Aims. Several previously identified patient-, injury-, and treatment-related factors are associated with the development of nonunion in distal femur fractures. However, the predictive value of these factors is not well defined. We aimed to assess the predictive ability of previously identified risk factors in the development of nonunion leading to secondary surgery in distal femur fractures. Methods. We conducted a retrospective cohort study of adult patients with traumatic distal femur fracture treated with lateral locking plate between 2009 and 2018. The patients who underwent secondary surgery due to fracture healing problem or plate failure were considered having nonunion. Background knowledge of risk factors of distal femur fracture nonunion based on previous literature was used to form an initial set of variables. A logistic regression model was used with previously identified patient- and injury-related variables (age, sex, BMI, diabetes, smoking, periprosthetic fracture, open fracture, trauma energy, fracture zone length, fracture comminution, medial side comminution) in the first analysis and with treatment-related variables (different surgeon-controlled factors, e.g. plate length, screw placement, and proximal fixation) in the second analysis to predict the nonunion leading to secondary surgery in distal femur fractures. Results. We were able to include 299 fractures in 291 patients. Altogether, 31/299 fractures (10%) developed nonunion. In the first analysis, pseudo-R. 2. was 0.27 and area under the receiver operating characteristic curve (AUC) was 0.81. BMI was the most important variable in the prediction. In the second analysis, pseudo-R. 2. was 0.06 and AUC was 0.67. Plate length was the most important variable in the prediction. Conclusion. The model including patient- and injury-related factors had moderate fit and predictive ability in the prediction of distal femur fracture nonunion leading to secondary surgery. BMI was the most important variable in prediction of nonunion. Surgeon-controlled factors had a minor role in prediction of nonunion. Cite this article: Bone Jt Open 2023;4(8):584–593

Aims. Restoring the pre-morbid anatomy of the proximal humerus is a goal of anatomical shoulder arthroplasty, but reliance is placed on the surgeon’s experience and on anatomical estimations. The purpose of this study was to present a novel method, ‘Statistical Shape Modelling’, which accurately predicts the pre-morbid proximal humeral anatomy and calculates the 3D geometric parameters needed to restore normal anatomy in patients with severe degenerative osteoarthritis or a fracture of the proximal humerus. Materials and Methods. From a database of 57 humeral CT scans 3D humeral reconstructions were manually created. The reconstructions were used to construct a statistical shape model (SSM), which was then tested on a second set of 52 scans. For each humerus in the second set, 3D reconstructions of four diaphyseal segments of varying lengths were created. These reconstructions were chosen to mimic severe osteoarthritis, a fracture of the surgical neck of the humerus and a proximal humeral fracture with diaphyseal extension. The SSM was then applied to the diaphyseal segments to see how well it predicted proximal morphology, using the actual proximal humeral morphology for comparison. Results. With the metaphysis included, mimicking osteoarthritis, the errors of prediction for retroversion, inclination, height, radius of curvature and posterior and medial offset of the head of the humerus were 2.9° (± 2.3°), 4.0° (± 3.3°), 1.0 mm (± 0.8 mm), 0.8 mm (± 0.6 mm), 0.7 mm (± 0.5 mm) and 1.0 mm (± 0.7 mm), respectively. With the metaphysis excluded, mimicking a fracture of the surgical neck, the errors of prediction for retroversion, inclination, height, radius of curvature and posterior and medial offset of the head of the humerus were 3.8° (± 2.9°), 3.9° (± 3.4°), 2.4 mm (± 1.9 mm), 1.3 mm (± 0.9 mm), 0.8 mm (± 0.5 mm) and 0.9 mm (± 0.6 mm), respectively. Conclusion. This study reports a novel, computerised method that accurately predicts the pre-morbid proximal humeral anatomy even in challenging situations. This information can be used in the surgical planning and operative reconstruction of patients with severe degenerative osteoarthritis or with a fracture of the proximal humerus. Cite this article: Bone Joint J 2017;99-B:927–33

Objectives. Our objective was to predict the knee extension strength and post-operative function in quadriceps resection for soft-tissue sarcoma of the thigh. Methods. A total of 18 patients (14 men, four women) underwent total or partial quadriceps resection for soft-tissue sarcoma of the thigh between 2002 and 2014. The number of resected quadriceps was surveyed, knee extension strength was measured with the Biodex isokinetic dynamometer system (affected side/unaffected side) and relationships between these were examined. The Musculoskeletal Tumor Society (MSTS) score, Toronto Extremity Salvage Score (TESS), European Quality of Life-5 Dimensions (EQ-5D) score and the Short Form 8 were used to evaluate post-operative function and examine correlations with extension strength. The cutoff value for extension strength to expect good post-operative function was also calculated using a receiver operating characteristic (ROC) curve and Fisher’s exact test. Results. Extension strength decreased when the number of resected quadriceps increased (p < 0.001), and was associated with lower MSTS score, TESS and EQ-5D (p = 0.004, p = 0.005, p = 0.006, respectively). Based on the functional evaluation scales, the cutoff value of extension strength was 56.2%, the equivalent to muscle strength with resection of up to two muscles. Conclusion. Good post-operative results can be expected if at least two quadriceps muscles are preserved. Cite this article: A. Tanaka, Y. Yoshimura, K. Aoki, M. Kito, M. Okamoto, S. Suzuki, T. Momose, H. Kato. Knee extension strength and post-operative functional prediction in quadriceps resection for soft-tissue sarcoma of the thigh. Bone Joint Res 2016;5:232–238. DOI: 10.1302/2046-3758.56.2000631

Aims. Accurate skeletal age and final adult height prediction methods in paediatric orthopaedics are crucial for determining optimal timing of growth-guiding interventions and minimizing complications in treatments of various conditions. This study aimed to evaluate the accuracy of final adult height predictions using the central peak height (CPH) method with long leg X-rays and four different multiplier tables. Methods. This study included 31 patients who underwent temporary hemiepiphysiodesis for varus or valgus deformity of the leg between 2014 and 2020. The skeletal age at surgical intervention was evaluated using the CPH method with long leg radiographs. The true final adult height (FH. TRUE. ) was determined when the growth plates were closed. The final height prediction accuracy of four different multiplier tables (1. Bayley and Pinneau; 2. Paley et al; 3. Sanders – Greulich and Pyle (SGP); and 4. Sanders – peak height velocity (PHV)) was then compared using either skeletal age or chronological age. Results. All final adult height predictions overestimated the FH. TRUE. , with the SGP multiplier table having the lowest overestimation and lowest absolute deviation when using both chronological age and skeletal age. There were no significant differences in final height prediction accuracy between using skeletal age and chronological age with PHV (p = 0.652) or SGP multiplier tables (p = 0.969). Adult height predictions with chronological age and SGP (r = 0.769; p ≤ 0.001), as well as chronological age and PHV (r = 0.822; p ≤ 0.001), showed higher correlations with FH. TRUE. than predictions with skeletal age and SGP (r = 0.657; p ≤ 0.001) or skeletal age and PHV (r = 0.707; p ≤ 0.001). Conclusion. There was no significant improvement in adult height prediction accuracy when using the CPH method compared to chronological age alone. The study concludes that there is no advantage in routinely using the CPH method for skeletal age determination over the simple use of chronological age. The findings highlight the need for more accurate methods to predict final adult height in contemporary patient populations. Cite this article: Bone Jt Open 2023;4(10):750–757

Prediction tools are instruments which are commonly used to estimate the prognosis in oncology and facilitate clinical decision-making in a more personalized manner. Their popularity is shown by the increasing numbers of prediction tools, which have been described in the medical literature. Many of these tools have been shown to be useful in the field of soft-tissue sarcoma of the extremities (eSTS). In this annotation, we aim to provide an overview of the available prediction tools for eSTS, provide an approach for clinicians to evaluate the performance and usefulness of the available tools for their own patients, and discuss their possible applications in the management of patients with an eSTS. Cite this article: Bone Joint J 2022;104-B(9):1011–1016

Aims. This study aimed to compare the performance of survival prediction models for bone metastases of the extremities (BM-E) with pathological fractures in an Asian cohort, and investigate patient characteristics associated with survival. Methods. This retrospective cohort study included 469 patients, who underwent surgery for BM-E between January 2009 and March 2022 at a tertiary hospital in South Korea. Postoperative survival was calculated using the PATHFx3.0, SPRING13, OPTIModel, SORG, and IOR models. Model performance was assessed with area under the curve (AUC), calibration curve, Brier score, and decision curve analysis. Cox regression analyses were performed to evaluate the factors contributing to survival. Results. The SORG model demonstrated the highest discriminatory accuracy with AUC (0.80 (95% confidence interval (CI) 0.76 to 0.85)) at 12 months. In calibration analysis, the PATHfx3.0 and OPTIModel models underestimated survival, while the SPRING13 and IOR models overestimated survival. The SORG model exhibited excellent calibration with intercepts of 0.10 (95% CI -0.13 to 0.33) at 12 months. The SORG model also had lower Brier scores than the null score at three and 12 months, indicating good overall performance. Decision curve analysis showed that all five survival prediction models provided greater net benefit than the default strategy of operating on either all or no patients. Rapid growth cancer and low serum albumin levels were associated with three-, six-, and 12-month survival. Conclusion. State-of-art survival prediction models for BM-E (PATHFx3.0, SPRING13, OPTIModel, SORG, and IOR models) are useful clinical tools for orthopaedic surgeons in the decision-making process for the treatment in Asian patients, with SORG models offering the best predictive performance. Rapid growth cancer and serum albumin level are independent, statistically significant factors contributing to survival following surgery of BM-E. Further refinement of survival prediction models will bring about informed and patient-specific treatment of BM-E. Cite this article: Bone Joint J 2024;106-B(2):203–211

Aims. This systematic review aims to identify 3D predictors derived from biplanar reconstruction, and to describe current methods for improving curve prediction in patients with mild adolescent idiopathic scoliosis. Methods. A comprehensive search was conducted by three independent investigators on MEDLINE, PubMed, Web of Science, and Cochrane Library. Search terms included “adolescent idiopathic scoliosis”,“3D”, and “progression”. The inclusion and exclusion criteria were carefully defined to include clinical studies. Risk of bias was assessed with the Quality in Prognostic Studies tool (QUIPS) and Appraisal tool for Cross-Sectional Studies (AXIS), and level of evidence for each predictor was rated with the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) approach. In all, 915 publications were identified, with 377 articles subjected to full-text screening; overall, 31 articles were included. Results. Torsion index (TI) and apical vertebral rotation (AVR) were identified as accurate predictors of curve progression in early visits. Initial TI > 3.7° and AVR > 5.8° were predictive of curve progression. Thoracic hypokyphosis was inconsistently observed in progressive curves with weak evidence. While sagittal wedging was observed in mild curves, there is insufficient evidence for its correlation with curve progression. In curves with initial Cobb angle < 25°, Cobb angle was a poor predictor for future curve progression. Prediction accuracy was improved by incorporating serial reconstructions in stepwise layers. However, a lack of post-hoc analysis was identified in studies involving geometrical models. Conclusion. For patients with mild curves, TI and AVR were identified as predictors of curve progression, with TI > 3.7° and AVR > 5.8° found to be important thresholds. Cobb angle acts as a poor predictor in mild curves, and more investigations are required to assess thoracic kyphosis and wedging as predictors. Cumulative reconstruction of radiographs improves prediction accuracy. Comprehensive analysis between progressive and non-progressive curves is recommended to extract meaningful thresholds for clinical prognostication. Cite this article: Bone Jt Open 2024;5(3):243–251

Aims. To develop prediction models using machine-learning (ML) algorithms for 90-day and one-year mortality prediction in femoral neck fracture (FNF) patients aged 50 years or older based on the Hip fracture Evaluation with Alternatives of Total Hip arthroplasty versus Hemiarthroplasty (HEALTH) and Fixation using Alternative Implants for the Treatment of Hip fractures (FAITH) trials. Methods. This study included 2,388 patients from the HEALTH and FAITH trials, with 90-day and one-year mortality proportions of 3.0% (71/2,388) and 6.4% (153/2,388), respectively. The mean age was 75.9 years (SD 10.8) and 65.9% of patients (1,574/2,388) were female. The algorithms included patient and injury characteristics. Six algorithms were developed, internally validated and evaluated across discrimination (c-statistic; discriminative ability between those with risk of mortality and those without), calibration (observed outcome compared to the predicted probability), and the Brier score (composite of discrimination and calibration). Results. The developed algorithms distinguished between patients at high and low risk for 90-day and one-year mortality. The penalized logistic regression algorithm had the best performance metrics for both 90-day (c-statistic 0.80, calibration slope 0.95, calibration intercept -0.06, and Brier score 0.039) and one-year (c-statistic 0.76, calibration slope 0.86, calibration intercept -0.20, and Brier score 0.074) mortality prediction in the hold-out set. Conclusion. Using high-quality data, the ML-based prediction models accurately predicted 90-day and one-year mortality in patients aged 50 years or older with a FNF. The final models must be externally validated to assess generalizability to other populations, and prospectively evaluated in the process of shared decision-making. Cite this article: Bone Jt Open 2023;4(3):168–181

Aims. The aim of this study was to evaluate the reliability and validity of a patient-specific algorithm which we developed for predicting changes in sagittal pelvic tilt after total hip arthroplasty (THA). Methods. This retrospective study included 143 patients who underwent 171 THAs between April 2019 and October 2020 and had full-body lateral radiographs preoperatively and at one year postoperatively. We measured the pelvic incidence (PI), the sagittal vertical axis (SVA), pelvic tilt, sacral slope (SS), lumbar lordosis (LL), and thoracic kyphosis to classify patients into types A, B1, B2, B3, and C. The change of pelvic tilt was predicted according to the normal range of SVA (0 mm to 50 mm) for types A, B1, B2, and B3, and based on the absolute value of one-third of the PI-LL mismatch for type C patients. The reliability of the classification of the patients and the prediction of the change of pelvic tilt were assessed using kappa values and intraclass correlation coefficients (ICCs), respectively. Validity was assessed using the overall mean error and mean absolute error (MAE) for the prediction of the change of pelvic tilt. Results. The kappa values were 0.927 (95% confidence interval (CI) 0.861 to 0.992) and 0.945 (95% CI 0.903 to 0.988) for the inter- and intraobserver reliabilities, respectively, and the ICCs ranged from 0.919 to 0.997. The overall mean error and MAE for the prediction of the change of pelvic tilt were -0.3° (SD 3.6°) and 2.8° (SD 2.4°), respectively. The overall absolute change of pelvic tilt was 5.0° (SD 4.1°). Pre- and postoperative values and changes in pelvic tilt, SVA, SS, and LL varied significantly among the five types of patient. Conclusion. We found that the proposed algorithm was reliable and valid for predicting the standing pelvic tilt after THA. Cite this article: Bone Joint J 2024;106-B(1):19–27

Aims. This study was designed to develop a model for predicting bone mineral density (BMD) loss of the femur after total hip arthroplasty (THA) using artificial intelligence (AI), and to identify factors that influence the prediction. Additionally, we virtually examined the efficacy of administration of bisphosphonate for cases with severe BMD loss based on the predictive model. Methods. The study included 538 joints that underwent primary THA. The patients were divided into groups using unsupervised time series clustering for five-year BMD loss of Gruen zone 7 postoperatively, and a machine-learning model to predict the BMD loss was developed. Additionally, the predictor for BMD loss was extracted using SHapley Additive exPlanations (SHAP). The patient-specific efficacy of bisphosphonate, which is the most important categorical predictor for BMD loss, was examined by calculating the change in predictive probability when hypothetically switching between the inclusion and exclusion of bisphosphonate. Results. Time series clustering allowed us to divide the patients into two groups, and the predictive factors were identified including patient- and operation-related factors. The area under the receiver operating characteristic (ROC) curve (AUC) for the BMD loss prediction averaged 0.734. Virtual administration of bisphosphonate showed on average 14% efficacy in preventing BMD loss of zone 7. Additionally, stem types and preoperative triglyceride (TG), creatinine (Cr), estimated glomerular filtration rate (eGFR), and creatine kinase (CK) showed significant association with the estimated patient-specific efficacy of bisphosphonate. Conclusion. Periprosthetic BMD loss after THA is predictable based on patient- and operation-related factors, and optimal prescription of bisphosphonate based on the prediction may prevent BMD loss. Cite this article: Bone Joint Res 2024;13(4):184–192

Aims. The purpose of this study was to develop a personalized outcome prediction tool, to be used with knee arthroplasty patients, that predicts outcomes (lengths of stay (LOS), 90 day readmission, and one-year patient-reported outcome measures (PROMs) on an individual basis and allows for dynamic modifiable risk factors. Methods. Data were prospectively collected on all patients who underwent total or unicompartmental knee arthroplasty at a between July 2015 and June 2018. Cohort 1 (n = 5,958) was utilized to develop models for LOS and 90 day readmission. Cohort 2 (n = 2,391, surgery date 2015 to 2017) was utilized to develop models for one-year improvements in Knee Injury and Osteoarthritis Outcome Score (KOOS) pain score, KOOS function score, and KOOS quality of life (QOL) score. Model accuracies within the imputed data set were assessed through cross-validation with root mean square errors (RMSEs) and mean absolute errors (MAEs) for the LOS and PROMs models, and the index of prediction accuracy (IPA), and area under the curve (AUC) for the readmission models. Model accuracies in new patient data sets were assessed with AUC. Results. Within the imputed datasets, the LOS (RMSE 1.161) and PROMs models (RMSE 15.775, 11.056, 21.680 for KOOS pain, function, and QOL, respectively) demonstrated good accuracy. For all models, the accuracy of predicting outcomes in a new set of patients were consistent with the cross-validation accuracy overall. Upon validation with a new patient dataset, the LOS and readmission models demonstrated high accuracy (71.5% and 65.0%, respectively). Similarly, the one-year PROMs improvement models demonstrated high accuracy in predicting ten-point improvements in KOOS pain (72.1%), function (72.9%), and QOL (70.8%) scores. Conclusion. The data-driven models developed in this study offer scalable predictive tools that can accurately estimate the likelihood of improved pain, function, and quality of life one year after knee arthroplasty as well as LOS and 90 day readmission. Cite this article: Bone Joint J 2020;102-B(9):1183–1193

Aims. To develop and internally validate a preoperative clinical prediction model for acute adjacent vertebral fracture (AVF) after vertebral augmentation to support preoperative decision-making, named the after vertebral augmentation (AVA) score. Methods. In this prognostic study, a multicentre, retrospective single-level vertebral augmentation cohort of 377 patients from six Japanese hospitals was used to derive an AVF prediction model. Backward stepwise selection (p < 0.05) was used to select preoperative clinical and imaging predictors for acute AVF after vertebral augmentation for up to one month, from 14 predictors. We assigned a score to each selected variable based on the regression coefficient and developed the AVA scoring system. We evaluated sensitivity and specificity for each cut-off, area under the curve (AUC), and calibration as diagnostic performance. Internal validation was conducted using bootstrapping to correct the optimism. Results. Of the 377 patients used for model derivation, 58 (15%) had an acute AVF postoperatively. The following preoperative measures on multivariable analysis were summarized in the five-point AVA score: intravertebral instability (≥ 5 mm), focal kyphosis (≥ 10°), duration of symptoms (≥ 30 days), intravertebral cleft, and previous history of vertebral fracture. Internal validation showed a mean optimism of 0.019 with a corrected AUC of 0.77. A cut-off of ≤ one point was chosen to classify a low risk of AVF, for which only four of 137 patients (3%) had AVF with 92.5% sensitivity and 45.6% specificity. A cut-off of ≥ four points was chosen to classify a high risk of AVF, for which 22 of 38 (58%) had AVF with 41.5% sensitivity and 94.5% specificity. Conclusion. In this study, the AVA score was found to be a simple preoperative method for the identification of patients at low and high risk of postoperative acute AVF. This model could be applied to individual patients and could aid in the decision-making before vertebral augmentation. Cite this article: Bone Joint J 2022;104-B(1):97–102

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. Methods. We included 53,099 patients with cemented THA due to osteoarthritis from the Swedish Hip Arthroplasty Registry for model derivation and internal validation, as well as 125,428 patients from England and Wales recorded in the National Joint Register for England, Wales, Northern Ireland, the Isle of Man, and the States of Guernsey (NJR) for external model validation. A model was developed using a bootstrap ranking procedure with a least absolute shrinkage and selection operator (LASSO) logistic regression model combined with piecewise linear regression. Discriminative ability was evaluated by the area under the receiver operating characteristic curve (AUC). Calibration belt plots were used to assess model calibration. Results. A main effects model combining age, sex, American Society for Anesthesiologists (ASA) class, the presence of cancer, diseases of the central nervous system, kidney disease, and diagnosed obesity had good discrimination, both internally (AUC = 0.78, 95% confidence interval (CI) 0.75 to 0.81) and externally (AUC = 0.75, 95% CI 0.73 to 0.76). This model was superior to traditional models based on the Charlson (AUC = 0.66, 95% CI 0.62 to 0.70) and Elixhauser (AUC = 0.64, 95% CI 0.59 to 0.68) comorbidity indices. The model was well calibrated for predicted probabilities up to 5%. Conclusion. We developed a parsimonious model that may facilitate individualized risk assessment prior to one of the most common surgical interventions. We have published a web calculator to aid clinical decision-making. Cite this article: Bone Joint J 2021;103-B(3):469–478

Early and accurate prediction of hospital length-of-stay (LOS) in patients undergoing knee replacement is important for economic and operational reasons. Few studies have systematically developed a multivariable model to predict LOS. We performed a retrospective cohort study of 1609 patients aged ≥ 50 years who underwent elective, primary total or unicompartmental knee replacements. Pre-operative candidate predictors included patient demographics, knee function, self-reported measures, surgical factors and discharge plans. In order to develop the model, multivariable regression with bootstrap internal validation was used. The median LOS for the sample was four days (interquartile range 4 to 5). Statistically significant predictors of longer stay included older age, greater number of comorbidities, less knee flexion range of movement, frequent feelings of being down and depressed, greater walking aid support required, total (versus unicompartmental) knee replacement, bilateral surgery, low-volume surgeon, absence of carer at home, and expectation to receive step-down care. For ease of use, these ten variables were used to construct a nomogram-based prediction model which showed adequate predictive accuracy (optimism-corrected R. 2. = 0.32) and calibration. If externally validated, a prediction model using easily and routinely obtained pre-operative measures may be used to predict absolute LOS in patients following knee replacement and help to better manage these patients. . Cite this article: Bone Joint J 2013;95-B:1490–6

Aims. We report the use of the distal radius and ulna (DRU) classification for the prediction of peak growth (PG) and growth cessation (GC) in 777 patients with idiopathic scoliosis. We compare this classification with other commonly used parameters of maturity. Patients and Methods. The following data were extracted from the patients’ records and radiographs: chronological age, body height (BH), arm span (AS), date of menarche, Risser sign, DRU grade and status of the phalangeal and metacarpal physes. The mean rates of growth were recorded according to each parameter of maturity. PG was defined as the summit of the curve and GC as the plateau in deceleration of growth. The rates of growth at PG and GC were used for analysis using receiver operating characteristic (ROC) curves to determine the strength and cutoff values of the parameters of growth. Results. The most specific grades for PG using the DRU classification were radial grade 6 and ulnar grade 5, and for GC were radial grade 9 and ulnar grade 7. The DRU classification spanned both PG and GC, enabling better prediction of these clinically relevant stages than other methods. The rate of PG (≥ 0.7 cm/month) and GC (≤ 0.15 cm/month) was the same for girls and boys, in BH and AS measurements. Conclusion. This is the first study to note that the DRU classification can predict both PG and GC, providing evidence that it may aid the management of patients with idiopathic scoliosis. Cite this article: Bone Joint J 2016;98-B:1689–96

Previously, we showed that case-specific non-linear finite element (FE) models are better at predicting the load to failure of metastatic femora than experienced clinicians. In this study we improved our FE modelling and increased the number of femora and characteristics of the lesions. We retested the robustness of the FE predictions and assessed why clinicians have difficulty in estimating the load to failure of metastatic femora. A total of 20 femora with and without artificial metastases were mechanically loaded until failure. These experiments were simulated using case-specific FE models. Six clinicians ranked the femora on load to failure and reported their ranking strategies. The experimental load to failure for intact and metastatic femora was well predicted by the FE models (R. 2. = 0.90 and R. 2. = 0.93, respectively). Ranking metastatic femora on load to failure was well performed by the FE models (τ = 0.87), but not by the clinicians (0.11 < τ < 0.42). Both the FE models and the clinicians allowed for the characteristics of the lesions, but only the FE models incorporated the initial bone strength, which is essential for accurately predicting the risk of fracture. Accurate prediction of the risk of fracture should be made possible for clinicians by further developing FE models.

Aims

To identify variables independently associated with same-day discharge (SDD) of patients following revision total knee arthroplasty (rTKA) and to develop machine learning algorithms to predict suitable candidates for outpatient rTKA.

Aims

Current levels of hip fracture morbidity contribute greatly to the overall burden on health and social care services. Given the anticipated ageing of the population over the coming decade, there is potential for this burden to increase further, although the exact scale of impact has not been identified in contemporary literature. We therefore set out to predict the future incidence of hip fracture and help inform appropriate service provision to maintain an adequate standard of care.

There is increasing popularity in the use of artificial intelligence and machine-learning techniques to provide diagnostic and prognostic models for various aspects of Trauma & Orthopaedic surgery. However, correct interpretation of these models is difficult for those without specific knowledge of computing or health data science methodology. Lack of current reporting standards leads to the potential for significant heterogeneity in the design and quality of published studies. We provide an overview of machine-learning techniques for the lay individual, including key terminology and best practice reporting guidelines.

Cite this article: Bone Joint J 2021;103-B(12):1754–1758.

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).

Objectives

Unicompartmental knee arthroplasty (UKA) is an alternative to total knee arthroplasty with isolated medial or lateral compartment osteoarthritis. However, polyethylene wear can significantly reduce the lifespan of UKA. Different bearing designs and materials for UKA have been developed to change the rate of polyethylene wear. Therefore, the objective of this study is to investigate the effect of insert conformity and material on the predicted wear in mobile-bearing UKA using a previously developed computational wear method.

Aims

The purpose of this study was to develop a prognostic model for predicting survival of patients undergoing surgery owing to metastatic bone disease (MBD) in the appendicular skeleton.

Hip displacement, defined in this study as a migration percentage (MP) of more than 40%, is a common, debilitating complication of cerebral palsy (CP). In this prospective study we analysed the risk of developing hip displacement within five years of the first pelvic radiograph.

All children with CP in southern and western Sweden are invited to register in the hip surveillance programme CPUP. Inclusion criteria for the two groups in this study were children from the CPUP database born between 1994 and 2009 with Gross Motor Function Classification System (GMFCS) III to V. Group 1 included children who developed hip displacement, group 2 included children who did not develop hip displacement over a minimum follow-up of five years. A total of 145 children were included with a mean age at their initial pelvic radiograph of 3.5 years (0.6 to 9.7).

The odds ratio for hip displacement was calculated for GMFCS-level, age and initial MP and head-shaft angle. A risk score was constructed with these variables using multiple logistic regression analysis. The predictive ability of the risk score was evaluated using the area under the receiver operating characteristics curve (AUC).

All variables had a significant effect on the risk of a MP > 40%. The discriminatory accuracy of the CPUP hip score is high (AUC = 0.87), indicating a high ability to differentiate between high- and low-risk individuals for hip displacement. The CPUP hip score may be useful in deciding on further follow-up and treatment in children with CP.

Cite this article: Bone Joint J 2015;97-B:1441–4.

We suggest that different mechanisms underlie joint pain at rest and on movement in osteoarthritis and that separate assessment of these two features with a visual analogue scale (VAS) offers better information about the likely effect of a total knee replacement (TKR) on pain. The risk of persistent pain after TKR may relate to the degree of central sensitisation before surgery, which might be assessed by determining the pain threshold to an electrical stimulus created by a special tool, the Pain Matcher. Assessments were performed in 69 patients scheduled for TKR. At 18 months after operation, separate assessment of pain at rest and with movement was again carried out using a VAS in order to enable comparison of pre- and post-operative measurements. A less favourable outcome in terms of pain relief was observed for patients with a high pre-operative VAS score for pain at rest and a low pain threshold, both features which may reflect a central sensitisation mechanism.

Aims

Our aim was to investigate the predictive factors for the development of a rebound phenomenon after temporary hemiepiphysiodesis in children with genu valgum.

Aims

Our aim was to examine the Elixhauser and Charlson comorbidity indices, based on administrative data available before surgery, and to establish their predictive value for mortality for patients who underwent hip arthroplasty in the management of a femoral neck fracture.

This study demonstrates a significant correlation between the American Knee Society (AKS) Clinical Rating System and the Oxford Knee Score (OKS) and provides a validated prediction tool to estimate score conversion.

A total of 1022 patients were prospectively clinically assessed five years after TKR and completed AKS assessments and an OKS questionnaire. Multivariate regression analysis demonstrated significant correlations between OKS and the AKS knee and function scores but a stronger correlation (r = 0.68, p < 0.001) when using the sum of the AKS knee and function scores. Addition of body mass index and age (other statistically significant predictors of OKS) to the algorithm did not significantly increase the predictive value.

The simple regression model was used to predict the OKS in a group of 236 patients who were clinically assessed nine to ten years after TKR using the AKS system. The predicted OKS was compared with actual OKS in the second group. Intra-class correlation demonstrated excellent reliability (r = 0.81, 95% confidence intervals 0.75 to 0.85) for the combined knee and function score when used to predict OKS.

Our findings will facilitate comparison of outcome data from studies and registries using either the OKS or the AKS scores and may also be of value for those undertaking meta-analyses and systematic reviews.

Cite this article: Bone Joint J 2015;97-B:503–9.

The Mangled Extremity Severity Score was applied to 152 patients with severely injured lower limbs. All cases with a score of seven or more required amputation; some with scores of less than seven eventually came to amputation. These observations are discussed.

A prospective study of factors which might help to predict mortality in patients with intracapsular fractures of the femoral neck has been undertaken. A multivariate analysis technique was used to analyse the collected data, and it was found that mental ability was the most significant variable; this factor had the greatest effect on outcome.

We measured the proximal migration of 265 acetabular cups over seven years and correlated the findings with clinical outcome and acetabular revision for aseptic loosening. Cups which eventually became aseptically loose were shown to migrate more rapidly than successful cups. The average proximal migration at two years postoperatively for four groups of cups showed a monotonic relationship to the acetabular revision rate for aseptic loosening at 6.5 years. We conclude that acetabular cups which develop aseptic loosening as evidenced by pain, revision or screw fracture show increased proximal migration by one year, and that the 'migration rate' at two years can be used to predict the acetabular revision rate from aseptic loosening at 6.5 years.

The uptake of 99mTc-MDP was studied in 73 patients after a tibial fracture. The image obtained five minutes after injection during a period between one and four weeks after fracture was found to be related to the incidence of non-union after six months. A ratio of 1.3 between the uptake at the fracture site and at normal bone adjacent to it predicted non-union in an individual patient with a sensitivity of about 70% and a specificity of 90%.

We have studied the correlation between the prevention of progressive collapse and the ratio of the intact articular surface of the femoral head, after transtrochanteric rotational osteotomy for osteonecrosis. We used probit analysis on 125 hips in order to assess the ratio necessary to prevent progressive radiological collapse over a ten-year period. The results show that a minimum postoperative intact ratio of 34% was required. This critical ratio may be useful for surgical planning and in assessing the natural history of the condition.

We report a prospective study of 232 consecutive patients with hip fractures. All were over 64 years of age and living independently before admission to a geriatric orthopaedic ward. We assessed the value, at admission, of predicting factors for independent living at one year after injury.

The most important factors were: (1) preinjury function in activities of daily living (grade A or B on the Katz et al (1963) scale); (2) absence of other medical conditions which would impair rehabilitation; and (3) cognitive function better than 7 on the Pfeiffer (1975) mental questionnaire. The odds ratios (95% CI) for these three predictors were 3.5 (1.3 to 9.1), 2.9 (1.3 to 6.1) and 2.4 (1.9 to 4.9), respectively. When all predictors were positive at admission, 92% were living independently at one year; with one, two or three negative predictors, the percentages living independently were 76, 61 and 27, respectively.

The median values of the total number of days in hospital, irrespective of diagnosis, during the first year were 12, 24, 29 and 149 days for the four groups. The mortality at one year was predictable on admission only by the number of medical conditions: with no other diagnosis than the fracture the mortality was 0%; with one or two additional conditions the mortality was 14%; and with three or more additional diagnoses it was 24%.

These simple and robust predictors can be used to optimise resources for rehabilitation.

We compared a group of 46 somatised patients with a control group of 41 non-somatised patients who had undergone elective surgery to the lumbar spine in an attempt to identify pre-operative factors which could predict the outcome. In a prospective single-centre study, the Distress and Risk Assessment method consisting of a modified somatic perception questionnaire and modified Zung depression index was used pre-operatively to identify somatised patients. The type and number of consultations were correlated with functional indicators of outcome, such as the Oswestry disability index and a visual analogue score for pain in the leg after follow-up for six and 12 months.

Similar improvements in the Oswestry disability index were found in the somatised and non-somatised groups. Somatised patients who had a good outcome on the Oswestry disability index had an increased number of orthopaedic consultations (50 of 83 patients (60%) vs 29 of 73 patients (39.7%); p = 0.16) and waited less time for their surgery (5.5 months) (sd 5.26) vs 10.1 months (sd 6.29); p = 0.026). No other identifiable factors were found. A shorter wait for surgery appeared to predict a good outcome. Early review by a spinal surgeon and a reduced waiting time to surgery appear to be of particular benefit to somatised patients.

In 139 patients with burst fractures of the thoracic, thoracolumbar or lumbar spine, the least sagittal diameter of the spinal canal at the level of injury was measured by computerised tomography. By multiple logistic regression we investigated the joint correlation of the level of the burst fracture and the percentage of spinal canal stenosis with the probability of an associated neurological deficit. There was a very significant correlation between neurological deficit and the percentage of spinal canal stenosis; the higher the level of injury the greater was the probability. The severity of neurological deficit could not be predicted.

We report the ten-year results for three designs of stem in 240 total hip replacements, for which subsidence had been measured on plain radiographs at regular intervals. Accurate migration patterns could be determined by the method of Einzel-Bild-Roentgen-Analyse-femoral component analysis (EBRA-FCA) for 158 hips (66%).

Of these, 108 stems (68%) remained stable throughout, and five (3%) started to migrate after a median of 54 months. Initial migration of at least 1 mm was seen in 45 stems (29%) during the first two years, but these then became stable. We revised 17 stems for aseptic loosening, and 12 for other reasons. Revision for aseptic loosening could be predicted by EBRA-FCA with a sensitivity of 69%, a specificity of 80%, and an accuracy of 79% by the use of a threshold of subsidence of 1.5 mm during the first two years. Similar observations over a five-year period allowed the long-term outcome to be predicted with an accuracy of 91%.

We discuss the importance of four different patterns of subsidence and confirm that the early measurement of migration by a reasonably accurate method can help to predict long-term outcome. Such methods should be used to evaluate new and modified designs of prosthesis.

Aims. Machine-learning (ML) prediction models in orthopaedic trauma hold great promise in assisting clinicians in various tasks, such as personalized risk stratification. However, an overview of current applications and critical appraisal to peer-reviewed guidelines is lacking. The objectives of this study are to 1) provide an overview of current ML prediction models in orthopaedic trauma; 2) evaluate the completeness of reporting following the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement; and 3) assess the risk of bias following the Prediction model Risk Of Bias Assessment Tool (PROBAST) tool. Methods. A systematic search screening 3,252 studies identified 45 ML-based prediction models in orthopaedic trauma up to January 2023. The TRIPOD statement assessed transparent reporting and the PROBAST tool the risk of bias. Results. A total of 40 studies reported on training and internal validation; four studies performed both development and external validation, and one study performed only external validation. The most commonly reported outcomes were mortality (33%, 15/45) and length of hospital stay (9%, 4/45), and the majority of prediction models were developed in the hip fracture population (60%, 27/45). The overall median completeness for the TRIPOD statement was 62% (interquartile range 30 to 81%). The overall risk of bias in the PROBAST tool was low in 24% (11/45), high in 69% (31/45), and unclear in 7% (3/45) of the studies. High risk of bias was mainly due to analysis domain concerns including small datasets with low number of outcomes, complete-case analysis in case of missing data, and no reporting of performance measures. Conclusion. The results of this study showed that despite a myriad of potential clinically useful applications, a substantial part of ML studies in orthopaedic trauma lack transparent reporting, and are at high risk of bias. These problems must be resolved by following established guidelines to instil confidence in ML models among patients and clinicians. Otherwise, there will remain a sizeable gap between the development of ML prediction models and their clinical application in our day-to-day orthopaedic trauma practice. Cite this article: Bone Jt Open 2024;5(1):9–19

We report the theoretical basis of a method to measure axial migration of femoral components of total hip replacements (THR). The use of the top of the greater trochanter and a lateral point on the collar of the stem, allowing for variations of up to 10 degrees rotation of the femur in any direction between successive radiographs, gave a maximum error of 0.37 mm. At a more realistic 5 degrees rotational variation, the error was only 0.13 mm. These data were confirmed in an experimental study using digitisation of points and special software. We also showed that the centre of the femoral head, the stem tip, and the lesser trochanter provided less accurate landmarks. In a second study we digitised a series of radiographs of 51 Charnley and 57 Stanmore THRs; the mean migration rates were found to be identical. We then studied 46 successful stems with a minimum follow-up of eight years and 46 stems which had failed by aseptic loosening at different times. At two years, the successful stems had migrated by a mean of 1.45 +/- 0.68 mm, but the failed cases had a mean migration of 4.32 +/- 2.58 mm (p < 0.0001). Of the successful cases 76% had migrated less than 2 mm, while in the failed group 84% had migrated more than 2 mm. For any particular case migration of more than 2.6 mm at two years had only a 5% chance of continuing success and would therefore merit special follow-up. Only 24% of the eventually successful stems showed migration at the stem-cement interface, but this had happened in every failed stem. We conclude that it would be possible to evaluate a new cemented design of femoral stem over a two-year period by the use of our method and to compare its performance against the reported known standard of the Charnley and Stanmore designs.

Aims. No predictive model has been published to forecast operating time for total knee arthroplasty (TKA). The aims of this study were to design and validate a predictive model to estimate operating time for robotic-assisted TKA based on demographic data, and evaluate the added predictive power of CT scan-based predictors and their impact on the accuracy of the predictive model. Methods. A retrospective study was conducted on 1,061 TKAs performed from January 2016 to December 2019 with an image-based robotic-assisted system. Demographic data included age, sex, height, and weight. The femoral and tibial mechanical axis and the osteophyte volume were calculated from CT scans. These inputs were used to develop a predictive model aimed to predict operating time based on demographic data only, and demographic and 3D patient anatomy data. Results. The key factors for predicting operating time were the surgeon and patient weight, followed by 12 anatomical parameters derived from CT scans. The predictive model based only on demographic data showed that 90% of predictions were within 15 minutes of actual operating time, with 73% within ten minutes. The predictive model including demographic data and CT scans showed that 94% of predictions were within 15 minutes of actual operating time and 88% within ten minutes. Conclusion. The primary factors for predicting robotic-assisted TKA operating time were surgeon, patient weight, and osteophyte volume. This study demonstrates that incorporating 3D patient-specific data can improve operating time predictions models, which may lead to improved operating room planning and efficiency. Cite this article: Bone Jt Open 2022;3(5):383–389

Aims. The aim of this study was to develop and internally validate a prognostic nomogram to predict the probability of gaining a functional range of motion (ROM ≥ 120°) after open arthrolysis of the elbow in patients with post-traumatic stiffness of the elbow. Methods. We developed the Shanghai Prediction Model for Elbow Stiffness Surgical Outcome (SPESSO) based on a dataset of 551 patients who underwent open arthrolysis of the elbow in four institutions. Demographic and clinical characteristics were collected from medical records. The least absolute shrinkage and selection operator regression model was used to optimize the selection of relevant features. Multivariable logistic regression analysis was used to build the SPESSO. Its prediction performance was evaluated using the concordance index (C-index) and a calibration graph. Internal validation was conducted using bootstrapping validation. Results. BMI, the duration of stiffness, the preoperative ROM, the preoperative intensity of pain, and grade of post-traumatic osteoarthritis of the elbow were identified as predictors of outcome and incorporated to construct the nomogram. SPESSO displayed good discrimination with a C-index of 0.73 (95% confidence interval 0.64 to 0.81). A high C-index value of 0.70 could still be reached in the interval validation. The calibration graph showed good agreement between the nomogram prediction and the outcome. Conclusion. The newly developed SPESSO is a valid and convenient model which can be used to predict the outcome of open arthrolysis of the elbow. It could assist clinicians in counselling patients regarding the choice and expectations of treatment. Cite this article: Bone Joint J 2022;104-B(4):486–494

Aims. Heterotopic ossification (HO) is a common complication after elbow trauma and can cause severe upper limb disability. Although multiple prognostic factors have been reported to be associated with the development of post-traumatic HO, no model has yet been able to combine these predictors more succinctly to convey prognostic information and medical measures to patients. Therefore, this study aimed to identify prognostic factors leading to the formation of HO after surgery for elbow trauma, and to establish and validate a nomogram to predict the probability of HO formation in such particular injuries. Methods. This multicentre case-control study comprised 200 patients with post-traumatic elbow HO and 229 patients who had elbow trauma but without HO formation between July 2019 and December 2020. Features possibly associated with HO formation were obtained. The least absolute shrinkage and selection operator regression model was used to optimize feature selection. Multivariable logistic regression analysis was applied to build the new nomogram: the Shanghai post-Traumatic Elbow Heterotopic Ossification Prediction model (STEHOP). STEHOP was validated by concordance index (C-index) and calibration plot. Internal validation was conducted using bootstrapping validation. Results. Male sex, obesity, open wound, dislocations, late definitive surgical treatment, and lack of use of non-steroidal anti-inflammatory drugs were identified as adverse predictors and incorporated to construct the STEHOP model. It displayed good discrimination with a C-index of 0.80 (95% confidence interval 0.75 to 0.84). A high C-index value of 0.77 could still be reached in the internal validation. The calibration plot showed good agreement between nomogram prediction and observed outcomes. Conclusion. The newly developed STEHOP model is a valid and convenient instrument to predict HO formation after surgery for elbow trauma. It could assist clinicians in counselling patients regarding treatment expectations and therapeutic choices. Cite this article: Bone Joint J 2022;104-B(8):963–971

Aims. The aims of this study were to assess mapping models to predict the three-level version of EuroQoL five-dimension utility index (EQ-5D-3L) from the Oxford Knee Score (OKS) and validate these before and after total knee arthroplasty (TKA). Methods. A retrospective cohort of 5,857 patients was used to create the prediction models, and a second cohort of 721 patients from a different centre was used to validate the models, all of whom underwent TKA. Patient characteristics, BMI, OKS, and EQ-5D-3L were collected preoperatively and one year postoperatively. Generalized linear regression was used to formulate the prediction models. Results. There were significant correlations between the OKS and EQ-5D-3L preoperatively (r = 0.68; p < 0.001) and postoperatively (r = 0.77; p < 0.001) and for the change in the scores (r = 0.61; p < 0.001). Three different models (preoperative, postoperative, and change) were created. There were no significant differences between the actual and predicted mean EQ-5D-3L utilities at any timepoint or for change in the scores (p > 0.090) in the validation cohort. There was a significant correlation between the actual and predicted EQ-5D-3L utilities preoperatively (r = 0.63; p < 0.001) and postoperatively (r = 0.77; p < 0.001) and for the change in the scores (r = 0.56; p < 0.001). Bland-Altman plots demonstrated that a lower utility was overestimated, and higher utility was underestimated. The individual predicted EQ-5D-3L that was within ± 0.05 and ± 0.010 (minimal clinically important difference (MCID)) of the actual EQ-5D-3L varied between 13% to 35% and 26% to 64%, respectively, according to timepoint assessed and change in the scores, but was not significantly different between the modelling and validation cohorts (p ≥ 0.148). Conclusion. The OKS can be used to estimate EQ-5D-3L. Predicted individual patient utility error beyond the MCID varied from one-third to two-thirds depending on timepoint assessed, but the mean for a cohort did not differ and could be employed for this purpose. Cite this article: Bone Jt Open 2022;3(7):573–581

Literature surrounding artificial intelligence (AI)-related applications for hip and knee arthroplasty has proliferated. However, meaningful advances that fundamentally transform the practice and delivery of joint arthroplasty are yet to be realized, despite the broad range of applications as we continue to search for meaningful and appropriate use of AI. AI literature in hip and knee arthroplasty between 2018 and 2021 regarding image-based analyses, value-based care, remote patient monitoring, and augmented reality was reviewed. Concerns surrounding meaningful use and appropriate methodological approaches of AI in joint arthroplasty research are summarized. Of the 233 AI-related orthopaedics articles published, 178 (76%) constituted original research, while the rest consisted of editorials or reviews. A total of 52% of original AI-related research concerns hip and knee arthroplasty (n = 92), and a narrative review is described. Three studies were externally validated. Pitfalls surrounding present-day research include conflating vernacular (“AI/machine learning”), repackaging limited registry data, prematurely releasing internally validated prediction models, appraising model architecture instead of inputted data, withholding code, and evaluating studies using antiquated regression-based guidelines. While AI has been applied to a variety of hip and knee arthroplasty applications with limited clinical impact, the future remains promising if the question is meaningful, the methodology is rigorous and transparent, the data are rich, and the model is externally validated. Simple checkpoints for meaningful AI adoption include ensuring applications focus on: administrative support over clinical evaluation and management; necessity of the advanced model; and the novelty of the question being answered. Cite this article: Bone Joint J 2022;104-B(12):1292–1303

The August 2023 Knee Roundup. 360. looks at: Curettage and cementation of giant cell tumour of bone: is arthritis a given?; Anterior knee pain following total knee arthroplasty: does the patellar cement-bone interface affect postoperative anterior knee pain?; Nickel allergy and total knee arthroplasty; The use of artificial intelligence for the prediction of periprosthetic joint infection following aseptic revision total knee arthroplasty; Ambulatory unicompartmental knee arthroplasty: development of a patient selection tool using machine learning; Femoral asymmetry: a missing piece in knee alignment; Needle arthroscopy – a benefit to patients in the outpatient setting; Can lateral unicompartmental knees be done in a day-case setting?

Aims. A substantial fraction of patients undergoing knee arthroplasty (KA) or hip arthroplasty (HA) do not achieve an improvement as high as the minimal clinically important difference (MCID), i.e. do not achieve a meaningful improvement. Using three patient-reported outcome measures (PROMs), our aim was: 1) to assess machine learning (ML), the simple pre-surgery PROM score, and logistic-regression (LR)-derived performance in their prediction of whether patients undergoing HA or KA achieve an improvement as high or higher than a calculated MCID; and 2) to test whether ML is able to outperform LR or pre-surgery PROM scores in predictive performance. Methods. MCIDs were derived using the change difference method in a sample of 1,843 HA and 1,546 KA patients. An artificial neural network, a gradient boosting machine, least absolute shrinkage and selection operator (LASSO) regression, ridge regression, elastic net, random forest, LR, and pre-surgery PROM scores were applied to predict MCID for the following PROMs: EuroQol five-dimension, five-level questionnaire (EQ-5D-5L), EQ visual analogue scale (EQ-VAS), Hip disability and Osteoarthritis Outcome Score-Physical Function Short-form (HOOS-PS), and Knee injury and Osteoarthritis Outcome Score-Physical Function Short-form (KOOS-PS). Results. Predictive performance of the best models per outcome ranged from 0.71 for HOOS-PS to 0.84 for EQ-VAS (HA sample). ML statistically significantly outperformed LR and pre-surgery PROM scores in two out of six cases. Conclusion. MCIDs can be predicted with reasonable performance. ML was able to outperform traditional methods, although only in a minority of cases. Cite this article: Bone Joint Res 2023;12(9):512–521

Aims. To determine the major risk factors for unplanned reoperations (UROs) following corrective surgery for adult spinal deformity (ASD) and their interactions, using machine learning-based prediction algorithms and game theory. Methods. Patients who underwent surgery for ASD, with a minimum of two-year follow-up, were retrospectively reviewed. In total, 210 patients were included and randomly allocated into training (70% of the sample size) and test (the remaining 30%) sets to develop the machine learning algorithm. Risk factors were included in the analysis, along with clinical characteristics and parameters acquired through diagnostic radiology. Results. Overall, 152 patients without and 58 with a history of surgical revision following surgery for ASD were observed; the mean age was 68.9 years (SD 8.7) and 66.9 years (SD 6.6), respectively. On implementing a random forest model, the classification of URO events resulted in a balanced accuracy of 86.8%. Among machine learning-extracted risk factors, URO, proximal junction failure (PJF), and postoperative distance from the posterosuperior corner of C7 and the vertical axis from the centroid of C2 (SVA) were significant upon Kaplan-Meier survival analysis. Conclusion. The major risk factors for URO following surgery for ASD, i.e. postoperative SVA and PJF, and their interactions were identified using a machine learning algorithm and game theory. Clinical benefits will depend on patient risk profiles. Cite this article: Bone Joint Res 2023;12(4):245–255