The classification algorithms to support the management of the patient with femur fracture.

Effectiveness in health care is a specific characteristic of each intervention and outcome evaluated. Especially with regard to surgical interventions, organization, structure and processes play a key role in determining this parameter. In addition, health care services by definition operate in a context of limited resources, so rationalization of service organization becomes the primary goal for health care management. This aspect becomes even more relevant for those surgical services for which there are high volumes. Therefore, in order to support and optimize the management of patients undergoing surgical procedures, the data analysis could play a significant role. To this end, in this study used different classification algorithms for characterizing the process of patients undergoing surgery for a femoral neck fracture. The models showed significant accuracy with values of 81%, and parameters such as Anaemia and Gender proved to be determined risk factors for the patient's length of stay. The predictive power of the implemented model is assessed and discussed in view of its capability to support the management and optimisation of the hospitalisation process for femoral neck fracture, and is compared with different model in order to identify the most promising algorithms. In the end, the support of artificial intelligence algorithms laying the basis for building more accurate decision-support tools for healthcare practitioners.

Femoral shaft displacement and Winquist classification provide predictive characteristics for combined femoral neck and femoral shaft fractures.

OBJECTIVES: The goal of this study is to establish radiographic features and characteristics of patient injury in cases of femoral shaft fractures that predict the presence of ipsilateral femoral neck fractures (IFNFs). METHODS: Patient data was retrospectively assessed from a single level I trauma center through the electronic health record using (Current Procedural Terminology) CPT codes for both isolated and combined ipsilateral femoral shaft and neck fractures. Demographic information, injury characteristics, and independently reviewed radiographic features were collected and compared against the same information from a group of isolated femoral shaft fractures. Multivariable logistic regression was performed to identify risk factors for concomitant IFNFs and their respective odds ratios. A probability algorithm for assessing ipsilateral femoral neck fractures based on independent multivariate predictors was constructed and used. RESULTS: A total of 113 patients with either isolated femoral shaft fractures or combined femoral shaft and IFNF (n = 33) met inclusion criteria and were identified for this study. Fracture displacement was most strongly associated with increased risk of combined injury with an aOR of 25.64 (95 %CI = 5.96-110.28) for every 100 % displacement. Motorcycle crash (MCC) was the mechanism associated with the highest risk of combined injury, with an aOR of 9.85 (95 % CI = 1.99-48.74). Combined injury was also correlated with lower Winquist score and presentation with a closed fracture, with aORs of 0.38 (95 %CI = 0.21 - 0.68) and 11.61 (95 %CI = 1.93-69.94), respectively. Presence of at least 3 of the statistically significant variables produced a positive predictive value (PPV) of ≥ 89 % for combined femoral shaft and IFNF. CONCLUSIONS: Identification of combined femoral shaft and IFNF is of critical importance when caring for orthopedic trauma patients. While diagnosis remains a challenging task, MCC mechanism, >100 % fracture displacement, and lower Winquist classification were found to be associated with combined injuries. The combination of these variables might assist in predicting the probability of combined injury and potentially guide decision making on the appropriateness of obtaining single sequence MRI or implementing prophylactic femoral neck fixation. LEVEL OF EVIDENCE: Level III.

The History of Classification Systems for Periprosthetic Femoral Fractures: A Literature Review.

Periprosthetic femoral fractures (PPFFs) following total hip arthroplasty (THA) present a significant clinical challenge due to their increasing incidence with an aging population and evolving surgical practices. Historically, classifications were primarily based on anatomical fracture location, the stability of the implant, and bone quality surrounding the implant. We critically analyzed 25 classification systems, highlighting the emergence and adaptations of key systems such as the Vancouver classification system (VCS) and the Unified classification system (UCS), which are lauded for their simplicity and effectiveness yet require further refinement. VCS, developed in 1995, categorizes fractures based on the site, implant stability, and bone quality, and remains widely used due to its robust applicability across different clinical settings. Introduced in 2014, UCS expands the VCS to encompass all periprosthetic fractures with additional fracture types, aiming for a universal application. Despite their widespread adoption, these systems exhibit shortcomings, including the incomplete inclusion of all PPFF types and the imprecise assessment of implant stability and surrounding bone loss. These gaps can result in misclassification and suboptimal treatment outcomes. This paper suggests the necessity for ongoing improvements in classification systems to include emerging fracture types and refined diagnostic criteria, ensuring that they remain relevant to contemporary orthopedic practices and continue to facilitate the precise tailoring of treatment to patient-specific circumstances. This comprehensive historical review serves as a foundation for future innovations in classification systems, ultimately aiming to standardize PPFF treatment and improve patient prognosis.

Ipsilateral proximal and shaft femoral fractures: Results of two treatment strategies / Fracturas ipsilaterales proximales y diafisarias del fémur: resultados de 2 estrategias de tratamiento

Introduction: Ipsilateral proximal and shaft femoral fractures typically occur in young adults after high-energy trauma. No consensus exists regarding the optimal internal fixation device or surgical strategy for these complex fractures. Our main objective is to identify differences on outcomes and complications between patients treated with one or combined implants. Material and method: This is a single-center retrospective cohort study in patients with associated fractures of the proximal (31 AO) and shaft femur (32 AO). We divided the patients into two groups according to the use of single (Group I) or combined implants (Group II). Demographic, clinical, radiological, surgical data and development of complications were collected.Results: We identified 28 patients (19 men and 9 women) with an average age of 43 years. We used an anterograde femoral nail in group I (17 patients) and a retrograde femoral nail or a plate associated with hip lag screws or sliding hip screw in Group II (11 patients). Patients were followed up for 26.28 (9.12–62.88) months. Osteonecrosis of the femoral head, osteoarthritis, infection or nonunion was found in 9 patients (32%). No significant differences (p 0.70) were found in complications between two groups or between definitive surgical fixation before or after the first 24h. Conclusions: No differences in the development of complications or timing of definitive fixation were found between the use of one or combined implants in ipsilateral proximal femur and shaft fractures. Regardless of the implant chosen, an appropriate osteosynthesis technique is crucial, even so high complication rates are expected.

Introducción: Las fracturas ipsilaterales proximales y diafisarias del fémur suelen ocurrir en adultos jóvenes después de un traumatismo de alta energía. No existe consenso sobre el dispositivo de fijación interna óptimo o la estrategia quirúrgica para estas fracturas complejas. Nuestro principal objetivo es identificar las diferencias en los resultados y complicaciones entre los pacientes tratados con un implante o combinados. Material y método: Este es un estudio de cohorte retrospectivo unicéntrico en pacientes con fracturas asociadas del fémur proximal (31 AO) y diafisarias (32 AO). Dividimos a los pacientes en 2 grupos según el uso de implantes únicos (grupo i) o combinados (grupo ii). Se recogieron datos demográficos, clínicos, radiológicos, quirúrgicos y complicaciones. Resultados: Se identificaron 28 pacientes (19 hombres y 9 mujeres) con una edad promedio de 43 años. Utilizamos un clavo femoral anterógrado en el grupo i (17 pacientes) y un clavo femoral retrógrado o una placa con tornillos a compresión o tornillo deslizante de cadera en el grupo ii (11 pacientes). Los pacientes fueron seguidos durante 26,28 (9,12-62,88) meses. Se encontró osteonecrosis de la cabeza femoral, osteoartritis, infección o seudoartrosis en 9 pacientes (32%). No se encontraron diferencias significativas (p=0,70) en las complicaciones entre los 2 grupos o entre la fijación quirúrgica definitiva antes o después de las primeras 24h. Conclusiones: No se encontraron diferencias en el desarrollo de complicaciones o el momento de la fijación definitiva entre el uso de un implante o combinado en fracturas ipsilaterales de fémur proximal y diafisario. Independientemente del implante elegido, una técnica de osteosíntesis adecuada es crucial; aun así son esperables altas tasas de complicaciones.(AU)

[Artículo traducido] Fracturas ipsilaterales proximales y diafisarias del fémur: resultados de dos estrategias de tratamiento / Ipsilateral proximal and shaft femoral fractures: Results of two treatment strategies

Introduction: Ipsilateral proximal and shaft femoral fractures typically occur in young adults after high-energy trauma. No consensus exists regarding the optimal internal fixation device or surgical strategy for these complex fractures. Our main objective is to identify differences on outcomes and complications between patients treated with one or combined implants. Material and method: This is a single-center retrospective cohort study in patients with associated fractures of the proximal (31 AO) and shaft femur (32 AO). We divided the patients into two groups according to the use of single (Group I) or combined implants (Group II). Demographic, clinical, radiological, surgical data and development of complications were collected.Results: We identified 28 patients (19 men and 9 women) with an average age of 43 years. We used an anterograde femoral nail in group I (17 patients) and a retrograde femoral nail or a plate associated with hip lag screws or sliding hip screw in Group II (11 patients). Patients were followed up for 26.28 (9.12–62.88) months. Osteonecrosis of the femoral head, osteoarthritis, infection or nonunion was found in 9 patients (32%). No significant differences (p 0.70) were found in complications between two groups or between definitive surgical fixation before or after the first 24h. Conclusions: No differences in the development of complications or timing of definitive fixation were found between the use of one or combined implants in ipsilateral proximal femur and shaft fractures. Regardless of the implant chosen, an appropriate osteosynthesis technique is crucial, even so high complication rates are expected.

Introducción: Las fracturas ipsilaterales proximales y diafisarias del fémur suelen ocurrir en adultos jóvenes después de un traumatismo de alta energía. No existe consenso sobre el dispositivo de fijación interna óptimo o la estrategia quirúrgica para estas fracturas complejas. Nuestro principal objetivo es identificar las diferencias en los resultados y complicaciones entre los pacientes tratados con un implante o combinados. Material y método: Este es un estudio de cohorte retrospectivo unicéntrico en pacientes con fracturas asociadas del fémur proximal (31 AO) y diafisarias (32 AO). Dividimos a los pacientes en 2 grupos según el uso de implantes únicos (grupo i) o combinados (grupo ii). Se recogieron datos demográficos, clínicos, radiológicos, quirúrgicos y complicaciones. Resultados: Se identificaron 28 pacientes (19 hombres y 9 mujeres) con una edad promedio de 43 años. Utilizamos un clavo femoral anterógrado en el grupo i (17 pacientes) y un clavo femoral retrógrado o una placa con tornillos a compresión o tornillo deslizante de cadera en el grupo ii (11 pacientes). Los pacientes fueron seguidos durante 26,28 (9,12-62,88) meses. Se encontró osteonecrosis de la cabeza femoral, osteoartritis, infección o seudoartrosis en 9 pacientes (32%). No se encontraron diferencias significativas (p=0,70) en las complicaciones entre los 2 grupos o entre la fijación quirúrgica definitiva antes o después de las primeras 24h. Conclusiones: No se encontraron diferencias en el desarrollo de complicaciones o el momento de la fijación definitiva entre el uso de un implante o combinado en fracturas ipsilaterales de fémur proximal y diafisario. Independientemente del implante elegido, una técnica de osteosíntesis adecuada es crucial; aun así son esperables altas tasas de complicaciones.(AU)

Treatment of B1 Distal Periprosthetic Femur Fractures.

The burden of periprosthetic distal femoral fractures is projected to increase accordingly with the increase in total knee arthroplasties (TKAs) performed globally in the future. Less invasive plating and intramedullary (IM) nailing techniques still seem to provide similar outcomes based on current literature. Double-plating and combination techniques may prove to be beneficial in the future pending further large-scale studies but currently have not demonstrated superiority over single plating and IM nailing based on current evidence. Distal femoral replacement may provide a useful option for future treatment, provided it is performed by a trained knee arthroplasty surgeon.

Reliability and validity of the Unified Classification System for postoperative periprosthetic femoral fractures around cemented polished taper-slip stems.

AIMS: This aim of this study was to assess the reliability and validity of the Unified Classification System (UCS) for postoperative periprosthetic femoral fractures (PFFs) around cemented polished taper-slip (PTS) stems. METHODS: Radiographs of 71 patients with a PFF admitted consecutively at two centres between 25 February 2012 and 19 May 2020 were collated by an independent investigator. Six observers (three hip consultants and three trainees) were familiarized with the UCS. Each PFF was classified on two separate occasions, with a mean time between assessments of 22.7 days (16 to 29). Interobserver reliability for more than two observers was assessed using percentage agreement and Fleiss' kappa statistic. Intraobserver reliability between two observers was calculated with Cohen kappa statistic. Validity was tested on surgically managed UCS type B PFFs where stem stability was documented in operation notes (n = 50). Validity was assessed using percentage agreement and Cohen kappa statistic between radiological assessment and intraoperative findings. Kappa statistics were interpreted using Landis and Koch criteria. All six observers were blinded to operation notes and postoperative radiographs. RESULTS: Interobserver reliability percentage agreement was 58.5% and the overall kappa value was 0.442 (moderate agreement). Lowest kappa values were seen for type B fractures (0.095 to 0.360). The mean intraobserver reliability kappa value was 0.672 (0.447 to 0.867), indicating substantial agreement. Validity percentage agreement was 65.7% and the mean kappa value was 0.300 (0.160 to 0.4400) indicating only fair agreement. CONCLUSION: This study demonstrates that the UCS is unsatisfactory for the classification of PFFs around PTS stems, and that it has considerably lower reliability and validity than previously described for other stem types. Radiological PTS stem loosening in the presence of PFF is poorly defined and formal intraoperative testing of stem stability is recommended. Cite this article: Bone Joint J 2021;103-B(8):1339-1344.

Outcomes of operatively treated interprosthetic femoral fractures.

AIMS: The prevalence of ipsilateral total hip arthroplasty (THA) and total knee arthroplasty (TKA) is rising in concert with life expectancy, putting more patients at risk for interprosthetic femur fractures (IPFFs). Our study aimed to assess treatment methodologies, implant survivorship, and IPFF clinical outcomes. METHODS: A total of 76 patients treated for an IPFF from February 1985 to April 2018 were reviewed. Prior to fracture, at the hip/knee sites respectively, 46 femora had primary/primary, 21 had revision/primary, three had primary/revision, and six had revision/revision components. Mean age and BMI were 74 years (33 to 99) and 30 kg/m2 (21 to 46), respectively. Mean follow-up after fracture treatment was seven years (2 to 24). RESULTS: Overall, 59 fractures were classified as Vancouver C (Unified Classification System (UCS) D), 17 were Vancouver B (UCS B). In total, 57 patients (75%) were treated with open reduction and internal fixation (ORIF); three developed nonunion, three developed periprosthetic joint infection, and two developed aseptic loosening. In all, 18 patients (24%) underwent revision arthroplasty including 13 revision THAs, four distal femoral arthroplasties (DFAs), and one revision TKA: of these, one patient developed aseptic loosening and two developed nonunion. Survivorship free from any reoperation was 82% (95% confidence interval (CI) 66.9% to 90.6%) and 77% (95% CI 49.4% to 90.7%) in the ORIF and revision groups at two years, respectively. ORIF patients who went on to union tended to have stemmed knee components and greater mean interprosthetic distance (IPD = 189 mm (SD 73.6) vs 163 mm (SD 36.7); p = 0.546) than nonunited fractures. Patients who went on to nonunion in the revision arthroplasty group had higher medullary diameter: cortical width ratio (2.5 (SD 1.7) vs 1.3 (SD 0.3); p = 0.008) and lower IPD (36 mm (SD 30.6) vs 214 mm (SD 32.1); p < 0.001). At latest follow-up, 95% of patients (n = 72) were ambulatory. CONCLUSION: Interprosthetic femur fractures are technically and biologically challenging cases. Individualized approaches to internal fixation versus revision arthroplasty led to an 81% (95% CI 68.3% to 88.6%) survivorship free from reoperation at two years with 95% of patients ambulatory. Continued improvements in management are warranted. Cite this article: Bone Joint J 2021;103-B(7 Supple B):122-128.

Selected Vancouver B2 periprosthetic femoral fractures around cemented polished femoral components can be safely treated with osteosynthesis.

AIMS: We aimed to compare the implant survival, complications, readmissions, and mortality of Vancouver B2 periprosthetic femoral fractures (PFFs) treated with internal fixation with that of B1 PFFs treated with internal fixation and B2 fractures treated with revision arthroplasty. METHODS: We retrospectively reviewed the data of 112 PFFs, of which 47 (42%) B1 and 27 (24%) B2 PFFs were treated with internal fixation, whereas 38 (34%) B2 fractures underwent revision arthroplasty. Decision to perform internal fixation for B2 PFFs was based on specific radiological (polished femoral components, intact bone-cement interface) and clinical criteria (low-demand patient). Median follow-up was 36.4 months (24 to 60). Implant survival and mortality over time were estimated with the Kaplan-Meier method. Adverse events (measured with a modified Dindo-Clavien classification) and 90-day readmissions were additionally compared between groups. RESULTS: In all, nine (8.01%) surgical failures were detected. All failures occurred within the first 24 months following surgery. The 24-month implant survival was 95.4% (95% confidence interval (CI) 89.13 to 100) for B1 fractures treated with internal fixation, 90% (95% CI 76.86 to 100) for B2 PFFs treated with osteosynthesis-only, and 85.8% (95% CI 74.24 to 97.36) for B2 fractures treated with revision THA, without significant differences between groups (p = 0.296). Readmissions and major adverse events including mortality were overall high, but similar between groups (p > 0.05). The two-year patient survival rate was 87.1% (95% CI 77.49 to 95.76), 66.7% (95% CI 48.86 to 84.53), and 84.2% (95% CI 72.63 to 95.76), for the B1 group, B2 osteosynthesis group, and B2 revision group, respectively (p = 0.102). CONCLUSION: Implant survival in Vancouver B2 PFFs treated with internal fixation was similar to that of B1 fractures treated with the same method and to B2 PFFs treated with revision arthroplasty. Low-demand, elderly patients with B2 fractures around well-cemented polished femoral components with an intact bone-cement interface can be safely treated with internal fixation. Cite this article: Bone Joint J 2021;103-B(7):1222-1230.

Deep neural networks with promising diagnostic accuracy for the classification of atypical femoral fractures.

Background and purpose - A correct diagnosis is essential for the appropriate treatment of patients with atypical femoral fractures (AFFs). The diagnostic accuracy of radiographs with standard radiology reports is very poor. We derived a diagnostic algorithm that uses deep neural networks to enable clinicians to discriminate AFFs from normal femur fractures (NFFs) on conventional radiographs.Patients and methods - We entered 433 radiographs from 149 patients with complete AFF and 549 radiographs from 224 patients with NFF into a convolutional neural network (CNN) that acts as a core classifier in an automated pathway and a manual intervention pathway (manual improvement of image orientation). We tested several deep neural network structures (i.e., VGG19, InceptionV3, and ResNet) to identify the network with the highest diagnostic accuracy for distinguishing AFF from NFF. We applied a transfer learning technique and used 5-fold cross-validation and class activation mapping to evaluate the diagnostic accuracy.Results - In the automated pathway, ResNet50 had the highest diagnostic accuracy, with a mean of 91% (SD 1.3), as compared with 83% (SD 1.6) for VGG19, and 89% (SD 2.5) for InceptionV3. The corresponding accuracy levels for the intervention pathway were 94% (SD 2.0), 92% (2.7), and 93% (3.7), respectively. With regards to sensitivity and specificity, ResNet outperformed the other networks with a mean AUC (area under the curve) value of 0.94 (SD 0.01) and surpassed the accuracy of clinical diagnostics.Interpretation - Artificial intelligence systems show excellent diagnostic accuracies for the rare fracture type of AFF in an experimental setting.

Progression of Asymptomatic Contralateral Femur in Patients with Complete Atypical Femoral Fracture, According to Initial Radiographic Findings.

BACKGROUND: Atypical femoral fracture (AFF) is associated with high prevalence rates of prodromal symptoms, bilateral involvement, complications, and the need for prophylactic fixation due to a risk of an impending fracture. Although most complete AFF cases have an asymptomatic contralateral femur at the initial presentation, there is lack of clarity on its progression. We evaluated the radiographic progression of asymptomatic contralateral femora in patients with a complete AFF and investigated the characteristics of these patients. METHODS: The medical records of 80 consecutive patients who had been treated for a complete AFF were retrospectively evaluated. We excluded 14 patients who had been lost to follow-up, 10 whose contralateral femur initially had been symptomatic and had been treated simultaneously, and 3 whose contralateral femur had previously been treated surgically. The remaining 53 patients were all women with an average age of 71.8 years. The average duration of bisphosphonate (BP) use was 63.6 months, and the mean follow-up duration was 48.9 months. All of the contralateral femora were asymptomatic and were divided into 2 grades according to the initial radiographic findings. We evaluated the prevalence of radiographic progression according to the grades and compared patient characteristics between the progression and non-progression groups. RESULTS: Radiographic progression was noted in 18 patients (34%) during the follow-up: 3 (12%) of 25 in grade 1 and 15 (53.6%) of 28 in grade 2 (p < 0.001). The mean time to progression for these 18 patients was 25.6 months, which also differed significantly depending on the grade (p = 0.02). Eleven and 9 (61.1% and 25.7%) of the patients received BP postoperatively in the progression and non-progression groups, respectively (p = 0.01). CONCLUSIONS: The prevalence of radiographic progression was relatively high, even though the contralateral femur was initially asymptomatic, and differed significantly according to the initial radiographic grade. The frequency of postoperative BP use was significantly higher in the progression group. LEVEL OF EVIDENCE: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

The inverted Vancouver C fracture. Case series of unstable proximal femur fractures above a knee revision stem treated by short cephalomedullary nail and lateral submuscular overlapping plate.

Unstable proximal femur fractures above a knee revision stem are an emerging complication that is especially difficult to treat. Since this pattern does not adapt to any previously reported classification, we named it "inverted Vancouver C fracture". In this single-centre case series, we pose a nail-plate combination for the treatment of such clinical picture. The incidence was low among proximal and implant-related femoral fractures. All the fractures healed without records of major local complications. Thus, we consider this technique safe and reproducible.

B2 or not B2? That is the question: a review of periprosthetic fractures around cemented taper-slip femoral components.

AIMS: Periprosthetic fractures (PPFs) around cemented taper-slip femoral prostheses often result in a femoral component that is loose at the prosthesis-cement interface, but where the cement-bone interface remains well-fixed and bone stock is good. We aim to understand how best to classify and manage these fractures by using a modification of the Vancouver classification. METHODS: We reviewed 87 PPFs. Each was a first episode of fracture around a cemented femoral component, where surgical management consisted of revision surgery. Data regarding initial injury, intraoperative findings, and management were prospectively collected. Patient records and serial radiographs were reviewed to determine fracture classification, whether the bone cement was well fixed (B2W) or loose (B2L), and time to fracture union following treatment. RESULTS: In total, 47 B2W fractures (54.0%) and one B3 fracture (1.1%) had cement that remained well-fixed at the cement-bone interface. These cases were treated with cement-in-cement (CinC) revision arthroplasty. Overall, 43 fractures with follow-up united, and two patients sustained further fractures secondary to nonunion and required further revision surgery. A total of 19 B2L fractures (21.8%) and 19 B3 fractures (21.8%) had cement that was loose at the cement-bone interface. These cases were managed by revision arthroplasty with either cemented or uncemented femoral components, or proximal femoral arthroplasty. One case could not be classified. CONCLUSION: We endorse a modification of the original Vancouver system to include a subclassification of B2 fractures around cemented femoral prostheses to include B2W (where cement is well-fixed to bone) and B2L (where the cement is loose). Fractures around taper-slip design stems are more likely to fracture in a B2W pattern compared to fractures around composite beam design stems which are more likely to fracture in a B2L pattern. B2W fractures can reliably be managed with CinC revision. Cite this article: Bone Joint J 2021;103-B(1):71-78.

Atypical periprosthetic femoral fracture associated with long-term bisphosphonate therapy.

BACKGROUND: Atypical femoral fracture is one of the many complications after the long-term use of bisphosphonates. The American Society for Bone and Mineral Research has officially excluded periprosthetic femoral fractures (PFFs) from the definition of atypical femoral fractures (AFFs). Several case reports found that PFFs can occur with characteristics similar to those of AFFs. The purpose of our study was to evaluate the proportion of atypical fractures among Vancouver type B1 fractures, and to determine the association between the long-term use of bisphosphonates and the occurrence of atypical periprosthetic femoral fractures (APFFs). METHODS: In this retrospective study, we reviewed 41 patients with Vancouver type B1 periprosthetic fractures between January 1, 2011 and December 31, 2018. We classified them into two groups, namely atypical and typical PFFs, based on the fracture morphology. We noted the proportion of atypical periprosthetic fractures among B1 fractures and identified risk factors. RESULTS: Among the 41 PFFs, 5 (13%) fractures were classified as atypical PFF based on the radiological characteristics. The longer duration of bisphosphonate use was probably the only independent risk factor that significantly increases the occurrence of APFF (p = 0.03, 0.08 (CI 0.008 - 0.16)). There were no significant differences in age, gender, body mass index, comorbidities, corticosteroid use, positioning of the femoral stem, the method of fixation (cemented or cementless) and time lapse from before the primary prosthesis implantation to the PFF in the development of atypical fracture type. CONCLUSIONS: There seems to be a correlation between the long-term intake of bisphosphonates and the atypical periprosthetic fracture. Atypical femoral fracture can also occur in the periprosthetic form. TRIAL REGISTRATION: Study number: 22/2019-SZTE, http://www.klinikaikutatas.hu/hu/kutatasetika/jovahagyott-vizsgalatok-koezerdeku-adatai/category/25-jovahagyott-vizsgalatok-kozerdeku-adatai-rkeb-2019.html?download=985:22-2019 .

Classification of femur fracture in pelvic X-ray images using meta-learned deep neural network.

In the medical field, various studies using artificial intelligence (AI) techniques have been attempted. Numerous attempts have been made to diagnose and classify diseases using image data. However, different forms of fracture exist, and inaccurate results have been confirmed depending on condition at the time of imaging, which is problematic. To overcome this limitation, we present an encoder-decoder structured neural network that utilizes radiology reports as ancillary information at training. This is a type of meta-learning method used to generate sufficiently adequate features for classification. The proposed model learns representation for classification from X-ray images and radiology reports simultaneously. When using a dataset of only 459 cases for algorithm training, the model achieved a favorable performance in a test dataset containing 227 cases (classification accuracy of 86.78% and classification F1 score of 0.867 for fracture or normal classification). This finding demonstrates the potential for deep learning to improve performance and accelerate application of AI in clinical practice.

Validity and reliability of the Unified Classification System applied to periprosthetic femur fractures: a comparison with the Vancouver system.

Objective: The Unified Classification System (UCS) presents itself as an evolution of the Vancouver Classification (VCS) for the evaluation of periprosthetic fractures of the proximal femur (PPF). The aim of our study was to highlight any loss of reproducibility or validity of the new classification system, compared to the previous one.Material and methods: We tested the interobserver and intraobserver agreement using 40 PPF clinical cases. Each classifying subtype of the UCS and VCS was present in at least two cases. Six experienced hip surgeons (Senior Surgeon, SS) and 5 surgeons in training (Junior Surgeon, JS) classified the clinical cases, using VCS and UCS. The validity of both classifications was then tested with intraoperative surveys.Results: The mean κ value for interobserver agreement for the VCS in the JS group was 0.65 and 0.81 for the SS group. The mean κ value for interobserver agreement for the UCS in the JS group was 0.63 and 0.65 for the SS group. The mean κ value for intraobserver agreement for the VCS in the JS group was 0.71 and 0.73 for the SS group. The mean κ value for intraobserver agreement for the UCS in the JS group was 0.72 and 0.7 for the SS group. Validity analysis showed a moderate agreement for the VCS and a good agreement for the UCS.Conclusion: The UCS completes the Vancouver classification, expanding it. It is reliable, despite the increase in classification categories and number of parameters to evaluate, with a slightly higher validity.

Expert Consensus for a Principle-based Classification for Treatment of Diaphyseal Pediatric Femur Fractures.

INTRODUCTION: Recent studies demonstrate considerable deviation from the American Academy of Orthopaedic Surgeons (AAOS) evidence-based guidelines for the treatment of pediatric diaphyseal femur fractures (PDFFs). This study aimed to determine if expert-consensus can be reached on a principle-based classification to be applied broadly to a wide variety of PDFF scenarios and if outcomes correspond to adherence to the classification. METHODS: A 2-stage study was performed. First, a survey of experts using a principle-based approach to PDFF. We conducted a survey of 17 thought-leaders (criteria≥20 y' experience+authors of the seminal pediatric femur fracture studies) who were asked to classify 15 cases of PDFF using the principle-based classification for agreement. Next, we conducted a retrospective review of 289 consecutive PDFF treated (2011-2015) at a level 1 pediatric trauma center. For each case, we compared the actual treatment and proposed "ideal" principle-based classification. We then compared clinical results and outcome data points including the length of stay, physician visits, and hospital charge data. RESULTS: A substantial (κ=0.7) expert-agreement was noted for assigning treatment principles with near-perfect (κ=0.93) agreement on conservative versus surgical management. We obtained agreement on employing a flexible implant (κ=0.84) rigid fixation (κ=0.75) and damage control philosophy (κ=0.64). Suboptimal clinical results were noted in 43% of the undertreated patients (24/56), 18.8% of the adequately treated, and 14.3% of overtreated (P<0.01) patients. An increasing trend for the length of hospital stay and a number of clinic visits was noted as the treatment class increased (P<0.01). Charges were 4.2 times higher for an episode of operative versus nonoperative care (P<0.01). Rigid fixation (class 4) had significantly (P=0.01) higher total and material charges than flexible fixation (class 3). DISCUSSION: The proposed classification has a substantial agreement among thought-leaders. Clinical results demonstrated significantly more suboptimal results in undertreated fractures, compared with ideally treated or more invasively treated fractures. More invasive treatments led to increased burden to families and the system in terms of length of stay and hospital charges. LEVEL OF EVIDENCE: Level III.

Precise proximal femur fracture classification for interactive training and surgical planning.

PURPOSE: Demonstrate the feasibility of a fully automatic computer-aided diagnosis (CAD) tool, based on deep learning, that localizes and classifies proximal femur fractures on X-ray images according to the AO classification. The proposed framework aims to improve patient treatment planning and provide support for the training of trauma surgeon residents. MATERIAL AND METHODS: A database of 1347 clinical radiographic studies was collected. Radiologists and trauma surgeons annotated all fractures with bounding boxes and provided a classification according to the AO standard. In all experiments, the dataset was split patient-wise in three with the ratio 70%:10%:20% to build the training, validation and test sets, respectively. ResNet-50 and AlexNet architectures were implemented as deep learning classification and localization models, respectively. Accuracy, precision, recall and [Formula: see text]-score were reported as classification metrics. Retrieval of similar cases was evaluated in terms of precision and recall. RESULTS: The proposed CAD tool for the classification of radiographs into types "A," "B" and "not-fractured" reaches a [Formula: see text]-score of 87% and AUC of 0.95. When classifying fractures versus not-fractured cases it improves up to 94% and 0.98. Prior localization of the fracture results in an improvement with respect to full-image classification. In total, 100% of the predicted centers of the region of interest are contained in the manually provided bounding boxes. The system retrieves on average 9 relevant images (from the same class) out of 10 cases. CONCLUSION: Our CAD scheme localizes, detects and further classifies proximal femur fractures achieving results comparable to expert-level and state-of-the-art performance. Our auxiliary localization model was highly accurate predicting the region of interest in the radiograph. We further investigated several strategies of verification for its adoption into the daily clinical routine. A sensitivity analysis of the size of the ROI and image retrieval as a clinical use case were presented.