The role of the angle of the fibularis longus tendon in foot arch support

Understanding the contribution of the fibularis longus tendon to the support of the midfoot arches has potential therapeutic applications. This cadaveric study sought to quantify this support across both the transverse arch and medial longitudinal arch and to establish whether a correlation exists between this support and the angle at which the tendon enters the sole.

Loss of the normal arches, in particular the medial longitudinal arch (MLA), is implicated in a number of important clinical conditions such as acquired pes planus. It is thought to occur in 20-30% of the population to some degree, and may often be an asymptomatic physiological variant in some individuals (Raj, DeCastro, & Kiel, 2019).
However, pes planus may be associated with symptoms such as pain on walking, and corresponding tendinitis of the long tendons which pass under the collapsed foot arches secondary to abnormal excursion. Pes planus is commonly caused by rupture of the plantar aponeurosis (Standring, 2005), and is often observed in women over the age of 40 due to natural tibialis posterior degeneration (Kohls-Gatzoulis et al., 2004). However, loss or weakness of any muscle supporting the arches may also result in this condition.
The tendons passing deep to the sole of the foot are important in maintaining the normal foot arches. Among these, the fibularis longus (FL) tendon is unique. The FL muscle lies in the lateral compartment of the leg, and forms a tendon that passes posterior to the lateral malleolus. It then courses into the sole of the foot, deep to the long plantar ligament, to cross the sole obliquely (Figure 1a,b). The tendon inserts onto the lateral aspects of the bases of the first metatarsal and medial cuneiform (Standring, 2005). Variations in its insertion have been noted, including slips onto the bases of the second and fifth metatarsals or a wide insertion across the base of the first metatarsal (Gomes et al., 2019). The FL primarily acts to plantarflex the ankle and first tarsometatarsal joints, as well as everting the foot at the midtarsal and subtalar joints.
The FL tendon is an extrinsic tendon that crosses the sole obliquely, generating tension across the joints of both transverse and longitudinal arches. Its oblique nature creates an angle between the FL tendon and the midline of the foot, which is referred to throughout as the angle of the FL tendon on the sole (θ), which may alter the support provided by FL to the transverse foot arch (TFA) and the MLA ( Figure 1b). While some evidence exists to suggest that the oblique passage of the FL tendon across the sole assists in supporting the F I G U R E 1 (a-d) left foot. (a) Lateral view. Origin of fibularis longus muscle and formation of its tendon (blue) posterior to lateral malleolus. Tendon passes deep to fibular retinacula (red). (b) Plantar view. Insertion of fibularis longus tendon, it passes deep to long plantar ligament (beige). θ, angle of fibularis longus tendon on sole, measured between midline of foot (green line) and line passing through fibularis longus tendon (orange line). (c) Plantar view. Placement of markers. A, midpoint of calcaneus posteriorly; B, medial aspect of medial cuneiform; C, midpoint of base of fifth metatarsal. AB = longitudinal distance between markers (A) and (B), representing medial longitudinal arch. BC = transverse distance between markers (B) and (C), representing transverse arch. (d) Plantar view. Deep dissection. Markers (pins) seen [Color figure can be viewed at wileyonlinelibrary.com] midfoot arches (Standring, 2005), it is often limited and the effects of the θ has not been extensively explored.
This preliminary cadaveric study aims to identify whether simulated FL tendon contraction acts to support the MLA and the TFA, and to quantify any effect that the θ has on this support.

| Specimen preparation
Eleven human cadaveric specimens of the lower leg and foot (distal to the midpoint of the tibia) were provided by Human Dissection Room, Anatomy Building, Department of Physiology, Development and Neuroscience, University of Cambridge, UK. The donors (5 male, 6 female; mean (SD) age = 78.5 (12.5) years, Table 1) had provided consent for anatomical research prior to decease in compliance with the Human Tissue Act 2004 and were embalmed using a vascular technique (with a solution of water, ethanol, formaldehyde, and menthol). None of the donors had undergone previous foot surgery or had any recorded foot pathologies. The foot specimens were dissected to a deep level, defined as removal of skin and fascia of the sole, all intrinsic plantar foot muscles, soft tissue from the dorsum of the foot and all long tendons except for the FL tendon. The FL tendon was sectioned proximal to the lateral malleolus and pretensioned with weights prior to testing.
All specimens were stored at the same temperature (20 C).

| Markers
Markers (pins) were inserted on the sole of each specimen to uniformly identify the MLA and the TFA (Figure 1c

| Experimental apparatus
In order to mimic conditions during the mid-to terminal-stance phase of gait, the calcaneus of each specimen was secured by a vice with the ankle dorsiflexed. Each specimen was secured with the sole facing upward which allowed observation of the markers and a modified vice allowed access to the FL tendon. The sole of the specimen was levelled in the horizontal plane, aligning the metatarsophalangeal joints with the T A B L E 1 Donor characteristics and parameter estimates Note: Parameter estimates were obtained using the means of two replicate data sets. CV represents the coefficient of variation for Δ dist150 calculated from the two replicate measurements separately. Bold italic indicates that the estimate from the second replicate is lower than the first and italic that the estimate from the second is higher than the first. FL, fibularis longus; θ, angle of fibularis longus tendon on sole; B, asymptotic baseline distance when force = ∞; k, force constant defining relationship between force and distance; Δ dist150 , predicted distance change between 0 and 150 N.
calcaneus as these are the major pressure points on weight-bearing over a wide range of weight-bearing percentages (Jones, 1941;Shelton et al., 2019). By the use of a laser level (Figure 2), each specimen was aligned into the anatomical position defined by Renton (1991), that is, a line passing from the midpoint of the calcaneus to between the heads of the second and third metatarsals. A camera (iPad) was positioned above the specimens to take scaled photographs of the marker positions, with a ruler included in each photograph, allowing for subsequent analysis of angles and distances.

| Simulating FL contraction
The proximal end of the sectioned FL tendon was secured to weighing scales, from which a bucket was hung. In addition to the force exerted by the bucket and scales on the FL tendon, a range of 0-150 N (in intervals of 10 N) was applied to the tendon by adding weights to the bucket. This range sought to simulate human physiological FL tendon contraction and was derived from studies demonstrating the peak eversion torque of the human foot (Kaminski, Perrin, & Gansneder, 1999;Zhao, Tsujimoto, Kim, Katayama, & Tanaka, 2018),  (Figure 3; Table S1), which required measurements of the angles at the directional changes and frictional coefficient at the bone-muscle interface. The angle of the (first) directional change at the ankle joint was fixed at 47.5 and the (second) directional change, as the FL tendon crosses onto the sole, was measured to the nearest five degrees. The frictional coefficient was estimated at 0.035 (Amadio, 2005;Uchiyama, Coert, Berglund, Amadio, & An, 1995).

| Measurements
Scaled photographs of the specimens were taken on each addition of weight and subsequently analyzed using Fiji 2.0.0 to allow for accurate measurements of the θ and arch distances (Schindelin et al., 2012). The AB longitudinal distance and BC transverse distance were measured at each weight interval, while the θ was measured at 0 N and assumed to remain constant during testing. The θ was measured from the midline of the foot to a line passing between the midpoints of the FL tendon at its insertion and as it bends onto the sole.
Only the sole was visible on the photographs, and the angles of the directional changes of the FL tendon were instead measured using a protractor. All measurements were made by the same observer. Seven sets of measurements were performed on the first specimen (male, right side, 68 years) to refine methodology and establish reproducibility. Two out of these seven measurement sets used the same protocol subsequently used for the remaining specimens and generated values for the predicted distance change provided by a force of 150 N (Δ dist150 ) as follows: for the TFA, 2.9 and 3.2 mm (coefficient of variance, CV = 6.0%); for the MLA, 7.3 and 7.4 mm (CV = 1.5%). This reproducibility and the observation that repeated measurements gradually damaged the specimens led us to decide that two replicates per specimen was an optimal practical design for the remaining specimens.

| Statistical analysis
Having measured the arch distances at 0-150 N on two separate occasions, the values were averaged and the force-arch (or AB and BC) distance relationships of each specimen were modeled using the following equation: y F = (y 0 − B).e −kx + B, where: y F = predicted distance when force = x, x = force applied, y 0 = predicted distance when x = 0, B = asymptotic baseline distance when x = ∞, k = a force constant that defines the relationship between force and distance ( Figure 4). The parameters y 0 , B and k were estimated using a least squares minimization approach (Table 1), and were used to plot the modeled line allowing comparisons of the arch distance changes across all specimens. The predicted change in distance caused by a force of 150 N (Δ dist150 ) was calculated for each specimen as Δ dist150 = y 0 −y 150 (Table 1). Simple linear regression (r) was performed on the θ and Δ dist150 for both arches, and the statistical significance of each correlation was determined. The significance level was p < .05. F I G U R E 5 Relationship between angle of fibularis longus tendon on sole (θ) and predicted distance change provided by a force of 150 N (Δ dist150 ) for both the transverse foot arch (TFA) and medial longitudinal arch (MLA). TFA: slope ± s.e. = 0.56 ± 0.13 mm/degree, r = .83, p = .002; MLA: 0.18 ± 0.18 mm/degree, r = .32, p = .33 (Konradsen, Voigt, & Hojsgaard, 1997). Therefore, the FL tendon may have an important role in stabilizing the foot arches. Bojsen-Moller (1979) described that when the foot joints are loosely packed, they are unstable and to this extent are unsupported; however, when the foot joints become tightly packed, they are supported. It is through this mechanism, by increasing the tension and apposition between the midfoot joints, that the FL tendon contributes to the maintenance of the normal foot arches.

Measurements
In this study, simulated FL tendon contraction in cadaveric specimens decreased the TFA and MLA arch distances confirming that the tendon supports both arches. The arch distance changes decreased with higher forces, most likely because the joints comprising the arches had reached the limits of their movements.

| Fibularis longus and medial longitudinal arch
The role of FL in supporting the MLA is controversial. It has been noted that the function of FL in raising the MLA is negligible mainly because the moment it produces across the joint between the first metatarsal and medial cuneiform is small compared with other leg muscles such as tibialis posterior (Angin, Crofts, Mickle, & Nester, 2014), and a cadaveric study has questioned the importance of this tendon in providing support to the MLA during the stance phase of gait (Dullaert et al., 2016). In addition, Sharkey, Ferris, and Donahue (1998) have shown no significant role of FL in supporting the foot arches at the end of the stance phase, and instead argued that the plantar aponeurosis is the key supporter of the arches. The plantar aponeurosis is thought to support the foot arches through the Windlass Mechanism and provides stiffness to the foot during locomotion (Bolgla & Malone, 2004;Hicks, 1954). Other studies suggest that the FL tendon supports the MLA. Thordarson, Schmotzer, Chon, and Peters (1995) showed that the FL tendon provides deforming forces to the MLA during the stance phase of gait, and its contraction causes locking of the first metatarsal ray and thus provides stabilization to the MLA (Bierman, Christensen, & Johnson, 2001;Johnson & Christensen, 1999). Electromyographic studies also confirm its role in MLA support during the stance phase of gait, as the activity of FL is reduced in patients with pes planus compared to patients with normal arches (Hunt & Smith, 2004;Murley, Menz, & Landorf, 2009).
Despite acquired flatfoot deformities (AFFDs) of the MLA often being associated with tibialis posterior or spring ligament defects (Arain, Harrington, & Rosenbaum, 2019), their pathogenesis and subsequently their management remain unclear and controversial (Tao, Chen, & Tang, 2019). This study suggests that the FL tendon supports the MLA, and may provide a novel therapeutic target in the management of AFFDs.

| Fibularis longus and transverse arch
Less controversial is the role of the FL tendon in supporting the TFA, which is also evident in this study. The oblique nature of the FL tendon, that is, more transverse than longitudinal in the sole, preferentially provides support in the transverse plane. The importance of the TFA has been recently highlighted by Venkadesan et al. (2020), as they have shown that it contributes to 40% of the stiffness of the foot, which aids to reduce flatfoot during locomotion. They note that this knowledge may help in the management of flatfoot disorders.
Therapeutically targeting the FL tendon in TFA arch collapse could help to ameliorate pes planus.

| The angle on the sole
A significant positive correlation was only observed between the θ and the TFA distance changes (r = .83, p = .002). Therefore, the θ may alter the support that the FL tendon provides to the TFA. Increases in the θ results in increases in the TFA distance changes and support provided to this arch by the FL tendon.
Knowledge of the θ allows consideration of novel therapeutic interventions to surgically re-position the FL tendon for maintenance of the TFA, for example, in cases of TFA collapse not responsive to muscle training or orthotic support. Basit, Eovaldi, and Siccardi (2019) indicated that the FL and fibularis brevis tendons are the most commonly dislocated in the ankle, and in these cases, it may be reasonable to surgically re-position and fix the FL tendon in favor of supporting the TFA.

| Limitations
This study is restricted by a number of limitations, most notably its small sample size. This study used formalin-fixed specimens, known to reduce the range of motion of human joints and have fixing effects on tendons (Balta et al., 2019;Balta, Cronin, Cryan, & O'mahony, 2015).
In addition, it has been shown that the Young's Modulus (stiffness) of formalin-fixed tendons is significantly higher than that of fresh-frozen specimens (Hohmann et al., 2019). Therefore, our study may not offer an ideal representation of the behavior of the FL tendon in nonembalmed cadavers and living subjects, and should be interpreted accordingly. This study was not able to account for the effects of weight loading which would otherwise act during the stance phase of gait. Loading the specimens with weight would flatten the MLA (Shelton et al., 2019), which would counteract the distance changes observed. By using cadaveric specimens, we were able to isolate the FL tendon, but this was at the expense of losing local surrounding structures such as the plantar aponeurosis. The effects of losing this aponeurosis, however, may have been circumvented as weightloading was not considered. The lateral aspect of the specimens was fixed using a plastic block, which helped to reduce any misalignment on progressive simulated FL tendon contraction. Loading the FL tendon at 150 N, however, stressed the experimental apparatus and some specimens lost the alignment provided by the laser level. The results of this study should be regarded as preliminary and should be verified with further rigorous testing.

| CONCLUSIONS
This study provides evidence that the FL tendon has a supportive effect on both the TFA and the MLA, as on its simulated contraction, the distance of both arches reduced. While the effect on the TFA is not controversial, the supportive effect of FL on the MLA is more contentious. This study indeed demonstrates that while the effect of FL on the TFA is related to the angle at which the tendon enters the sole, the same does not appear to hold true for the MLA. Targeting the FL tendon may provide a novel physical (or surgical) method in the management of AFFDs. This preliminary study provides a foundation to further investigate this unique tendon and its actions on arch support.
Future studies should quantify its effects on the arches in live participants during gait and weight-loading activities, and compare its supportive effect to other key arch supporters, namely tibialis posterior and the plantar aponeurosis.

ACKNOWLEDGMENTS
We thank the donors for allowing cadaveric research to be possible.