Aims. The aim of this study was to determine whether fixation, as opposed to revision arthroplasty, can be safely used to treat reducible Vancouver B type fractures in association with a cemented collarless polished tapered femoral stem (the Exeter). Methods. This retrospective cohort study assessed 152 operatively managed consecutive unilateral Vancouver B fractures involving Exeter stems; 130 were managed with open reduction and internal fixation (ORIF) and 22 with revision arthroplasty. Mean follow-up was 6.5 years (SD 2.6; 3.2 to 12.1). The primary outcome measure was revision of at least one component. Kaplan–Meier survival analysis was performed. Regression analysis was used to identify risk factors for revision following ORIF. Secondary outcomes included any reoperation, complications, blood transfusion, length of hospital stay, and mortality. Results. Fractures (B1 n = 74 (49%); B2 n = 50 (33%); and B3 n = 28 (18%)) occurred at median of 4.2 years (interquartile range (IQR) 1.2 to 9.2) after primary total hip arthroplasty (THA) (n = 138) or hemiarthroplasty (n = 14). Rates of revision and reoperation were significantly higher following revision arthroplasty compared to ORIF for B2 (p = 0.001) and B3 fractures (p = 0.050). Five-year survival was significantly better following ORIF: 92% (95% confidence interval (CI) 86.4% to 97.4%) versus 63% (95% CI 41.7% to 83.3%), p < 0.001. ORIF was associated with reduced blood transfusion requirement and reoperations, but there were no differences in medical complications, hospital stay, or mortality between surgical groups. No independent predictors of revision following ORIF were identified: where the bone-cement interface was intact, fixation of B2 or B3 fractures was not associated with an increased risk of revision. Conclusion. When the bone-cement interface was intact and the fracture was anatomically reducible, all Vancouver B fractures around Exeter stems could be managed with fixation as opposed to revision arthroplasty. Fixation was associated with reduced need for blood transfusion and lower risk of revision surgery compared with revision arthroplasty. Cite this article: Bone Joint J 2021;103-B(2):309–320

Aims. Dual mobility (DM) bearings are an attractive treatment option to obtain hip stability during challenging primary and revision total hip arthroplasty (THA) cases. The purpose of this study was to analyze data submitted to the American Joint Replacement Registry (AJRR) to characterize utilization trends of DM bearings in the USA. Methods. All primary and revision THA procedures reported to AJRR from 2012 to 2018 were analyzed. Patients of all ages were included and subdivided into DM and traditional bearing surface cohorts. Patient demographics, geographical region, hospital size, and teaching affiliation were assessed. Associations were determined by chi-squared analysis and logistic regression was performed to assess outcome variables. Results. A total of 406,900 primary and 34,745 revision THAs were identified, of which 35,455 (8.7%) and 8,031 (23.1%) received DM implants respectively. For primary THA, DM usage increased from 6.7% in 2012 to 12.0% in 2018. Among revision THA, DM use increased from 19.5% in 2012 to 30.6% in 2018. Patients < 50 years of age had the highest rates of DM implantation in every year examined. For each year of increase in age, there was a 0.4% decrease in the rate of DM utilization (odds ratio (OR) 0.996 (95% confidence interval (CI) 0.995 to 0.997); p < 0.001). Females were more likely to receive a DM implant compared to males (OR 1.077 (95% CI 1.054 to 1.100); p < 0.001). Major teaching institutions and smaller hospitals were associated with higher rates of utilization. DM articulations were used more commonly for dysplasia compared with osteoarthritis (OR 2.448 (95% CI 2.032 to 2.949); p < 0.001) during primary THA and for instability (OR 3.130 (95% CI 2.751 to 3.562) vs poly-wear; p < 0.001) in the revision setting. Conclusion. DM articulations showed a marked increase in utilization during the period examined. Younger patient age, female sex, and hospital characteristics such as teaching status, smaller size, and geographical location were associated with increased utilization. DM articulations were used more frequently for primary THA in patients with dysplasia and for revision THA in patients being treated for instability. Cite this article: Bone Joint J 2020;102-B(7 Supple B):27–32

Aims. We aimed to evaluate the long-term outcome of highly cross-linked polyethylene (HXLPE) cemented acetabular components and assess whether any radiolucent lines (RLLs) which arose were progressive. Methods. We retrospectively reviewed 170 patients who underwent 187 total hip arthroplasties at two hospitals with a minimum follow-up of ten years. All interventions were performed using the same combination of HXLPE cemented acetabular components with femoral stems made of titanium alloy. Kaplan-Meier survival analysis was performed for the primary endpoint of acetabular component revision surgery for any reason and secondary endpoint of the appearance of RLLs. RLLs that had appeared once were observed over time. We statistically assessed potential relationships between RLLs and a number of factors, including the technique of femoral head autografting and the Japanese Orthopaedic Association score. Results. The mean follow-up period was 13.0 years (10.0 to 16.3). Femoral head autografting was performed on 135 hips (72.2%). One acetabular component was retrieved because of deep infection. No revision was performed for the aseptic acetabular loosening. The Kaplan-Meier survival curve for the primary and secondary endpoints were 98.2% (95% confidence interval (CI) 88.6% to 99.8%) and 79.3% (95% CI 72.8% to 84.6%), respectively. RLLs were detected in 38 hips (21.2%), at a mean of 1.7 years (1 month to 6 years) postoperatively. None of the RLLs were progressive, and the presence of RLLs did not show a significant association with the survival and clinical score. RLLs were more frequently observed in hips without femoral head autografts than in those with autografts. Conclusion. The use of HXLPE cemented acetabular components in total hip arthroplasty demonstrated excellent clinical outcomes after ten years, and no RLLs were progressive, and their presence did not affect the outcome. Femoral head autografting did not negatively impact the acetabular component survival or the appearance of RLLs. Cite this article: Bone Joint J 2021;103-B(10):1604–1610

Aims. The number of convolutional neural networks (CNN) available for fracture detection and classification is rapidly increasing. External validation of a CNN on a temporally separate (separated by time) or geographically separate (separated by location) dataset is crucial to assess generalizability of the CNN before application to clinical practice in other institutions. We aimed to answer the following questions: are current CNNs for fracture recognition externally valid?; which methods are applied for external validation (EV)?; and, what are reported performances of the EV sets compared to the internal validation (IV) sets of these CNNs?. Methods. The PubMed and Embase databases were systematically searched from January 2010 to October 2020 according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The type of EV, characteristics of the external dataset, and diagnostic performance characteristics on the IV and EV datasets were collected and compared. Quality assessment was conducted using a seven-item checklist based on a modified Methodologic Index for NOn-Randomized Studies instrument (MINORS). Results. Out of 1,349 studies, 36 reported development of a CNN for fracture detection and/or classification. Of these, only four (11%) reported a form of EV. One study used temporal EV, one conducted both temporal and geographical EV, and two used geographical EV. When comparing the CNN’s performance on the IV set versus the EV set, the following were found: AUCs of 0.967 (IV) versus 0.975 (EV), 0.976 (IV) versus 0.985 to 0.992 (EV), 0.93 to 0.96 (IV) versus 0.80 to 0.89 (EV), and F1-scores of 0.856 to 0.863 (IV) versus 0.757 to 0.840 (EV). Conclusion. The number of externally validated CNNs in orthopaedic trauma for fracture recognition is still scarce. This greatly limits the potential for transfer of these CNNs from the developing institute to another hospital to achieve similar diagnostic performance. We recommend the use of geographical EV and statements such as the Consolidated Standards of Reporting Trials–Artificial Intelligence (CONSORT-AI), the Standard Protocol Items: Recommendations for Interventional Trials–Artificial Intelligence (SPIRIT-AI) and the Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis–Machine Learning (TRIPOD-ML) to critically appraise performance of CNNs and improve methodological rigor, quality of future models, and facilitate eventual implementation in clinical practice. Cite this article: Bone Jt Open 2021;2(10):879–885

Aims. The COVID-19 pandemic drastically affected elective orthopaedic services globally as routine orthopaedic activity was largely halted to combat this global threat. Our institution (University College London Hospital, UK) previously showed that during the first peak, a large proportion of patients were hesitant to be listed for their elective lower limb procedure. The aim of this study is to assess if there is a patient perception change towards having elective surgery now that we have passed the peak of the second wave of the pandemic. Methods. This is a prospective study of 100 patients who were on the waiting list of a single surgeon for an elective hip or knee procedure. Baseline characteristics including age, American Society of Anesthesiologists (ASA) grade, COVID-19 risk, procedure type, and admission type were recorded. The primary outcome was patient consent to continue with their scheduled surgical procedure. Subgroup analysis was also conducted to define if any specific patient factors influenced decision to continue with surgery. Results. Overall, 88 patients (88%) were happy to continue with their scheduled procedure at the earliest opportunity. Patients with an ASA grade I were most likely to agree to surgery, followed by patients with ASA grades II, then those with grade III (93.3%, 88.7%, and 78.6% willingness, respectively). Patients waitlisted for an injection were least likely to consent to surgery, with just 73.7% agreeing. In all, there was a large increase in the proportion of patient willingness to continue with surgery compared to our initial study during the first wave of the pandemic. Conclusion. As COVID-19 lockdown restrictions are lifted after the second peak of the pandemic, we are seeing greater willingness to continue with scheduled orthopaedic surgery, reinforcing a change in patient perception towards having elective surgery. However, we must continue with strict COVID-19 precautions in order to minimize viral transmission as we increase our elective orthopaedic services going forward. Cite this article: Bone Jt Open 2021;2(10):865–870

Aims. To compare rates of serious adverse events in patients undergoing revision knee arthroplasty with consideration of the indication for revision (urgent versus elective indications), and compare these with primary arthroplasty and re-revision arthroplasty. Methods. Patients undergoing primary knee arthroplasty were identified in the national Hospital Episode Statistics (HES) between 1 April 1997 to 31 March 2017. Subsequent revision and re-revision arthroplasty procedures in the same patients and same knee were identified. The primary outcome was 90-day mortality and a logistic regression model was used to investigate factors associated with 90-day mortality and secondary adverse outcomes, including infection (undergoing surgery), pulmonary embolism, myocardial infarction, and stroke. Urgent indications for revision arthroplasty were defined as infection or fracture, and all other indications (e.g. loosening, instability, wear) were included in the elective indications cohort. Results. A total of 939,021 primary knee arthroplasty procedures were included (939,021 patients), of which 40,854 underwent subsequent revision arthroplasty, and 9,100 underwent re-revision arthroplasty. Revision surgery for elective indications was associated with a 90-day rate of mortality of 0.44% (135/30,826; 95% confidence interval (CI) 0.37 to 0.52) which was comparable to primary knee arthroplasty (0.46%; 4,292/939,021; 95% CI 0.44 to 0.47). Revision arthroplasty for infection was associated with a much higher mortality of 2.04% (184/9037; 95% CI 1.75 to 2.35; odds ratio (OR) 3.54; 95% CI 2.81 to 4.46), as was revision for periprosthetic fracture at 5.25% (52/991; 95% CI 3.94 to 6.82; OR 6.23; 95% CI 4.39 to 8.85). Higher rates of pulmonary embolism, myocardial infarction, and stroke were also observed in the infection and fracture cohort. Conclusion. Patients undergoing revision arthroplasty for urgent indications (infection or fracture) are at higher risk of mortality and serious adverse events in comparison to primary knee arthroplasty and revision arthroplasty for elective indications. These findings will be important for patient consent and shared decision-making and should inform service design for this patient cohort. Cite this article: Bone Joint J 2021;103-B(10):1578–1585

Aims. Knee arthroplasty surgery is a highly effective treatment for arthritis and disorders of the knee. There are a wide variety of implant brands and types of knee arthroplasty available to surgeons. As a result of a number of highly publicized failures, arthroplasty surgery is highly regulated in the UK and many other countries through national registries, introduced to monitor implant performance, surgeons, and hospitals. With time, the options available within many brand portfolios have grown, with alternative tibial or femoral components, tibial insert materials, or shapes and patella resurfacings. In this study we have investigated the effect of the expansion of implant brand portfolios and where there may be a lack of transparency around a brand name. We also aimed to establish the potential numbers of compatible implant construct combinations. Methods. Hypothetical implant brand portfolios were proposed, and the number of compatible implant construct combinations was calculated. Results. A simple knee portfolio with cemented cruciate-retaining (CR) and posterior-stabilized (PS) components, with and without a patella, has four combinations. If there are two options available for each component, the numbers double for each option, resulting in 32 combinations. The effect of adding a third option multiplies the number by 1.3. Introducing compatible uncemented options, with the effect of hybrids, multiplies the number by 4. An implant portfolio with two femoral components (both in CR and PS), with two insert options and a patella, all in cemented and uncemented versions leads to 192 possible compatible implant construct combinations. There are implant brands available to surgeons with many more than two options. Conclusion. This study demonstrates that the addition of multiple variants within a knee brand portfolio leads to a large number (many hundreds) of compatible implant construct combinations. Revision rates of implant combinations are not currently reviewed at this level of granularity, leading to the risk of camouflage of true outcomes. Cite this article: Bone Joint J 2021;103-B(10):1555–1560

Aims. This study used an artificial neural network (ANN) model to determine the most important pre- and perioperative variables to predict same-day discharge in patients undergoing total knee arthroplasty (TKA). Methods. Data for this study were collected from the National Surgery Quality Improvement Program (NSQIP) database from the year 2018. Patients who received a primary, elective, unilateral TKA with a diagnosis of primary osteoarthritis were included. Demographic, preoperative, and intraoperative variables were analyzed. The ANN model was compared to a logistic regression model, which is a conventional machine-learning algorithm. Variables collected from 28,742 patients were analyzed based on their contribution to hospital length of stay. Results. The predictability of the ANN model, area under the curve (AUC) = 0.801, was similar to the logistic regression model (AUC = 0.796) and identified certain variables as important factors to predict same-day discharge. The ten most important factors favouring same-day discharge in the ANN model include preoperative sodium, preoperative international normalized ratio, BMI, age, anaesthesia type, operating time, dyspnoea status, functional status, race, anaemia status, and chronic obstructive pulmonary disease (COPD). Six of these variables were also found to be significant on logistic regression analysis. Conclusion. Both ANN modelling and logistic regression analysis revealed clinically important factors in predicting patients who can undergo safely undergo same-day discharge from an outpatient TKA. The ANN model provides a beneficial approach to help determine which perioperative factors can predict same-day discharge as of 2018 perioperative recovery protocols. Cite this article: Bone Joint J 2021;103-B(8):1358–1366

Aims. A multicentre, randomized, clinician-led, pragmatic, parallel-group orthopaedic trial of two surgical procedures was set up to obtain high-quality evidence of effectiveness. However, the trial faced recruitment challenges and struggled to maintain recruitment rates over 30%, although this is not unusual for surgical trials. We conducted a qualitative study with the aim of gathering information about recruitment practices to identify barriers to patient consent and participation to an orthopaedic trial. Methods. We collected 11 audio recordings of recruitment appointments and interviews of research team members (principal investigators and research nurses) from five hospitals involved in recruitment to an orthopaedic trial. We analyzed the qualitative data sets thematically with the aim of identifying aspects of informed consent and information provision that was either unclear, disrupted, or hindered trial recruitment. Results. Recruiters faced four common obstacles when recruiting to a surgical orthopaedic trial: patient preferences for an intervention; a complex recruitment pathway; various logistical issues; and conflicting views on equipoise. Clinicians expressed concerns that the trial may not show significant differences in the treatments, validating their equipoise. However, they experienced role conflicts due to their own preference and perceived patient preference for an intervention arm. Conclusion. This study provided initial information about barriers to recruitment to an orthopaedic randomized controlled trial. We shared these findings in an all-site investigators’ meeting and encouraged researchers to find solutions to identified barriers; this led to the successful completion of recruitment. Complex trials may benefit for using of a mixed-methods approach to mitigate against recruitment failure, and to improve patient participation and informed consent. Cite this article: Bone Jt Open 2021;2(8):631–637