Lateral ankle sprains are among the most common injuries in athletics (7). Many ankle injuries occur while participating in athletic activities that require jumping and landing, such as basketball, soccer, and volleyball (18). It has been estimated that between 23,000 and 27,000 ankle sprains occur every day in the United States (14). However, 55% of individuals suffering a lateral ankle sprain may not seek treatment from a health care professional (19); thus, the actual incidence of ankle injuries could be vastly underestimated. Yeung et al. (29) noted that in sports such as basketball, the reinjury rate for lateral ankle sprains has been reported to exceed 70%. Approximately 40–75% of individuals suffering from a lateral ankle sprain may develop residual symptoms or chronic ankle dysfunction (8).
Chronic ankle dysfunction typically includes complaints of pain during activity, recurrent swelling, a feeling of giving way, and repetitive injury (22). These residual symptoms were first defined by Freeman et al. (6) and fall into two categories, mechanical and functional instability. Mechanical instability refers to anatomical changes, most notably joint laxity of the lateral ligaments and capsule. Functional ankle instability (FAI) is defined as impaired proprioception, strength, postural control, and neuromuscular control without ligamentous laxity (6). These instabilities are not mutually exclusive, most likely having some degree of overlap. For example, Tropp et al. (26) found that only 42% of FAI was associated with mechanical instability and only 36% of mechanical instability was associated with FAI.
Recently, sports medicine researchers have attempted to find both clinical and laboratory measures that could detect and explain deficits in athletes with FAI. Previous research has examined proprioception 13), muscle latency 27), and postural control (26); however, none of these studies were conducted in a dynamic setting. Static conditions may fail to elicit postural control deficiencies due to the ease of the testing procedure (24). Therefore, a dynamic measure would be more challenging and potentially more effective for detecting deficits in subjects with joint instability. This would benefit clinicians by aiding in the detection of instabilities that would otherwise go undetected (25).
Reimann et al. (24) developed a dynamic clinical measure of subjectively assessing postural control during a functional performance task. The authors used a multiple, single-leg hop test because of its association with the requirements of a sports-related function. The criterion measure was error scores assessed by the tester. The reliability of the outcome measures ranged from 0.70 to 0.92. A newer objective measure of dynamic stability is time to stabilization (TTS). This measure is based upon previous studies that have used single leg stance assessments of individuals with lower-extremity injuries (26). TTS is a functional examination of stability by definition, forcing subjects to maintain balance through a transition from a dynamic to a static state (9). Colby et al. (3) used the measure and examined deficiencies in dynamic stability in ACL-deficient and reconstructed patients. More recently, TTS has been used in preliminary studies examining FAI (25).
Two protocols have been used to measure TTS: a step down and a jump protocol. Depending on the protocol, the subject would either step down off an elevated platform or jump a set distance with a minimum height requirement and land on the force plate and regain their balance (3,25). Both protocols have been shown to be highly reliable methods of determining TTS in all directions for both the dominant and nondominant limbs (3). Furthermore, Birmingham (1) indicates that objective measures of balance after a maximal forward jump would be appropriate to distinguish between group means.
Colby et al. (3) and Ross and Guskiewicz (25) have used different methods of analysis to measure TTS. Both methods are designed to determine the point where the ground reaction forces (GRF) in the anterior/posterior, medial/lateral, and vertical directions decay to and stay within the normal values of their respective static baselines but are very different in how they determine their values. Other studies have used combinations of the two protocols and two methods of analysis (28). As a result, large variations in TTS scores exist and a consistent basis for comparison is lacking. Therefore, the purpose of this investigation is to determine which combination of protocol and analysis would be the most effective in detecting differences in dynamic stability between healthy subjects and subjects with FAI.
Fifty-eight subjects participated in this investigation: 29 healthy individuals (21.2 ± 2.7 yr, 170.5 ± 7.6 cm, 67.9 ± 12.6 kg) and 29 individuals with FAI (21.8 ± 2.3 yr, 173.6 ± 11.4 cm, 77.8 ± 17.6 kg). Each subject was tested during a single test session, at which time the subjects read and signed the informed consent approved by the university institutional review board, completed a medical history and ankle stability questionnaire to determine eligibility, and were placed in their respective groups (healthy or FAI). All subjects were free from lower-extremity and head injury for the previous 3 months, and did not suffer from any equilibrium disorders. Additionally, healthy subjects were free from previous ankle injuries, whereas FAI subjects had to meet the criteria set by Hubbard and Kaminski (13). This criterion establishes that each subject has sensations of weakness, “looseness,” and episodes of giving way during daily activity without any history of fractures to the ankle, injury within the past 3 months, and no formal rehabilitation of the affected ankle.
Subjects reported to a university research laboratory for testing. Each subject’s test leg was determined. For FAI subjects, this was the ankle suffering from instability. Healthy subjects were selected as matched controls by leg dominance. Subjects then had their tested ankle measured for stiffness and laxity followed by performing the maximum vertical jump (Vertmax) testing.
Baseline measures of static stance were then recorded as a single leg stance on a force plate, during a 5-s window (25). Testing protocols were conducted in a counterbalanced fashion to minimize any potential learning or fatigue effect. Subjects either completed three successful step protocol trials followed by jump protocol trials or completed three successful jump protocol trials followed by the step trials. If a subject lost balance and touched the floor with the contralateral limb, the trial was discarded and repeated. Likewise, if a short additional hop occurred upon landing, the trial was discarded and repeated. The average of the three successful trials was then used for analysis.
Stiffness and laxity.
Ligament function was measured using a LigMaster (Sport Tech, Inc., Charlottesville, VA). Subjects were placed in a side lying position so that the leg to be tested was slightly flexed at the knee around a counter bearing and the heel was fixed against another counter bearing (Fig. 1). The primary investigator applied force to the anterior tibia via an actuator, stressing the anterior talo-fibular ligament. Force was applied up to 150 N, and the two trial averages of the slope of the curve (stiffness) and displacement at 30 N were recorded. Harukazu et al. (10) indicated that an anterior drawer test performed at 30 N of force in subjects not under anesthesia was more sensitive to detecting differences than other force values.
Maximum vertical leap.
Subjects’ maximal vertical leap was measured using a Vertec vertical jump device (Sports Imports, Columbus, OH). Maximal vertical leap was determined as the difference between the subjects’ maximum standing reach height and the subjects’ maximum leap height. Measurements were made at 1.27-cm increments. The largest difference between the standing reach and leap height of three trials was used during subject testing.
Time to stabilization.
A Bertec triaxial force plate (Bertec Corporation, Columbus, OH) was used to measure duration of instability at a frequency of 180 Hz (25). The force plate data underwent an analog to digital conversion and was stored on a PC-type computer using the DATAPAC 2000 (Run Technologies, Laguna Hills, CA) analog data acquisition, processing, and analysis system. All subjects completed two protocols, a step down and jump protocol in a random counterbalanced order. The step down protocol followed the protocol used by Colby et al. (3). Each subject started atop a 20-cm-high platform facing the force plate. Subjects were instructed to step off the platform and onto the force plate leading with the test leg (Fig. 2). All subjects were instructed to stabilize as quickly as possible, place their hands on their hips, and look straight ahead for 20 s. Upon completion, the subject returned to the platform and completed two additional trials. The jump protocol followed the protocol used by Ross and Guskiewicz (25). Subjects started in a standing position 70 cm from the center of the force plate. Each subject was required to jump off both legs and touch a overhead marker placed at a position equivalent to 50% of the subject’s maximum vertical leap before landing on the force plate (Fig. 3). Each subject was to land on the test leg, stabilize as quickly as possible, and balance for 20 s with their hands on their hips and looking straight ahead. Upon completion, the subject returned to the start position and completed two more trials. All subjects were instructed to jump with their head up and hands in a position to touch the designated marker.
Self-report perception of difficulty.
At the conclusion of his/her test session, each subject was required to subjectively assess which task they perceived as most difficult. Subjects were asked to report their perception of which task (step protocol or jump protocol) was more difficult to complete.
Time to stabilization (TTS).
Data from each of the two completed events were thoroughly evaluated using both methods of analysis (UTOP and SE). Thus, four combinations of task and method of analysis were possible. All data were initially analyzed using the Colby et al. (3) method. Stabilization times in the medial/lateral and anterior/posterior directions were determined using the technique of sequential estimation. The technique incorporates an algorithm to calculate a cumulative average of the data points in a series by successively adding in one point at a time (3). This cumulative average was compared against the overall series mean, and the individual series was considered stable when the sequential average remained within 0.25 standard deviation of the overall series mean (Fig. 4). The series consists of all data points within the first 3 s of touch down. These stabilization times are calculated in the same plane or direction as postural sway measures.
Additionally, a stabilization time was calculated in the vertical plane or direction. Vertical TTS was established as the time when the vertical force component reached and stayed within 5% of the subject’s body weight after landing. A subject’s body weight was established before data collection and was calculated as the average of the variation of the vertical GRF during a 5-s static stance.
The data from each completed trial were then evaluated through the method of analysis used by Ross and Guskiewicz (25). Vertical, medial/lateral, and anterior/posterior stabilization times were determined by calculating the time it took for the initial components of the GRF of a jump landing to become similar to the components of the GRF from a static single leg stance (Fig. 5). To make this determination, the vertical, medial/lateral, and anterior/posterior data are rectified and starting at the peak GRF, an unbounded third order polynomial is fitted to each of the GRF components. The TTS was determined as the point at which the unbounded polynomial transects the static horizontal lines.
Subject demographics were compared using independent sample t-tests. Dependent variables were initially compared using an independent sample t-test using a Bonferroni adjustment to rule out leg dominance differences. Because no leg dominance differences were noted, separate three-way (group × protocol × method of analysis) ANOVA with repeated measures on the last two factors were conducted for the TTS scores. An alpha level of 0.05 was used for all statistical tests.
Ankle joint stiffness and laxity.
The group means and standard deviations for ankle joint stiffness and displacement under 30 N of force can be seen in Table 1. No significant group differences for stiffness (T(56) = −1.318, P = 0.193) or laxity (T(56) = −0.882, P = 0.382) were observed.
Time to Stabilization
Anterior/posterior TTS revealed a significant protocol by analysis by group interaction (F(1,56) = 6.9, P = 0.011). Likewise, a significant protocol by group interaction (F(1,56) = 4.4, P = 0.042) and protocol by analysis interaction (F(1,56) = 14.1, P < 0.001) were noted. A significant protocol main effect (F(1,56) = 13.0, P < 0.001) was observed as the step protocol (3517.1 ± 90.0 ms) produced significantly longer TTS scores than the jump protocol (3230.4 ± 42.7 ms). An analysis main effect (F(1,56) = 58.6, P < 0.001) was also observed as the UTOP method (3850.5 ± 117.9 ms) produced significantly longer TTS scores than the SE method (2897.0 ± 13.3 ms). No significant analysis by group interaction (F(1,56) = 0.5, P = 0.498) was observed for anterior/posterior TTS. There was also no significant group (healthy and FAI) main effect observed (F(1,56) = 0.003, P = 0.954). Because there were no group differences, the data was collapsed and can be seen in Figure 6.
Medial/lateral TTS also revealed no significant differences between healthy and FAI groups (F(1,56) = 3.1, P = 0.083); thus, the data have been collapsed across groups (Fig. 6). A significant protocol main effect (F(1,56) = 11.5, P < 0.001) was observed as the step protocol (3224.0 ± 158.9 ms) produced significantly longer TTS scores than the jump protocol (2713.8 ± 150.8 ms). An analysis main effect (F(1,56) = 58.9, P < 0.001) was also observed as the UTOP method (3915.4 ± 250 ms) produced significantly greater TTS scores than the SE method (2022.5 ± 41.8 ms). No significant protocol by group (F(1,56) = 1.5, P = 0.226), analysis by group (F(1,56) = 2.5, P = 0.120), or protocol by analysis (F(1,56) = 0.3, P = 0.592) interactions were noted. Similarly, no significant protocol by analysis by group interaction (F(1,56) = 0.3, P = 0.592) was observed for medial/lateral TTS.
Vertical TTS indicated that there were no differences between healthy and FAI groups (F(1,56) = 1.9, P = 0.176); the data were collapsed across groups and appear in Figure 6. A significant protocol main effect (F(1,56) = 112.0, P < 0.001) was observed as the jump protocol (2381.7 ± 36.5 ms) produced significantly longer TTS than the step protocol (1533.5 ± 71.8 ms). A significant analysis main effect (F(1,56) = 176.6, P < 0.001) was also observed as the UTOP method (2554.4 ± 68.7 ms) produced significantly longer TTS scores than the SE method (1360.8 ± 52.1 ms). No significant protocol by group (F(1,56) = 0.1, P = 0.727), analysis by group (F(1,56) = 0.1, P = 0.719), or protocol by analysis (F(1,56) = 0.4, P = 0.529) interactions were noted. Likewise, no significant protocol by analysis by group interaction (F(1,56) = 1.3, P = 0.257) was observed for vertical TTS.
The purpose of this investigation was to determine which combination of protocol and analysis would be the most effective for detecting differences in dynamic stability between subjects without a history of ankle sprains and subjects with FAI. There were no significant mechanical differences (stiffness or laxity) detected between the healthy and FAI groups. This finding was necessary to ensure that our results could be focused directly on FAI. This finding is supported by previous research that has shown individuals with FAI may not be mechanically unstable (10,26). Therefore, we are confident that the subjects recruited were free of mechanical instability.
Anterior/posterior TTS scores did reveal a significant protocol by group interaction. This interaction indicates that the jump protocol, regardless of method of analysis, will identify dynamic stability deficits in functionally unstable ankles as compared with a step down protocol. It was expected that the jump-landing task used in the current study would produce longer TTS because it was a more stressful test of dynamic stability. It required the subject to jump instead of step down, producing increased GRF and greater range in the amplitude of those forces (5). Caulfield and Garrett (2) suggest that subjects with previous ankle injuries will protect their lateral ligament complex by landing in a more dorsiflexed position, causing increased GRF and an improper landing technique. It is believed that an interaction between central programming and peripheral feedback is responsible for dynamic control of the ankle during functional activities (16). Whereas this is only conjecture, as the current study did not examine kinematics, it is a possible explanation for the increased TTS scores seen in subjects with FAI.
Residual symptoms such as chronic ankle swelling may also explain the effectiveness of the jump protocol in determining group differences. Hopkins and Palmieri (12) indicate that during a step down test, peroneal muscle activity decreased just before touch down, as a result of joint effusion. Subjects were injury free for at least the last 3 months; however, there is no guarantee that minor joint effusion was not present in the functionally unstable ankles. Joint effusion may have decreased afferent information and subsequently prolonged muscle response time. Hoffman and Payne (11) attribute the diminished ability to detect motion and joint position sense to a decreased accuracy of afferent input and efferent output. Joint swelling and chronic ligament disruption could have caused deafferentation of the ankle region. This deafferentation would alter the interpretation of the joint’s position and movement in both closed and open kinetic chains. Thus, a deafferentation could also increase the latency period of the reacting joint musculature (15). This result is similar with previous studies that have found deficits in neuromuscular control in subjects suffering from FAI. It may be possible that anterior/posterior TTS will also be in agreement with McGuine et al. (18), who indicate that deficits in proprioception and postural sway can predict future ankle injury. The TTS scores may be able to help predict increased risk of reinjury and recommend functional rehabilitation to minimize that risk.
Deficits in medial/lateral TTS scores between the healthy and FAI groups were not significant, but did indicate that the jump protocol and the UTOP method of analysis were more likely to detect deficits in dynamic stability despite lack of statistical significance. This finding is consistent with previous studies that have examined FAI using a jump protocol and the UTOP method of analysis (25). However, it is possible that no statistical differences were observed between groups because of altered balance strategies. According to Pintsaar et al. (23), individuals suffering from FAI utilize a hip strategy to compensate for lack of support in the ankle during a single leg balance task. Therefore, it is possible that our subjects were able to successfully compensate for decreased neuromuscular control in their ankle because they were utilizing a hip strategy technique as a compensatory mechanism to stabilize after the jump. However, this is speculative, as EMG and joint kinematics were not examined in this investigation.
When examining vertical TTS, the jump protocol produced longer TTS scores, as did the UTOP method of analysis. These results were expected as the GRF are an integral part of the TTS value (25), and the jump protocol (4.03 ± 0.81 BW) produced higher peak vertical GRF than did the step down test (2.06 ± 0.53 BW). The longer TTS times associated with the UTOP method of analysis could be due to the application of a line of best fit, expressed as an unbounded third order polynomial when dealing with the vertical GRF. Colby et al. (3) determines TTS as the point where the vertical GRF reached and stayed within ± 5% of the subject’s body weight, over a 3-s window. Ross and Guskiewicz (25) use a 20-s window that decreases the slope of the line of best fit and prolongs its intercept with the horizontal line representing the vertical GRF baseline.
Whereas studies have examined anterior/posterior and medial/lateral TTS, no previous studies have examined vertical TTS as a tool to detect dynamic stability deficits in subjects with FAI. Colby et al. (3) found that using vertical GRF in conjunction with a step down protocol, deficits in dynamic stability could be identified in ACL reconstructed patients. During the step and jump landings, lower extremity musculature is responsible for decelerating and stabilizing the individual. Therefore, it could be inferred that eccentric strength and neuromuscular control of the lower extremity are vital to the safety of the lower extremity during landings. In a study conducted by Kovacs et al. (17), it was reported that the highest mean power was produced by the ankle plantarflexors during forefoot landings, followed by the knee and hip extensors. It may be that the tibialis anterior and triceps surae muscle groups play a greater role in the stabilization of the lower leg just before and after landing than the evertors and invertors when landing from a jump in a forward direction (20). Results from a previous study investigating ankle brace use and muscle activity suggests that the tibialis anterior may aid in maintaining the position of the foot relative to the ground at initial contact (4). Interestingly, few studies examine strength deficits in the gastrocnemius and the anterior tibialis (21) muscles for subjects with FAI. In the studies that have examined strength in these muscles (21), no deficits have been found, a possible explanation for the lack of differences in vertical TTS as a measure of FAI.
TTS scores were longer (greater) in the medial/lateral and anterior/posterior directions after performing the step down protocol. Multiple main effects were observed for the calculations of the medial/lateral and anterior/posterior TTS scores. The scores for both directions were significantly higher during the step test regardless of analysis and with the UTOP method of analysis regardless of protocol. Whereas both protocols are more dynamic than previously mentioned measures of neuromuscular control (13,21), the degree of difficulty in completing the jump protocol is higher because it requires greater eccentric strength, coordination, and stability of the ankle joint. The apparent ease of the step down test as described by the self-report of perception of difficulty may be responsible for the longer TTS scores. McKinely and Pedotti (20) noted that the subjects with the best ability to stabilize during a jump-landing protocol have greater and earlier cocontraction around the ankle joint before landing. This muscle cocontraction created greater muscle stiffness in the ankle joint and allowed faster reactions to the landing surface. Subject self-reported perception of difficulty during the step down protocol was much lower than during the hop test. With the perceived ease of the protocol, a subject’s muscle activation may have been delayed or even failed to exist in the prelanding timeframe. McKinely and Pedotti (20) associate an anticipatory contraction of the lower leg musculature to a decrease in reaction time while landing from a jump. This could explain the higher TTS scores during the step down test; however, this is only speculative, as EMG data were not collected during this investigation.
The longer TTS scores by method of analysis were partially expected due to a fundamental difference in technique. Sequential estimation is a running average and thus acts as a smoothing technique. This technique is then compared with the overall average of the 3-s collection period. Because the smoothing is compared to the raw data average, it is only reasonable to assume that stabilization will occur more quickly than when calculated by the method used by Ross and Guskiewicz (25). Ross and Guskiewicz (25) compare a line of best fit from the raw data against an average, independent of the jump landing. Their time of collection, 20 s, is also significantly longer than the 3-s collection period. This increased number of data points will prolong the intersection of the GRF curves and baseline measures, which determine the point of stability.
SUMMARY AND CONCLUSIONS
Based upon the results of this investigation and review of the literature, we believe that when comparing FAI with healthy ankles, a jump protocol will be more able to successfully detect differences in dynamic stability than a step down protocol. We also conclude that the UTOP method of analysis is better suited to detect these differences, as it does not smooth the data and therefore potentially mask deficits in regards to FAI. We are confident in our conclusion despite the lack of group differences for three main reasons.
First, the higher GRF exhibited during the jump protocol more closely mimic those seen during athletic activity and are an integral part of the TTS calculation. Additionally, subjects reported that the jump protocol, which more closely resembles an athletic activity, was more difficult to complete than the step protocol, which mimics daily activities. Therefore, the jump protocol should be used whenever possible, to better study the mechanisms of injury, as they would occur during athletic events. Second, TTS varied greatly based on protocol and method of analysis. However, with statistically significant main effects for each (e.g., protocol and analysis) in all directions (e.g., vertical, medial/lateral, and anterior/posterior), group main effects were potentially masked because the protocol and analysis scores were collapsed to determine group means. Third and most importantly, we are confident in our conclusions because of the three-way interaction noted with anterior/posterior TTS. This interaction indicates that deficits in dynamic stability can be detected in subjects with FAI through the TTS measure.
Therefore, we recommend that researchers and clinicians utilize the jump protocol and UTOP method of analysis in order to detect dynamic stability deficits in individuals with FAI.
1. Birmingham, T. Test-retest reliability of lower extremity functional instability measures. Clin. J. Sports Med.
2. Caulfield, B., and M. Garrett. Functional instability of the ankle: Differences in patterns of ankle and knee movement prior to and post landing in a single leg jump. Int. J. Sports Med.
3. Colby, S., R. Hintermeister, M. Torry, and J. Steadman. Lower limb stability with ACL impairment. J. Orthop. Sports Phys. Ther.
4. Cordova, M. L., C. W. Armstrong, J. M. Rankin, and R. A. Yeasting. Ground reaction forces and EMG activity with ankle bracing during inversion stress. Med. Sci. Sports Exerc.
5. Dufek, J., and B. Bates. Biomechanical factors associated with injury during landing in jump sports. Sports Med.
6. Freeman, M., M. Dean, and I. Hanham. The etiology and prevention of functional instability of the foot. J. Bone Joint Surg. Br.
7. Garrick, J. The frequency of injury, mechanism of injury, and epidemiology of ankle sprains. Am. J. Sports Med.
8. Gerber, J., G. Williams, C. Scoville, et al. Persistent disability associated with ankle sprains: a prospective examination of an athletic population. Foot Ankle Int.
9. Goldie, P., T. Bach, and O. Evans. Force platform measures for evaluating postural control: reliability and validity. Arch. Phys. Med. Rehabil.
10. Harukazu, T., K. Yasuda, Y. Ohkoshi, B. Beynon, and P. Renstrom. Anterior drawer test for acute anterior talofibular ligament injuries of the ankle: how much load should be applied during the test? Am. J. Sports Med.
11. Hoffman, M., and V. Payne. The effects of proprioceptive ankle disk training on healthy subjects. J. Orthop. Sports Phys. Ther.
12. Hopkins, T., and R. Palmieri. Effects of ankle joint effusion on lower leg function. Clin. J. Sports Med.
13. Hubbard, T., and T. Kaminski. Kinesthia is not affected by FAI status. J. Athl. Train.
14. Kannus, P., and P. Renstrom. Treatment for acute tears of the lateral ligaments of the ankle: operation, cast, or early controlled mobilization. J. Bone Joint Surg.
15. Konradsen, L., and J. Ravn. Ankle instability caused by prolonged peroneal reaction time. Acta Orthop. Scand.
16. Konradsen, L. Factors contributing to chronic ankle instability: kinesthesia and joint position sense. J. Athl. Train.
17. Kovács, I., J. Tihanyi, P. Devita, L. Rácz, J. Barrier, and T. Hortobágyi. Foot placement modifies kinematics and kinetics during drop jumping. Med. Sci. Sports Exerc.
18. McGuine, T., J. Greene, T. Best, and G. Leverson. Balance as a predictor of ankle injuries in high school basketball players. Clin. J. Sports Med.
19. McKay, G., P. Goldie, W. Payne, and B. Oaks. Ankle injuries in basketball: injury rate and risk factors. Br. J. Sports Med.
20. McKinely, P., and A. Pedotti. Motor strategies in landing from a jump: the role of skill in task execution. Exp. Brain Res.
21. McNight, C., and C. Armstrong. The role of ankle strength in functional ankle instability. J. Sport Rehabil.
22. Nitz, A., J. Dobner, and D. Kersey. Nerve injuries and grades II and III ankles sprains. Am. J. Sports Med.
23. Pinstaar, A., J. Brynhildsen, and H. Troop. Postural corrections after standardized perturbations of single limb stance; effect of training and orthotic devices in patients with ankle instability. Br. J. Sports Med.
24. Reimann, B., N. Caggiano, and S. Lephart. Examination of a clinical method of assessing postural control during a functional performance task. J. Sport Rehabil.
25. Ross, S., and K. Guskiewicz. Time to stabilization
: a method for analyzing dynamic postural stability
. Athletic Therapy Today
26. Tropp, H., P. Odenrick, and J. Gillquist. Stabilometry recordings in functional and mechanical instability of the ankle joint. Int. J. Sports Med.
27. Vaes, P., W. Duquet, and B. Gheluwe. Peroneal reaction time and eversion motor response in healthy and unstable ankles. J. Athl. Train.
28. Wikstrom, E., M. Powers, and M. Tillman. Dynamic stabilization time following isokinetic and functional fatigue. J. Athl. Train.
29. Yeung, M., K. Chan, C. So, and W. Yuan. An epidemiological survey on ankle sprain. Br. J. Sports Med.