Journal Logo

Clinical Sciences: Clinically Relevant

Ground reaction forces and EMG activity with ankle bracing during inversion stress


Author Information
Medicine& Science in Sports & Exercise: September 1998 - Volume 30 - Issue 9 - p 1363-1370
  • Free


Ankle sprains have been identified as the most common injury occurring during athletic participation (7,10,11). Because of the frequency of ankle injuries, a considerable amount of epidemiological research has been conducted to examine the causes as well as the effects of various methods used to prevent such injuries (7-10,31-33,37). It has been shown that both prophylactic ankle braces and taping appear to be useful in reducing the incidence of ankle injury (3,7,9,10,31,33,36). Numerous studies have demonstrated that external support of the ankle, such as that provided by bracing and taping, can reduce passive inversion of the foot (12-16,25,27). Because of their mechanical effects, it has been proposed that these appliances reduce stress placed on the surrounding soft tissue (31-33,37). This suggests that bracing and taping may act to attenuate the external inversion forces that are applied to the foot. However, no empirical data exists to support this theory.

The external forces placed on the foot and ankle are absorbed by the passive structures (tendon, ligament, fascia, bone) and the contractile tissues(muscle) surrounding these joints. Independent of any external mechanical support provided by ankle bracing, the musculature controlling the foot and ankle act to provide a dynamic restraint. In particular, the peroneus longus muscle is most important in resisting an inversion moment at the foot (2). Evaluation of peroneus longus latency or reaction time has been done comparing functionally unstable and normal ankles through the use of an inversion platform under a quasi-static position (4,18,19,21,22,26). Some of these studies have concluded that functional instability of the ankle results partly from mechanical instability and prolonged peroneal reaction time (21,22). Conversely, others have reported no such deficit in peroneal latency between healthy and unstable ankles (4,18,19,26). A significant limitation common to all of the aforementioned studies is that none involved assessment of peroneal muscle activity during a dynamic lateral movement that is similar to that which often precipitates ankle injury.

The combined role of external ankle support and peroneus longus activity, whether under a simulated ankle sprain (20,34) or during a dynamic movement (35), has not been well studied. Furthermore, the specific role of other leg muscles controlling the foot and ankle (i.e., tibialis anterior and gastrocnemius) during a dynamic task remains unquestioned. No known published data exists demonstrating the EMG amplitude of the peroneus longus, or any other leg muscle, present in a braced condition compared with an unbraced condition during a well-controlled lateral dynamic task. Additionally, knowledge of the EMG activity of selected muscles controlling the foot and ankle at specific loading points during a lateral dynamic activity would provide insight into the relationship between external prophylactic restraint and dynamic support provided by selected leg muscles. Simultaneously analyzing ground reaction forces and EMG activity of the peroneus longus, tibialis anterior, and medial gastrocnemius during a dynamic lateral movement may provide greater insight into the efficacy of prophylactic ankle bracing. Thus, the purposes of this study were to evaluate the effects of external ankle support on the ground reaction forces and myoelectrical activity of selected lower extremity muscles at specific loads of force during dynamic inversion stress.


Subjects and design

Twenty-four physically active, healthy male college students (age= 23.3 ± 3.4 yr, height = 184.0 ± 7.4 cm, and mass = 92.5 ± 14.3 kg) participated in this study. Subjects had incurred no ankle trauma within the past 2 yr and were free from any trouble with vision, inner ear problems, neuromuscular dysfunction, or any orthopaedic condition that causes difficulty with running. They were accepted to participate after giving informed written consent in accordance with the policy statement set forth by the American College of Sports Medicine. Subjects reported to the laboratory on two separate occasions. The first day consisted of an orientation session in which the subject became familiar with the equipment used in the study as well as the movement pattern to be performed during the lateral shuffling task. On the second day, subjects were tested under three ankle brace conditions while performing a lateral shuffling movement.


Two separate data acquisition systems were used in this study. A standard strain gauge force platform measured ground reaction forces in three orthogonal planes. A telemetered electromyography (EMG) system recorded the electrical activity of three lower extremity muscles. Each device delivered an analog signal (voltage) which was then converted into a digital signal via an analog-to-digital converter board positioned in a microcomputer. The separate signals were time matched through a voltage spike sent from the force plate amplifier to the EMG system at the beginning of force plate sampling. This provided a time base common to both data sets.

Force platform dynamometry. An A.M.T.I. strain gauge force platform (model OR5-1, Newton, MA) centered in a walkway platform was used to gather kinetic data. Six raw voltage signals were sampled (500 Hz), electronically processed and amplified (gain set at 2000, frequency response set at 1050 Hz) by an A.M.T.I. Signal Conditioner (model SGA6-1). The amplifier was interfaced through a 12-bit analog-to-digital converter (Data Translations Inc., model 2801, Marlborough, MA) positioned in an IBM microcomputer to provide three orthogonal ground reaction force components (Fx, Fy, Fz). Customized software digitally filtered the raw data (fourth order Butterworth lowpass filter with a cut-off set at 6 Hz) and determined peak impact force, maximum loading force, and maximum propulsion force along the mediolateral axis. These kinetic variables were normalized to a percentage of one's body weight and interpolated to 200 points, representing 100% of the contact time on the force plate.

Electromyography. An eight-channel telemetered EMG system (Noraxon USA, Telemyo System; Scottsdale, AZ) recorded the electrical activity of the three selected muscles. Three channels of the system were used to record the activity of the peroneus longus, tibialis anterior, and medial gastrocnemius. A fourth channel was used to record an analog signal coming from the force platform amplifier. This signal identified the initial time of contact on the force plate, which allowed for synchronization of the EMG and ground reaction force data. A standard bipolar electrode arrangement was used. EMG signals were transmitted to a receiver interfaced to a controlling desktop computer. These signals were digitally converted at 500 Hz and processed by the accompanying Myosoft software (Noraxon USA) using a linear envelope detector. The linear envelope detector rectified the raw signal and low-pass filtered it with a cutoff set at 10 Hz. The cutoff frequency was determined by performing a spectral analysis on the raw EMG data (39). Custom software was used to ensemble the five trials for each condition and interpolate the processed EMG data for each of the four channels to 100 points, representative of 100% of the contact time on the force plate. The averaged and interpolated data were normalized to the average EMG activity for each muscle and represented a percentage of the average EMG activity produced for the control condition. This normalization process was performed to help reduce intersubject variability (40). The normalized EMG values of each muscle were time matched to the ground reaction force data and statistically analyzed.

Experimental Protocol

Orientation session. Potential subjects were first screened to determine whether they fit the necessary inclusion criteria. Subjects were then asked to read and sign the informed consent form after all questions had been answered. The equipment used in the study was demonstrated and explained to all subjects. Immediately following, subjects practiced the lateral shuffling movement at their desired speed until they felt comfortable with the movement pattern. This movement required a direction change in which subjects accelerated from an established starting point to the force platform and back to the original starting position. The lateral shuffling movement allowed for dynamic foot inversion during force plate contact. Subjects performed five maximal shuffles and the average speed was recorded. The orientation session terminated after a convenient time for the test session had been scheduled.

Testing session. Before any testing took place, each subject thoroughly warmed up and stretched. Subjects were then prepared for electrode placement; this involved removing hair from the areas of electrode placement followed by cleansing the area with a rubbing alcohol solution on a mild abrasive pad. Self-adhesive gel electrodes were placed over the muscle bellies of the peroneus longus, tibialis anterior, and medial gastrocnemius of the right leg as previously described (6). A reference electrode was placed over the right lateral femoral epicondyle.

All subjects were tested under three ankle brace conditions: 1) Control-no brace; 2) Aircast Sport-Stirrup brace (Aircast Inc, Summit, NJ); and 3) Active Ankle Training brace (Active Ankle Systems, Inc, Louisville, KY) in a balanced random order. Each subject completed five to seven lateral shuffles between 80-90% of their maximal shuffle speed before striking the force platform. Shuffling speed was controlled with the use of two photocells. The first photocell was positioned 3 m from the initial starting position. The second photocell was positioned 1 m ahead of the first and 1 m from the force platform (Fig. 1). Subjects were instructed to contact the force platform such that their foot remained flat and perpendicular to the mediolateral axis. Force plate sampling started when the subject contacted the plate; this ensured consistency for all trials. Following initial foot contact, subjects were encouraged to rapidly accelerate off the force platform toward the original starting point. Five trials were performed for each condition, and subjects received a 10-min rest between conditions. Subjects were tested no later than one week following the orientation session.

Figure 1
Figure 1:
Schematic representation of experimental setup and coordinate system used. The force platform(FP) is located 5 m from the starting point. Photo-cell 1 (PC1) is located 3 m from the starting point, while photo-cell 2 (PC2) is located 1 m forward of the FP. A 1 m distance separates PC1 from PC2.

Statistical Analyses

A doubly multivariate design was employed in this study. The single within-subjects factor was ankle brace with three levels (control-no brace, Aircast Sport-Stirrup brace, Active Ankle brace). The dependent variables associated with the ground reaction forces were: lateral peak impact force (LPIF), lateral maximum loading force (LMLF), and lateral peak propulsion force (LPPF). The dependent variables affiliated with the EMG data were: average EMG activity of the peroneus longus, tibialis anterior, and medial gastrocnemius at the point of LPIF, LMLF, and LPPF. Five trials of each subject for each condition were averaged for each dependent variable and used for statistical analysis.

Two separate one-way repeated measures multivariate ANOVA(MANOVA) models were analyzed. The first model determined whether differences existed across ankle brace on the linear combination of lateral peak impact force, lateral maximum loading force, and lateral peak propulsion force. A second repeated measures MANOVA analysis was performed to see whether differences existed across ankle brace on the linear combination of average EMG activity of the peroneus longus, tibialis anterior, and medial gastrocnemius at the point of LPIF, LMLF, and LPPF. UnivariateF-tests and Tukey multiple comparison procedures were used post hoc to locate specific group differences. The level of significance was set a priori at P ≤ 0.05.


Ground Reaction Forces

Amplitude values for the ground reaction force (GRF) data for each brace condition are presented in Table 1. The GRF time-history for each brace condition is illustrated in Figure 2. There was no overall multivariate effect of the brace condition on the linear combination of the GRF variables (Wilks' Lambda F (8,86) = 1.07, P > 0.05). When considered univariately, there was no effect of brace on LPIF (F(2,46) = 1.57, P > 0.05); LMLF (F (2,46) = 0.373, P> 0.05); and LPPF (F (2,46) = 1.60, P > 0.05).

Summary of ground reaction force variables by condition (mean± SD).
Figure 2
Figure 2:
Average ground reaction force curve for each condition.


Values for each EMG variable under all brace conditions are presented in Table 2. The EMG curves of each muscle for each condition are also graphically displayed in Figures 3-5. There was no overall multivariate effect attributed to the brace condition on the average EMG variables when they were considered linearly (Wilks' Lambda F (18,76) = 1.01, P > 0.05). Single univariate F-tests revealed a significant main effect for brace on average EMG activity of the peroneus longus at LPIF (F (2,46) = 7.4, P< 0.05). Tukey multiple comparison procedure showed reduced EMG activity in the Aircast and Active Ankle brace conditions compared with the control, but no difference existed between either brace condition (Fig. 6). No differences among brace conditions were found for peroneus longus muscle activity at the point of LMLF (F (2,46) = 0.401, P > 0.05) and LPPF (F(2,46) = 0.626, P > 0.05). With respect to the other lower extremity muscles, EMG activity of the tibialis anterior at the point of LPIF (F(2,46) = 0.293, P > 0.05); LMLF (F (2,46) = 0.548, P > 0.05); and LPPF (F (2,46) = 0.380, P > 0.05) did not differ among brace conditions (Fig. 7). The EMG activity of the medial gastrocnemius did not differ among brace conditions at the time of LPIF (F (2,46) = 0.604, P > 0.05); LMLF (F(2,46) = 0.547, P > 0.05); and LPPF (F (2,46) = 0.272,P > 0.05), as well (Fig. 8).

Summary of average EMG for each muscle by condition (Mean ± SD).
Figure 3
Figure 3:
Average EMG curve for each muscle during the control condition.
Figure 4
Figure 4:
Average EMG curve for each muscle during the Aircast condition.
Figure 5
Figure 5:
Average EMG curve for each muscle during the Active ankle condition.
Figure 6
Figure 6:
Effect of different ankle braces on EMG activity of the peroneus longus at each force.
Figure 7
Figure 7:
Effect of different ankle braces on EMG activity of the tibialis anterior at each force. There were no statistical differences between the control and brace conditions.
Figure 8
Figure 8:
Effect of different ankle braces on EMG activity of the medial gastrocnemius at each force. There were no statistical differences between the control and brace conditions.


Ground Reaction Forces

General GRF characteristics. The force curve time-histories displayed in Figure 2 are very similar in shape to those reported by Luethi et al. (23) who evaluated the effects of two different tennis shoes on lower extremity kinematics and mechanical loading. The GRF curves for each condition demonstrated comparable patterns, in that three distinct peaks existed in each. The initial sharp rise to the first peak represents the impact peak. Following the impact peak, a small decline occurs, which is immediately followed by a second peak termed "maximum loading force." The subsequent fall in the maximum loading force is later followed by the peak propulsion force.

Ankle bracing and GRF variables. LPIF was defined as the greatest amount of force present during the first 10% of contact time on the force platform. This value essentially represents the maximum impact force at foot contact. The average magnitude of peak impact force for each condition ranged from 0.92 to 0.99 BW. These values were the largest with respect to the other force variables measured (Table 1). Unlike the hypothesis, neither ankle brace altered LPIF compared with the control condition. This observation is clinically relevant as it suggests that these semirigid ankle stabilizers do not aid in absorbing force placed on the foot and ankle immediately following contact. This result is not in agreement with previously reported data (23,35). Stuessi et al. (35) showed the Aircast brace to yield a smaller average lateral peak force produced over the entire contact time compared with the control condition in 7 of 11 total subjects. It should be noted that their subjects did not perform a lateral shuffling movement, but rather a specially designed running shoe was fabricated to create a supinatory movement at the foot. The reduction in lateral force was attributed to a decrease in the inversion angle and mediolateral velocity at touchdown. In a similar study(23), the effects of a flexible and rigid tennis shoe on vertical and lateral ground reaction forces during a lateral movement were compared. The data showed that subjects wearing the flexible shoe had higher average maximum forces and a larger inversion angle compared with those wearing the rigid shoe. Although this study did not evaluate the role of external ankle support, these results may help to support the theory that mechanical support offered by rigid ankle braces may aid in attenuating applied forces.

LMLF describes the greatest amount of force as the foot and ankle are loaded after initial contact. This value was defined to occur between 11-30% of total contact. Among the three ankle brace conditions evaluated in this study, virtually no change in maximum loading force occurred (Fig. 9). Based on the interpretation of the peak impact force, it appears that the semirigid ankle braces do not absorb external inversion forces early in stance. Therefore, these forces may be absorbed and transferred to the supportive elements of the lateral foot and ankle (muscle, connective tissue, bone) and possibly up the lower extremity kinetic chain to the knee. It has been advocated that the major function of a brace is to restrict foot inversion and not act as a force bypass during loading (1). The fact that ankle bracing failed to significantly reduce peak impact and maximum loading force experienced at the foot and ankle may have several explanations. The first concerns the relationship between the mechanical property of the brace and the range of motion present in the subtalar joint. Under the control condition, total inversion range of motion during the lateral movement would significantly increase if the muscles controlling inversion could not limit this motion. Increasing the joint motion would create a longer period of time for the external forces to be absorbed by the bones, muscles, tendons, and ligaments of the foot and ankle. This would result in overloading these structures with the large internal forces created (23). Conversely, severely decreasing the range of motion at a joint through the use of a rigid ankle brace would limit the time period for the external forces to act over the available joint motion. Given this situation, insufficient joint range of motion would lead to large external forces being absorbed by associated foot and ankle structures. Furthermore, greater compressive forces would be created and may lead to a greater potential for injury.

Figure 9
Figure 9:
Effect of different ankle braces on lateral ground reaction force variables. There were no statistical differences between the control and experimental conditions.

Another explanation as to why no significant differences existed between levels of brace on peak impact and maximum loading force may be the result of insufficient inversion stress created during this movement pattern. In this controlled clinical trial, it was felt that subjects shuffling between 80-90% of their maximum speed would be sufficient enough to produce a large inversion moment. Although specific joint moments were not measured in this study, it is possible that not enough force was created, thereby not stressing the mechanical stiffness of the braces. It has been suggested that an inversion moment up to 420 Nm can be applied to the foot before the lateral structures fail (30). Perhaps the mechanical effect of the braces would have been demonstrated if a greater load had been applied. Data concerning the stress-strain relationship of semirigid ankle braces under a controlled load are lacking. However, it is reasonable to expect that either of the two semirigid braces that were tested would exhibit a greater yield strength compared with adhesive tape. Furthermore, it may be speculated that the external forces created within this system (combined foot/ankle structures and brace) acted within a safety boundary resulting in no apparent force attenuation.

There is little argument that ankle braces limit motion. However, the efficiency with which these braces achieve this may be under question. The implication that semirigid ankle braces may be too rigid and lack sufficient stretching properties certainly exists. It emerges that a compromise between too little and too much joint range of motion needs to exist in an effort to reduce the load placed on the supporting structures. Anderson et al. (1) reported that the inversion angle, time to inversion, and calcaneal inversion velocity significantly decreased during sudden inversion with a soft nonrigid ankle brace compared with a control condition. They concluded that strain of the lateral structures of the foot and ankle may be reduced by increasing the time of inversion and decreasing the rate at which the calcaneus inverts. Based on their data, it appears that a softer external ankle support creates a large enough impulse by lengthening the time at which the external force acts, thereby decreasing the momentum of the foot as it inverts.

LPPF identified the highest amount of force that occurred during the last 70% of contact time on the force plate; in other words, the most amount of force produced when accelerating from the force plate. It was hypothesized that bracing would restrict the subject's ability to accelerate off the force platform as they returned to the starting position, thereby reducing the propulsion force. However, no significant differences existed between the ankle brace conditions on peak propulsion force (Fig. 9). This result is advantageous, as it suggests that the Aircast and Active Ankle braces do not impair functional ability. In the past, concern has been expressed about the negative influence of bracing on functional performance. Under the conditions of the present study, this did not occur. This result does support previously reported data that these semirigid ankle stabilizers do not hinder functional performance (5,14,28,29,38), although some studies have reported an inhibitory effect of such braces (12,24).


Ankle bracing and peroneus longus EMG. Ankle bracing reduced the average EMG activity of the peroneus longus at the point of lateral peak impact force compared with the control condition(Fig. 6). However, there was no significant difference between the two braces. The fact that bracing had an effect on peroneus longus activity at peak impact force can be interpreted two ways. The first interpretation suggests that external support reduces the strain or load that is placed on the muscle that dynamically limits forced inversion of the foot. Although bracing did not alter LPIF, it may be speculated that bracing does in fact act to reduce the impact force placed on the soft tissue surrounding the foot and ankle as evidenced by decreased EMG activity. Since a positive correlation exists between the amount of electrical activity in a muscle and the tension developed in that muscle (17), it is reasonable to consider that decreased peroneus longus EMG activity is indicative of force attenuation by the brace. Conversely, it may be viewed that limiting the activity of the peroneus longus through the use of a rigid ankle brace is detrimental because the muscle used to control forced inversion is being inhibited. Not enough data exists to support either conclusion.

No effect of ankle bracing was shown on peroneus longus activity at the point of maximum loading force, in contrast to the hypothesis (Fig. 6). With respect to each brace condition, maximum loading force occurred between 22-30% of total contact. The fact that peroneus longus activity was reduced during the first 10% of foot contact (LPIF) with bracing and not between 11-30% (LMLF) possibly suggests that bracing may only be effective early in supporting the peroneus longus and stabilizing the foot against forced inversion. Following initial impact, the residual force may be absorbed by the peroneus longus and passive structures of the foot and ankle. No differences were observed between all levels of brace condition on peroneus longus activity at the point of peak propulsion force (Fig. 6). At this point during the lateral movement, acceleration from the force plate is occurring and the role of the peroneus longus has changed from eccentrically controlling foot inversion to assisting in producing an ankle plantar flexion and a foot eversion moment. The absence of a brace effect on peroneus longus activity at this point during contact is important, as unrestricted function of the muscle is maintained.

Ankle bracing and tibialis anterior EMG. Average EMG activity of the tibialis anterior during peak impact force, maximum loading force, and peak propulsion force remained unchanged under the three brace conditions(Fig. 7). The primary role of the tibialis anterior muscle is to dorsiflex the ankle and control plantar flexion. Its secondary role is to assist with inversion of the foot. The mechanical objective of an ankle brace is to restrict frontal plane motion and not sagittal plane motion; as a result, it would be expected that an ankle brace would not alter the EMG activity of this muscle. The role of the tibialis anterior in attenuating force during a lateral dynamic movement appears limited. However, this muscle may aid in maintaining the position of the foot relative to the ground at initial contact.

Ankle bracing and medial gastrocnemius EMG. Average EMG activity of the medial gastrocnemius at the point of peak impact force, maximum loading force, and peak propulsion force remained unaffected by bracing as well (Fig. 8). The medial gastrocnemius is responsible for plantar flexing the ankle through its physical attachment to the calcaneus via the Achilles tendon. However, it also functions to assist the triplanar supinatory motion at the foot and ankle. As stated earlier, the brace functions to limit subtalar joint motion; as a result, one would not expect to see changes in this muscle's activity during this task. Based on interpretation of the EMG curve, it appears that the medial gastrocnemius is most active during the propulsion phase of this dynamic movement pattern. The fact that bracing did not affect the peak propulsion force, a force predominantly produced by this muscle, and the EMG activity at this specific point implies that its normal function remains unaltered. These data indicate that ankle bracing does not affect the function of the surrounding musculature that controls sagittal plane joint movement.


The data presented in this investigation represented an attempt to evaluate the role of ankle bracing on the external forces created on the foot and ankle along with the electrical activity of selected lower extremity muscles during a dynamic inversion task. Overall, bracing did not appear to affect peak impact force, maximum loading force, or peak propulsion force. With respect to the electrical activity of the muscles, bracing reduced the EMG activity of the peroneus longus during peak impact force. However, bracing did not alter the EMG activity of the peroneus longus at the point of maximum loading force and peak propulsion force. Furthermore, bracing did not alter the EMG activity of the tibialis anterior or medial gastrocnemius during peak impact force, maximum loading force, and peak propulsion force. Because this study is unique, little data exist in terms of supporting or refuting these findings. Understanding the complex relationship between ankle bracing and EMG activity of selected musculature during dynamic loading warrants further study.


1. Anderson, D. L., D. J. Sanderson, and E. M. Hennig. The role of external nonrigid ankle bracing in limiting ankle inversion. Clin. J. Sport Med. 5:18-24, 1995.
2. Anderson, M. K. and S. J. Hall.Sports Injury Management. Baltimore: Williams & Wilkins, 1995, pp. 243-244.
3. Bahr, R., R. Karlsen, O. Lian, and R. V. Ovrebo. Incidence and mechanisms of acute ankle inversion injuries in volleyball: a retrospective cohort study. Am. J. Sports Med. 22:595-600, 1994.
4. Beckman, S. M. and T. S. Buchanan. Ankle inversion injury and hypermobility: effect on hip and ankle muscle electromyography onset latency. Arch. Phys. Med. Rehabil. 76:1138-43, 1995.
5. Beriau, M. R., W. B. Cox, and J. Manning. Effects of ankle braces upon agility course performance in high school athletes. J. Athl. Train. 29:224-230, 1994.
6. Delagi, E. F. and A. Perroto. Anatomic Guide for the Electromyographer: The Limbs. 2nd Ed. Springfield, IL: Charles C. Thomas, 1981, pp. 146-159.
7. Garrick, J. G. and R. K. Requa. The epidemiology of foot and ankle injuries in sports. Clin. Sports Med. 7:29-36, 1988.
8.Garrick, J. G. Characterization of the patient population in a sports medicine facility. Physician Sportsmed. 13:4-5, 1985.
9.Garrick, J. G. Epidemiologic Perspective. Clin. Sports Med. 1:13-18, 1982.
10. Garrick, J. G. The frequency of injury, mechanism of injury, and epidemiology of ankle sprains. Am. J. Sports Med. 5:241-242, 1977.
11. Glick, J. M., R. B. Gordon, and D. Nishimoto. The prevention and treatment of ankle injuries. Am. J. Sports Med. 4:136-141, 1976.
12. Greene, T. A. and C. R. Wight. A comparative support evaluation of three ankle orthoses before, during and after exercise. J. Orthop. Sports Phys. Ther. 11:453-466, 1990.
13.Greene, T. A. and S. K. Hillman. Comparison of support provided by a semirigid orthosis and adhesive ankle taping before, during and after exercise. Am. J. Sports Med. 18:498-506, 1990.
14. Gross, M. T., J. R. Everts, S. E. Roberson, D. S. Roskin, and K. D. Young. Effect of Donjoy ankle ligament protector and aircast sport-stirrup orthoses on functional performance.J. Orthop. Sport Phys. Ther. 19:150-156, 1994.
15.Gross, M. T., A. K. Lapp, and J. M. Davis. Comparison of Swede-o-universal ankle support and aircast sport-stirrup orthoses and ankle tape in restricting eversion-inversion before and after exercise. J. Orthop. Sports Phys. Ther. 13:11-19, 1991.
16. Gross, M. T., M. K. Bradshaw, L. C. Ventry, and K. H. Weller. Comparison of support provided by ankle taping and semirigid orthosis. J. Orthop. Sports Phys. Ther. 9:33-39, 1987.
17. Inman, V. T., H. J. Ralston, J. B. Saunder, B. Feinstein, and E. W. Wright. Relationship of human EMG to muscular tension. Electroencephalogr. Clin. Neurophysiol. 4:187-194, 1952.
18. Isakov, E., J. Mizrahi, P. Solzi, Z. Susak, and M. Lotem. Response of the peroneal muscles to sudden inversion of the ankle during standing. Int. J. Sport Biomech. 2:100-109, 1986.
19. Johnson, M. B. and C. L. Johnson. Electromyographic response of peroneal muscles in surgical and nonsurgical injured ankles during sudden inversion.J. Orthop. Sports Phys. Ther. 18:497-501, 1993.
20.Karlsson. J. and G. O. Andreasson. The effect of external ankle support in chronic lateral ankle joint instability: an electromyographic study. Am. J. Sports Med. 20:257-261, 1992.
21. Karlsson, J., L. Peterson, G. O. Andreasson, and C. Hogfors. The unstable ankle: a combined EMG and biomechanical modeling study. Int. J. Sport Biomech. 8:129-144, 1992.
22. Konradsen, L. and J. B. Ravn. Prolonged reaction time in ankle instability. Int. J. Sports Med. 12:290-292, 1994.
23.Luethi, S. M., E. C. Frederick, M. R. Hawes, and B. M. Nigg. Influence of shoe construction on lower extremity kinematics and load during lateral movements in tennis.Int. J. Sport Biomech. 2:166-174, 1986.
24. MacKean, L. C., G. Bell, and R. S. Burnham. Prophylactic ankle bracing vs taping: effects on functional performance in female basketball players. J. Orthop. Sports Phys. Ther. 22:77-81, 1995.
25. Martin, N. and R. A. Harter. Comparison of inversion restraint provided by ankle prophylactic devices before and after exercise. J. Athl. Train. 28:324-329, 1993.
26. Nawoczenski, D. A., M. G. Owen, M. L. Ecker, B. Altman, and M. Epler. Objective evaluation of peroneal response to sudden inversion stress. J. Orthop. Sports Phys. Ther. 7:107-109, 1985.
27. Paris, D. L., V. Vardaxis, and J. Kokkaliaris. Ankle ranges of motion during extended activity periods while taped and braced. J. Athl. Train. 30:223-8, 1995.
28. Paris, D. L. The effects of the Swede-O, New Cross, and McDavid ankle braces and adhesive ankle taping on speed, balance, agility and vertical jump.J. Athl. Train. 27:253-256, 1992.
29. Pienkowski, D., M. McMorrow, R. Shapiro, D. N. Caborn, and J. Stayton. The effect of ankle stabilizers on athletic performance: a randomized prospective study. Am. J. Sports Med. 23:757-762, 1995.
30. Pope, M. H., P. Renstrom, D. Donnermeyer, and S. Morgenstern. A comparison of ankle taping methods. Med. Sci. Sports Exerc. 19:143-147, 1987.
31. Rovere, G. D., T. J. Clarke, C. S. Yates, and K. Burley. Retrospective comparison of taping and ankle stabilizers in preventing ankle injuries. Am. J. Sports Med. 16:228-232, 1988.
32. Sitler, M. R. and M. Horodyski. Effectiveness of prophylactic ankle stabilizers for prevention of ankle injuries. Sports Med. 20:53-57, 1995.
33. Sitler, M. R, J. Ryan, B. Wheeler, et al. The efficacy of a semirigid ankle stabilizer to reduce acute ankle injuries in basketball: a randomized clinical study at West Point. Am. J. Sports Med. 22:454-461, 1994.
34. Sprigings, E. J., J. D. Pelton, and B. R. Brandell. An EMG analysis of the effectiveness of external ankle support during ankle sudden inversion. Can. J. Appl. Sci. 6:72-75, 1981.
35. Stuessi, E., V. Tiegermann, H. Gerber, H. Raemy, and A. Stacoff. A biomechanical study of the stabilization effect of the aircast ankle brace. In: Biomechanics X, B. Jonsson (Ed.). Champaign, IL: Human Kinetics, 1987, pp. 159-164.
36. Surve, I., M. P. Schwellnus, T. Noakes, and C. Lombard. A fivefold reduction in the incidence of recurrent ankle sprains in soccer players using the sport-stirrup orthosis. Am. J. Sports Med. 22:601-606, 1994.
37. Tropp, H., C. Askling, and J. Gillquist. Prevention of ankle sprains. Am. J. Sports Med. 13:259-262, 1985.
38. Verbrugge, J. D. The effects of semirigid air-stirrup bracing vs. adhesive ankle taping on motor performance. J. Orthop. Sports Phys. Ther. 23:320-325, 1996.
39. Winter, D. A. Biomechanics and Motor Control of Human Movement. 2nd Ed. New York, NY: John Wiley & Sons, Inc., 1990, pp. 36-38.
40. Yang, J. F. and D. A. Winter. Electromyographic amplitude normalization methods: improving their sensitivity as diagnostic tools in gait analysis. Arch. Phys. Med. Rehabil. 65:517-521, 1984.


©1998The American College of Sports Medicine