A 3d Real-time Micromechanical Study of Bone-cement Interface

INTRODUCTION The lasting integrity of the bond between the bone and the cement is essential to the success and longevity of cemented total hip replacements [1]. It is well known how the strength of the bone-cement interface is related to the amount of bone interdigitated with the cement [2] and although numerous studies have been carried out on bone-cement interfacial strength at the continuum level, relatively little is known about the micromechanical behaviour of the interface. The influence of the local stress distribution on the interface; whether bone or cement fails first; the effect of bone quality and morphology on the interfacial strength, are all important in the understanding of the aseptic loosening process [3, 4]. As a result of substantial heterogeneity, anisotropy, intraspecimen variability, and the difficulty in preparing samples, bone mechanical properties exhibit substantial variation. For this reason several bone substitutes, such as open-cell foams, have been used in order to simulate a range of cancellous bone properties by controlling the morphological parameters in bone analogue materials. Image-guided failure assessment (IGFA) is a technique that allows 3D progressive analysis of damage. It involves the use of a micro-compression device to apply and maintain a given displacement during the microcomputed tomography (microCT) imaging. The specimen is compressed and timelapsed imaged, allowing the assessment of damage initiation and progression not only in the elastic region but also in the plastic region [5]. In this paper the micromechanical behaviour of the bonecement interface has been studied using both experimental and numerical approaches. Foam-cement interface samples were created and compared with those of bovine bone-cement samples. The objectives of the study were: (1) To use a novel micromechanical loading stage (LS) and microCT to assess local deformation and damage at the interface under selected levels of uniaxial compression; (2) to correlate microstructural stresses estimated by an image-based finite element analysis with regions of bone/foam-cement interface microdamage and (3) to evaluate how the morphology of the interdigitated area affects the micromechanical response of the interface. MATERIALS AND METHODS AlSi7Mg (45ppi, pores per inch) aluminum foam (M-pore, Germany) and bovine trabecular bone taken from fresh bovine iliac crest were used to bond with acrylic bone cement (Simplex P) to create interface samples. The rectangular bone/foam coupons were machined to size using a low speed diamond saw with constant water irrigation. Each coupon was then placed into the lower half of a custom made mold and bone/foam-cement interface stripes were prepared by applying a constant pressure of 60kPa to the cement through the top of the mold following the procedure reported by [6]. Bone/foamcement specimens with dimensions of approx 10×16×5mm [2] were cut out from the stripes. Step-wise testing of bone/foam cement samples (n=2) in combination with time-lapsed microCT imaging was performed. A novel micro-mechanical loading stage (LS), equipped with a 3kN miniature load cell, was used to measure the loaded composite specimens directly into the microCT chamber. The samples were glued on the lower compressive platen inside the LS and a preload of 5N was applied through the top platen connected with the actuator. A complete CT acquisition was performed at a rotational step of 0.19o, voltage of 55kV, current of 145μA and a voxel size of 20μm. The specimens were then step-wise compressed at displacement levels of 0.3mm and 0.6mm (Fig.1). After each displacement step the acquisition procedure was repeated.