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

Types of implant

Over 65 000 metal-on-metal prostheses were implanted in the UK, and, although most continue to function well, more than 17 000 patients have ongoing symptoms and a large number have required revision surgery.3 The best way to monitor and advise patients is still debated. Not all metal-on-metal implants are the same, and essentially there are two classes: non-stemmed surface replacements, which have metal bearing surfaces that can produce wear particles from the articulation, and stemmed replacements, which have a metal taper junction between head and stem and a metal bearing surface. In stemmed replacements metal debris can arise both from wear at the articulation and from wear and corrosion at the taper junction.

Both classes were developed for younger, more active patients, to improve longevity compared with conventional replacements. Some selected surface replacements with larger heads (>48 mm) are still used in men as they facilitate high levels of activity and have good outcomes for long periods, but smaller head sizes have largely been discontinued. By contrast larger head sizes (>36 mm) in stemmed replacements are at greater risk of complications, and their use has been abandoned worldwide.

Within these two broad classes, some designs of implant have greater risk of causing adverse tissue response—for example, DePuy’s articular surface replacement.4 Women have a significantly higher risk of soft tissue responses than men. Although case reports exist of patients with metal-on-metal implants developing cardiac failure and cancer, a direct causative link to systemic complications has not been established and seems unlikely from the evidence to date.5 Indeed, other evidence suggests that matched patients with metal-on-metal replacements have reduced mortality at 10 years compared with those receiving conventional replacements.6

Although plain radiography can identify some abnormalities, metal artefact reduction series (MARS) magnetic resonance imaging (MRI) or ultrasonography is needed to detect early changes. Patients are likely to achieve better outcomes after revision surgery if the soft tissue changes are less advanced. High circulating metal ion concentrations are associated with a greater likelihood of adverse soft tissue response, but the absolute level associated with risk is not established.

Regulatory changes

Arguably, this problem has arisen partly because the regulatory approval pathways did not require evidence of safety and effectiveness before metal-on-metal hips were marketed.7 Regulatory changes have subsequently been made to the introduction and monitoring of these implants. The US Food and Drug Administration issued a safety communication in 2013 recommending regular clinical evaluation of patients with symptoms and consideration of soft tissue imaging and metal ion testing.8 The FDA also issued a final order requiring evidence of clinical outcomes to be submitted by 18 May 2016 if a manufacturer wanted to continue marketing its metal-on-metal total hip replacement or market a new metal-on-metal implant.9

In 2012, the MHRA issued a medical device alert that recommended annual clinical review, MARS MRI or ultrasound imaging, and assessment of metal ion concentrations in all symptomatic patients with high risk implants. For asymptomatic patients at high risk, clinical and blood metal ion testing was recommended, with imaging reserved for patients with those with rising metal ions concentrations. Asymptomatic patients with low risk implants did not require regular follow-up.10

More stringent follow-up

The recent MHRA update introduces recommendations for much more stringent follow-up.1 This includes annual use of patient reported outcome measures (the Oxford hip score) to assess pain and function; annual universal blood ion testing regardless of patient symptoms; and imaging with plain radiography, MARS MRI, or ultrasonography. The evidence surrounding the level of increase of metal ions and deterioration in the Oxford hip score required to trigger a change in management is unclear.

The MHRA’s imaging recommendations seem inconsistent, particularly for symptomatic patients: all those with lower risk implants have MRI or ultrasonography whereas patients with higher risk implants do so only if the blood ion concentrations and the Oxford hip score are worsening.

It remains to be seen whether this substantial expansion of patient surveillance will identify soft tissue complications at an earlier stage, so that outcomes can be improved. Regular assessment of asymptomatic patients will require careful handling of patients’ concerns and expectations. The new requirement for review of more than 45 000 asymptomatic patients will also be costly.11 There is a strong case for manufacturers bearing some, if not all, of this burden.