Evaluation of the Stress-Shielding Effect of a PEEK Knee Prosthesis. A Finite Element Study

Abstract

Background: The long-term success of total knee arthroplasty (TKA) is often compromised by stress shielding, which can lead to bone resorption and even implant loosening. This study employs finite element analysis (FEA) to compare the stress-shielding effects of a knee prosthesis made from polyether ether ketone (PEEK) with a traditional titanium Ti6Al4V implant on an osteoporotic tibial bone model. Methods: Stress distribution and the stress-shielding factor (SSF) were evaluated at seven critical points in the proximal tibia under physiological loading conditions. Results: Results indicate that the PEEK prosthesis yields a more uniform stress transmission, with von Mises stress levels within the optimal 2–3 MPa range for bone maintenance and consistently negative or near-zero SSF values, implying minimal stress shielding. Conversely, titanium implants exhibited significant stress shielding with high positive SSF values across all points. Additionally, stress concentrations on the polyethylene liner were lower and more evenly distributed in the PEEK model, suggesting reduced wear potential. Conclusions: These findings highlight the biomechanical advantages of PEEK in reducing stress shielding and preserving bone integrity, supporting its potential use to improve implant longevity in TKA. Further experimental and clinical validation are warranted.

1. Introduction

Total knee arthroplasty (TKA) is currently the most effective treatment for advanced knee osteoarthritis [1,2]. Among the various surgical techniques, kinematic alignment has shown potential biomechanical benefits. However, recent evidence suggests that these advantages do not always translate into clinically significant improvements in functional outcomes [3,4,5].

Over the past decade, the number of TKA procedures has increased substantially and is expected to exceed 3 million per year in the United States by 2030 [6]. Parallel to this trend, there has also been an increase in the rate of revision surgeries, primarily due to prosthetic failure. Several studies have investigated factors that contribute to such failures, including both pre- and postoperative conditions as well as implant design [7,8,9,10]. The main causes of revision include infection, instability, aseptic loosening, polyethylene wear, stiffness, and periprosthetic fractures [11,12,13]. Additionally, patient-related factors such as sex, age, body mass index (BMI) [14], comorbidities, socioeconomic status, and surgical variables like implant alignment and positioning play a critical role in the success of TKA [15,16,17].

One of the most relevant biomechanical challenges in TKA is stress shielding, a phenomenon that can significantly compromise the longevity of the implant and the quality of the surrounding bone tissue [18,19,20,21]. Stress shielding arises due to the high Young’s modulus of the metallic components, which alters the physiological distribution of mechanical loads. This leads to reduced mechanical stimulation of the peri-prosthetic bone compared to healthy conditions [22,23].

According to Wolff’s Law, bone remodels in response to mechanical stress. Therefore, the reduction of stress caused by stress shielding induces local bone resorption and structural weakening [24,25,26]. In the case of the tibial component, which is typically rigid, a large portion of the load is transferred through the implant, bypassing adjacent bone regions—particularly the proximal tibial metaphysis. This load imbalance can lead to periprosthetic osteoporosis, compromising both the primary and long-term stability of the implant, especially in osteoporotic patients [27,28,29].

Stress shielding may also result in morphological changes in the tibia and femur, complicating future revision surgeries due to the reduced quality and quantity of residual bone [30]. Several factors influence the extent of stress shielding, including the material and geometry of the prosthetic components, the method of fixation (cemented or uncemented), and the accuracy of their alignment [31]. For instance, tibial components with high Young’s modulus may exacerbate load deviation, while misalignment may increase localized underloading or overloading, further worsening the effects of stress shielding [32,33].

To address these issues, polyether ether ketone (PEEK) has recently emerged as a promising alternative material for knee prostheses. Thanks to its biomechanical properties, which more closely resemble those of natural bone, PEEK offers a potential solution to reduce stress shielding [34]. Its lower elastic modulus compared to metals such as cobalt-chromium or titanium enables better load transmission to the surrounding bone and helps preserve bone density, promoting long-term implant stability [35,36,37,38].

Biomechanical studies and finite element simulations have demonstrated that PEEK components reduce the overall stiffness of the bone–implant system, decreasing abnormal stress concentrations beneath the implant [39,40,41,42]. For example, Yu et al. [43] observed a significant reduction in stress shielding when using PEEK tibial components compared to traditional metallic ones. Similarly, Zhang et al. [36], in a cadaveric study, reported improved preservation of bone microarchitecture. Zhao et al. [44] confirmed the clinical viability of PEEK prostheses, showing comparable short-term outcomes to those of cobalt-chromium alloys.

The finite element method (FEM) is widely used to assess stress shielding in preclinical stages by simulating biomechanical responses under physiological loading conditions [18,45]. Wilczyński et al. [46] highlighted how using elastic materials and uncemented fixation in FEM models can promote more physiological load transfer and protect peri-prosthetic bone.

Despite growing interest in PEEK, the literature lacks well-validated FEM studies specifically focused on modeling its mechanical interaction with osteoporotic bone in TKA. To fill this gap, it is necessary to combine numerical simulations with experimental data to evaluate the actual effectiveness of PEEK in reducing stress shielding and enhancing implant performance.

The objective of this study is to numerically investigate the effects of using PEEK as a material for the tibial component in total knee arthroplasty. Specifically, we aim to evaluate the extent to which PEEK can reduce stress shielding compared to traditional metallic materials in osteoporotic bone conditions. To this end, a finite element model of the proximal tibia with an implanted prosthesis was developed. The model includes both metallic (Ti6Al4V) and polymeric (PEEK) tibial components, and simulations were carried out under physiological loading conditions. Stress and strain distributions within the periprosthetic bone were then analyzed to quantify the impact of different material properties on stress shielding.

2. Materials and Methods

The study focuses on comparing a knee prosthetic component made of titanium Ti6Al4V with one made of PEEK. To evaluate the influence of these materials on the transmission of stress to the proximal part of the tibia, a finite element model was developed based on previously validated works [46,47,48,49,50]. The model analyzes bone stress after implantation and loading of the Ti6Al4V component compared to the PEEK component. Hypothetically, the PEEK component can transfer more load to the bone, resulting in higher bone stresses due to the less stiff mechanical properties of PEEK compared to titanium (lower Young’s modulus).

2.1. Geometries

The study examined two femoral prostheses with fixed tibial platforms, identical in design, but differing in materials and manufacturing technologies. The experimental device includes a femoral and tibial component, both entirely made with 3D printing in PEEK and using titanium alloy casting, respectively, along with a polymer insert positioned between the two components (see Figure 1).

The simulation of total knee arthroplasty (TKA) was performed based on the anatomical dimensions of the tibial bone of a male individual, taking as reference the study conducted by Carman et al. [45].

The average cortical bone thickness was 3 mm. A simplification was made regarding the 3D model of the tibia by referencing the study conducted by Wliczyński et al. [46], which allowed for a reduced complexity of the tibial geometry while preserving essential anatomical features relevant for the simulation. All 3D models were created using the Autodesk Inventor 2024 software Figure 2.

2.2. Materials

Bone is generally composed of cortical bone on the outer part and trabecular bone on the inner part [50], both characterized by anisotropic behavior. By definition, a material with anisotropic behavior has 21 elastic constants describing its elastic properties. However, it has been demonstrated that bone exhibits transverse isotropy in the YZ and XZ planes Figure 2. Consequently, it can be modeled as an orthotropic material with 9 elastic constants. To model osteoporotic bone, the Young’s modulus of cortical bone was reduced by 32%, and that of trabecular bone was reduced by 66%. According to the study by Galas et al. [51], the mechanical properties of bone considered are reported in

Table 1. Mechanical properties of osteoporotic bone; the subscripts x and y refer to the orthogonal directions of the tibia, while the subscript z represents the direction parallel to the length of the tibia.

Material Properties	Cortical Osteoporotic Bone	Cancellous Osteoporotic Bone
Young’s modulus (MPa)	Ex = 7820	Ex = 6252
	Ey = 7820	Ey = 2.380
	Ez = 11,560	Ez = 2.890
Poisson’s ratio	${\nu }_{xy}=0.12$	${\nu }_{xz}=0.3$
	${\nu }_{yz}=0.37$	${\nu }_{yz}=0.27$
	${\nu }_{xz}=0.14$	${\nu }_{xz}=0.19$

[50,51,52].

In the case of prosthetic materials of the tibial component, the insert, femoral component, isotropic and linearly elastic mechanical properties have been considered, as reported in

Table 2. Mechanical properties of prosthetic materials.

Material Properties	Ti6Al4V	PEEK	UHMWPE
Young’s modulus (MPa)	E = 110,000	E = 18,000	E = 920
Poisson’s ratio	ν = 0.3	ν = 0.3	ν = 0.3

[49,50,51,52].

2.3. Finite Element Model

Two finite element models were created using Ansys Workbench R 2023 (Southpointe 2600 Ansys Drive, Canonsburg, PA 15317, USA) software, one for each material of the tibial and femoral components. A mesh sensitivity analysis was performed with the aim of defining the optimal element size in terms of computational cost and result accuracy. Meshes with element sizes ranging from 1.8 mm to 3 mm were tested. The main results of the analysis are summarized in the following

Table 3. Sensitivity analysis of the mesh.

Mesh Element Size (mm)	von Mises Stress	Computation Time (min)
3.0	15 MPa	5
2.5	18 MPa	8
2.0	22 MPa	14
1.8	22.3 MPa	22

:

As can be observed, the variation in results between the 2 mm mesh and the 1.8 mm mesh is negligible (<0.2%), while the computational time increases significantly. Therefore, a mesh with an element size of 2 mm was chosen as the optimal compromise between accuracy and computational cost. The total number of elements was 36,326 for the tibial component, 45,369 for the insert, and 231,123 for the tibial bone (Figure 3).

2.4. Contact, Loading, and Boundary Conditions

The interactions between the implant, bone, and implant components were modeled using a “surface–surface” contact formulation, applying specific friction coefficients based on the materials involved: 0.2 for Ti6Al4V-UHMWPE and 0.25 for PEEK-UHMWPE [51]. A fixed contact type was used between the bone and the tibial component through the sharing of contact elements. Subsequently, a vertical compressive load of 1150 N (along the Z axis, approximately twice the body weight; Figure 4) was applied to the midpoint of the transepicondylar axis of the femur to simulate the load during the gait cycle with the knee in a straight position at 0° of flexion [51]. This load condition represents the most severe condition a patient who has undergone TKA could encounter.

Meanwhile, the lower part of the tibia was constrained in all directions.

To evaluate the phenomenon of stress shielding, we analyzed the contact stress distribution between the tibial component and the surrounding bone using the von Mises criterion. This stress measure, derived from plasticity theory, provides a scalar value that integrates the principal stresses into an equivalent state of stress. It is particularly useful for assessing complex, multidirectional loading patterns acting on bone tissue, which is inherently anisotropic and heterogeneous. As established in the literature, stress shielding occurs when the mechanical stimuli applied to the bone fall below a critical threshold, leading to bone resorption. This bone loss compromises implant fixation, reduces mechanical stability, and increases the risk of early or late loosening. In addition, it may complicate revision surgeries due to the diminished quality and quantity of residual bone. Numerous studies indicate that a threshold stress of approximately 2 MPa is critical: below this value, bone tends to undergo resorption, while stresses between 2 and 5 MPa are generally considered necessary to maintain bone mass—often referred to as maintenance stress [50,53,54,55,56,57]. Therefore, in this study, von Mises stress values in the peri-prosthetic bone were interpreted with respect to these thresholds to evaluate the potential of alternative materials to mitigate stress-shielding effects [58,59,60,61]. Furthermore, to quantify stress shielding, we calculated the stress-shielding factor (SSF) using Equation (1), which compares the stress in the peri-prosthetic bone with and without the implant. This approach follows the methodology proposed by Ceddia et al. [62] whose FEA-based procedure has been validated for simulating orthopedic implant–bone interactions. By applying this method to prosthetic components made from materials with different elastic moduli, we were able to assess the degree of mechanical stimulus reduction and its correlation with the stiffness of the tibial component.

$$Stress shielding factor={\displaystyle \frac{{\sigma }_{bone}^{ref}-{\sigma }_{bone}^{with implant}}{{\sigma }_{bone}^{ref}}}$$

(1)

3. Results

3.1. Stress Analysis on the Insert

The finite element analysis allowed the evaluation of the von Mises stress on the upper surface of the polyethylene liner in model a, with femoral and tibial components made of titanium, and in model b, with both femoral and tibial components made of PEEK (Figure 5). It can be observed that the highest stress value was 16.324 MPa for model a and 7.412 MPa for model b.

The reduction of contact stress on the polyethylene liner surface is essential to limit material wear and extend the lifespan of the prosthetic implant. High pressure peaks cause localized stress concentrations, leading to plastic deformations and micro-damage, which promote the formation of micro-cracks and the release of wear particles. A more uniform and lower stress distribution, as shown in the model with PEEK components (Figure 5, model b), reduces overload areas, thereby limiting the risk of mechanical damage and friction between the joint surfaces. From a tribological perspective, lower stress values help reduce both abrasion and fatigue from cyclic loading, which are the main causes of polyethylene surface degradation. Controlling the level of contact stress thus helps preserve the integrity of the liner, decreasing the generation of debris and improving the overall durability of the implant.

3.2. Stress Analysis on Prosthetic Components

The highest stress concentrations are mainly observed in the prosthesis made of titanium (model a, Figure 6). This is due to the significantly higher Young’s modulus of titanium compared to the surrounding tibial bone, causing the prosthesis to absorb most of the applied loads, thereby shielding the bone from stress. However, from a mechanical strength standpoint, these stress levels are not critical, remaining well below the yield strength of titanium. In contrast, the model with the PEEK prosthesis shows a more uniform stress distribution, with loads spread over a larger surface area. Thanks to its lower Young’s modulus compared to titanium, PEEK promotes better stress transfer to the surrounding bone, encouraging a more physiological load distribution. Even in this case, the stress values in the PEEK prosthesis are below its yield strength, confirming its mechanical safety under the applied loading conditions.

3.3. Stress Analysis on Bone

The analysis of stress transmitted to the tibial bone in the two prosthesis configurations, one made of titanium and the other of PEEK, shows significant differences related to the mechanical properties of the materials. The stiffness of titanium, with a Young’s modulus significantly higher than that of bone, causes a phenomenon known as “stress shielding,” which is the protection of the bone from mechanical load. In practice, most of the load is absorbed by the prosthesis itself, resulting in reduced stress transfer to the surrounding bone. Consequently, the stress values in the bone under the titanium tibial insert lower, with a maximum of 1.532 MPa. This low level of stress can lead to decreased mechanical stimulation necessary to maintain bone mass and health, increasing the risk of bone atrophy and long-term compromise of stability. On the contrary, the PEEK prosthesis, characterized by lower Young’s modulus, allows greater transmission of mechanical forces to the bone. This results in a more uniform stress distribution on the bone, with higher stress values reaching a maximum of 3.176 MPa (Figure 7). This stress range is consistent with that favorable for bone maintenance and formation (2–3 MPa), suggesting that the PEEK prosthesis promotes more physiological and beneficial mechanical stimulation for the bone. The increased transmission of stress helps maintain bone remodeling and density, reducing the risk of bone loss associated with stress shielding.

3.4. Stress Shielding Evaluation

Four cross sections (labeled 1–4 in Figure 8) were examined for the stress-shielding analysis. For each section, we calculated the average stress in the bone at 1 mm from the implant and presented it in

Table 4. Bone stress values measured in the four sections for models a and b.

Sections	von Mises Stress (MPa) Model A	von Mises Stress (MPa) Model B
1	0.965	1.243
2	1.065	1.136
3	1.764	2.269
4	2.046	2.663

.

From Table 4, it can be observed that the lowest stress in the bone occurs in section 1 for model A, with an average value of 0.965 MPa, and in section 2 for model B, with 1.136 MPa. Conversely, the highest stress is recorded in the lower part of the implant (section 4), measuring 2.046 MPa for model A and 2.663 MPa for model B. Overall, the PEEK model (model B) exhibits higher stress values compared to the titanium model.

By applying Equation (1) to calculate the stress-shielding factor (SSF), the results presented in Figure 9 were obtained. A positive SSF value indicates a reduction in stress levels within the bone following prosthesis implantation, a condition that may lead to bone loss. At all investigated sections, the SSF values for the PEEK prosthesis were significantly lower. Conversely, for the Ti6Al4V titanium alloy, negative stress-shielding factor values were observed. These findings allow us to conclude that the PEEK prosthesis for TKA shows no clear evidence of stress shielding at any of the seven analyzed sections.

The analysis of the results clearly shows that the titanium prosthesis induces a stress-shielding effect at all analyzed points, with stress-shielding-factor values consistently positive and particularly high at points 2, 6, and 7. In contrast, PEEK systematically exhibits negative or near-zero values, indicating a substantial absence of stress shielding throughout the entire examined tibial area. The differences are especially pronounced at points 1, 6, and 7, where PEEK presents significantly negative values (up to −1.7), compared to the correspondingly high values observed for titanium (up to +1.3). This behavior suggests that PEEK is capable of transferring load more homogeneously and physiologically to the surrounding bone, thereby reducing the risk of pathological bone remodeling and potential long-term bone resorption. In summary, the results highlight that the use of PEEK for TKA prosthesis fabrication effectively minimizes the stress-shielding effect, indicating a greater biomechanical compatibility with bone tissue compared to titanium.

4. Discussion

The loosening of the tibial prosthesis after total knee arthroplasty (TKA) remains an unresolved issue in knee surgery [56]. Several hypotheses have been proposed to explain early loosening, including insufficient fixation, inadequate cementation technique, the phenomenon of stress shielding, or incorrect implant positioning [58,59,60]. Additionally, implant design factors such as the presence of a cement pocket under the tibial plate and surface roughness have prompted design modifications aimed at reducing failure rates. Despite these improvements, tibial loosening continues to be frequently reported in clinical practice.

A study by Chaiyaki et al. [61] highlighted how prostheses with a thicker tibial base (≥4 mm), especially when made of materials with high elastic modulus such as cobalt-chromium (CoCr), are associated with increased bone resorption in the proximal tibial region. The greater stiffness and thickness of these components reduce load transmission to the bone, particularly in the metaphyseal region, promoting stress shielding. In line with Wolff’s Law, this insufficient mechanical stimulation leads to bone resorption and decreased bone density, thereby compromising implant stability.

In this context, optimizing the prosthesis design to mitigate stress shielding is crucial. Eltlhawy et al. [62] investigated porous tibial stems based on triply periodic minimal surfaces (TPMS), such as Schwarz, diamond, and gyroid structures, made of Ti6Al4V. Finite element analyses under static and dynamic loading showed that TPMS stems increased von Mises stress on the bone surface beneath the tibial plate by 14–24% and 15–36%, respectively, when compared to solid stems. This increase is considered beneficial as it stimulates the bone and reduces the risk of resorption. Moreover, TPMS stems reduced stress concentrations at the stem tip by up to 21.3%, improving overall biomechanical integration and decreasing potential complications.

Garabano et al. [19], in a 10-year follow-up study, found a significantly higher incidence of medial tibial bone resorption with titanium tibial bases (24.41%) compared to all-polyethylene bases (1.25%). These findings support the idea that the Young’s modulus of the implant material plays a major role in stress shielding. Materials like titanium or cobalt-chromium, due to their high elastic modulus, are associated with more pronounced bone loss in the proximal tibia.

Recently, the use of polymeric materials such as PEEK (polyether ether ketone), which have mechanical properties closer to those of native bone, has emerged as a promising solution. PEEK reduces the stiffness mismatch between the implant and bone, resulting in improved load transfer and reduced stress shielding [49,63,64,65,66]. Zhao et al. [44] reported that PEEK tibial components led to increased stress transmission to the tibial tray and proximal tibia while reducing the area affected by stress shielding. Similar benefits were observed in osteoporotic bone models, confirming the potential of PEEK in complex clinical cases.

Our findings corroborate these results. The finite element analysis conducted in this study demonstrated that PEEK prosthetic components significantly reduce stress shielding compared to titanium Ti6Al4V. The von Mises stress values in the proximal tibial region with PEEK remained within the 2–3 MPa range—favorable for maintaining bone mass—while titanium components led to a more pronounced reduction in local mechanical stimulus. The calculated stress-shielding factor (SSF) in PEEK-based models was consistently close to zero or negative, indicating effective stress transmission, while titanium-based models showed high SSF values, especially in the metaphyseal region.

These results align with the existing literature and reinforce the notion that PEEK components promote more physiological stress distribution, enhance osseointegration, and may ultimately improve implant longevity. Additionally, PEEK may represent a viable alternative for patients with metal allergies, offering improved biocompatibility and reduced risk of hypersensitivity reactions [49,67,68,69,70,71].

Limitations

Despite the promising results, this study has several limitations. First, the finite element model is based on idealized boundary conditions and material properties, and it assumes perfect bonding between the implant and the bone, which may not fully reflect the complex biomechanical environment of the human knee. A fundamental limitation in applying realistic loads in finite element analysis (FEA) of knee prostheses lies in the variability of patient-specific factors, such as gait patterns, muscle strength, and joint kinematics. Moreover, replicating dynamic and multidirectional loading in simplified static or dynamic simulations can lead to oversimplification. Future studies should consider a wider range of load cycles to improve modeling accuracy.

Second, the model does not account for patient-specific anatomical variations and biological responses, which may affect the generalizability of the findings. Third, the analysis focuses solely on the proximal tibial region and does not consider long-term bone remodeling or fatigue effects over time. Future work should include experimental validation and in vivo data to better replicate physiological conditions and strengthen the reliability of numerical predictions. Finally, the effects of micromotions at the bone–implant interface, as well as the differences between cemented and uncemented fixation methods, were not evaluated in this study.

5. Conclusions

This finite element analysis study demonstrates that PEEK (polyetheretherketone) knee prosthetic components substantially reduce the stress-shielding effect compared to conventional titanium Ti6Al4V implants in total knee arthroplasty. The PEEK prosthesis provides a more uniform and physiological transmission of mechanical stresses to the proximal tibial bone, with von Mises stress values maintained within the favorable range for bone maintenance and formation (2–3 MPa). The calculated stress-shielding factor (SSF) for PEEK components consistently showed negative or near-zero values at all evaluated points, indicating minimal risk of bone resorption and pathological remodeling, in contrast to the titanium model which exhibited a pronounced stress-shielding effect. Moreover, stress concentrations on the polyethylene liner were lower and more evenly distributed with the PEEK model, suggesting potential improvements in implant longevity through reduced wear. These findings support the superior biomechanical compatibility of PEEK material for tibial components in knee prostheses, with potential clinical benefits including better preservation of bone microarchitecture and enhanced implant durability, even in osteoporotic patients. Further experimental validation and long-term clinical studies are recommended to corroborate these promising computational results.