Professor Mike Reed, Chair, Editorial Committee
Mr Tim Wilton, NJR Medical Director
Much as we reported last year, there has been a significant impact of the COVID-19 pandemic on the volume of all joint procedures. Activity in 2020 was roughly half that of normal, and in 2021 was found to be around 70-85%. Interestingly, hip replacement activity was the most preserved of all the joints (85%) and this may indicate these patients being given higher priority.
We were pleased to be able to support the BASK, BESS and BHS annual conferences with a presentation on NJR’s work from senior NJR clinicians. Each session was followed by a lively open question session with delegates who were interested in hearing how the NJR is supporting the work of the orthopaedic sector to deliver high quality care to their patients. Considerable interest has been generated by our finding of significant differences in revision rates between variants within the same brand of implants. The NJR regional compliance team was also present at each event, manning our information stand and offering support to surgeons in using our new reporting portal, NJR Connect – Data Services.
Meanwhile the work of the NJR Editorial Committee has continued with development of the strategy and style of the report. All members take responsibility for producing a report that is rigorously edited, taking almost a full year to prepare for, write and review. The Committee brings together experts on data collection and reporting as well as generous input from a patient perspective, clinicians from specialist societies and members of the NJR Management Team. Each year we aim to make progress in reporting on our rich data resource, making data easily accessible to best inform patient-centred care.
This year we present an additional section on Patient Reported Outcome Measures (PROMs) which showcases our current ideas on reporting hip and knee PROMs at the implant brand level in future years. This year we describe the completeness of the data available from the National PROMs programme, explore whether the available data are generalisable to the entire cohort, describe a non-inferiority approach to the analysis and the issues with that approach in the context of routine reporting and interpretation and also propose a model for future reporting. We are keen to get feedback from all stakeholders, but particularly surgeons, patients and industry. We also share a case study from South West London Elective Orthopaedic Centre showing the benefits of using NJR data in improving patient outcomes.
Although it is not clearly reflected in the data, surgical teams have been reporting increased complexity and comorbidity, as patients have been waiting longer for surgery in comparison to 2019. This may affect outcomes in future years.
We launch the report each September at the British Orthopaedic Association (BOA) Congress and this year the report has been published online only. Increasingly there is considerable additional information available online and we would encourage you to explore the NJR’s dedicated annual report website at reports.njrcentre.org.uk. The website offers a helpful interactive platform for the descriptive NJR data, with many supporting appendices.
Commentary on findings
Hip replacement
We are now at 18.75 years since NJR inception and thus we have reported at up to 18 years follow-up although in some areas the numbers at risk are low. Surgeons are now performing a median of 59 primary procedures over three years, and 498 per unit. This is a huge drop over the timescale of the COVID-19 pandemic with the last unaffected reporting period being the 2019 annual report.
What are we implanting?
In terms of fixation type and bearing, cemented metal-on-polyethylene (MoP) hip replacement still dominates across the life of the registry, but in 2021 ceramic-on-polyethylene (CoP) hybrid hips were the most often implanted (see Table 3.H2). Dual mobility operations continue to increase with 2021 being the biggest year for their use despite the pandemic (Figure 3.H1 (b)). Although now more common than resurfacings and almost as common as reverse hybrid hips, it is still a minor player in terms of numbers overall (Figure 3.H2 (b)).
Within fixation types, the use of CoP implants continues to grow and now clearly dominates in both uncemented and hybrid hips. Ceramic-on-ceramic (CoC) use is dropping in all uncemented and hybrid fixations (Figures 3.H3 (b) and (c)). In both uncemented and hybrid CoP hips, both 32mm and 36 mm heads have been similarly popular in 2021 (Figure 3.H3 (e)).
What are the results?
At ten years and greater CoP bearings now outperform MoP in revision rates for cemented, uncemented and hybrid hips (although at very extended follow-up it is not always significant) (Table 3.H5). This is good news given their increasing popularity, although these data may be confounded by more frequent use of highly cross-linked polyethylene in these modern bearings which could be partially responsible for the observed pattern.
A key message from the hip section of our report is that the five-year revision rates of metal-on-polyethylene-on-metal (MoPoM) dual mobility bearings appear higher than most cemented and uncemented unipolar combinations (apart from metal-on-metal (MoM)). Ceramic-on-polyethylene-on-metal (CoPoM) may be better but it is difficult to know at this point. Dual mobility patients are of a different demographic to those having a unipolar total hip replacement (THR) and this may be a reason for difference in results (although older patients generally are found to have lower rates of revision in the registry - Table 3.H5). Infection and periprosthetic fracture appear to be common causes of revision (Table 3.H9).
As noted before, younger patients have a higher rate of revision than older ones, and this difference is most marked in women, with younger women having the highest revision rates as a group, and older women the lowest. (Figure 3.H9 (a))
CoP bearings seem to have a particular advantage over MoP in younger age groups (Table 3.H6). We should note that resurfacing in males under 55 has a revision rate of double that of some of the alternatives out at 15 years, even though that is the group for whom resurfacing is still said by some to be appropriately indicated.
Head size and bearing construct revision rates are complex and the details are shown in Figures 3.H10 (a) to (l). Broadly though, head sizes for both cemented MoP and CoP hips seem to be optimal at 28mm and 32 mm with sizes outside of these ranges having higher revision rates. In uncemented hips with a CoP bearing the head sizes of 32mm or 36mm appear optimal (Figure 3.H10 (e)).
Overall, there are excellent results for many different brands of hip replacements, many of these have been used in large numbers and have failure rates substantially lower than 5% at ten years. Tables 3.H7 and 3.H8 give revision rates for different brands of primary hip replacements by fixation and bearing and are worth careful consideration. Costs are a factor in our decision-making and Table 3.11 has clear information on the lifespan of patient age groups by gender. The current NHS England Best Practice Tariff recommendation supporting cemented or hybrid prostheses for most patients over 70 can be supported by these data. Costs are a factor in our decision-making and Table 3.H11 has clear information on the lifespan of patient age groups by gender. The current NHS England Best Practice Tariff recommendation supporting cemented or hybrid prostheses for most patients over 70 can be supported by these data.
In procedures for acute trauma, numbers of patients receiving dual mobility and total hip replacements have not recovered post-pandemic possibly because of other factors like the publication of the HEALTH trial (Bhandari M, et al., 2019). In contrast to elective practice, dual mobility hips don’t appear to have an increased revision rate in trauma patients, although specific trials are underway to study this.
In procedures for acute trauma, numbers of patients receiving dual mobility and total hip replacements have not recovered post-pandemic possibly because of other factors like the publication of the HEALTH trial (Bhandari M, et al., 2019). In contrast to elective practice, dual mobility hips don’t appear to have an increased revision rate in trauma patients, although specific trials are underway to study this.
Knee replacement
As with hip replacement we now report knee replacement outcomes up to 18 years, based upon over 1.4 million primaries.
What are we implanting?
Across the life of the registry, unconstrained cemented fixed bearing total knee replacement (TKR) still dominates the registry (58%) with posterior stabilised cemented fixed bearing TKR being the next biggest group at 20%, although the use of posterior stabilised implants has been dropping over the last ten years. Both cemented (mainly fixed bearing) and uncemented/hybrid (mainly mobile bearing) unicondylar knee replacements are on the increase as a proportion of knee replacement, moving from a total of 8% to 13% over the last ten years. In terms of actual numbers of procedures, unicondylar knees have increased rapidly since 2013 until the COVID- 19 pandemic began. Surgeons performing total knee replacement over the last three years have carried out a median of 86 cases and those performing unicondylar a median of 19; pre-pandemic the figures were 109 and 21 respectively. These three-year averages are likely to drop again in the registry and this will be reported in the next annual report. Only a small proportion of unicondylar knee procedures are being done by a surgeon who performs fewer than 12 per year. NICE standards in 2022 report on the need for a discussion around unicondylar versus total knee replacement for isolated medial osteoarthritis and this will likely drive its use in the future.
What are the results?
Since 2008, revision rates have continued to improve out to ten years. It is very difficult to ascribe a particular reason for the early increased revision rate pre-2008, although it may be attributable to the improved linkage between primary and revision cases over the last 14 years which appears to coincide with the advent of NJR’s Clinician Feedback reporting. Reassuringly the current dominant choice of cemented unconstrained is validated in the revision data at ten years and beyond although the monobloc polyethylene tibia TKRs also perform well (in a slightly different group). Overall, some high-volume brands of total knee replacement are showing 2.5% revision rates or less at ten years. The best performing and popular unicondylar knee brands are around double that, although in patients five to six years younger (Table 3.K9 (a)). Patellofemoral replacement revision rates come in around five times higher than a cemented TKR at 10 or 18 years.
Knee replacements in males under 55 fail earlier than for other groups with a revision rate of over 12% at 18 years, and figures show that early revision means a higher chance of re-revision. For women having unicondylar knee replacement however, they are more likely to fail than their age equivalent male counterpart. Detail of knee replacement by brand and by patella resurfacing are presented and readers are encouraged to read the detail relevant to their implant choices. NICE has recently concluded that surgeons should offer patella resurfacing (NICE guidance NG157) but revision outcomes appear to be brand/construct specific. Surgeons are encouraged to examine the performance data of the specific constructs that they use.
Revision of a knee replacement is a significant risk factor for further revision, but this is less pronounced if the primary was a patellofemoral joint replacement. Looking at Figure 3.K6 (b), it is worth noting that uncemented unicompartmental knee replacements that are revised appear to behave like revised total knee replacements with respect to re-revision, not “like a primary” as some may propose.
Ankle replacement
For ankle replacement, data collection began in 2010, with this report now being based on almost 8,000 ankle replacements. All ankle replacement implant brands are uncemented designs but previously some surgeons added cement to the construct. In the latest data this trend appears to have stopped.
Apart from the COVID-19 impacted years, ankle replacement has been increasing rapidly since 2015. This increase has been largely delivered by higher volume surgeons doing more than one case a month. In 2021, with COVID-19, the median number of ankle replacements performed was three. Only 3% of consultants performed 20 or more primary ankle replacements and a further 14.8% performed between 10 and 19 primary ankle replacements.
A single implant brand – INFINITY – now captures over 65% of the market.
Overall revision rates run at around 5.5% at five years (Table 3.A3), with revision rates much higher in the under 65s. The market leader is now tracking at less than a 3% risk of revision at five years, so this is promising but still does not reach parity with total hip or knee replacement. With wide variance of failure rates of differing implants, it is important that surgeons use an implant with either a good track record or one that is being introduced responsibly.
Aseptic loosening, then infection, dominate the indications for ankle revision. With revision, the most common outcome was a single stage revision. Amputation and conversion to arthrodesis are reported as revisions but there remains some concern about under-reporting of both of these.
Elbow replacement
During 2021-22 the NJR Data Quality Committee has overseen a large and detailed exercise to identify missing data and source check for mismatched data. The project was conducted in collaboration with the British Orthopaedic Trainees Association (BOTA), British Elbow and Shoulder Society (BESS), Royal College of Surgeons of England (RCS Eng) and the British Orthopaedic Association (BOA). As a result of the audit, there is a significant increase in the numbers of elbows available for analysis on this report – over 7,000. These new data are mainly radial head replacements but also include previously ‘missing’ total elbow replacements.
Across the life of the registry there are similar numbers for trauma and elective indications. In 2021, trauma (mostly radial head replacements) dominated with volume activity being relatively consistent with previous years. The use of distal humeral hemiarthroplasty in trauma cases continues to grow. The volume of primary total elbow replacements has marginally increased over the last five years (except for 2020 and 2021 due to the impact of COVID-19) with the number of surgeons that are performing one or two procedures falling annually since 2018. Elective activity in 2020 and 2021 dropped to around 50-60% of normal volumes.
In elective practice, around 50% of total elbow replacements in the registry have used a single implant brand (Coonrad Morrey). In radial head replacement for trauma, the Anatomic brand dominates.
The Total Elbow Replacement revision rate is just over 1% per year for elective indications. Beyond two years, revisions of those primaries performed for trauma appear to be less frequent, but we see low numbers at this point so estimates are less certain.
Shoulder replacement
As a result of a review of the shoulder data in 2018 and 2019, new classification and component attributes are used in the annual report. The analysis team has cross-checked the implanted construct with the indicated procedure at the time of the surgery and positively confirmed the implanted construct matches the reported procedure in just over 90% of cases. A total of 56,312 primary shoulder replacements were available for analysis.
There has been a changing scene in shoulder replacement over the nine years of data reported. There was a continued increasing preference for reverse polarity total shoulder replacement year-on-year until 2019 which has since plateaued, but it is not clear if this is a true plateau or a secondary effect of the impact of COVID-19.
In elective practice, resurfacing humeral hemiarthroplasty, stemmed humeral hemiarthroplasty, stemmed total shoulder replacement, and resurfacing total shoulder replacements have all declined since the start of data collection while stemless total shoulder replacements have steadily increased and the volume of stemmed reverse polarity total shoulder replacements has increased substantially.
In trauma cases, the popularity of stemmed humeral hemiarthroplasty has reduced in recent years while the popularity of stemmed reverse polarity total shoulder replacements continues to increase steadily albeit allowing for the impact of COVID-19 in 2020 and 2021.
Broadly, revision rates for trauma and elective shoulder replacement are a little less than 1% per year, with rates for trauma indications appearing to be even better where follow-up is beyond three years. Reassuringly the increasingly popular elective stemmed reverse polarity total shoulder replacement appears to have revision rates around 0.5% per year. However, these implants are more difficult to revise than other shoulder devices and it remains to be seen whether this is truly a definite advantage of the reverse polarity implant. There are groups, for instance males under 55, who fair much worse with elective shoulder replacement with revision rates over 14% at seven years.
A pre-op PROMs questionnaire is collected by the NJR for shoulder replacement and around 25% of elective patients return these within the correct timescale. Around a third of patients complete a further six-month PROMs but engagement is poor beyond that. At six months following surgery, 5.3% of patients reported a score worse than they did pre-operatively. Patients with a high pre-op score – over 40 on the Oxford Should Score (OSS) – had an equal chance of being better or worse at six months.
Concluding acknowledgements
The NJR continues to work collaboratively with our many stakeholders; the most important, of course, are the patients we serve, and whom we would like to thank for allowing us to use their data.
The NJR operational collaboration is a huge team effort – this year managed almost exclusively by work performed remotely online. Elaine Young, NJR Director of Operations has demonstrated the great versatility of her leadership and her team.
Many thanks also to the following without which the NJR could not function:
All members of the NJR Steering Committee
Members of the NJR sub-committees:
Executive
Data Quality
Editorial
Implant Scrutiny
Medical Advisory
Regional Clinical Coordinators
Research
Surgical Performance
Members of the Data Access Review Group
Members of the NJR Patient Network
Other organisations:
Medicines and Healthcare products Regulatory Agency (MHRA)
Care Quality Commission (CQC)
NHS England / Improvement
NHS Digital
Getting It Right First Time (GIRFT)
British Orthopaedic Association (BOA)
British Hip Society (BHS)
British Association for Surgery of the Knee (BASK)
British Elbow and Shoulder Society (BESS)
British Orthopaedic Foot and Ankle Society (BOFAS)
European Orthopaedic Research Society (EORS)
Healthcare Quality Improvement Partnership (HQIP)
Confidentiality Advisory Group (CAG)
Association of British HealthTech Industries (ABHI)
We are most grateful to our NJR delivery contractors for their very valuable input into the NJR Annual Report, and many other functions. NEC Software Solutions, University of Bristol and University of Oxford teams help us refine and improve each year.
We offer our personal thanks to Vicky McCormack, Report Project Manager, NEC; Deirdra Taylor, Associate Director of Communication and Stakeholder Engagement and Oscar Espinoza, Design and Communication Manager for the NJR, for getting the final report into shape.
Bhandari M et al.; Total Hip Arthroplasty or Hemiarthroplasty for Hip Fracture. Value Health. N Engl J Med 2019; 381:2199-2208.