Practice Practice Pointer

Diagnosing and investigating adverse reactions in metal on metal hip implants

BMJ 2011; 343 doi: (Published 29 November 2011) Cite this as: BMJ 2011;343:d7441
  1. Camdon Fary, clinical fellow1, consultant orthopaedic surgeon2,
  2. Geraint Emyr Rhys Thomas, clinical research fellow3,
  3. Adrian Taylor, consultant orthopaedic surgeon 1, honorary senior clinical lecturer3,
  4. David Beard, professor of musculoskeletal sciences3,
  5. Andrew Carr, Nuffield professor of orthopaedic surgery13,
  6. Sion Glyn-Jones, consultant orthopaedic surgeon1, clinical senior lecturer3
  1. 1Nuffield Orthopaedic Centre, Oxford OX3 7LD, UK
  2. 2Royal Melbourne and Western Hospitals, Melbourne, Victoria, Australia
  3. 3Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, NIHR Biomedical Research Unit into Musculoskeletal Disease, University of Oxford, Nuffield Orthopaedic Centre, Oxford, OX3 7LD, UK
  1. Correspondence to: C Fary, Western Hospital, Footscray, Victoria, Australia camfary{at}
  • Accepted 11 October 2011

Use of metal bearings in hip replacements and resurfacing operations has led to adverse reactions to metal debris in many patients. This article examines how to assess and manage any patient with a metal on metal total hip replacement who presents with potential joint failure

Summary points

  • Large diameter, metal on metal bearings may be used in hip replacement and hip resurfacing operations

  • Early evidence suggested that metal on metal implants were more resistant to wear and tear and dislocation; consequently they were often used in younger and more active patients

  • Adverse reactions to metal debris can result in the formation of a mass lesion that can be locally invasive around the joint. Such lesions may be difficult to diagnose as many are silent

  • Symptomatic patients may complain of discomfort in the hip or buttock; a swelling in the hip region; pain in the hip; or clicking or “giving way” of the hip

  • Ultrasound scanning by a skilled operator, followed by magnetic resonance imaging when a lesion is detected, will lead to a diagnosis in most cases

  • Revision surgery is usually indicated. Revision after adverse reaction to metal debris has a poorer outcome than revision for other indications

Promising five year results in 2005 for large diameter “metal on metal” hip replacement surgery1 led to a rapid increase in the number of surgeons performing this procedure and in orthopaedic companies competing to produce their own resurfacing designs. Adverse reactions to metal debris were not anticipated or predicted.

Over 250?000 large diameter, metal on metal, articulating total hip replacements (most of which are resurfacing procedures) have now been performed worldwide.2 UK and Australian national joint registries have reported statistically significant increased revision rates for specific metal on metal total hip replacements compared with conventional metal on plastic total hip replacements.3 4 5 Adverse soft tissue reactions to the debris released by metal on metal implants (metal wear debris) may result in the early failure of these implants and a need for revision surgery. Although adverse reactions to such debris can present with few symptoms, lesions may be highly locally destructive, which makes revision surgery challenging.

Recently a specific metal on metal implant was recalled internationally, which has resulted in a review of the evaluation, introduction, and regulation of new medical devices (BMJ 2011;342:d2905, doi:10.1136/bmj.d2905). However, some patients with failed implants may not yet have presented to practitioners with silent or symptomatic adverse reactions to devices, and the incidence of failed metal on metal hip replacements is likely to be cumulative. This may become an important health and economic burden on patients and commissioners of healthcare respectively.

We review the assessment and management of any patient with a metal on metal total hip replacement who presents with potential joint failure. We draw from recent guidelines, the evidence in the literature (which is limited to case-control, cohort, and case studies), and our own experience.

What is a metal on metal hip replacement?

A metal on metal hip replacement refers to the type of articulating bearing surface that was designed as an alternative to conventional metal on polyethylene bearings. The metal is specially hardened and composed of cobalt chrome alloy.

The metal on metal articulating bearing is used in two distinctly different femoral prosthetic designs: as a large metal femoral head on a standard femoral stem or as a large metal femoral head cemented on to the native bone of the femoral neck (hip resurfacing).

Why have metal on metal articulating surfaces been used?

The metal on metal bearing surface has been used to treat younger and more active patients, with excellent results reported in some case series.6 Reported benefits of hip resurfacing include lower theoretical rate of dislocation associated with large diameter heads than with conventional metal on polyethylene total hip replacement; low wear; bone conservation; physiological femoral loading; and improved restoration of joint mechanics. Case series have suggested that revising a hip resurfacing procedure to a standard total hip replacement conserves more bone than a revision of a conventional hip replacement.7

What are the problems associated with metal on metal articulating surfaces?

Inflammatory changes

Observed soft tissue inflammatory reactions to metal wear debris have been called inflammatory pseudotumour,8 adverse reaction to metal debris,9 aseptic lymphocytic vasculitis associated lesion,10 and metallosis.11 Inflammatory pseudotumour is a clinical description of a mass (hence tumour) that may be cystic or solid (hence pseudo) in association with clinical, radiological, and histolopathological findings consistent with inflammation.

Subtle differences between different prosthetic designs are believed to play a role in increased wear and subsequent metal debris. One particular hip resurfacing device, the articular surface replacement (ASR) hip resurfacing device, has been shown in a large multicentre cohort study to have a much higher rate of failure attributed to adverse reactions to metal debris than other resurfacing devices.9 The ASR hip resurfacing device has a shallower acetabular component, resulting in greater edge loading and wear than with other designs.


The findings of a case series suggest that a spectrum of necrotic and inflammatory changes occur in response to the deposition of particles released by cobalt-chrome prosthetic wear into the periprosthetic tissues.12 The pathogenesis of these changes is uncertain but may involve both a cytotoxic response and a delayed hypersensitivity (type IV) response to cobalt-chrome particles.[10 ]Adverse reactions can be caused by large amounts of wear debris but have also been shown to occur around implants with small amounts of wear debris.9 Retrieval studies of failed components show that most failures of components are the result of increased wear, suggesting a toxic aetiology. A smaller number of failures may result from hypersensitivity reaction with normal wear.9

How common are adverse reactions?

The true incidence of adverse reactions to metal wear debris is not known as they are difficult to diagnose and are therefore probably under-reported. Four per cent of masses are thought to be asymptomatic initially,13 and magnetic resonance imaging studies have shown that “silent” pathology exists (25% of patients with a best possible Oxford hip score had evidence of adverse reactions to metal debris confirmed by magnetic resonance imaging).14

Designer series (series performed by surgeons involved in the design process) suggest a cumulative revision rate of <0.5% at five years after implantation.1 However, independent series suggest a higher rate of between 1% and 9.8% depending on the device studied.9 A cohort study using a Cox’s proportional hazards model has shown that female patients and age <40 years increased the risk of revision for adverse reaction to metal debris by eight and a three times respectively.8

In 2011 the National Joint Registry for England and Wales reported that the seven year revision rate for any cause of primary total hip replacement (285?600) regardless of prosthesis type was 4.7%.3 Within this group the worst performers were resurfacing procedures and large head, metal on metal total hip replacements (revision rates of 11.8% and 13.6% respectively). Metal on metal articulations were more often revised in women than in men. Device brand also seemed to be important; the five year revision rate for the most commonly inserted hip replacement (the Birmingham hip replacement) was 3.4% compared with 9.6% for the ASR hip resurfacing device and 11.3% for the ASR total hip replacement. The Australian Joint Registry, too, has flagged the ASR hip resurfacing device and the ASR total hip replacement as having higher than anticipated rates of revision.

The cumulative revision rate for adverse reactions may increase progressively with time.8 The incidence of adverse reaction to metal debris may be greater than that predicted by examining the revision rate.

How might patients with adverse reactions to metal wear debris present?

Patients may complain of discomfort in the groin, the lateral aspect of the hip, or the buttock and may report the sensation of a mass or a visible swelling around the hip.8 Less commonly, often after a period of discomfort, more obvious signs such as late dislocation and instability occur; patients report clicking or clunking of the hip, or say that the hip “gives way.” According to a cohort study of 1224 patients, the mean time to presentation from the primary replacement is 17 months.8

Rarely patients present with symptoms and signs of serious adverse events, such as femoral nerve and arterial damage.15 Patients with nerve involvement complain of progressive thigh pain, weakness, and numbness, usually caused by a direct pressure effect from the inflammatory mass or associated fluid.10 Pressure from the mass on the femoral artery results in stenosis and fibrosis of the artery; patients present with symptoms of intermittent claudication (pain in the legs that increases with activity and is relieved by rest) combined with femoral neurological symptoms.

How to investigate patients with a suspected inflammatory pseudotumour

Any patient who has a symptomatic hip resurfacing or large diameter metal on metal total hip replacement and presents with any of the symptoms and signs mentioned above should be referred to their orthopaedic specialist without delay. Given that adverse reactions to metal debris may be more common than previously thought, being vigilant when examining a patient who has had a hip arthroplasty and referring early are wise precautions. We have compiled an algorithm for follow-up (box) based on advice from the British Orthopaedic Association and the Medicines and Healthcare Products Regulatory Agency.16

Follow-up of patients implanted with metal on metal hip replacements*

  • Follow up patients at least annually for five years postoperatively and more frequently in the presence of symptoms. Beyond five years, follow up patients in accordance with locally agreed protocols

  • Investigate patients with painful metal on metal hip replacements. Specific tests should include evaluation of cobalt and chromium ion concentrations in the patient’s blood and cross sectional imaging, including magnetic resonance imaging or ultrasound scanning

  • Consider measuring cobalt and chromium ion concentrations in the blood and/or cross sectional imaging for the following patient groups:

    • -Patients with radiological features associated with adverse outcomes including (a) component position and (b) small component size (hip resurfacing only)

    • -Cases where the patient or surgeon is concerned about the metal on metal hip replacement

    • -Cohorts of patients where there is concern about higher than expected rates of failure

  • If cobalt or chromium ion concentrations are raised above seven parts per billion (7 µg/L), then conduct a second test three months after the first to identify patients who need closer surveillance, which may include cross sectional imaging

  • If imaging shows soft tissue reactions, fluid collections, or tissue masses then consider revision surgery

  • *Based on advice from the British Orthopaedic Association and the Medicines and Healthcare Products Regulatory Agency (MHRA).16

Excluding other diagnoses

Investigations for other common causes for failure of hip prostheses—such as periprosthetic fracture (usually of the femoral neck in resurfacing procedures), infection, component loosening, and acute dislocation—must form part of any investigatory process by a generalist clinician or emergency doctor. Exclude pain referred from another source (such as the spine or pelvis).

An adequate anteroposterior radiographic image of the pelvis and the lateral hip will show a periprosthetic fracture and an acute dislocation. Viewing sequential radiographs from implantation is the best way of detecting loosening and migration of components and is best done as part of an orthopaedic review.

Exclusion of infection as a cause of failure is important as patients with adverse reactions to metal debris rarely present with symptoms or signs of systemic infection and routine investigations for markers of infection are usually normal. We recommend full blood examination, erythrocyte sedimentation rate, and C reactive protein as baseline investigations.

What tests might confirm an adverse reaction to metal debris?

Are metal ion concentrations useful?

A cohort study has shown that plasma chromium and cobalt metal ion concentrations in patients with well functioning metal on metal implants are not usually close to toxic levels,17 and in a prospective, controlled, longitudinal study concentrations are similar to those in metal on polyethylene implants.18 Metal ion concentrations of at least an order of magnitude higher than those measured in the serum have been found in the joint fluid and surrounding area of a poorly functioning metal on metal bearing.19 A safe level of metal ions is difficult to define for metal on metal prostheses. The 33 year follow-up of first generation metal on metal total hip replacement showed acceptable serum cobalt concentrations but had five patients in the final follow-up.20 Extrapolation with current metal on metal prostheses is not possible.

The Medicines and Healthcare Products Regulatory Agency has advised on safe plasma metal ion concentrations, although these are underpinned only by weak evidence. The agency advises that patients with serum concentrations of cobalt or chromium ions above 7 µg/L should be further investigated and ion measurements repeated as part of closer follow-up; this advice was derived from a study in which the cut-off concentration of 7 µg/L for either metal ion was found to have a 90% specificity but only a 50% sensitivity to predict hip failure.21 Other studies using 4 µg/L as a cut-off increased the detection of cases of adverse reaction to debris from 75% to 90%.22 It is not clear, however, whether patients with “high” plasma metal ion concentrations should have a revision of their hip replacement.10 Metal ion concentrations may be an indicator of the amount of wear that is taking place. But a high serum concentration does not always indicate that a pseudotumour is forming.10Interpretation of plasma metal ion concentrations is difficult and is best done as part of an orthopaedic assessment.

Metal ions are mainly excreted via the renal route. Urine metal ion concentrations fluctuate depending on many factors and are not reliable indicators of serum metal ion concentrations nor are they to be used to indicate or rule out the presence of an adverse reaction.

Is imaging useful?

Plain radiography

Plain radiographs are of little use in screening during the early stages of adverse reactions to metal debris23 as they typically appear normal and may therefore falsely reassure. Only clinically advanced lesions complicated by osteolysis and/or soft tissue reaction may be visible on plain radiographs.

Ultrasound scanning

At our institution all adverse reactions to metal debris were detected initially by screening with ultrasound.21 This is our preferred initial screening tool. The anterior, lateral, and posterior aspects of the joint can be carefully examined for masses or fluid filled cysts that connect with the joint. However, ultrasound scanning is user dependent and requires an experienced musculoskeletal radiologist to interpret the images. Smaller masses and collections are easily missed.

Computed tomography

Computed tomography is better than plain radiography in evaluating bone stock around a hip replacement and can detect an inflammatory pseudotumour.21 However, it is not suitable as a screening tool and osteolysis is often a late presentation.

Magnetic resonance imaging

Metal artefact reduction sequences have considerably improved the resolution of magnetic resonance imaging in patients with hip implants. Small volumes of tissue or fluid can now be measured and followed with sequential scanning if suspicious, particularly if the patient is asymptomatic.14 After a symptomatic clinical presentation and ultrasound findings suggestive of a mass, magnetic resonance imaging is a suitable tool for making a definitive diagnosis of a mass resulting from an adverse reaction to metal debris. We rely on this approach in our practice.

Is revision surgery always indicated?

The natural course of adverse reactions to metal debris is unclear, but two case series have suggested the likelihood of progression; both report that revision was required because symptoms deteriorated after an initial period of conservative management.24 25

What are the outcomes after revision surgery?

A case-control study found that outcomes (measured by the Oxford hip score) for patients who had revision of hip resurfacing for adverse reaction to metal debris were significantly worse than for patients who had revisions for fractures or other reasons.15 The rate of complications after revisions for the former was also higher. The same study, using conventional primary total hip replacements as controls, found that the outcome for patients who had revisions of resurfacing for adverse reaction to metal debris was also significantly worse than that of age and sex matched controls, whereas the outcome for patients who had revisions for fracture or other causes was not significantly different from that of their control group. Outcome may be influenced by the degree of tissue destruction at the time of surgery. Small case series have reported good results in patients who had early revision surgery in the absence of soft tissue destruction.24 25 We consider it important to diagnose adverse reactions to metal debris and consider revision surgery early to limit the extent of soft tissue destruction and osteolysis, especially as metal bearings are traditionally used in younger and more active patients.


Fig 1 A: Metal on metal hip resurfacing device. B: Anteroposterior radiograph showing hip resurfacing device. C: Large head, metal on metal bearing on a standard uncemented femoral stem. D: Anteroposterior radiograph showing large head, metal on metal total hip replacement


Fig 2 A: Coronal T2 weighted magnetic resonance image showing large fluid collection adjacent to the left hip (indicated by arrow; fluid appears white), suggesting adverse reaction to metal debris. B: In the same patient, after initial incision and incision of fascia lata only, spontaneous release of the fluid collection (brown). C: In the same patient, the image shows the dislocated prosthetic femoral head attached to the remaining 50% of the bony femoral neck, with necrotic tissue inferiorly (black arrow) and osteolysis of most of the greater trochanter and its muscular attachments superiorly (white arrow)


Cite this as: BMJ 2011;343:d7441


  • Contributors: SG-J had the idea for the article. CF wrote the initial draft and is the guarantor. CF, GERT, AT, and SG-J obtained, reviewed, and discussed relevant literature and contributed to revisions of the manuscript. All authors contributed to the final version.

  • Competing interests: All authors have completed the ICMJE uniform disclosure form at (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Not commissioned; externally peer reviewed.


View Abstract