Current advice regarding implant choice is based on estimates of cost-benefit derived from implant survival to an endpoint of revision. Current estimates do not account for many implant failures which are treated with non-revision surgery and may not be accurate. The aim of this study was to estimate survival of major stem implant design groups to an endpoint of reoperation.

Primary total hip replacement and linked revision form the National Joint Registry (NJR) and Hospital Episode Statistics (HES) data linked by unique identifier were used. Survival of femoral implant groups (cemented stainless steel polished taper [PTSS], cemented cobalt chrome polished taper [PTCC], cemented composite beam [CB], collarless cementless [NCOL] and collared cementless [COL]) was estimated using Kaplan-Meier method.

809,832 patients with valid NJR and HES data from England, were included. Cumulative failure at ten years for PTSS increased overall from 2.9% (95%CI 2.8–2.9) to 3.6% (95%CI 3.6–3.7) after inclusion of reoperations. Cumulative failure at ten years for PTSS increased from 2.5% (95%CI 2.5–2.6) to 3.3% (95%CI 3.2–3.4), for PTCC increased from 3.8% (95%CI 3.5–4.0) to 5.4% (95%CI 5.1–5.6), for CB increased from 3.1% (95%CI 2.9–3.3) to 4.1% (95%CI 3.8–4.3), for NCOL increased from 3.4% (95%CI 3.3–3.5) to 3.9% (95%CI 3.8–4.0), and for COL increased from 2.5% (95%CI 2.4–2.6) to 3.1% (95%CI 2.9–3.2), after inclusion of reoperations.

Re-operation for internal fixation is as significant life event for the patient as revision. When a more inclusive metric is used, the patient and clinician's perspective on what constitutes a GIRFT implant may not be the same. Further work is required to update implant selection guidance in view of the change in implant performance.

Current estimates of periprosthetic fracture risk associated with femoral implants is mostly limited to revision only estimates and does not accurately represent stem performance. The aim of this study was to estimate the risk of surgically treated post-operative periprosthetic femoral fracture (POPFF) more accurately associated with frequently used femoral implants used for total hip arthroplasty (THA).

A cohort study of patients who underwent primary THA in England between January 1, 2004, and December 31, 2020. Periprosthetic fractures were identified from prospectively collected revision records and national procedure coding records. Survival modelling was used to estimate POPFF incidence rates, adjusting for potential confounders. Subgroup analyses were performed for patients over 70 years, with non-osteoarthritic indications, and neck of femur fracture.

POPFF occurred in 0.6% (5100/809,832) of cases during a median (IQR) follow up of 6.5 (3.9 to 9.6) years. The majority of POPFF were treated with fixation after implantation of a cemented stem. Adjusted patient time incidence rates (PTIR) for POPFF varied by stem design, regardless of cement fixation. Cemented composite beam stems (CB stems) demonstrated the lowest risk of POPFF. Collared cementless stems had an equivalent or lower rate of POPFF versus the current gold standard polished taper slip cemented stem.

POPFF account for a quarter of all revisions following primary THA. Cemented CB stems are associated with the lowest POPFF risk. Stem design is strongly associated with POPFF risk, regardless of the presence of cement. Surgeons, policymakers, and patients should consider these findings when recommending femoral implants in those most at risk of POPFF.

The documentation of deep infection rates in joint replacement is fraught with multiple difficulties. Deep infections acquired in theatre may present late, but some later presenting deep infections are clearly haematogenous, and not related to surgical management. The effect of Ultra Clean Air on infection rates was published by Charnley in 1972 (CORR,87:167–187). The data is valuable because large numbers of THRs were performed in standard and Ultra Clean theatres, and detailed microbiology of the air was also recorded. No IV antibiotics were used, so only the effect of air quality was studied.

We extracted the data on theatre type and numbers from Table 3, and numbers and intervals from surgery of deep infections from Table 7. Theatre types with 300 air changes per hour and 3.5 CFU/M³ were classified as Ultra Clean. A logistic regression model was used to examine the effect of theatre type and time elapsed after procedure on the probability of becoming infected.

The model suggests that, controlling for time period, Ultra Clean Air is associated with a significantly lower probability of infection, with an OR of 0.30, p = 2.74 × 10⁻⁶. The effect is larger earlier post-surgery, but it does persist. The results are best reviewed as a graphic, which shows that Ultra Clean Air clearly affects the deep infection rate for up to four years post-surgery.

Ultra Clean Air reduces infection rates for up to four years post-surgery, so it is safe to assume that infections presenting after this are haematogenous. Ultra Clean Air does not eliminate early deep infection, so some early infections are not related to air quality. It is not practical to undertake widespread detailed retrospective analyses of cases. When monitoring infection rates there needs to be a balance between failing to record infections related to surgical technique and waiting many years to record low numbers of very late presenting problems. We suggest that registries should regard infections documented within three years of surgery as treatment complications.

For any figures or tables, please contact the authors directly.

Successful estimation of postoperative PROMs prior to a joint replacement surgery is important in deciding the best treatment option for a patient. However, estimation of the outcome is associated with substantial noise around individual prediction. Here, we test whether a classifier neural network can be used to simultaneously estimate postoperative PROMs and uncertainty better than current methods.

We perform Oxford hip score (OHS) estimation using data collected by the NJR from 249,634 hip replacement surgeries performed from 2009 to 2018. The root mean square error (RMSE) of the various methods are compared to the standard deviation of outcome change distribution to measure the proportion of the total outcome variability that the model can capture. The area under the curve (AUC) for the probability of the change score being above a certain threshold was also plotted.

The proposed classifier NN had a better or equivalent RMSE than all other currently used models. The standard deviation for the change score for the entire population was 9.93, which can be interpreted as the RMSE that would be achieved for a model that gives the same estimation for all patients regardless of the covariates. However, most of the variation in the postoperative OHS/OKS change score is not captured by the models, confirming the importance of accurate uncertainty estimation. The threshold AUC shows similar results for all methods close to a change score of 20 but demonstrates better accuracy of the classifier neural network close to 0 change and greater than 30 change, showing that the full probability distribution performed by the classifier neural network resulted in a significant improvement in estimating the upper and lower quantiles of the change score probability distribution. Consequently, probabilistic estimation as performed by the classifier NN is the most adequate approach to this problem, since the final score has an important component of uncertainty.

This study shows the importance of uncertainty estimation to accompany postoperative PROMs prediction and presents a clinically-meaningful method for personalised outcome that includes such uncertainty estimation.

Whilst total hip replacement (THR) is generally safe and effective, pre-existing medical conditions, particularly those requiring inpatient admission, may increase the risk of post-operative mortality. Delaying elective surgery may reduce the risk, but it is unclear how long a delay is sufficient.

We analysed 958,145 primary THRs performed for solely osteoarthritis April 2003-December 2018, in the NJR linked to Hospital Episodes Statistics to identify inpatient admissions prior to elective THR for 17 conditions making up the Charlson index including myocardial infarction, congestive heart failure, cerebrovascular disease and diabetes. Crude analyses used Kaplan-Meier and adjusted analyses used Cox modelling. Patients were categorised for each co-morbidity into one of four groups: not recorded in previous five-years, recorded between five-years and six-months before THR, recorded six-months to three-months before THR, and recorded between three-months and day before surgery.

90-day mortality was 0.34% (95%CI: 0.33–0.35). In the 432 patients who had an acute MI in the three months before THR, this figure increased to 18.1% (95%CI 14.8, 22.0). Cox models observed 63 times increased hazard of death within 90-days if patients had an acute MI in the 3-months before their THR, compared to patients who had not had an MI in the five years before their THR (HR 63.6 (95%CI 50.8, 79.7)) This association reduced as the time between acute MI and THR increased. For congestive cardiac failure, the hazard in the same scenario was 18-times higher with a similar protective effect of delaying surgery.

Linked NJR and HES data demonstrate an association between inpatient admission for acute medical co-morbidities and death within 90-days of THR. This association is greatest in MI, congestive cardiac failure and cerebrovascular disease with smaller associations observed in several other conditions including diabetes. The hazard reduces when longer delays are seen between the admission for acute medical conditions and THR in all diagnoses. This information will help patients with previous medical admissions and surgeons to determine optimal timing for surgery.

In a recent phase 2 superiority clinical trial we demonstrated that a single dose of 60mg of the human monoclonal antibody denosumab inhibits osteolytic lesion activity in patients undergoing revision total hip arthroplasty (THA), demonstrating proof of biological efficacy for this clinical application. Here, we examined the effect that denosumab has on disease biology at the osteolysis tissue level.

Osteolytic tissue taken from the prosthesis-bone lesion interface at revision surgery in patients with osteolysis (n=10 participants that had received a single 60 mg dose of denosumab 8 weeks prior to revision surgery and n=10 that had received placebo) was examined for total genetic message activity and protein levels using whole genome sequencing and mass spectrometry, respectively.

The top five upregulated enriched pathways with denosumab treatment included inflammatory response, myeloid cell activation, myeloid leukocyte migration, neutrophil and granulocyte activation (p<6.26 × 10⁻²⁸). Cell morphogenesis was amongst the most downregulated pathways (p<3.42 ×10⁻²³). Finally, comparison of the trial mRNA and protein data versus mouse single cell RNA sequencing data of the same pathway blockade in mouse tibia showed the same direction of effect, suggesting that giving the drug causes then cells responsible for osteolysis to disperse into a more immature form (128 of 189 genes (z=4.87, P<0.0001) disease and functional pathways at the mRNA level and 10 of 11 (z=2.72, P=0.0065) at the protein level).

In this first-in-man study we identify multiple genes and pathways within periprosthetic osteolysis tissue that are affected by denosumab treatment. The dominant pathways involved upregulation of innate inflammatory signaling and downregulation of cell morphogenesis. We also found enrichment of similar disease and functional pathways at both the mRNA and protein levels versus mRNA pathway enrichment found in mouse osteomorphs. These data provide the first human data of the mechanistic effect of denosumab treatment on inflammatory osteolytic lesion activity after joint replacement that is necessary to support its clinical application.

∗Winner of The Bone & Joint Journal prize∗

Revision total hip arthroplasty (rTHA) can be complex and associated with significant cost, with an increasing burden within the UK and globally. Regional rTHA networks have been proposed aiming to improve outcomes, reduce re-revisions and therefore costs. The aim of this study was to accurately quantify the cost and reimbursement for the rTHA service and to assess the financial impact of case complexity at a tertiary referral centre within the NHS.

A retrospective analysis of all revision hip procedures was performed over two consecutive financial years (2018–2020). Cases were classified according to the Revision Hip Complexity Classification (RHCC) and by mode of failure; infected or non-infected. Patients of ASA grade of 3 or greater or BMI over 40 are considered “high-risk” by the RHCC. Costs were calculated using PLICS and remuneration based on the HRG data. The primary outcome was the financial difference between tariff and cost per episode per patient. Comparisons between groups were analysed using analysis of variance and two-tailed unpaired t-test.

199 revision episodes were identified in 168 patients: 25 (13%) least complex revisions (H1), 110 (55%) complex revisions (H2) and 64 (32%) most complex revisions (H3). 76 (38%) cases were due to infection. 78 (39%) of patients were in the “high-risk” group. Median length of stay increased with case complexity from 4, to 6 to 8 days (p=0.17) and significantly for revisions performed for infection (9 vs 5 days; p=0.01). Cost per episode increased significantly between complexity groups (p=0.0002) and for infected revisions (p=0.003). All groups demonstrated a mean deficit, but this significantly increased with revision complexity (£301, £1,820 and £4,757 per case; p=0.02) and for infected failure (£4,023 vs £1,679; p=0.02). The total deficit to the trust for the two-years was £512,202.

Current NHS reimbursement for rTHA is inadequate and should be more closely aligned to complexity. An increase in the most complex rTHA at major revision centres (MRC) will likely place a greater financial burden on these units.

Aims

To develop and externally validate a parsimonious statistical prediction model of 90-day mortality after elective total hip arthroplasty (THA), and to provide a web calculator for clinical usage.