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. 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. Given the possible radiation damage and inaccuracy of radiological investigations, particularly in children, ultrasound and superb microvascular imaging (SMI) may offer alternative methods of evaluating new bone formation when limb lengthening is undertaken in paediatric patients. The aim of this study was to assess the use of ultrasound combined with SMI in monitoring new bone formation during limb lengthening in children. Methods. In this retrospective cohort study, ultrasound and radiograph examinations were performed every two weeks in 30 paediatric patients undergoing limb lengthening. Ultrasound was used to monitor new bone formation. The number of vertical vessels and the blood flow resistance index were compared with those from plain radiographs. Results. We categorized the new bone formation into three stages: stage I (early lengthening), in which there was no obvious callus formation on radiographs and ultrasound; stage II (lengthening), in which radiographs showed low-density callus formation with uneven distribution and three sub-stages could be identified on ultrasound: in Ia punctate callus was visible; in IIb there was linear callus formation which was not yet connected and in IIc there was continuous linear callus. In stage III (healing), the bone ends had united, the periosteum was intact, and the callus had disappeared, as confirmed on radiographs, indicating healed bone. A progressive increase in the number of vertical vessels was noted in the early stages, peaking during stages IIb and IIc, followed by a gradual decline (p < 0.001). Delayed healing involved patients with a prolonged stage IIa or those who regressed to stage IIa from stages IIb or IIc during lengthening. Conclusion. We found that the formation of new bone in paediatric patients undergoing limb lengthening could be reliably evaluated using ultrasound when combined with the radiological findings. This combination enabled an improved assessment of the prognosis, and adjustments to the lengthening protocol. While SMI offered additional insights into angiogenesis within the new bone, its role primarily contributed to the understanding of the microvascular environment rather than directly informing adjustments of treatment. Cite this article: Bone Joint J 2024;106-B(7):751–758

Aims. To investigate if preoperative CT improves detection of unstable trochanteric hip fractures. Methods. A single-centre prospective study was conducted. Patients aged 65 years or older with trochanteric hip fractures admitted to Stavanger University Hospital (Stavanger, Norway) were consecutively included from September 2020 to January 2022. Radiographs and CT images of the fractures were obtained, and surgeons made individual assessments of the fractures based on these. The assessment was conducted according to a systematic protocol including three classification systems (AO/Orthopaedic Trauma Association (OTA), Evans Jensen (EVJ), and Nakano) and questions addressing specific fracture patterns. An expert group provided a gold-standard assessment based on the CT images. Sensitivities and specificities of surgeons’ assessments were estimated and compared in regression models with correlations for the same patients. Intra- and inter-rater reliability were presented as Cohen’s kappa and Gwet’s agreement coefficient (AC1). Results. We included 120 fractures in 119 patients. Compared to radiographs, CT increased the sensitivity of detecting unstable trochanteric fractures from 63% to 70% (p = 0.028) and from 70% to 76% (p = 0.004) using AO/OTA and EVJ, respectively. Compared to radiographs alone, CT increased the sensitivity of detecting a large posterolateral trochanter major fragment or a comminuted trochanter major fragment from 63% to 76% (p = 0.002) and from 38% to 55% (p < 0.001), respectively. CT improved intra-rater reliability for stability assessment using EVJ (AC1 0.68 to 0.78; p = 0.049) and for detecting a large posterolateral trochanter major fragment (AC1 0.42 to 0.57; p = 0.031). Conclusion. A preoperative CT of trochanteric fractures increased detection of unstable fractures using the AO/OTA and EVJ classification systems. Compared to radiographs, CT improved intra-rater reliability when assessing fracture stability and detecting large posterolateral trochanter major fragments. Cite this article: Bone Jt Open 2024;5(6):524–531

Aims. Acetabular retroversion is a recognized cause of hip impingement and can be influenced by pelvic tilt (PT), which changes in different functional positions. Positional changes in PT have not previously been studied in patients with acetabular retroversion. Methods. Supine and standing anteroposterior (AP) pelvic radiographs were retrospectively analyzed in 69 patients treated for symptomatic acetabular retroversion. Measurements were made for acetabular index (AI), lateral centre-edge angle (LCEA), crossover index, ischial spine sign, and posterior wall sign. The change in the angle of PT was measured both by the sacro-femoral-pubic (SFP) angle and the pubic symphysis to sacroiliac (PS-SI) index. Results. In the supine position, the mean PT (by SFP) was 1.05° (SD 3.77°), which changed on standing to a PT of 8.64° (SD 5.34°). A significant increase in posterior PT from supine to standing of 7.59° (SD 4.5°; SFP angle) and 5.89° (SD 3.33°; PS-SI index) was calculated (p < 0.001). There was a good correlation in PT change between measurements using SFP angle and PS-SI index (0.901 in the preoperative group and 0.815 in the postoperative group). Signs of retroversion were significantly reduced in standing radiographs compared to supine: crossover index (0.16 (SD 0.16) vs 0.38 (SD 0.15); p < 0.001), crossover sign (19/28 hips vs 28/28 hips; p < 0.001), ischial spine sign (10/28 hips vs 26/28 hips; p < 0.001), and posterior wall sign (12/28 hips vs 24/28 hips; p < 0.001). Conclusion. Posterior PT increased from supine to standing in patients with symptomatic acetabular retroversion. The features of acetabular retroversion were less evident on standing radiographs. The low PT angle in the supine position is a factor in the increased appearance of acetabular retroversion. Patients presenting with symptoms of hip impingement should be assessed by supine and standing pelvic radiographs to highlight signs of acetabular retroversion, and to assist with optimizing acetabular correction at the time of surgery. Cite this article: Bone Joint J 2022;104-B(7):786–791

Aims. Total hip arthroplasty (THA) and total knee arthroplasty (TKA) are common orthopaedic procedures requiring postoperative radiographs to confirm implant positioning and identify complications. Artificial intelligence (AI)-based image analysis has the potential to automate this postoperative surveillance. The aim of this study was to prepare a scoping review to investigate how AI is being used in the analysis of radiographs following THA and TKA, and how accurate these tools are. Methods. The Embase, MEDLINE, and PubMed libraries were systematically searched to identify relevant articles. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for scoping reviews and Arksey and O’Malley framework were followed. Study quality was assessed using a modified Methodological Index for Non-Randomized Studies tool. AI performance was reported using either the area under the curve (AUC) or accuracy. Results. Of the 455 studies identified, only 12 were suitable for inclusion. Nine reported implant identification and three described predicting risk of implant failure. Of the 12, three studies compared AI performance with orthopaedic surgeons. AI-based implant identification achieved AUC 0.992 to 1, and most algorithms reported an accuracy > 90%, using 550 to 320,000 training radiographs. AI prediction of dislocation risk post-THA, determined after five-year follow-up, was satisfactory (AUC 76.67; 8,500 training radiographs). Diagnosis of hip implant loosening was good (accuracy 88.3%; 420 training radiographs) and measurement of postoperative acetabular angles was comparable to humans (mean absolute difference 1.35° to 1.39°). However, 11 of the 12 studies had several methodological limitations introducing a high risk of bias. None of the studies were externally validated. Conclusion. These studies show that AI is promising. While it already has the ability to analyze images with significant precision, there is currently insufficient high-level evidence to support its widespread clinical use. Further research to design robust studies that follow standard reporting guidelines should be encouraged to develop AI models that could be easily translated into real-world conditions. Cite this article: Bone Joint J 2022;104-B(8):929–937

Aims. Ankle fracture fixation is commonly performed by junior trainees. Simulation training using cadavers may shorten the learning curve and result in a technically superior surgical performance. Methods. We undertook a preliminary, pragmatic, single-blinded, multicentre, randomized controlled trial of cadaveric simulation versus standard training. Primary outcome was fracture reduction on postoperative radiographs. Results. Overall, 139 ankle fractures were fixed by 28 postgraduate year three to five trainee surgeons (mean age 29.4 years; 71% males) during ten months' follow-up. Under the intention-to-treat principle, a technically superior fixation was performed by the cadaveric-trained group compared to the standard-trained group, as measured on the first postoperative radiograph against predefined acceptability thresholds. The cadaveric-trained group used a lower intraoperative dose of radiation than the standard-trained group (mean difference 0.011 Gym. 2. , 95% confidence interval 0.003 to 0.019; p = 0.009). There was no difference in procedure time. Conclusion. Trainees randomized to cadaveric training performed better ankle fracture fixations and irradiated patients less during surgery compared to standard-trained trainees. This effect, which was previously unknown, is likely to be a consequence of the intervention. Further study is required. Cite this article: Bone Jt Open 2023;4(8):594–601

Aims. The primary aim of this study was to report the radiological outcomes of patients with a dorsally displaced distal radius fracture who were randomized to a moulded cast or surgical fixation with wires following manipulation and closed reduction of their fracture. The secondary aim was to correlate radiological outcomes with patient-reported outcome measures (PROMs) in the year following injury. Methods. Participants were recruited as part of DRAFFT2, a UK multicentre clinical trial. Participants were aged 16 years or over with a dorsally displaced distal radius fracture, and were eligible for the trial if they needed a manipulation of their fracture, as recommended by their treating surgeon. Participants were randomly allocated on a 1:1 ratio to moulded cast or Kirschner wires after manipulation of the fracture in the operating theatre. Standard posteroanterior and lateral radiographs were performed in the radiology department of participating centres at the time of the patient’s initial assessment in the emergency department and six weeks postoperatively. Intraoperative fluoroscopic images taken at the time of fracture reduction were also assessed. Results. Patients treated with surgical fixation with wires had less dorsal angulation of the radius versus those treated in a moulded cast at six weeks after manipulation of the fracture; the mean difference of -4.13° was statistically significant (95% confidence interval 5.82 to -2.45). There was no evidence of a difference in radial shortening. However, there was no correlation between these radiological measurements and PROMs at any timepoint in the 12 months post-injury. Conclusion. For patients with a dorsally displaced distal radius fracture treated with a closed manipulation, surgical fixation with wires leads to less dorsal angulation on radiographs at six weeks compared with patients treated in a moulded plaster cast alone. However, the difference in dorsal angulation was small and did not correlate with patient-reported pain and function. Cite this article: Bone Jt Open 2024;5(2):132–138

Aims. In the Netherlands, general practitioners (GPs) can request radiographs. After a radiologically diagnosed fracture, patients are immediately referred to the emergency department (ED). Since 2020, the Máxima Medical Centre has implemented a new care pathway for minor trauma patients, referring them immediately to the traumatology outpatient clinic (OC) instead of the ED. We investigated whether this altered care pathway leads to a reduction in healthcare consumption and concomitant costs. Methods. In this retrospective cohort study, patients were included if a radiologist diagnosed a fracture on a radiograph requested by the GP from August to October 2019 (control group) or August to October 2020 (research group), on weekdays between 8.30 am and 4.00 pm. The study compared various outcomes between groups, including the length of the initial hospital visit, frequency of hospital visits and medical procedures, extent of imaging, and healthcare expenses. Results. A total of 634 patients were included. The results show a median reduction of 25 minutes in duration of initial hospital visits, one fewer hospital visit, overall fewer medical procedures, and a decrease in healthcare costs of €303.40 per patient in the research group compared to the control group. No difference was found in the amount of imaging. Conclusion. The implementation of the new care pathway has resulted in a substantial reduction in healthcare use and costs. Moreover, the pathway provides advantages for patients and helps prevent crowding at the ED. Hence, we recommend immediately referring all minor trauma patients to the traumatology OC instead of the ED. Cite this article: Bone Jt Open 2023;4(10):728–734

Aims. The aims of this study were to describe the epidemiology of metacarpal shaft fractures (MSFs), assess variation in treatment and complications following standard care, document hospital resource use, and explore factors associated with treatment modality. Methods. A multicentre, cross-sectional retrospective study of MSFs at six centres in the UK. We collected and analyzed healthcare records, operative notes, and radiographs of adults presenting within ten days of a MSF affecting the second to fifth metacarpal between 1 August 2016 and 31 July 2017. Total emergency department (ED) attendances were used to estimate prevalence. Results. A total of 793 patients (75% male, 25% female) with 897 MSFs were included, comprising 0.1% of 837,212 ED attendances. The annual incidence of MSF was 40 per 100,000. The median age was 27 years (IQR 21 to 41); the highest incidence was in men aged 16 to 24 years. Transverse fractures were the most common. Over 80% of all fractures were treated non-surgically, with variation across centres. Overall, 12 types of non-surgical and six types of surgical treatment were used. Fracture pattern, complexity, displacement, and age determined choice of treatment. Patients who were treated surgically required more radiographs and longer radiological and outpatient follow-up, and were more likely to be referred for therapy. Complications occurred in 5% of patients (39/793). Most patients attended planned follow-up, with 20% (160/783) failing to attend at least one or more clinic appointments. Conclusion. MSFs are common hand injuries among young, working (economically active) men, but there is considerable heterogeneity in treatment, rehabilitation, and resource use. They are a burden on healthcare resources and society, thus further research is needed to optimize treatment. Cite this article: Bone Jt Open 2024;5(8):652–661

Aims. To determine whether side-bending films in scoliosis are assessed for adequacy in clinical practice; and to introduce a novel method for doing so. Methods. Six surgeons and eight radiographers were invited to participate in four online surveys. The generic survey comprised erect and left and right bending radiographs of eight individuals with scoliosis, with an average age of 14.6 years. Respondents were asked to indicate whether each bending film was optimal (adequate) or suboptimal. In the first survey, they were also asked if they currently assessed the adequacy of bending films. A similar second survey was sent out two weeks later, using the same eight cases but in a different order. In the third survey, a guide for assessing bending film adequacy was attached along with the radiographs to introduce the novel T1-45B method, in which the upper endplate of T1 must tilt ≥ 45° from baseline for the study to be considered optimal. A fourth and final survey was subsequently conducted for confirmation. Results. Overall, 12 (86%) of 14 respondents did not use any criteria to assess the bending film adequacy; the remaining two each described a different invalidated method. In total, 12 (86%) of the respondents felt T1-45B was easy to learn and apply. There was fair to substantial intra-rater reliability (k = 0.25 to 0.88) which improved to fair to almost perfect (k = 0.38 to 0.88) post-introduction of the guide. Inter-rater reliability varied considerably among the rater groups but similarly increased following introduction of the guide (k. S1. = 0.19 to 0.34, k. S2. = 0.33 to 0.43 vs k. S3. = 0.49 to 0.5, k. S4. = 0.35 to 0.43). Conclusion. Many surgeons and radiographers do not assess spinal bending films for adequacy. We propose that the change in the plane of the upper endplate of T1 on side-bending can be used in this evaluation. In the T1-45B method, a change of ≥ 45° on side bending qualifies as an adequate bend effort. Cite this article: Bone Jt Open 2023;4(9):689–695

Aims. Large bone defects resulting from osteolysis, fractures, osteomyelitis, or metastases pose significant challenges in acetabular reconstruction for total hip arthroplasty. This study aimed to evaluate the survival and radiological outcomes of an acetabular reconstruction technique in patients at high risk of reconstruction failure (i.e. periprosthetic joint infection (PJI), poor bone stock, immunosuppressed patients), referred to as Hip Reconstruction In Situ with Screws and Cement (HiRISC). This involves a polyethylene liner embedded in cement-filled bone defects reinforced with screws and/or plates for enhanced fixation. Methods. A retrospective chart review of 59 consecutive acetabular reconstructions was performed by four surgeons in a single institution from 18 October 2018 to 5 January 2023. Cases were classified based on the Paprosky classification, excluding type 1 cases (n = 26) and including types 2 or 3 for analysis (n = 33). Radiological loosening was evaluated by an orthopaedic surgeon who was not the operating surgeon, by comparing the immediate postoperative radiographs with the ones at latest follow-up. Mean follow-up was 557 days (SD 441; 31 to 1,707). Results. Out of the 33 cases analyzed, six (18.2%) constructs required revision, with four revisions due to uncontrolled infection, one for dislocation, and one for aseptic loosening. Among the 27 non-revised constructs, only one showed wider radiolucencies compared to immediate postoperative radiographs, indicating potential loosening. Patients who underwent revision (n = 6) were significantly younger and had a higher BMI compared to those with non-revised constructs (p = 0.016 and p = 0.026, respectively). Sex, race, ethnicity, American Society of Anesthesiologists grade, infection status (patients with postoperative PJI diagnosis (septic) vs patients without such diagnosis (aseptic)), and mean follow-up did not significantly differ between revised and non-revised groups. Conclusion. The HiRISC technique may serve as a feasible short-term (about one to two years) alternative in patients with large acetabular defects, particularly in cases of PJI. Longer follow-up is necessary to establish the long-term survival of this technique. Cite this article: Bone Joint J 2024;106-B(5 Supple B):82–88

Aims. Radiological residual acetabular dysplasia (RAD) has been reported in up to 30% of children who had successful brace treatment of infant developmental dysplasia of the hip (DDH). Predicting those who will resolve and those who may need corrective surgery is important to optimize follow-up protocols. In this study we have aimed to identify the prevalence and predictors of RAD at two years and five years post-bracing. Methods. This was a single-centre, prospective longitudinal cohort study of infants with DDH managed using a published, standardized Pavlik harness protocol between January 2012 and December 2016. RAD was measured at two years’ mean follow-up using acetabular index-lateral edge (AI-L) and acetabular index-sourcil (AI-S), and at five years using AI-L, AI-S, centre-edge angle (CEA), and acetabular depth ratio (ADR). Each hip was classified based on published normative values for normal, borderline (1 to 2 standard deviations (SDs)), or dysplastic (> 2 SDs) based on sex, age, and laterality. Results. Of 202 infants who completed the protocol, 181 (90%) had two and five years’ follow-up radiographs. At two years, in 304 initially pathological hips, the prevalence of RAD (dysplastic) was 10% and RAD (borderline) was 30%. At five years, RAD (dysplastic) decreased to 1% to 3% and RAD (borderline) decreased to < 1% to 2%. On logistic regression, no variables were predictive of RAD at two years. Only AI-L at two years was predictive of RAD at five years (p < 0.001). If both hips were normal at two years’ follow-up (n = 96), all remained normal at five years. In those with bilateral borderline hips at two years (n = 21), only two were borderline at five years, none were dysplastic. In those with either borderline-dysplastic or bilateral dysplasia at two years (n = 26), three (12%) were dysplastic at five years. Conclusion. The majority of patients with RAD at two years post-brace treatment, spontaneously resolved by five years. Therefore, children with normal radiographs at two years post-brace treatment can be discharged. Targeted follow-up for those with abnormal AI-L at two years will identify the few who may benefit from surgical correction at five years’ follow-up. Cite this article: Bone Joint J 2024;106-B(7):744–750

Aims. Although the short stem concept in hip arthroplasty procedure shows acceptable clinical performance, we sometimes get unexplainable radiological findings. The aim of this retrospective study was to evaluate changes of radiological findings up to three years postoperatively, and to assess any potential contributing factors on such radiological change in a Japanese population. Methods. This is a retrospective radiological study conducted in Japan. Radiological assessment was done in accordance with predetermined radiological review protocol. A total of 241 hips were included in the study and 118 hips (49.0%) revealed radiological change from immediately after surgery to one year postoperatively; these 118 hips were eligible for further analyses. Each investigator screened whether either radiolucent lines (RLLs), cortical hypertrophy (CH), or atrophy (AT) appeared or not on the one-year radiograph. Further, three-year radiographs of eligible cases were reviewed to determine changes such as, disappeared (D), improved (I), stable (S), and progression (P). Additionally, bone condensation (BC) was assessed on the three-year radiograph. Results. CH was observed in 49 hips (21.1%), AT was observed in 63 hips (27.2%), and RLLs were observed in 34 hips (14.7%) at one year postoperatively. Among 34 hips with RLLs, 70.6% showed change of either D or I on the three-year radiograph. BC was observed in younger patients more frequently. Conclusion. The Fitmore stem works well in a Japanese population with favourable radiological change on hips with RLLs. Longer-term follow-up is required to determine clinical relevance. Cite this article: Bone Jt Open 2022;3(1):20–28