Six-Week Remodeled Bike Pedal Training Improves Dynamic Control of Lateral Shuffling in Athletes With Functional Ankle Instability

Abstract

Background:

Remodeled bicycle pedal training with multidirectional challenges through muscle strengthening and neuromuscular facilitation may increase dynamic postural control and performance during lateral shuffling for athletes with functional ankle instability (FAI).

Hypothesis:

The 6-week remodeled bicycle pedal training is effective on the ankle joint control and muscle activation, and especially that of the ankle evertor muscle co-contraction to improve dynamic postural control during lateral shuffling for athletes with FAI.

Study Design:

Laboratory randomized controlled trial.

Level of Evidence:

Level 2.

Methods:

Fourteen healthy athletes (healthy group) and 26 athletes with FAI aged 18 to 30 years were included in the study. The athletes with FAI were randomly assigned to either the training group (FAI-T group) or the nontraining group (FAI-NT group). The athletes in the FAI-T group underwent 6 weeks of remodeled bicycle pedal training, whereas those in the FAI-NT group did not undergo any intervention. Muscle co-contraction index and muscle activation in the initial contact (IC) and propulsion phases, and ankle joint angle in the IC and propulsion phases were measured during lateral shuffling before and after 6 weeks of training.

Results:

After remodeled bicycle pedal training, the FAI-T group demonstrated greater muscle activation in the hamstring (P = 0.01), greater muscle coactivation of the tibialis anterior (TA) and the peroneus longus (P = 0.01), and greater ankle eversion angle in the IC phase. Significantly greater muscle activation of the TA (P = 0.01), greater coactivation of quadriceps and hamstring (P = 0.03), and a smaller ankle inversion angle (P = 0.04) in the propulsion phase were observed in the FAI-T group after training compared with those in the FAI-NT group.

Conclusion:

Remodeled bicycle pedal training facilitates the TA and peroneus longus activation and the coactivation of the quadriceps and hamstring muscles during lateral shuffling and resulted in enhanced ankle and knee joint stability. In addition, a better ankle movement strategy during a dynamic task can be achieved via a 6-week remodeled pedal training program.

Clinical Relevance:

This remodeled bicycle pedal training can be effective for rehabilitating athletes with FAI to recover lateral dynamic movement capability.

The tearing or stretching of ankle ligaments in the ankle ligament complex is known as ankle sprain.^14,15,31,43 Ankle sprains accounts for 15% to 75% of all sport injuries, and they can be transformed into chronic ankle instability (CAI) and functional disability. ²⁴ CAI can be divided into 2 etiologies: of functional ankle Instability (FAI) and mechanical ankle instability (MAI). ¹⁸ Mechanical instability causes anatomic abnormalities of the ankle, generally associated with ligament laxity, while functional instability resulting from postural defects or tendon and muscle adjustment, usually related to a proprioceptive deficit.^2,18,41

Altered muscle activation patterns and kinematics may be seen in athletes with ankle instability. A study reported that athletes with FAI demonstrate lower muscle activation of the ankle evertor after inversion perturbation in their injured ankle during a dynamic task. ¹⁹ Athletes with CAI exhibited a smaller coactivation index of the tibialis anterior (TA) and the peroneus longus (PL) during a landing task, which implies reduced ankle joint stability. ²³ Altered muscle activation patterns might result in changes of the lower extremity kinematics. Athletes with an unstable ankle demonstrate greater inversion angle and inversion variability than those of healthy people during the prejump period. ⁶ Moreover, the lateral and anterior peak ground-reaction forces occur significantly earlier in athletes with FAI during the postimpact of a single-leg drop ^4,7 and the loading rate in athletes with ankle instability is greater than that in healthy controls. ⁴⁶ As a result, the alterations in kinematics, kinetics, and muscle activation patterns of the jumping could lead to the recurrence of the ankle sprain in athletes with FAI.

Muscle strengthening, balance training, and proprioceptive training have been shown to be effective intervention modules for FAI. ³⁰ It is found that FAI causes reduced muscles strength and delayed muscles response time in the ankle^32,33 and increase of ankle muscle strength may increase functional stability.^22,34 disclosed that muscles strength was significantly lower for those with functional unstable ankles compared with healthy ankle and the isokinetic exercise effectively increase muscles strength in functional unstable ankles. ³⁴ So, ankle muscles co-contraction may be increased through muscles activation after 6-week pedal training, and consequently the ankle stability may be improved. However, current rehabilitation programs focus on only static or relatively low-intensity training. A customized bidirectional bicycle was developed for athletes with ankle instability to facilitate specific muscle contraction to maintain ankle stability during rehabilitation with dynamic movement. ¹⁶ After a 6-week bidirectional bicycle training program, participants with ankle instability spent significant less time in the figure-of-8 running test than the baseline measurement. However, the bidirectional bicycle exercise requires force to be exerted in only the sagittal and frontal planes, which does not reflect practical sport movements.

A remodeled 3-directional pedal has been designed so that the ankle could move freely in 3 planes to simulate ankle instability during dynamic movement. ⁵ Athletes with FAI who underwent 6-week remodeled pedal training demonstrated significant improvements in passive joint position sense and static postural control. ⁵ However, the training effects of the remodeled pedals on muscle activation patterns and dynamic control for athletes with FAI were not examined. Hence, the aim of the present study was to investigate the effects of remodeled bicycle pedal training on ankle joint kinematics and muscle activation during dynamic movement for athletes with FAI. It was hypothesized that ankle inversion angle would decrease muscle activation, especially that of the ankle evertor, and the magnitude of muscle co-contraction would increase to improve ankle stability during dynamic movement after 6-week remodeled bicycle pedal training.

Methods

Participants

This study recruited 42 athletes aged from 18 to 30 years. Among them, 28 of them had FAI and the remaining 14 athletes were healthy (H group). The healthy group reported no musculoskeletal dysfunctions in the lower extremities and their self-reported Cumberland Ankle Instability Tool (CAIT) score was 28 or higher.^26,27

The inclusion criteria for athletes with FAI were as follows. The subject had (1) a history of at least 1 acute ankle inversion sprain that caused pain and led to protected weight bearing/dysfunction of the injured ankle in the 6 months before this investigation^3,27; (2) experienced ankle giving way in the last 3 months before this investigation; (3) at least 1 recurrent ankle sprain in the last 3 months before this investigation^3,17; (4) a self-reported CAIT score of less than 24, indicating severe ankle instability ⁹ ; and (5) negative clinical anterior drawer and talar tilt tests. The exclusion criteria for all groups were as follows. The subject had (1) any history of lower extremity fractures or serious orthopaedic injury that would affect performance, (2) any neurological disorder, (3) any acute inflammation at the ankle joint, (4) any history of equilibrium and balance control disorder, and (5) any head injury at the start of this investigation. Recruited athletes exercised on a regular basis (at least 1-2 hours a day, 2-3 times a week) and engaged in sports that involve lateral shuffling.

Athletes with FAI were randomly assigned to either the training group (FAI-T group) or the nontraining group (FAI-NT group). Participants in the FAI-T group underwent 6-week remodeled bicycle pedal training, whereas those in the FAI-NT group received no intervention. Athletes in the FAI-T group who were unable to complete at least 80% of the training program would be excluded from the study. Athletes in the FAI-NT group who did not regularly exercise (at least 1 hour a day and 2 days a week based on their training record and a daily logbook, respectively) would be excluded from the study. Two participants in total were excluded from this study. One participant in the FAI-T group was unable to meet the requirement of at least 80% of the training program and 1 participant in the FAI-NT had national sports competition that hindered him from postassessment and data collection. Therefore, the total number for FAI-T, FAI-NT, and H groups were 13, 13, and 14, respectively.

Before data collection, the participants were informed of the experimental methods and signed informed consent forms. The study procedures and consent forms were approved by the Institutional Review Board of National Cheng Kung University Hospital (IRB No. B-ER-101-172).

Instrumentation

To imitate the perturbations experienced by an athlete during exercise, a remodeled bicycle pedal was designed to tilt sideways in a range of 0° to 40° (with 20° of inversion and 20° of eversion) during the load cycle. The remodeled bicycle pedal moves freely in 3 dimensions to simulate ankle instability and induce the activation of the PL during remodeled pedal cycling. ⁵

A surface electromyography (sEMG) system (BTS FREEEMG 300, BTS Bioengineering) was used to record muscle activity during tasks at a sampling rate of 1000 Hz. The sEMG electrodes were placed on the muscle belly of the TA, PL, gastrocnemius medialis (GM), gastrocnemius lateralis (GL), biceps femoris (BF), semitendinosus (ST), rectus femoris (RF), vastus intermedius (VI), vastus medialis (VM), and vastus lateralis (VL). To minimize skin impedance, the skin was shaved and cleaned with alcohol. The sEMG electrodes were placed on the affected side in the FAI-T and FAI-NT groups and on the dominant leg in the control group. The dominant leg was determined as the leg used for kicking a ball for a maximum distance with the greatest amount of force. ⁸

The 8 infrared digital cameras of a VICON motion capture system (Oxford Metrics) were used to record the real-time 3-dimensional trajectory of reflective markers at a sampling rate of 200 Hz. A total of 44 lightweight reflective markers from a modified Helen-Hayes marker set were attached to the bony landmarks of each participant. All participants were requested to wear specific shoes (JUMP, Lu-Tung Corporation) that had small holes to enable the placement of markers directly on the bony landmarks of the midpoints of the first and fifth metatarsal heads and posterior heel.

The motion capture system and sEMG system were time-synchronized during the lateral shuffling task.

Lateral Shuffling Task

The lateral shuffling task (Figure 1) consisted of (1) knees flexing approximately 60°, with feet shoulder-width apart and trunk erect, and body weight balanced between feet; (2) shifting body weight toward unaffected limb of FAI-T and FAI-NT groups or nondominant limb of H group, picking up affected limb and then explosively pushing affected foot against ground to initiate sideway moving; (3) maintaining the body posture and quick pace throughout the drill. Data of the affected limb (FAI-T and FAI-NT groups) or dominant limb (H group) were collected from the initial contact (IC) phase through the propulsion phase in a given trial.

Figure 1.

The lateral shuffling task.

Procedures

First, investigators placed the sEMG electrodes on the affected leg (FAI-T group and FAI-NT group) or dominant leg (H group) and then recorded the maximum voluntary isometric contraction (MVIC) for each muscle site. Second, the reflective markers were placed on each participant, who then performed the lateral shuffling task five times. After the preassessment, the FAI-T group received 6-week remodeled bicycle pedal training and the FAI-NT and the H group received no additional training but maintained their regular daily training. All participants in the FAI-T and FAI-NT groups conducted a postassessment after 6 weeks. The healthy group conducted only a preassessment. A 10-minute warm-up session was conducted before data collection. Three practices with maximum effort of lateral shuffling task were held to familiarize participants with the procedure of data collection in the laboratory setting. Overall, data from 3 successful trials with maximum effort were collected. A rest period of at least 30 seconds was given between trials.

Training Program

The training intensity was evaluated by monitoring the heart rate. The training session commenced at a bicycle pedal rate of 50 to 70 rpm ¹⁶ to reach the target training intensity. Each training session started with a 5-minute warm-up period at 50% of the target heart rate. Participants exercised with low resistance (level 1 of the bicycle) during cycling for 1 minute and the resistance was then increased to 0.077 body weight of the corresponding participant ⁴⁰ for a further 3 minutes and then returned to level 1. The training program included 3 sessions per week with 40 minutes per session. ⁵

Data Analysis

The raw sEMG signals were processed with a bandpass filter at 40 to 400 Hz with a fourth-order Butterworth filter and a full-wave rectified. The average muscle activation of the BF and ST represented the hamstring (HAM). The average muscle activation of the RF, VI, VM, and VL represented the quadriceps (QUAD). The average muscle activation of the GM and GL represented the gastrocnemius (GAS). The sEMG data were normalized to the corresponding MVIC to assess the corresponding level of muscle activation. The root-mean-square value and the muscle co-contraction index (CI) were calculated to quantify the muscle activation level and muscle coactivation level, respectively.

The CI was calculated based on a pair of integrated agonist and antagonist activities. ²¹ 100% means full coactivation and 0% indicates lack of coactivation. Three CIs were calculated in the present study, namely those for knee sagittal plane motion (ie, QUAD/HAM), ankle frontal plane motion (ie, TA/PL), and ankle sagittal plane motion (ie, TA/GAS). The CI was calculated as follows:

$$\mathrm{CI}=\frac{2{\mathrm{I}}_{\mathrm{antagonist}}}{{\mathrm{I}}_{\mathrm{total}}}\times 100\%$$

where I_antagonist is the integrated sEMG signal of the antagonist muscle used to represent the antagonist muscle activity during the period. The antagonist muscle scale was multiplied by 2 because of an equivalent increment in the agonist muscle activity. ¹¹ I_total is the sum of the integrated sEMG signal of the antagonist muscle and agonist muscle activities.

The dependent variables were derived at the time point of IC and the time the athletes pushed against the ground and were determined based on vertical ground-reaction force. The IC phase was determined as vertical force appeared while the propulsion phase was determined as the time point the maximal vertical force appeared to the time point the vertical force disappeared.

Statistical Analysis

One-way analysis of variance with least significant difference post hoc tests or Kruskal-Wallis tests was used to test for differences among the FAI-T, FAI-NT, and H groups in terms of basic demographics and the difference of performance parameters among 3 groups after training. Comparisons of the intervention effects between the FAI-T and FAI-NT groups were made using the analysis of covariance with the pretest baseline as a covariate.

The dependent variables were ankle joint angles, muscle activation levels, and muscle CI. The alpha level was set to 0.05. SPSS version 17.0 (IBM Corp) was used for all statistical analyses.

Results

Participant Demographics

There was no significant difference among the 3 groups with regard to their age, height, body weight, and body mass index (P > 0.05). There was also no significant difference between the FAI-T and FAI-NT groups in terms of CAIT scores (P > 0.05). The demographics of the participants are shown in Table 1.

Table 1.

Demographic data for each study group

	FAI-T	FAI-NT	Healthy	P a
Sex				—
Male	9	9	10
Female	4	4	4
Unstable ankle	7R6L	7R6L	10R4L	—
Right	7	7	10
Left	6	6	4
Age, y	22.6 ± 62.47	22.86 ± 1.78	21.37 ± 0.75	0.06
Height, cm	168.75 ± 6.73	166.90 ± 7.95	170.84 ± 7.53	0.37
Body weight, kg	67.88 ± 13.68	62.22 ± 8.83	65.14 ± 10.90	0.41
BMI, kg/m2	23.73 ± 3.95	22.31 ± 2.65	22.19 ± 2.22	0.34
Baseline CAIT	16.38 ± 5.65	16.47 ± 5.03	29.36 ± 0.93	<0.01

BMI, body mass index; CAIT, Cumberland Ankle Instability Tool; FAI, functional ankle instability; NT, nontraining group; T, training group.

^aValue in boldface indicates statistical significance (P < 0.05).

Electromyography

After training, the activation levels of HAM and TA were significantly greater in the FAI-T group than those in the FAI-NT group in the IC and propulsion phases, respectively (P = 0.01 and 0.011) (Table 2). Regarding muscle coactivation, the CIs of the TA-PL in the IC phase and the QUAD-HAM in the propulsion phase were significantly greater in the FAI-T group than those in the FAI-NT group (P = 0.02 and 0.03) (Table 2).

Table 2.

ANCOVA of sEMG variables for training group and nontraining group during lateral shuffling in IC phase and propulsion phase after 6 weeks with baseline measurement as covariate

	Group		ANCOVA (Pretest as Covariate)
Outcome Variable	FAI-T (Mean ± SD)	FAI-NT (Mean ± SD)	F statistic	P a
RMS
IC phase
HAM	0.59 ± 0.14	0.44 ± 0.18	7.555	0.01
QUAD	1.40 ± 0.42	1.29 ± 0.50	0.711	0.40
GAS	0.88 ± 0.23	0.85 ± 0.48	0.031	0.86
TA	0.27 ± 0.08	0.24 ± 0.10	0.805	0.38
PL	0.27 ± 0.08	0.29 ± 0.14	0.917	0.35
Propulsion phase
HAM	0.25 ± 0.09	0.19 ± 0.14	2.037	0.16
QUAD	0.46 ± 0.17	0.41 ± 0.19	1.137	0.29
GAS	0.45 ± 0.31	0.49 ± 0.29	0.092	0.76
TA	0.36 ± 0.17	0.23 ± 0.10	7.211	0.01
PL	0.25 ± 0.11	0.19 ± 0.08	2.644	0.11
Co-contraction index, %
IC phase
QUAD-HAM	64.29 ± 7.78	56.23 ± 15.44	4.053	0.05
TA-GAS	63.02 ± 14.83	57.65 ± 21.26	0.782	0.38
TA-PL	92.74 ± 4.10	87.47 ± 8.07	6.521	0.02
Propulsion phase
QUAD-HAM	63.87 ± 8.57	54.93 ± 14.03	5.132	0.03
TA-GAS	57.71 ± 16.33	56.44 ± 16.36	0.087	0.77
TA-PL	86.11 ± 8.61	88.50 ± 9.96	0.518	0.48

ANCOVA, analysis of covariance; FAI, functional ankle instability; GAS, gastrocnemius; HAM, hamstring; IC, initial contact; NT, nontraining group; PL, peroneus longus; QUAD, quadriceps; RMS, root mean square; T, training group; TA, tibialis anterior.

^aValues in boldface indicate statistical significance (P < 0.05).

After training, comparisons of the muscle activation levels in the IC phase among the groups revealed that the muscle activities of HAM were significantly greater in the FAI-T group than in the FAI-NT group (Table 3). The FAI-T group demonstrated significant muscle activation of GAS and TA in both the IC and propulsion phases compared with those for the healthy group (Table 3). The CIs in the FAI-T group were greater than those in the FAI-NT group, but the difference did not reach statistical significance between the groups during either the IC or propulsion phase (Table 3).

Table 3.

One-way ANOVA and post hoc test results for EMG variables of 3 groups during lateral shuffling at IC phase and propulsion phase after 6 weeks

	Group
Outcome Variable	FAI-T (Mean ± SD)	FAI-NT (Mean ± SD)	Healthy (Mean ± SD)	P	Post-comparisons
RMS
IC phase
HAM	0.59 ± 0.14	0.44 ± 0.18	0.52 ± 0.18	0.05	a
QUAD	1.40 ± 0.42	1.29 ± 0.50	1.14 ± 0.40	0.19	—
GAS	0.88 ± 0.23	0.85 ± 0.48	0.68 ± 0.21	0.04	b
TA	0.27 ± 0.08	0.24 ± 0.10	0.21 ± 0.08	0.05	b
PL	0.27 ± 0.08	0.29 ± 0.14	0.23 ± 0.07	0.23	—
Propulsion phase
HAM	0.25 ± 0.09	0.19 ± 0.14	0.19 ± 0.09	0.18	—
QUAD	0.46 ± 0.17	0.41 ± 0.19	0.41 ± 0.29	0.48	—
GAS	0.45 ± 0.31	0.49 ± 0.29	0.27 ± 0.16	0.04	b, c
TA	0.36 ± 0.17	0.23 ± 0.10	0.27 ± 0.12	0.02	a, b
PL	0.25 ± 0.11	0.19 ± 0.08	0.20 ± 0.08	0.25	—
Co-contraction index, %
IC phase
QUAD-HAM	64.29 ± 7.78	56.23 ± 15.44	64.94 ± 13.62	0.09	—
TA-GAS	63.02 ± 14.83	57.65 ± 21.26	65.07 ± 14.58	0.43	—
TA-PL	92.74 ± 4.10	87.47 ± 8.07	91.06 ± 6.56	0.18	—
Propulsion phase
QUAD-HAM	63.87 ± 8.57	54.93 ± 14.03	62.93 ± 12.65	0.06	—
TA-GAS	57.71 ± 16.33	56.44 ± 16.36	60.93 ± 13.68	0.68	—
TA-PL	86.11 ± 8.61	88.50 ± 9.96	91.32 ± 6.97	0.11	—

ANOVA, analysis of variance; EMG, electromyography; FAI, functional ankle instability; GAS, gastrocnemius; HAM, hamstring; IC, initial contact; NT, nontraining group; PL, peroneus longus; QUAD, quadriceps; RMS, root mean square; T, training group; TA, tibialis anterior.

a, b, and c, respectively, represent significant differences between FAI-T and FAI-NT, FAI-T and healthy, and FAI-NT and healthy groups.

Kinematics

The FAI-T group exhibited ankle eversion, whereas the FAI-NT group exhibited ankle inversion in the IC phase (P = 0.01) (Table 4). The ankle inversion angle was significantly smaller in the FAI-T group after training than that in the FAI-NT group (P = 0.04) (Table 4).

Table 4.

ANCOVA of ankle joint angles (degrees) for FAI-T group and FAI-NT group during lateral shuffling in IC phase and propulsion phase after 6 weeks with baseline measurement as covariate

	Group		ANCOVA (Pretest as Covariate)
Joint Angle, deg	FAI-T (Mean ± SD)	FAI-NT (Mean ± SD)	F Statistic	P a
IC phase
Dorsiflexion (+)/ plantarflexion (−)	8.75 ± 4.96	4.79 ± 7.61	0.898	0.35
Eversion (+) / inversion (−)	3.38 ± 3.41	−1.20 ± 8.05	6.883	0.01
External rotation (+) / internal rotation (−)	12.32 ± 4.95	8.74 ± 5.97	2.765	0.11
Propulsion phase
Dorsiflexion (+) / Plantarflexion (−)	−9.46 ± 5.00	−14.86 ± 11.58	3.496	0.07
Eversion (+) / inversion (−)	−9.16 ± 5.48	−10.46 ± 4.01	4.420	0.04
External rotation (+) / internal rotation (−)	−1.73 ± 5.70	0.98 ± 5.39	0.923	0.34

ANCOVA, analysis of covariance; FAI, functional ankle instability; IC, initial contact; NT, nontraining group; T, training group.

^aValues in boldface indicate statistical significance (P < 0.05).

Comparisons of the joint angle in the propulsion phase among the groups after training indicate that the angle of ankle plantar flexion was the largest in the FAI-NT group, followed by the FAI-T group and the healthy group, with the differences reaching statistical significance among groups (Table 5).

Table 5.

One-way ANOVA and post hoc test results for ankle joint angles (degrees) of 3 groups in IC phase and propulsion phase after 6 weeks

	Group
Joint angle, deg	FAI-T (Mean ± SD)	FAI-NT (Mean ± SD)	Healthy (Mean ± SD)	P	Post Hoc Comparisons
IC phase
Dorsiflexion (+) / plantarflexion (−)	8.75 ± 4.96	4.79 ± 7.61	6.89 ± 9.57	0.31	—
Eversion (+) / inversion (−)	3.38 ± 3.41	−1.20 ± 8.05	2.06 ± 7.07	0.10	—
ER (+) / IR (−)	12.32 ± 4.95	8.74 ± 5.97	10.00 ± 4.26	0.10	—
Propulsion phase
Dorsiflexion (+) / plantarflexion (−)	−9.46 ± 5.00	−14.86 ± 11.58	−4.37 ± 9.78	<0.01	a, b, c
Eversion (+) / inversion (−)	−9.16 ± 5.48	−10.46 ± 4.01	−6.70 ± 8.32	0.08	—
ER (+) / IR (−)	−1.73 ± 5.70	0.98 ± 5.39	2.94 ± 12.25	0.13	—

ANCOVA, analysis of covariance; ER, external rotation; FAI, functional ankle instability; IC, initial contact; IR, internal rotation; NT, nontraining group; T, training group.

a, b, and c, respectively, represent significant differences between FAI-T and FAI-NT, FAI-T and healthy, and FAI-NT and healthy groups.

Discussion

The purpose of this study was to examine the effect of this 6-week training with a remodeled bicycle pedal on the muscle activities, muscle co-contraction, and kinematics of the lower extremity in the athletes with FAI. The 3-dimensional remodeled bicycle pedal with tilts 40^o sideways during loaded cycles had capability of enhancing joint control. The current study design using true control group (FAI-NT) proved that the 6-week training with a remodeled bicycle pedal provided better outcomes than that of no-intervention (FAI-NT). Participants in the FAI-T group were given sudden ankle plantar flexion and inversion during pedaling to elicit reaction of PL muscles to avoid ankle over inversion. ⁵ By simulating the movement of ankle sprain but under control, the participants could practice how to control ankle movement during dynamic training program. Thus, this design is a different training/intervention from traditional strengthening programs. Additionally, the results again supported that the FAI-T improved not only the muscle activation and co-contraction but also the ankle eversion angle at IC and propulsion phases. Both improved muscle activation and ankle kinematics suggested that the effects were attributed from remodeled bicycle pedal instead of strengthening of pedaling.

Lateral movements are crucial in sports such as volleyball, tennis, and basketball. ²⁵ These sports require rapid changes of direction, such as lateral shuffling and side cuts. ⁴⁴ Lateral shuffling is more challenging and may cause ankle sprains because of rapid changes in direction. Hence, the investigation of ankle dynamic control ability during lateral shuffling is important for preventing ankle sprains. This study hypothesized that the muscle activation of the ankle evertor and the magnitude of the coactivation will increase to improve ankle stability during dynamic movement after training. The results support our hypotheses, showing that the CI of TA-PL was significantly greater in the FAI-T group than that in the FAI-NT group after training. After remodeled bicycle pedal training, the FAI-T group demonstrated a significantly smaller ankle inversion angle compared with that in the FAI-NT group.

Muscle Activation and Coactivation in the IC Phase

Shock absorption is important in the IC phase. It is achieved via plantar flexor muscle activation before and after impact and contributes to eccentric contraction to absorb the impact force and reduce joint load. In the IC phase, the TA muscle activates eccentrically to decelerate the ankle dorsiflexion to reduce joint loading at touchdown. The FAI-T group after training demonstrated a significantly greater amount of co-contraction between TA and PL than that in the FAI-NT group. This provided greater joint stability during lateral shuffling to potentially reduce injury at touchdown. Our findings confirm that the unstable condition created by the 3-dimensional pedal of the remodeled bicycle could improve ankle stability by enhancing the coactivation of muscles.

In addition, subjects in the FAI-T group after training demonstrated a greater activation level of HAM muscles as well as a tendency of greater coactivation between QUAD and HAM muscles in the IC phase compared with the subjects in the FAI-NT group. The primary function of the QUAD and HAM muscles is to eccentrically decelerate the center of mass and stabilize the knee after impact. ²⁸ In general, a larger CI indicates a greater activation rate in the antagonist muscle than the agonist muscle, and thus greater joint stiffness. Previous studies have shown that those who have anterior cruciate ligament deficiency may undergo excessive anterior translation of the tibial plateau, and thus the greater CI of QUAD-HAM could result in an increase in the compression force of the knee joint during close kinetic chain exercise^{12,13,37,42,48} to increase the knee joint stability and reduce the anteroposterior displacement of the tibia. ¹

Muscle Activation and Coactivation in the Propulsion Phase

The TA activation level was enhanced significantly in the FAI-T group after training compared with both the FAI-NT and healthy groups. TA is an important muscle in impeding ankle sprain, especially during propulsion, by preventing excessive ankle plantarflexion. Son et al ³⁶ proposed that subjects with CAI might adopt a hip-dominant gait strategy for power production based on findings that 22% more hip concentric power attributed to 11% less ankle concentric power because decreased plantar-flexor function was produced during the preswing. Therefore, remodeled bicycle pedal training might improve power generation for forward acceleration during the preswing for the TA muscle. The CI of QUAD-HAM was significantly greater in the FAI-T group after training, indicating improved knee stabilization. It has been reported that the vasti muscles and HAM muscles are both activated in the propulsion phase in healthy participants during a lateral shuffling task. ²⁸ The FAI-T group after training had a significantly greater CI of QUAD-HAM than the FAI-NT group, indicating that remodeled pedal training improved not only ankle stability but also knee stability during a lateral shuffling task. An unstable ankle limits energy transfer via the kinetic chain of the lower extremity ³⁸ and this may increase the rate of noncontact knee injuries. ³⁹ Hence, pedaling training may improve ankle stability and prevent the occurrence of noncontact knee injuries.

Kinematic Outcomes in the IC Phase

The FAI-T group exhibited ankle eversion, whereas the FAI-NT group exhibited ankle inversion in the IC phase during lateral shuffling. The healthy group exhibited ankle eversion, which is a function of a lateral weight shift to adjust the center of mass to the appropriate position, and prevent unfavorable excessive medial shift of the center of mass during medial landing. ⁴⁷ Combining the kinematics and the sEMG profiles measured in the present study, the HAM muscle activation was significantly greater in the FAI-T group after training compared with that in the FAI-NT group. HAM muscles assist tibia stabilization, which could prevent excessive tibia rotation and reduce the chances of ankle sprain. ²⁹ In lateral shuffling, the ankle joint performs greater joint motions in the coronal plane. Thus, the significantly greater CI of TA-PL in the FAI-T group than that in the FAI-NT group indicates greater ankle joint stability in the IC phase in the FAI-T group after training. The ankle exhibited external rotation in the IC phase in the FAI-T group after training, although the difference did not reach statistical significance. These results indicate that the ankle was more stable after remodeled bicycle pedal training, which might reduce the rate of ankle sprains.

Kinematic Outcomes in the Propulsion Phase

The FAI-T group after training demonstrated a smaller ankle plantar flexion angle and a significantly smaller inversion angle in the propulsion phase than those in the FAI-NT group. This might have resulted from improved neuromuscular control after training. The continuous challenge to the ankle joint during remodeled pedaling affects the muscle response. Remodeled pedal training has been reported to improve ankle joint position sense and postural sway. ⁵ The injured ligament may present proprioception deficits and a delayed muscle reaction time. ⁴⁵ The impaired proprioception transmits incorrect information to the central nervous system, which in turn influences the ability of muscles and joint adapting to surface and results in misplaced foot position at landing. ³⁵ Hence, ankle control may be improved to avoid excessive ankle plantar flexion and inversion after remodeled pedal training. Ankle plantar flexion was exhibited in the propulsion phase to change direction during lateral shuffling movement. Coupling with the muscle activities, the GAS muscle activates to transfer knee power to the ankle joint to perform powerful propulsion, ²⁰ which might explain why the ankle joint performed inversion and plantarflexion in this phase during lateral shuffling. This is the critical time for an ankle sprain.

Limitations

Gender differences might influence kinematic performance. According to a previous study, women demonstrated a greater ankle joint angle in the frontal plane than men did. ¹⁰ Further investigation is needed to address this issue. While the EMG differences between the groups were statistically significant, their clinical significance remains unknown.

Conclusion

Remodeled bicycle pedal training helps normalize the gait neuromechanics that may contribute to recurrent sprains and improves knee stability by facilitating TA activation and the coactivation of the QUAD and HAM muscles during the propulsion phase. Furthermore, the reduced ankle plantar flexion and inversion angle in the propulsion phase after training confirm that neuromuscular control can be improved through remodeled bicycle pedal training. In addition to the improvement in static balance control shown in our previous study, the present study demonstrates that a better movement strategy of the lower extremities during a dynamic task can be achieved via a 6-week remodeled pedal training program. Therefore, incorporation of a remodeled pedal training program into a rehabilitation program is warranted for those with FAI.