Elsevier

Journal of Biomechanics

Volume 47, Issue 14, 7 November 2014, Pages 3509-3516
Journal of Biomechanics

Stress-shielding induced bone remodeling in cementless shoulder resurfacing arthroplasty: a finite element analysis and in vivo results

https://doi.org/10.1016/j.jbiomech.2014.08.029Get rights and content

Abstract

Cementless surface replacement arthroplasty (CSRA) of the shoulder was designed to preserve the individual anatomy and humeral bone stock. A matter of concern in resurfacing implants remains the stress shielding and bone remodeling processes. The bone remodeling processes of two different CSRA fixation designs, conical-crown (Epoca RH) and central-stem (Copeland), were studied by three-dimensional (3-D) finite element analysis (FEA) as well as evaluation of contact radiographs from human CSRA retrievals. FEA included one native humerus model with a normal and one with a reduced bone stock quality. Compressive strains were evaluated before and after virtual CSRA implantation and the results were then compared to the bone remodeling and stress-shielding pattern of eight human CSRA retrievals (Epoca RH n=4 and Copeland n=4). FEA revealed for both bone stock models increased compressive strains at the stem and outer implant rim for both CSRA designs indicating an increased bone formation at those locations. Unloading of the bone was seen for both designs under the central implant shell (conical-crown 50–85%, central-stem 31–93%) indicating high bone resorption. Those effects appeared more pronounced for the reduced than for the normal bone stock model. The assumptions of the FEA were confirmed in the CSRA retrieval analysis which showed bone apposition at the outer implant rim and stems with highly reduced bone stock below the central implant shell. Overall, clear signs of stress shielding were observed for both CSRAs designs in the in vitro FEA and human retrieval analysis. Especially in the central part of both implant designs the bone stock was highly resorbed. The impact of these bone remodeling processes on the clinical outcome as well as long-term stability requires further evaluation.

Introduction

Shoulder arthroplasty has become an efficient treatment option for different pathologies of the gleno-humeral joint. Although good clinical results are achieved, complications as periprosthetic fractures, glenohumeral instability or aseptic loosening can compromise the long-term success (Sperling et al., 2013). To improve long-term strategies for patients and ease upcoming revision surgeries, bone conserving implants have been developed (Ballas and Beguin, 2013, Burgess et al., 2009, Delaney et al., 2013, Huguet et al., 2010).

Cementless surface replacement arthroplasty (CSRA) of the humeral head has recently gained popularity as it conserves the individual humeral anatomy and preserves humeral bone stock (Burgess et al., 2009). Studies have reported about a good short- and mid-term functional outcome and no increased revision rates for CSRA (Al-Hadithy et al., 2012, Alizadehkhaiyat et al., 2013, Levy and Copeland, 2004, Raiss et al., 2009). However, from hip resurfacing arthroplasty (HRA) it is well known that resurfacing implants affect the load transfer and induce stress shielding (Gupta et al., 2006, Ong et al., 2006a). According to Wolff´s law, stress shielding can lead to excessive bone resorption affecting the clinical outcome and leading to premature implant failure (Ruben et al., 2012).

Recent clinical studies have provided evidence that stress shielding also occurs in conventional shoulder replacement arthroplasty (Nagels et al., 2003, Verborgt et al., 2007). For CSRA of the humerus only little data is available, as the bone is covered by the radiopaque implant shell and long-term results are not yet available. However, in a current CRSA retrieval study, we found clear signs of stress shielding below the implant (Schmidutz et al., 2012).

Therefore, this study aimed to further evaluate the load transfer and bone remodeling processes of CSRA implants considering two different geometrical designs. Finite element (FE) models of a normal and a reduced bone stock were created and the stress-shielding pattern were analyzed before and after virtual implantation of the two different CSRA implants. Furthermore, the results were then compared to human CSRA implant retrievals. Our hypothesis was that CSRA implants cause significant stress shielding and bone remodeling processes below the implant.

Section snippets

Implant geometry and fixation

Two CSRA designs with different fixation philosophies, one with a conical-crown shaped ring (Epoca RH, DePuy Synthes, USA) and one with a central-stem (Copeland, Biomet, USA) (Fig. 1, Fig. 2), were evaluated. Both designs are commercially available and in clinical use. They are anchored cementless and the central stem intends to provide sufficient primary stability after insertion. Both, the Epoca RH and Copeland have a spherical joint surface and a hydroxyapatite coating at the inner surface

Finite element analysis

The compressive strains for the native humeral models with a normal and reduced BMD and after virtual implantation of the CSRA implants are depicted in Fig. 2, Fig. 3. The results for the absolute strains in the eight regions are given in Table 2. The changes [%] of the compressive strains compared to the native humeral models are given in Table 3 and depicted in Fig. 4.

Discussion

Minimization of stress-shielding is an important factor for the long-term success of implants, as it can cause excessive bone resorption and compromise the outcome of an implant (Decking et al., 2006, Ruben et al., 2012). The present study evaluated the stress-shielding and bone remodeling pattern of two different CSRA designs using FEA and human retrieval analysis.

Our results provide evidence that CSRA causes an inhomogeneous strain distribution in the humeral head and induces

Conflict of interest statement

All authors declare that they have no potential conflicts of interest.

Acknowledgments

This study was performed at the AO Research Institute Davos. The authors thank Biomet and Synthes for providing the CAD data for the CSRA implants and the AO Research Institute for funding the research fellowship of F. Schmidutz.

References (29)

Cited by (0)

View full text