Physical Therapy Korea

Eversion is the movement of the ankle joint that rotates the foot outward along the anterior-posterior axis. This complex motion involves the concerted action of talocrural, subtalar, and transverse tarsal joints, with a predominant kinematic pattern observed in the subtalar joint [1]. The peroneus longus (PL) and peroneus brevis (PB) are attached to the 1st metatarsal bone and 5th metatarsal bones, respectively, and function as primary muscles of eversion [2]. In general, structural stability is ensured on the medial side of the ankle through deltoid ligament [3], whereas on the lateral side of the ankle, the PL and PB provides primary dynamic defense against the lateral ankle sprain [4,5]. Therefore, when considering evertor muscle strengthening methods for ankle stability, exercise method to effectively use PL and PB was needed.

PL and PB muscles crossing the ankle joint are attached to the tibia; thus, both muscles pull the foot outward and form a mechanical coupling between the talus horizontal rotation and the leg rotation [1]. Previous study has reported that tibial rotatory torque produced by biceps femoris (BF) muscle activation can influence evertor torque output [6]. Due to its anatomical structure, the BF rotates the tibia laterally when knee is flexed [7,8]. As the BF is a knee flexor, muscle activation of the BF can be affected by the knee joint position [6,9]. According to Lentell et al. [6], BF activation and isokinetic eversion torque were both higher in the knee flexion (KF) position compared to the knee extension (KE) position during isokinetic eversion exercise. At KE position, ligamentous stability could be increased, and accessory rotatory motion of the knee joint could be diminished [10]. However, tibial rotation also differs based on the varying ankle positions during eversion exercise. In the ankle plantarflexed (PF) position, calcaneal inversion and tibial external rotation increased by about 3°–5° [11]. Lee et al. [12] investigated BF and peroneal muscle activation during side-lying isometric eversion exercise, reporting that the higher PL and PB muscle activation in PF position compared to the ankle neutral (N) position with low-moderate BF muscle activation. Therefore, both knee and ankle position can impact ankle eversion strength measurements. However, previous studies did not consider whether the ankle and knee position can affect the muscle activation of the BF and peroneal muscles during isometric eversion exercise [6,12,13].

Isometric eversion exercise can be useful in the early stages of the ankle rehabilitation [14]. To perform an effective isometric eversion exercise, there are many options for exercise positions. Several studies have selected the sitting position [12,15], long-sitting [16] or supine position [17], or side-lying position [18]. Among them, side-lying position can generate more muscle activation compared to sitting position [12]. The side-lying position is more convenient to provide resistance than the supine or sitting positions because the examiner utilizes the direction of gravity when providing resistance. Therefore, a study to identify appropriate ankle and knee position to activate the peroneal muscles would be valuable for implementing effective side-lying ankle eversion exercise.

The purpose of this study was to identify an effective side-lying eversion position for activating the PL and PB and to compare BF muscle activation based on ankle positions (N and PF) and knee positions (KF and KE) during side-lying isometric eversion exercise. We hypothesized that the PL and PB activation would be higher in the PF position, and that differences in muscle activation of the PL, PB, and BF would exist according to knee positions.

1. Participants

Thirty healthy individuals (16 males, 14 females; mean age = 24.8 ± 3.1 years; mean height = 170.0 ± 7.2 cm; mean body mass = 66.7 ± 10.0 kg) participated in the study. The inclusion criteria required participants to have no history of lateral ankle sprain within the last 6 months. The Ankle Joint Functional Assessment Tool (AJFAT) was used to determine the extent of ankle joint instability. Individuals achieving an AJFAT score of 22 or above are categorized as having no apparent indications of instability [19,20]. The average AJFAT score of participants was 28.1 ± 6.2. When both feet met the criteria, the foot kicking the ball was selected for measurement [21]. The number of recruited ankles was 26 on the dominant side and four on the non-dominant side, translating to 25 right feet and 5 left feet. The exclusion criteria were inability to complete the test due to any illness or pathology affecting neuromuscular control, a history of lower extremity surgery [22], a history of ankle sprain more than twice, recent lower extremity injuries within the past 6 months, or a body mass index of ≥ 30 kg/m2 [23].

Sample size for this study was determined using the G*Power software (ver. 3.1.9.2; Heinrich Heine University D_üsseldorf), with an effect size of 0.046 (partial eta squared of the knee × ankle interaction effect), derived from PL activation data from a pilot experiment involving seven participants (effect size of 0.22, α = 0.05, and power of 0.80). Since PL is the primary muscle of eversion and interaction effect size is judged to be an important value in obtaining information about the interaction between knee and ankle positions, the knee × ankle positions interaction effect size of PL muscle activation was used in sample size calculation. It was estimated that 30 participants were needed. The present study protocol was reviewed and approved by the Yonsei University Mirae Institutional Review Board (IRB no. 1041849-202309-BM-177-03). All participants were apprised of the protocols and objectives and a signed consent form was obtained from them when they were enrolled.

2. Smart KEMA Strength Measurements

For the strength test of ankle eversion muscles, the Smart KEMA strength sensor (KOREATECH, Inc.) was used (Figure 1A). The screen of the strength sensor displays the numerical tension (kgf) pulling from the top and bottom of the sensor, with good to high intra-rater reliability (ICC_3,1 > 0.85, ICC_2,1 > 0.85) [24,25]. The measured value is transmitted to the tablet via Bluetooth and displayed in real-time using the Smart KEMA application (KOREATECH, Inc.) (Figure 1B). In the application, the value of strength sensor was collected for 5 seconds, and the average of the middle 3 seconds was calculated. Each end of the strength sensor was connected to a 5-cm wide strap and length-adjustable nonelastic belt. The strap was fastened around the metatarsal bone, and the belt was fixed to the floor using an adsorber. The belt was tightened until a force of 2 kgf was applied to control initial tension.

Figure 1. Smart KEMA strength sensor (KOREATECH, Inc.) for measurement of evertor strength (A) and Smart KEMA application (KOREATECH, Inc.) displaying real-time values of the strength sensor (B).

3. Surface Electromyography

Muscle activation of the PL, PB, and BF were recorded using Tele-Myo DTS electromyography (EMG) with a wireless telemetry system (Noraxon Inc.) at 1,000 Hz. Data analysis was conducted using MyoResearch XP Master Edition software (Noraxon Inc.). Before electrode placement, the attachment sites of the skin were shaved and cleaned with alcohol swabs. Separate bipolar (Ag/AgCl) surface electrodes were attached to the PL, PB, and BF muscles at 2 cm intervals, following the recommendations of the Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles recommendations [17,26]. For the PL and PB, electrodes were positioned at one-fourth and three-fourths of the fibular length, respectively. Muscle activation during ankle eversion in N and PF positions was identified to ensure activation in different muscles (PL and PB) [17]. For BF, electrodes were placed on the lateral thigh, two-thirds of the distance between the greater trochanter and knee joint [23]. Data were band-pass filtered (10–500 Hz) and smoothed using a 150-ms moving window [18,27].

4. Procedures

Participants were instructed to perform a 3-minute warm up involving walking at a self-selected speed. Four eversion measurements were taken based on knee and ankle positions as follows: knee flexion and ankle neutral (KFN), knee flexion and ankle plantarflexed (KFPF), knee extension and ankle neutral (KEN), and knee extension and ankle plantarflexed (KEPF); the order of measurements was randomized (Figure 2). Participants were instructed to perform side-lying eversion with the medial malleolus as the fulcrum to prevent hip rotation or abduction. In the side-lying position, the measurement side was set as the upper leg. The KF position was defined as adoption of hip and knee joint angles of 90°. The KE position was defined as fully extended knee joint and minimal hip flexion without overlapping of the lower leg, with a pillow placed between the legs. Ankle angles were designated as the N position (0° of dorsiflexion and 0° of eversion) and PF position (50° plantarflexion and 0° of eversion) [17]. The initial positions of the hip, knee and ankle were confirmed using a standard goniometer. During eversion, participants maintained their toes in a N position without any flexion or extension.

Figure 2. Participant set up for eversion in a side-lying position with knee flexion (A), and with knee extension (B).

Initially, the four types of maximal eversion were randomly performed twice for 5 seconds each. Maximal eversion strength and muscle activation measurements were collected simultaneously. Then, 70% of the lowest eversion strength value was determined as the submaximal contraction among four different strength tests. Participants were provided with a tablet displaying real-time strength sensor values. The four types of submaximal eversion were randomly performed twice for 5 seconds each. Muscle activation was recorded during the same weight of submaximal eversion. Comparing muscle activation of PL, PB and BF under the same strength with submaximal isometric eversion exercise can reduce the possibility that muscle activation differences may be due to the effort level of participants and could allow effective understanding of the alterations of the mechanism based on different ankle and knee positions. Therefore, data regarding muscle activation of PL, PB, and BF were collected during submaximal isometric contraction in this study. To mitigate potential muscle fatigue effects, a 10-second intertrial interval was provided, and a 1-minute break was allowed between position changes. The maximal eversion strength value was normalized to body mass. The EMG data collected during submaximal isometric eversion exercise were normalized to the maximal voluntary isometric contraction (MVIC). MVIC was determined as the highest value among the four types of maximal muscle activation.

5. Statistical Analysis

The IBM SPSS Statistics software (ver. 23.0, IBM Co.) was used to perform statistical analyses. A 2 × 2 repeated measures analysis of variance was performed to investigate differences in muscle activation and strength. Within-group factors include the knee position (KF and KE positions) and ankle position (N and PF positions). If significant interaction effect was found, a paired t-test was used to compare the specific differences between variables. The level of significance was set at p < 0.05. To identify the size of the influence of the independent variable on the dependent variable, partial eta squared (${\eta }_p^2$) was reported to estimate the effect size (small effect: 0.01–0.06, medium effect: 0.06–0.14, and large effect: > 0.14) [28].

The PL and PB muscle activation showed no significant interaction effects between knee and ankle positions (Table 1). The PL muscle activation showed significant main effects based on knee and ankle positions (Figure 3A), and PB muscle activation showed significant main effects based on knee and ankle positions (Figure 3B). Both PL and PB muscle activation were significantly greater in the KE and PF positions compared to KF and N positions (p < 0.05) (Table 2). The ${\eta }_p^2$ for the main effects of PL and PB showed large effects.

Table 1 . Main and interaction effects according to knee and ankle positions.

Variable	Main effect: knee position				Main effect: ankle position				Interaction effect: knee × ankle
	F(1,29)	p-value	ηp2		F(1,29)	p-value	ηp2		F(1,29)	p-value	ηp2
PL	5.627	0.025*	0.163		78.001	<0.05*	0.729		1.967	0.171	0.064
PB	7.501	0.010*	0.205		29.708	<0.05*	0.506		1.559	0.222	0.051
BF	5.304	0.029*	0.155		43.973	<0.05*	0.603		12.914	0.001*	0.308
Maximal evertor strength	0.946	0.339	0.032		66.886	<0.05*	0.698		0.235	0.632	0.008

PL, peroneus longus; PB, peroneus brevis; BF, biceps femoris. *p < 0.05..

Table 2 . Muscle activation and maximal evertor strength during four types of ankle eversion exercise.

Variable	KFN	KFPF	KEN	KEPF
PL	30.7 ± 14.8	49.4 ± 22.8	34.9 ± 20.6	59.1 ± 23.6
PB	41.0 ± 15.8	52.2 ± 16.1	46.0 ± 14.7	61.4 ± 17.1
BF	17.5 ± 10.2	23.0 ± 11.9	20.0 ± 16.4	33.0 ± 23.4
Maximal evertor strength	19.0 ± 7.32	12.6 ± 3.9	18.1 ± 7.0	12.2 ± 5.0

Values are presented as mean ± standard deviation. KFN, knee flexion and ankle neutral; KFPF, knee flexion and ankle plantarflexed; KEN, knee extension and ankle neutral; KEPF, knee extension and ankle plantarflexed; PL, peroneus longus; PB, peroneus brevis; BF, biceps femoris..

Figure 3. Muscle activation according to knee and ankle positions. (A) PL, (B) PB, and (C) BF. PL, peroneus longus; PB, peroneus brevis; BF, biceps femoris; PF, plantarflexion; N, neutral; MVIC, maximal voluntary isometric contraction; KF, knee flexion; KE, knee extension.

BF muscle activation showed a significant interaction effect between knee and ankle positions (Table 1). The ${\eta }_p^2$ for the main and interaction effects of the BF showed large effects. Post-hoc paired t-test showed that BF muscle activation was significantly greater in the PF position than in the N position, regardless of knee position (p < 0.05) (Figure 3C). In the PF position, a significant difference was found between the KF and KE position (p < 0.05). However, in the N position, no significant difference was found between knee positions (p > 0.05). BF muscle activation was greater in the KEPF position than in the KFPF position (p < 0.05) (Table 2).

Evertor strength showed a significant effect only in the ankle positions (Table 1). The ${\eta }_p^2$ for the main effects of ankle position on evertor strength showed a large effect. Evertor strength was greater in the N position than in the PF position (Figure 4).

Figure 4. Maximal evertor strength according to knee and ankle positions. PF, plantarflexion; N, neutral; KF, knee flexion; KE, knee extension.

The purpose of this study was to investigate the differences in PL, PB, and BF muscle activations according to ankle and knee positions. Overall, we found that PL and PB muscle activation were higher in the PF and KE positions than in the N and KF positions. However, maximal eversion strength was greater in the N position than in the PF position. BF muscle activation was higher in the PF position than in the N position, regardless of knee positions and in the KEPF position than in the KFPF position.

To calculate the sample size, we used the knee and ankle positions interaction effect of PL muscle activation as the effect size. Since the PL muscle length varies depending on the degree of tibial rotation, and tibial external rotation can induced by the ankle joint PF [11], we thought there would be an interaction effect between the knee and ankle positions. However, there was no significant interaction effect between the knee and ankle positions in PL muscle activation. The reason may be that tibial external rotation by the ankle joint PF was insufficient to elicit significant changes in PL muscle length. Muscle activation can be influenced by changes in muscle length, and 3°–5° of tibial rotation caused by PF may not induce significant differences in PL muscle activation (PL muscle length).

Our study showed that peroneal muscles activation exhibited approximately 10–20 %MVIC greater in the PF position than in the N position. These results were similar to the results of previous study that reported a 9.2 %MVIC greater peroneal muscle activation and 0.37 Nm/kg lesser eversion force in the PF position than in the N position during isometric eversion [17]. Donnelly et al. [17] measured eversion torque using a hand-held dynamometer, and the eversion performance of participants may have depending on the level of resistance provided by the examiner. This may dilute the difference of peroneal muscle activation between the N and PF positions. Similarly, Hintermann et al. [29] found that evertor function was reduced in the N position than in the PF position due to decreased evertor moment arm, as the ankle is dorsiflexed with the distal and lateral displacement of the instantaneous rotation axis of the ankle joint.

However, contrary to muscle activation, evertor maximal strength was greater in the N position compared to the PF position. Considering these results, we propose that the influence of the peroneus tertius or extensor digitorum longus, serving as foot evertor and ankle dorsiflexor, may contribute to this phenomenon. In our study, we attempted to eliminate the effect of the extensor digitorum longus by preventing participants from extending their toes; however, the peroneus tertius could not be controlled in the N position. Therefore, it is possible that peroneus tertius muscle activation increased compared to the level of PL and PB muscle activation in the N position to achieve the same tension of eversion. Although conflicting opinions as to whether peroneus tertius has a meaningful effect on the eversion function exist [30], results similar to our study have been consistently reported [12,13,17]. Additionally, in the N position, the length of the PL and PB muscles increases due to the tendon arrangement that passes behind the lateral malleolus [31]. As PL and PB muscle is lengthened, peroneal muscle fibers may act like an elastic band that passively produces increased force with stretch. An increase in the force produced by passive muscle component makes a decrease in force developed by the contractile elements [32]. Therefore, fewer motor units would be needed due to the reduced requirements in the N position to perform a submaximal eversion of same tension. This may explain the mechanism of higher eversion strength and lower peroneal muscle activation in the N position.

In our study, the PL and PB muscle activation exhibited approximately 5–10 %MVIC greater in the KE position than in the KF position. In the KE position, locking of the knee joint changes the hip-knee-ankle complex movements, which may lead to differences in peroneal muscle activation. For the KE eversion to withstand the same weight as KF eversion, a relatively movable hip or ankle joint may be rotated instead of the fixed knee joint. Souza et al. [33] reported that hip internal rotation mobility has a significant relationship with weight-bearing rearfoot eversion. Although we did not investigate the hip and subtalar joint rotation ratio, it is possible that an internally rotated hip joint may cause more ankle eversion. Further investigation will be needed to investigate the influence of the hip and ankle joint rotation during side-lying eversion.

BF muscle activation exhibited approximately 10–15 %MVIC greater in the KEPF position than in other positions; notably, in the PF position, the BF showed higher muscle activation with KE than with KF. During KE, the “screw-home” locking mechanism in the knee joint, characterized by tibial external rotation, adopts a locked position [1]. This locking mechanism increases knee joint congruence and ensures joint stability by creating a contact area of approximately 375 mm² at the medial tibiofemoral joint and 275 mm² at the lateral tibiofemoral joint [34]. Accordingly, a tibial external rotation of approximately 10° is generated [35]; thus, BF muscle activation could be increased as a tibial external rotator in KE. However, a previous study reported that BF muscle activation exhibited approximately 6–12 %MVIC higher in KF than in KE during isokinetic eversion [6]. Our study results showed that BF muscle activation exhibited 2.5–10 %MVIC greater in the KE position than in the KF position. This disparity in findings may be attributed to the influence of ankle position. Eversion is accompanied by a slight amount of foot abduction [36], talus external rotation-tibial external rotation showed a strong relationship [37]. In the N position, the locking-in effect of the talus on ankle mortise is reduced [29]. Lentell et al. [6] compared the BF muscle activation during KFN and KEN eversions, an unlocked talocrural joint (N position) would require relatively less tibial rotation. In other words, our results can be demonstrated by the hypothesis that locked talocrural joint in the PF position may have required more tibial external rotation to achieve the same weight of isometric eversion as in the N position.

The KEPF position showed the highest PL, PB, and BF muscle activation in the present study. Muscle activation and contraction are closely related [23], and four classifications have been proposed based on the level of muscle activation as follows: < 20 %MVIC = low muscle activation, 20–40 %MVIC = moderate muscle activation, 41–60 %MVIC = high muscle activation, > 60 %MVIC = very high muscle activation [38,39]. Analysis of four positions (KFN, KFPF, KEN, and KEPF) according to these criteria demonstrated that the PL and PB muscle activation was at a moderate level in the KFN position, moderate to high level in the KFPF and KEN positions, and very high level in the KEPF position. BF muscle activation showed a low level in the KFN position and moderate level in the KFPF, KEN, and KEPF positions. Although BF muscle activation was highest in the KEPF position (33.0 ± 23.4 %MVIC), it remained at a moderate level, similar to other positions. In comparison, the PL and PB muscle activation in the KEPF position was at a very high level, it could be expected to have an effect of improving eversion muscle strength [40,41]. Therefore, we propose the KEPF position to be useful for side-lying isometric eversion measurement method.

This study has several limitations. First, the amount of actual ankle and tibial rotation was not examined. Thus, the occurrence of ankle or tibial rotation may have influenced our interpretation of the difference in BF muscle activation between the KF and KE positions. However, talus rotation or locking of the talocrural joint were difficult to confirm without an in vitro study design. In future studies, the movement of the tibial external rotation should be analyzed through the movement of landmarks, such as the lateral malleolus, to ensure a more reasonable interpretation. Second, it was not certain whether surface EMG signals were collected only from the PL and PB. Since surface EMG electrodes do not penetrate muscles, signals from adjacent muscles, such as extensor digitorum longus, could be collected. To minimize this effect, participants performed only limited toe flexion or extension during eversion. Third, the results of this study could be valid for static eversion function. Ankle instability usually occurs in dynamic situations, and eversion strength or muscle activation should be considered in dynamic measurements.

In this study, evertor strength and muscle activation of the PL, PB, and BF were compared during isometric side-lying eversion. In the KEPF position, evertor strength was the lowest, but muscle activation was highest for all three muscles. In particular, although BF muscle activation was at a moderate level, it was also the only position where the PL and PB muscle activation were at very high levels. In clinical practice, the effects of the knee-ankle complex should be considered rather than simply measuring muscle strength to assess ankle evertor function. Based on these results, we propose the KEPF position as a means of effective isometric eversion measurement.

None.

None to declare.

No potential conflicts of interest relevant to this article are reported.

Conceptualization: DL, OK. Data curation: DL. Formal analysis: DL, JK. Investigation: DL, SH. Methodology: DL, OK. Project administration: DL. Resources: DL, SH. Supervision: JK. Visualization: DL. Writing - original draft: DL. Writing - review & editing: JK, OK.