View commentaries on this research

This is a plain English summary of an original research article

Large data sets from national registries and case series show that about 82% of total knee replacements and 70% of partial (unicompartmental) knee replacements last for 25 years.

This NIHR-funded review of that data is the first to demonstrate longer-term effectiveness of these procedures. This means that the operation, which aims to relieve pain for osteoarthritis, may not need to be delayed to reduce the chance of a patient needing further surgery at a later date. It is an encouraging finding for an ageing population.

This study didn’t include data from the UK, as the National Joint Register of England, Wales, Northern Ireland and the Isle of Man didn’t go back far enough for this analysis. However, there is no reason to suspect that UK data, which is being collected, would be much different.

Why was this study needed?

Knee replacement surgery is a routine operation to replace a worn or damaged knee joint with an artificial one. The most common reason for needing surgery is osteoarthritis. Total knee replacement is where both sides of the knee joint are replaced. Unicompartmental knee replacement involves just one side of the knee joint being replaced. A variety of components and methods of fixing can be used.

In 2017 there were 106,334 knee replacement procedures carried out in England, Wales and Northern Ireland. The average age of patients was 69 years. There were also 6,502 knee revision procedures, where people needed a second operation. Further surgery on elderly patients should be avoided if possible, so it is important to know how long a knee replacement can be expected to last. Research to date only gives estimates for 15 and 20 years. This study looked at longer-term data.

What did this study do?

This study was made up of a systematic review and meta-analysis of published case series, and a meta-analysis of data from national joint registries. The authors found 33 case series, with a total of 7,232 knee replacements. Twenty-six case series looked at total replacement, while seven reported survival of unicompartmental replacement. Data from the Australian and Finnish national registries were used, with a total of 299,291 total replacements, and 7,714 unicompartmental.

The quality of the published case series was low. And only the Finnish registry had outcomes data at 20 and 25 years.

What did it find?

Data from the national registries provided the most reliable results:

  • For total knee replacements, 93.0% were still intact at 15 years (95% CI 92.8 to 93.1; Australian and Finnish registries with 299,291 replacements). At 20 years, 90.1% of replacements had survived (95% CI 89.7 to 90.4; Finnish registry with 88,532 replacements). 82.3% of replacements were still intact at 25 years (95% CI 81.3 to 83.2; Finnish registry with 76,651 replacements).
  • For unicompartmental replacements, 76.5% were still intact at 15 years (95% CI 75.2 to 77.7; Finnish registry with 7,714 procedures). At 20 years, 71.6% had survived (95% CI 69.6 to 73.6; Finnish registry with 3,935 replacements). 69.8% were still intact at 25 years (95% CI 67.6 to 72.1; Finnish registry with 3,935 replacements).

What does current guidance say on this issue?

The British Orthopaedic Association and Royal College of Surgeons published a commissioning guide for painful osteoarthritis of the knee in 2017.

It says that total knee replacements should only be considered for patients whose pain has not been adequately controlled by three months of non-surgical management (based on NICE guidance for managing osteoarthritis).

It states that joint survival is 95% at seven years. It also says that unicompartmental knee replacement has a lower joint survival than total replacement.

What are the implications?

This study provides evidence that knee replacement surgery has better long-term outcomes than was previously thought. The average age of someone in the UK having a total replacement is around 70 years. The procedure is often delayed, to reduce the chance of needing revision surgery at a later date.

This study shows that at least 80% of total replacements last for 25 years. This could mean that people are offered surgery younger.

Ideally, outcome data from a UK registry could include reporting on the situation here along with functional outcomes and satisfaction. This study was unable to look at problems that were not severe enough to lead to revision surgery.

Citation and Funding

Evans JT, Walker R, Evans W et al. How long does a knee replacement last? A systematic review and meta-analysis of case series and national registry reports with more than 15 years of follow-up. Lancet. 2019;393:655-63.

This study was supported by the NIHR Biomedical Research Centre at the University Hospitals Bristol NHS Foundation Trust and the University of Bristol.


BOA and RCS. Commissioning guide: painful osteoarthritis of the knee. London: Royal College of Surgeons; 2017.

NHS website. Knee replacement. London: Department of Health and Social Care; 2016.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre

  • Share via:
  • Print article


Expert commentary

This article makes for a very good headline. It brings together data from various sources and lets commissioners and patients know that knee replacement is a very durable intervention.

However, the data behind it lacks depth, and the key outcomes of function and satisfaction are not captured. The authors have extrapolated from country to country and to a greater timespan than that for which they have robust data. The main issue to be wary of is that their endpoint is revision. An unhappy patient with an unsuccessful operation who does not have a further knee replacement operation is counted as a success.

The success of a knee replacement will depend on multiple patient, surgical and implant-related variables – each patient still needs to be considered carefully as an individual. Patients should be optimistic about these results but will still need to engage carefully with their carers in order for them to have realistic expectations.

Fares S Haddad, Professor of Orthopaedic and Sports Surgery; Consultant Orthopaedic Surgeon, University College Hospital, London

The commentator declares no conflicting interests


Back to top