Introduction
Calcaneal fractures account for 2% of all body fractures and 60% of foot fractures, 75% of which are displaced intra-articular fractures [
1]. Calcaneal fractures are usually caused by an impact on the calcaneus during a fall from a height. High energy impact on the calcaneus causes collapse of the posterior talar articular surface [
2]. The calcaneal trabeculae are arranged in the direction of the tension and pressure to which they are subjected, including the tension trabeculae and the pressure trabeculae. The pressure trabeculae are dense below the posterior talar articular surface [
3]. Due to the specific anatomy of the calcaneus, the main weight-bearing area is the posterior talar articular surface, and this is where most calcaneal fractures occur [
4]. Calcaneal fractures are mainly manifested by heel tenderness and the inability to walk on the ground, which affects people’s health and life. And the best way to treat calcaneal fractures is still unclear [
2,
5].
Most clinical classifications of calcaneal fractures use the Sanders classification proposed by Sanders [
6] in 1990. This classification mainly reflects the degree of injury to the posterior talar articular surface of the calcaneus and has important implications for the clinical treatment and prognosis of calcaneal fractures. Depending on the degree of fracture and soft tissue damage, treatment options for calcaneal fractures include non-surgical and surgical treatments [
7,
8]. Non-surgical treatment, also known as conservative treatment, is indicated for patients with comminuted calcaneal fractures, poor soft tissue conditions or other serious diseases. The treatment does not reset the fracture sites but allows for immobilisation in a cast and early weight bearing. The pain and swelling of the patient’s foot are reduced by applying ice, simple activities and appropriate functional exercises. Surgical treatment options include open reduction and internal fixation (ORIF) and percutaneous minimally invasive fixation (PMIF).
ORIF is suitable for displaced calcaneal fractures, particularly intra-articular fractures of the calcaneus, and is effective in restoring the force line bearing a heavy burden on the hindfoot [
9]. This treatment achieves good results in terms of reliable fixation and accurate restoration through the use of implants such as plates and screws, with a cure rate of over 75% [
10]. Surgical treatment aims to restore the geometry of the calcaneus and the flatness of the articular surface. The fracture blocks can be repositioned to accurately restore the posterior talar articular surface and the anatomical relationship between the three subtalar articular surfaces, the degree of which is closely related to the clinical efficacy of the calcaneal fracture [
11]. ORIF requires extensive dissection of the soft tissue to fully expose the fracture, and the soft tissue incision margins may become infected or even necrotic after surgery, which is more damaging to the body [
12]. PMIF is indicated for mild calcaneal fractures with partial loss of calcaneal height and mild widening, which can be repositioned with the aid of steel nails and Kirschner wires. This method does not require extensive soft tissue dissection, avoids soft tissue necrosis, reduces intraoperative bleeding and reduces the risk of post-operative complications [
13]. The treatment of displaced intra-articular calcaneal fractures continues to be a challenge [
14‐
16]. Research showed that conventional ORIF is more effective but has a higher complication rate of 11.7% to 35%. In recent years, PMIF technology has significantly reduced the post-operative complication rate to 0–13% [
17].
In recent years, with the continuous development of computer technology, biomechanics, medicine, and physics, finite element simulation techniques have been increasingly used in areas such as fracture fixation. Hanbin Ouyang [
18] designed a calcaneal locking plate with good biomechanical properties by topology optimisation and compared it with a conventional locking plate using the finite element method. The new design was found to have a more reasonable stress distribution and demonstrated superiority in fatigue strength. Ming Ni [
19] used finite element software to simulate the biomechanical properties of calcaneal fractures fixed with absorbable screws and to calculate the stress and displacement distribution of the calcaneus and implants. The results showed that the stress and displacement in the calcaneus were within acceptable limits, which was similar to the results of Haowei Zhang [
20]. Ching-Hsuan Chen [
21] analysed the use of the whole locking, nonlocking, and hybrid screws to fix Sanders type II calcaneal fractures and found that hybrid screws had a biomechanical advantage over whole locking screws. Qing-Jiang Pang [
22] took a finite element approach to investigate the need for talus support screws in the fixation of calcaneal fractures and found that they were key to achieving biomechanical stability. Several clinical surgeries have demonstrated that minimally invasive screws can accurately restore the anatomical structure of the calcaneus and achieve a good prognosis [
23,
24]. In most of these, the traction of the ligaments, plantar fascia, and soft tissue on the bones was not taken into account. Ming Zhang and Duo Wai-Chi Wong [
25] developed a methodological quality assessment instrument for single-subject finite element analysis (MQSSFE) in computational orthopaedics. The instrument contained 37 items with 6 domains such as model reconstruction and configuration, model verification, and model assumption and validity. The MQSSFE was compared with the modified Down and Black instrument [
26] to demonstrate that the MQSSFE was valid and reliable.
The calcaneus and the implants have complex mechanical behaviour, both in the early stages of surgery and in the post-operative healing period. This is due to a combination of the geometry of the bones, the fracture site, and the configuration of the implants. Traditional open reduction surgery is more effective but has a higher complication rate. Because of the development of minimally invasive techniques, calcaneal fracture complications can be effectively reduced. However, due to the complexity of calcaneal fracture, the surgeons need to choose the appropriate treatment plan according to the characteristics of patients and specific conditions, so as to achieve better post-operative efficacy [
27].
In summary, this work compared the biomechanical properties of the ORIF and PMIF surgical approaches to fixing calcaneal fractures using finite element analysis (FEA). The aim was to investigate the differences in biomechanical properties. This study: (1) the disadvantage of the irreversibility of clinical surgery was overcome, experimental consumables were reduced and costs were saved to a certain extent, the simulation parameters can be optimised several times through clinical feedback, and the results obtained are useful for exploring the pathogenesis of human orthopaedic diseases and for refining models; (2) research on the mechanics of the osteomuscular system, the development of implants and other orthopaedic rehabilitation devices and surgical optimisation are assisted; and (3) through parametric mechanical simulations, clinicians are provided with references to design and refine treatment plans and rehabilitation programmes.
Discussion
Minimally invasive surgery is considered to be the new trend in the treatment of calcaneal fractures. Calcaneal fractures have a high incidence and are difficult to treat clinically. This work investigated the differences in biomechanical properties between ORIF and PMIF for the treatment of calcaneal fractures (Sanders type II and type III).
Imaging data of the lower limb were collected, and a full-foot model was established. In most of the previous finite element studies of calcaneal fractures, a single calcaneus model was used without taking into account the traction of the ligaments, plantar fascia, and soft tissue on the bones [
18,
33,
34]. A three-dimensional solid model of the foot–ankle complex including bones, soft tissue, cartilages, ligaments, and plantar fascia was established on this basis in this paper. Biological tissues generally exhibit nonlinear properties, but in biomechanical simulations of the bone and muscle system, the stress and deformation on the tissue are in a linear relationship when the loading is small, and the stress and strain can show a good linear elastic relationship. Therefore, the nonlinear properties of tissues can be somewhat simplified and linearly elastic material parameters can be used, which can improve the computational efficiency without affecting the computational accuracy [
22,
35]. Professor Guangrong Yu [
36], Department of Orthopaedic Surgery, Tongji Hospital, Tongji University School of Medicine, proposed the principle of three-point fixation of calcaneal fractures; namely, the sustentaculum tali, the anterior process, and the calcaneal tuberosity were fixed by screws. Among them, the sustentaculum tali was the most reliable fulcrum for fixing the posterior talar articular surface. It can meet the clinical need for calcaneal fracture treatment, which is of great significance in the fixation of calcaneal fractures.
The finite element results indicated that there were significant differences in the stress distribution between the two surgical approaches for the treatment of calcaneal fractures. In both the Sanders type II and III calcaneal fracture models, the calcaneal stress was distributed in the medial cortex and the posterior talar articular surface, with no significant stress concentration. This was consistent with the conclusion that the posterior talar articular surface was the main weight-bearing site. The maximum stress on the calcaneus was higher in Sanders type III than in Sanders type II for the same surgical approach. The main reason for this difference was that Sanders type III was a three-part fracture of the posterior talar articular surface, whereas Sanders type II was a two-part fracture of the posterior talar articular surface. In type III fractures, the calcaneus was more comminuted, so the calcaneus would be more stressed by the pulling forces of the plantar fascia and Achilles tendon. The maximum stress on the screws in type III fractures was higher than in type II fractures in both the ORIF and PMIF groups for the same surgical approach. In the ORIF group, the screws directly below the posterior talar articular surface were subjected to greater stress in the type II model and the sustentaculum tali screws in the type III model. The maximum stress in the PMIF group was concentrated in the sustentaculum tali screws, which was caused by the concentration of the fracture lines in the type III model from the middle of the posterior talar articular surface to the sustentaculum tali. The stress distribution of the plates was similar in type II and III fracture models in the ORIF group, suggesting that the different fracture patterns had less influence on the stress distribution of the plates. And there was no obvious stress concentration to avoid breakage of the plates due to stress fatigue.
In terms of stability of the fixation system, the PMIF achieved better stability after calcaneal fracture surgery. The maximum displacement of the calcaneus in both type II and III fracture models in the ORIF group was concentrated in the anterior process, whereas the maximum displacement in the PMIF group was concentrated in the posterior talar articular surface. This difference was related to the forces and fixation on the calcaneus in the foot–ankle composite model. The ORIF group relied on three screws for fixation of the posterior talar articular surface, with two screws driven from the outside into the inside of the anterior process. In the PMIF group, two screws were used to fix the posterior talar articular surface and the sustentaculum tali, and two screws were inserted from the end of the calcaneal tuberosity to fix the tuberosity and the anterior process. The calcaneal tuberosity screws, sustentaculum tali screws, and anterior process screws maintained the length, width and height of the calcaneus, respectively. The maximum displacement on the screws in both models was concentrated in the sustentaculum tali screws and the anterior process screws, with the maximum displacement on the screws in the ORIF group being less than 0.22 mm and in the PMIF group being less than 0.15 mm. The maximum displacement of the plates in the ORIF group appeared in the middle of the plates, which was lower than 0.13 mm, providing sufficient stability to the fixation system. Compared to ORIF, PMIF offered better clinical results for the safety and stability of the fixation system and reduced post-operative complications.
Because the stiffness of the implants (plates and screws) was much greater than the stiffness of the bone, the implants were subjected to more loading and the bone to less, thus predisposing the patient to stress shielding effect after surgery. In clinical practice, it was found that during the rehabilitation phase after internal fixation, the bone tissue was not mechanically stimulated enough due to a prolonged low-stress environment, which could easily lead to osteoporosis and secondary fractures. Resorbable implants provided adequate fixation in the initial post-operative period and gradually degraded as the bone tissue continued to heal, eliminating the need for re-surgical removal. At this stage, the challenge of reconciling the rate of implant degradation with the rate of bone healing is an urgent one.
In summary, the stress and displacement of the implants were within the permissible stress and elastic deformation. The maximum stress and displacement of the calcaneus and the implants were slightly lower in the PMIF group than in the ORIF group when fixing Sanders type II and III calcaneal fractures. Fatigue injuries are one of the main causes of fractures and implant breakage. When fatigue loading occurs, localised structural damage to the calcaneus and implants can occur. When local stresses exceed the threshold, cracks will develop at the stress concentrations, resulting in stress fractures of the calcaneus and breakage of the implants. Early post-operative weight bearing should be avoided as this may lead to loosening and breakage of the implants and displacement of the fracture fragments. Appropriate functional exercises can be performed under medical supervision. The present study still has some limitations.
(1)
Model: the model still needs further refinement, the cortical and cancellous bones were not segmented in this study’s model, the mechanical properties of tissues such as nerves and blood vessels were not considered, and for the time being, isotropic linear elastic material was used to simulate soft tissue and the mechanical properties of muscles were simplified;
(2)
Boundary conditions: static studies of balanced standing have been carried out, and complex dynamic properties (e.g. normal walking, running, impact, loading) have not been studied. The complete gait cycle consists of a stance phase and a swing phase, with the first contact of the heel with the ground being the start of the stance phase. During the stance phase, the foot changes from plantar flexion to dorsiflexion and the weight-loading area shifts from the heel to the whole foot. During heel release, the loading is distributed to the metatarsals and transferred to the phalanxes. Ankle reaction force varies with the cadence. It can reach a peak of 3–5 times body weight when walking and up to 13 times body weight when running.
(3)
Model validation: the measurement and FEA results have some inaccuracies due to experimental conditions and discrepancies in the model building;
(4)
Simulation: the contradiction between computational accuracy and efficiency has always existed. This study took into account accuracy and efficiency to the maximum extent possible under the available computing conditions. The selection of simulation parameters relied on previous studies and the actual situation of the model to obtain, and there was still some shortage in parameter optimisation. However, according to the clinical analysis, the simulation mainly plays the role of pre-assistance and subsequent validation and cannot replace clinical practice.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.