Comparison of lower-leg muscle activation and establishment of muscle activation patterns during single-leg stance under various instability conditions in healthy active subjects: a cross-sectional study

Participants

An a priori sample size estimation was performed using G * Power v.3.1.9.2 software. Based on Alfuth & Gomoll (2018), who observed large muscle activation effect sizes, an expected effect size of F = 0.4 with α = 0.05 indicated a required sample size of 19 for >80% power. To increase measure power, the sample size was raised to 30 participants (21 male, nine female (mean ± SD) age = 22.73 ± 2.69 years, height = 1.73 ± 0.08 m, body mass = 70.13 ± 12.39 kg, Body Mass Index = 23.41 ± 2.72, time of weekly physical activity = 7.3 ± 2.84 h).

To be eligible, participants had to be between 18 and 30 years of age (to ensure a homogeneous young adult sample and, therefore, minimizing the potential variability associated with age-related neuromuscular adaptations and physical activity levels), not to have experienced lower limb injury or pain within the last year prior to the study, and self-report as physically active, such that they perform at least 90 min of weekly physical activity, assessed using the International Physical Activity Questionnaire (IPAQ). Exclusion criteria included previous participation in any lower limb balance or proprioception exercise program and the presence of known balance impairments such as vertigo, vestibular or central alterations. All participants provided written informed consent and completed a basic information form prior to data collection, which included demographic (age and sex) and anthropometric (height and weight) measures. Participant recruitment was conducted through volunteer calls at the Faculty of Physiotherapy of the University of Valencia by a convenience sampling, since participants were selected based on availability and willingness to participate in the study.

Ethics committee

This study was approved by the Ethics Committee of University of Valencia (approval number 1271077), and was conducted following the STROBE Statement for cross-sectional studies to ensure transparency and completeness in reporting (von Elm et al., 2008).

Procedures

Specific instability device

The BB (Blackboard Training, Innenstadt, Germany) (Fig. 1) is a training device designed for monopodal stability work, composed of two wooden boards that, connected by a strap, are rectangular in shape. Its base has a Velcro surface where wooden half-cylinders can be freely placed. Depending on the placement, different types of instability are achieved. This study intended to analyze seven instability conditions (Fore, Sup, Pro, Rear, Inver, Ever, Total) detailed in Fig. 2, and the floor as a stable condition. In Fore, Sup and Pro conditions the BB had the rearfoot fixed and the forefoot was unstable with bar placed in the center, laterally or medially, forcing pronosupination, supination or pronation, respectively. In Rear, Inver and Ever conditions the BB had the forefoot fixed and the rearfoot was unstable with the bar placed in the center, laterally or medially, forcing inversion-eversion of the rearfoot, only inversion or only eversion, respectively. Finally, in the Total configuration both the forefoot and rearfoot were unstable with the bars placed in the center, forcing the pronosupination and inversion-eversion movements of the entire foot.

Electromyographic signal processing and maximum voluntary isometric contractions

The chosen technique to record muscle activity was surface electromyography (EMG). The analyzed muscles included: soleus (SOL), gastrocnemius medialis (GM), gastrocnemius lateralis (GL), TA, PL, and PB (Fig. 3). After the area was shaved and cleaned with alcohol to minimize skin impedance, adhesive pre-gelled Ag/AgCl EMG electrodes (BlueSensor M; Ambu Ballerup, Denmark; 40 × 34 mm) were placed on the middle region of the muscle belly aligned with the fiber orientation, according to the SENIAM guidelines (Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles; www.seniam.org). Verification of electrode placement was done by palpating the muscle while the subject performed a contraction against a resistance provided by the examiner. The lower limb to measure was randomized.

The MuscleLab 4020e (Ergotest Technology AS) EMG device was used to record muscle activity. Raw EMG signals were processed using custom-written scripts in MATLAB (version R2019b; The Mathworks, Natick, MA, USA). First, signals were sampled at 1,000 Hz. Then, they were band-pass filtered using a fourth-order Butterworth filter with a cutoff frequency of 20 and 500 Hz, rectified, integrated, and converted to root-mean-square signals using a hardware circuit network (frequency response: 450 kHz; averaging constant: 12 ms; total error ±0.5%). To minimize the influence of transient spikes, the EMG signals were visually inspected, and any trials exhibiting abrupt, non-physiological peaks were excluded from analysis.

To allow between-condition comparisons and with normalization purposes, the highest activation of each muscle during a maximum voluntary isometric contraction (MVIC) was recorded. This normalization method has been widely used to compare muscle activation across conditions and individuals (Besomi et al., 2019). After a practice trial, each muscle was tested three times for 5 s each, with 2-min intervals between trials. Verbal encouragement was provided during testing (Harput et al., 2013).

Statistical analysis

To normalize EMG data, the trial with the highest activation from the MVICs was selected. Of the 20-s trials recorded for each exercise, the maximum signal amplitude within the middle 10 s was extracted and divided by the corresponding MVIC value to obtain the normalized EMG value (nEMG). The average nEMG across the three trials, expressed as a percentage of MVIC, was used for final analysis (Harput et al., 2013).

SPSS Statistics 28 (IBM Corporation, Chicago, IL, USA) was used to perform all statistical analyses. Descriptive statistics (means, 95% confidence intervals, standard deviations) were defined. Normality was assessed using the Shapiro-Wilk test. A two-way repeated-measures ANOVA was employed to determine differences between muscles, conditions, and their interaction. Effect size was evaluated with ηp² (partial Eta-squared): small (0.01 < ηp² < 0.06), medium (0.06 < ηp² < 0.14), or large (ηp² > 0.14) (Cohen, 1977).

Additionally, when a significant main effect or interaction was detected, pairwise comparisons were conducted using post hoc t-tests with Bonferroni correction. The number of pairwise comparisons was 28 within conditions and 15 within muscles, resulting in a corrected significance thereshold of p < 0.00015 to control for Type I error. However, as SPSS provides adjusted p-values through a mathematically equivalent correction, results were interpreted using the conventional significance threshold of p < 0.05. Effect sizes for pairwise comparisons were reported using Cohen’s d: small (d = 0.20–0.49), medium (d = 0.50– 0.79), or large (d $\ge$ 0.80) (Cohen, 1977).

Within-muscle differences between conditions

Group means for each muscle under each condition are presented in Table 1. Between-conditions mean differences and p values, as well as the detailed post-hoc results of the significant with effect sizes, are displayed respectively in Table S3 and S4 of the supplementary material. In relation to the most relevant findings: all configuration showed higher activation than the floor and Rear for the SOL, with Total and Fore configurations producing the highest activation (d = 1.37 and d = 1.56, respectively, when comparing to the floor). In the case of GM, the Total configuration showed higher activation than Rear (d = 0.64), but no further significant differences were observed between conditions. For the GL, both Total and Fore conditions showed higher activation compared to most other configurations, with Total producing the highest activation overall (d = 1.43 when compared to Rear). Regarding the TA, the Total, Fore, and Sup configurations showed higher activation than the floor and all rearfoot conditions, with Total producing the highest activation (d = 1.86 when comparing to the floor). For the PL, the Total, Fore, Pro, and Ever configurations showed higher activation than the floor, Rear, and Sup, with Pro leading to the highest activation (d = 1.47 when compared to Rear). Finally, in the case of the PB, the Total, Fore, Pro, and Ever configurations produced higher activation than the floor, Rear, and Sup, with Total showing the highest activation (d = 2.21 when comparing to Rear).

Figure S1 of the supplementary material shows an ensemble of the six muscles with their between-conditions differences.

Within-condition muscle activation patterns

All between-muscles mean differences and p values, as well as the detailed post-hoc comparison analyses of the significant with effect sizes, are displayed in Tables S5 and S6, respectively. Figure 4 shows muscle activation pattern for each condition.

In relation to the main relevant findings: in the Ever configuration, both PL and PB showed higher activation than GM (d = 0.98 and 0.68), GL (d = 0.81 and 0.66), and TA (d = 0.89 and 0.69), with PL also showing higher activation than SOL (d = 0.64). In the Sup condition, the TA showed higher activation compared to GM, GL, PL, and PB (d = 1.16, 0.95, 0.73, and 0.59, respectively). For the Pro condition, PL and PB showed higher activation compared to SOL (d = 0.99 and 0.95), GM (d = 1.49 and 1.21), and GL (d = 0.80 and 0.71). In the Fore condition, TA and PB showed higher activation than SOL (d = 0.67 and 0.59), GM (d = 1.35 and 1.01), and GL (d = 0.66 and 0.68), and PL showed also higher activation than GM and GL (d = 1.10 and 0.77). Finally, in the Total instability condition, TA showed higher activation than GM and GL (d = 1.30 and 0.74), PL was higher than GM (d = 0.87), and PB showed higher activation than SOL, GM, GL, and PL (d = 0.75, 1.43, 1.01, and 0.64, respectively).

Muscle activation between conditions

One of the main findings of the present study was that the TA, PL, and PB exhibited the most significant differences between configurations and achieved the highest levels of activation during the single-leg stance exercise (approximately 60% to 80% of MVIC). In addition, the configurations of the BB that seem to be significatively more demanding in terms of muscle activity are Pro, Fore, and Total (forefoot pronation, forefoot pronosupination and pronosupination and inversion-eversion of the entire foot, respectively), with Sup (forefoot supination) and Ever (rearfoot eversion) configurations in certain cases. It was found that, in the case of the TA, the most effective configurations are Sup, Fore, and Total, whilst for activating the peroneus (i.e., PL and PB), Sup configuration is not recommended, and Pro or Ever were chosen instead. Specifically, the Sup configuration promotes tibialis anterior activation by inducing a forefoot supination movement, which requires increased TA engagement to control and stabilize the medial arch. Similarly, the Ever configuration emphasizes peroneus activation by promoting rearfoot eversion, which stimulates the peroneal muscles to counteract instability. This may be related to the type of device being analyzed and plane of motion in which the action takes place. The BB offers a small and rigid base of support, which is partially unstable or firm depending on the configuration. In addition, the configurations under study were specifically designed to target these stabilizers (i.e., TA, PL, and PB), thereby inducing movement in the frontal plane. Mayer et al. (2023), analyzed kinematics of the leg muscles in a single-leg stance on both sides of the Togu® Jumper® (bladder and flat), suggesting that standing on the bladder side produces greater ankle movement in the sagittal plane, while the flat side does so in the frontal plane (Mayer et al., 2023). Strøm et al. (2016), had previously observed a similar effect in terms of inversion-eversion number of directional changes when standing on the Wobble board, which proved to be superior in terms of muscle activation to the BOSU® and Airex. The study proposed that reducing the base of support may be the main cause of this effect (Strøm et al., 2016). This aligns with the idea that an unstable but firm small surface can induce movement and elevate activation levels in the mediolateral plane, making therefore the BB an interesting tool to achieve this. Furthermore, the findings of the present study were consistent with the results of Alfuth & Gomoll (2018), who observed differences in the TA and PL with rearfoot stable and forefoot unstable in the frontal plane (equivalent to our Fore configuration) in relation to the floor. However, their study did not analyze variations in forefoot instability from Fore to Sup or Pro (i.e., inducing supination or pronation of the forefoot), nor did they include Total instability configuration or the variations of the rearfoot instability configurations (i.e., Inver or Ever). Therefore, the results presented in this study suggest that other configurations, such as Sup or Total for TA, Pro or Ever for PL, and Pro or Total for PB, could be as effective as Fore in activating each muscle, and may have some additional specific clinical applications, as discussed below.

Muscular patterns between conditions

There are two relevant findings regarding the specificity of Sup (forefoot supination) and Ever (rearfoot eversion) muscular patterns. First, the choice of Sup configuration seems appropriate when the aim was to activate TA while maintaining activation levels of the peroneus, its antagonists, similar to those produced on the floor (Fig. 3–E). This is significant since, to the best of our knowledge, there are no other unstable devices capable of producing a selective activation of TA without, at the same time, the increased activation of the peroneus muscles. Second, the activation of the peroneus muscles in Ever was higher than that of the other muscles under study, especially in the case of PL, which was significantly higher than TA (Fig. 3–D). This observation is similar to the findings of Alfuth & Gomoll (2018) who reported differences in PL activation between a stable forefoot and an unstable rearfoot (similar to our Rear configuration) compared to the floor. In the present study, the differences in Ever configuration were not only existent compared to the floor but also to Rear configuration (inversion-eversion of the rearfoot). It is important to highlight that maintaining balance on the BB configured in Ever induced a demand on the subject similar to that generated in Pro condition (forefoot pronosupination), as the movement primarily requires foot pronation and eversion. However, the peroneus muscles in Ever showed greater activation than TA, which did not occur in Pro configuration. Thus, the findings of this study suggest that an eversion action with a fixed forefoot facilitates the specific activation of PL, minimizing the activity of the other stabilizing muscles. Therefore, Ever configuration acts in a specific way and may be clinically relevant when selective activation of the peroneus is desired, above that of the other muscles under study. Thus, although the configurations Fore, Pro, and Total may be effective for activating TA and the peroneus muscles, working on the BB in its Sup instability variation or with the rearfoot configured in Ever may offer advantages that should be considered in clinical practice.

Strengths, limitations and clinical applications

Despite the interesting findings of the present study, several limitations need to be acknowledged. First, the study relied solely on EMG data and analyzed only six superficial lower-leg muscles. Including kinematic data from the BB or analyzing additional muscles, such as the tibialis posterior, would have been of great interest. Second, the sample consisted only of healthy subjects, making it controversial to extrapolate the results to injured populations, as it is known that ankle stabilizers in individuals with chronic ankle instability (CAI) exhibit different activation patterns compared to healthy individuals (Delahunt et al., 2010; Suda & Sacco, 2011). Third, the analysis was performed only during a single-leg stance exercise. Clinicians should ensure that subjects maintain the position to use these findings as a reliable reference. Lastly, since our criteria did not include any specific requirements regarding participants’ vision, it is possible that some may have had difficulty seeing the fixed point during the measurement, which could have altered their balance during the tasks, although no such issues were reported. Thus, future research should aim to replicate these findings with injured samples and include additional muscles while also analyzing movement. Including other exercises in the analysis, such as single-leg squats or forward lunges, could maximize the applicability of the results.

The variety of findings obtained in this study suggests diverse clinical applications. The ability to specifically activate TA could be useful in cases of TA weakness or in foot drop condition (Stewart, 2008; Waseem et al., 2023). Furthermore, the established relationship between delayed electromyographic activation of the peroneal muscles and ankle sprains (Linford et al., 2006; Hopkins et al., 2009), as well as the lack of peroneal activation in patients with CAI (Hopkins et al., 2009), suggests that incorporating the BB in a prevention training program could increase the activation of these muscles. In addition, a recent study suggests the conservative treatment of hallux valgus through contraction of PL (Ikuta et al., 2024), so the BB could be an interesting tool for addressing this condition. Additionally, the selective activation of desired muscles may prevent overuse and fatigue of the others, which is crucial during the early stages of rehabilitation or for patients with significant muscle weakness. Moreover, physiotherapists can utilize the findings from this study to determine how to work with the BB, specifically by selecting the appropriate configuration to achieve the desired activation pattern of the muscles. Overall, the findings of this study demonstrate that the BB’s specific design effectively activates foot and ankle stabilizers, making it a valuable tool in rehabilitation programs that require controlled and selective muscle activation.