Effects of stemmed and nonstemmed hip replacement on stress distribution of proximal femur and implant
- 1.7k Downloads
Despite improvements in shape, material, and coating for hip stem, both stress shielding and aseptic loosening have been the major drawbacks of stemmed hip arthroplasty. Some nonstemmed systems were developed to avoid rasping off the intramedullary canal and evacuating the bone marrow due to stem insertion.
In this study, the finite-element models of one intact, one stemmed, and two nonstemmed femora with minimal removal of the healthy neck were investigated to evaluate their biomechanical effects. The resurfacing (ball-shaped) and fitting (neck-shaped) systems were respectively selected as the representative of the ready- and custom-made nonstemmed implants. The stress distribution and interface micromotion were selected as the comparison indices.
The results showed that stress distributions of the two nonstemmed femora are consistently more similar to the intact femur than the stemmed one. Around the proximal femur, the stem definitely induces the stress-shielding phenomenon of its counterparts. The fitting system with the anatomy-shaped cup can make intimate contact with the neck cortex and reduce the bone-cup micromotion and the implant stress. Comparatively, the reamed femoral head provides weaker support to the resurfacing cup causing higher interfacial micromotion.
The reserved femoral neck could act as the load-transferring medium from the acetabular cup, femoral neck, to the diaphysial bone, thus depressing the stress-shielding effect below the neck region. If the hip-cup construct can be definitely stabilized, the nonstemmed design could be an alternative of hip arthroplasty for the younger or the specific patients with the disease limited only to the femoral head.
KeywordsHip Stemmed Nonstemmed Stress-shielding Finite-element
Total hip replacement
Total hip replacement (THR) with the intramedullary stem has been successfully employed to treat various hip diseases of the patients over the age of forty [1, 2, 3]. The surgical method of the stemmed THR is to amputate the intact femoral neck entirely, rasp off the intramedullary canal, and evacuate the bone marrow for the stem insertion. However, stem insertion into the intramedullary canal necessitates both blood transfusion and bone reaming; and thus might lead to infection [1, 2, 3, 4, 5]. If the hip defects are limited to the superficial regions of the femoral head, some studies suggested preserving the bridging bone stock (femoral neck) between the femoral head and the diaphysial shaft [6, 7, 8, 9].
There are two types of stem-induced complication that affect the long-term results of THR. One is bone resorption (osteolysis) with the wear debris inducing chemical reaction of the immune system . The other is bone loss (osteoporosis) in response with the stress of the proximal femur shielded by the inserted stem [11, 12, 13, 14]. A great many of the clinical results have demonstrated that the stress-shielding effect leads to bone remodeling of the stemmed femur and the implant failure (loosening and cracking) [15, 16, 17]. For the diaphysial cortex, the impingement of the stem end has been reported as the stress raiser that induces bone hypertrophy and even fractures .
For younger or more active patients, the removal of the more proximal bone stock makes the revision surgery unreliable and even leads to subsequent surgery with increasingly difficult problems and poorer results [1, 5, 15, 17]. In the literatures, some nonstemmed systems have been designed and evaluated by numerical or experimental methods [11, 12, 18, 19, 20]. These studies consistently revealed that the proximal region of the nonstemmed femur shows more physiological stress distribution patterns than the stemmed one [11, 12, 18]. The reported biomechanical advantages of the nonstemmed systems over their counterparts were attributed to the preservation of both the femoral neck and the intramedullary canal.
The reported nonstemmed systems can be divided into two types: femoral ball- and neck-shaped cups [18, 21]. Both types preserve the femoral neck and cover the head and neck with the ready- or custom-made cups. The resurfacing system (Durom™, Zimmer, Inc., Indiana, USA) is the typical representative of the ball-shaped cup. Prior to use, the femoral head is reamed as a cylinder to fit the ready-made resurfacing cup. The cancellous cylinder of the reamed ball serves as the load bearer of the superimposed cup. From the biomechanical viewpoint, the femoral neck is subjected to the combined loads, potentially leading to sliding, bending, and twisting of the superimposed cup. Without stem support, hence, the mechanical failure (e.g. loosening and breaking) and the construct instability at the bone-implant interfaces have been reported as the major concerns of hip resurfacing system [22, 23].
For the neck-shaped cup, the CT images can be used as the shape reference for designing the custom-made cup that can intimately fit the peripheral cortex of the preserved neck. Without stem insertion, the bone-implant constructs of the two cup systems are stabilized by screws. From the biomechanical viewpoint, the performance of the nonstemmed system is closely related to the load-transferring mechanism of the hip cup, interfacial fitness of the bone-cup construct, and mechanical strength of the underlaid bone (cancellous or cortical). According to the authors’ knowledge, however, there is still no study investigating extensively the biomechanical differences between the ready-made ball- and the custom-made neck-shaped cups.
This study aimed to investigate three subjects of hip arthroplasty: 1) the stem-induced stress-shielding effect on the proximal femur, 2) the biomechanical influence of preserving the femoral neck on the proximal femur, and 3) the micromotion difference between the ball- and neck-shaped cups at the bone-implant interfaces. Based on the CT images, the finite-element model of the intact femur was three-dimensionally reconstructed and the stemmed and nonstemmed hip implants were instrumented. The results of this study would provide insight into the load-transferring and interface-slipping mechanisms of both stemmed and nonstemmed systems.
Femoral model and hip prostheses
Only the proximal femur was analyzed and the distal end was rigidly fixed (Figure 1A). Both hip compression and muscular contractions were simulated as the physiological loads onto the proximal femur. The X-Y-Z coordinate system with the origin at the knee intercondyle was set to define the muscular contractions. The X, Y, and Z axis were mediolaterally, anteroposteriorly, and superoinferiorly directed, respectively (Figure 1A). The insertion points and force directions of the gluteus, piriformis, iliopsoas, ilio-tibial tract, vastus lateralis, and adductor were cited from the literature [25, 26, 27]. The vectorial components of the hip and muscle forces on the femur were selected at the static position of one leg stance [18, 28, 29, 30]. The geometric models of one stemmed and two nonstemmed systems were developed in the SolidWorks environment (Figure 1).
The stress distribution, interfacial micromotion, and implant failure were the main concerns of this study (Figures 1A and 2). Four lines from aa to dd were drawn and projected onto the femoral surfaces to evaluate the implant-related effect on the cortical stress (Figure 1A). For the nonstemmed systems, the highly stressed bar and screws make them the potentials for mechanical failure that were evaluated by the stress distribution along the three lines from ee to gg (Figure 2A and B). The interfacial micromotion was employed to indicate the instability of the bone-cup construct and was determined along lines hh and ii (Figure 2C and D). The von Mises and maximum principle stresses were respectively used as the comparison indices of the bone and implant stresses.
The Young’s modulus, Poisson’s ratio and bone density were respectively assigned the values of 17.5 GPa, 0.3 and 1.9 gcm-3 for the cortical bone and 1.5 GPa, 0.12 and 0.8 gcm-3 for the cancellous bone [30, 31, 32, 33]. The Co-Cr-Mo alloy with Young’s modulus (=234 GPa) and Poisson’s ratio (=0.3) was used as the material of the stemmed and nonstemmed implants. The friction coefficients were assigned as 0.2 between cup and bone and 0.32 between metallic prostheses, respectively [34, 35]. All materials were assumed to have linearly elastic, homogeneous, and isotropic material properties throughout. The calculated stresses of all implants were compared with the yielding strength (=450 MPa) of the Co-Cr-Mo alloy to validate the assumption of linear elasticity. The interaction of touching surfaces between the bone-implant interfaces were simulated with the surface-to-surface contact elements to allow interfacial slippage.
By using the automatic mesh algorithm, the bone-implant constructs were meshed by the software Simulation, version 2011 (SolidWorks Corporation, Concord, MA, USA). Four models were meshed by the 10-node tetrahedral solid elements with the curved element boundary, thus eliminating the sharp discontinuities to induce unrealistically high stress concentration. According to aspect ratio and Jacobian checks, all elements were within acceptable distortion limits to maximize the accuracy of the results. There was a refinement of the mesh in the vicinity of the aforementioned lines, cups and screws, to improve the modeling accuracy until excellent monotonic convergence behavior (h-adaptive method) with < 1% difference in the total strain energy was achieved. A special element density, which is twice averagely that for the rest of the model, was used for the cups and screws. The overall average element size was 3.5 mm. The numbers of elements of the four models were 92,911 (intact), 74,552 (stemmed), 89,703 (resurfacing), and 100,683 (fitting), respectively.
Stress distribution of surrounding bones
Micromotion at bone-cup interface
Stress distribution of nonstemmed implants
Stress distribution of surrounding bones
From the biomechanical viewpoint, the inserted stem inevitably redistributes the stress values of the proximal femur [11, 13, 14]. Subsequently, the reduced stress and induced debris around the stem make the surrounding bones prone to bone osteoporosis (stress shield) and resorption (immune reaction). These two bone reactions often deteriorate the stem-stabilizing environment. Hence, the weakened bone strength and increased interface micromotion cannot support the external loads through the inserted stem [1, 11, 12, 13]. Figure 3 show significant differences in cortical stresses on the medial side between the intact and the stemmed femora. For the nonstemmed femora, however, the cortical stresses within the proximal and diaphysial regions were similar to those of the intact femur.
For the medial cortex below the line jj (Figure 3A), the decrease in cortical stress showed that most of the proximal loads were transferred from the stem to the diaphysial cortex . According to Wolff's law, the reduction in bone loads causes the bone to adapt itself by reducing the mechanical strength (material remodeling) or thinning the trabecular size (structural remodeling). The results of the current study were consistent with the reported stem-induced effect on bone stress [11, 13, 14]. This indicates potentially higher aseptic loosening between the intramedullary stem and cancellous bone compared with that of the nonstemmed implant [11, 12, 37], which is consistent with the phenomenon observed for cortical stress. The results of this study demonstrated the stem-induced stress problem on the proximal femur.
Micromotion at bone-cup interfaces
Stress distribution of nonstemmed implants
Contrary to the stemmed system, the ball-shaped cup is supported by the underlaid bone and further stabilized by the central bar. However, this makes the central bar as the heavily stressed component (Figure 5). Except for the cup-bar junction, stress concentration also occurs at approximately the middle region of the central bar where is about projective site of cup rim onto the bar (Figure 5A and B). It can be explained by the fact that, outside the cup, the bar begins to bear the loads. For the fitting system, the intercomponent junctions and cup slots were the stress-concentrated sites. Owing to shape discontinuity, the highly stressed sites of the central and locking screws are the interlocking hole and the diameter-reduced corner (Figure 5C). The designs should be modified to avoid plastic yielding or fatigue cracking at these sites. Results of this study showed that the long-term stabilization of the bone-cup construct is the major concern of the nonstemmed system without stem insertion.The load-transferring mechanisms of the two nonstemmed systems are compared in Figure 6B and C. Hip compression is the dominant load at the femoral head while anteroposterior loads are comparatively minor, resulting in comparable values of cortical stress between the resurfacing and fitting systems (Figure 3B). For the resurfacing system, hip compression was transferred into the proximal femur through the cup interior (red arrows) and central bar (black arrow) (Figure 6B). For the fitting system, the load-transferring paths are the cup interior (red arrows) and the central screw (black arrow) (Figure 6C). The load through the upper contact surface of the fitting cup is further transferred through the locking screw. The load through the central screw is concentrated at the intersecting site and transferred into the lateral cortex. Due to support of lateral cortex and longer bone-screw contact, the peak stress at the intersecting site of the central screw was somewhat higher than that of the central bar (Figure 5A).
The stem insertion induces stress redistribution of the proximal and diaphysial femur. Less removal of the proximal bone stock makes the load-transferring pattern of the nonstemmed femur more physiologically similar to that of the intact femur. Without the inserted stem, the central bar of the resurfacing system was highly stressed, making it prone to mechanical failure. Both contact situation and bone strength should be determinant factors of long-term stability of the nonstemmed system. For the cup-shaped system, the cup should be the custom-made and the underlaid cortex could be preserved to provide more intimate and stronger support to the superimposed cup. If the cup can be definitely stabilized, the nonstemmed design could be an alternative for the younger or the specific patients with the disease limited only to the femoral head.
We would like to thank Chi-Pin Hsu for his help in three-dimensionally reconstruction of the proximal femur and Wen-Hsien Chuang for his participation in scanning in vivo using computed tomography.
- 5.Epinette JA, Manley MT: Fifteen years of clinical experience with hydroxyapatite coating in joint arthroplasty. Springer, Health and Medicine - Medical Specialties. 2003, New York: Clinical, 217-224.Google Scholar
- 13.Chen WP, Tai CL, Lee MS, Lee PC, Liu CP, Shih CH: Comparison of stress shielding among different cement fixation modes of femoral stem in total hip arthroplasty – A three-dimensional finite element analysis. J Med Biol Eng. 2004, 24: 183-187.Google Scholar
- 14.Tai CL, Chen WP, Lee MS, Lee PC, Shih CH: Comparison of stress shielding between straight and curved stem in cementless total hip arthroplasty – an in vitro experimental study. J Med Biol Eng. 2004, 24: 177-181.Google Scholar
- 20.Chen TH, Lung CY, Cheng CK: Biomechanical comparison of a new stemless hip prosthesis with different shapes-a finite element analysis. J Med Biol Eng. 2009, 29: 108-113.Google Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2474/15/312/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.