Editorial | The National Joint Registry of England and Wales - 9th Annual Report | Which Medical Device

A review from Mr. Alasdair Mackie, Orthopaedic Specialist Registrar, Northern Deanery.

The 9th annual report of the National Joint Registry (NJR) for England and Wales has been published in the midst of ongoing controversy about metal-on-metal bearings (1). The Registry now includes data on over 1.2 million joint replacements and continues to pursue its mission to collect high quality and relevant data about joint replacement surgery in order to provide an early warning of issues relating to patient safety…and ensure the quality and cost effectiveness of joint replacement surgery.

The NJR continues to expand: ankle replacements have made their first appearance this year, with shoulder and elbow to follow shortly, and from October, data from Northern Ireland will be included. Compliance, consent and linkability have continued to rise: between 93 and 100% of procedures are now registered in the NJR, consent to the storage of data has increased to 90.4% from 88.9% over the last year, and 95.5% of revision procedures are now linked to the primary procedure.

In 2011, 95% of the 411 orthopaedic units in England and Wales submitted data about at least one procedure, an average of 204 hip and 219 knees per unit.

Hip Replacement

2011 saw 80,314 hip replacements performed, up 5% from 2010. These were 71,672 primary and 8,641 revision surgeries, reflecting a revision burden of 11.1%. Despite the impression that the demographics of joint replacement might be changing, age and gender of patients undergoing hip replacement have seen little change since 2003, with a mean age of 67.2 years and the majority (60%) female. Similarly, mean BMI (28.5), and ASA grades have not changed.

30 day mortality after primary and revision hip arthroplasty remains low at 0.2%, but is strongly age related.  8 year mortality for men under 55 is 3.8% but for over 80s, 81.9%.

Indications for hip replacement were osteoarthritis in 93%, avascular necrosis in 2%, fracture in 2%, congenital dislocation in 2% and inflammatory arthritis in 1% although these indications also vary with age. Both mechanical and chemical thromboprophylaxis were used in 90% of cases (TED stockings in 66% and LMWH in 71% respectively).

Of primary hip arthroplasties, 36% were all cemented, and 44% were uncemented, unchanged from previous years. 18% were hybrid or reverse hybrid and the remaining 2% hip resurfacing. There were 142 brands of femoral stem on the market, of which the Exeter V40 stem was the most popular cemented stem (64% of the market) and the Contemporary cup the most popular acetabular component (just under 35% of cases). The most popular cementless components were the Corail stem (47% of cases) and the Pinnacle socket (33%). Hip resurfacing continued to decline 4,350 (2009) to 1,801 (2011) but the Birmingham hip resurfacing device remained the market leader.

Metal-on-polyethylene remained the most common articulation. The poor results of some metal-on-metal articulations have been followed by a significant decline in their use and a consequent increase in the adoption of ceramic-on-ceramic bearings (from 2,000 in 2003 to 17,000 in 2011). Large diameter bearings have also decreased in popularity, with just fewer than 30% of implantations having a head size of 36mm or greater. 

The NHS continues to shoulder the burden of revision surgery, performing 84% of revisions. Indications for revision were infection in 12% and adverse tissue reactions in 11%, the latter highlighting the impact of metal-on-metal revision surgery. The lowest rates of revision were associated with cemented metal or ceramic on polyethylene combinations.  The 8 year revision rate of all cemented hips was 2.3%, compared with 5.1% for uncemented implants. 

Update on Metal-on-Metal

The NJR data continue to highlight poor results with some metal-on-metal bearings. The size of this problem is substantial: 8 % of stemmed THR procedures in the NJR used metal-on-metal bearings, representing over 30,000 patients.  The overall five-year revision rate of 6.2% is far higher than any other bearing combination, although revision rates are lower when smaller head sizes are used.  This year, data from the Registry lead to publications calling for an end to stemmed metal-on-metal hip replacements, and refuting the suggestion that metal-on-metal bearings were associated with an increased risk of cancer at 7 years (2), (3).

Knee Replacements

84,653 knee replacements were recorded on the NJR in 2011, 3.3% up on 2010. These included 5,137 revisions (6.1%). 86% of these were cemented total knee replacements, 4% cementless, 1% hybrid, 1% patello-femoral replacement and 8% unicondylar. The majority were unconstrained fixed bearing knees (68%), the next most common being posterior stabilised fixed bearing knees (25%). Mobile bearing devices comprised only 7% of implantations. These proportions have changed little since 2005. The mean age of patients was just over 68, 56% being female.

Patellar replacement at primary knee replacement remains an area of controversy and opinion about best practice is divided.  Figures from the NJR reflect this with 37% of patients undergoing cemented knee replacement having patellar replacement at the time of the primary procedure. 

The PFC knee was the most common total knee replacement used (37%) although the Stryker Triathlon knee and the Genesis II from Smith and Nephew are becoming more popular. The Oxford partial knee remains the market leader in unicondylar replacements (70%) and the Avon patello-femoral replacement is still market leader in its class.

5,135 revision operations were performed in 2011, up 1% from 2010, the main indications being aseptic loosening (35%) and infection (23%). The 8 year revision rate for cemented unconstrained fixed bearing knees was 2.6% with little difference identified between the other bicondylar knees.  Unicondylar and patello-femoral implants did not fare as well, with 8 year revision rates of 10.1% for fixed and 11.1% for mobile bearings and 14.7% after patello-femoral replacement.  30 day mortality after knee replacement was 0.2%. Mortality after 8 years was 3.1% for females under 55, and 51.8% for those over 80.  

Ankle Replacements

This was the first year in which ankle replacements were represented in the Registry. Unsurprisingly as a result, the compliance rate for ankles was very low at 64%, but seems likely to improve. 492 total ankle replacements were recorded, 21 (4.2%) of which were revisions.  75% of primary procedures were performed within the NHS and the majority were uncemented.  The main surgical indication was osteoarthritis at 88%, 19% of which was posttraumatic. The average age was 68 with a slight higher male proportion (56%). The market leader was the Mobility ankle from DePuy, used in 57% of primary procedures. 

Indications for the 21 revision surgeries were wide ranging and included malalignment (24%), polyethylene wear (29%), undiagnosed pain (29%), aseptic tibial loosening, infection and osteolysis of the tibia.

Once more therefore, the National Joint Registry has produced a large volume of data about joint replacement, which is of great interest and which provides headline data about joint replacement practice in England and Wales. The Registry has proved to be an excellent resource in describing problems with metal-on-metal bearings in particular, and has provided a great deal of supporting evidence when problems appear. The future lies in collecting more of the same data as it expands its reach, but must also involve closer scrutiny of higher resolution data in order that we improve outcomes for individual patients.

User comments: The National Joint Registry of England and Wales - 9th Annual Report

References: