Introduction
Subtalar instability is a common functional talocalcaneal instability always coupled with lateral ankle instability. It was reported that about 10–25% of patients with lateral ankle instability suffered from subtalar instability [
1,
2]. Subtalar instability mainly results in varus tilt and anterior translation of the calcaneus. However, its clinical manifestation is often covered by the lateral ankle instability, for which, it is only recognized as a separate clinical condition that needs specific treatment in recent years [
3].
The exact aetiology of subtalar instability is still unknown. Several cadaveric studies reported that the main ligaments which stabilize the subtalar joint were the calcaneofibular ligament (CFL), the interosseous talocalcaneal ligament (ITCL) and the cervical ligament (CL) [
4‐
7]. For chronic subtalar instability, conservative therapies such as proprioceptive training, stretching of the Achilles tendon, and prescription of orthosis are the primary treatment. When these conservative treatments fail, surgery is indicated. The earlier operation adopted for subtalar instability was duplicated from the procedures for stabilizing the lateral ankle. Most of them, such as the Elmslie [
8], Chrisman-Snook [
9] and Watson-Jones [
10] procedures, attempt to recreate the CFL and anterior talofibular ligament (ATFL). Theoretically, CFL recreation is effective in stabilizing the subtalar joint because the CFL can bridge the posterior facet of the subtalar joint. However, the ITCL and CL, which play an important role in subtalar stability, were not recreated in these approaches.
Therefore, new procedures including the Pisani ITCL reconstruction [
11], Schon CL reconstruction [
12], Choisne CFL reconstruction [
13], Schon triligamentous reconstruction [
12] and Mann triligamentous reconstruction [
14] were developed.
Any procedures for joint stabilization should be evaluated for biomechanical effectiveness. The goal of these tenodesis reconstructions is generally to mimic the functions of ligaments as far as possible. To the best of our knowledge, for these newly developed tenodesis procedures, their effects in stabilizing the subtalar joint have not been compared before.
Some cadaver studies have evaluated the effects of different ligaments in stabilizing the subtalar joint [
4‐
7]. However, to compare these different types of tenodesis reconstruction, the cadavers to be tested should be strictly the same in terms of their material properties and morphologies, in order to ensure the comparability of the experimental results. Unfortunately, a cadaveric model could be used for only two or three times for the bony framework between each bone tunnels would be unstable after repeated use. The finite element analysis (FEA) is a reliable tool for the quantitative evaluation of biomechanical performance. By using FEA, one model can be tested on the basis of the same material properties and morphologies under several different loading settings. Moreover, FEA can also derive a number of invaluable outputs such as the contact pressure and internal stress, and therefore, it is useful for preoperative planning and surgery assessment. Referencing to a previous study [
15], an FE model of the foot and ankle containing 28 bones (tibia, fibula and 26 ft bones), sesamoids, plantar fascia, 24 main ligaments, and cartilage was developed and used to compare 5 different tenodesis reconstructions over their biomechanical behaviors. The goal of this study is to provide reference for the optimal design of subtalar stabilization.
Discussion
The subtalar joint plays an important role in hindfoot stability. Specifically, the ligamentous contents in the subtalar joint are main structures contributing to its stability. Taillard et al. found that the CFL ruptured first with an inversional force exerted on the cadaveric model, followed by the CL and then the ITCL [
28]. Several procedures have been developed for restoring subtalar instability, including CFL reconstruction, CL reconstruction, ITCL reconstruction, and triligamentous reconstruction. However, no comprehensive comparison on their biomechanical behaviors has been conducted yet. In view of this, we compared these 5 representative procedures in this numerical analysis, and the results suggested that the Choisne CFL and Mann triligamentous reconstructions were most effective in stabilizing the subtalar joint.
Previous cadaveric studies have compared the stabilizing effect of different subtalar ligaments. However, which ligament contributes the most remains controversial. The objective of the present study was not to examine the effect of injury in one of the mentioned ligaments. Thus, the ITCL, CFL and CL were removed simultaneously to simulate subtalar instability. Nevertheless, the results of different tenodesis reconstructions can reflect the contribution of these ligaments.
Pisani stressed the importance of ITCL and developed a procedure to recreate the ITCL in1996 [
11]. A one-half peroneus brevis graft is woven through two calcaneal and two talar tunnels to recreate ITCL. ITCL is a V-shaped ligament composed of two bands, with the fiber oriented from superomedial to inferolateral [
17]. It occupies nearly half of the medial part of the tarsal canal. There is still no agreement regarding the mechanical contribution of ITCL to subtalar instability. Some researchers compared it with the cruciate ligaments of the knee in terms of the stabilizing effect on the subtalar joint [
11,
29]. Knudson reported that the ITCL contributed substantially to supination stability [
30]. Tochigi et al. suggested that the ITCL restrained the subtalar joint in inversion and prevented anteromedial displacement [
31]. However, several publications questioned the importance of the ITCL [
4,
6]. For example, Li et al. pointed out that the most important ligament of the tarsal sinus was the CL, while the ITCL, although always present, was a thin single-band ligament [
32]. Smith and Cahill [
33,
34] highlighted that the ITCL was not an important component for subtalar stability. In our simulation, the Pisani ITCL reconstruction was found to reinforce the inversional stability, but had little effect on rotational stability. Tochigi suggested that the ITCL counteracted against the drawer force on the calcaneus [
31]. The ITCL was located between the posterior and middle facet of the subtalar joint, similar to the cruciate ligaments of the knee. The cruciate ligaments mainly restrain the anterior and posterior translation of the tibia. Similarly, the ITCL may primarily restrain the translational displacement of calcaneus, but contribute little to rotational stability.
The Schon CL reconstruction was developed in 1991 [
12]. One-half peroneus brevis tendon was used in this procedure. Schon believed this procedure was appropriate for mid-to-moderate subtalar instability. The CL is another important ligament of the subtalar joint. Its fiber is oriented from the superolateral calcaneal surface to an inferolateral tubercle on the talus neck. Some researchers believed that the CL was more important than the ITCL in stabilizing the subtalar joint [
2,
3,
32]. Earlier studies found that the CL could restrain the inversion and external rotation of the subtalar joint [
7,
33,
34]. Our study also revealed that the Schon CL reconstruction contributed more to the rotational stability than the ITCL reconstruction did, but its effect on inversional stabilization was the weakest of the three procedures.
Choisne described a tenodesis reconstruction procedure to recreate the CFL in 2016 [
13]. The entire peroneus brevis tendon was used in this procedure. The CFL originates from the fibular tip to the lateral posterior calcaneus and bridges the posterior facet of the subtalar joint subtalar joint. It is widely accepted as an important ligament for ankle stabilization, and its function in relation to subtalar stability has been gradually revealed and confirmed in recent years. Weindel and Karlsson supposed that the CFL played a key role in subtalar stability [
3,
6]. Choisne et al. reported that subtalar instability occurred after section of CFL in isolation in their experiment [
35]. Pellegrini et al. found that the CFL disruption could lead to an increase in inversion and external rotation, which might be detectable during a manual examination [
4]. Our study showed that the Choisne CFL reconstruction was more effective than the Pisani ITCL and Schon CL reconstructions in alleviating the instability of subtalar joint. However, the Choisne CFL reconstruction needs to use the entire peroneus brevis tendon as graft, while the Pisani ITCL and Schon CL reconstructions use only half of a peroneus brevis tendon. This may enhance the stabilizing effect of Choisne CFL reconstruction.
After a comparison of these three procedures, we found that the stress in reconstructed grafts of the 3 single ligament reconstruction all increased compared to the corresponding ligament of intact model in inversional test. On the contrary, the force of ATFL in reconstructed models all decreased. It suggest the reconstructed grafts withstood more stress than the normal ligament. In the long term, these grafts may not be able to withstand the increased internal stress.
For subtalar instability usually coexists with lateral ankle instability [
1,
2], the triligamentous reconstruction procedure was developed for recreating the ATFL, CFL and CL. Schon et al. described a triligamentous reconstruction [
12]. The entire peroneus brevis is weaved through 4 bone tunnels to recreate the CFL, ATFL and CL. Thomas et al. described a triligamentous reconstruction in Mann’s surgery of the foot and ankle [
14]. This technique was designed to correct symptomatic severe subtalar instability with a hamstring tendon graft. In our simulation, the triligamentous reconstruction improved the subtalar stability very effectively without recreating the ITCL. The contact characteristics of the subtalar joint and the stress of grafts in Mann triligamentous reconstruction was more similar to the intact model. The results showed that the Mann procedure had a stronger correcting power in controlling subtalar instability than the Schon procedure did. This may be attributed to the difference in their grafts. The elastic modulus and cross-sectional area of semitendinosus are 1036 MPa and 20mm
2, respectively [
36,
37]. The peroneus brevis tendon has an elastic modulus value of149.7 MPa and a cross-sectional area of about 19.5mm
2 [
25]. Obviously, the strength of semitendinosus is much higher than that of peroneus brevis tendon. However, according to our simulation results, the hamstring tendon is so stiff that it restrains the mobility of the subtalar joint. This might result in detrimental long-term effects on the cartilage [
38]. The increased internal stress in the subtalar joint could accumulate over time, ultimately leading the cartilage degeneration and hind foot stiffness. On the other hand, the peroneus brevis tendon is too elastic for the subtalar joint and the model remains a certain degree of laxity under the Schon procedure. The ideal graft should have the same material properties as the native ligament and mimic the same biomechanical behaviors as the original joint [
36]. Therefore, both types of graft have defects.
In general, although several experiments have evaluated the biomechanical function of subtalar ligaments, the conclusions remain controversial. Therefore, there is no consensus on which ligament should be recreated first by the tenodesis procedure. At the same time, the injury mechanism and clinical manifestation of subtalar instability are similar to that of ankle instability, and therefore, its diagnosis is often misinterpreted. As a result, the publications of surgical reconstruction regarding subtalar instability are very limited. In addition, most of these studies are retrospective case series with limited numbers of patients [
9,
11‐
13,
29]. Thomas et al. [
39] has summarized the results of these reports. The results showed that their clinical results seem to be similar. On the other side, there is no comprehensive comparison targeting the correcting power of these tenodesis procedures so far. The results of our study showed that none of these tenodesis procedures could restore the biomechanical behaviors of the subtalar joint to normal. Fortunately, the stability of subtalar joint was supported by the bony geometry and ligaments jointly. In a cadaveric research on hindfoot stability, Cass et al. found that the subtalar joint appeared to be stable when the foot was loaded, and they believed that the stability of subtalar joint largely came from the bony constrains [
40]. In our simulation, the subtalar joint in rotational test was loaded without weight bearing. In this condition, the tested procedures still managed to restore the subtalar stability effectively. Even though the joint was subject to a certain degree of laxity with these procedures, with the inherent stability of the joint surface congruency, the subtalar instability would be further improved in weight bearing.
There are several limitations in the present study. First, some simplifications were introduced in our model. The material properties of the ligaments and grafts were determined in accordance with the literature rather than actual measurements. Different material models were used for soft tissues, including both linear hyperelastic and non-linear hyperelastic models, and the viscoelastic properties of the soft tissues were not considered. These data from published literatures introduced uncertainty to this study. And, some parts of the grafts that pass through the bone tunnels were also simplified. This may also affect the forces within the ligaments. Despite these simplifications, the validation test showed that our simulation results were very close to the experimental measurements of previous cadaveric studies. Second, our FE model was only simulated based on an young male volunteer. However, the validation comparator used were cadaver specimens from old male and female person. Therefore, when we use the conclusion of this article, was must take note of its applicability. Moreover, only one model was simulated in this study. However, the contact mechanics and contact patterns would vary from person to person. Therefore, subject-specific finite element models were needed to improve the accuracy of our analysis in future.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.