Elsevier

Bone

Volume 42, Issue 2, February 2008, Pages 358-364
Bone

Microstimulation at the bone–implant interface upregulates osteoclast activation pathways

https://doi.org/10.1016/j.bone.2007.09.055Get rights and content

Abstract

Peri-implant bone resorption after total joint arthroplasty is a key parameter in aseptic loosening. Implant wear debris and biomechanical aspects have both been demonstrated to be part of the bone resorption process. However, neither of these two parameters has been clearly identified as the primary initiator of peri-implant bone resorption. For the biomechanical parameters, micromotions were measured at the bone–implant interface during normal gait cycles. The amplitude of the micromotions was shown to trigger differentiation of bone tissues. So far no data exists directly quantifying the effect of micromotion and compression on human bone. We hypothesize that micromotion and compression at the bone–implant interface may induce direct activation of bone resorption around the implant through osteoblasts–osteoclasts cell signaling in human bone. This hypothesis was tested with an ex vivo loading system developed to stimulate trabecular bone cores and mimic the micromotions arising at the bone–implant interface. Gene expression of RANKL, OPG, TGFB2, IFNG and CSF-1 was analyzed after no mechanical stimulation (control), exposure to compression or exposure to micromotions. We observed an 8-fold upregulation of RANKL after exposure to micromotions, and downregulation of OPG, IFNG and TGFB2. The RANKL:OPG ratio was upregulated 24-fold after micromotions. This suggests that the micromotions arising at the bone–implant interface during normal gait cycles induce a bone resorption response after only 1 h, which occurs before any wear debris particles enter the system.

Introduction

After total joint arthroplasty, a radiolucent zone is frequently observed at the interface between bone and implant [4], [29]. This radiolucent zone is associated with a progressive peri-implant bone resorption. The implant fixation is affected, therefore inducing a risk of aseptic loosening. This becomes a serious problem as aseptic loosening accounts for more than two-third of hip revisions in Sweden, a country where an extensive implant register has been set up for many years [16].

Two hypotheses are generally used to explain peri-implant bone resorption. The first hypothesis focuses on a biological reaction to wear particles. Numerous studies have shown that the debris after implant wear induces inflammatory reactions in the tissues surrounding the implant [2], [7]. Bone formation may also be impaired by the presence of particles, such as titanium debris which were shown to induce apoptosis to osteoblasts culture in vitro[24]. In all cases, particulate debris accumulate in the tissue surrounding the implant. Upon accumulation, a chain of cellular events is triggered within the tissue leading to periprosthetic osteolysis and implant loosening [15]. The second hypothesis used to explain peri-implant bone resorption is based on biomechanical considerations. A stiff metallic implant in a load bearing bone considerably changes the mechanical state of the bone. Based on a numerical approach, Huiskes and Nunamaker showed that bone resorption around the implant is associated with high peak stresses immediately postoperatively [8]. The bone structure is affected by the new stress patterns around an implant. Numerical models predict bone loss around the implant based only on these modified mechanical patterns [33].

Mechanical effects play certainly also a crucial role at the bone–implant interface. The pumping action of the implant during gait cycle causes load fluctuation within the hip joint fluid. Using a numerical model, the interfacial compressive stress involved was found to be between 2 × 10 3 and 0.1 MPa [27]. Beside compressive stress, micromotions at the interface have been suspected to play a key role in tissue differentiation around the implant. It has been calculated that micromotions between 5 and 100 μm occur at the bone–hip implant interface during normal gait cycles [27]. Mandell et al. analyzed the case of conical-collared intramedullary hip stem and reported micromotions up to 163 μm in the worst design [17]. From an experimental point of view, Baleani et al. quantified bone–implant micromotions under torsional load with position transducers in a hip implant model. A maximum of 56 μm was measured in uncemented stems [1]. Finally, with in vivo models, Jasty et al. found that micromotions lower than 40 μm favor bone formation in dog, while micromotions higher than 100 μm lead to the creation of a fibrous tissue [10].

Based on clinical observations, numerical and experimental biomechanical analysis and in vivo experiments, there is strong evidence to support the theory that micromotions and compressive stress at the bone–implant interface play an important role in the process of peri-implant bone resorption. However, so far no data exist quantifying directly the effect of micromotion and compression on human bone. Therefore we hypothesize that micromotion and compression at the bone–implant interface may induce direct activation of bone resorption around the implant through osteoblasts–osteoclasts cells signaling in human bone. This hypothesis was tested with an ex vivo loading system using human bone samples.

Section snippets

Bone samples preparation

Twenty-five human femoral heads were obtained from the Hôpital Orthopédique de la Suisse Romande following total hip prosthesis procedures (Ethical Protocol 51/01, University of Lausanne). In the next 4 h following the sample collection, each femoral head was fixed axially in a custom fixation device and a central section of 6 mm thick slice was extracted with a surgical saw. Then, 4 to 16 trabecular bone cores of 3 mm radius and 6 mm height were extracted from the slice with a biopsy puncher

Results

In the following results, we report normalized gene expressions as mean ± standard error of the mean (SEM).

Discussion

A clinical study showed that up to 14% bone loss arose during the first 3 months after total hip arthroplasties [34]. In parallel, rapid early migrations have been detected by roentgen stereophotogrammetry in many asymptotic hips, often as early as 4 months postoperatively [11], [18]. The migrations have been found to predict an increased risk of clinical loosening. The fate of an orthopedic implant seems then to be determined at an early stage, probably before any wear particles are produced,

Acknowledgments

Project no. 04-P2 was supported by the AO Research Fund of the AO Foundation, Davos, Switzerland. We thank Marc-Olivier Montjovent and Sandra Jaccoud for technical assistance and Tyler Thacher for English editing.

References (35)

  • N.L. Fazzalari et al.

    The ratio of messenger RNA levels of receptor activator of nuclear factor kappaB ligand to osteoprotegerin correlates with bone remodeling indices in normal human cancellous bone but not in osteoarthritis

    J Bone Miner Res

    (2001)
  • M.A.R. Freeman et al.

    Observations upon the interface between bone and polymethylmethacrylate cement

    J Bone Joint Surg Br

    (1982)
  • K. Hirose et al.

    Recombinant interferon-gamma is a potent inhibitor of osteoblastic cell functions

    Meikai Daigaku Shigaku Zasshi

    (1989)
  • M. Horikoshi et al.

    Comparison of interface membranes obtained from failed cemented and cementless hip and knee prostheses

    Clin Orthop Relat Res

    (1994)
  • R. Huiskes et al.

    Local stresses and bone adaption around orthopedic implants

    Calcif Tissue Int

    (1984)
  • K. Janssens et al.

    Transforming growth factor-beta1 to the bone

    Endocr Rev

    (2005)
  • M. Jasty et al.

    In vivo skeletal responses to porous-surfaced implants subjected to small induced motions

    J Bone Joint Surg Am

    (1997)
  • Cited by (0)

    View full text