Optimal biomechanical choice of implant placement in various pilon fracture types: a finite element study (2024)

Background

The research on the biomechanical characteristics of individual implant placement for pilon fractures based on the different initial direction of fracture displacement is still insufficient. This study’s aim is to compare the stress distribution in bones and implants with various pilon fracture types.

Methods

Varus, valgus, dorsiflexion, and plantarflexion type fractures were categorized as type I, II, III, and IV, respectively. The buttress plate was placed medially in subtypes I_A and II_B, whereas it was placed anterolaterally in subtypes I_B and II_A; The anterior or posterior buttress plate was utilized in subtypes III_A and IV_A, the lag screws were applied in subtypes III_B and IV_B. The maximum equivalent stress of tibia (TI-S_max) and implants (IF-S_max), stress of fracture fragments (S_fe), and axial displacement values of the fracture fragments (AD_fe) in each subtype were analyzed when the ankle was in a neutral position, 15° of varus and valgus in types I and II, 15° of dorsiflexion and plantarflexion in types III and IV.

Results

Under the same axial stress loading conditions, TI-S_max, S_fe, AD_fe of subtypes I_A and II_A were significantly lower than subtypes I_B and II_B, while IF-S_max of subtypes I_A and II_A were obviously larger than subtypes I_B and II_B. Additionally, TI-S_max, S_fe, AD_fe of subtypes III_A and IV_A were considerably lower as IF-S_max met expectations compared to subtypes III_B and IV_B.

Conclusion

Based on these results, when making decisions for open reduction and internal fixation in various pilon fractures, the choice and placement of the implant can be recommended as follows: the medial buttress plate for varus types; the anterolateral plate for valgus types; the anterior plate for dorsiflexion types; the posterior plate for plantarflexion types.

Optimal choices and placements of implants in individual pilon fracture types still pose a challenge for orthopedic surgeons, particularly with complex types [1, 2]. In Rüedi and Allgower’ s fundamental surgical principles, the placement for a buttress plate is recommended on the medial side of the distal tibia [3]. Sirkin, however, noted that a medial buttress plate may inadequately support the displaced fragments in the valgus force injury type fractures [4]. Topliss and other experts are also concerned that the position of the foot during injuries might be a decisive factor of fracture patterns [5, 6]. The placement of the implant should be based on the orientation of the major fracture lines [6]. According to the initial fracture displacement at the time of injury, five types of pilon fractures were originally proposed, including: varus, valgus, dorsiflexion, plantarflexion and neutral [7]. The modified plate placement principles were also recommended as follows: the medial plate was chosen for varus types; the anterolateral plate was used for valgus types; the anterior plate was utilized for dorsiflexion types; the posterior plate was applied for plantarflexion types. Previous clinical data also showed promising outcomes when applying these principles [7, 8]. However, the research on biomechanical characteristics of optimal choices of implant placement for various pilon fractures based on the different initial direction of fracture displacement is still insufficient and relevant biomechanical studies are necessary.

Therefore, a finite element analysis of different choices of implant placements in the varus, valgus, dorsiflexion, and plantarflexion pilon fracture types was conducted. The objectives were to compare the stress distribution at the bone and individual implants, and to investigate the displacement values of the fragments of different fracture types during mechanical loading.

This study was approved by our institutional review board (No. [2022]027–01) and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained.

Finite element modeling

Pre-disposal

Computed tomography (CT) images of the right lower extremity from a healthy male volunteer (age: 27Y, height: 175cm, weight: 65kg) without tumors, fractures or deformities were utilized. Scanning parameters included 120kV, 125 mA, slice thickness of 0.6 mm, and spacing of 1.0mm, with each layer scanned for 200ms. The digital imaging and communication of medicine (DICOM) data sets were imported into Mimics 14.11 software (Materialise, Leuven, Belgium). The tibia, fibula, and talus were distinguished from other tissues using threshold segmentation, and their 3D geometries were reconstructed and saved as.STL files after smoothing.

Bone modeling

The.STL files were processed in Geomagic Studio12.0 for image optimization and surface fitting, and then converted into initial graphics exchange specification (IGES) format. These were imported into UG NX10.0 software (Siemens Corp., Germany) to assemble the detailed bone model, using a Hounsfield unit threshold to differentiate cortical (HU > 700) and cancellous bone (HU ≤ 700) [9].

Modeling of plates and screws

The implants (Smith & Nephew Inc, USA) included 4.5mm anatomically contoured locking plates, 3.5mm reconstructed locking plates, 3.5mm and 4.5mm locking screws, and 4.5mm lag screws. Designs were created using Solidworks 2013 (Dassault Systemes Simulia Corp., Providence, RI, USA), which accurately represented the plates and screws based on physical models, simplifying the screws into cylindrical shapes.

Pilon fractures modeling

The varus and valgus types were modeled together, feature osteotomy lines from superior-medial to inferior-lateral or vice versa, creating similarly sized medial or anterolateral articular surface fragments across the middle line of the articular surface (Fig.1A,B). Dorsiflexion and plantarflexion types were also modeled together, each osteotomized to generate equally sized anterior or posterior fracture fragments (Fig.1C,D).

The UG NX10.0 software simulated fracture reduction and assembled the various pilon fracture types with their respective implants. The fixation models for the tibia plate included three screws on each side of the fracture line. Similarly, three lag screws, positioned in a triangle shape, from anterior to posterior in the dorsiflexion models or from posterior to anterior in the plantarflexion models. Additionally, all fibular plate models utilized three screws each, located 5 cm proximal to the fibula tip.

Experimental categorizing

Pilon fractures were categorized into four types (Fig.2A) based on initial fracture displacements: varus, valgus, dorsiflexion, and plantarflexion. Due to the complex injury mechanisms and diverse fracture patterns, the neutral types were not included. The detailed experimental categorizing is shown in Fig.2B. Type I was divided into subtypes I_A and I_B, with the main buttress plate placed medially or anterolaterally, based on the fragment location. Similarly, type II was divided into subtypes II_A and II_B. Additional subtypes I_A1, I_B1, II_A1, and II_B1 were identified under 15°varus stress loading; subtypes I_A2, I_B2, II_A2, and II_B2 under 15°valgus stress loading. Type III included subtype III_A with anterior buttress plates and III_B with only lag screws. Subtypes IV_A and IV_B used posterior buttress plates and lag screws, respectively. Subtypes III_A1, III_B1, IV_A1, and IV_B1 were classified under 15°dorsiflexion stress loading; Subtypes III_A2, III_B2, IV_A2, and IV_B2 under 15°plantarflexion stress loading.

Boundary conditions

An initial mesh size of 1.2mm was chosen, and after several rounds of reduction by 0.15mm, a uniform mesh size of 0.6mm was considered converged due to increased stress concentration near implants without significant deformation changes [10]. Materials for implants and bones were modeled as elastic, linear, and isotropic, ignoring soft tissues like ligaments and cartilages to focus on axial loading, the primary load type in weight-bearing bones [11]. The models used second-order hexahedral units [12], settling on a mesh size of 0.6mm, resulting in 36,485 nodes and 21,265 elements. Mechanical properties were assigned as follows: cortical bone, cancellous bone, and implants (Ti6Al4V) had elastic moduli of 17GPa, 700 MPa, and 112GPa, respectively, with Poisson's ratios of 0.3, 0.2, and 0.3 [13]. Fracture ends were modeled as complete but unconnected fractures with a friction coefficient of 0.4 [14]. Simplified screws were treated as locking plate-screws to simulate the locking mechanism effectively.

Loading and stress analysis

To simulate human walking, two phases were implemented in the loading tests: (1) Loading was applied to the model fixed in a neutral position, simulating the normal physiological condition; (2) Individual axial loading tests were conducted with the ankle joint at 15°of varus or valgus for the corresponding varus and valgus fracture models, and at 15° of dorsiflexion or plantarflexion for the respective dorsiflexion and plantarflexion models. Stress distribution and displacement values for medial, anterolateral, anterior, and posterior fragments were measured and compared at the same points under each loading condition. The loading simulation model, as shown in Fig.3, was secured at the tibial plateau, constraining all six degrees of freedom to prevent any movement in the X, Y, and Z axes at the proximal ends of the tibia and fibula. A rigid surface was placed at the bottom of the talus, where a distributed load of 600N (equivalent to approximately 1 × body weight) was applied vertically. This setup ensured the stress loading surface was nearly parallel to the fixed surface, eliminating torsional loads from the test setup. The simulation did not account for forces exerted by muscles and ligaments around the ankle or the frictional forces from joint cartilage and capsule, focusing solely on bone and implant responses under specified loads [15].

Evaluation metrics

During various mechanical loading tests, we recorded the maximum equivalent stress of the tibia (TI-S_max: MPa, accurate to 0.01), the internal fixation or implants (IF-S_max: MPa, accurate to 0.01), and the axial displacement of the fracture ends or fragments (AD_fe: mm, accurate to 0.01). Stress in the fracture fragments (S_fe: MPa, accurate to 0.01) was also measured for each type.

Data analysis

We compiled and analyzed the Von Mises stress and displacement values for the fracture fragments and implants in each subtype across different loading conditions. The inter-rating data was recorded and compared for each fracture pattern to preliminarily assess mechanical stability and implant efficacy. Due to the limitations of finite element analysis (FEA), very little differences were documented in the values of repeated measurements under one loading condition in the same sample. Therefore, we only recorded single data samples in each test.

Stress diagrams of individual implant placement for each type of pilon fracture models are shown in Fig.4. Under the same axial stress loading conditions in the neutral position, TI-S_max, AD_fe, and S_fe of subtypes I_A, II_A, III_A, and IV_A were considerably lower than subtypes I_B, II_B, III_B, and IV_B. Conversely, IF-S_max of subtypes I_A, II_A, III_A, and IV_A were obviously greater than subtypes I_B, II_B, III_B, and IV_B(Fig.5 and Table1).

In the 15° of varus position, TI-S_max, AD_fe, and S_fe of subtypes I_A1 and II_A1 were significantly lower than subtypes I_B1 and II_B1. However, IF-S_max of subtypes I_A1 and II_A1 was conversely larger than subtypes I_B1 and II_B1. Similarly, in the 15° of valgus position, TI-S_max, AD_fe, and S_fe of subtypes I_A2 and II_A2 were significantly lower than subtypes I_B2 and II_B2. On the contrary, IF-S_max of subtypes I_A2 and II_A2 was higher than subtypes I_B2 and II_B2(Fig.6 and Table2).

In the 15° of dorsiflexion position, TI-S_max, AD_fe, and S_fe of subtypes III_A1 and IV_A1 were moderately lower than subtypes III_B1 and IV_B1. However, IF-S_max of subtypes III_A1 and IV_A1 was larger than subtypes III_B1 and IV_B1. Likewise, in the 15° of plantarflexion position, TI-S_max, AD_fe, and S_fe of subtypes III_A2 and IV_A2 were marginally lower than subtypes III_B2 and IV_B2. On the other hand, IF-S_max of subtypes III_A2 and IV_A2 was higher than subtypes III_B2 and IV_B2(Fig.7 and Table3).

The main finding of the present study is that there is an obvious correlation between the stress distribution of the bone and the implants as well as the displacement values of the fracture fragments in individual pilon fracture types. The optimal goals of pilon fracture treatment are the anatomical reconstruction, stable internal fixation, and early targeted exercise [16, 17]. A deeper understanding of the injury mechanism of pilon fractures are effective for selecting the most reliable implants. FEA is still a reliable method for internal fixation biomechanics analysis of different fractures [15, 18]. Recently, the subject-specific prediction of soft tissue models emerged to improve the accuracy of the analysis [19]. When plates are placed on fractures, better biomechanical stability can be achieved when the implants are subjected to greater stress and fracture fragments are less displaced [18]. In the present study, FEA results showed that the main buttress plate placed medially obtained lower TI-S_max, AD_fe and S_fe as well as a larger IF-S_max than placed anterolaterally in the varus type. Conversely in the valgus type, the main buttress plate placed medially showed a lower IF-S_max as well as lager TI-S_max, AD_fe, and S_fe than placed anterolaterally. Moreover, in the dorsiflexion and plantarflexion type, the plate fixation gained a larger IF-S_max as well as lower TI-S_max, AD_fe, and S_fe than the lag screws. These results support that the main buttress plate should be placed according to the initial displacement direction of pilon fractures. There is no consensus between AO/ASIF which proposed to use a medial plate, and other scholars who recommended to position the main plate based on the orientation of the major fracture lines [3, 4, 6]. In the present study, when the ankle was in a neutral position, 15° of varus, or valgus position, the medial buttress plate in the varus type can obtain more biomechanical stability, compared to the anterolateral buttress plate. Similar results showed that in the valgus type, compared with the medial buttress plate, the anterolateral plate had a more reliable fixation.

Our results also showed that the biomechanical properties of the buttress plates were superior to the lag screws whether in the dorsiflexion or plantarflexion pilon fractures. Currently, buttress plates or lag screws are two standard internal fixation methods for these fractures [20, 21]. Erdem et al. [22] reported that there was little difference between buttress plates and lag screws for posterior ankle fractures. According to Hoekstra’s [23] study, the fragments of posterior pilon fractures could be easily fixed with a buttress plate through the posterior medial approach. In the present study, when the ankle was in a neutral, 15°of dorsiflexion or plantarflexion position, TI-S_max, S_fe and AD_fe with a buttress plate in each model were smaller than the lag screw fixation, whereas IF-S_max with buttress plates was higher whether in the dorsiflexion or plantarflexion fracture type.

Based on these results, when making decisions for open reduction and internal fixation in various pilon fractures, the choice and placement of the implant can be recommended as follows: the medial buttress plate for varus types; the anterolateral plate for valgus types; the anterior plate for dorsiflexion types; the posterior plate for plantarflexion types.

There are some other limitations in this study: First, the designed fracture lines and fragments were relatively simplified while the muscles and ligaments were not considered. Second, the length of the buttress plates and the angle of screws, were also not included. Finally, due to the limitations of FEA and experimental conditions, the results of various subcategorizations in different loading conditions were solely obtained from one sample, corresponding statistical analysis was difficult to finish. Consequently, the mechanical experiments on frozen fresh cadaveric specimens in vitro are necessary in the future to understand this topic deeply.

CT:

Computed tomography

DICOM:

Digital imaging and communication of medicine

IGES:

Initial graphics exchange specification

TI-S_max :

The maximum equivalent force of the tibia

IF-S_max :

The maximum equivalent force of the implants

Ad_fe :

The axial displacement value of the fracture fragments

S_fe :

The stress of the fracture fragments

FEA:

Finite element analysis

AO/ASIF:

Association for the Study of Internal Fixation

Author notes

1. Peizhao Liu, Xianzhong Mei and Zhixiang Wang these authors contributed equally to this work and should be considered co-first authors.

Authors and Affiliations

1. Department of Orthopedics, General Hospital of Central Theater Command of PLA, (Wuhan General Hospital of Guangzhou Command, Previously), Hubei Province, NO. 627, Wuluo Road, Wuhan, 430030, P.R. China

   Peizhao Liu,Feng Xu&Shijun Wei

2. Department of Orthopedics, Shenzhen Pingle Orthopedic Hospital, Shenzhen, Guangdong, P.R. China

   Xianzhong Mei

3. Department of Orthopedics, Wuhan No.1 Hospital, Wuhan, Hubei, P.R. China

   Zhixiang Wang

4. Department of Orthopedics, South China Hospital, Medical School, Shenzhen University, Shenzhen, Guangdong Province, 518116, P.R. China

   Xianhua Cai

5. Department of Orthopedics, The First College of Clinical Medical School, China Three Gorges University and Yichang Central People’s Hospital, Yichang, Hubei, P.R. China

   Kangquan Shou

6. The First Clinical Medical School of Southern Medical University, Guangzhou, Guangdong, P.R. China

   Shijun Wei

Contributions

Liu Peizhao. Mei Xianzhong and Wei Shijun. Cai Xianhua wrote the main manuscript text and Wang Zhixiang. Shou Kangquan prepared figures. Xu Feng reviewed the manuscript.

Corresponding authors

Correspondence to Xianhua Cai or Shijun Wei.

Ethics approval and consent to participate

Ethics approval was obtained from the institutional ethical committee of General Hospital of Central Theater Command (No. [2022]027–01). The study was in accordance with relevant guidelines and regulations. The participant gave written informed consent to participate in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions