Association of cord blood vitamin D with child neurodevelopment at 7 years of age.

AIM: The role of fetal vitamin D [25-hydroxyvitamin D (25(OH)D)], one of the nuclear steroid transcription regulators, and brain development is unclear. We previously found a weak but persistent association between cord blood 25(OH)D and child language abilities at 18 months and 4 years of age, but no association with cognition or behaviour. The aim of this study was to investigate the association between cord blood 25(OH)D and a range of neurodevelopmental outcomes in these same children at 7 years of age. METHODS: Cord blood samples from 250 Australian mother-child pairs were analysed for 25(OH)D by mass spectroscopy. Children underwent tests of cognition, language, academic abilities and executive functions with a trained assessor at 7 years of age. Caregivers completed questionnaires to rate their child's behaviour and executive functioning in the home environment. Associations between standardised 25(OH)D and outcomes were assessed using regression models, taking into account possible social and demographic confounders. RESULTS: Standardised 25(OH)D in cord blood was not associated with any test or parent-rated scores. Nor was there any association with the risk of having a poor test or parent-rated score. Likewise, cord blood 25(OH)D categorised as <25, 25-50 and >50 nmol/L was not associated with test scores or parent-rated scores. CONCLUSIONS: There was no evidence that cord blood vitamin D concentration or deficiency was associated with cognition, language, academic abilities, executive functioning or behaviour at 7 years of age.

Categorisation of continuous covariates for stratified randomisation: How should we adjust?

To obtain valid inference following stratified randomisation, treatment effects should be estimated with adjustment for stratification variables. Stratification sometimes requires categorisation of a continuous prognostic variable (eg, age), which raises the question: should adjustment be based on randomisation categories or underlying continuous values? In practice, adjustment for randomisation categories is more common. We reviewed trials published in general medical journals and found none of the 32 trials that stratified randomisation based on a continuous variable adjusted for continuous values in the primary analysis. Using data simulation, this article evaluates the performance of different adjustment strategies for continuous and binary outcomes where the covariate-outcome relationship (via the link function) was either linear or non-linear. Given the utility of covariate adjustment for addressing missing data, we also considered settings with complete or missing outcome data. Analysis methods included linear or logistic regression with no adjustment for the stratification variable, adjustment for randomisation categories, or adjustment for continuous values assuming a linear covariate-outcome relationship or allowing for non-linearity using fractional polynomials or restricted cubic splines. Unadjusted analysis performed poorly throughout. Adjustment approaches that misspecified the underlying covariate-outcome relationship were less powerful and, alarmingly, biased in settings where the stratification variable predicted missing outcome data. Adjustment for randomisation categories tends to involve the highest degree of misspecification, and so should be avoided in practice. To guard against misspecification, we recommend use of flexible approaches such as fractional polynomials and restricted cubic splines when adjusting for continuous stratification variables in randomised trials.

Forecasting the future burden of primary total shoulder replacement in Australia.

OBJECTIVE: To forecast the number of primary total shoulder replacements (TSR) in Australia to the year 2035, and associated costs. METHODS: De-identified TSR data for 2009-2019 were obtained from the Australian Orthopaedic Association National Joint Replacement Registry. Population data, including population projections to 2035, were obtained from the Australian Bureau of Statistics. Three forecasting scenarios were used: constant TSR rates from 2019 onwards (Scenario 1, conservative); continued growth in TSR rates using negative binomial regression (Scenario 2, exponential); and continued growth using negative binomial regression with monotone B-splines (Scenario 3, moderate). Healthcare costs were estimated using TSR projections and average procedure costs, inflated to 2035 Australian dollars. RESULTS: The use of TSR increased by 242% in Australia from 2009 to 2019 (from 1983 to 6789 procedures for people ≥40 years). Under Scenario 1, the incidence of TSR is conservatively projected to rise to 9676 procedures by 2035 (43% increase from 2019), at a cost of $AUD 312.6 million to the health system. Under Scenario 2, TSR incidence would increase to 45,295 procedures by 2035 (567% increase), costing $AUD 1.46 billion. Under Scenario 3, 28,257 TSR procedures are forecast in 2035 (316% increase) at a cost of $913 million. CONCLUSIONS: Recent growth in TSR likely relates to prosthesis improvements, greater surgeon proficiency, and expanded clinical indications. Under moderate and exponential scenarios that consider rising TSR rates and population projections, Australia would face three- to five-fold growth in procedures by 2035. This would have profound implications for the healthcare budget, clinical workforce, and infrastructure.

Short Message Service Reminder Nudge for Parents and Influenza Vaccination Uptake in Children and Adolescents With Special Risk Medical Conditions: The Flutext-4U Randomized Clinical Trial.

Importance: Children with chronic medical conditions are at increased risk of severe influenza. Uptake of influenza vaccination in children and adolescents with these identified special risk medical conditions (SRMCs) is suboptimal. Objective: To assess the effectiveness of Flutext-4U, a parent short message service (SMS) reminder nudge intervention, in increasing influenza immunization in children and adolescents with SRMCs. Design, Setting, and Participants: This randomized clinical trial was conducted at a tertiary pediatric hospital in Adelaide, South Australia, from April 15 to September 30, 2021. Children and adolescents aged 6 months to younger than 18 years with SRMCs and a subspecialist outpatient appointment over a 5-month period during the Australian seasonal influenza vaccination season (April-August 2021) were eligible to participate. Follow-up was until September 30, 2021. Interventions: Participants were randomly assigned (1:1 ratio) to control: clinician nudges (hospital vaccine availability, ease of access, and recommendation from hospital subspecialists) or SMS intervention (control conditions plus an additional SMS reminder nudge to parents), with randomization stratified by age group (<5 years, 5-14 years, or >14 to <18 years). Main Outcomes and Measures: The primary outcome was influenza vaccination, as confirmed by the Australian Immunisation Register. Results: A total of 600 participants (intervention group: 298 [49.7%]; mean [SD] age, 11.5 [4.6] years; 162 female participants [54.4%]; control group: 302 [50.3%]; mean [SD] age, 11.4 [4.7] years; 155 female participants [51.3%]) were included. Influenza vaccination was 38.6% (113 of 293) in the SMS intervention group compared with 26.2% (79 of 302) in the control group (adjusted odds ratio [aOR], 1.79; 95% CI, 1.27-2.55; P = .001). Time to vaccine receipt was significantly lower among SMS participants (adjusted hazard ratio, 1.67; 95% CI, 1.25-2.22; P < .001). For participants randomly assigned by June 15, a significantly greater proportion receiving the SMS intervention were vaccinated during the optimal delivery period April to June 30 (SMS group: 40.0% [76 of 190] vs 25.4% [50 of 197]; aOR, 1.97; 95% CI, 1.28-3.06; P = .002). Conclusions and Relevance: Results of this randomized clinical trial suggest that an additional SMS reminder nudge for parents delivered in the tertiary care hospital setting to children and adolescents with SMRCs resulted in higher influenza vaccine uptake compared with clinician nudges alone. Trial Registration: ANZCTR Identifier: ACTRN12621000463875.

A comparison of survival models for prediction of eight-year revision risk following total knee and hip arthroplasty.

BACKGROUND: There is increasing interest in the development and use of clinical prediction models, but a lack of evidence-supported guidance on the merits of different modelling approaches. This is especially true for time-to-event outcomes, where limited studies have compared the vast number of modelling approaches available. This study compares prediction accuracy and variable importance measures for four modelling approaches in prediction of time-to-revision surgery following total knee arthroplasty (TKA) and total hip arthroplasty (THA). METHODS: The study included 321,945 TKA and 151,113 THA procedures performed between 1 January 2003 and 31 December 2017. Accuracy of the Cox model, Weibull parametric model, flexible parametric model, and random survival forest were compared, with patient age, sex, comorbidities, and prosthesis characteristics considered as predictors. Prediction accuracy was assessed using the Index of Prediction Accuracy (IPA), c-index, and smoothed calibration curves. Variable importance rankings from the Cox model and random survival forest were also compared. RESULTS: Overall, the Cox and flexible parametric survival models performed best for prediction of both TKA (integrated IPA 0.056 (95% CI [0.054, 0.057]) compared to 0.054 (95% CI [0.053, 0.056]) for the Weibull parametric model), and THA revision. (0.029 95% CI [0.027, 0.030] compared to 0.027 (95% CI [0.025, 0.028]) for the random survival forest). The c-index showed broadly similar discrimination between all modelling approaches. Models were generally well calibrated, but random survival forest underfitted the predicted risk of TKA revision compared to regression approaches. The most important predictors of revision were similar in the Cox model and random survival forest for TKA (age, opioid use, and patella resurfacing) and THA (femoral cement, depression, and opioid use). CONCLUSION: The Cox and flexible parametric models had superior overall performance, although all approaches performed similarly. Notably, this study showed no benefit of a tuned random survival forest over regression models in this setting.

Do Dual-mobility Cups Reduce Revision Risk in Femoral Neck Fractures Compared With Conventional THA Designs? An International Meta-analysis of Arthroplasty Registries.

BACKGROUND: Dual-mobility cups in THA were designed to reduce prosthesis instability and the subsequent risk of revision surgery in high-risk patients, such as those with hip fractures. However, there are limited data from clinical studies reporting a revision benefit of dual-mobility over conventional THA. Collaboration between anthroplasty registries provides an opportunity to describe international practice variation and compare between-country, all-cause revision rates for dual-mobility and conventional THA. QUESTIONS/PURPOSES: We summarized observational data from multiple arthroplasty registries for patients receiving either a dual-mobility or conventional THA to ask: (1) Is dual-mobility use associated with a difference in risk of all-cause revision surgery compared with conventional THA? (2) Are there specific patient characteristics associated with dual-mobility use in the hip fracture population? (3) Has the use of dual-mobility constructs changed over time in patients receiving a THA for hip fracture? METHODS: Six member registries of the International Society of Arthroplasty Registries (from Australia, Denmark, Sweden, the Netherlands, the United Kingdom, and the United States) provided custom aggregate data reports stratified by acetabular cup type (dual-mobility or conventional THA) in primary THA for hip fracture between January 1, 2002, and December 31, 2019; surgical approach; and patient demographic data (sex, mean age, American Society of Anesthesiologists class, and BMI). The cumulative percent revision and mortality were calculated for each registry. To determine a global hazard ratio of all-cause revision for dual-mobility compared with conventional THA designs, we used a pseudoindividual patient data approach to pool Kaplan-Meier prosthesis revision data from each registry and perform a meta-analysis. The pseudoindividual patient data approach is a validated technique for meta-analysis of aggregate time-to-event survival data, such as revision surgery, from multiple sources. Data were available for 15,024 dual-mobility THAs and 97,200 conventional THAs performed for hip fractures during the study period. RESULTS: After pooling of complete Kaplan-Meier survival data from all six registries, the cumulative percent revision for conventional THA was 4.3% (95% confidence interval [CI] 4.2% to 4.5%) and 4.7% (95% CI 4.3% to 5.3%) for dual-mobility THA at 5 years. We did not demonstrate a lower risk of all-cause revision for patients receiving dual-mobility over conventional THA designs for hip fracture in the meta-analysis once between-registry differences were adjusted for (HR 0.96 [95% CI 0.86 to 1.06]). A lower proportion of dual-mobility procedures were revised for dislocation than conventional THAs (0.9% versus 1.4%) but a higher proportion were revised for infection (1.2% versus 0.8%). In most registries, a greater proportion of dual-mobility THA patients were older, had more comorbidities, and underwent a posterior approach compared with conventional THA (p < 0.001). The proportion of dual-mobility THA used to treat hip fractures increased in each registry over time and constituted 21% (2438 of 11,874) of all THA procedures in 2019. CONCLUSION: The proportion of dual-mobility THAs in patients with hip fractures increased over time, but there was large variation in use across countries represented here. Dual-mobility cups were not associated with a reduction in the overall risk of revision surgery in patients with hip fractures. A randomized controlled trial powered to detect the incidence of dislocation and subsequent revision surgery is required to clarify the efficacy of dual-mobility cups to treat hip fractures. LEVEL OF EVIDENCE: Level III, therapeutic study.

Periprosthetic fracture as a late mode of failure of the Anatomique Benoist Girard II femoral prosthesis.

AIM: The Anatomique Benoist Girard (ABG) II femoral implant was a commonly used stem for primary total hip replacement (THR) at our institution (Launceston, Tasmania Australia). We identified peri-prosthetic fracture as the main cause of late failure. METHODS: The late periprosthetic fracture rate for ABG II implants was reviewed with national statistics, using Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) data. National revision rates for periprosthetic fracture were used to compare ABG II with all other cementless femoral stems. RESULT: ABG II stems accounted for 1% (2719 implants) of all femoral stem implants in Australia during the 12-year review period, compared to 23% (587 implants) in Launceston Hospitals. Although the Launceston cumulative percent revision rate for the ABG II stem was lower than the National rate at all time points, the reasons for revision were similar. The most common reason for revision of ABG II was fracture (56.8%), followed by loosening (15.3%). This differs from the reasons for revision in other cementless prostheses (loosening 23.9%, fracture 20.8%, dislocation 18.7%). Cumulative percent revision rates from late periprosthetic fracture, were higher for the ABG II stem than other cementless femoral prostheses. CONCLUSION: This review of the AOANJRR has confirmed a local and national higher revision rate of the ABG II stem due to late periprosthetic fracture compared with other cementless stems. Stem design must be considered to reduce the risk of late periprosthetic fracture.

Bipolar Hemiarthroplasty Does Not Result in a Higher Risk of Revision Compared with Total Hip Arthroplasty for Displaced Femoral Neck Fractures: An Instrumental Variable Analysis of 36,118 Procedures from the Australian Orthopaedic Association National Joint Replacement Registry.

BACKGROUND: Previous randomized studies have suggested that there is no short-term difference between the risk of revision following total hip arthroplasty (THA) and hemiarthroplasty (HA) for hip fracture in elderly patients. The aim of the present study was to compare the long-term revision rates of primary THA and HA for femoral neck fracture in order to determine whether unipolar or bipolar HA increases the all-cause risk of revision in patients 50 to 79 years old. METHODS: Data for 36,188 patients who underwent primary arthroplasty, including 13,035 unipolar and 8,220 bipolar HAs and 14,863 THAs, from September 1, 1999, to December 31, 2019, were obtained from the Australian Orthopaedic Association National Joint Replacement Registry. Unadjusted analyses were performed, as well as analyses adjusted for age, sex, femoral cement, and procedure year. The primary outcome was time to first revision for any cause. Secondary analyses were performed for the reason for revision (i.e., infection, dislocation, and periprosthetic fracture). Instrumental variable analysis of hospital preference (for either HA or THA) was performed in order to mitigate the effect of any unmeasured confounding. All analyses were restricted to hospitals performing at least 10 procedures in the prior year. RESULTS: A total of 18,955 procedures were available for the comparison of modular unipolar HA to THA. Both the adjusted analysis performed with use of Cox proportional hazards (hazard ratio [HR], 1.94; 95% confidence interval [CI], 1.64 to 2.31; p < 0.001) and the instrumental variable analysis (HR, 2.82; 95% CI, 1.89 to 4.22; p < 0.001) demonstrated a higher risk of revision following modular unipolar HA compared with THA from 3 months postoperatively. A total of 13,168 procedures were available for the comparison of bipolar HA to THA. The adjusted analysis performed with use of Cox proportional hazards showed a significantly higher risk of revision for bipolar HA (HR, 1.29; 95% CI, 1.08 to 1.54; p = 0.01). The instrumental variable analysis showed a similar effect size that was not significant (HR, 1.27; 95% CI, 0.91 to 1.78; p = 0.16). CONCLUSIONS: Bipolar HA and THA demonstrated no significant difference in revision risk at long-term follow-up. Unipolar HA demonstrated higher risk of revision from 3 months postoperatively compared to THA. The higher risk of revision for dislocation observed following THA may be offset by the higher risk of revision for acetabular erosion or pain following bipolar HA, resulting in more equivalent revision risk. LEVEL OF EVIDENCE: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

An empirical analysis of dealing with patients who are lost to follow-up when developing prognostic models using a cohort design.

BACKGROUND: Researchers developing prediction models are faced with numerous design choices that may impact model performance. One key decision is how to include patients who are lost to follow-up. In this paper we perform a large-scale empirical evaluation investigating the impact of this decision. In addition, we aim to provide guidelines for how to deal with loss to follow-up. METHODS: We generate a partially synthetic dataset with complete follow-up and simulate loss to follow-up based either on random selection or on selection based on comorbidity. In addition to our synthetic data study we investigate 21 real-world data prediction problems. We compare four simple strategies for developing models when using a cohort design that encounters loss to follow-up. Three strategies employ a binary classifier with data that: (1) include all patients (including those lost to follow-up), (2) exclude all patients lost to follow-up or (3) only exclude patients lost to follow-up who do not have the outcome before being lost to follow-up. The fourth strategy uses a survival model with data that include all patients. We empirically evaluate the discrimination and calibration performance. RESULTS: The partially synthetic data study results show that excluding patients who are lost to follow-up can introduce bias when loss to follow-up is common and does not occur at random. However, when loss to follow-up was completely at random, the choice of addressing it had negligible impact on model discrimination performance. Our empirical real-world data results showed that the four design choices investigated to deal with loss to follow-up resulted in comparable performance when the time-at-risk was 1-year but demonstrated differential bias when we looked into 3-year time-at-risk. Removing patients who are lost to follow-up before experiencing the outcome but keeping patients who are lost to follow-up after the outcome can bias a model and should be avoided. CONCLUSION: Based on this study we therefore recommend (1) developing models using data that includes patients that are lost to follow-up and (2) evaluate the discrimination and calibration of models twice: on a test set including patients lost to follow-up and a test set excluding patients lost to follow-up.

Acetabular screws do not improve early revision rates in primary total hip arthroplasty. An instrumented registry analysis.

BACKGROUND: Initial stability of uncemented acetabular components in total hip arthroplasty (THA) is important for osseointegration and potentially enhanced by screw fixation. We used Australian Orthopaedic Association National Joint Replacement Registry data to determine whether screw usage influences uncemented acetabular component survival. METHODS: Primary THA with uncemented acetabular components performed for osteoarthritis from 1999 to 2018 was included. Survivorship was calculated using Kaplan-Meier estimates of cumulative percent revision (CPR). Comparisons used Cox proportional hazards method. An instrumental variable analysis adjusted for surgeon preference for screws as a confounding factor was used. RESULTS: Three hundred thirty thousand one hundred ninety-two THAs were included (31.8% with screws, 68.2% without). Two hundred twenty thousand six hundred seven were included in the instrumental variable analysis. Revision rate of acetabular components (all causes) was higher with screws during the first six years (hazard ratio (HR) = 1.45 (95% CI 1.34, 1.57), p < 0.001) and lower thereafter (HR = 0.81 (95% CI 0.67, 0.98), p = 0.027). Revision rate of acetabular components for loosening was higher with screws over the entire study period (HR = 1.73 (95% CI 1.51, 1.98), p < 0.001). Overall THA revision rate was higher with screws during the first six years (HR = 1.20 (95% CI 1.15, 1.26), p < 0.001) but lower thereafter (HR = 0.89 (95% CI 0.81, 0.98), p = 0.020). Revision rate for dislocation was higher with screws over the entire period (HR = 1.16 (95% CI 1.06, 1.26), p < 0.001). Instrumental variable analysis revealed higher revision rates with acetabular screws in the first six years. (HR = 1.18 (95% CI 1.09-1.29), p < 0.001). CONCLUSION: Screws did not confer a protective effect against acetabular loosening and were not associated with long-term negative consequences.

Revision Risk After Unipolar or Bipolar Hemiarthroplasty for Femoral Neck Fractures: An Instrumental Variable Analysis of 62,875 Procedures from the Australian Orthopaedic Association National Joint Replacement Registry.

BACKGROUND: There remains much international practice variation regarding the choice of a unipolar or bipolar prosthesis design for displaced femoral neck fractures that are treated with hemiarthroplasty. The purpose of the present study was to compare revision rates following primary hemiarthroplasty for femoral neck fracture to determine if the unipolar hemiarthroplasty design increases the risk of revision arthroplasty for all causes. METHODS: Instrumental variable analysis was performed with use of data that had been entered into the Australian Orthopaedic Association National Joint Replacement Registry from September 1, 1999, to December 31, 2018. Sixty-two thousand, eight hundred and seventy-five patients with femoral neck fractures that were treated with primary modular unipolar or bipolar hemiarthroplasty procedure were analyzed. Hospital preference for prosthesis design in the 12 months prior to the index procedure was used as an instrument to adjust for unmeasured confounding. The primary outcome was time to first revision for any cause. Secondary analyses were performed on the reason for revision (infection, dislocation, periprosthetic fracture, or acetabular erosion), the use of cement femoral stem fixation, and the type of stem (polished or matte). RESULTS: Modular unipolar hemiarthroplasty was associated with a higher rate of revision at >2.5 years (hazard ratio [HR], 1.86; 95% confidence interval [CI], 1.46 to 2.36; p < 0.001), but there was no difference between the groups before 2.5 years (HR, 0.98; 95% CI, 0.85 to 1.13; p = 0.79). Protective factors for revision included female sex (HR, 0.82; 95% CI, 0.74 to 0.9), use of cemented fixation (HR, 0.69; 95% CI, 0.62 to 0.77), and surgery performed in a public hospital setting (HR 0.79; 95% CI, 0.70 to 0.89). Modular unipolar prostheses had a greater risk of revision for acetabular erosion, particularly in later time periods (HR at ≥5.5 years, 5.10; 95% CI, 2.40 to 10.83; p < 0.001), while being protective against periprosthetic fractures (HR, 0.72; 95% CI, 0.59 to 0.87; p < 0.001) at all time points. There was no difference in terms of the risk of revision for infection, dislocation, or stem type. CONCLUSIONS: Bipolar hemiarthroplasty designs resulted in a lower risk of revision than unipolar designs. Unipolar hemiarthroplasties are justified for patients with femoral neck fracture and a shorter life expectancy (≤2.5 years). LEVEL OF EVIDENCE: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

What Is the Effect of Using a Competing-risks Estimator when Predicting Survivorship After Joint Arthroplasty: A Comparison of Approaches to Survivorship Estimation in a Large Registry.

BACKGROUND: There is increasing interest in the development of statistical models that can be used to estimate risk of adverse patient outcomes after joint arthroplasty. Competing risk approaches have been recommended to estimate risk of longer-term revision, which is often likely to be precluded by the competing risk of death. However, a common approach is to ignore the competing risk by treating death as a censoring event and using standard survival models such as Cox regression. It is well-known that this approach can overestimate the event risk for population-level estimates, but the impact on the estimation of a patient's individualized risk after joint arthroplasty has not been explored. QUESTIONS/PURPOSES: We performed this study to (1) determine whether using a competing risk or noncompeting risk method affects the accuracy of predictive models for joint arthroplasty revision and (2) determine the magnitude of difference that using a competing risks versus noncompeting risks approach will make to predicted risks for individual patients. METHODS: The predictive performance of a standard Cox model, with competing risks treated as censoring events, was compared with the performance of two competing risks approaches, the cause-specific Cox model and Fine-Gray model. Models were trained and tested using data pertaining to 531,304 TKAs and 274,618 THAs recorded in the Australian Orthopaedic Association National Joint Replacement Registry between January 1, 2003 and December 31, 2017. The registry is a large database with near-complete capture and follow-up of all hip and knee joint arthroplasty in Australia from 2003 onwards, making it an ideal setting for this study. The performance of the three modeling approaches was compared in two different prediction settings: prediction of the 10-year risk of all-cause revision after TKA and prediction of revision for periprosthetic fracture after THA. The calibration and discrimination of each approach were compared using the concordance index, integrated Brier scores, and calibration plots. Calibration of 10-year risk estimates was further assessed within subgroups of age by comparing the observed and predicted proportion of events. Estimated 10-year risks from each model were also compared in three hypothetical patients with different risk profiles to determine whether differences in population-level performance metrics would translate into a meaningful difference for individual patient predictions. RESULTS: The standard Cox and two competing risks models showed near-identical ability to distinguish between high-risk and low-risk patients (c-index 0.64 [95% CI, 0.64 to 0.64] for all three modeling approaches for TKAs and 0.66 [95% CI 0.66 to 0.66] for THA). All models performed similarly in patients younger than 75 years, but for patients aged 75 years and older, the standard Cox model overestimated the risk of revision more than the cause-specific Cox and Fine-Gray model did. These results were echoed when predictions were made for hypothetical individual patients. For patients with a low competing risk of mortality, the 10-year predicted risks from the standard Cox, cause-specific Cox, and Fine-Gray models were similar for TKAs and THAs. However, a larger difference was observed for hypothetical 89-year-old patients with increased mortality risk. In TKAs, the revision risk for an 89-year-old patient was so low that this difference was negligible (0.83% from the cause-specific Cox model versus 1.1% from the standard Cox model). However, for THAs, where older age is a risk factor for both death and revision for periprosthetic fracture, a larger difference was observed in the 10-year predicted risks for a hypothetical 89-year-old patient (3.4% from the cause-specific Cox model versus 5.2% from the standard Cox model). CONCLUSION: When developing models to predict longer-term revision of joint arthroplasty, failing to use a competing risks modeling approach will result in overestimating the revision risk for patients with a high risk of mortality during the surveillance period. However, even in an extreme instance, where both the frequency of the event of interest and the competing risk of death are high, the largest absolute difference in predicted 10-year risk for an individual patient was just 1.8%, which may not be of consequence to an individual. Despite these findings, when developing or using risk prediction models, researchers and clinicians should be aware of how competing risks were handled in the modeling process, particularly if the model is intended for use populations where the mortality risk is high. LEVEL OF EVIDENCE: Level III, therapeutic study.

The three-year survivorship of robotically assisted versus non-robotically assisted unicompartmental knee arthroplasty.

AIM: There has been a significant reduction in unicompartmental knee arthroplasty (UKA) procedures recorded in Australia. This follows several national joint registry studies documenting high UKA revision rates when compared to total knee arthroplasty (TKA). With the recent introduction of robotically assisted UKA procedures, it is hoped that outcomes improve. This study examines the cumulative revision rate of UKA procedures implanted with a newly introduced robotic system and compares the results to one of the best performing non-robotically assisted UKA prostheses, as well as all other non-robotically assisted UKA procedures. METHODS: Data from the Australian Orthopaedic Association National Joint Arthroplasty Registry (AOANJRR) for all UKA procedures performed for osteoarthritis (OA) between 2015 and 2018 were analyzed. Procedures using the Restoris MCK UKA prosthesis implanted using the Mako Robotic-Arm Assisted System were compared to non-robotically assisted Zimmer Unicompartmental High Flex Knee System (ZUK) UKA, a commonly used UKA with previously reported good outcomes and to all other non-robotically assisted UKA procedures using Cox proportional hazard ratios (HRs) and Kaplan-Meier estimates of survivorship. RESULTS: There was no difference in the rate of revision when the Mako-assisted Restoris UKA was compared to the ZUK UKA (zero to nine months: HR 1.14 (95% CI 0.71 to 1.83; p = 0.596) vs nine months and over: HR 0.66 (95% CI 0.42 to 1.02; p = 0.058)). The Mako-assisted Restoris had a significantly lower overall revision rate compared to the other types of non-robotically assisted procedures (HR 0.58 (95% confidence interval (CI) 0.42 to 0.79); p < 0.001) at three years. Revision for aseptic loosening was lower for the Mako-assisted Restoris compared to all other non-robotically assisted UKA (entire period: HR 0.34 (95% CI 0.17 to 0.65); p = 0.001), but not the ZUK prosthesis. However, revision for infection was significantly higher for the Mako-assisted Restoris compared to the two comparator groups (ZUK: entire period: HR 2.91 (95% CI 1.22 to 6.98; p = 0.016); other non-robotically assisted UKA: zero to three months: HR 5.57 (95% CI 2.17 to 14.31; p < 0.001)). CONCLUSION: This study reports comparable short-term survivorship for the Mako robotically assisted UKA compared to the ZUK UKA and improved survivorship compared to all other non-robotic UKA. These results justify the continued use and investigation of this procedure. However, the higher rate of early revision for infection for robotically assisted UKA requires further investigation. Cite this article: Bone Joint J 2020;102-B(3):319-328.

Postmarket surveillance of arthroplasty device components using machine learning methods.

PURPOSE: While joint arthroplasty is generally a safe and effective procedure, there are concerns that some devices are at increased risk of failure. Early identification of total hip arthroplasty devices with increased risk of failure can be challenging because devices consist of multiple components, hundreds of distinct components are currently used in surgery, and any estimated effect needs to address confounding due to device and patient factors. The purpose of this study was to assess the effectiveness of machine learning approaches at identifying recalled components listed by the US Food and Drug Administration using data from a US total joint arthroplasty registry. METHODS: An open cohort study was conducted using data (January 1, 2001, to December 31, 2015) from 74 520 implantations and 348 unique components in the Kaiser Permanente Total Joint Replacement Registry. Exposures of interest were device components used in elective primary total hip arthroplasty. The outcome was time to first revision surgery, defined as exchange, removal, or addition of any component. Machine learning methods included regularized/unregularized Cox models and random survival forest. RESULTS: Among the recalled components detected were ASR acetabular shell/large femoral head, Durom acetabular shell/Metasul large femoral head, and Rejuvenate modular neck stem. The three components not identified were characterized by small numbers of devices recorded in the registry. CONCLUSIONS: The novel approaches to signal detection may improve postmarket surveillance of frequently used arthroplasty devices, which in turn will improve public health.

The Effect of Surgeon Preference for Selective Patellar Resurfacing on Revision Risk in Total Knee Replacement: An Instrumental Variable Analysis of 136,116 Procedures from the Australian Orthopaedic Association National Joint Replacement Registry.

BACKGROUND: The optimum strategy regarding resurfacing the patella in total knee replacement (TKR) remains debated, with wide national and international variability. To minimize the confounders of a direct comparison of cases with or without resurfacing, we performed an instrumental variable analysis examining revision risk on the basis of surgeon preference for patellar resurfacing in TKR, specifically examining the subsequent patellar revision risk and all-cause revision risk among cases of surgeons preferring to perform resurfacing on a selective basis compared with those whose preference is to routinely resurface or who infrequently perform resurfacing. METHODS: Data from the Australian Orthopaedic Association National Joint Replacement Registry from 1999 to 2016 were obtained to assess the cumulative percent revision, hazard ratio, and revision diagnoses for 3 cohorts, grouped according the surgeons' patellar-resurfacing preferences: infrequently (<10% of the time), selectively (10% to <90% of the time), or routinely (≥90% of the time). To avoid confounding from implant design or surgeon performance, only minimally stabilized TKRs using hybrid or cemented fixation and performed by surgeons performing ≥50 TKR procedures per year were included. RESULTS: A total of 136,116 procedures were included, with patients in all 3 cohorts having similar demographics. The selectively resurfaced cohort had a higher patellar revision risk compared with the routinely resurfaced cohort, with this risk being greatest in the first 4.5 years following the primary procedure, including 306% higher in the first 1.5 years, and remaining 50% higher after 4.5 years. In a subgroup analysis, increased revision risk was observed among males, females, those <65 years of age, and those ≥65 years of age. The infrequently resurfaced cohort had the highest risk of subsequent patellar revision, up to 482% higher, as seen in the first 1.5 years after the primary procedure, compared with the routinely resurfaced cohort. Additionally, the risk of all-cause revision was 20% higher for the selectively resurfaced cohort compared with routinely resurfaced. CONCLUSIONS: Surgeons who preferred selective resurfacing of the patella had a higher risk of patellar revision than those who had preference for routine resurfacing of the patella. Overall, a greater preference for resurfacing resulted in a lower risk of patellar revision. LEVEL OF EVIDENCE: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Are Hooded, Crosslinked Polyethylene Liners Associated with a Reduced Risk of Revision After THA?

BACKGROUND: Hooded acetabular liners and head sizes ≥ 32 mm have both shown to have a beneficial effect on the revision rate for dislocation in THA. Experience with noncrosslinked polyethylene (nonXLPE) raised concerns regarding the risk of impingement damage, loosening, and osteolysis with hooded liners; however, the evidence for this in crosslinked polyethylene (XLPE) is inconclusive. The interaction between different femoral head sizes and hooded liners is not well understood, and it is unclear whether hooded XLPE liners have a beneficial effect on overall long-term survivorship. QUESTIONS/PURPOSES: We analyzed a large national joint registry to ask: (1) Is the use of hooded XLPE liners associated with a reduced revision rate for dislocation compared with nonhooded liners? (2) Is there a difference in the revision rate for aseptic loosening/osteolysis? (3) Is head size associated with any difference in the revision rate between hooded and nonhooded liners? METHODS: The Australian Orthopaedic Association National Joint Replacement Registry longitudinally maintains data on all primary and revision joint arthroplasties with nearly 100% capture. We analyzed all conventional primary THAs performed from registry inception in September 1999 until December 31, 2016 in patients with a diagnosis of osteoarthritis who had nonhooded or hooded XLPE bearings in a cementless acetabular shell. The study group included 192,659 THA procedures with XLPE liners, of which 67,904 were nonhooded and 124,755 were hooded. The mean age of patients receiving nonhooded liners was 70 years (range, 11-100 years); 44% were males. This was similar to the patients with hooded liners, who had a mean age of 70 years (range, 16-100 years); 45% were males. The main outcome measure was the cumulative percent revision at 15 years of the THA using Kaplan-Meier estimates of survivorship. We examined reasons for revision and and performed multivariable analysis to control for the confounding factors of three head size groups (< 32mm, 32mm, and > 32mm) and for the method of femoral fixation. RESULTS: There was a higher revision rate for dislocation for patients with nonhooded liners at all times to 15 years (HR, 1.31; 95% CI, 1.17-1.47; p < 0.001). There was a higher revision rate for the diagnosis of aseptic loosening/osteolysis with patients with nonhooded liners compared with hooded liners (HR, 1.19; 95% CI, 1.05-1.34; p = 0.006). Head sizes of 32 mm or larger were independently associated with a lower comparative revision rate between hooded and nonhooded liners, but this was not apparent for head sizes smaller than 32 mm. It appeared that the main driver of the finding in larger heads was a reduced dislocation risk with hooded liners for 32 mm heads (HR, 1.50; 95% CI, 1.23-1.80; p < 0.001) and for heads larger than 32 mm (HR, 1.50; 95% CI, 1.20-1.89; p < 0.001). CONCLUSIONS: Prior research has suggested that hooded acetabular liners may be associated with impingement, loosening, and osteolysis; however, in this large, registry-based report we found that XLPE hooded liners are not associated with an increased revision rate for aseptic loosening/osteolysis. Although there are many potential confounding variables in this registry analysis, if anything, surgeons using larger femoral heads and hooded liners likely did so in patients with a higher perceived dislocation risk. Patients with larger heads and XLPE hooded liners were, however, less likely to experience revision for dislocation. These liners therefore appear reasonable to use in primary THA at the surgeon's discretion. LEVEL OF EVIDENCE: Level III, therapeutic study.

What Is the Risk of Repeat Revision When Patellofemoral Replacement Is Revised to TKA? An Analysis of 482 Cases From a Large National Arthroplasty Registry.

BACKGROUND: Patellofemoral replacements (PFRs) have a higher rate of revision than unicompartmental knee arthroplasty or TKA. However, there is little information regarding why PFRs are revised, the components used for these revisions, or the outcome of the revision procedure. Some contend that PFR is a bridging procedure that can easily be revised to a TKA with similar results as a primary TKA; however, others dispute this suggestion. QUESTIONS/PURPOSES: (1) In the setting of a large national registry, what were the reasons for revision of PFR to TKA and was the level of TKA constraint used in the revision associated with a subsequent risk of rerevision? (2) Is the risk of revision of the TKA used to revise a PFR greater than the risk of revision after a primary TKA and greater than the risk of rerevision after revision TKA? METHODS: Data were obtained from the Australian Orthopaedic Association Joint Replacement Registry through December 31, 2016, for TKA revision procedures after PFR. Because revisions for infection may be staged procedures resulting in further planned operations, for the revision analyses, these were excluded. There were 3251 PFRs, 482 of which were revised to TKA during the 17-year study period. The risk of second revision was calculated using Kaplan-Meier estimates of survivorship for PFRs revised to TKAs, and that risk was compared with the risk of first revision after TKA and also with the risk of a second revision after revision TKA. Hazard ratios (HRs) from Cox proportional hazards models were used to compare second revision rates among the different levels of prosthesis constraint used in the index revision after PFR (specifically, cruciate-retaining versus cruciate-substituting). RESULTS: The main reasons for revising a PFR to TKA were progression of disease (56%), loosening (17%), and pain (12%). With the numbers available for analysis, there was no difference in the risk of a second revision when a PFR was revised to a cruciate-retaining TKA than when it was revised to a cruciate-substituting TKA (HR, 1.24 [0.65-2.36]; p = 0.512). A total of 204 (42%) of the PFR revisions had the patella component revised when the PFR was converted to a TKA. There difference in rates of second revision when the patella component was revised or not revised (HR, 1.01 [0.55-1.85]; p = 0.964). When we eliminated the devices that ceased to be used before 2005 (older devices), we found no change in the overall risk of repeat revision. The risk of a PFR that was revised to a TKA undergoing a second revision was greater than the risk of TKA undergoing a first revision (HR, 2.39 [1.77-3.24]; p < 0.001), but it was less than the risk of a revision TKA undergoing a second revision (HR, 0.60 [0.43-0.81]; p = 0.001). CONCLUSIONS: The risk of second revision when a PFR is revised is not altered if cruciate-retaining or posterior-stabilized TKA is used for the revision nor if the patella component is revised or not revised. The risk of repeat revision after revision of a PFR to a TKA was much higher than the risk of revision after a primary TKA, and these findings did not change when we analyzed only devices in use since 2005. When PFR is used for the management of isolated patellofemoral osteoarthritis, patients should be counselled not only about the high revision rate of the primary procedure, but also the revision rate after TKA. Further studies regarding the functional outcomes of these procedures may help clarify the value of PFRs and subsequent revisions. LEVEL OF EVIDENCE: Level III, therapeutic study.

Outcomes of hip and knee replacement surgery in private and public hospitals in Australia.

BACKGROUND: This study determined the contributing factors of hospital sector (private versus public) variation in revision rates after elective total hip replacement (THR) for hip fracture, and elective total knee replacement (TKR). METHODS: Using data from a large national arthroplasty registry, funnel plots for hospitals were generated, displaying the proportion of revised primary procedures. The proportion of outliers for each distribution was defined as the proportion outside the upper 99.7% confidence limit. Survival analyses determined differences between hospital sector revision rates separately for implants with the lowest revision rate, and for all other implants. Multivariate Cox regression determined the role of hospital sector in revision, adjusting for possible confounders. RESULTS: For THR performed for osteoarthritis, 17.4% of private and 4.4% of public hospitals were outliers. For TKR performed for osteoarthritis, 19.6% of private and 10.0% of public hospitals were outliers. For THR for fractured neck of femur, 8.1% of private and 0.0% of public hospitals were outliers. Adjusted and unadjusted Kaplan-Meier analyses showed higher THR revision rates in private hospitals for osteoarthritis and fractured neck of femur, but no difference when restricted to the 10 prostheses with the lowest revision rate. The Kaplan-Meier analysis of TKR showed higher revision rates for private hospitals, with the association reversing when restricted to prostheses with the lowest revision rate. CONCLUSIONS: Considerable variation was seen in the revision rate after THR and TKR between hospital sectors in Australia. The variation was largely due to differences in prosthesis selection.

Impact of Image-Derived Instrumentation on Total Knee Arthroplasty Revision Rates: An Analysis of 83,823 Procedures from the Australian Orthopaedic Association National Joint Replacement Registry.

BACKGROUND: Computer navigation and image-derived instrumentation (IDI) are technology-based methods developed to improve outcomes and potentially reduce revision total knee arthroplasty (TKA). IDI refers to the use of manufactured, patient-specific surgical jigs. Conflicting reports exist on IDI-associated improvements in outcomes. The primary aim of the current study was to compare the rates of revision among TKA cases in which components were initially implanted with use of IDI, computer navigation, or neither of these methods ("other" TKA). The secondary aim was to determine whether the outcomes of IDI differed for specific subgroups. METHODS: Data were obtained from the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) for the 3 TKA groups: IDI, computer-navigated, and other TKA. The study period was from the first IDI procedure recorded by the AOANJRR (April 2010) to December 31, 2016. The analysis was restricted to primary TKA cases undertaken for osteoarthritis and involving patellar resurfacing and the use of a cross-linked polyethylene insert. Subanalyses were performed to evaluate the effects of age, sex, implantation method, IDI manufacturer, prosthetic design, and prosthesis type on the rates of revision. Kaplan-Meier estimates of survivorship described the time to first revision. Hazard ratios (HRs, Cox proportional hazards models) with adjustment for age and sex were used to compare revision rates. RESULTS: IDI was used in 5,486 primary TKA procedures. There was no significant difference among the groups in the cumulative percent revision (CPR) at 5 years: 3.3% (95% confidence interval [CI], 2.4% to 4.6%) for IDI, 2.4% (95% CI, 2.2% to 2.7%) for the computer-navigated group, and 2.5% (95% CI, 2.3% to 2.7%) for other TKA. Posterior-stabilized TKA with use of the IDI method had a significantly higher rate of revision at >3 months (HR, 1.45 [95% CI, 1.02 to 2.04]; p = 0.036), as did IDI TKA in the ≤65-year-old patient cohort (HR, 1.52 [95% CI, 1.10 to 2.09]; p = 0.010), compared with computer-navigated TKA. Patellar revision was significantly more likely in the IDI group. CONCLUSIONS: IDI TKA demonstrated no overall difference in early to mid-term revision rates compared with standard implantation methods. However, elevated rates of revision were seen with posterior-stabilized TKA, in patients ≤65 years of age, and for patellar revision, meaning that this method should be used with some caution and requires further study. LEVEL OF EVIDENCE: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Unicompartmental Knee Arthroplasty Revision to TKA: Are Tibial Stems and Augments Associated With Improved Survivorship?

BACKGROUND: Some surgeons contend that unicompartmental knee arthroplasty (UKA) can easily be revised to a TKA when revision is called for, whereas others believe that this can be complex and technically demanding. There has been little research regarding the efficacy or rationale of using metal augmentation and tibial stem extensions when revising a UKA to a TKA. QUESTION/PURPOSES: (1) Is the use of stem extensions for the tibial component associated with increased survival when revising a UKA to a TKA? (2) Is the addition of modular augments associated with increased survival compared with stem extensions alone? (3) Is TKA design (minimally stabilized versus posterior-stabilized) or (4) tibial fixation (cemented versus cementless) associated with differences in survivorship? METHODS: Data from the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) were used to analyze implant survival after revision of a UKA to a TKA, comparing results in which tibial components were used with and without modular components. The groups analyzed were TKA without a stem extension, those in which a tibial stem extension was used, and those in which a tibial stem extension was used together with an augment. There were 4438 revisions of UKAs to TKAs available for analysis. The mean duration of followup of patients having the TKA revisions was 5 years (SD, 3.5 years). There were 2901 (65%) procedures in which a tibial stem extension was not used, 870 (20%) procedures with a tibial stem extension, and 667 (15%) with a tibial stem extension and metallic augment. Kaplan-Meier estimates of survivorship were calculated and hazard ratios (HRs) from Cox proportional hazard models, adjusting for age and sex, were used to compare the rate of revision among groups. The overall 10-year cumulative percent revision (CPR) for UKA revised to a TKA was 16%. RESULTS: At 10 years, the CPR was increased when a stem extension was not used (19%; 95% confidence interval [CI],16.5-20.7 without a stem extension compared with 13%; 95% CI, 9.2-17.0 with a stem extension; entire period HR, 1.44; 95% CI, 1.10-1.89; p = 0.007). There was no difference in the 10-year CPR when an augment was used together with a stem extension compared with a stem extension alone (HR, 1.26; 95% CI, 0.85-1.86; p = 0.251). When minimally stabilized and posterior-stabilized TKAs were compared, there was no difference in survivorship. Minimally stabilized TKA designs without stem extensions showed higher CPR compared with when stem extensions were used (HR, 1.77; 95% CI, 1.16-2.70; p = 0.007), whereas posterior-stabilized designs without stem extensions showed higher CPR only when compared with when stem extensions and augments were both used (HR, 2.16; 95% CI, 1.24-3.77; p = 0.006). Cementless fixation of the tibial component resulted in a higher CPR than when cement was used (HR, 1.36; 95% CI 1.08-1.71; p = 0.008). CONCLUSIONS: In this registry study, the risk of repeat revision after revision of a UKA to a TKA was lower when a tibial stem extension was used, but no such difference was found with respect to augments. Our study did not account for the degree of bone loss or surgeon preference when considering stems and augments. Further research to establish the degree of bone loss associated with UKA to TKA revision procedures will help clarify these findings. LEVEL OF EVIDENCE: Level III, therapeutic study.