroneus longus muscle (labeled 1 in Figure), being the time interval between the start of the inversion and the onset of EMG. The criterion for the onset of EMG of the peroneus longus muscle during the sudden ankle inversion was an increase in the signal to more than twice the noise level.

  • The total inversion time (labeled 2 in the Figure), being the total time between the start (indicated by an initial vertical course of the graph) and the end of the tilting of the inversion platform (indicated by the vertical upward course of the graph).

  • The mean inversion speed, calculated as the ratio of the total inversion time to the total angular displacement during the inversion (50°).

  • The maximum inversion speed, calculated using the following formula: maximum inversion speed (°/s) = maximum inversion speed (in m/s) × 57.3/0.157 m, with 57.3 being the conversion factor from radials into degrees and 0.157 m being the distance from the accelerometer position to the axis of platform rotation.

  • The time of the first decelerating moment, being the time interval between the start of the tilting and the first upward deflection of the velocity graph.

  • The angular position of the first decelerating moment, corresponding to the time of the first decelerating moment and read from the goniometric curve.

  • The time of the second decelerating moment, being the time interval between the start of the tilting of the platform and the second upward deflection of the velocity graph during the sudden inversion.

  • The angular position of the second decelerating moment, corresponding to the time of the second decelerating moment and read from the goniometric curve.

  • The electromechanical delay of the peroneus longus muscle was assessed separately before the perturbation test. At present, the electromechanical delay can only be determined during a concentric, voluntary contraction, so it was measured with the subject seated, with the lower leg supported by a chair and the foot in a relaxed position, not touching the chair. The accelerometer was attached to the forefoot (fixed with tape). After an auditory signal, the subject was asked to move the foot as quickly as possible from a resting position into eversion. The electromechanical delay of the peroneus longus muscle was then measured as the time elapsed between the start of the EMG and the beginning of the movement of the foot into eversion.

  • The motor response time of the peroneus longus muscle, calculated as the sum of the peroneal latency time and its electromechanical delay.

Data Analysis

The results of the left ankles correlated significantly with the results of the right ankles for the following variables: latency time and motor response time of the peroneus longus muscle, maximum inversion speed, and the time of the second deceleration point (Pearson r > .70, P < .05). The subsequent statistical tests were performed for the left and right ankles separately for all variables. As the data of all variables were normally distributed (Kolmogorov-Smirnov goodness-of-fit test, P < .05), correlations were calculated using an intraclass correlation (ICC) model (3,1) after a repeated-measures analysis of variance with the significance level set at P < .05. Also, standard errors of the measured variables were calculated (SEM = SD×√1 − r).

RESULTS

The ICCs of the measured variables ranged from .17 to .89, representing poor to good reliability (Tables 1 and 2). The maximum inversion speed showed the highest reliability coefficient (ICC = .89, SEM = 34.8°/s) and the electromechanical delay of the peroneus longus muscle, the lowest (ICC = .17, SEM = 2.7 milliseconds).

Table 1. Reliability Results for the Variables in the Right Ankle.

graphic file with name i1062-6050-42-1-60-t01.jpg

Open in a new tab

Table 2. Reliability Results for the Variables in the Left Ankle.

graphic file with name i1062-6050-42-1-60-t02.jpg

Open in a new tab

DISCUSSION

We attempted to conduct this reliability study under the best possible methodologic conditions. A 1-week interval was chosen to minimize potential confounding effects (ie, minimizing learning effects, fatigue effects, or true improvement in condition or function). When assessing test-retest reliability, a time interval of between 2 and 14 days is advocated.15 To reduce sources of measurement error as much as possible, test conditions were kept constant. Therefore, all measurements (test or retest) were taken at the same time of day and were evaluated by the same blinded rater. Intrarater reliability has been shown to be excellent for all measured variables (reliability coefficients range from .92 to .99).25 As a general guideline, reliability coefficients below .50 represent poor reliability, those between .50 and .75 reflect moderate to good reliability, and values above .75 indicate good to excellent reliability.16 Reliability coefficients represent the amount of variation between a set of scores that can be explained by true variance,16,26 and the SEM represents the range of the estimated measurement error around an observed score.

Peroneus Longus Latency Time

We observed no significant difference between the 2 test occasions for the peroneus longus muscle latency time. This finding is in agreement with the results of Lynch et al19 and Benesch et al.18 The ICC values for the latency time in our study (.71 and .83) indicate moderate to good reliability and are obviously higher than the correlation coefficient reported by Benesch et al18 (Spearman ρ = .00). The ICC of the latency time of the peroneus longus muscle of the left ankle was .83; thus, 83% of its variability can be ascribed to true variation among subjects, and 17% was caused by measurement error. The SEMs of the peroneus longus muscle in this study are between 6.4 and 8.3 seconds, meaning that we are 95% confident the true scores of the latency time of the left ankle lie between 49.7 and 74.9 milliseconds and those of the right ankle between 47.6 and 81.2 milliseconds.

From the study of Vaes et al12 it is clear that controversy exists over whether the latency time of the peroneal muscles in patients with unstable ankle joints is significantly delayed compared with that of subjects with healthy ankle joints. None of the cited authors1–12 reported SEMs of peroneal latency time in subjects with healthy ankle joints or patients with unstable ankle joints. If we knew the SEM of the peroneus longus muscle latency time in those populations, we might better understand why some authors did find significant differences in peroneal latency times between healthy ankle joints and chronic unstable ankle joints6–8,10,11 and others did not.1–5,9,12

Motor Response Time and Electromechanical Delay

The motor response time of the peroneus longus muscle has been calculated because the peroneal latency time only represents the start of EMG activity in the muscle (62 to 66 milliseconds after the start of the sudden ankle inversion). However, a supplementary time period (the electromechanical delay) must occur before sufficient motor units are recruited to generate muscle force that can induce or control motion. So when we take this electromechanical delay of the peroneus longus muscle into account along with the latency time, we have more reliable information about the exact moment at which the peroneus longus muscle builds sufficient power to decelerate sudden ankle inversions.

The electromechanical delay values of the peroneus longus muscle display low ICCs. This is probably due to the low range of scores (Tables 1 and 2). Reliability is based upon the proportion of the total observed variance in scores that can be attributed to error.16 When the variance in a set of scores is decreased, the proportion of the error variance to the total observed variance increases, resulting in a decrease in the magnitude of the ICC. In contrast with the latency time, the ICC value of the electromechanical delay of the peroneus longus of the left ankle was considerably lower than that of the right ankle. This cannot be explained by a difference in the range of scores between the left and right ankles, as these were comparable (between 15 and 24 milliseconds for the left ankle and between 16 and 23 milliseconds for the right ankle). Correlation coefficients can also be subjected to sampling error. So the difference in magnitude between the ICC values may result by chance, as these ICC values were not significant, and, thus, may not be a good estimate of the real population correlation coefficient for these variables.

Despite the low ICCs of the electromechanical delay of the peroneus longus muscle, the ICCs of the motor response time of the muscle still represent moderate to good reliability. The SEMs of the electromechanical delay are rather small, thereby influencing the variability of the motor response time values in a minor way. Moreover, the motor response time values of the peroneus longus muscle are characterized by rather low SEMs (6.5 to 9.3 milliseconds), making them useful for future researchers. Measuring the motor response time of the peroneus longus muscle in both subjects with healthy ankle joints and patients with chronic ankle instability may reveal the presence or absence of a possibly delayed effective counteraction of the peroneus longus muscle during sudden ankle inversion in these patients.

Inversion Speed and Inversion Time

The ICCs of the maximum inversion speed, the mean inversion speed, and the total inversion time indicate good consistency between the 2 test occasions. Compared with the other variables we measured, the inversion speed values have larger SEMs. This can be explained by the wider range of scores of the inversion speed values between subjects (see SDs in Tables 1 and 2). Several authors measured inversion times or inversion speed values during sudden ankle inversions in standing subjects with healthy ankle joints.12,27,28 Shorter inversion times and higher inversion speed values indicate less control of sudden ankle inversions. Vaes et al11 reported a significantly shortened total inversion time in subjects with chronically unstable ankle joints compared with healthy ankle joints. Vaes et al12 could not confirm these findings. Knowing the SEM of the total inversion time values in both subjects with healthy ankle joints and patients with chronic ankle instability may help us to interpret these conflicting results.

Ankle supports are thought to prevent injuries to the ankle joint by restricting excessive ankle inversion amplitude and inversion speed. Anderson et al28 and Scheuffelen et al27 observed that ankle supports lengthened the total inversion time28 and reduced the mean27 or maximum28 inversion speed in subjects with healthy ankle joints compared with controls (those wearing no ankle support). These authors did not investigate the test-retest reliability of the measured variables. Lacking reliability coefficients and the SEM of inversion time or inversion speed values impedes the interpretation of their results, as statistical significance is not the same as clinical significance.

The mean inversion speed values assessed by Vaes et al12 and Anderson et al28 (400°/s to 430°/s) were comparable with those in our study (429.1 ± 43.5°/s to 438.1 ± 48.5°/s). The maximum inversion speed values we observed ranged from 383°/s to 668°/s (±82°/s) and are substantially higher than those reported by Anderson et al28 (324 ± 111.9°/s). This may be due to differences in test protocols, such as the amount of ankle inversion (22° versus 50°) or the control of anticipatory muscle activity in the peroneal muscles.

Timing and Angular Position of Decelerating Moments

The ICCs of the times and angular positions of the first and second decelerating moments represent poor to good reliability. The lower ICCs of these variables may be affected by their lower variance among subjects' scores, as the SDs of these values are much lower when compared with the total inversion time and the mean and maximum inversion speed values. When looking at both the ICC and SEM, the time values of the first and second decelerating moments are more reliable than their corresponding angles.

As with the electromechanical delay of the peroneus longus muscle, the ICC value of the angular position of the second decelerating moment of the left ankle (ICC = .37) is considerably lower than for the right ankle (ICC = .70). The lower range of scores of the left ankle (between 42° and 48° for the left ankle and between 40° and 50° for the right ankle) may explain the observed difference in magnitude between ICC values. Moreover, the ICC value of the left ankle is not significant (P = .16). If we repeat the reliability testing with another sample, we might observe a different correlation coefficient that better approximates the ICC value of the right ankle.

Evaluating times and angular positions of decelerating moments may help us to understand the mechanisms of ankle stabilization during excessive sudden ankle inversions. The first decelerating moment (at 31 to 35 milliseconds after the start of the sudden ankle inversion) may be of a passive nature, resulting from a very fast stretch of the capsule and ligaments (especially the talofibular ligament) of the ankle joint, subtalar joint, and collagen tissue in the peroneal muscle-tendon unit. It cannot be explained by an active intervention, as the motor response time of the peroneus longus muscle starts far too late (after 82 to 86 milliseconds). In addition, no EMG activity of the peroneus longus muscle could be observed in the 100-millisecond time period preceding the sudden ankle perturbation.

The second decelerating moment may be partially ascribed to an eccentric action of the peroneus longus muscle, controlling ankle joint motion in the late phase of the sudden ankle inversion. The motor response time of the peroneus longus muscle (ie, the time delay before the muscle can generate sufficient muscle force to induce or control ankle motion and starting at 82 to 86 milliseconds after the onset of the sudden ankle inversion) clearly precedes the occurrence of the second decelerating moment (103 to 109 milliseconds after the onset of the sudden ankle inversion). However, in 3 subjects, no second decelerating moment could be observed. Hence, the meaning of this second decelerating moment remains unclear, and its usefulness for further research can be questioned.

CONCLUSIONS

Based upon the ICC and the SEM values, the reliability of the latency time and motor response time of the peroneus longus muscle, the total inversion time, the mean and maximum inversion speeds, and the time values of the first and second decelerating moments is acceptable in subjects with healthy ankle joints, supporting further research into the test-retest reliability of sudden ankle inversion measurements in patients with chronic ankle instability. The lower reliability of the electromechanical delay of the peroneus longus muscle and the angular positions of decelerating moments (of the left ankle in particular) brings into question the usefulness of these measures in further research.

REFERENCES

  1. Beckman SM, Buchanan TS. Ankle inversion injury and hypermobility: effect on hip and ankle muscle electromyography onset latency. Arch Phys Med Rehabil. 1995;76:1138–1143. doi: 10.1016/s0003-9993(95)80123-5. [DOI] [PubMed] [Google Scholar]
  2. Ebig M, Lephart SM, Burdett RG, Miller MC, Pincivero DM. The effect of sudden inversion stress on EMG activity of the peroneal and tibialis anterior muscles in the chronically unstable ankle. J Orthop Sports Phys Ther. 1997;26:73–77. doi: 10.2519/jospt.1997.26.2.73. [DOI] [PubMed] [Google Scholar]
  3. Fernandes N, Allison GT, Hopper D. Peroneal latency in normal and injured ankles at varying angles of perturbations. Clin Orthop Rel Res. 2000;375:193–201. doi: 10.1097/00003086-200006000-00023. [DOI] [PubMed] [Google Scholar]
  4. Isakov E, Mizrahi J, Solzi P, Susak Z. Response of the peroneal muscles to sudden inversion of the ankle during standing. Int J Sports Biomech. 1986;2:100–109. [Google Scholar]
  5. Johnson MB, Johnson CL. Electromyographic response of peroneal muscles in surgical and nonsurgical injured ankles during sudden inversion. J Orthop Sports Phys Ther. 1993;18:497–501. doi: 10.2519/jospt.1993.18.3.497. [DOI] [PubMed] [Google Scholar]
  6. Karlsson J, Andreasson GO. The effect of external ankle support in chronic lateral ankle joint instability: an electromyographic study. Am J Sports Med. 1992;20:257–261. doi: 10.1177/036354659202000304. [DOI] [PubMed] [Google Scholar]
  7. Konradsen L, Ravn JB. Prolonged peroneal reaction time in ankle instability. Int J Sports Med. 1991;12:290–292. doi: 10.1055/s-2007-1024683. [DOI] [PubMed] [Google Scholar]
  8. Löfvenberg R, Kärrholm J, Sundelin G, Ahlgren O. Prolonged reaction time in patients with chronic lateral instability of the ankle. Am J Sports Med. 1995;23:414–417. doi: 10.1177/036354659502300407. [DOI] [PubMed] [Google Scholar]
  9. Nawoczenski DA, Owen MG, Ecker ML, Altman B, Epler M. Objective evaluation of peroneal response to sudden inversion stress. J Orthop Sports Phys Ther. 1985;7:107–109. doi: 10.2519/jospt.1985.7.3.107. [DOI] [PubMed] [Google Scholar]
  10. Rosenbaum D, Becker HP, Gerngross H, Claes L. Peroneal reaction times for diagnosis of functional ankle instability. Foot Ankle Surg. 2000;6:31–38. [Google Scholar]
  11. Vaes P, Van Gheluwe B, Duquet W. Control of acceleration during sudden ankle supination in people with unstable ankles. J Orthop Sports Phys Ther. 2001;31:741–752. doi: 10.2519/jospt.2001.31.12.741. [DOI] [PubMed] [Google Scholar]
  12. Vaes P, Duquet W, Van Gheluwe B. Peroneal reaction times and eversion motor response in healthy and unstable ankles. J Athl Train. 2002;37:475–480. [PMC free article] [PubMed] [Google Scholar]
  13. Eils E, Rosenbaum D. A multi-station proprioceptive exercise program in patients with ankle instability. Med Sci Sports Exerc. 2001;33:1991–1998. doi: 10.1097/00005768-200112000-00003. [DOI] [PubMed] [Google Scholar]
  14. Fleiss J. The Design and Analysis of Clinical Experiments. New York, NY: John Wiley and Sons; 1986.
  15. Fitzpatrick R, Davey C, Buxton MJ, Jones DR. Evaluating patient-based outcome measures for use in clinical trials. Health Technol Assess. 1998;2:1–74. [PubMed] [Google Scholar]
  16. Portney L, Watkins M. Foundation of Clinical Research. Application to Clinical Practice. 2nd ed. Upper Saddle River, NJ: Prentice Hall Health; 2000:505–528.
  17. Hopper D, Allison G, Fernandes N, O'Sullivan L, Wharton A. Reliability of the peroneal latency in normal ankles. Clin Orthop Rel Res. 1998;350:159–165. [PubMed] [Google Scholar]
  18. Benesch S, Lipke K, Rosenbaum D, Becker H. Reliability of peroneal reaction time measurements. Clin Biomech (Bristol, Avon) 2000;15:21–28. doi: 10.1016/s0268-0033(99)00026-1. [DOI] [PubMed] [Google Scholar]
  19. Lynch SA, Eklund U, Gottlieb D, Renstrom PA, Beynnon B. Electromyographic latency changes in the ankle musculature during inversion moments. Am J Sports Med. 1996;24:362–369. doi: 10.1177/036354659602400319. [DOI] [PubMed] [Google Scholar]
  20. Woo SL, Peterson RH, Ohland KJ, Sites TJ, Danto MI. The effects of strain rate on the properties of the medial collateral ligament in skeletally immature and mature rabbits: a biomechanical and histological study. J Orthop Res. 1990;8:712–721. doi: 10.1002/jor.1100080513. [DOI] [PubMed] [Google Scholar]
  21. Danto MI, Woo SL. The mechanical properties of skeletally mature rabbit anterior cruciate ligament and patellar tendon over a range of strain rates. J Orthop Res. 1993;11:58–67. doi: 10.1002/jor.1100110108. [DOI] [PubMed] [Google Scholar]
  22. Attarian DE, McCrackin HJ, Devito DP, McElhaney JH, Garrett WE., Jr. Biomechanical characteristics of human ankle ligaments. Foot Ankle. 1985;6:54–58. doi: 10.1177/107110078500600202. [DOI] [PubMed] [Google Scholar]
  23. Noyes FR, DeLucas JL, Torvik PJ. Biomechanics of anterior cruciate ligament failure: an analysis of strain-rate sensitivity and mechanisms of failure in primates. J Bone Joint Surg Am. 1974;56:236–253. [PubMed] [Google Scholar]
  24. Crowninshield RD, Pope MH. The strength and failure characteristics of rat medial collateral ligaments. J Trauma. 1976;16:99–105. doi: 10.1097/00005373-197602000-00004. [DOI] [PubMed] [Google Scholar]
  25. Eechaute C. The Effect of a Six Week During Wobble Board Training Program on a Sudden Inversion in the Standing Position and on the Presence of Disabilities and Handicaps in Patients With Chronic Ankle Instability [master's thesis] Brussels, Belgium: Vrije Universiteit Brussel; 2000.
  26. Drouin J. How should we determine a measurement is appropriate for clinical practice? Athl Ther Today. 2003;8((4)):2–4. [Google Scholar]
  27. Scheuffelen C, Rapp W, Gollhofer A, Lohrer H. Orthotic devices in functional treatment of ankle sprain. Int J Sports Med. 1993;14:140–149. doi: 10.1055/s-2007-1021158. [DOI] [PubMed] [Google Scholar]
  28. Anderson DL, Sanderson DJ, Hennig EM. The role of external nonrigid ankle bracing in limiting ankle inversion. Clin J Sport Med. 1995;5:18–24. doi: 10.1097/00042752-199501000-00004. [DOI] [PubMed] [Google Scholar]

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

RESOURCES