Loosening and Osteolysis Associated with Metal-on-Metal Bearings: A Local Effect of Metal Hypersensitivity?

Over the last several years, many orthopaedic surgeons have embraced so-called alternative bearing surfaces in total hip arthroplasty, largely on the basis of the hypothesis that a reduction in the volumetric wear rates afforded by these improved articulating couples will reduce the prevalence of osteolysis and aseptic loosening, which in turn will improve implant survivorship. The articles by Korovessis et al.1 and Milosev et al.2 in this issue of The Journal provide valuable new data on the intermediate-term performance of one metal-on-metal-bearing hip-replacement system. These studies show that this particular metal-on-metal system is not immune to osteolysis and aseptic loosening and further suggest that osteolysis and aseptic loosening may be mediated, at least in part, by an adaptive immune response (metal hypersensitivity) independent of, or in concert with, the relatively well-characterized innate immune response (particle-stimulated macrophage, fibroblast, and osteoblast-mediated inflammatory bone loss) typically associated with periprosthetic osteolysis3. Metal hypersensitivity-induced osteolysis and aseptic loosening represent underappreciated and incompletely understood mechanisms of implant failure.

How strong is the evidence that the failures in these two cohorts at the time of intermediate-term follow-up (at a mean of seventy-seven months and eighty-five months) were due to metal hypersensitivity? It should be clearly understood that, at present, the evidence linking osteolysis and aseptic loosening with metal hypersensitivity is circumstantial: cause and effect have not been established. In fact, a similar association was postulated three decades ago, in the era of the early-generation metal-on-metal bearings4,5. The issue has yet to be resolved. Part of the difficulty in establishing cause and effect is the lack of a gold standard for the diagnosis of metal hypersensitivity. The histological findings of a delayed-type hypersensitivity-like reaction (termed aseptic lymphocyte-dominated vasculitis-associated lesion by Willert et al.6) are suggestive, but not pathognomonic, of a metal hypersensitivity response. Furthermore, metal allergy testing is an evolving science. While patch testing is a reliable method for evaluating contact dermatitis, its utility for testing for deep-tissue metal allergy is suspect7. In vitro testing, such as lymphocyte transformation testing, has its advantages, but clinical validation of the accuracy of these tests is lacking. Nonetheless, there have now been multiple reports that the histopathological characteristics associated with failed metal-on-metal bearings are distinct from those associated with other bearing-surface combinations1,2,6,8. In addition, there is increasing evidence that hypersensitivity may play a role in the pathogenesis of osteolysis9.

To what extent can the findings reported in this issue of The Journal be generalized to other metal-on-metal bearing systems? This is an important consideration since there are numerous differences among the various metal-on-metal bearing couples available for clinical use. These differences involve the metallurgy (carbon content, method of fabrication, and heat treatment), geometry (clearance, sphericity, surface finish, and head size), acetabular component design (monoblock acetabulum compared with modular acetabulum with or without an interpositional layer of polyethylene), fixation surface (threaded or porous socket), and femoral component design. Since the causes of osteolysis and aseptic loosening are multifactorial, any of these differences theoretically may play a role in the differential prevalence of these entities. While the role of each of these factors can be definitively ascertained only in comparative clinical trials, the phenomenon of early osteolysis and loosening associated with a distinctive histological pattern (aseptic lymphocyte-dominated vasculitis-associated lesion) has been reported in association with multiple metal-on-metal systems1,2,6,8,10. It is premature to attribute osteolysis and aseptic loosening to one characteristic (such as carbon content) or to conclude that all metal-on-metal bearing systems are equally susceptible to these phenomena. Willert et al. concluded that the prevalence of clinically important metal hypersensitivity reactions in patients with metal-on-metal bearings “appears to be low,” on the basis of their retrieval study6. However, the true prevalence is unknown. Long-term high-quality clinical studies are required to address this question.

Why is the prevalence of osteolysis associated with the metal-on-metal bearing system reported on in this issue of The Journal seemingly similar to that associated with conventional bearing couples, such as those consisting of metal on polyethylene, despite the dramatically lower volumetric wear rate? Even though the volumetric wear rate associated with metal-on-metal bearings is dramatically lower than that associated with metal-on-polyethylene bearings, the number of particles generated from metal-on-metal bearings is actually higher because the particles are an order of magnitude smaller11. These particles have a high specific surface area, facilitating the dissolution (corrosion) of the metal constituents releasing metal ions into the surrounding tissues. This suggests that the volumetric wear rate alone does not dictate the prevalence of osteolysis; other characteristics of the degradation products (such as size, shape, and chemical form) are important in the determination of bioreactivity. Implant hypersensitivity reactions (metal sensitivity or metal allergy) are generally associated with Type-IV, delayed-type hypersensitivity, a lymphocyte-mediated response. It is likely that metal-stimulated lymphocytes participate in the pathogenesis of aseptic osteolysis given that activated lymphocytes release powerful cytokines such as IL-2 (interleukin-2), IFN-γ (interferon-gamma), and RANKL (receptor activator of NF-κB ligand), which can directly and indirectly promote osteoclast activity and inhibit osteoblast activity12. Alternatively, there may be specific design, geometric, and/or manufacturing parameters that contribute to the failures, independent of the composition of the bearing couple.

These studies1,2 demonstrate that contemporary alternative bearing surfaces are not a panacea and that detailed comparative clinical studies are required to determine the factors associated with implant performance and survivorship. Furthermore, the role of metal hypersensitivity in implant performance needs to be further defined in studies of metal-on-metal bearings, with an emphasis on the potential for preoperative screening and postoperative noninvasive diagnosis.