Biomechanical study of using patient-specific diaphyseal femoral cone in revision total knee arthroplasty (rTKA)

Abstract

Background

The primary objective of revision total knee surgery is to achieve solid bone fixation. Generally, this could be accomplished using sleeves and long stems, which require substantial remaining bone stock and may increase the risk of stem tip pain. An alternative approach involves the use of customized diaphyseal cones, which can preserve the integrity of the bone canal. This study evaluates the impact of employing femoral diaphyseal cones with various stem lengths on stress distribution and relative motion.

Methods

CT scan data from five patients were used to generate the 3D model of the femur, cement, customized stems, and cones, along with assigning patient-specific material for each candidate's femur. Three different stem lengths, both with and without the customized cone, were assessed under three gait loading conditions to compare the resulting Von Mises stress distribution and relative motion.

Results

Analysis indicated that the use of customized femoral cones moderately increases stress distribution values up to 30 % while significantly reducing relative motion at the femoral canal-cone interface by nearly 60 %. The presence of the cone did not significantly alter relative motion with varying stem lengths, although stem length variation without a cone substantially affected these values.

Conclusion

Incorporating cones alongside stems enhances metaphyseal fixation, reduces stress shielding, potentially allowing for the use of shorter stems. Furthermore, cones promote osseointegration by minimizing relative motion, ultimately improving prosthetic stability.

1. Introduction

The goal of revision total knee arthroplasty (rTKA) is to create a stable prosthesis that allows for pain-free movement within an acceptable range of motion and preserves bone tissue while addressing defects.¹ Achieving solid fixation is crucial for expediting post-operative mobility and rehabilitation, as well as enhancing the prosthesis's longevity in revision knee arthroplasty.^2,3 However, this can be challenging in cases with inadequate bone stock.⁴

One of the key approaches to enhancing stability and alignment in revision arthroplasties involves the use of stems.⁵ The use of intramedullary stems can improve implant fixation by reducing stress on the implant surface through load sharing, thereby increasing stability.^6,7 Diaphysis fixation can also be achieved through the use of stems, which reduces the load on the metaphysis. Metaphyseal fixation can be achieved through the use of cement, sleeves or various cone techniques.⁸

Using porous cones increases the contact surface between the implant and the bone, optimizing stress distribution at the bone-implant interface. The high friction coefficient of the porous structure on the cone's surface, when in contact with the bone, also decreases micro-motions.⁹ Additionally, due to their porous structure, titanium cones possess the same elastic modulus as cancellous bone, thereby mitigating the occurrence of stress shielding phenomenon.¹⁰ The porous structure also facilitates bone ingrowth and osseointegration, thereby promoting biological fixation.¹¹ By leveraging these advantages, the use of a metaphyseal cone in rTKA permits the use of a shorter stem to attain a stable and durable fixation.¹² This feature reduces the risk of stem tip pain in patients after rTKA.¹³ Furthermore, experimental cadaver studies and finite element analysis (FEA) have shown that the use of a tibial cone and stem reduces the risk of cement-implant interference debonding, especially in demanding activities such as stair descent.¹⁴

This study aims to evaluate the biomechanical implications of utilizing patient-specific femoral meta-diaphyseal cones and varying stems lengths on implant stability during rTKA, under standard loading scenarios (normal walking, stair descent, and stair ascent), through FEA.

2. Methods

2.1. 3D Modelling and planning

Three-dimensional DICOM images of five rTKA candidates were used to construct the geometrical model of the femurs from the distal endpoint to the midshaft. The age of the patients ranged from 59 to 76 years including two males and three females. Personalized stems and solid diaphyseal cone augments matching the contour of the femoral canal were designed for each patient (Fig. 1). All Cones were individually tailored based on the computed tomography (CT) scans to match each patient's unique femoral canal anatomy. The lengths of the cones were standardized at 2.5 cm, conforming to the dimensions of commercially available standard cones.

Fig. 1.

Femoral pre-operative planning for finite element modeling, depicting the distal femur sectioning to facilitate the modeling process.

The diameter of each stem was selected to fit the specific dimensions of the patient's medullary canal. To evaluate the influence of the stem length on the resulting relative motion and stress distribution, the distance between the proximal region of the cone and the stem tip was considered variable with three increments measured at 3 cm, 4 cm, and 5 cm. Consequently, stems measuring 65 mm, 75 mm, and 85 mm were examined (Fig. 1). Additionally, the cement geometry was modeled to fill all remaining voids from the distal end to a point 3 mm proximal to the stem tip.

2.2. Finite element Modelling

For each patient, the simulation permutations consisted of: a) two treatment strategies, utilizing either the presence or absence of a customized cone; b) three variations in stem length; and c) three distinct gait loading conditions. This configuration yielded a total of 90 simulations. Dynamic implicit simulations were conducted using ABAQUS software (Simulia, Rhode Island, United States) to determine the Von Mises stress distribution within the femoral canal, as well as the relative motion values at the contact interfaces across all scenarios for each patient.

2.3. Material assignment

Mechanical properties of cone, stem, and cement were considered solid homogeneous.^15,16 For Femurs, however, the gray value-based material assignment was used to generate the density and Young's modulus of each patient's bone according to the Hounsfield unit values.¹⁷ The gray value ranges for each patient were divided into 8 main material sections (One section of bone marrow, two for cancellous bone, and five for cortical bone) and 10 subsections which resulted in a total of 80 materials. The mechanical properties of the models are shown in Table 1.

Table 1.

Mechanical properties of different parts of the model.

Material	Density ($\frac{\mathit{g}}{\mathit{c}{\mathit{m}}^{\mathbf{3}}}$)	Young's modulus (MPa)	Poisson's ratio (v)
Tibia Marrow	0.000464*HU+1	1	0.3
Cancellous Bone	0.000464*HU+1	1904*ρ^1.64	0.3
Cortical Bone	0.000464*HU+1	2065*ρ^3.09	0.3
Cone/Stem (Ti6Al4V)	4.43	110000	0.3
Cement	1.2	2000	0.46

2.4. Contact Definition

Contacts were defined established using standard surface-to-surface (Master-Slave) techniques with hard contact properties in the normal direction and penalty method for tangential behavior across all contact pairings. These pairings were assigned different coefficients of friction as outlined in Table 218, 19, 20

Table 2.

Coefficient of friction values for Contact Pairs in models.

Contact Pairs	Coefficient of Friction
Bone-Cone	0.9
Bone-Cement	0.7
Stem-Cement	0.2
Cone-Cement	0.2

2.5. Load and boundary conditions

For each patient, a Cartesian coordinate system was defined according to the stem's orientation (X: lateral, Y: anterior, and, Z: the axis of the stem) to accurately model the loading condition (Fig. 2a). To optimize simulation efficiency, the femoral component was not included in the model. Instead, forces and moments were applied to a reference point situated at the center of the distal stem's cross-section, where it connects to the femoral component (Fig. 2a–and b). This reference point was kinematically linked to the stem's cross-section surface. Three different gait cycle amplitudes—representative of normal walking, stair ascent, and stair descent—were applied to the reference point, utilizing data from the Orthoload database.²¹ A fixed boundary condition was imposed on the proximal end of the femur to simulate its in vivo constraint (Fig. 2b).

Fig. 2.

Load application at the reference point and the fixed boundary conditions (BCs) applied to the proximal femur.

2.6. Mesh construction

For all models, a uniform surface mesh was created with a maximum edge length of 1.5 mm. A non-structured finite element volume mesh comprising three-dimensional tetrahedral elements with four nodes (C3D4) was generated, maintaining the maximum triangle edge at 2 mm (Fig. 3). To ensure the mesh was sufficiently refined for accurate stress analysis, a convergence study was performed on the femur of patient #1. Aimed to determine an appropriate mesh density, the convergence study resulted in average von Mises stress values at the cone-femoral canal interface area that did not deviate by more than 3.44 % from the converged value. The results of this mesh study are summarized in Table 3.

Fig. 3.

Mesh construction for models.

Table 3.

Mesh convergence study results on the femur of patient #1, using a 3 cm stem and cone, considering a normal walking gait.

	Mesh Size (mm)	Number of Nodes	Maximum stress at the femoral canal and cone interface area (MPa)	Average stress at the femoral canal and cone interface area (MPa)
Mesh #1	1	57893	5.58	1.10
Mesh #2	1.25	50131	4.76	1.12
Mesh #3	1.5	44502	4.9	1.16
Mesh #4	1.75	39715	4.79	1.16
Mesh #5	2	35965	4.35	1.17

3. Results

The maximum amounts of Von Mises stress and relative motion during the gait cycles at the bone-cone contact area were recorded for each patient across all simulations (see Appendix A). The average values derived from these data points for the cohort of five patients are presented in Table 4. Furthermore, 2D plots of these results are depicted in Fig. 4, Fig. 5 for visual reference.

Table 4.

Average of the maximum observed von Mises stress (MPa) and relative motion (micrometers) at the femoral canal and cone interface area for patients across three gait cycles.

Average Results	Cone usage?	Stem 3 cm			Stem 4 cm			Stem 5 cm
		Normal	Down	Up	Normal	Down	Up	Normal	Down	Up
Max Stress (MPa)	Yes	6.26	7.82	7.16	7.85	8.94	8.49	6.41	7.68	5.91
	No	5.67	6.94	6.40	5.46	6.59	5.76	4.38	5.23	4.28
Max relative motion (Microns)	Yes	15.61	17.20	16.58	17.03	14.66	17.04	14.39	13.48	14.57
	No	44.81	66.92	59.37	45.92	59.94	58.69	21.57	27.93	29.75

Fig. 4.

Average maximum Von Mises stress experienced by five patients at the femoral canal and cone interface area during gait cycles, categorized by stem length (in MPa).

Fig. 5.

Average maximum relative motion measured in micrometers for the same five patients during the gait cycle, categorized by stem length (Microns).

As indicated by Table 4 and Fig. 4, stress levels within the femoral canal when a cone was used were nearly 30 % higher compared to when it was not used. The highest observed stress within the femoral canal reached 8.94 MPa in models incorporating the cone, as opposed to 6.94 MPa in those without. It was also noted that as the stem length increased, stress values in models not using cones marginally decreased. However, models with cones showed that the 4 cm stem generated the highest stress values. With respect to the type of loading, descending stairs induced the greatest stress, while ascending stairs and normal walking resulted in lower stresses.

Fig. 5 demonstrates the significant reduction in relative motion—by approximately 60%—afforded by the inclusion of a cone. Analogous to the stress results, lengthening the stem decreased the relative motion in models lacking a cone, but this effect was not as pronounced in models equipped with a cone.

To succinctly demonstrate the variations in stress and motion, 3D contours of the maximum Von Mises stress and relative motion for two patients were visualized, comparing the scenarios with and without the utilization of a cone under a specified gait load. Fig. 6 illustrates the stress contours for patient #1 using a 5 cm stem during a normal walking gait load. The results showed that stress values in the proximal region, near the stem tip, remained consistent at approximately 4–5 MPa. Closer to the bone-cone interface in the distal region, there was a discernible increase in stress for the models incorporating cones.

Fig. 6.

Maximum Von-Mises stress distribution contours (MPa) in patient #1 under normal walking gait load, a) with and b) without cone.

Fig. 7 presents a parallel analysis of relative motion 3D contours under the same gait condition for patient #4. Here, the relative motion did not surpass 80 μm when a cone was present, with the least motion (7–15 μm) observed directly at the bone-cone interface. Conversely, without a cone, relative motion escalated to about 40 μm in the corresponding area, highlighting the cone's efficacy in minimizing motion.

Fig. 7.

Maximum relative motion contours (mm) in patient #4 under normal walking gait load, a) with and b) without cone.

4. Dicussion

According to the concept of zonal fixation, achieving stable fixation in at least two out of the three zones (epiphysis, metaphysis, and diaphysis) is crucial. This consideration should be meticulously integrated into preoperative planning and implant selection.²² Diaphyseal fixation can be facilitated by stems, while metaphyseal fixation can be achieved using cement, sleeves, or cone techniques.⁸ In this study, we investigated the impact of employing a diaphyseal femoral cone with a stem on prosthesis relative motion and bone stress in rTKA through FEA in five patients. Specifically, we assessed the effects of employing a cone with stems of varying lengths (65 mm, 75 mm, and 85 mm) under different loading conditions (normal walking, descending stairs, and ascending stairs).

Although several biomechanical studies have explored the impact of using tibia metaphyseal cones, to our knowledge, this study is the first to investigate the effect of employing femur diaphyseal cones. Cones are commonly used for their ability to promote osseointegration and to distribute stress uniformly within the bone.²³ The applied load and the stress on the bone are critical factors for the longevity of the reconstructive technique in rTKA.²⁴ Our results indicated that the utilization of a cone, as opposed to not using one in rTKA, leads to an approximate 30 % increase in bone stress.

Increasing bone stress can be beneficial in preventing stress shielding, as Wolff's law suggests that reducing the load transferred to the bone can result in bone weakening and subsequent loosening over the long term.²⁵ Moreover, the ability to design and produce a cone with a porous structure helps in ensuring that the elastic modulus of the implant and bone become more similar, thereby mitigating the risk of stress shielding.²⁶ Similarly, in the tibia, employing a cone alongside a stem creates a rigid structure that prevents stem migration and offers favorable conditions for bone ingrowth and prosthesis fixation.²⁷ The use of a stem in rTKA assists in bypassing the joint surface and metaphyseal bone, thus distributing the increased stress caused by constrained articular fixation.²⁸

The simulations revealed that for all three loading modes (normal walking, ascending, and descending stairs) and regardless of whether a cone was used, bone stress decreased with an increase in stem length. This decrease is attributed to the stem's role in transferring load from the metaphysis to the diaphysis. However, using longer stems introduces greater surgical complexity and concerns about stress shielding, which in turn may elevate the risk of long-term fractures.²⁹ Research suggests that using a short stem in tandem with a cone not only achieves safer metaphyseal fixation but also mitigates stress concentration and the occurrence of stress risers, thereby enhancing stability, similar to findings in the tibia³⁰ Moreover, employing a short stem (without diaphyseal engagement) alongside a cone has been shown to perform comparably to long stems (with diaphyseal engagement), making the combined use of a cone and stem a dependable method for achieving fixation.¹² A significant benefit of opting for a short stem over a long one is the lower risk of end-of-stem pain 13.

In total knee arthroplasty, the consideration for stem use extends beyond length to include diameter. Achieving optimal stem stability requires a canal fill ratio exceeding 0.85, thus ensuring a stable structure.³¹ This crucial factor was diligently accounted for in the pre-operative planning for the models in this study, with stem diameters tailored to the unique medullary canal of each patient. Porous cones play a vital role in providing essential metaphyseal fixation for the new implant, which, in turn, supports the use of a short stem.³²

The findings from this study reveal that using a cone, irrespective of stem length, significantly reduces relative motion by approximately 60 %. Notably, when a cone is used, extending the stem length does not further reduce relative motion. This finding contrasts with scenarios where the cone is not employed, as relative motion diminishes with an increase in stem length. Thus, the application of a cone allows for the use of shorter stems, which is particularly advantageous in patients with diaphyseal bowing, deformity, or ipsilateral hardware, making the combination of a short stem and cone a preferred method for achieving fixation.

Experimental evidence has also shown that the use of a metaphyseal sleeve alongside a stem in the tibia during rTKA lessens relative motion.³³ Prior studies indicate that to enhance osseointegration at the implant-bone interface, relative motion should be maintained below 50 μm.^34,35 In this study, the observed relative motion with the use of a cone across all loading conditions was around 20 μm, suggesting that porous cones significantly aid in bone ingrowth and biological fixation.³⁶ Additionally, the porous structure of the cones not only facilitates osseointegration but also contributes to reducing implant stiffness and preventing stress shielding due to its elastic modulus being similar to that of bone.³⁷

For finite element modeling and analysis, a customized cone based on CT scans of each patient was utilized, ensuring it was tailored to individual anatomical requirements. Previous research has indicated that relative motion with customized cones is lower in the medial, valgus, and posterior directions compared to standard, off-the-shelf cones.³⁸

5. Limitations & Strength

This study has several limitations, including a relatively small sample size, which restricts the ability to draw more definitive conclusions. A more extensive analysis involving a larger cohort could provide more conclusive results. Another limitation is the omission of the time-dependent response of cement in the analysis, an aspect that could be explored in future research. Additionally, the modeling of the cone as solid, without considering the porosity of its surface in the FEA, suggests that future investigations could benefit from incorporating porous cone modeling to better understand and simulate osseointegration.

6. Conclusion

The utilization of cones and stems is a recognized method for securing fixation in revision total knee arthroplasty. The findings of the finite element analysis in this study demonstrate that incorporating a cone alongside the stem not only establishes metaphyseal fixation, but also mitigates stress shielding, allowing for the potential use of shorter stems. Additionally, the use of cones facilitates osseointegration by reducing relative motion, consequently enhancing the stability of the prosthesis.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Ethical APPROVAL and patient consent

Access to the patient CT scan electronic data was approved by the Research Ethics Committees of School of Medicine, Tehran University of Medical Sciences (ID: IR.TUMS.MEDICINE.REC.1401.444).

All procedures performed in studies involving human participants were conducted according to the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments.

Declaration of patient consent FORM

The study was conducted using CT scan images obtained as part of routine clinical practice, thus no patient consent was required.

CRediT authorship contribution statement

Reza Nourishirazi: Software, Methodology. Ghazaleh Moradkhani: Methodology, Validation, Writing – original draft. Arash SharafatVaziri: Supervision, Conceptualization. Hamidreza Nematy: Writing – original draft. Ramin Shayan-moghadam: Data curation, Writing – original draft. Morad Karimpour: Project administration, Supervision, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is/are the supplementary data to this article.