Metal-on-metal bearings for total hip arthroplasty have a long history of use that dates back to the introduction of the McKee-Farrar prosthesis in the 1960s. Metal-on-metal bearings ultimately fell out of favor because of the excellent clinical results of the Charnley low-friction arthroplasty concept, which employed metal-on-UHMWPE (ultra-high molecular weight polyethylene) bearing surfaces. However, the use of metal-on-metal bearings was revisited in the 1990s as polyethylene-associated periprosthetic osteolysis emerged as a major clinical problem limiting the longevity of hip reconstructions in high-demand patient populations.
Within the last decade, numerous metal-on-metal designs for both total hip arthroplasty and hip resurfacing arthroplasty have been introduced into the marketplace. However, over the last several years it has become evident that contemporary metal-on-metal bearings are not immune to some of the problems, including periprosthetic osteolysis, that have plagued other bearing surface combinations. In addition, investigators described a distinct histological pattern, termed “aseptic lymphocyte-dominated vasculitis-associated lesion” (ALVAL), that was associated with symptomatic metal-on-metal bearings. More recently, investigators described local soft-tissue reactions termed “pseudotumors” in association with symptomatic metal-on-metal devices. Initially, studies of adverse local tissue reactions (including osteolysis, ALVAL, and pseudotumors) in retrospective studies of metal-on-metal implants did not indicate the prevalence. However, case-control studies and registries began to document that certain metal-on-metal bearing designs were associated with an unacceptably high failure rate, leading to high-profile device recalls that captured the attention of the worldwide media. In some cases, the adverse local tissue response (ALTR) was found to result in extensive soft-tissue necrosis leading to compromise of the soft-tissue envelope about the hip, which increased the complexity of revision surgery and its complication rate1.
Some authors have stressed the importance of identifying ALTRs before they progress to the stage at which there is substantial compromise of the soft-tissue envelope about the hip1. Patients with an ALTR will characteristically complain of hip pain with or without swelling, occasionally accompanied by compression of adjacent neurovascular structures. Furthermore, studies of patients with implant failure due to an ALTR have suggested that such failures are more likely to occur in situations in which there is accelerated corrosion and/or wear of metallic surfaces, as evidenced by direct measurement of retrieved devices and/or by elevated levels of cobalt and chromium in the bloodstream. Thus, measurement of metal ion levels in serum or whole blood has been advocated in the evaluation of patients with symptomatic metal-on-metal bearings.
Imaging studies have also been advocated to identify patients with an ALTR; ultrasonography and metal artifact reduction sequence magnetic resonance imaging (MARS MRI) have been considered important screening and/or diagnostic tools. Therefore, in order to noninvasively identify individuals with an ALTR, the clinician must currently rely on patient symptomatology, imaging studies such as ultrasonography and MARS MRI, and metal concentrations in the serum or whole blood. Although this testing can provide useful information, clinicians have been struggling with determining the sensitivity and specificity of such tests performed to predict which ALTRs may lead to a substantial and clinically important compromise of the periprosthetic tissues.
The study by Hart et al. provides important information regarding the utility of imaging studies in the management of patients with metal-on-metal bearings. Most notably, this study shows that so-called cystic pseudotumors identified by MARS MRI were equally prevalent in patients with and patients without hip pain following metal-on-metal hip replacement. Furthermore, there was no correlation between the presence of a pseudotumor and the position of the acetabular component as determined with use of three-dimensional computed tomography (CT). This study calls into question the utility of MARS MRI in the management of patients with metal-on-metal bearings. It further calls into question the relationship between ALTRs, as determined with use of MARS MRI, and hip pain.
It should be pointed out that variability exists in the MARS technique. Although MRI performed with appropriate protocols has already been shown to be efficacious in assessing metal-on-polyethylene implants2, newer pulse sequences are being devised to markedly reduce artifacts and to enable an evaluation of the bone-implant interface that is not limited to just the areas of synovial expansion3. Attention to imaging detail is imperative in order to optimize soft-tissue contrast, to yield an accurate evaluation of the extent of synovial expansion, and to assess for subtle areas of periprosthetic bone loss4.
What is a clinician to do? Unfortunately, this study does not provide the guidance that we are seeking, and it may further complicate the management of patients with metal-on-metal bearings. For now, clinicians must employ their clinical judgment, integrating findings from the history and physical examination, radiographs, advanced imaging studies such as MARS MRI and ultrasonography, laboratory studies including metal ion levels, and the track record of the implant to guide clinical decision-making. What is clearly needed is natural history studies regarding the evolution of ALTRs and correlative studies that are able to distinguish particular MARS MRI features that may be associated with progressive soft-tissue lesions and/or clinical symptoms. For example, the authors do point out that the volume of the pseudotumor on MARS MRI was correlated with the presence of pain even though the simple presence of a pseudotumor was not. In addition, standardization and optimization of the MARS technique is urgently needed to maximize the clinical utility of MRI in this setting.
It would have been of interest to compare serum or whole blood metal levels in the two groups of patients to determine whether, alone or in concert with the MRI findings, they were correlated with clinical symptoms. Others have shown a correlation between metal levels and the presence of an ALTR5. It would also have been of interest to correlate the character of the local soft-tissue reaction with the pattern of damage on the metal bearing surfaces and modular connections of retrieved devices. One possible explanation for the observed lack of a relationship between component position as determined by CT and the presence of an ALTR is that a major source of debris is corrosion at the head-neck junction. Finally, it would have been of interest to characterize the prevalence of ALTRs documented with use of MARS MRI in symptomatic and asymptomatic patients with contemporary metal-on-polyethylene devices.
Clearly, additional research is needed in order to guide the clinician in defining the appropriate indications for revision of arthroplasties with metal-on-metal bearings. In particular, the natural history of the asymptomatic ALTRs documented with MARS MRI needs to be elucidated.