Aims. The aim of this study was to identify factors associated with five-year cancer-related mortality in patients with limb and trunk soft-tissue sarcoma (STS) and develop and validate machine learning algorithms in order to predict five-year cancer-related mortality in these patients. Methods. Demographic, clinicopathological, and treatment variables of limb and trunk STS patients in the Surveillance, Epidemiology, and End Results Program (SEER) database from 2004 to 2017 were analyzed. Multivariable logistic regression was used to determine factors significantly associated with five-year cancer-related mortality. Various machine learning models were developed and compared using area under the curve (AUC), calibration, and decision curve analysis. The model that performed best on the SEER testing data was further assessed to determine the variables most important in its predictive capacity. This model was externally validated using our institutional dataset. Results. A total of 13,646 patients with STS from the SEER database were included, of whom 35.9% experienced five-year cancer-related mortality. The random forest model performed the best overall and identified tumour size as the most important variable when predicting mortality in patients with STS, followed by M stage, histological subtype, age, and surgical excision. Each variable was significant in logistic regression. External validation yielded an AUC of 0.752. Conclusion. This study identified clinically important variables associated with five-year cancer-related mortality in patients with limb and trunk STS, and developed a predictive model that demonstrated good accuracy and predictability. Orthopaedic oncologists may use these findings to further risk-stratify their patients and recommend an optimal course of treatment. Cite this article: Bone Joint J 2023;105-B(6):702–710

Aims. The aim of this study was to evaluate the ability of a machine-learning algorithm to diagnose prosthetic loosening from preoperative radiographs and to investigate the inputs that might improve its performance. Methods. A group of 697 patients underwent a first-time revision of a total hip (THA) or total knee arthroplasty (TKA) at our institution between 2012 and 2018. Preoperative anteroposterior (AP) and lateral radiographs, and historical and comorbidity information were collected from their electronic records. Each patient was defined as having loose or fixed components based on the operation notes. We trained a series of convolutional neural network (CNN) models to predict a diagnosis of loosening at the time of surgery from the preoperative radiographs. We then added historical data about the patients to the best performing model to create a final model and tested it on an independent dataset. Results. The convolutional neural network we built performed well when detecting loosening from radiographs alone. The first model built de novo with only the radiological image as input had an accuracy of 70%. The final model, which was built by fine-tuning a publicly available model named DenseNet, combining the AP and lateral radiographs, and incorporating information from the patient’s history, had an accuracy, sensitivity, and specificity of 88.3%, 70.2%, and 95.6% on the independent test dataset. It performed better for cases of revision THA with an accuracy of 90.1%, than for cases of revision TKA with an accuracy of 85.8%. Conclusion. This study showed that machine learning can detect prosthetic loosening from radiographs. Its accuracy is enhanced when using highly trained public algorithms, and when adding clinical data to the algorithm. While this algorithm may not be sufficient in its present state of development as a standalone metric of loosening, it is currently a useful augment for clinical decision making. Cite this article: Bone Joint J 2020;102-B(6 Supple A):101–106

Aims. Accurate identification of the ankle joint centre is critical for estimating tibial coronal alignment in total knee arthroplasty (TKA). The purpose of the current study was to leverage artificial intelligence (AI) to determine the accuracy and effect of using different radiological anatomical landmarks to quantify mechanical alignment in relation to a traditionally defined radiological ankle centre. Methods. Patients with full-limb radiographs from the Osteoarthritis Initiative were included. A sub-cohort of 250 radiographs were annotated for landmarks relevant to knee alignment and used to train a deep learning (U-Net) workflow for angle calculation on the entire database. The radiological ankle centre was defined as the midpoint of the superior talus edge/tibial plafond. Knee alignment (hip-knee-ankle angle) was compared against 1) midpoint of the most prominent malleoli points, 2) midpoint of the soft-tissue overlying malleoli, and 3) midpoint of the soft-tissue sulcus above the malleoli. Results. A total of 932 bilateral full-limb radiographs (1,864 knees) were measured at a rate of 20.63 seconds/image. The knee alignment using the radiological ankle centre was accurate against ground truth radiologist measurements (inter-class correlation coefficient (ICC) = 0.99 (0.98 to 0.99)). Compared to the radiological ankle centre, the mean midpoint of the malleoli was 2.3 mm (SD 1.3) lateral and 5.2 mm (SD 2.4) distal, shifting alignment by 0.34. o. (SD 2.4. o. ) valgus, whereas the midpoint of the soft-tissue sulcus was 4.69 mm (SD 3.55) lateral and 32.4 mm (SD 12.4) proximal, shifting alignment by 0.65. o. (SD 0.55. o. ) valgus. On the intermalleolar line, measuring a point at 46% (SD 2%) of the intermalleolar width from the medial malleoli (2.38 mm medial adjustment from midpoint) resulted in knee alignment identical to using the radiological ankle centre. Conclusion. The current study leveraged AI to create a consistent and objective model that can estimate patient-specific adjustments necessary for optimal landmark usage in extramedullary and computer-guided navigation for tibial coronal alignment to match radiological planning. Cite this article: Bone Jt Open 2022;3(10):767–776

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. To propose a new method for evaluating paediatric radial neck fractures and improve the accuracy of fracture angulation measurement, particularly in younger children, and thereby facilitate planning treatment in this population. Methods. Clinical data of 117 children with radial neck fractures in our hospital from August 2014 to March 2023 were collected. A total of 50 children (26 males, 24 females, mean age 7.6 years (2 to 13)) met the inclusion criteria and were analyzed. Cases were excluded for the following reasons: Judet grade I and Judet grade IVb (> 85° angulation) classification; poor radiograph image quality; incomplete clinical information; sagittal plane angulation; severe displacement of the ulna fracture; and Monteggia fractures. For each patient, standard elbow anteroposterior (AP) view radiographs and corresponding CT images were acquired. On radiographs, Angle P (complementary to the angle between the long axis of the radial head and the line perpendicular to the physis), Angle S (complementary to the angle between the long axis of the radial head and the midline through the proximal radial shaft), and Angle U (between the long axis of the radial head and the straight line from the distal tip of the capitellum to the coronoid process) were identified as candidates approximating the true coronal plane angulation of radial neck fractures. On the coronal plane of the CT scan, the angulation of radial neck fractures (CTa) was measured and served as the reference standard for measurement. Inter- and intraobserver reliabilities were assessed by Kappa statistics and intraclass correlation coefficient (ICC). Results. Angle U showed the strongest correlation with CTa (p < 0.001). In the analysis of inter- and intraobserver reliability, Kappa values were significantly higher for Angles S and U compared with Angle P. ICC values were excellent among the three groups. Conclusion. Angle U on AP view was the best substitute for CTa when evaluating radial neck fractures in children. Further studies are required to validate this method. Cite this article: Bone Joint J 2024;106-B(9):964–969

Limb alignment in total knee arthroplasty (TKA) influences periarticular soft-tissue tension, biomechanics through knee flexion, and implant survival. Despite this, there is no uniform consensus on the optimal alignment technique for TKA. Neutral mechanical alignment facilitates knee flexion and symmetrical component wear but forces the limb into an unnatural position that alters native knee kinematics through the arc of knee flexion. Kinematic alignment aims to restore native limb alignment, but the safe ranges with this technique remain uncertain and the effects of this alignment technique on component survivorship remain unknown. Anatomical alignment aims to restore predisease limb alignment and knee geometry, but existing studies using this technique are based on cadaveric specimens or clinical trials with limited follow-up times. Functional alignment aims to restore the native plane and obliquity of the joint by manipulating implant positioning while limiting soft tissue releases, but the results of high-quality studies with long-term outcomes are still awaited. The drawbacks of existing studies on alignment include the use of surgical techniques with limited accuracy and reproducibility of achieving the planned alignment, poor correlation of intraoperative data to long-term functional outcomes and implant survivorship, and a paucity of studies on the safe ranges of limb alignment. Further studies on alignment in TKA should use surgical adjuncts (e.g. robotic technology) to help execute the planned alignment with improved accuracy, include intraoperative assessments of knee biomechanics and periarticular soft-tissue tension, and correlate alignment to long-term functional outcomes and survivorship

Aims. Periacetabular osteotomy (PAO) is widely recognized as a demanding surgical procedure for acetabular reorientation. Reports about the learning curve have primarily focused on complication rates during the initial learning phase. Therefore, our aim was to assess the PAO learning curve from an analytical perspective by determining the number of PAOs required for the duration of surgery to plateau and the accuracy to improve. Methods. The study included 118 consecutive PAOs in 106 patients. Of these, 28 were male (23.7%) and 90 were female (76.3%). The primary endpoint was surgical time. Secondary outcome measures included radiological parameters. Cumulative summation analysis was used to determine changes in surgical duration. A multivariate linear regression model was used to identify independent factors influencing surgical time. Results. The learning curve in this series was 26 PAOs in a period of six months. After 26 PAO procedures, a significant drop in surgical time was observed and a plateau was also achieved. The mean duration of surgery during the learning curve was 103.8 minutes (SD 33.2), and 69.7 minutes (SD 18.6) thereafter (p < 0.001). Radiological correction of acetabular retroversion showed a significant improvement after having performed a total of 93 PAOs, including anteverting PAOs on 35 hips with a retroverted acetabular morphology (p = 0.005). Several factors were identified as independent variables influencing duration of surgery, including patient weight (β = 0.5 (95% confidence interval (CI) 0.2 to 0.7); p < 0.001), learning curve procedure phase of 26 procedures (β = 34.0 (95% CI 24.3 to 43.8); p < 0.001), and the degree of lateral correction expressed as the change in the lateral centre-edge angle (β = 0.7 (95% CI 0.001 to 1.3); p = 0.048). Conclusion. The learning curve for PAO surgery requires extensive surgical training at a high-volume centre, with a minimum of 50 PAOs per surgeon per year. This study defined a cut-off value of 26 PAO procedures, after which a significant drop in surgical duration occurred. Furthermore, it was observed that a retroverted morphology of the acetabulum required a greater number of procedures to acquire proficiency in consistently eliminating the crossover sign. These findings are relevant for fellows and fellowship programme directors in establishing the extent of training required to impart competence in PAO. Cite this article: Bone Joint J 2024;106-B(4):336–343

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

Objectives. The aim of this study was to assess the clinical application of, and optimize the variables used in, the BACH classification of long-bone osteomyelitis. Methods. A total of 30 clinicians from a variety of specialities classified 20 anonymized cases of long-bone osteomyelitis using BACH. Cases were derived from patients who presented to specialist centres in the United Kingdom between October 2016 and April 2017. Accuracy and Fleiss’ kappa (Fκ) were calculated for each variable. Bone involvement (B-variable) was assessed further by nine clinicians who classified ten additional cases of long bone osteomyelitis using a 3D clinical imaging package. Thresholds for defining multidrug-resistant (MDR) isolates were optimized using results from a further analysis of 253 long bone osteomyelitis cases. Results. The B-variable had a classification accuracy of 77.0%, which improved to 95.7% when using a 3D clinical imaging package (p < 0.01). The A-variable demonstrated difficulty in the accuracy of classification for increasingly resistant isolates (A1 (non-resistant), 94.4%; A2 (MDR), 46.7%; A3 (extensively or pan-drug-resistant), 10.0%). Further analysis demonstrated that isolates with four or more resistant test results or less than 80% sensitive susceptibility test results had a 98.1% (95% confidence interval (CI) 96.6 to 99.6) and 98.8% (95% CI 98.1 to 100.0) correlation with MDR status, respectively. The coverage of the soft tissues (C-variable) and the host status (H-variable) both had a substantial agreement between users and a classification accuracy of 92.5% and 91.2%, respectively. Conclusions. The BACH classification system can be applied accurately by users with a variety of clinical backgrounds. Accuracy of B-classification was improved using 3D imaging. The use of the A-variable has been optimized based on susceptibility testing results. Cite this article: A. J. Hotchen, M. Dudareva, J. Y. Ferguson, P. Sendi, M. A. McNally. The BACH classification of long bone osteomyelitis. Bone Joint Res 2019;8:459–468. DOI: 10.1302/2046-3758.810.BJR-2019-0050.R1

Aims. The purpose of this study was to compare the radiological outcomes of manual versus robotic-assisted medial unicompartmental knee arthroplasty (UKA). Methods. Postoperative radiological outcomes from 86 consecutive robotic-assisted UKAs (RAUKA group) from a single academic centre were retrospectively reviewed and compared to 253 manual UKAs (MUKA group) drawn from a prior study at our institution. Femoral coronal and sagittal angles (FCA, FSA), tibial coronal and sagittal angles (TCA, TSA), and implant overhang were radiologically measured to identify outliers. Results. When assessing the accuracy of RAUKAs, 91.6% of all alignment measurements and 99.2% of all overhang measurements were within the target range. All alignment and overhang targets were simultaneously met in 68.6% of RAUKAs. When comparing radiological outcomes between the RAUKA and MUKA groups, statistically significant differences were identified for combined outliers in FCA (2.3% vs 12.6%; p = 0.006), FSA (17.4% vs 50.2%; p < 0.001), TCA (5.8% vs 41.5%; p < 0.001), and TSA (8.1% vs 18.6%; p = 0.023), as well as anterior (0.0% vs 4.7%; p = 0.042), posterior (1.2% vs 13.4%; p = 0.001), and medial (1.2% vs 14.2%; p < 0.001) overhang outliers. Conclusion. Robotic system navigation decreases alignment and overhang outliers compared to manual UKA. Given the association between component placement errors and revision in UKA, this strong significant improvement in accuracy may improve implant survival. Level of Evidence: III. Cite this article: Bone Jt Open 2021;2-3:191–197

Aims. Natural Language Processing (NLP) offers an automated method to extract data from unstructured free text fields for arthroplasty registry participation. Our objective was to investigate how accurately NLP can be used to extract structured clinical data from unstructured clinical notes when compared with manual data extraction. Methods. A group of 1,000 randomly selected clinical and hospital notes from eight different surgeons were collected for patients undergoing primary arthroplasty between 2012 and 2018. In all, 19 preoperative, 17 operative, and two postoperative variables of interest were manually extracted from these notes. A NLP algorithm was created to automatically extract these variables from a training sample of these notes, and the algorithm was tested on a random test sample of notes. Performance of the NLP algorithm was measured in Statistical Analysis System (SAS) by calculating the accuracy of the variables collected, the ability of the algorithm to collect the correct information when it was indeed in the note (sensitivity), and the ability of the algorithm to not collect a certain data element when it was not in the note (specificity). Results. The NLP algorithm performed well at extracting variables from unstructured data in our random test dataset (accuracy = 96.3%, sensitivity = 95.2%, and specificity = 97.4%). It performed better at extracting data that were in a structured, templated format such as range of movement (ROM) (accuracy = 98%) and implant brand (accuracy = 98%) than data that were entered with variation depending on the author of the note such as the presence of deep-vein thrombosis (DVT) (accuracy = 90%). Conclusion. The NLP algorithm used in this study was able to identify a subset of variables from randomly selected unstructured notes in arthroplasty with an accuracy above 90%. For some variables, such as objective exam data, the accuracy was very high. Our findings suggest that automated algorithms using NLP can help orthopaedic practices retrospectively collect information for registries and quality improvement (QI) efforts. Cite this article: Bone Joint J 2020;102-B(7 Supple B):99–104

Aims. The preoperative diagnosis of periprosthetic joint infection (PJI) remains a challenge due to a lack of biomarkers that are both sensitive and specific. We investigated the performance characteristics of polymerase chain reaction (PCR), interleukin-6 (IL6), and calprotectin of synovial fluid in the diagnosis of PJI. Methods. We performed systematic search of PubMed, Embase, The Cochrane Library, Web of Science, and Science Direct from the date of inception of each database through to 31 May 2021. Studies which described the diagnostic accuracy of synovial fluid PCR, IL6, and calprotectin using the Musculoskeletal Infection Society criteria as the reference standard were identified. Results. Overall, 31 studies were identified: 20 described PCR, six described IL6, and five calprotectin. The sensitivity and specificity were 0.78 (95% confidence interval (CI) 0.67 to 0.86) and 0.97 (95% CI 0.94 to 0.99), respectively, for synovial PCR;, 0.86 (95% CI 0.74 to 0.92), and 0.94 (95% CI 0.90 to 0.96), respectively, for synovial IL6; and 0.94 (95% CI 0.82 to 0.98) and 0.93 (95% CI 0.85 to 0.97), respectively, for synovial calprotectin. Likelihood ratio scattergram analyses recommended clinical utility of synovial fluid PCR and IL6 as a confirmatory test only. Synovial calprotectin had utility in the exclusion and confirmation of PJI. Conclusion. Synovial fluid PCR and IL6 had low sensitivity and high specificity in the diagnosis of PJI, and is recommended to be used as confirmatory test. In contrast, synovial fluid calprotectin had both high sensitivity and specificity with utility in both the exclusion and confirmation of PJI. We recommend use of synovial fluid calprotectin studies in the preoperative workup of PJI. Cite this article: Bone Joint J 2022;104-B(3):311–320

Aims. Accurate diagnosis of chronic periprosthetic joint infection (PJI) presents a significant challenge for hip surgeons. Preoperative diagnosis is not always easy to establish, making the intraoperative decision-making process crucial in deciding between one- and two-stage revision total hip arthroplasty (THA). Calprotectin is a promising point-of-care novel biomarker that has displayed high accuracy in detecting PJI. We aimed to evaluate the utility of intraoperative calprotectin lateral flow immunoassay (LFI) in THA patients with suspected chronic PJI. Methods. The study included 48 THAs in 48 patients with a clinical suspicion of PJI, but who did not meet European Bone and Joint Infection Society (EBJIS) PJI criteria preoperatively, out of 105 patients undergoing revision THA at our institution for possible PJI between November 2020 and December 2022. Intraoperatively, synovial fluid calprotectin was measured with LFI. Cases with calprotectin levels ≥ 50 mg/l were considered infected and treated with two-stage revision THA; in negative cases, one-stage revision was performed. At least five tissue cultures were obtained; the implants removed were sent for sonication. Results. Calprotectin was positive (≥ 50 mg/l) in 27 cases; out of these, 25 had positive tissue cultures and/or sonication. Calprotectin was negative in 21 cases. There was one false negative case, which had positive tissue cultures. Calprotectin showed an area under the curve of 0.917, sensitivity of 96.2%, specificity of 90.9%, positive predictive value of 92.6%, negative predictive value of 95.2%, positive likelihood ratio of 10.6, and negative likelihood ratio of 0.04. Overall, 45/48 patients were correctly diagnosed and treated by our algorithm, which included intraoperative calprotectin measurement. This yielded a 93.8% concordance with postoperatively assessed EBJIS criteria. Conclusion. Calprotectin can be a valuable tool in facilitating the intraoperative decision-making process for cases in which chronic PJI is suspected and diagnosis cannot be established preoperatively. Cite this article: Bone Joint J 2024;106-B(5 Supple B):118–124

Aims. Synovial fluid white blood cell (WBC) count and percentage of polymorphonuclear cells (%PMN) are elevated at periprosthetic joint infection (PJI). Leucocytes produce different interleukins (IL), including IL-6, so we hypothesized that synovial fluid IL-6 could be a more accurate predictor of PJI than synovial fluid WBC count and %PMN. The main aim of our study was to compare the predictive performance of all three diagnostic tests in the detection of PJI. Methods. Patients undergoing total hip or knee revision surgery were included. In the perioperative assessment phase, synovial fluid WBC count, %PMN, and IL-6 concentration were measured. Patients were labeled as positive or negative according to the predefined cut-off values for IL-6 and WBC count with %PMN. Intraoperative samples for microbiological and histopathological analysis were obtained. PJI was defined as the presence of sinus tract, inflammation in histopathological samples, and growth of the same microorganism in a minimum of two or more samples out of at least four taken. Results. In total, 49 joints in 48 patients (mean age 68 years (SD 10; 26 females (54%), 25 knees (51%)) were included. Of these 11 joints (22%) were infected. The synovial fluid WBC count and %PMN predicted PJI with sensitivity, specificity, accuracy, PPV, and NPV of 82%, 97%, 94%, 90%, and 95%, respectively. Synovial fluid IL-6 predicted PJI with sensitivity, specificity, accuracy, PPV, and NPV of 73%, 95%, 90%, 80%, and 92%, respectively. A comparison of predictive performance indicated a strong agreement between tests. Conclusions. Synovial fluid IL-6 is not superior to synovial fluid WBC count and %PMN in detecting PJI. Level of Evidence: Therapeutic Level II. Cite this article: Bone Jt Open 2020;1-12:737–742

Aims. This study aimed to evaluate the accuracy of implant placement with robotic-arm assisted total hip arthroplasty (THA) in patients with developmental dysplasia of the hip (DDH). Methods. The study analyzed a consecutive series of 69 patients who underwent robotic-arm assisted THA between September 2018 and December 2019. Of these, 30 patients had DDH and were classified according to the Crowe type. Acetabular component alignment and 3D positions were measured using pre- and postoperative CT data. The absolute differences of cup alignment and 3D position were compared between DDH and non-DDH patients. Moreover, these differences were analyzed in relation to the severity of DDH. The discrepancy of leg length and combined offset compared with contralateral hip were measured. Results. The mean values of absolute differences (postoperative CT-preoperative plan) were 1.7° (standard deviation (SD) 2.0) (inclination) and 2.5° (SD 2.1°) (anteversion) in DDH patients, and no significant differences were found between non-DDH and DDH patients. The mean absolute differences for 3D cup position were 1.1 mm (SD 1.0) (coronal plane) and 1.2 mm (SD 2.1) (axial plane) in DDH patients, and no significant differences were found between two groups. No significant difference was found either in cup alignment between postoperative CT and navigation record after cup screws or in the severity of DDH. Excellent restoration of leg length and combined offset were achieved in both groups. Conclusion. We demonstrated that robotic-assisted THA may achieve precise cup positioning in DDH patients, and may be useful in those with severe DDH. Cite this article: Bone Joint Res 2021;10(10):629–638

Aims. Intraoperative 3D navigation (ION) allows high accuracy to be achieved in spinal surgery, but poor workflow has prevented its widespread uptake. The technical demands on ION when used in patients with adolescent idiopathic scoliosis (AIS) are higher than for other more established indications. Lean principles have been applied to industry and to health care with good effects. While ensuring optimal accuracy of instrumentation and safety, the implementation of ION and its associated productivity was evaluated in this study for AIS surgery in order to enhance the workflow of this technique. The aim was to optimize the use of ION by the application of lean principles in AIS surgery. Methods. A total of 20 consecutive patients with AIS were treated with ION corrective spinal surgery. Both qualitative and quantitative analysis was performed with real-time modifications. Operating time, scan time, dose length product (measure of CT radiation exposure), use of fluoroscopy, the influence of the reference frame, blood loss, and neuromonitoring were assessed. Results. The greatest gains in productivity were in avoiding repeat intraoperative scans (a mean of 248 minutes for patients who had two scans, and a mean 180 minutes for those who had a single scan). Optimizing accuracy was the biggest factor influencing this, which was reliant on incremental changes to the operating setup and technique. Conclusion. The application of lean principles to the introduction of ION for AIS surgery helps assimilate this method into the environment of the operating theatre. Data and stakeholder analysis identified a reproducible technique for using ION for AIS surgery, reducing operating time, and radiation exposure. Cite this article: Bone Joint J. 2020;102-B(1):5–10

Aims. The aim of this study was to establish a reliable method for producing 3D reconstruction of sonographic callus. Methods. A cohort of ten closed tibial shaft fractures managed with intramedullary nailing underwent ultrasound scanning at two, six, and 12 weeks post-surgery. Ultrasound capture was performed using infrared tracking technology to map each image to a 3D lattice. Using echo intensity, semi-automated mapping was performed to produce an anatomical 3D representation of the fracture site. Two reviewers independently performed 3D reconstructions and kappa coefficient was used to determine agreement. A further validation study was undertaken with ten reviewers to estimate the clinical application of this imaging technique using the intraclass correlation coefficient (ICC). Results. Nine of the ten patients achieved union at six months. At six weeks, seven patients had bridging callus of ≥ one cortex on the 3D reconstruction and when present all achieved union. Compared to six-week radiographs, no bridging callus was present in any patient. Of the three patients lacking sonographic bridging callus, one went onto a nonunion (77.8% sensitive and 100% specific to predict union). At 12 weeks, nine patients had bridging callus at ≥ one cortex on 3D reconstruction (100%-sensitive and 100%-specific to predict union). Presence of sonographic bridging callus on 3D reconstruction demonstrated excellent reviewer agreement on ICC at 0.87 (95% confidence interval 0.74 to 0.96). Conclusion. 3D fracture reconstruction can be created using multiple ultrasound images in order to evaluate the presence of bridging callus. This imaging modality has the potential to enhance the usability and accuracy of identification of early fracture healing. Cite this article: Bone Joint Res 2021;10(12):759–766

Aims. This study aimed to explore whether serum combined with synovial interleukin-6 (IL-6) measurement can improve the accuracy of prosthetic joint infection (PJI) diagnosis, and to establish the cut-off values of IL-6 in serum and synovial fluid in detecting chronic PJI. Methods. Patients scheduled to have a revision surgery for indications of chronic infection of knee and hip arthroplasties or aseptic loosening of an implant were prospectively screened before being enrolled into this study. The Musculoskeletal Infection Society (MSIS) definition of PJI was used for the classification of cases as aseptic or infected. Serum CRP, ESR, IL-6, and percentage of polymorphonuclear neutrophils (PMN%) and IL-6 in synovial fluid were analyzed. Statistical tests were performed to compare these biomarkers in the two groups, and receiver operating characteristic (ROC) curves and area under the curve (AUC) were analyzed for each biomarker. Results. A total of 93 patients were enrolled. There was no difference in demographic data between both groups. Synovial fluid IL-6, with a threshold of 1,855.36 pg/ml, demonstrated a mean sensitivity of 94.59% (95% confidence interval (CI) 81.8% to 99.3%) and a mean specificity of 92.86% (95% CI 82.7 to 98.0) for detecting chronic PJI. Then 6.7 pg/ml was determined to be the optimal threshold value of serum IL-6 for the diagnosis of chronic PJI, with a mean sensitivity of 97.30% (95% CI 85.8% to 99.9%) and a mean specificity of 76.79% (95% CI 63.6% to 87.0%). The combination of synovial IL-6 and serum IL-6 led to improved accuracy of 96.77% in diagnosing chronic PJI. Conclusion. The present study identified that a combination of IL-6 in serum and synovial IL-6 has the potential for further improvement of the diagnosis of PJI. Cite this article: Bone Joint Res 2020;9(9):587–592

Aims. The aim of this study was to further evaluate the accuracy of ten promising synovial biomarkers (bactericidal/permeability-increasing protein (BPI), lactoferrin (LTF), neutrophil gelatinase-associated lipocalin (NGAL), neutrophil elastase 2 (ELA-2), α-defensin, cathelicidin LL-37 (LL-37), human β-defensin (HBD-2), human β-defensin 3 (HBD-3), D-dimer, and procalcitonin (PCT)) for the diagnosis of periprosthetic joint infection (PJI), and to investigate whether inflammatory joint disease (IJD) activity affects their concentration in synovial fluid. Methods. We included 50 synovial fluid samples from patients with (n = 25) and without (n = 25) confirmed PJI from an institutional tissue bank collected between May 2015 and December 2016. We also included 22 synovial fluid samples aspirated from patients with active IJD presenting to Department of Rheumatology, the first Medical Centre, Chinese PLA General Hospital. Concentrations of the ten candidate biomarkers were measured in the synovial fluid samples using standard enzyme-linked immunosorbent assays (ELISA). The diagnostic accuracy was evaluated by receiver operating characteristic (ROC) curves. Results. BPI, LTF, NGAL, ELA-2, and α-defensin were well-performing biomarkers for detecting PJI, with areas under the curve (AUCs) of 1.000 (95% confidence interval, 1.000 to 1.000), 1.000 (1.000 to 1.000), 1.000 (1.000 to 1.000), 1.000 (1.000 to 1.000), and 0.998 (0.994 to 1.000), respectively. The other markers (LL-37, HBD-2, D-dimer, PCT, and HBD-3) had limited diagnostic value. For the five well-performing biomarkers, elevated concentrations were observed in patients with active IJD. The original best thresholds determined by the Youden index, which discriminated PJI cases from non-PJI cases could not discriminate PJI cases from active IJD cases, while elevated thresholds resulted in good performance. Conclusion. BPI, LTF, NGAL, ELA-2, and α-defensin demonstrated excellent performance for diagnosing PJI. However, all five markers showed elevated concentrations in patients with IJD activity. For patients with IJD, elevated thresholds should be considered to accurately diagnose PJI. Cite this article: Bone Joint J 2021;103-B(1):32–38

In recent years, machine learning (ML) and artificial neural networks (ANNs), a particular subset of ML, have been adopted by various areas of healthcare. A number of diagnostic and prognostic algorithms have been designed and implemented across a range of orthopaedic sub-specialties to date, with many positive results. However, the methodology of many of these studies is flawed, and few compare the use of ML with the current approach in clinical practice. Spinal surgery has advanced rapidly over the past three decades, particularly in the areas of implant technology, advanced surgical techniques, biologics, and enhanced recovery protocols. It is therefore regarded an innovative field. Inevitably, spinal surgeons will wish to incorporate ML into their practice should models prove effective in diagnostic or prognostic terms. The purpose of this article is to review published studies that describe the application of neural networks to spinal surgery and which actively compare ANN models to contemporary clinical standards allowing evaluation of their efficacy, accuracy, and relatability. It also explores some of the limitations of the technology, which act to constrain the widespread adoption of neural networks for diagnostic and prognostic use in spinal care. Finally, it describes the necessary considerations should institutions wish to incorporate ANNs into their practices. In doing so, the aim of this review is to provide a practical approach for spinal surgeons to understand the relevant aspects of neural networks. Cite this article: Bone Joint J 2021;103-B(9):1442–1448