WO2023065049A1 - Talus implant comprising a polycarbonate-urethane coating - Google Patents

Abstract

Customized talus implants are expensive and time-consuming to produce; to overcome some limitations of customized talus implants, universal talus implants are used. However, direct contact between the biological cartilage and the metal implant surface universal talus implants may be problematic. In order to improve the universal talus implant contact characteristics, a polymeric coating is added onto said implants; the coating may also be used on customized talus implants. The present disclosure relates generally to artificial implants comprising a coating, wherein said coating is a polycarbonate-urethane (PCU), a high density polyethylene poly(aryl-ether-etherketone) (PEEK) or a poly(aryl-ether-ketone-ether-ketoneketone) (PEKEKK); said artificial implant is preferably a talus implant.

Description

TALUS IMPLANT COMPRISING A POLYCARBONATE-URETHANE COATING

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to United States Patent Provisional Application US 63/270,898, filed October 22, 2021 , and United States Patent Provisional Application US 63/417,765, filed October 20, 2022, the entire contents both of which is hereby incorporated by reference.

FIELD

[0002] The present disclosure relates generally to Polycarbonate-Urethane Coating can Significantly Improve Talus Implant Contact Characteristics

BACKGROUND

[0003] Hemiarthroplasty is a common surgical procedure with specific indications, used to treat conditions of the joint (e.g. avascular necrosis, fractures). In ankle hemiarthroplasty, customized talus implants have widely reported clinical successes, and are regarded as a promising treatment method for patients suffering from avascular necrosis (Ando et al., 2016; Bowes et al., 2019; Harnroongroj and Harnroongroj, 2014; Katsui et al., 2019; Tanaka et al., 2003; Taniguchi and Tanaka, 2019). However, the individualized nature of fabrication of customized implants lead them to be expensive and time-consuming. To overcome the limitations of customized talus implants, the concept of universal talus implants has been proposed, with initial support from numerical and clinical feasibility studies (Bowes et al., 2019; Liu et al., 2020; Trovato et al., 2017). However, all proposed talus implants are made of metal or stiff material such as ceramic or cobalt-chrome.

SUMMARY

[0004] In one aspect, the present disclosure relates generally to polycarbonateurethane coating that can significantly improve talus implant contact characteristics.

[0005] In one aspect there is provided a talus implant comprising:

[0006] a body section;

[0007] a neck section; and

[0008] a head, [0009] the talus implant having an outer surface comprising a coating comprising a polycarbonate-urethane (PCU), a high density polyethylene and poly(aryl-ether-ether- ketone) (PEEK), poly(aryl-ether-ketone-ether-ketoneketone (PEKEKK).

[0010] In one example, at least a portion of the outer surface comprises a lattice.

[0011] In one example, the coating has a thickness of between about 0.5 to about

20 mm.

[0012] In one example, the coating has a thickness of between about 0.5 mm to about 6 mm.

[0013] In one example, the coating has a thickness of between about 1 mm to about 3 mm.

[0014] In one example, the coating has a thickness of between about 1 .5 mm to about 2 mm.

[0015] In one example, the talus implant comprises titanium, ceramic, or CoCr.

[0016] In one example, having a shape which is customized.

[0017] In one example, having a shape which is universal.

[0018] In one example, the shape is selected using a cartilage-exclusive sizing guide.

[0019] In one example, having a shape which is a statistical shape model (SSM).

[0020] In one example, the shape is selected using a cartilage-exclusive sizing guide.

[0021] In one example, the talus implant comprises titanium, ceramic, or cobaltchrome (CoCr).

[0022] In one example, having average peak contact pressures that are similar to that of a biological implant.

[0023] In one example, having similar contact areas that are similar to that of a biological implant.

[0024] In one aspect there is provided a method of reconstructing a subjects ankle comprising: implanting a talus implant as described herein.

[0025] In one example, the patient is a human.

[0026] In one aspect, there is provided an artificial implant comprising:

[0027] a body, at least a portion said body having an outer surface comprising a coating comprising a polycarbonate-urethane (PCU), a high density polyethylene and poly(aryl-ether-ether-ketone) (PEEK), poly(aryl-ether-ketone-ether-ketoneketone (PEKEKK). [0028] In one example, the artificial implant is a humeral head implant, a radial head implant, a distal humerus implant, a-scaphoid implant, a lunate implant, femoral head implant, a patella implant, a talus implant, a metatarsal head implant, a phalanx implant, or a plug implant.

[0029] In one example, the plug implant comprises metal, titanium, ceramic, or CoCr.

[0030] In one aspect there is provided a method of implanting an artificial implant in a subject, comprising: providing an artificial implant as described herein, and implanting the artificial implant in the subject

[0031] In one example, the subject is a human.

BRIEF DESCRIPTION OF THE FIGURES

[0032] Embodiments of the present disclosure will now be described, by way of example only, with reference to the attached Figures.

[0033] Fig. 1 . Overview of finite element models with different types of talus implants for a subject in neutral standing.

[0034] Fig. 2. Boundary conditions of FE models (as displayed on the biological model).

[0035] Fig. 3A-D. Contact stress distribution on the cartilage for four subjects in neutral standing posture (the color scale was normalized to that of the biological model for each cartilage surface).

[0036] Fig. 4. A graphical abstract.

[0037] Fig. 5 Shape of the biological talus bone and three artificial implants (customized, universal, and SSM) for subject 1 (in order to more clearly illustrate the differences between scale and offset, models with deviation contours are in the bottom two rows taking original shape as the reference.

[0038] Fig. 6 Finite model of the talus joint.

[0039] Fig. 7 Boundary conditions of the finite element model (displayed on the biological model).

[0040] Fig. 8 Overview of FE models of the talus joint (DF: dorsiflexion; NS: neutral standing; PF: plantar Flexion).

[0041] Fig. 9 contact stress on the tibiofibular cartilage for the biological talus of four subjects in neutral standing posture (contact pressure in the medial malleoli region of the tibial cartilage is highlighted). [0042] Fig. 10A-F. Scaled contact stress distribution on the cartilage for subject 1 in DF(a); NS(c) and PF (e); and the contact area percentage of the adjacent cartilage with different ranges of contact pressure in DF (b), NS (d) and PF (f).

[0043] Fig.11 comparison of total contact area for all implant models of subject 1.

[0044] Fig. 12 Number of models having more than 5% of cartilage contact area with a contact pressure greater than 17.5 MPa using different talus implants for all subjects (e.g., the first bar shows that there are 2 customized + implants that have more than 5% contact area with a contact pressure greater than 17 Mpa.

[0045] Fig. 13 total contact area of the talus bone in four subjects.

[0046] Fig. 14A-F. Scaled contact stress distribution on the cartilage for subject 2 in DF (a); NS(c); and PF (e); the contact area of the adjacent cartilage surface and the area percentage increasing in 0.25 MPa intervals in DF(b), NS(d) and PF(f).

[0047] Fig. 15A-F Scaled contact stress distribution on the cartilage for subject 3 in DF(a); NS (c); and PF(e); the coated area of the adjacent cartilage surface and area percentage increasing in 0.25 MPa intervals in DF(b), NS(d) and PF(f).

[0048] Fig. 16A-F. Scaled contact stress distribution on the cartilage for subject 4 in DF(a); NS(c); and PF(e); the contact area of the adjacent cartilage surface and area percentage increasing in 0.25 MPa in DF(b), NS(d) and PF(f).

[0049] Fig. 17 Diamond shaped lattice disk before and after coating.

DETAILED DESCRIPTION

[0050] In one aspect, the present disclosure relates generally to polycarbonateurethane coating that can significantly improve talus implant contact characteristics.

[0051] In one aspect there is provided a talus implant comprising:

[0052] a body section;

[0053] a neck section; and

[0054] a head,

[0055] the talus implant having an outer surface comprising a coating comprising a polycarbonate-urethane (PCU), a high density polyethylene and poly(aryl-ether-ether- ketone) (PEEK), poly(aryl-ether-ketone-ether-ketoneketone (PEKEKK).

[0056] It will be appreciated that is some examples, it may be possible to construct the talus implant entirely from a polycarbonate-urethane (PCU), a high density polyethylene and poly(aryl-ether-ether-ketone) (PEEK), poly(aryl-ether-ketone-ether- ketoneketone (PEKEKK). In other examples, it may be possible to have a smaller core of metal/ceramic and larger amount of a polycarbonate-urethane (PCU), a high density polyethylene and poly(aryl-ether-ether-ketone) (PEEK), poly(aryl-ether-ketone-ether- ketoneketone (PEKEKK).

[0057] In one example, at least a portion of the outer surface comprises a lattice.

[0058] In one example, the coating has a thickness of between about 0.5 to about

20 mm.

[0059] In one example, the coating has a thickness of between about 0.5 mm to about 6 mm.

[0060] In one example, the coating has a thickness of between about 1 mm to about 3 mm.

[0061] In one example, the coating has a thickness of between about 1 .5 mm to about 2 mm.

[0062] In one example, the talus implant comprises titanium, ceramic, or CoCr.

[0063] In one example, having a shape which is customized.

[0064] In one example, having a shape which is universal.

[0065] In one example, the shape is selected using a cartilage-exclusive sizing guide.

[0066] In one example, having a shape which is a statistical shape model (SSM).

[0067] In one example, the shape is selected using a cartilage-exclusive sizing guide.

[0068] In one example, the talus implant comprises titanium, ceramic, or cobaltchrome (CoCr).

[0069] In one example, having average peak contact pressures that are similar to that of a biological implant.

[0070] In one example, having similar contact areas that are similar to that of a biological implant.

[0071] In one aspect there is provided a method of reconstructing a subjects ankle comprising: implanting a talus implant as described herein.

[0072] The term “subject”, as used herein, refers to an animal, and can include, for example, domesticated animals, such as cats, dogs, etc., livestock (e.g., cattle, horses, pigs, sheep, goats, etc.), laboratory animals (e.g., mouse, rabbit, rat, guinea pig, etc.), mammals, non-human mammals, primates, non-human primates, rodents, birds, reptiles, amphibians, fish, and any other animal. The subject may be an infant, a child, an adult, or elderly. In a specific example, the subject is a human. [0073] In one example, the patient is a human.

[0074] In one aspect, there is provided an artificial implant comprising:

[0075] a body, at least a portion said body having an outer surface comprising a coating comprising a polycarbonate-urethane (PCU), a high density polyethylene and poly(aryl-ether-ether-ketone) (PEEK), poly(aryl-ether-ketone-ether-ketoneketone (PEKEKK).

[0076] In one example, the artificial implant is a humeral head implant, a radial head implant, a distal humerus implant, a-scaphoid implant, a lunate implant, femoral head implant, a patella implant, a talus implant, a metatarsal head implant, a phalanx implant, or a plug implant.

[0077] In some examples, a plug implant is around and resurface a small portion of the joint.

[0078] In one example, the plug implant comprises metal, titanium, ceramic, or CoCr.

[0079] In one aspect there is provided a method of implanting an artificial implant in a subject, comprising: providing an artificial implant as described herein, and implanting the artificial implant in the subject.

[0080] In one example, the subject is a human.

[0081] Methods of the invention are conveniently practiced by providing the compounds and/or compositions used in such method in the form of a kit. Such kit preferably contains the composition. Such a kit preferably contains instructions for the use thereof.

[0082] To gain a better understanding of the invention described herein, the following examples are set forth. It should be understood that these examples are for illustrative purposes only. Therefore, they should not limit the scope of this invention in anyway.

[0083] EXAMPLES

[0084] EXAMPLE 1

[0085] Introduction

[0086] Hemiarthroplasty is a common surgical procedure with specific indications, used to treat conditions of the joint (e.g. avascular necrosis, fractures). In ankle hemiarthroplasty, customized talus implants have widely reported clinical successes, and are regarded as a promising treatment method for patients suffering from avascular necrosis (Ando et al., 2016; Bowes et al., 2019; Harnroongroj and Harnroongroj, 2014; Katsui et al., 2019; Tanaka et al., 2003; Taniguchi and Tanaka, 2019). However, the individualized nature of fabrication of customized implants lead them to be expensive and time-consuming. To overcome the limitations of customized talus implants, the concept of universal talus implants has been proposed, with initial support from numerical and clinical feasibility studies (Bowes et al., 2019; Liu et al., 2020; Trovato et al., 2017). However, all proposed talus implants are made of metal or stiff material such as ceramic or cobalt-chrome.

[0087] A documented concern with hemiarthroplasty procedures using metal implants is the potential for cartilage wear and bone fractures. For instance, acetabular wear after hip hemiarthroplasty involving metal implants has been widely reported in the literature (Guyen, 2019; Hedbeck et al., 2011 ; Jia et al., 2015). Both experimental and numerical studies have shown that mechanical loading changes caused by stiff implant surface material will have an impact on the surrounding bones, where an increased load can affect bone remodeling and bone density changes (van Lenthe et al., 2002, 1997), and a decreased load can cause resorption and osteopenia, thereby leading to an increased risk of bone fractures (Jarvenpaa et al., 2014; Lavernia et al., 2014). These existing issues in hemiarthroplasty stem from direct contact between the biological cartilage and the metal implant surface (e.g. cobalt-chrome) (Gadkari et al., 2013; Ruatti et al., 2017; Tracey et al., 2019). Metal-based talus implants can cause similar issues (Ruatti et al., 2017). Attempts to improve the metal-biological interactions in hemiarthroplasty have shown compliant materials to be more adept at transmitting forces more physiologically than stiff materials (de Ruiter et al., 2017; Rankin et al., 2016). In particular, polycarbonate-urethane (PCU) has been noted to be a commonly used compliant material, for instance in Lazic et al. (2019) and Alley et al., (2020), where free- floating PCU pieces were surgically inserted in patients’ knee and hip joints, respectively. [0088] However, attempts to reduce the potential risk of wear and fracture in ankle hemiarthroplasty have never been made in the literature. In order to further evaluate the potential risks of metal implants and provide possible solution to improve the interaction between the biological cartilage and metal talus implants, the current study aimed to quantify the contact characteristics of three different types of metal implants with and without a PCU coating, using a previously validated finite element (FE) model of the talus (Liu et al., 2021). We hypothesized that when under a static load, the PCU-coated talus implants would have contact characteristics more similar to the biological talus than would the non-PCU-coated implant. [0089] Methods and materials

[0090] Overview of the talus implants

[0091] In this study, we adopted three types of talus implants (customized, SSM [statistical shape model], and universal) from previous studies (Liu et al., 2020; Trovato et al., 2017), which defined these talus shapes as follows: the customized implant template was directly reconstructed using CT scans of subjects’ talus; the SSM implant template was based on the average shape of a sample of subjects (n = 98 tali), calculated using a statistical shape modeling approach (Liu et al., 2020); and the universal implant template was the talus, selected from a sample of n = 98 tali, that had the least total deviation from the rest of the tali (Trovato et al., 2017). The different implant types were evaluated with and without PCU covering the cartilage regions by replacing the biological talus (Figure 1).

[0092] Implant selection for each subject

[0093] This study was based on data from four embalmed cadaveric subjects, with CT scans their tali taken in neutral standing (Liu et al., 2021). Of a total dataset of 10 embalmed cadavers, four were selected at random (6 female and 4 male with age at death 84.5±12), with ethical approval from the university research ethics board (No. Pro00026057). A high-resolution Somatom definition flash scanner was employed with the following specifications: 0.8 mm pitch, 0° gantry tilt, 300 effective mAs, 80 kV voltage, and 1.0 s rotation time. These images (512x512 pixels) were imported into MIMICS software (Materialize, NV, Belgium) for segmentation.

[0094] For each subject, we tested the three implant types (Customized, SSM, Universal) with and without a PCU coating. For convenience, PCU coated implants will herein be referred to as PCU implants and non-PCU coated implants that have their original metal (Cobalt-Chrome [CoCr]) surface will be referred to as CoCr implants (Figure 1). As such, we constructed the following models from each of our four subjects: PCU-coated models of the three implant types (Customized-PCU, SSM-PCU, Universal- PCU), non-PCU coated models of the three implant types (Customized-COCr, SSM- CoCr, Universal-CoCr). In addition to the implant type models, each subject also had a biological model, directly derived and reconstructed from CT scans of the cadaveric subject, which served as a point of comparison. In this way, we had seven models (6 implant types, 1 biological model) for each of the four subjects, for a total of 28 models.

[0095] Each subject’s implant types were processed with reference to that subject in order to promote implant fit, though the procedures were consistent and as follows. The metal portion of the Customized-PCU implants had the same volume and shape as that of the biological talus, as customized implants are individualized based on the CT scans of each subject. Universal-PCU implants were selected using a previously developed cartilage-exclusive sizing guide (Trovato et al. 2017), whereby tali were divided into 10 size ranges based on volume. The appropriate size was selected by finding the size that matches the biological talus volume. The SSM-PCU implants were chosen using the same sizing guide as the Universal-PCU implants. For CoCr implants, there was a need to account for the loss of cartilage volume, so we used different scaling methods for each implant type. The appropriate size of Universal-CoCr implant was selected by referring to the cartilage-inclusive talus sizing chart by Trovato et al. (2017) to find the volume range that each subjects’ talus would fall in. Customized and SSM implants were scaled to the same volume as that of the universal implant using scale factors in Geomagic (3D Systems®, Morrisville, USA). As a result of scaling to account for loss of cartilage volume, the metal parts of CoCr implants are larger as compared to the metal parts of PCU implants.

[0096] Finite element model

[0097] The 3D reconstructed models of the cadaver bones were removed of spikes using Geomagic software (3D Systems®, Morrisville, USA) and meshed using Hypermesh (Altair®, Troy, USA). We developed a total of seven models for each subject: one biological model directly reconstructed from CT scans, and six implant models in which the talar implant replaced the biological talus bone. These implants were aligned with the biological talus using a best-fit alignment tool in Geomagic (3D Systems®, Morrisville, USA) before replacement and meshing. Then, the adjacent bones (tibia, fibula, calcaneus, and navicular) were translated a small distance away from the talus to leave room for the creation of the cartilage layers.

[0098] The current study used two sets of FE meshes: one for the biological and PCU models, and the other for the CoCr models. For the biological and PCU models, the articular regions of all cortical bones were first meshed with 4-node shell elements which were then extruded 1 .5 mm along the normal direction of the surface to form 4-layers of solid elements representing the natural cartilage or PCU (Anderson et al., 2007). The remaining cortical surfaces for all bones were meshed using 3-node shell elements. All shell elements in these models had 1 mm thickness. For CoCr models, the cartilage regions on the cortical bones of the tibia, fibula, calcaneus, and navicular were meshed using 4-node shell elements first, and then extruded along the normal direction for 1.5 mm to create cartilage; their corresponding cartilage surfaces on the talus implants were meshed using 4-node shell elements only. Lastly, the remaining cortical bones were modelled with 3-node shell elements. The 3-node shell elements were 1 mm in thickness The cancellous bone was not considered in these models. A detailed description of each component is shown in Table 1.

Table 1. Element type and material properties of the FE model

Components type Material behavior Mechanical properties Reference

Cortical bone S3 Linear elastic E=19,000 MPa, v=0.3 Mondal and Ghosh (2019)

Cartilage C3D8 Hyperelastic (Ogden) p=2.43, a=12.45, D=0.176 Brown et al. (2009)

Talar Implant (CoCr) S3 Linear elastic E=210,000 MPa, v=0.3 Al Jabbari (2014)

PCU C3D8 Hyperelastic (Ogden) p=3.23, CF3.31, 0=0.2 IDSM corp.

(E: Youngs modulus; v: passion ratio; Strain energy potential of Ogden hyperelastic model is: U = — (Z + Z> + Z> - 3) + - (J^el - I)²; where p, a and D are material constants; Z is deviatoric principal stretches; J^el is elastic volume strain)

[0099] Four thin layers of the most distal elements of the bones adjacent to the talus were defined as rigid bodies (Figure 2). The fibula was constrained to move together with the tibia along the Z-axis (superior-inferior), as both bones tend to move cranially and caudally together in real-life movements. The model was also subjected to three consecutive static steps: (1) moving adjacent bones back to their CT scan positions to create initial contact while fixing the talus; (2) releasing all 6 degrees of freedom (DOF) for the talus and maintaining adjacent bone positions as in step 1 ; (3) allowing translational DOF along the Z-axis for the tibia and fibula, and a rotational DOF along the Z-axis for the fibula. Lastly, we applied a compressive force of 2000 N to the reference points of the tibia bone along the Z-axis to simulate standing. The entire simulation was performed in ABAQUS (Dassault Systemes®, Johnston, USA). Contact characteristics including contact pattern, contact area, and peak contact pressure in all the cartilage surfaces of the four subjects using both CoCr and PCU talus implants were extracted and compared to their corresponding biological models.

[00100] Results

[00101 ] Contact pattern

[00102] A visual comparison of CoCr and PCU models showed an intra-subject and intra-implant type trends. For the same subject and with same implant type (customized, SSM, or universal), there was a similar contact pattern in the tibial and calcaneal cartilage in both CoCr and PCU implants, but slight differences were noted in

10

SUBSTITUTE SHEET (RULE 26) the fibular and navicular cartilage. When comparing, each implant type to the biological counterpart of each subject for individual cartilage contact patterns, it was found that for all subjects, the customized type implants had the most similar contact pattern to the biological, followed by SSM and universal types (Figure 3).

[00103] When we visually compared the contact patterns of each cartilage and implant model across subjects, it was observed that contact patterns varied with subjects. To illustrate, we compared the fibular cartilage in the customized type model for subjects 1-4, then the fibular cartilage in SSM type models, etc., and found that these patterns varied from one subject to another (Figure 3).

[00104] Peak contact pressure

[00105] In the biological models, cartilage peak contact pressure generally varied between 6.03 MPa and 15.44 MPa, with a few exceptions: Subject 3’s fibular cartilage reached 22.59 MPa, Subject 4’s fibular cartilage reached 21 .70 MPa, and Subject 4’s calcaneal cartilage reached 22.18 MPa.

[00106] To compare the implant models with their respective biological models, the peak contact pressure for each cartilage in each subject was normalized to their biological counterparts (Table 2). When compared to the biological, our findings showed a general trend: CoCr implants tended to have larger peak contact pressures than the biological, whereas PCU implants had similar peak values as the biological. More specifically, the Universal-CoCr implant had the most pronounced average peak contact pressure, at 4.4 times that of the biological model; this was followed by the SSM-CoCr at an average peak 3.9 times that of the biological, and the Customized-CoCr at an average of 2.8 times. PCU implants had average peak contact pressures that more closely matched that of the biological, with Customized-PCU at 0.9 times the biological, SSM- PCU at 1.1 times, and Universal-PCU at 1.4 times.

Table 2. Heatmap of the normalized peak contact pressure of four subjects

_ . . . . . Customized SSM Universal

Subject Name _{CoCr pcu CoCr pcu CoCr pcu}

Fibula 0.9 0.9 2.1 1.3 1.5 1.1 . Tibia 2.6 0.9 2.8 1.1 3.8 1.1 calcaneus 2.0 0.8 3.1 1.0 2.7 0.9

Navicular 2.6 0.8 2.3 1.3 3.6 1.3

Fibula 3.5 0.8 2.4 0.8 3.4 1.0 e Tibia 1.9 1.0 10.3 3.2 21.0 7.0 ouujuui i calcaneus 3.5 0.8 8.1 1.5 8.7 1.4

Navicular 3.0 1.0 1.5 0.8 2.2 1.1

Fibula 2.5 0.8 1.0 0.4 0.8 0.5 Tibia 5.6 1.0 5.6 1.4 2.2 1.2 C_a|_{caneus 1 4 0 9} 2.8 1.0 2.9 1.2

Navicular 3.0 1.0 1.4 0.9 1.7 0.8

Fibula 2.8 0.8 0.6 0.3 0.4 0.3 Tibia 4.2 1.0 8.4 1.2 6.1 0.9 calcaneus 2.8 0.8 3.0 0.6 2.3 0.6

Navicular 2.9 0.9 7.2 1.4 7.9 1.9

Mean 2.8 0.9 3.9 1.1 4.4 1.4

(The maximum values are highlighted in red and the minimum values are in blue)

[00107] Contact area

[00108] We grouped the cartilage regions based on location and compared the contact areas across the four subjects (Table 3). For all subjects, the biological model had a relatively large cartilage contact area, with the exception of the fibular cartilage in Subject 1. The average cartilage contact areas in the biological models were: fibula at 281.97 mm², tibia at 1093.47 mm², calcaneus at 870.47 mm², and navicular at 505.42 mm².

12

SUBSTITUTE SHEET (RULE 26) Table 3. Heatmap of the contact area (mm²) of four subjects in each individual cartilage

Cart uiilaayaec CuKiar'tc Ri l nir'nl Customized SSM Universal

Name ^M CoCr PCU CoCr PCU CoCr PCU

Subject 1 212.66 146.41 217.59 154.74 218.88 207.22 265.05

Subject 2 353.85 286.63 365.06 261.82 350.65 243.03 330.42

Fibula Subject s 295.26 297.3 314.23 220.24 251.45 206.88 332.26

Subject 4 266.12 174.36 281.79 120.7 231.32 112.18 208.96

Mean 281.97 226.18 294.67 189 38 263.08 192.33 284.17

Subject 1 1114.48 872.05 940.47 827.71 933.66 614.23 878.84

Subject 2 1119 853.56 775.86 683.3 736.74 408.18 566.09

Tibia Subject s 1150 9 837.87 830.34 752 71 744.11 818.71 731.97

Subject 4 989.48 765.83 722.25 762.03 1018.79 732.42 1004.24

Mean 1093.47 832.33 817.23 756.44 858.33 643.39 795.29

Subject 1 796.91 597.79 801.7 513 62 669.01 493.21 757.81

Subject s 964.17 621.87 955.65 596.18 844.03 686.84 907.04

Calcaneus Subject s 1006.04 887.91 984.43 640.65 806.62 760.26 912.01

Subject 4 714.77 496.21 705.79 452 18 667.97 514.96 756.74

Mean 870.47 650.95 861.89 550.66 746.91 613.82 833.40

Subject 1 437.98 363.84 437.49 293.88 348.18 203.19 301.28

Subject s 559.64 461.71 551.86 537 78 592.37 496.44 529.68

Navicular Subject s 542.69 377.32 440.74 302.06 411.26 362.28 487.09

Subject 4 481.35 384.89 471.25 248.38 442.61 197.89 367.3

Mean 505.42 396.94 475.34 345 53 448.61 314.95 421.34

(The gradation of colors on the heatmap were derived relative to each individual cartilage type. The maximum values in each cartilage are highlighted in blue and the minimum values were in red)

[00109] Then, we compared the average contact area of each cartilage in the biological model with their CoCr counterparts. In general, CoCr implant models had lower contact areas. To illustrate, the Customized-CoCr had the largest contact area amongst all CoCr models based on the dark shades of blue in Table 3, with 226.18 mm² in the fibula, 832.33 mm² in the tibia, 650.95 mm² in the calcaneus and 396.94 mm² in the navicular. Next was SSM-CoCr with the second largest contact area, with 189.38 mm² in the fibula, 756.44 mm² in the tibia, 550.66 mm² in the calcaneus, 345.53 mm² in the navicular; and lastly the Universal-CoCr with 192.33 mm² in the fibula, 643.39 mm² in the tibia, 613.82 mm² in the calcaneus, 314.95 mm² in the navicular, noting that these values were all less than the biological model.

[00110] In contrast, when we compared the average cartilage contact areas between the biological and their PCU counterparts, we found that PCU models showed

13

SUBSTITUTE SHEET (RULE 26) more similar contact areas to the biological. For a combined average of the four subjects, the biological-PCU contact area variations for each individual cartilage ranged as follows: from 2.20 mm² (0.78%) to 18.90 mm² (6.70%) in the fibula, 235.14 mm² (21.50%) to 298.18 mm² (27.27%) in the tibia, 8.58 mm² (0.99%) to 123.57 mm² (14.20%) in the calcaneus, and 30.08 mm² (5.95%) to 84.04 mm² (16.64%) in the navicular.

[00111] Discussion

[00112] The current study aimed to quantify the effect of adding a layer of compliant material (PCU) over a stiff material (cobalt chromium) in talus implants. To do so, we used a validated FE model of the ankle joint to evaluate the contact characteristics of a set talus implants previously designed by our group. There were three previously designed implant types (customized, SSM, and universal), and each implant was subjected to two conditions (with or without PCU) for a total of six implant models per subject. For the four cadaveric subjects used in this study, we used computational simulations to investigate the contact pattern, peak contact pressure, and contact area of the implant models and compared them with their biological counterparts.

[00113] Biological FE models of the four subjects have been previously validated by replicating the experiment conducted by Anderson et al., (2007), where contact characteristics in tibiotalar cartilage was measured using a sensor under an axial load of 600 N. We used the same joint structures and applied the same loading, boundary conditions, and postures reported in Anderson et al. (2007) to our biological FE models of tibiotalar joint for four subjects. Our predicted contact pattern, mean contact pressure, and contact areas were comparable with the experimental results of Anderson et al. (2007). More information on the validation can be found in Liu et al. (2021).

[00114] Despite the significant role the natural cartilage plays in redistributing loading (van Lenthe et al., 1997) and lubricating joint contact surfaces (Sophia Fox et al., 2009; Xia et al., 2016), there is no knowledge to date of talus implant designs that have incorporated natural cartilage-like materials. Instead, all talus implants have been made of stiff material such as titanium, ceramic, and CoCr (Tanaka et al., 2003; Tonogai et al., 2017; Tracey et al., 2019; Wagener et al., 2017). To our best knowledge, the load distributing ability of stiff material talus implants has not yet been examined in literature.

[00115] While stiff material-based implants have gained clinical success (Angthong, 2014; Bowes et al., 2019; Gadkari et al., 2013; Tonogai et al., 2017), our study found that there are still pronounced contact characteristic variations when they are compared to their biological counterparts. A general trend of increased peak contact pressure (Table 2) and reduced contact area (Table 3) was noticed in CoCr models, thereby suggesting that stiff talus implants could potentially lead to contact stress concentrations. Thus, under the same magnitude of load, CoCr implants would produce larger contact stresses. Our results further showed that contact patterns varied with implant type, but for the same implant type, both PCU and CoCr models showed a similar contact pattern at most cartilage locations. It may be a potential interpretation that contact pattern is more influenced by implant shape than by PCU coating, though further investigations will be needed (Figure 3).

[00116] Interestingly, all subjects’ PCU coated implants showed similar peak contact pressure and contact area to their respective biological models. Not only did this show the important role of cartilage in the ankle joint, but also supported our hypothesis that PCU covered implants would produce contact characteristics similar to those of natural cartilage under static loading, as opposed to CoCr implants.

[00117] A further observation is that in the implant models, the Universal-CoCr and Universal-PCU models had the highest peak contact pressure in the tibial cartilage of Subject 2, at 151.60 MPa and 50.11 MPa respectively. Unlike the tibial contact pattern of other implant models, these two models showed a localized contact contour which might due to the fact that universal talus implants have a poorer fit with adjacent bones in Subject 2, as the universal was originally selected based on the criterion of “least difference” with the rest of the tali in the dataset (Trovato et al., 2017).

[00118] After evaluating CoCr and PCU talus implants in subjects 1-4, our study found that CoCr implants all had a high peak contact pressure, which could lead to adjacent cartilage wear and bone fractures in the long-term. In terms of contact pattern and contact area similarity to the biological talus, Customized-CoCr was the closest in the CoCr models, followed by the SSM and then the universal implants. It should be noted that the geometry of customized models used in this study are similar to that of the original talus, while in clinical practice, customized implant is created using the geometry of the opposite talus that is mirror-imaged due to the deformity of the affected talus. Thus, the custom talus results of this study are most likely better than those of clinical practice, as it has been shown that there is some minor deviation when comparing contralateral tali (Angthong et al., 2020; Islam et al., 2014). In contrast, all PCU coated implants showed similar peak contact pressure and contact area to their biological counterparts, and showed greater resemblance to the biological than did CoCr models. This indicates that PCU implants have an advantage over CoCr implants when it comes to load distribution ability. While Customized-PCU had the most similar contact characteristics to the biological (Table 2), SSM-PCU had the second most comparable contact characteristics to the biological, and it has additional advantages over other implants (i.e. high manufacturing cost of customized implants; possibility of localized contact area in universal implants).

[00119] It should also be noted that in our study, the cartilage is assumed to be uniform on the bones due to the difficulty of replicating non-uniform cartilage thickness in computational modelling. The assumption and use of uniform cartilage have been seen in other studies as well (e.g. Anderson et al., 2007; 2006). Applying uniform cartilage in the model could influence the accuracy with which contact pressures and areas are predicted. Notably, when Anderson et al. (2007) compared the contact characteristics between using uniform cartilage talus models with their experimental data from in-vitro testing, they found there was only a conservative estimate of 10-15% of variations between the two methods.

[00120] Conclusion

[00121] This study aimed to quantify the effect PCU coating can have on three previously designed metal-based implants, by evaluating the contact characteristics of adjacent cartilage when the implants are coated or not with PCU. Our findings showed that coating metal talus implants (PCU models) can improve peak contact pressures and contact areas, and lead to implant contact characteristics that are similar to that of the biological talus. These results are valuable and confirmed the benefit of compliant material coating on contact characteristics of metal-based talus implant, suggesting that clinical benefits may be derived from a compliant material coating on existing metalbased implant technology and can provide valuable insight on the future design of talus implants.

[00122] References

[00123] Anderson, D.D., Goldsworthy, J.K., Li, W., James Rudert, M., Tochigi, Y., Brown, T.D., 2007. Physical validation of a patient-specific contact finite element model of the ankle. Journal of Biomechanics 40, 1662-1669. https://doi.Org/10.1016/j.jbiomech.2007.01.024

[00124] Anderson, D.D., Goldsworthy, J.K., Shivanna, K., Grosland, N.M., Pedersen, D.R., Thomas, T.P., Tochigi, Y., Marsh, J.L., Brown, T.D., 2006. Intra-articular contact stress distributions at the ankle throughout stance phase-patient-specific finite element analysis as a metric of degeneration propensity. Biomech Model Mechanobiol 5, 82-89. https://doi.Org/10.1007/S10237-006-0025-2

[00125] Al Jabbari, Y.S., 2014. Physico-mechanical properties and prosthodontic applications of Co-Cr dental alloys: a review of the literature. J Adv Prosthodont 6, 138- 145. https://doi.Org/10.4047/jap.2014.6.2.138

[00126] Ando, Y., Yasui, T., Isawa, K., Tanaka, S., Tanaka, Y., Takakura, Y., 2016. Total Talar Replacement for Idiopathic Necrosis of the Talus: A Case Report. J Foot Ankle Surg 55, 1292-1296. https://doi.Org/10.1053/j.jfas.2015.07.015

[00127] Angthong, C., 2014. Anatomic Total Talar Prosthesis Replacement Surgery and Ankle Arthroplasty: An Early Case Series in Thailand. Orthop Rev (Pavia) 6. https://doi.org/10.4081/or.2014.5486

[00128] Angthong, C., Rajbhandari, P., Veljkovic, A., Piyaphanee, A., Stufkens, S.A.S., Wibowo, R., 2020. Morphometric geometric differences between right and left human tali: A cadaveric study of fluctuating asymmetry via systematic measurement and three-dimensional scanning. PLoS One 15, e0232012. https://doi.org/10.1371/journal.pone.0232012

[00129] Bowes, J., Adeeb, S., Grosvenor, A., Beaupre, L, Jomha, N.M., 2019. Development and Implantation of a Universal Talar Prosthesis. Front. Surg. 6. https://doi.Org/10.3389/fsurg.2019.00063

[00130] Brown, C.P., Nguyen, T.C., Moody, H.R., Crawford, R.W., Oloyede, A., 2009. Assessment of common hyperelastic constitutive equations for describing normal and osteoarthritic articular cartilage. Proc Inst Meeh Eng H 223, 643-652. https://doi.org/10.1243/09544119JEIM546

[00131] de Ruiter, L, Janssen, D., Briscoe, A., Verdonschot, N., 2017. The mechanical response of a polyetheretherketone femoral knee implant under a deep squatting loading condition. Proc Inst Meeh Eng H 231 , 1204-1212. https://doi.Org/10.1177/0954411917738805

[00132] de Ruiter, L, Rankin, K., Browne, M., Briscoe, A., Janssen, D., Verdonschot, N., 2021. Decreased stress shielding with a PEEK femoral total knee prosthesis measured in validated computational models. Journal of Biomechanics 118, 110270. https://doi.Org/10.1016/j.jbiomech.2021 .110270 [00133] DSM Corp., 2021 . [WWW Document], URL https://www.dsm.com/corporate/home.html (accessed 5.5.2021). [00134] Elsner, J. J., McKeon, B.P., 2017. Orthopedic Application of Polycarbonate

Urethanes: A Review. Techniques in Orthopaedics 32, 132-140. https://doi.org/10.1097/BTO.0000000000000216

[00135] Elsner, J. J., Portnoy, S., Zur, G., Guilak, F., Shterling, A., Linder-Ganz, E.,

2010. Design of a free-floating polycarbonate-urethane meniscal implant using finite element modeling and experimental validation. J Biomech Eng 132, 095001. https://doi.Org/10.1115/1 .4001892

[00136] Gadkari, K.P., Anderson, J.G., Bohay, D.R., Maskill, J.D., Padley, M.A., Behrend, L.A., 2013. An Eleven-Year Follow-up of a Custom Talar Prosthesis After Open Talar Extrusion in an Adolescent Patient: A Case Report. JBJS Case Connect 3, e118. https://doi.Org/10.2106/JBJS.CC.L.00331

[00137] Guyen, O., 2019. Hemiarthroplasty or total hip arthroplasty in recent femoral neck fractures? Orthopaedics & Traumatology: Surgery & Research, 2018 Instructional Course Lectures (SoFCOT) 105, S95-S101. https://doi.Org/10.1016/j.otsr.2018.04.034

[00138] Harnroongroj, Thos, Harnroongroj, Thossart, 2014. The Talar Body Prosthesis: Results at Ten to Thirty-six Years of Follow-up. J Bone Joint Surg Am 96, 1211-1218. https://doi.Org/10.2106/JBJS.M.00377

[00139] Hedbeck, C.J., Blomfeldt, R., Lapidus, G., Tbrnkvist, H., Ponzer, S., Tidermark, J., 2011 . Unipolar hemiarthroplasty versus bipolar hemiarthroplasty in the most elderly patients with displaced femoral neck fractures: a randomised, controlled trial. International Orthopaedics (SICOT) 35, 1703-1711.

[00140] Islam, K., Dobbe, A., Komeili, A., Duke, K., El-Rich, M., Dhillon, S., Adeeb, S., Jomha, N.M., 2014. Symmetry analysis of talus bone. Bone & Joint Research 3, 139— 145. https://doi.Org/10.1302/2046-3758.35.2000264

[00141] Jarvenpaa, J., Soininvaara, T., Kettunen, J., Miettinen, H., Kroger, H., 2014. Changes in bone mineral density of the distal femur after total knee arthroplasty: a 7-year DEXA follow-up comparing results between obese and nonobese patients. Knee 21 , 232-235. https://doi.Org/10.1016/j.knee.2013.03.004

[00142] Jia, Z., Ding, F., Wu, Y., Li, W., Li, H., Wang, D., He, Q., Ruan, D., 2015.

Unipolar versus bipolar hemiarthroplasty for displaced femoral neck fractures: a systematic review and meta-analysis of randomized controlled trials. Journal of Orthopaedic Surgery and Research 10, 8. https://doi.org/10.1186/s13018-015-0165-0 [00143] Katsui, R., Takakura, Y., Taniguchi, A., Tanaka, Y., 2019. Ceramic Artificial Talus as the Initial Treatment for Comminuted Talar Fractures [Formula: see text]. Foot Ankle Int 1071100719875723. https://doi.org/10.1177/1071100719875723 [00144] Lavernia, C.J., Rodriguez, J.A., lacobelli, D.A., Hungerford, D.S., Krackow, K.A., 2014. Bone mineral density of the femur in autopsy retrieved total knee arthroplasties. J Arthroplasty 29, 1681-1686. https://doi.Org/10.1016/j.arth.2014.03.010 [00145] Liu, T., Jomha, N.M., Adeeb, S., El-Rich, M., Westover, L, 2020. Investigation of the Average Shape and Principal Variations of the Human Talus Bone Using Statistic Shape Model. Front. Bioeng. Biotechnol. 8. https://doi.org/10.3389/fbioe.2020.00656

[00146] Liu, T., Jomha, N.M., Adeeb, S., El-Rich, M., Westover, L, 2021 . The evaluation of artificial talus implant on ankle joint contact characteristics: A finite element study of four subjects. Medical & Biological Engineering & Computing. (Under review) [00147] Mondal, S., Ghosh, R., 2019. Effects of implant orientation and implant material on tibia bone strain, implant-bone micromotion, contact pressure, and wear depth due to total ankle replacement. Proc Inst Meeh Eng H 233, 318-331. https://doi.Org/10.1177/0954411918823811

[00148] Rankin, K.E., Dickinson, A.S., Briscoe, A., Browne, M., 2016. Does a PEEK Femoral TKA Implant Preserve Intact Femoral Surface Strains Compared With CoCr? A Preliminary Laboratory Study. Clin Orthop Relat Res 474, 2405-2413. https://doi.org/10.1007/s11999-016-4801-8

[00149] Ruatti, S., Corbet, C., Boudissa, M., Kerschbaumer, G., Milaire, M., Merloz, P., Tonetti, J., 2017. Total Talar Prosthesis Replacement after Talar Extrusion. J Foot Ankle Surg 56, 905-909. https://doi.Org/10.1053/j.jfas.2017.04.005

[00150] Sophia Fox, A.J., Bedi, A., Rodeo, S.A., 2009. The Basic Science of Articular Cartilage. Sports Health 1 , 461-468. https://doi.org/10.1177/1941738109350438 [00151] Tanaka, Y., Takakura, Y., Kadono, K., Taniguchi, A., Hayashi, K., lida, J., Sugimoto, K.I., Tohma, Y., Ohgushi, H., 2003. Alumina Ceramic Talar Body Prosthesis for Idiopathic Aseptic Necrosis of the Talus. KEM 240-242, 805-808. https://doi.org/10.4028/www.scientific.net/KEM.240-242.805

[00152] Taniguchi, A., Tanaka, Y., 2019. An Alumina Ceramic Total Talar Prosthesis for Avascular Necrosis of the Talus. Foot Ankle Clin 24, 163-171. https://doi.Org/10.1016/j.fcl.2018.10.004 [00153] Tochigi, Y., Rudert, M.J., Amendola, A., Brown, T.D., Saltzman, C.L., 2005. Tensile engagement of the peri-ankle ligaments in stance phase. Foot Ankle Int 26, 1067-1073. https://doi.Org/10.1177/107110070502601212

[00154] Tonogai, I., Hamada, D., Yamasaki, Y., Wada, K., Takasago, T., Tsutsui, T., Goto, T., Sairyo, K., 2017. Custom-Made Alumina Ceramic Total Talar Prosthesis for Idiopathic Aseptic Necrosis of the Talus: Report of Two Cases [WWW Document], Case Reports in Orthopedics, https://doi.org/10.1155/2017/8290804

[00155] Tracey, J., Arora, D., Gross, C.E., Parekh, S.G., 2019. Custom 3D-Printed Total Talar Prostheses Restore Normal Joint Anatomy Throughout the Hindfoot. Foot & Ankle Specialist 12, 39-48. https://doi.org/10.1177/1938640018762567

[00156] Trovato, A., El-Rich, M., Adeeb, S., Dhillon, S., Jomha, N., 2017.

Geometric analysis of the talus and development of a generic talar prosthetic. Foot and Ankle Surgery, Total ankle Replacement 23, 89-94. https://doi.Org/10.1016/j.fas.2016.12.002

[00157] van Lenthe, G.H., de Waal Malefijt, M.C., Huiskes, R., 1997. Stress shielding after total knee replacement may cause bone resorption in the distal femur. The Journal of Bone and Joint Surgery. British volume 79-B, 117-122. https://doi.Org/10.1302/0301 -620X.79B1 .0790117

[00158] van Lenthe, G.H., Willems,

Verdonschot, N., de Waal Malefijt, M.C., Huiskes, R., 2002. Stemmed femoral knee prostheses: effects of prosthetic design and fixation on bone loss. Acta Orthop Scand 73, 630-637. https://doi.Org/10.1080/000164702321039589

[00159] Wagener, J., Gross, C.E., Schweizer, C., Lang, T.H., Hintermann, B., 2017. Custom-made total ankle arthroplasty for the salvage of major talar bone loss. Bone Joint J 99-B, 231-236. https://doi.org/10.1302/0301-620X.99B2.BJJ-2016-0504.R2

[00160] Wan, L., Asia, R.J. de, Rubash, H.E., Li, G., 2008. In vivo cartilage contact deformation of human ankle joints under full body weight. Journal of Orthopaedic Research 26, 1081-1089. https://doi.org/10.1002/jor.20593

[00161] Xia, Y., Momot, K.I., Chen, Z., Chen, C.T., Kahn, D., Badar, F., 2016. CHAPTER 1 Introduction to Cartilage 1-43. https://doi.org/10.1039/9781782623663- 00001

[00162] EXAMPLE 2

[00163] Abstract [00164] Customized talus implants have been regarded as a better treatment alternative to talus avascular necrosis than traditional surgical fusion because of its ability to maintain joint mobility while ameliorating pain. Despite the use of ankle hemiarthroplasty clinically, the cartilage contact characteristics of adjacent bones remain unclear. This study aims to use finite element modelling to evaluate the contact characteristics of three cobalt-chrome talus implants in three postures, in four subjects. This study also compared the contact area, contact pressure, and peak contact pressure of the implant models with a reference biological model. Amongst the various biological and implant models, our results showed that the biological models generally had the largest contact areas and smallest peak contact pressures, whereas the implant type models had smaller contact areas and relatively larger peak contact pressure. Moreover, amongst the three implant types, customized-scale models showed a larger total contact area than that of the SSM-scale and universal-scale models, but their variation was relatively limited. The results from this study can have significance in future endeavors into ankle joint modelling, and as well as being able to improve implant design to enhance recovery outcomes for patients who may benefit from talar replacement.

[00165] 1. Introduction

[00166] Avascular necrosis is when inadequate blood supply results in the death of bone. There may be instances where bones crack and eventually collapse due to this lack of circulation [1], Avascular necrosis in the talus can be the result of varied etiologies, such as medication effects, idiopathic, or talus fractures, and can cause debilitating pain and mobility limitations for the individuals afflicted [2, 3], In terms of treatment, tibial-talar- calcaneal fusion has been performed, but often leads to mobility limitations [4, 5], As an emerging and promising treatment method, customized talus implants have shown greater improvements to pain and mobility than the traditional method of surgical fusion [6, 7], However, producing a custom implant can require a considerable amount of time, which increases waiting time between bone injury and surgery resulting in worse clinical outcomes [2, 3], Furthermore, custom-made talus implants have the disadvantage of high costs due to their individualized nature. To capitalize on the benefits of talar replacement over other surgical approaches, while reducing issues related to wait-time and cost, a series of universal implants with different sizes using a single shape template has been developed [8, 9], Universal implants can improve cost-effectivenss and production speed; and in doing so, promote improved clinical outcomes while maintaining a degree of customized-fit for patients with regards to size [8, 9] [00167] Mechanical stress in intra-articular cartilage of diarthrodial joints can be considered as an indirect measure of the joint’s level of well-being [10], In order to understand the mechanical environment of the joint surrounding the talus, earlier experimental studies attempted to measure in-vitro contact characteristics of ankle joints under various postures and magnitudes of loading using cadavers. These attempts include Kimizuka et al. (1980) [11], who measured the contact area in neutral standing under 1500 N and concluded that the position of the contact area at the talocrural joint was mainly anterolateral in neutral standing. Millington et al. (2007) [12] also measured the contact area at the ankle joint under a physiological load magnitude of 1000N, and found that dorsiflexion had the largest contact area of the postures tested (neutral standing, 20° dorsiflexion, plantarflexion, supination and pronation). Moreover, the contact characteristics associated with fractured ankle joints have been investigated. It was reported that contact area decreased when the size of the posterior malleolar fragment increased [13], and that the contact stress pattern that occurred in displaced lateral malleolar fractures was mild and not linked to late secondary degeneration [14], These studies have pointed to the importance of mechanical stress and contributed valuable insights into the understanding of the mechanical environment of natural ankle joints. However, while hemi-arthroplasty has been tested using clinical trials for ankle joint [7, 15, 16], the contact characteristics of natural cartilage and metal implants remains unclear in the joints surrounding the talus.

[00168] Due to the limitations of direct measurements, finite element (FE) models have emerged as an effective way of investigating the mechanical environment of the ankle because of their non-invasiveness and cost-effectiveness. For instance, previous studies have used FE to investigate the effect of degenerative propensity on the tibio-talar cartilage [10] and the effect of focal resurfacing with a metallic implant on ankle contact mechanics [17], Furthermore, several studies have investigated the performance of ankle prostheses [18-20], However, little is known about the effect of talus implants on the cartilage at the level of the entire talus joint using FE modeling. Thus, the current study aims to develop a a high-fidelty FE model of the talus and surrounding joints and investigate the contact characteristics of various implants that were previously designed [4].

[00169] 2. Material and Methods

[00170] 2.1 Geometry acquisition [00171] The current study used four embalmed cadavers for initial investigation of the effect of talus implants. These four cadavers were randomly selected from a dataset of 10 embalmed cadavers (6 female and 4 male with age at death 84.5±12) .obtained from the University of Alberta Anatomy Department with ethical approval from the university research ethics board. The rationale for reserving preliminary investigations to four cadavers was largely due to the high computational load involved in the development of FE models, with plans to expand investigations to other cadavers in future research. [00172] In the previously published experimental cadaver study [21], the foot was first dissected at a point approximately 100 mm from the geometric center of the ankle joint, and then CT scanned under three postures: 20° dorsiflexion (DF), neutral standing (NS) and -20° plantarflexion (PF). A high-resolution Somatom definition flash scanner with the following specifications was used: 0.8 mm pitch, 0° gantry tilt, 300 effective mAs, 80 kV voltage and 1 .0 s rotation time. Images (512x512 pixels) with a slice thickness of 0.6 mm were obtained at 0.1 mm increments. These images were imported into MIMICS (Materialize, NV, Belgium, Version 20.0). Bones of the talus joint were segmented and digitized into STL file format. These files were then cleaned of spikes and smoothened using built in command (“relax polygons”) in Geomagic (3D Systems®, Morrisville, USA, Version 2014) in order to obtain a detailed 3D geometry for geometric comparision and finite element analysis.

[00173] 2.2 Biological talus and artificial implants

[00174] The current study evaluated one biological model and three implant types (customized, universal [9], and statistical shape model (SSM) [4]) for each cadaveric subject. The customized implant was directly reconstructed from CT scans of subjects’ talus; the universal implant was a talus selected from 98 tali that had the least total deviation from the rest of the tali [9]; and the SSM implant was the average shape of a sample of subjects (n=98 tali), obtained via statistical shape modeling [4, 22], For subject 1 , each implant type included three models: original, scale, and offset. Together, these make up a total of 1 biological and 9 implant models. When implant “type” is used, we refer inclusively to all three original, scale, and offset models within a type; and to refer to a specific model (e.g. the original in the customized type), we will use the nomenclature type-model for clarity (e.g. customized-original). Note that subjects 2-4 used a reduced set of models with one model of each implant type as described in Section 2.4.

[00175] The biological model was obtained from the CT scans of subject 1 , with a layer of 1 .5 mm thick cartilage [22] added in FE. The customized-original was also obtained from CT scans of subject 1 , while the universal and SSM- originals were obtained from previous studies [4, 9], Scale and offset implants were developed by modifying their respective original implants. We used scale and offset (bulit-in functions in Geomagic software) to increase the entire volume of the three original talus implants to compensate for the layer of cartilage that exists in a natural talus bone. For offset implants, we offset the surface of the talus 0.5 mm along the normal direction of each triangle surface of the bone (Figure 5). For scaled implants, we used a scale factor to increase the volume by one-third of the volume of the cartilage, which is also the volume of an 8-node solid elements in the FE model of the biological talus (Figure 6).

[00176] 2.3 Finite element model

[00177] The 3D geometry of the biological ankle joint (tibia, fibula, calcaneus, navicular and talus bones) that we obtained for each subject was meshed using HyperMesh (Altair®, Troy, USA, Version 2019). For the biological models, all bones were meshed directly. For implant models, it was aligned with the biologcial talus first using the ‘best-fit’ command in Geomagic (3D Systems®, Morrisville, USA) before meshing. The adjacent bones were translated a small distance away from the talus in order to create cartilage layers and avoid the interference of any cartilage surfaces and bones.

[00178] For the biological model, the articular surfaces of the adjacent bones (tibia, fibula, calcaneus and navicular bones) were used to create solid cartilage on the bones. Thus, they were first meshed with 4-node shell elements and then extruded at 1 .5 mm [22] for all talar articular surface elements to create 4 layers of solid cartilage. The non- articular surfaces of the adjacent bones were meshed using 3-node shell elements.

Similarly, the articular surfaces of the biological talus were meshed using a 4-node shell elements that was then extruded to create a layer of cartilage comprised of 4 layers of solid elements with a total thickness of 1 .5 mm in the biological models [23], For the implant models, the same procedures were repeated for the adjacent bones. However, the talus implants were meshed differently than that of the biological talus due to the fact that the cartilage layers do not exist on talus implants. The cartilage surfaces of the talus implants were meshed with 4-node shell elements and the non-cartilage surfaces were meshed with 3-node shell elements. After meshing, the tibia, fibula, calcaneus and navicular bones were moved away from the talus bone with a small displacement to clear the penetration of cartilage in order to form initial contact. Five pairs of potential contact areas were defined as surfaces first and then five pairs of “surface-to-surface” contacts were established between the talus and surrounding bones with a frictional coefficient 0.01 [10], The cortical bones and implants were modeled with linear elastic materials and the cartilage was modelled with hyperelastic materials (Table 1) [24-27], All the cortical bones were modelled as shell elements (1 mm) in both the biological and implant models and the talus implants were made out of metal (CoCr) (Figure 6).

Table 1. Element type and material properties of the FE model

C o m po ne nt s Elem ent ty e Mate ri al be h a vior Meeh an I c a 1 pro e rt les Refe ren ce

[00179] We assigned the most distal elements of adjacent bones with reference to the talus bone center as rigid bodies, and applied boundary conditions to the reference points of these rigid bodies. In order to mimic the realistic combined motions of the tibia and fibula, an equation constraint was added to both bones to allow them to move together along the Z-axis during loading (Figure 7). Next, the model was simulated in three consecutive static general time steps: (1) establishing initial contact by moving adjacent bones back to their CT scan position, and fixing the talus; (2) allowing the talus bone to freely adjust in all 6 degrees of freedom (dof), while adjacent bones remain in the same position as in step 1 ; (3) unconstraining the translational dof along the Z-axis of both the tibia and fibula bones, and unconstraining the rotational dof of the fibula bone along the Z-axis. Then, a compressive force of 2000N was applied to the reference point of the tibia bone, along the direction of gravity (i.e. Z-direction). After defining boundary conditions and loading scenarios as above, we analyzed the model in ABAQUS (Dassault Systemes®, Johnston, USA, Version 6.14).

[00180] 2.4 Model setup

[00181] In subject 1 , we used ten models for each posture (DF, NS, PF) for a total of 30 models (Figure 8). The 10 models consisted of one biological model serving as a reference, and nine implant models. As aforementioned, there were three implant types (customized, universal, and SSM), and each type contained three models: an original, offset, and scale model. After the ten models were simulated under three common postures (neutral standing; NS, 20° dorsiflexion; DF, 20° plantarflexion; PF), we evaluated thecontact characteristics of the fibular, tibial, calcaneal, and navicular cartilage. Contact characteristics are defined as a collection of contact pattern, contact pressure, and contact area between contacting surfaces of implants and cartilages. [00182] Considering the computational complexity involved in the production of 30 models, we aimed to reduce the number of models for subjects #2-4 by selecting one model with the best performance from each of the three implant types. Our criterion for selection was based on total contact area on the talus bone, and choosing the models with the largest contact area. Based on this, we selected the customized-scale, universalscale, and SSM-scale models for further analysis in subjects #2-4. We then used the same method as in subject 1 , simulating these three implant models in addition to a biological model in each posture (DF, NS, PF) for a total of 12 models, and recorded the contact characteristics for the fibular, tibial, calcaneal, and navicular cartilage.

[00183] 2.5 Convergence analysis

[00184] A mesh convergence study was performed using the biological model of subject 1 under dorsiflexion posture. The element size of the tibial cartilage size varied from 0.68 elements/mm² to 10.08 elements/ mm² while the rest of the cartilage maintained an element size of 10.08 elements/ mm². The boundary and loading conditions of the convergence analysis was exactly the same as the biological and implant models in the study.

[00185] 2.6 Model validation

[00186] In model validation, we isolated the tibia and talus bones from the biological model in neutral standing; this is because Anderson et al. (2007)’s experiments examined the tibia and talus bones of two subjects in neutral standing and measured the contact characteristics of the bones [22], As their work was on the left leg and our study on the right, we further mirrored the right leg cartilage after simulation so as to be comparable to Anderson et al. (2007). The same boundary conditions and magnitude of load were also applied to the model. The contact area and contact pattern were compared to that of Anderson et al. (2007) for purposes of validation [22], [00187] 3. Results

[00188] 3.1 Model validation

[00189] For model validation, we compared the contact pattern, contact area, peak contact pressure, and mean contact pressure (see Table 2) of our biological models with Anderson et al. (2017)’s experiments [22], The experimental data was measured through a calibrated pressure sensor that was inserted in the ankle joint and fixed through two 1 mm diameter stainless-steel K wires. Comparing our biological models of subjects 1-4 in NS to Anderson et al. (2017) [22], we observed a similar contact pattern, but our study had a lower contact pressure at the center of the tibial cartilage. The total contact area is the sum of two contact areas: the dome region and the medial malleoli region. However, Anderson et al. (2017) did not attempt to measure the contact stresses in the medial malleolus region which we reported here (Figure 9). Our model had a slightly larger contact area at the talus dome, ranging from 503.92 to 566.61 mm² when compared to Anderson et al. 's 493.6 mm² [23], Furthermore, the current study had a larger maximum contact pressure ranging from 4.49 MPa to 5.47 MPa, and a comparable mean contact stress from 1.09 MPa to 1.17 MPa. This is in comparison to Anderson et al.’s study, which had a maximum contact pressure varying from 2.92 MPa to 3.69 MPa, and a mean stress from 1.15 MPa to 1.96 MPa. The maximum contact pressure ranging from 4.49 MPa to 5.47 MPa is slightly larger than Anderson et al.,(2017)’s work but showed a good agreement with the experimental work (5.13±1 .16 MPa) of Wang et al. (1995) [28], Moreover, comparing the contact area within different contact pressure ranges in Anderson et al. (2017)’s experiements [22], our FE models showed a similar pattern and comparable area engagement within each range, with an exception of subject 2’s contact area at 0.5 MPa.

Table 2. Comparison of results between experimental work of Anderson et al. (2017} and numerical results from the current study

[00190] 3.2 Convergence analysis

[00191] The current study used the maximum contact pressure and total contact area in the tibial cartilage as metrics of convergence. When mesh size was increased by 35% (from 7.49 to 10.08 elements/mm²), the absolute change in the value of maximum contact pressure and contact area were around 0.15 MPa and 9.5 mm², respectivley; thus, the current study deemed a mesh size of 10.8 elements/mm² as adequate. The computational time (Dell Core(TM)i7-9700 CPU@3.00GHz and a RAM of 16 GB) varied from 8 hours (0.68 elements/mm²) to 18 hours (10.08 elements/mm²). The cartilage of all models in the study were meshed with an element size of 10.8 elements/mm2, with the biological models having around 119,880 elements in cartilage, and the implant models having around 91 ,000 elements in the cartilage.

[00192] 3.3 Contact stress distribution on the cartilage

[00193] For subject 1 , we first compared the peak contact pressures between the biological model and implant models in DF, NS, and PF. Our results showed that the biological model had the lowest peak pressure in the tibial, calcaneal, and navicular cartilages, ranging from 3.59 MPa to 18.61 MPa. Though when it came to fibular cartilage, the biological model had a greater peak contact pressure when compared to SSM-original in DF, customized-original in DF, universal-original in NS, SSM-original in NS, and all customized type models in NS (Figure 10). This is likely because that the horizonal displacements of the fibula and the tibia were based on CT scans, such that the horizontal distance between them remained the same for the entire simulation. In the biological model, the cartilage on the biological talus bone reduced the space between the fibula and the tibia, and therefore increased contact stress. In contrast, the lack of cartilage on the talus implant provided space for the cartilage on the fibula bone and decreased contact stress.

[00194] Next, we focused on the contact pressure areas of subject 1 ’s implant models, evaluating the pressure area of each individual cartilage in DF, NS, and PF (Figures 10b, 6d, and 6f). First, in DF, we found that less than 5% of the fibular, tibial, posterior calcaneal, and navicular cartilage contact areas had contact pressure of greater than 17.5 MPa in the implant models. A threshold of 17.5 MPa was used as ankle bone would fracture under a stress of 18 MPa [10], However, exceptions were observed in the middle calcaneal cartilage, whereby SSM and universal type models had approximately 9% of their contact areas at 17.5+ (>17.5) MPa. Second, in NS, we found that less than 5% of contact area had a contact pressure of more than 17.5 MPa in the fibular, tibial, calcaneal (posterior and middle), and navicular cartilage. It should be noted that there were multiple outliers. In the universal-offset, universal-scale and SSM-scale models, the navicular cartilage had approximately 6.8%, 8.3% and 7.2% contact area respectively; in the universal and SSM types, the posterior calcaneal cartilage had an approximate average of 12.3% and 7% of contact area with pressure greater than 17.5 MPa respectively. Lastly in PF, we found that similar to the previous postures, most cartilage surfaces had less than 5% of their contact areas at a pressure of more than 17.5 MPa, with some exceptions in the posterior calcaneal and navicular cartilage. In the posterior calcaneal cartilage, the SSM type had an approximate average of 7% of its contact area with a contact pressure of 17.5+ MPa. And in the navicular cartilage, the customized, universal and SSM types had an approximate average of 6%, 11%, and 10% of contact area greater than a pressure of 17.5 MPa respectively.

[00195] We also investigated cartilage contact areas of the biological and implant models in subject 1. While the biological model consistently had the largest contact area for all cartilage surfaces and postures, the contact area for the implant models were more varied. Instead of evaluating individual cartilage surfaces in each model, we combined the contact areas based on the cartilage type, model, and posture in order to produce a total contact area of the talus (Figure 11); for example, the total contact area for the customized-original model in DF would be the combined contact areas of the fibular, tibial, calcaneal, and navicular cartilage. Our findings showed that in all postures, the customized type had the largest combined contact area of all implant models; the SSM type had the second largest combined area; and lastly the universal type had the smallest combined area. In addition, we found that the scaled models (i.e. customized-scale, universal-scale, SSM-scale) typically had a larger contact area than the offset ones, leading us to attribute the larger area to the effects of scaling. Two exceptions to the case where the offset models had a larger contact area than scaled were found in the SSM- offset in NS and SSM-offset in PF. There were also two instances when the offset had a smaller contact area than the original, namely in the customized-offset in NS and customized-offset in PF. Taking into consideration the above findings, the current study focused on the three best-fitting implant models (as defined by largest combined contact area) for further investigation in order to reduce computational complexity. As such, the remaining three subjects were evaluated using the customized-scale, SSM-scale, and universal- scale models.

[00196] 3.4 Comparison of contact characteristics in subjects 1-4

[00197] In subject 1 , we simulated an original, scale, and offset model for each implant type (customized, universal, SSM) in DF, NS, and PF postures. As mentioned before, we focused on the three implant models with the best performance for further investigations in subjects #2-4; as such, in subjects 2-4, we simulated the scale models for each implant type in the three postures. We also simulated a biological model for each subject in those postures. In order to evaluate the contact characteristics of all four models, we took the biological and scaled models (i.e. customized-scale, universal-scale, SSM-scale) from all postures for subject 1 , and compared it to that of subjects 2-4. The contact pattern, peak contact pressure, and detailed percentage-area contact pressure were extracted from subjects 2-4 (Fig. 14-16).

[00198] In terms of peak contact pressure, when compared to their implant models, we found that subject 1-4’s biological models had a lower peak contact pressure in all cartilage surfaces, varying from 3.59 MPa to 22.59 MPa. The two exceptions to the case were in the tibial cartilage of subject #3 (30.03 MPa contact pressure) and subject #4 (39.24 MPa contact pressure). In contrast to the generally lower peak contact pressure of the biological models, the peak contact pressure in the scaled implant models varied more significantly, with a high of 158.5 MPa.

[00199] This study also compared the performance of customized, universal, and SSM scale models for subjects 1-4 in DF, NS, and PF postures, for a combined 36 models. We set a threshold of ‘5% of the cartilage contact area with a contact pressure of 17.5 MPa’. In this way, models with less than 5% of their cartilage contact area exceeding 17.5 MPa would be considered to have better performance than those with more than 5% contact area exceeding 17.5 MPa. We noticed that within some models, there may be multiple individual cartilage surfaces that met the established threshold; for example, the universal-scale model of subject 1 in NS had both its navicular and middle calcaneal cartilages exceed the threshold (Figure 10d). In order to capture the number of cartilage surfaces that meet the threshold in each of the 36 models, we tallied the number based on cartilage types, cutting across postures and subjects. To continue with the previous example using the universal-scale model of subject 1 in NS, the navicular and middle calcaneal cartilage both met the threshold, so one tally would be added to the bar graph under ‘navicular’ and one under ‘middle calcaneus’. Using this method, we found that the customized-scale had the least number of models where more than 5% of its contact area had a contact pressure of 17.5+ MPa, followed by universal-scale, and then SSM-scale. It should be noted that of the 12 SSM-scale implant models, five met the threshold and were tallied under the ‘post-calcaneus’ bar on the graph, and no tallies were added from either the customized or universal-scale models (Figure 12).

[00200] In addition, as there are five potential contact areas on a talus bone (at the tibial, fibular, posterior calcaneal, middle calcaneal, and navicular cartilage) and the contact area of each cartilage varies with posture and subject, the current study summed all the contact cartilage areas and used it as a measure to evaluate the appropriateness of the talus implant (Figure 13). Our findings showed that the biological talus had the largest total contact area among all the subjects and postures, which was on average 588±127.6 mm² larger than the customized-scale implant models. We also observed that the differences in contact area among the implant models were relatively smaller (Figure 13). The average contact area of SSM-scale implant models was around 239±271.2 mm² smaller than that of the customized-scale models, and was around 25±155.5 mm² larger than that of the universal-scale models.

[00201] 4 Discussion

[00202] Using FE modelling, the current study investigated the contact pressure and distribution and the contact area generated around the talus joint under axial loading using customized, universal, and SSM implant designs in three common postures (DF, NS, PF). This study also used two methods, scale and offset, to adjust the talus implant size in order to account for cartilage thickness and evaluate their influences on the contact characteristics of cartilage surfaces. Furthermore, we evaluated the impacts that three different CoCr implants had on contact pressure values and distribution on cartilage surfaces. The results of this study can contribute to knowledge of the mechanical impacts of using metal talus implants on adjacent cartilage, so as to provide information on cartilage damage- patterns and propensities. This information can also be used to inform the design of talus implants to produce a better fit for patients.

[00203] Model Validation

[00204] The predicted contact stress distribution in the validated models of subjects 1-4 were continuous and relatively uniform in contact patches (Figure 9). The validity of our FE models was tested by comparing contact pattern, contact area, peak contact pressure, and mean contact pressure with the experimental results [22], We found an overall comparable contact pattern between Anderson et al. (2017) [ 22] and the current work in talus dome region, but a relatively lower contact pressure in the middle tibial cartilage and an additional area of contact pressure in the medial malleoli region of the tibial cartilage in the current work (Figure 9). It is likely that Anderson et al. (2017) did not have an area of contact pressure in the medial malleoli region because sensors were not placed on the side of the tibial cartilage in their study [23], The contact on the side of the tibia cartilage likely also relieved the contact pressure in the middle tibial cartilage region. In addition, we found a slightly higher peak contact pressure and contact area in our FE study as compared to that of Anderson et al., (2007) [22], The higher peak contact pressure and area can likely be attributed to individual variability, as the studies used different subjects who might have had natural variations in talus joint geometries. The difference in the peak contact pressure could also be the result of placing the sensor into the joint space. A stiff sensor can lead to 10-26% peak pressure error [29], and pressure truncation from the sensor can result in errors in peak pressure and 5-10% errors in contact force [30], Additionally, the discretization of the three-dimensional surfaces in the finite element model could lead to excessive stress concentrations caused by element discretization resulting in excessively higher peak stresses. Thus, we used the general contact area trend within each contact pressure range, and compared it to Anderson et al. (2017) [22], We found that the trend and the average stresses were in good agreement with the experimental data, thereby lending support to the validity of our FE models. [00205] Methods to consider the cartilage thickness in implants [00206] When it comes to the design of customized talus implants, existing studies have used the exact shape and size of the biological talus from patients, but the thickness of biological talus cartilage has not been consistently taken into consideration [2, 3, 7]. Our study considered adjustment to the talus implant to account for cartilage thickness in subject 1 using two methods: offset and scaling. The offset method adds a uniform thickness (0.5 mm) to the implant, but may produce slight alterations to the shape of the original talus, which is not entirely covered with cartilage. In contrast, a scaling method can maintain the bony shape of the implant, but there could be variations in deviation between the original and scaled models (Figure 5). Our findings showed that neither offset nor scaling as a means of accounting for cartilage thickness affected the contact pattern on cartilage surfaces. However, the contact areas for some cartilage surfaces were increased when compared to implants that did not account for cartilage thickness; this effect was comparatively more pronounced in the scaled models (Figure 10d). A potential explanation for the increased contact areas is because when cartilage thickness is added to a bone, the implant volume increases and can push on the space formed by the implant and adjacent bones, thereby creating more contact areas between the two surfaces. An alternate viewpoint could be that an implant that only considered the shape and size of the original bone excluding cartilage is slightly too small, precipitating in smaller contact area. When we investigated the effect of scale versus offset in subject 1 , we found that scaling is better able to account for cartilage thickness with respect to the total contact area. Thus, we focused on the scaling method in subjects 2-4 to reduce computational complexity.

[00207] Contact characteristics of three implants

[00208] This study applied an axial load of 2000N on the tibia in FE, an amount around three to four times a person’s body weight and the load amount that has consistently been used in literature [14, 23, 31 , 32], Consistent with experimental findings using cadaveric ankle joints [10, 14, 31 , 32], we found that peak contact pressure in the tibial cartilage of biological models varied from 7.21 MPa to 13.74 MPa amongst subjects 1-4. However, three outliers exist: the tibial cartilage contact stress of subject 3 in DF (30.03MPa), subject 4 in DF (35.46 MPa), and subject 4 in PF (39.24 MPa). These significantly higher contact pressures only occurred in a small region, and may in part be due to spikes on the surface of the bone from the process of recreating geometries from CT scan images. As opposed to smoothing these spikes, we used minimal degrees of smoothing in the process of preparing geometries from CT scans, operating from a conservative perspective of trying to preserve the at times incongruous surface geometries [10], In order to limit the effects that these incongruous surfaces from the smoothing process may have, our study considered the histograms of surface stresses when comparing different models, rather than focusing on maximum contact stresses which could be artefacts.

[00209] We observed that extremely high peak contact pressure was found in all three implant types, and could reach a high of 158.5 MPa, as was the case of the calcaneal cartilage in SSM-scale in subject 4 under DF. Area percentage histograms (FIGS. 14-15) showed that implant models, regardless of subject and posture, had a larger percent of cartilage with pressure greater than 17.5 MPa than the biological models. A greater percentage of high-pressured contact areas may be a shared limitation of metal-based talus implants, which is likely due to a combination of the stiffness of metal talus implants, and the inadequate cartilage thickness between the articular surfaces of adjacent bones and metal-based implants, together resulting in stress concentration in certain areas of the cartilage. This further stresses the importance of a cartilage layer on limiting contact pressure. As such, future talus implant designs can look at incorporating a more cartilage-like surface layer on top of implants to avoid extreme high peak contact pressures. In alignment with the literature, it has been shown that hemiarthroplasty implants with a very low stiffness can lead to pronounced improvements in cartilage contract stress and protect adjacent cartilages [33],

[00210] A pressure of around 18.0 MPa has been shown in the literature to be a threshold contact pressure, past which bones at the ankle joint could fracture [10]; as such, our study adopted 17.5 MPa as our reference for evaluation of implant performance. We added a further threshold as to contact area, assuming that when a cartilage surface has more than 5% of its contact area over a pressure of 17.5 MPa that it may result in fractures in natural bones. It should be noted that a review of literature did not find a convention pertaining to percent of contact area, but it was imperative that a threshold be established for the purpose of comparing implant performance. As such, the 5% contact area threshold should be interpreted as relative and not absolute, i.e. not all cartilage surfaces that have more than 5% of its contact area with a contact pressure of 17.5+ MPa will result in fractures. Instead, it more offers a relative value to compare and rank the implants, in that even if the contact area threshold was changed to another similar value, the ranking of implant performance would remain similar.

[00211] The current study selected solid elements for all cartilage tissues because it focused on the contact characteristics in the cartilage, and the model involved contact and large deformation of the cartilage in the simulation. The cortical bone and implants were modelled as shell elements in order to reduce the computational complexity. This combination will not only maintain contact interaction accuracy and stability, but also allow flexibility in the cortical bones and simplify the simulation. The cartilage element type used in this study was also used in the works of Anderson et al. (2006, 2007) [10, 23],

[00212] 5. Conclusion

[00213] The current study evaluated three different types of CoCr talus implants (customized, universal, and SSM) and quantified the resulting contact characteristics when the implants load against the adjacent articular cartilages, namely the fibular, tibial, calcaneal, and navicular surfaces. Contact characteristic effects were investigated using FE models in DF, NS, and PF postures for four subjects. Our findings showed that not only did the biological talus model have the largest contact area, but also the least amount of peak pressure. In contrast, the three implant type models had smaller contact areas and significantly larger peak contact pressures. Of the three implant types, customized-scale appeared to be better than SSM-scale and universal-scale implants with respect to total contact area. However, the variation in contact areas between customized-scale and universal-scale/SSM-scale implants were minimal, so that the actual differences in their performance may be limited. Further investigation should focus on quantifying the impact of this variation among the talus implant types and improving the performance of metal implants. The results of this study can contribute to the developing understanding of the mechanical environment of talus implants and the natural talus. This knowledge can not only provide insight into ankle joint modelling, but also improve implant design and subsequently post-implant surgery recovery outcomes. [00214] References

[00215] 1. Pearce DH, Mongiardi CN, Fornasier VL, Daniels TR (2005)

Avascular Necrosis of the Talus: A Pictorial Essay. RadioGraphics 25:399-410. https://doi.org/10.1148/rg.252045709

[00216] 2. Katsui R, Takakura Y, Taniguchi A, Tanaka Y (2019) Ceramic

Artificial Talus as the Initial Treatment for Comminuted Talar Fractures [Formula: see text]. Foot Ankle Int 1071100719875723. https://doi.org/10.1177/1071100719875723 [00217] 3. Shnol H, LaPorta GA (2018) 3D Printed Total Talar Replacement:

A Promising Treatment Option for Advanced Arthritis, Avascular Osteonecrosis, and Osteomyelitis of the Ankle. Clinics in Podiatric Medicine and Surgery 35:403-422. https://doi.Org/10.1016/j.cpm.2O18.06.002

[00218] 4. Liu T, Jomha NM, Adeeb S, et al (2020) Investigation of the

Average Shape and Principal Variations of the Human Talus Bone Using Statistic Shape Model. Front Bioeng Biotechnol 8:. https://doi.org/10.3389/fbioe.2020.00656

[00219] 5. Fortin PT, Balazsy JE (2001) Talus Fractures: Evaluation and

Treatment. Journal of the American Academy of Orthopaedic Surgeons 9:14

[00220] 6. Imam MA, Matthana A, Kim JW, Nabil M (2017) A 24-Month

Follow-Up of a Custom-Made Suture-Button Assembly for Syndesmotic Injuries of the Ankle. The Journal of Foot and Ankle Surgery 56:744-747. https://doi.Org/10.1053/j.jfas.2017.02.010

[00221] 7. Tonogai I, Hamada D, Yamasaki Y, et al (2017) Custom-Made

Alumina Ceramic Total Talar Prosthesis for Idiopathic Aseptic Necrosis of the Talus: Report of Two Cases. In: Case Reports in Orthopedics. https://www.hindawi.com/journals/crior/2017/8290804/. Accessed 26 Dec 2019

[00222] 8. Islam K, Dobbe A, Duke K, et al (2014) Three-dimensional geometric analysis of the talus for designing talar prosthetics. Proc Inst Meeh Eng H 228:371-378. https://doi.Org/10.1177/0954411914527741

[00223] 9. Trovato A, El-Rich M, Adeeb S, et al (2017) Geometric analysis of the talus and development of a generic talar prosthetic. Foot and Ankle Surgery 23:89- 94. https://doi.Org/10.1016/j.fas.2016.12.002

[00224] 10. Anderson DD, Goldsworthy JK, Shivanna K, et al (2006) Intraarticular contact stress distributions at the ankle throughout stance phase-patient-specific finite element analysis as a metric of degeneration propensity. Biomech Model Mechanobiol 5:82-89. https://doi.org/10.1007/s10237-006-0025-2 [00225] 11. Kimizuka M, Kurosawa H, Fukubayashi T (1980) Load-bearing pattern of the ankle joint. Arch Orth Traum Surg 96:45-49. https://doi.org/10.1007/BF01246141

[00226] 12. Millington S, Grabner M, Wozelka R, et al (2007) A stereophotographic study of ankle joint contact area. Journal of Orthopaedic Research 25: 1465-1473. https://doi.org/10.1002/jor.20425

[00227] 13. Macko VW, Matthews LS, Zwirkoski P, Goldstein SA (1991) The joint-contact area of the ankle. The contribution of the posterior malleolus. J Bone Joint Surg Am 73:347-351

[00228] 14. Brown TD, Hurlbut PT, Hale JE, et al (1994) Effects of imposed hindfoot constraint on ankle contact mechanics for displaced lateral malleolar fractures. J Orthop Trauma 8:511-519

[00229] 15. Stevens BW, Dolan CM, Anderson JG, Bukrey CD (2007) Custom talar prosthesis after open talar extrusion in a pediatric patient. Foot Ankle Int 28:933- 938. https://doi.Org/10.3113/FAI.2007.0933

[00230] 16. Wagener J, Gross CE, Schweizer C, et al (2017) Custom-made total ankle arthroplasty for the salvage of major talar bone loss. Bone Joint J 99-B:231- 236. https://doi.org/10.1302/0301-620X.99B2.BJJ-2016-0504.R2

[00231] 17. Anderson DD, Tochigi Y, Rudert MJ, et al (2010) Effect of implantation accuracy on ankle contact mechanics with a metallic focal resurfacing implant. J Bone Joint Surg Am 92:1490-1500. https://doi.org/10.2106/JBJS.L00431

[00232] 18. Reggiani B, Leardini A, Corazza F, Taylor M (2006) Finite element analysis of a total ankle replacement during the stance phase of gait. Journal of Biomechanics 39:1435-1443. https://doi.Org/10.1016/j.jbiomech.2005.04.010

[00233] 19. Espinosa N, Walti M, Favre P, Snedeker JG (2010) Misalignment of total ankle components can induce high joint contact pressures. J Bone Joint Surg Am 92: 1179-1187. https://doi.Org/10.2106/JBJS.1.00287

[00234] 20. Terrier A, Larrea X, Guerdat J, Crevoisier X (2014) Development and experimental validation of a finite element model of total ankle replacement. Journal of Biomechanics 47:742-745. https://doi.Org/10.1016/j.jbiomech.2013.12.022

[00235] 21. Trovato AN, Bornes TD, El-Rich M, et al (2018) Analysis of a generic talar prosthetic with a biological talus: A cadaver study. Journal of Orthopaedics 15:230-235. https://doi.Org/10.1016/j.jor.2018.01 .015 [00236] 22. Liu T, Ead M, Cruz SDV, et al (2022) Polycarbonate-urethane coating can significantly improve talus implant contact characteristics. Journal of the Mechanical Behavior of Biomedical Materials 125:104936. https://doi.Org/10.1016/j.jmbbm.2021.104936

[00237] 23. Anderson DD, Goldsworthy JK, Li W, et al (2007) Physical validation of a patient-specific contact finite element model of the ankle. Journal of Biomechanics 40: 1662-1669. https://doi.Org/10.1016/j.jbiomech.2007.01 .024

[00238] 24. Akrami M, Qian Z, Zou Z, et al (2018) Subject-specific finite element modelling of the human foot complex during walking: sensitivity analysis of material properties, boundary and loading conditions. Biomech Model Mechanobiol 17:559-576. https://doi.Org/10.1007/sl 0237-017-0978-3

[00239] 25. Al Jabbari YS (2014) Physico-mechanical properties and prosthodontic applications of Co-Cr dental alloys: a review of the literature. J Adv Prosthodont 6:138-145. https://doi.Org/10.4047/jap.2014.6.2.138

[00240] 26. Mondal S, Ghosh R (2019) Effects of implant orientation and implant material on tibia bone strain, implant-bone micromotion, contact pressure, and wear depth due to total ankle replacement. Proc Inst Meeh Eng H 233:318-331 . https://doi.Org/10.1177/0954411918823811

[00241] 27. Robinson DL, Kersh ME, Walsh NO, et al (2016) Mechanical properties of normal and osteoarthritic human articular cartilage. J Meeh Behav Biomed Mater 61 :96-109. https://doi.Org/10.1016/j.jmbbm.2016.01.015

[00242] 28. Wang CL, Cheng CK, Chen CW, et al (1995) Contact areas and pressure distributions in the subtalar joint. J Biomech 28:269-279. https://doi.Org/10.1016/0021 -9290(94)00076-g

[00243] 29. Wu JZ, Herzog W, Epstein M (1998) Effects of inserting a pressensor film into articular joints on the actual contact mechanics. J Biomech Eng 120:655-659. https://doi.Org/10.1115/1 .2834758

[00244] 30. Fregly BJ, Sawyer WG (2003) Estimation of discretization errors in contact pressure measurements. J Biomech 36:609-613. https://doi.org/10.1016/s0021- 9290(02)00436-0

[00245] 31 . Steffensmeier SJ, Saltzman CL, Berbaum KS, Brown TD (1996)

Effects of medial and lateral displacement calcaneal osteotomies on tibiotalar joint contact stresses. J Orthop Res 14:980-985. https://doi.org/10.1002/jor.1100140619 [00246] 32. Vrahas M, Fu F, Veenis B (1994) Intraarticular contact stresses with simulated ankle malunions. J Orthop Trauma 8:159-166. https://doi.Org/10.1097/00005131-199404000-00014

[00247] 33. Berkmortel C, Langohr GDG, King G, Johnson J (2020)

Hemiarthroplasty implants should have very low stiffness to optimize cartilage contact stress. J Orthop Res 38:1719-1726. https://doi.org/10.1002/jor.24610

[00248] 34. Bahr R, Pena F, Shine J, et al (1998) Ligament force and joint motion in the intact ankle: a cadaveric study. Knee Surg Sports Traumatol Arthrosc 6:115-121. https://doi.Org/10.1007/S001670050083

[00249] EXAMPLE 3

[00250] The talus bone in the human ankle joint has a limited blood supply, making it vulnerable to osteonecrosis when injured or fractured which can lead to severe ankle pain ¹. Talar replacement is a surgical procedure where the talus bone is replaced with an artificial prosthesis which can preserve ankle functionality and reduce pain ². Metal-based implants are commonly used as they demonstrate good biocompatibility. However, they may lead to joint degeneration due to excessive wear ³.

[00251] To represent a better physiological joint environment, researchers have utilized polycarbonate-urethane (PCU) as an artificial material that can function as cartilage in the knee and hip joints and have reported good outcomes ^4,5 Utilizing PCU in the anklejoint has not yet been investigated and therefore, the current study aims to explore this. Coating talus implants with a layer of PCU to act as cartilage can potentially reduce the likelihood of wear in the ankle joint. The first phase of this project involves designing an interface for the PCU to attach to the implant.

[00252] Methods

[00253] A lattice structural interface with two different designs and two cell sizes was modelled using nTopology® software. These four lattice models were metal 3D printed in the shape of disks with a 4 mm solid stainless steel layer and a 2 mm lattice stainless steel layer (Figure 17). Since injection molding will be used to add the PCU layer, a custom mold was designed to hold the lattice disks and inject PCU (Bionate® II 80A - DSM Biomedical, Pennsylvania, USA) on top such that it creates a 2 mm layer of solid PCU (Figure 17). The hypothesis is that the high pressure will force the PCU to flow through the lattice and act as a bonding layerbetween the disk (implant) and the layer of solid PCU used as a cartilage substitute. In order to quantifiably compare the four lattice disks as bonding interfaces for the PCU layer, the volume of PCU that flowed through the lattices will be determined by scanning the disks with a 3D X-ray microscope. The 3D models produced will then be exported to an image processing software (ORS Dragonfly®) where the volume of voids and PCU can be extracted.

[00254] Results

[00255] The adherence of PCU to the samples was manually inspected by attempting to detach the PCU layer and it appeared to be strongly bonded to the metal. Preliminary scans were performed on a sample plastic lattice disk coated with PCU. Results indicated that the2 mm lattice layer embedded with PCU consisted of: 26% plastic lattice, 28% voids, and 46% PCU.

[00256] Conclusions

[00257] A smaller volume of voids in the lattice layer is more desirable as the solid PCU layer will more likely adhere to the implant. The next stage of the project will involve wear testing the PCU against articular cartilage.

[00258] References

[00259] 1 . Delanois, R E et al., J Bone Jt Surg, 80:529- 36, 1998.

[00260] 2. Taniguchi, A et al., J Bone Jt Surg - Am Vol, 97:1348-1353, 2015.

[00261] 3. Hussein, M A et al., Materialsvol. 82749- 2768, 2015.

[00262] 4. Alley, R et al., Orthop J Sport Med, 8:2020. [5] Lazic, S et al., HIP Int,

30:303-308, 2020.

[00263] The embodiments described herein are intended to be examples only. Alterations, modifications and variations can be effected to the particular embodiments by those of skill in the art. The scope of the claims should not be limited by the particular embodiments set forth herein, but should be construed in a manner consistent with the specification as a whole.

[00264] All publications, patents and patent applications mentioned in this Specification are indicative of the level of skill those skilled in the art to which this invention pertains and are herein incorporated by reference to the same extent as if each individual publication patent, or patent application was specifically and individually indicated to be incorporated by reference.

[00265] The invention being thus described, it will be obvious that the same may be varied in many ways. Such variations are not to be regarded as a departure from the spirit and scope of the invention, and all such modification as would be obvious to one skilled in the art are intended to be included within the scope of the following claims.

Claims

WHAT IS CLAIMED IS:

1. A talus implant comprising: a body section; a neck section; and a head, the talus implant having an outer surface comprising a coating comprising a polycarbonate-urethane (PCU), a high density polyethylene and poly(aryl-ether-ether- ketone) (PEEK), poly(aryl-ether-ketone-ether-ketoneketone (PEKEKK).

2. The talus implant of claim 1 , wherein at least a portion of the outer surface comprises a lattice.

3. The talus implant of claim 1 or 2, wherein the coating has a thickness of between about 0.5 to about 20 mm.

4. The talus implant of claim 1 or 2, wherein the coating has a thickness of between about 0.5 mm to about 6 mm.

5. The talus implant of claim 1 or 2, wherein the coating has a thickness of between about 1mm to about 3 mm.

6. The talus implant of claim 1 or 2, wherein the coating has a thickness of between about 1 .5 mm to about 2 mm.

7. The talus implant of any one of claims 1 to 6, where the talus implant comprises titanium, ceramic, or CoCr.

8. The talus implant of any one of claims 1 to 7, having a shape which is customized.

9. The talus implant of any one of claims 1 to 7, having a shape which is universal.

10. The talus implant of claim 9, wherein the shape is selected using a cartilageexclusive sizing guide.

11 . The talus implant of any one of claims 1 to 7, having a shape which is a statistical shape model (SSM).

12. The talus implant of claim 11 , wherein the shape is selected using a cartilageexclusive sizing guide.

13. The talus implant of any one of claims 1 to 12, wherein the talus implant comprises titanium, ceramic, or cobalt-chrome (CoCr).

14. The talus implant of any one of claims 1 to 11 , having average peak contact pressures that are similar to that of a biological implant.

15. The talus implant of any one of claims 1 to 11 , having similar contact areas that are similar to that of a biological implant.

16. A method of reconstructing a subjects ankle comprising: implanting a talus implant of any one of claims 1 to 15.

17. The method of claim 16, wherein the patient is a human.

18. An artificial implant comprising: a body, at least a portion said body having an outer surface comprising a coating comprising a polycarbonate-urethane (PCU), a high density polyethylene and poly(aryl- ether-ether-ketone) (PEEK), poly(aryl-ether-ketone-ether-ketoneketone (PEKEKK).

19. The artificial implant of claim 18, wherein the artificial implant is a humeral head implant, a radial head implant, a distal humerus implant, a-scaphoid implant, a lunate implant, femoral head implant, a patella implant, a talus implant, a metatarsal head implant, a phalanx implant, or a plug implant.

20. The artificial implant of claim 19, wherein the plug implant comprises metal, titanium, ceramic, or CoCr.

21 . A method of implanting an artificial implant in a subject, comprising: providing an artificial implant of any one of claims 18 to 20, and implanting the artificial implant in the subject

22. The method of claim 21 , wherein the subject is a human.