Journal of Sport Rehabilitation, 2014, 23, 134-144 http://dx.doi.org/10.1123/JSR.2013-0021 © 2014 Human Kinetics, Inc.

www.JSR-Journal.com ORIGINAL RESEARCH REPORT

Lower-Extremity Electromyography Measures During Walking With Ankle-Destabilization Devices Luke Donovan, Joseph M. Hart, and Jay Hertel Context: Ankle-destabilization devices are rehabilitation tools that may improve neuromuscular control by increasing lower-extremity muscle activation. Their effects should be tested in healthy individuals before being implemented in rehabilitation programs. Objective: To compare EMG activation of lower-extremity muscles during walking while wearing 2 different ankle-destabilization devices. Design: Crossover. Setting: Laboratory. Participants: 15 healthy young adults (5 men, 10 women). Intervention: Surface EMG activity was recorded from the anterior tibialis, peroneus longus, lateral gastrocnemius, rectus femoris, biceps femoris, and gluteus medius as subjects walked on a treadmill shod, with an ankle-destabilization boot (ADB), and an ankle-destabilization sandal (ADS). Main Outcome Measures: Normalized amplitudes 100 ms before and 200 ms after initial heel contact, time of onset activation relative to initial contact, and percent of activation time across the stride cycle were calculated for each muscle in each condition. Results: The precontact amplitudes of the peroneus longus and lateral gastrocnemius and the postcontact amplitudes of the lateral gastrocnemius were significantly greater in the ADB and ADS conditions. In the ADB condition, the rectus femoris and biceps femoris postcontact amplitudes were significantly greater than shod. The peroneus longus and lateral gastrocnemius were activated significantly earlier, and the anterior tibialis, lateral gastrocnemius, and rectus femoris were activated significantly longer across the stride cycle in the ADB and the ADS conditions. In addition, the peroneus longus was activated significantly longer in the ADB condition when compared with shod. Conclusions: Both ankle-destabilization devices caused an alteration in muscle activity during walking, which may be favorable to an injured patient. Therefore, implementing these devices in rehabilitation programs may be beneficial to improving neuromuscular control. Keywords: chronic ankle instability, gait, rehabilitation Lateral ankle sprains are among the most common musculoskeletal injury seen in sports1,2 and in people who are recreationally active.3 It is estimated that approximately 47% to 74% of people who suffer a lateral ankle sprain will go on to have recurrent sprains 6 to 18 months after their first ankle sprain.4–6 Furthermore, it is estimated that about 30% of people who suffer an ankle sprain will go on to develop chronic ankle instability (CAI).5,7 CAI is a pathology characterized by residual symptoms of ankle instability and recurrent feelings of the ankle “giving way” that lasts longer than 1 year after the initial sprain.8,9 Although the cause of CAI remains unclear, multiple characteristics have been identified to be different in groups with CAI compared with those who do not. These characteristics include, but are not limited to, impaired proprioception,10–14 decreased neuromuscular control,15–19 decreased range of motion,20–22 decreased strength,10,15,23 and altered gait.24–28 Multiple studies have shown patients with CAI to have decreased The authors are with the Kinesiology Program, University of Virginia, Charlottesville, VA. Address author correspondence to Luke Donovan at [email protected].

134

dorsiflexion20,27 and increased inversion during the late phases of swing.24,25,28,29 Finally, multiple studies found neuromuscular deficits in patients after acute ankle sprains and in those with CAI in more proximal muscles such as the hip,30,31 quadriceps, and hamstring musculature.32,33 It is thought that a combination of these characteristics leads to repetitive bouts of instability.34 CAI is commonly treated conservatively.35 Rehabilitation programs often consist of exercises to improve range of motion, increase strength, and improve proprioception and neuromuscular control. In addition to therapeutic exercises, clinicians often implement devices designed to improve various aspects associated with CAI. These devices, such as the biomechanical ankle platform system (BAPS), DynaDisc, foam padding, and wobble boards, allow for multiplanar motion at the ankle and have been shown to improve postural control in patients with CAI.17,36–38 The unstable nature of these rehabilitation devices most likely enhances muscle activation by placing increased external demands on the sensorimotor system. Even though these devices cause improvements in postural control, they are limited to relatively nonfunctional rehabilitation exercises and have not been shown to alter gait patterns of patients with CAI. A recent case

Ankle-Destabilization Devices  135

report showed that an extensive dynamic neuromuscular training program caused an alteration in landing patterns and ground-reaction forces during gait in a single participant with CAI.39 The program consisted of posturalcontrol exercises, strength exercises, plyometrics, and speed/agility exercises. That case report may indicate that neuromuscular training that goes beyond static balancing may be an essential part of improving functional movement patterns. Ankle-destabilization devices were developed to provide clinicians with a method to address these neuromuscular deficits associated with CAI, while the patient performs more functional tasks such as walking, running, and jumping. Past research has not described a specific definition of ankle-destabilization devices, but for the purpose of this article we will operationally define them as devices that consist of either a shoe or a sandal with an articulator below the heel designed to mimic the motion that occurs at both the subtalar and talocrural joint during walking. The goal of these devices is to force the patient using the device into plantar flexion, inversion, and internal rotation in a controlled manner while completing functional tasks. It is thought that by causing an anticipated perturbation at the ankle, the evertors will contract before initial contact (IC) to pull the foot out of the pathological position. In addition, pre-IC and post-IC muscle activity, which can be measured by surface electromyography (sEMG), is essential in providing dynamic stability to an unstable joint.25,40 Specifically, when the peroneal musculature is contracted, the musculotendinous unit generates stiffness, which helps dynamically stabilize the joint.9,40 A recent study has shown that an ankle-destabilization device (Myolux Athletik, Cevres Santé, Le Bourgetdu-Lac, France) caused an increase in EMG amplitudes of the anterior tibialis, peroneus longus, and peroneus brevis during walking and caused a pre-IC activation of both the peroneus longus and peroneus brevis in people with no history of ankle sprain.41 However, that study only examined muscles in the lower leg, did not examine amplitude before heel contact, and had the subject walk for only 6 m, only wearing the device on 1 foot. We believe that before recommending using these devices in an injured population it is important to understand the influence they have on lower-extremity muscle activation both before and after heel strike in healthy individuals. Therefore, the purpose of this study was to compare EMG activation of several lower-leg and thigh muscles during treadmill walking in healthy subjects wearing 2 different ankle-destabilization devices.

Methods Design We performed a crossover descriptive laboratory study. The independent variable was condition (shod, ankledestabilization boot [ADB], and ankle-destabilization sandal [ADS]). The dependent variables were sEMG

amplitudes before and after IC, timing of muscle onset relative to IC, and percent activation time across the stride cycle for the anterior tibialis, peroneus longus, lateral gastrocnemius, rectus femoris, biceps femoris, and gluteus medius.

Participants Fifteen healthy physically active young adults (10 women and 5 men) completed this study (Table 1). All participants reported to have no history of previous ankle sprain, no history of lower-extremity injury or surgery, and no history of any illness that might influence neuromuscular control or impede their ability to complete the walking tasks. All subjects provided written informed consent, and the study was approved by the university’s institutional review board.

Instruments Disposable, pregelled 10-mm-round Ag-AgCl surface EMG electrodes were used for all participants. The signal was amplified with a high-gain, differential-input biopotential amplifier with a gain of 1000 and digitized with a 16-bit data-acquisition system (MP 150, Biopac Systems, Inc, Goleta, CA) at 2000 Hz with a commonmode rejection ratio of 110 dB, an input impedance of 1.0 MΩ, and a noise voltage of 0.2 mV. Data collection and reduction were completed using Acqknowledge software (v 4.0, Cambridge, England). During treadmill walking, foot contacts were synchronized with sEMG data by using a 2-pronged foot switch (Biopac Systems Inc, Santa Barbara, CA) that identified positive and negative pressure. All participants wore a standard athletic shoe (New Balance, Brighton, MA, Model X755WB) during the control shod condition.

Ankle-Destabilization Devices We used 2 separate ankle-destabilization devices: the ADB (Figure 1) and the ADS (Figure 2). The ADB was previously used by Forestier et al.41 This device consists of a half-shoe with an articulator located beneath the heel and a puck that is worn beneath the metatarsal heads. The articulator allows for approximately 45° of combined inversion and plantar flexion. The puck decreases the Table 1  Subject Demographics Mean (SD) Age (y)

22.9 (3.4)

Height (cm)

173 (9.4)

Mass (kg)

70.8 (18)

Godin score

84 (40)

FAAM activities of daily living %

100 (0)

FAAM Sport %

100 (0)

Abbreviations: FAAM, Foot and Ankle Ability Measure.

(a)

(b)

(c)

Figure 1 — The ankle-destabilization boot (a) in a neutral position; (b) rotating on the axis of the articulator, causing a plantar-flexed and inverted position; and (c) being worn with the metatarsal puck.

(a)

(b)

(c) Figure 2 — (a) Lateral and (b) posterior view of the ankle-destabilization sandal. (c) The ankle-destabilization sandal being worn.

136

Ankle-Destabilization Devices  137

amount of plantar flexion of the foot. The ADS (Myolux II, Cevres Santé, Le Bourget-du-Lac, France) is a fulllength sandal and allows for approximately 30° of inversion and plantar flexion. The articulator is different from that of the ADB; not only does it allow less motion, but also the articulator has a smaller lever arm and a cradle shape. The ADS was designed for earlier stages in functional rehabilitation, while the ADB was intended for later phases of rehabilitation and sport performance.

Testing Procedures Day 1.  On the first day of the study, participants

completed the informed-consent form, a general healthhistory questionnaire, the Foot and Ankle Ability Measure activity of daily living and sport scales, and the Godin Leisure Time Physical Activity Scale. Next, they were fitted for shoes, the ADB, and the ADS. They were acclimated to each of the 3 conditions by walking on the treadmill at 4.83 km/h for approximately 5 minutes.

Day 2.  Participants returned to the laboratory within 7

days of the screening visit. Testing leg was determined via coin flip before the start of the study. All surface EMG-electrode setup procedures were completed as recommended by past research.42–44 Skin was shaved, debrided, and cleansed with isopropyl alcohol over each electrode site to minimize impedance. Each site was determined by palpation of the muscle belly during voluntary contraction of the muscle. Surface electrodes were placed 2 cm apart and oriented parallel to the muscle-fiber orientation over the midline of the muscle belly.43,44 Electrodes were placed over the anterior tibialis, peroneus longus, lateral gastrocnemius, rectus femoris, biceps femoris, and gluteus medius as recommended by previous research.42–45 Once sEMG electrodes were set up, we visually evaluated for crosstalk between muscles by having the individual contract each muscle at separate times. Walking.  Participants walked on a treadmill at a

4.83 km/h pace at a 0% incline in each condition for 30 seconds. Data were recorded after the participants reached the designated pace and stated that they felt that they were walking normally. Condition order was counterbalanced via a Latin square. IC and terminal stance were identified via foot switches placed on the heel and toe.

Data Processing All sEMG data were filtered (10- to 500-Hz band pass) and integrated using a 10-sample moving-window rootmean-square (RMS) algorithm and normalized to sEMG data collected during quiet standing. Quiet Standing.  A 500-millisecond time epoch during the middle portion of quiet standing was selected, filtered (10- to 500-Hz band pass), and processed using a 10-sample moving-average RMS algorithm. The mean

RMS value and standard deviation were calculated and later used for normalization purposes. Pre-IC Amplitude.  The area under the RMS curve was

calculated for a 100-millisecond time epoch immediately before IC. This value was normalized by dividing it by the mean RMS value of each corresponding muscle during quiet standing. A total of 9 strides were processed (3 consecutive steps were taken during the first, middle, and last 10 seconds of the treadmill walking). The average amplitude for the 9 trials was calculated and subjected to statistical analysis. Our method of calculating amplitude is consistent with previous authors25 and with recommendations for sEMG.43,44 Post-IC Amplitude.  A 200-millisecond time epoch

immediately after IC was used to calculate the area under the RMS curve using the same normalization procedures and number of strides as previously described. We chose a 100- and 200-millisecond pre/post IC because Forestier et al41 found that while using ankle-destabilization devices, the peroneus longus was activated as early as 77 milliseconds before IC and as late as 144 milliseconds after IC without the devices. We wanted to be sure we captured sufficient sEMG activity around these event points.

Time of Activation Relative to IC.  Each muscle was determined to be activated when the RMS value was 10 SDs above the mean RMS value recorded during quiet standing. A time epoch (s) was measured from IC to when the muscle first activated. If the muscle was not activated until after IC, a positive value represented post-IC activation (Figure 3[a]). However, if the muscle was activated before IC, a negative value represented preactivation (Figure 3[b]). As with the precontact and postcontact amplitudes, 9 strides were analyzed. Percent of Activation Time.  Activation was again

determined if the RMS value exceeded 10 SDs of the quiet standing mean for each muscle. RMS values that exceeded the 10-SD threshold were assigned a value of 1, and any RMS value that did not exceed the threshold was assigned a value of 0. The mean value across each stride was then calculated for each muscle and multiplied by 100 to form a percentage. The percentage represents how long the muscle was activated during a given stride cycle.

Statistical Analysis Data were analyzed using Statistical Package for Social Sciences (SPSS) Version 20.0 (SPSS, Inc, Chicago, IL). For each of the 4 main outcome measures (pre-IC amplitude, post-IC amplitude, time to activation, and percent activation time), a MANOVA was performed to assess the effects of device condition collectively on the activity of all 6 muscles. Specifically, for each outcome measure, a 1-within-factor MANOVA was performed with the factor being condition at 3 levels (shod, ADB, ADS). This factor was considered to have repeated measures because all subjects were tested in each of 3

138  Donovan, Hart, and Hertel

(a)

(b) Figure 3 — (a) Example of a typical peroneus longus surface electromyographic (sEMG) signal during walking in the shod condition. The muscle is considered to be activated after initial contact. Time was calculated from initial contact until the sEMG signal exceeded 10 SDs above the mean signal during quiet standing. (b) Example of a typical peroneus longus sEMG signal during walking in the ankle-destabilization-boot condition. The muscle is considered to be preactivated before initial contact. Time was calculated from initial contact until the instant the sEMG signal was 10 SDs above the mean signal during quiet standing.

conditions. The level of significance for the MANOVA was set a priori at P ≤ .05. If there was a significant condition main effect for the MANOVA for a given outcome measure, univariate ANOVAs were performed to assess the impact of condition on the activity of each of the 6 muscles. If the condition main effect for the ANOVA

was significant for a particular muscle with a Bonferronicorrected P value of

Lower-extremity electromyography measures during walking with ankle-destabilization devices.

Ankle-destabilization devices are rehabilitation tools that may improve neuromuscular control by increasing lower-extremity muscle activation. Their e...
385KB Sizes 0 Downloads 0 Views