Aims. Computer-assisted 3D preoperative planning software has the potential to improve postoperative stability in total hip arthroplasty (THA). Commonly, preoperative protocols simulate two functional positions (standing and relaxed sitting) but do not consider other common positions that may increase postoperative impingement and possible dislocation. This study investigates the feasibility of simulating commonly encountered positions, and positions with an increased risk of impingement, to lower postoperative impingement risk in a CT-based 3D model. Methods. A robotic arm-assisted arthroplasty planning platform was used to investigate 11 patient positions. Data from 43 primary THAs were used for simulation. Sacral slope was retrieved from patient preoperative imaging, while angles of hip flexion/extension, hip external/internal rotation, and hip abduction/adduction for tested positions were derived from literature or estimated with a biomechanical model. The hip was placed in the described positions, and if impingement was detected by the software, inspection of the impingement type was performed. Results. In flexion, an overall impingement rate of 2.3% was detected for flexed-seated, squatting, forward-bending, and criss-cross-sitting positions, and 4.7% for the ankle-over-knee position. In extension, most hips (60.5%) were found to impinge at or prior to 50° of external rotation (pivoting). Many of these impingement events were due to a prominent ischium. The mean maximum external rotation prior to impingement was 45.9° (15° to 80°) and 57.9° (20° to 90°) prior to prosthetic impingement. No impingement was found in standing, sitting, crossing ankles, seiza, and downward dog. Conclusion. This study demonstrated that positions of daily living tested in a CT-based 3D model show high rates of impingement. Simulating additional positions through 3D modelling is a low-cost method of potentially improving outcomes without compromising patient safety. By incorporating CT-based 3D modelling of positions of daily living into routine preoperative protocols for THA, there is the potential to lower the risk of postoperative impingement events. Cite this article: Bone Jt Open 2023;4(6):416–423

Aims. Traditionally, total hip arthroplasty (THA) templating has been performed on anteroposterior (AP) pelvis radiographs. Recently, additional AP hip radiographs have been recommended for accurate measurement of the femoral offset (FO). To verify this claim, this study aimed to establish quantitative data of the measurement error of the FO in relation to leg position and X-ray source position using a newly developed geometric model and clinical data. Methods. We analyzed the FOs measured on AP hip and pelvis radiographs in a prospective consecutive series of 55 patients undergoing unilateral primary THA for hip osteoarthritis. To determine sample size, a power analysis was performed. Patients’ position and X-ray beam setting followed a standardized protocol to achieve reproducible projections. All images were calibrated with the KingMark calibration system. In addition, a geometric model was created to evaluate both the effects of leg position (rotation and abduction/adduction) and the effects of X-ray source position on FO measurement. Results. The mean FOs measured on AP hip and pelvis radiographs were 38.0 mm (SD 6.4) and 36.6 mm (SD 6.3) (p < 0.001), respectively. Radiological view had a smaller effect on FO measurement than inaccurate leg positioning. The model showed a non-linear relationship between projected FO and femoral neck orientation; at 30° external neck rotation (with reference to the detector plane), a true FO of 40 mm was underestimated by up to 20% (7.8 mm). With a neutral to mild external neck rotation (≤ 15°), the underestimation was less than 7% (2.7 mm). The effect of abduction and adduction was negligible. Conclusion. For routine THA templating, an AP pelvis radiograph remains the gold standard. Only patients with femoral neck malrotation > 15° on the AP pelvis view, e.g. due to external rotation contracture, should receive further imaging. Options include an additional AP hip view with elevation of the entire affected hip to align the femoral neck more parallel to the detector, or a CT scan in more severe cases. Cite this article: Bone Jt Open 2022;3(10):795–803

Aims. Aseptic loosening is a leading cause of uncemented arthroplasty failure, often accompanied by fibrotic tissue at the bone-implant interface. A biological target, neutrophil extracellular traps (NETs), was investigated as a crucial connection between the innate immune system’s response to injury, fibrotic tissue development, and proper bone healing. Prevalence of NETs in peri-implant fibrotic tissue from aseptic loosening patients was assessed. A murine model of osseointegration failure was used to test the hypothesis that inhibition (through Pad4-/- mice that display defects in peptidyl arginine deiminase 4 (PAD4), an essential protein required for NETs) or resolution (via DNase 1 treatment, an enzyme that degrades the cytotoxic DNA matrix) of NETs can prevent osseointegration failure and formation of peri-implant fibrotic tissue. Methods. Patient peri-implant fibrotic tissue was analyzed for NETs biomarkers. To enhance osseointegration in loose implant conditions, an innate immune system pathway (NETs) was either inhibited (Pad4-/- mice) or resolved with a pharmacological agent (DNase 1) in a murine model of osseointegration failure. Results. NETs biomarkers were identified in peri-implant fibrotic tissue collected from aseptic loosening patients and at the bone-implant interface in a murine model of osseointegration failure. Inhibition (Pad4-/-) or resolution (DNase 1) of NETs improved osseointegration and reduced fibrotic tissue despite loose implant conditions in mice. Conclusion. This study identifies a biological target (NETs) for potential noninvasive treatments of aseptic loosening by discovering a novel connection between the innate immune system and post-injury bone remodelling caused by implant loosening. By inhibiting or resolving NETs in an osseointegration failure murine model, fibrotic tissue encapsulation around an implant is reduced and osseointegration is enhanced, despite loose implant conditions. Cite this article: Bone Joint J 2021;103-B(7 Supple B):135–144

Aims. The aims of this study were to develop an in vivo model of periprosthetic joint infection (PJI) in cemented hip hemiarthroplasty, and to monitor infection and biofilm formation in real-time. Methods. Sprague-Dawley rats underwent cemented hip hemiarthroplasty via the posterior approach with pre- and postoperative gait assessments. Infection with Staphylococcus aureus Xen36 was monitored with in vivo photoluminescent imaging in real-time. Pre- and postoperative gait analyses were performed and compared. Postmortem micro (m) CT was used to assess implant integration; field emission scanning electron microscopy (FE-SEM) was used to assess biofilm formation on prosthetic surfaces. Results. All animals tolerated surgery well, with preservation of gait mechanics and weightbearing in control individuals. Postoperative in vivo imaging demonstrated predictable evolution of infection with logarithmic signal decay coinciding with abscess formation. Postmortem mCT qualitative volumetric analysis showed high contact area and both cement-bone and cement-implant interdigitation. FE-SEM revealed biofilm formation on the prosthetic head. Conclusion. This study demonstrates the utility of a new, high-fidelity model of in vivo PJI using cemented hip hemiarthroplasty in rats. Inoculation with bioluminescent bacteria allows for non-invasive, real-time monitoring of infection. Cite this article: Bone Joint J 2021;103-B(7 Supple B):9–16

Aims. The practice of overlapping surgery has been increasing in the delivery of orthopaedic surgery, aiming to provide efficient, high-quality care. However, there are concerns about the safety of this practice. The purpose of this study was to examine the safety and efficacy of a model of partially overlapping surgery that we termed ‘swing room’ in the practice of primary total hip (THA) and knee arthroplasty (TKA). Methods. A retrospective review of prospectively collected data was carried out on patients who underwent primary THA and TKA between 2006 and 2017 in two academic centres. Cases were stratified as partially overlapping (swing room), in which the surgeon is in one operating room (OR) while the next patient is being prepared in another, or nonoverlapping surgery. The demographic details of the patients which were collected included operating time, length of stay (LOS), postoperative complications within six weeks of the procedure, unplanned hospital readmissions, and unplanned reoperations. Fisher's exact, Wilcoxon rank-sum tests, chi-squared tests, and logistic regression analysis were used for statistical analysis. Results. A total of 12,225 cases performed at our institution were included in the study, of which 10,596 (86.6%) were partially overlapping (swing room) and 1,629 (13.3%) were nonoverlapping. There was no significant difference in the mean age, sex, body mass index (BMI), side, and LOS between the two groups. The mean operating time was significantly shorter in the swing room group (58.2 minutes) compared with the nonoverlapping group (62.8 minutes; p < 0.001). There was no significant difference in the rates of complications, readmission and reoperations (p = 0.801 and p = 0.300, respectively) after adjusting for baseline American Society of Anesthesiologists scores. Conclusion. The new ‘swing room’ model yields similar short-term outcomes without an increase in complication rates compared with routine single OR surgery in patients undergoing primary THA or TKA. Cite this article: Bone Joint J 2020;102-B(7 Supple B):112–115

Aims. Current treatments of prosthetic joint infection (PJI) are minimally effective against Staphylococcus aureus biofilm. A murine PJI model of debridement, antibiotics, and implant retention (DAIR) was used to test the hypothesis that PlySs2, a bacteriophage-derived lysin, can target S. aureus biofilm and address the unique challenges presented in this periprosthetic environment. Methods. The ability of PlySs2 and vancomycin to kill biofilm and colony-forming units (CFUs) on orthopaedic implants were compared using in vitro models. An in vivo murine PJI model of DAIR was used to assess the efficacy of a combination of PlySs2 and vancomycin on periprosthetic bacterial load. Results. PlySs2 treatment reduced 99% more CFUs and 75% more biofilm compared with vancomycin in vitro. A combination of PlySs2 and vancomycin in vivo reduced the number of CFUs on the surface of implants by 92% and in the periprosthetic tissue by 88%. Conclusion. PlySs2 lysin was able to reduce biofilm, target planktonic bacteria, and work synergistically with vancomycin in our in vitro models. A combination of PlySs2 and vancomycin also reduced bacterial load in periprosthetic tissue and on the surface of implants in a murine model of DAIR treatment for established PJI. Cite this article: Bone Joint J 2020;102-B(7 Supple B):3–10

Aims. Osteolysis, secondary to local and systemic physiological effects, is a major challenge in total hip arthroplasty (THA). While osteolytic defects are commonly observed in long-term follow-up, how such lesions alter the distribution of stress is unclear. The aim of this study was to quantitatively describe the biomechanical implication of such lesions by performing subject-specific finite-element (FE) analysis on patients with osteolysis after THA. Patients and Methods. A total of 22 hemipelvis FE models were constructed in order to assess the transfer of load in 11 patients with osteolysis around the acetabular component of a THA during slow walking and a fall onto the side. There were nine men and two women. Their mean age was 69 years (55 to 81) at final follow-up. Changes in peak stress values and loads to fracture in the presence of the osteolytic defects were measured. Results. The von Mises stresses were increased in models of those with and those without defects for both loading scenarios. Although some regions showed increases in stress values of up to 100%, there was only a moderate 11.2% increase in von Mises stress in the series as a whole. The site of fracture changed in some models with lowering of the load to fracture by 500 N. The most common site of fracture was the pubic ramus. This was more frequent in models with larger defects. Conclusion. We conclude that cancellous defects cause increases in stress within cortical structures. However, these are likely to lead to a modest decrease in the load to fracture if the defect is large (> 20cm. 3. ) or if the patient is small with thin cortical structures and low bone mineral density. Cite this article: Bone Joint J 2018;100-B:1455–62

Aims. Fixation of osteoporotic proximal humerus fractures remains challenging even with state-of-the-art locking plates. Despite the demonstrated biomechanical benefit of screw tip augmentation with bone cement, the clinical findings have remained unclear, potentially as the optimal augmentation combinations are unknown. The aim of this study was to systematically evaluate the biomechanical benefits of the augmentation options in a humeral locking plate using finite element analysis (FEA). Methods. A total of 64 cement augmentation configurations were analyzed using six screws of a locking plate to virtually fix unstable three-part fractures in 24 low-density proximal humerus models under three physiological loading cases (4,608 simulations). The biomechanical benefit of augmentation was evaluated through an established FEA methodology using the average peri-screw bone strain as a validated predictor of cyclic cut-out failure. Results. The biomechanical benefit was already significant with a single cemented screw and increased with the number of augmented screws, but the configuration was highly influential. The best two-screw (mean 23%, SD 3% reduction) and the worst four-screw (mean 22%, SD 5%) combinations performed similarly. The largest benefits were achieved with augmenting screws purchasing into the calcar and having posteriorly located tips. Local bone mineral density was not directly related to the improvement. Conclusion. The number and configuration of cemented screws strongly determined how augmentation can alleviate the predicted risk of cut-out failure. Screws purchasing in the calcar and posterior humeral head regions may be prioritized. Although requiring clinical corroborations, these findings may explain the controversial results of previous clinical studies not controlling the choices of screw augmentation

Objectives. The primary purpose of this meta-analysis was to determine whether statin usage could reduce the risk of glucocorticoid-related osteonecrosis in animal models. Methods. A systematic literature search up to May 2015 was carried out using the PubMed, Ovid, EBM reviews, ISI Web of Science, EBSCO, CBM, CNKI databases with the term and boolean operators: statins and osteonecrosis in all fields. Risk ratio (RR), as the risk estimate of specific outcome, was calculated along with 95% confidence intervals (CI). The methodological quality of individual studies was assessed using a quantitative tool based on the updated Stroke Therapy Academic Industry Roundtable (STAIR) recommendations. Results. A total of 11 eligible studies were included according to predetermined criteria. The pooled data demonstrated that animals with statin usage, either alone or combined with other treatments, were at a decreased risk of developing glucocorticoid-related osteonecrosis (RR = 2.06, 95% confidence interval (CI) 1.71 to 2.50). Moreover, subgroup analysis revealed that compared with statins alone, statins combined with other treatments significantly decreased the risk of osteonecrosis (RR = 1.23, 95% CI 1.02 to 1.47). However, we could find no significant risk difference for different gender, or for different time points. Conclusions. The present study suggests that statins combined with other treatments are efficient in preventing the development of glucocorticoid-related osteonecrosis in animals. These results might shed light on clinical practice when glucocorticoids are prescribed, and could be further investigated in high-quality clinical trials. Cite this article: Z. Yang, H. Liu, D. Li, X. Xie, T. Qin, J. Ma, P. Kang. The efficacy of statins in preventing glucocorticoid-related osteonecrosis in animal models: A meta-analysis. Bone Joint Res 2016;5:393–402. DOI: 10.1302/2046-3758.59.2000500

Aims

When performing revision total hip arthroplasty using diaphyseal-engaging titanium tapered stems (TTS), the recommended 3 to 4 cm of stem-cortical diaphyseal contact may not be available. In challenging cases such as these with only 2 cm of contact, can sufficient axial stability be achieved and what is the benefit of a prophylactic cable? This study sought to determine, first, whether a prophylactic cable allows for sufficient axial stability when the contact length is 2 cm, and second, if differing TTS taper angles (2° vs 3.5°) impact these results.

Aims

Periprosthetic hip fractures (PPFs) after total hip arthroplasty are difficult to treat. Therefore, it is important to identify modifiable risk factors such as stem selection to reduce the occurrence of PPFs. This study aimed to clarify differences in fracture torque, surface strain, and fracture type analysis between three different types of cemented stems.

Pre-operative planning for total hip replacement (THR) is challenging in hips with severe acetabular deformities, including those with a hypoplastic acetabulum or severe defects and in the presence of arthrodesis or ankylosis. We evaluated whether a Rapid Prototype (RP) model, which is a life-sized reproduction based on three-dimensional CT scans, can determine the feasibility of THR and provide information about the size and position of the acetabular component in severe acetabular deformities. THR was planned using an RP model in 21 complex hips in five men (five hips) and 16 women (16 hips) with a mean age of 47.7 years (24 to 70) at operation. An acetabular component was implanted successfully and THR completed in all hips. The acetabular component used was within 2 mm of the predicted size in 17 hips (80.9%). All of the acetabular components and femoral stems had radiological evidence of bone ingrowth and stability at the final follow-up, without any detectable wear or peri-prosthetic osteolysis. The RP model allowed a simulated procedure pre-operatively and was helpful in determining the feasibility of THR pre-operatively, and to decide on implant type, size and position in complex THRs. Cite this article: Bone Joint J 2013;95-B:1458–63

Aims

Cementless acetabular components rely on press-fit fixation for initial stability. In certain cases, initial stability is more difficult to obtain (such as during revision). No current study evaluates how a surgeon’s impaction technique (mallet mass, mallet velocity, and number of strikes) may affect component fixation. This study seeks to answer the following research questions: 1) how does impaction technique affect a) bone strain generation and deterioration (and hence implant stability) and b) seating in different density bones?; and 2) can an impaction technique be recommended to minimize risk of implant loosening while ensuring seating of the acetabular component?

Aims. Iliopsoas impingement occurs in 4% to 30% of patients after undergoing total hip arthroplasty (THA). Despite a relatively high incidence, there are few attempts at modelling impingement between the iliopsoas and acetabular component, and no attempts at modelling this in a representative cohort of subjects. The purpose of this study was to develop a novel computational model for quantifying the impingement between the iliopsoas and acetabular component and validate its utility in a case-controlled investigation. Methods. This was a retrospective cohort study of patients who underwent THA surgery that included 23 symptomatic patients diagnosed with iliopsoas tendonitis, and 23 patients not diagnosed with iliopsoas tendonitis. All patients received postoperative CT imaging, postoperative standing radiography, and had minimum six months’ follow-up. 3D models of each patient’s prosthetic and bony anatomy were generated, landmarked, and simulated in a novel iliopsoas impingement detection model in supine and standing pelvic positions. Logistic regression models were implemented to determine if the probability of pain could be significantly predicted. Receiver operating characteristic curves were generated to determine the model’s sensitivity, specificity, and area under the curve (AUC). Results. Highly significant differences between the symptomatic and asymptomatic cohorts were observed for iliopsoas impingement. Logistic regression models determined that the impingement values significantly predicted the probability of groin pain. The simulation had a sensitivity of 74%, specificity of 100%, and an AUC of 0.86. Conclusion. We developed a computational model that can quantify iliopsoas impingement and verified its accuracy in a case-controlled investigation. This tool has the potential to be used preoperatively, to guide decisions about optimal cup placement, and postoperatively, to assist in the diagnosis of iliopsoas tendonitis. Cite this article: Bone Jt Open 2023;4(1):3–12

Aims. Total hip arthroplasty (THA) is a common procedure to address pain and enhance function in hip disorders such as osteoarthritis. Despite its success, postoperative patient recovery exhibits considerable heterogeneity. This study aimed to investigate whether patients follow distinct pain trajectories following THA and identify the patient characteristics linked to suboptimal trajectories. Methods. This retrospective cohort study analyzed THA patients at a large academic centre (NYU Langone Orthopedic Hospital, New York, USA) from January 2018 to January 2023, who completed the Patient-Reported Outcomes Measurement Information System (PROMIS) pain intensity questionnaires, collected preoperatively at one-, three-, six-, 12-, and 24-month follow-up times. Growth mixture modelling (GMM) was used to model the trajectories. Optimal model fit was determined by Bayesian information criterion (BIC), Vuong-Lo-Mendell-Rubin likelihood ratio test (VLMR-LRT), posterior probabilities, and entropy values. Association between trajectory groups and patient characteristics were measured by multinomial logistic regression using the three-step approach. Results. Among the 1,249 patients, a piecewise GMM model revealed three distinct pain trajectory groups: 56 patients (4.5%) in group 1; 1,144 patients (91.6%) in group 2; and 49 patients (3.9%) in group 3. Patients in group 2 experienced swift recovery post-THA and minimal preoperative pain. In contrast, groups 1 and 3 initiated with pronounced preoperative pain; however, only group 3 exhibited persistent long-term pain. Multinomial regression indicated African Americans were exceedingly likely to follow trajectory groups 1 (odds ratio (OR) 2.73) and 3 (OR 3.18). Additionally, odds of membership to group 3 increased by 12% for each BMI unit rise, by 19% for each added postoperative day, and by over four if discharged to rehabilitation services (OR 4.07). Conclusion. This study identified three distinct pain trajectories following THA, highlighting the role of individual patient factors in postoperative recovery. This emphasizes the importance of preoperatively addressing modifiable risk factors associated with suboptimal pain trajectories, particularly in at-risk patients. Cite this article: Bone Jt Open 2024;5(3):174–183

Aims. This study aimed to develop and validate a fully automated system that quantifies proximal femoral bone mineral density (BMD) from CT images. Methods. The study analyzed 978 pairs of hip CT and dual-energy X-ray absorptiometry (DXA) measurements of the proximal femur (DXA-BMD) collected from three institutions. From the CT images, the femur and a calibration phantom were automatically segmented using previously trained deep-learning models. The Hounsfield units of each voxel were converted into density (mg/cm. 3. ). Then, a deep-learning model trained by manual landmark selection of 315 cases was developed to select the landmarks at the proximal femur to rotate the CT volume to the neutral position. Finally, the CT volume of the femur was projected onto the coronal plane, and the areal BMD of the proximal femur (CT-aBMD) was quantified. CT-aBMD correlated to DXA-BMD, and a receiver operating characteristic (ROC) analysis quantified the accuracy in diagnosing osteoporosis. Results. CT-aBMD was successfully measured in 976/978 hips (99.8%). A significant correlation was found between CT-aBMD and DXA-BMD (r = 0.941; p < 0.001). In the ROC analysis, the area under the curve to diagnose osteoporosis was 0.976. The diagnostic sensitivity and specificity were 88.9% and 96%, respectively, with the cutoff set at 0.625 g/cm. 2. . Conclusion. Accurate DXA-BMD measurements and diagnosis of osteoporosis were performed from CT images using the system developed herein. As the models are open-source, clinicians can use the proposed system to screen osteoporosis and determine the surgical strategy for hip surgery. Cite this article: Bone Joint Res 2023;12(9):590–597

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. 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. Precise implant positioning, tailored to individual spinopelvic biomechanics and phenotype, is paramount for stability in total hip arthroplasty (THA). Despite a few studies on instability prediction, there is a notable gap in research utilizing artificial intelligence (AI). The objective of our pilot study was to evaluate the feasibility of developing an AI algorithm tailored to individual spinopelvic mechanics and patient phenotype for predicting impingement. Methods. This international, multicentre prospective cohort study across two centres encompassed 157 adults undergoing primary robotic arm-assisted THA. Impingement during specific flexion and extension stances was identified using the virtual range of motion (ROM) tool of the robotic software. The primary AI model, the Light Gradient-Boosting Machine (LGBM), used tabular data to predict impingement presence, direction (flexion or extension), and type. A secondary model integrating tabular data with plain anteroposterior pelvis radiographs was evaluated to assess for any potential enhancement in prediction accuracy. Results. We identified nine predictors from an analysis of baseline spinopelvic characteristics and surgical planning parameters. Using fivefold cross-validation, the LGBM achieved 70.2% impingement prediction accuracy. With impingement data, the LGBM estimated direction with 85% accuracy, while the support vector machine (SVM) determined impingement type with 72.9% accuracy. After integrating imaging data with a multilayer perceptron (tabular) and a convolutional neural network (radiograph), the LGBM’s prediction was 68.1%. Both combined and LGBM-only had similar impingement direction prediction rates (around 84.5%). Conclusion. This study is a pioneering effort in leveraging AI for impingement prediction in THA, utilizing a comprehensive, real-world clinical dataset. Our machine-learning algorithm demonstrated promising accuracy in predicting impingement, its type, and direction. While the addition of imaging data to our deep-learning algorithm did not boost accuracy, the potential for refined annotations, such as landmark markings, offers avenues for future enhancement. Prior to clinical integration, external validation and larger-scale testing of this algorithm are essential. Cite this article: Bone Jt Open 2024;5(8):671–680

Aims. To evaluate how abnormal proximal femoral anatomy affects different femoral version measurements in young patients with hip pain. Methods. First, femoral version was measured in 50 hips of symptomatic consecutively selected patients with hip pain (mean age 20 years (SD 6), 60% (n = 25) females) on preoperative CT scans using different measurement methods: Lee et al, Reikerås et al, Tomczak et al, and Murphy et al. Neck-shaft angle (NSA) and α angle were measured on coronal and radial CT images. Second, CT scans from three patients with femoral retroversion, normal femoral version, and anteversion were used to create 3D femur models, which were manipulated to generate models with different NSAs and different cam lesions, resulting in eight models per patient. Femoral version measurements were repeated on manipulated femora. Results. Comparing the different measurement methods for femoral version resulted in a maximum mean difference of 18° (95% CI 16 to 20) between the most proximal (Lee et al) and most distal (Murphy et al) methods. Higher differences in proximal and distal femoral version measurement techniques were seen in femora with greater femoral version (r > 0.46; p < 0.001) and greater NSA (r > 0.37; p = 0.008) between all measurement methods. In the parametric 3D manipulation analysis, differences in femoral version increased 11° and 9° in patients with high and normal femoral version, respectively, with increasing NSA (110° to 150°). Conclusion. Measurement of femoral version angles differ depending on the method used to almost 20°, which is in the range of the aimed surgical correction in derotational femoral osteotomy and thus can be considered clinically relevant. Differences between proximal and distal measurement methods further increase by increasing femoral version and NSA. Measurement methods that take the entire proximal femur into account by using distal landmarks may produce more sensitive measurements of these differences. Cite this article: Bone Jt Open 2022;3(10):759–766