Metal-on-metal hip prostheses: where are we now?

In the late 1950s John Charnley, working in the north of England, introduced Teflon as a joint bearing surface material in his then visionary “low friction” hip joint replacement. Initial clinical success was soon followed by catastrophic mechanical failure and local tissue destruction owing to foreign body soft tissue reactions, and the use of Teflon in prostheses was abandoned by 1962.1 Fifty years later this cycle of promising innovation followed by unforeseen complications seems to be repeating itself with metal-on-metal hip prostheses that use large diameter (≥36 mm) bearings.

Charnley’s solution was to use high density polyethylene, and this is still a good choice of bearing material for older patients. However, in young patients with high physical demands, polyethylene prostheses wear out too quickly and often need to be revised. The introduction of hard wearing contemporary metal-on-metal bearings in the 1990s promised an end to prosthesis failure related to wear debris and better biomechanical function owing to the larger more natural diameter of the replaced joint. Initial enthusiasm led to about a million of these bearings being inserted worldwide during the past decade. However, reports of poor survival of prostheses, destructive local tissue reactions, and raised concentrations of cobalt and chromium metals in the blood of patients receiving these implants have now burst this bubble of optimism.

This episode has provoked reflection on the marketing authorisation process for new medical devices and raises several clinical questions. Does the clinical performance of these devices in most patients justify their role in the treatment of hip arthritis? What are the actual health risks of the technology? What forms of surveillance should we use to detect adverse effects, and how should we treat them?

The five year survival of large diameter metal-on-metal bearings placed above a standard femoral prosthesis has been substantially poorer than that of conventional hip replacement with a smaller femoral head component,2 and their use has been largely abandoned in the United Kingdom and elsewhere. The short term survivorship of metal-on-metal hip resurfacing, in which only the hip joint surface (and not the head) is replaced, is also poorer than for conventional hip replacement.3 However, the finding that the longevity of hip resurfacing in younger men with large (≥54 mm) hip joints is similar to conventional hip replacement for some brands of prosthesis suggests a place for this technology in selected patients.3 4 But short term evidence is no substitute for long term survivorship data, and ongoing use of this technology will ultimately require demonstration of long term prosthesis survival at least equivalent to conventional hip replacement in this patient group.5 6

A recent clinical trial found that the 12 month functional outcomes of hip resurfacing are similar to those of conventional hip replacement and do not support claims of better hip function,7 although the measurement tools used may have been subject to ceiling effects. Further randomised studies of longer duration—that target younger men and use outcome measures that are sensitive to performance differences at the higher end of the functional spectrum—are needed to clarify the efficacy of this intervention and inform models of its cost effectiveness.

What of exposure to metal debris? Toxicological studies and evidence from accidental exposure show that high concentrations of cobalt or chromium have genotoxic and other effects on multiple organ systems.8 The consequences of prolonged systemic exposure to the mildly raised concentrations of these metals in patients with a well functioning prosthesis remain unclear. The Food and Drug Administration in the United States has responded by instructing prosthesis manufacturers to conduct cross sectional studies up to eight years after implantation to quantitate the adverse health effects associated with metal exposure (FDA, May 2012). The methods chosen will need to be appropriate and robust enough to confidently exclude general health effects that are likely to be subtle over this short exposure time. Recent findings that the short term risk of cancer and all cause mortality do not seem to be higher in patients with metal-on-metal bearing surfaces than in those with other bearings may instil confidence.9 10 However, the anticipated long service life of these prostheses and the lead time for the development of disease mean that definitive answers will emerge only through ongoing surveillance.

The Medicines and Healthcare Products Regulatory Agency (MRHA) in the UK recommends measuring blood cobalt or chromium concentrations to detect the malfunction of prostheses in selected patient groups, followed by cross sectional imaging of the hip in patients with concentrations greater than 7 μg/L.11 However, the value of this screening tool is unproven because the sensitivity of the chosen threshold is low (52%).12 In addition, it is unclear how asymptomatic patients found to have raised metal concentrations or non-destructive lesions on imaging of the hip should be treated. The reliability of metal concentrations also needs to be assessed, given the lack of standardisation of collection and assay methods between laboratories. Furthermore, the cost of this additional device specific surveillance must be included in assessments of the cost effectiveness of this technology.

Metal-on-metal hip resurfacing may have a role to play in treating a subset of patients with high physical demands. However, investment in defining its long term outcomes and safety is needed before its cost effectiveness can be established. Meanwhile, other bearing materials continue to be developed and may provide an alternative solution to the problem of prosthesis wear and failure. These technologies will require similar scrutiny before we can be confident of their clinical value,13 as history has a way of repeating itself in the innovation of joint arthroplasty.