Background

Fibrous dysplasia is a developmental anomaly of bone formation that may exist in a monostotic or polystotic form. Surgical treatment is considered advisable only with presence of significant or progressive deformity or persistent pain. Early surgery is indicated before the tumor expands or fracture occurs.

The most important outcome predictor of Legg-Calvé-Perthes disease (LCPD) is the shape of the healed femoral head. However, the deformity of the femoral head is currently evaluated by non-reproducible, categorical, and qualitative classifications. In this regard, recent advances in computer vision might provide the opportunity to automatically detect and delineate the outlines of bone in radiographic images for calculating a continuous measure of femoral head deformity. This study aimed to construct a pipeline for accurately detecting and delineating the proximal femur in radiographs of LCPD patients employing existing algorithms. To detect the proximal femur, the pretrained stateof-the-art object detection model, YOLOv5, was trained on 1580 manually annotated radiographs, validated on 338 radiographs, and tested on 338 radiographs. Additionally, 200 radiographs of shoulders and chests were added to the dataset to make the model more robust to false positives and increase generalizability. The convolutional neural network architecture, U-Net, was then employed to segment the detected proximal femur. The network was trained on 80 manually annotated radiographs using real-time data augmentation to increase the number of training images and enhance the generalizability of the segmentation model. The network was validated on 60 radiographs and tested on 60 radiographs. The object detection model achieved a mean Average Precision (mAP) of 0.998 using an Intersection over Union (IoU) threshold of 0.5, and a mAP of 0.712 over IoU thresholds of 0.5 to 0.95 on the test set. The segmentation model achieved an accuracy score of 0.912, a Dice Coefficient of 0.937, and a binary IoU score of 0.854 on the test set. The proposed fully automatic proximal femur detection and segmentation system provides a promising method for accurately detecting and delineating the proximal femoral bone contour in radiographic images, which is necessary for further image analysis

Gunshot-induced fractures of the proximal femur typically present with severe comminution and bone loss. These fractures may also be associated with local damage to soft tissue, neurovascular structures and injuries to abdominal organs. The aim was to evaluate the outcomes of civilian gunshot injuries to the proximal femur at a major trauma center in South Africa. A retrospective review of all patients who sustained gunshot-induced proximal femur fractures between January 2014 and December 2017 was performed. Patients with gunshot injuries involving the hip joint, neck of femur or pertrochanteric fractures were included. Patient demographics, clinical- treatment and outcome data were collected. Results are reported as appropriate given the distribution of continuous data or as frequencies and counts. Our study included 78 patients who sustained 79 gunshot-induced proximal femur fractures. The mean age of patients was 31 ± 112, and the majority of patients were male (93.6%). Pertrochantenteric fractures were the most common injuries encountered (73.4%). Treatment included cephalomedullary nail (60.8%), arthrotomy and internal fixation (16.4%) and interfragmentary fixation with cannulated screws (6%). One case of complete neck of femur fracture had fixation failure, which required conversion to total hip arthroplasty. The overall union rate was 69.6%, and 6.3% of patients developed a fracture-related infection in cases who completed follow-up. The study shows an acceptable union rate when managing these fractures and a low risk of infection. As challenging as they are, individual approaches for each fracture and managing each fracture according to their merits yield acceptable outcomes

Resection of the proximal femur raises several challenges to the orthopedic oncology surgeon. Among these is the re-establishment of the abductor mechanism that might impacts on hip function. Extent of tumor resection and surgeons’ preferences dictate the reconstruction method of the abductors. While some surgeons advocate the necessity of greater trochanter (GT) preservation whenever possible, others attempt direct soft tissues reattachment to the prosthesis. Sparse data in the literature evaluated the outcomes of greater trochanter fixation to the proximal femur megaprosthesis. This is a retrospective monocentric study. All patients who received a proximal femoral replacement after tumor resection between 2005 and 2021 with a minimum follow-up of three months were included. Patients were divided into two groups: (1) those with preserved GT reattached to the megaprosthesis and (2) those with direct or indirect (tenodesis to fascia lata) abductor muscles reattachment. Both groups were compared for surgical outcomes (dislocation and revision rates) and functional outcomes (Trendelenburg gait, use of walking-assistive device and abductor muscle strength). Additionally patients in group 1 were subdivided into patients who received GT reinsertion using a grip and cables and those who got direct GT reinsertion using suture materials and studied for GT displacement at three, six and 12 months. Time to cable rupture was recorded and analyzed through a survival analysis. Fifty-six patients were included in this study with a mean follow-up of 45 months (3-180). There were 23 patients with reinserted GT (group 1) and 33 patients with soft tissue repair (group 2). Revision rate was comparable between both groups(p=0.23); however, there were more dislocations in group 2 (0/23 vs 6/33; p=0.037). Functional outcomes were comparable, with 78% of patients in group 1 (18/23) and 73% of patients in group 2 (24/33) that displayed a Trendelenburg gait (p=0.76). In group 1, 70% (16/23) used walking aids compared to 79% of group 2 (27/33) (p=0.34). Mean abductor strength reached 2.7 in group 1 compared to 2.3 in group 2 (p=0.06). In group 1, 16 of the 23 patients had GT reinsertion with grip and cables. Median survival of cables for these 16 patients reached 13 months in our series. GT displacement reached a mean of two mm, three mm, and 11 mm respectively at three, six and 12 months of follow-up in patients with grip and cables compared to 12 mm, 24 mm and 26 mm respectively at the same follow-up intervals in patients with GT stand-alone suture reinsertion(p<0.05). Although GT preservation and reinsertion did not improve functional outcomes after proximal femur resection and reconstruction with a megaprosthesis, it was significantly associated with lower dislocation rate despite frequent cable failure and secondary GT migration. No cable or grip revision or removal was recorded. Significantly less displacement was observed in patients for whom GT reattachment used plate and cables rather than sutures only. Therefore we suggest that GT should be preserved and reattached whenever possible and that GT reinsertion benefits from strong materials such as grip and cables

Advances in cancer therapy have prolonged cancer patient survival even in the presence of disseminated disease and an increasing number of cancer patients are living with metastatic bone disease (MBD). The proximal femur is the most common long bone involved in MBD and pathologic fractures of the femur are associated with significant morbidity, mortality and loss of quality of life (QoL). Successful prophylactic surgery for an impending fracture of the proximal femur has been shown in multiple cohort studies to result in patients more likely to walk after surgery, longer survival, lower transfusion rates and shorter post-operative hospital stays. However, there is currently no optimal method to predict a pathologic fracture. The most well-known tool is Mirel's criteria, established in 1989 and is limited from guiding clinical practice due to poor specificity and sensitivity. The goal of our study is to train a convolutional neural network (CNN) to predict fracture risk when metastatic bone disease is present in the proximal femur. Our fracture risk prediction tool was developed by analysis of prospectively collected data for MBD patients (2009–2016) in order to determine which features are most commonly associated with fracture. Patients with primary bone tumors, pathologic fractures at initial presentation, and hematologic malignancies were excluded. A total of 1146 patients comprising 224 pathologic fractures were included. Every patient had at least one Anterior-Posterior X-ray. The clinical data includes patient demographics, tumor biology, all previous radiation and chemotherapy received, multiple pain and function scores, medications and time to fracture or time to death. Each of Mirel's criteria has been further subdivided and recorded for each lesion. We have trained a convolutional neural network (CNN) with X-ray images of 1146 patients with metastatic bone disease of the proximal femur. The digital X-ray data is converted into a matrix representing the color information at each pixel. Our CNN contains five convolutional layers, a fully connected layers of 512 units and a final output layer. As the information passes through successive levels of the network, higher level features are abstracted from the data. This model converges on two fully connected deep neural network layers that output the fracture risk. This prediction is compared to the true outcome, and any errors are back-propagated through the network to accordingly adjust the weights between connections. Methods to improve learning included using stochastic gradient descent with a learning rate of 0.01 and a momentum rate of 0.9. We used average classification accuracy and the average F1 score across test sets to measure model performance. We compute F1 = 2 x (precision x recall)/(precision + recall). F1 is a measure of a test's accuracy in binary classification, in our case, whether a lesion would result in pathologic fracture or not. Five-fold cross validation testing of our fully trained model revealed accurate classification for 88.2% of patients with metastatic bone disease of the proximal femur. The F1 statistic is 0.87. This represents a 24% error reduction from using Mirel's criteria alone to classify the risk of fracture in this cohort. This is the first reported application of convolutional neural networks, a machine learning algorithm, to an important Orthopaedic problem. Our neural network model was able to achieve impressive accuracy in classifying fracture risk of metastatic proximal femur lesions from analysis of X-rays and clinical information. Our future work will aim to validate this algorithm on an external cohort

Advances in cancer therapy have prolonged patient survival even in the presence of disseminated disease and an increasing number of cancer patients are living with metastatic bone disease (MBD). The proximal femur is the most common long bone involved in MBD and pathologic fractures of the femur are associated with significant morbidity, mortality and loss of quality of life (QoL). Successful prophylactic surgery for an impending fracture of the proximal femur has been shown in multiple cohort studies to result in longer survival, preserved mobility, lower transfusion rates and shorter post-operative hospital stays. However, there is currently no optimal method to predict a pathologic fracture. The most well-known tool is Mirel's criteria, established in 1989 and is limited from guiding clinical practice due to poor specificity and sensitivity. The ideal clinical decision support tool will be of the highest sensitivity and specificity, non-invasive, generalizable to all patients, and not a burden on hospital resources or the patient's time. Our research uses novel machine learning techniques to develop a model to fill this considerable gap in the treatment pathway of MBD of the femur. The goal of our study is to train a convolutional neural network (CNN) to predict fracture risk when metastatic bone disease is present in the proximal femur. Our fracture risk prediction tool was developed by analysis of prospectively collected data of consecutive MBD patients presenting from 2009–2016. Patients with primary bone tumors, pathologic fractures at initial presentation, and hematologic malignancies were excluded. A total of 546 patients comprising 114 pathologic fractures were included. Every patient had at least one Anterior-Posterior X-ray and clinical data including patient demographics, Mirel's criteria, tumor biology, all previous radiation and chemotherapy received, multiple pain and function scores, medications and time to fracture or time to death. We have trained a convolutional neural network (CNN) with AP X-ray images of 546 patients with metastatic bone disease of the proximal femur. The digital X-ray data is converted into a matrix representing the color information at each pixel. Our CNN contains five convolutional layers, a fully connected layers of 512 units and a final output layer. As the information passes through successive levels of the network, higher level features are abstracted from the data. The model converges on two fully connected deep neural network layers that output the risk of fracture. This prediction is compared to the true outcome, and any errors are back-propagated through the network to accordingly adjust the weights between connections, until overall prediction accuracy is optimized. Methods to improve learning included using stochastic gradient descent with a learning rate of 0.01 and a momentum rate of 0.9. We used average classification accuracy and the average F1 score across five test sets to measure model performance. We compute F1 = 2 x (precision x recall)/(precision + recall). F1 is a measure of a model's accuracy in binary classification, in our case, whether a lesion would result in pathologic fracture or not. Our model achieved 88.2% accuracy in predicting fracture risk across five-fold cross validation testing. The F1 statistic is 0.87. This is the first reported application of convolutional neural networks, a machine learning algorithm, to this important Orthopaedic problem. Our neural network model was able to achieve reasonable accuracy in classifying fracture risk of metastatic proximal femur lesions from analysis of X-rays and clinical information. Our future work will aim to externally validate this algorithm on an international cohort

Introduction: There is a clear need for the development of more sensitive risk assessment tools for clinical predictors of fractures. Bone densitometries are limited in the ability to account for complex geometry, architecture, and heterogeneity of bone. Quantitative computed tomography (QCT)-based finite element (FE) Methods: (QCT/FEM) are able to perform structural analyses taking these factors into consideration to accurately predict bone strength. However, no basic data have been available regarding predicted strength (PS) of the proximal femur by QCT/FEM with reference to age in a normal population. The purpose of this study was thus to create a database on PS in a normal population as a preliminary trial. With these data, parameters that affect PS were also analyzed. Methods: Participants in this study comprised individuals who participated in a health checkup program with computed tomography (CT) at our hospital in 2008. Participants included 487 men and 237 women (age range, 40–87 years). Exclusion criteria were provided. Scan data of the proximal femur were isolated and taken from overall data from CT of each participant with simultaneous scans of a calibration phantom containing hydroxyapatite rods. A FE model was constructed from the isolated data using Mechanical Finder software. For each of the FE models, loading and boundary conditions as well as the definition of PS were exactly the same as described by Bessho et al. (Bone 2009). For each participant, height, weight, and abdominal circumference (AC) were measured. The analyses included linear regression analysis relating age and PS, one-way analysis of variance to compare average PS among the groups of participants who were divided into 5-year age brackets, and multiple regression analysis to determine how PS was affected by age, height, weight, and AC. Differences were considered significant for values of p< 0.05. Result: The following results were obtained. First, average PS was lower in women than in men for all age ranges. Second, PS in men under stance configuration, and those in women under stance and fall configurations significantly decreased with age. Third, weight positively affected PS in both men and women. Discussion: This was the first study to investigate changes in PS with age in a normal population. Whether PS by QCT/FEM correlates more closely with fracture risk for osteoporotic patients in comparison to other bone densitometries remains unclear, but the our results did not contradict any existing concept of risk factors for fragility fracture. More baseline data for PS in normal populations need to be accumulated by increasing the number of participants in studies like this

Case Report: Metastatic deposits in the proximal femur commonly result in pathological fracture. Conventionally these fractures are treated surgically, by internal fixation or arthroplasty. The emphasis in treating these fractures is on restoring stability to the proximal femur and relieving pain. We present two cases in which pathological fractures of the proximal femur secondary to metastatic renal carcinoma were treated conservatively with excellent functional outcomes. In both cases, the medical condition of the patient precluded surgery. A 68 year old male with a subcapital fracture of the proximal femur was treated with bedrest and mobilisation. At 6 months he was able to mobilise with crutches, swim, and had returned to almost all normal activities despite non-union of the fracture. A 63 year old male had a pathological fracture of the proximal femur treated by DCS fixation. The fracture failed to unite and the plate fractured. Despite this the patient was able to walk with crutches, pain free. Discussion: After a pathological fracture of the proximal femur conservative management can lead to satisfactory analgesia, function and therefore quality of life

Introduction: The anatomy of the proximal femur is an important factor in the design of uncemented femoral prostheses for which the quality of fixation and the associated bony remodelling depend on the primary stability and optimal transmission of forces to the proximal femur. This study looks at the variation in the diameter of the proximal femur with age and sex in a homogeneous population. Materials and Methods: We studied standardised pre-operative antero-posterior radiographs of the proximal femur of 2,777 patients who have undergone total hip arthroplasty using a custom implant over a 10 year period. The radiographs were corrected for magnification and a measurement made of the endosteal diameter at the narrowest point of the proximal femur. These measurements were used in the design and manufacture of the custom femoral implant. Results: Of 2777 patients, 1588 were female and 1189 male. The mean age for females was 69.9 years (Range 30–92) and for males 67.2 years (Range 34–92). The mean proximal femoral canal diameter was 12.67mm for females and 13.36mm for males. The mean diameter of the proximal femur increased from 12.99mm in males less than 60 years to 13.47mm in those of over 60. This increase was not statistically significant (p-value 0.064, 95% CI). In females there was a statistically significant increase in the mean diameter from 11.38mm in the under 60 age group to 12.90mm in those over 60 ( p-value 0.000, 95% CI). Conclusions: The increase in the diameter of the proximal femur with age especially in females presents a significant challenge to the design and long-term survivability of uncemented femoral components. This is more so when viewed against the already good long term results available for cemented femoral implants

Objective. To three-dimensionally reconstruct the proximal femur of DDH (Developmental dysplasia of the hip) and measure the related anatomic parameters, so that we could have a further understanding of the morphological variation of the proximal femur of DDH, which would help in the preoperative planning and prosthesis design specific for DDH. Methods. From Jan.2012 to Dec.2014, 38 patients (47 hips) of DDH were admitted and 30 volunteers (30 hips) were selected as controls. All hips from both groups were examined by CT scan and radiographs. The Crowe classification method was applied. The CT data were imported into Mimics 17.0. The three-dimensional models of the proximal femur were then reconstructed, and the following parameters were measured: neck-shaft angle, neck length, offset, height of the centre of femoral head, height of the isthmus, height of greater trochanter, the medullary canal diameter of isthmus(Di), the medullary canal diameter 10mm above the apex of the lesser trochanter(DT+10), the medullary canal diameter 20mm below the apex of the lesser trochanter(DT-20), and then DT+10/Di, DT-20/Di and DT+10/DT-20 were calculated. Results. There is no significant difference in neck-shaft angle between Crowe I-III DDH and the control group, while the neck-shaft angle is much smaller in Crowe IV DDH. The neck length of Crowe IV DDH is much smaller than those of Crowe I-III DDH. As for Di there is neither significant difference between Crowe I DDH and the control group, nor significant difference between CroweII-III and Crowe IV, but the difference is significant between the first two groups and the latter two groups. DT+10/DT-20 and the offset have no significant difference between the control group and DDH groups. DT-20, DT+10, DT+10/Di and DT-20/Di are much smaller in Crowe IV DDH than that in Crowe I-III and the control groups. Height of greater trochanter in Crowe IV is larger than those in Crowe I-III and the control group. Height of the centre of femoral head in Crowe IV DDH is smaller than those in Crowe I-III DDH and the control group. The height of the isthmus in Crowe IV is much smaller than those in Crowe I-III DDH and the control group. Conclusion. The neck-shaft angle in DDH groups is not larger than that in the control group, while in contrast, it's much smaller in Crowe IV DDH than that in the control group. Comparing to Crowe I-III DDH and the control group, Crowe IV DDH has a dramatic change in the intramedullary and extramedullary parameters. The isthmus and the great trochanter are higher and there is apparent narrowing of the medullary canal around the level of the lesser trochanter

Introduction: Estimates suggest that 50% of new cases of invasive cancer diagnosed each year will eventually metastasise to bone. The proximal end of the femur is the most common site of long bone involvement by metastatic disease. Accepted principles for the treatment of metastatic disease of the proximal femur have been published. The results of 31 consecutive patients treated with a long intramedullary hip screw for metastatic disease of the proximal femur are reported. Patients and Methods: Retrospective case note review of all patients that had a long intramedullary hip screw for metastatic disease affecting the proximal femur over a four-year period 1998–2002. Results: The case notes of 31 patients (33 femurs) were reviewed. There were 21 females and 12 males with a mean age of 71 years. 31 femurs were Zickel group Ia or Ib, the remaining 2 were impending pathological fractures (Zickel group II) that were fixed prophylactically. Post operatively all patients were allowed to fully weight bear. 70 % of patients regained their initial level of mobility or increased their level of dependence by a factor of one. Mean hospital stay was 20.8 days (mode 7 days). Patients that died post operatively had a mean survival of 299 days (range 2–1034). Those patients that were still alive at the last follow up had a mean survival of 475 days (range 7–1384). There were no cases of fixation or implant failure. There was one case of deep infection that was treated by implant removal. Conclusion: On the basis of these findings, the long intramedullary hip screw fulfils the principles for treatment of metastatic disease and can be recommended for the treatment of pathological or impending pathological fractures of the proximal femur

Introduction. The human body is a complex and continually adapting organism. It is theorised that the morphology of the proximal femur is closely related to that of the distal femur. Patients that have abnormal anatomy in the proximal femur, such as a high femoral neck anteversion angle, may have abnormal anatomy in the distal femur to overcome proximal differences. This phenomenon is of key interest when performing Total Hip Replacement (THR) or Total Knee Replacement (TKR) surgery. The current design and placement of existing hip and knee implants does not account for any correlation between the anatomical parameters of the proximal and distal femur, where bone anatomy may have adapted to compromise for abnormalities. A preliminary study of 21 patients has been carried out to assess the relationship between the proximal and distal femur. The difficulties in defining and measuring key anatomical parameters on the femur have been widely discussed in the literature [1] due to its complex three dimensional geometry. Using CT scans of healthy octogenarians, it was possible to mark key anatomical landmarks which could be used to define various anatomical axes throughout the femur. Correlation analyses could then be carried out on these parameters to assess the relationship between proximal and distal femur morphology. Methods. Each femur was initially realigned along the mechanical axis (MA); defined by joining the centre of the femoral head (FHC) to the centre of the intercondylar notch (INC) [2]. All anatomical landmarks were then identified using the Materialise Mimics v12 software (Figure 1 and 2) and exported into Microsoft Excel for analysis. Key anatomical parameters which were derived from these landmarks included the femoral neck axis (FNA), femoral neck anteversion angle (FNAA) [1–4], condylar twist angle, clinical transepicondylar axis (TEA), trochlea sulcus angle and medial and lateral trochlea twist. A correlation analysis was carried out on SPSS Statistics v20 (IBM) to assess the relationship between proximal and distal anatomical parameters. Results. The correlation analysis displayed a positive linear correlation between the FNAA and the clinical TEA (adjusted R squared = 0.471, p < 0.001) indicating that an abnormally high FNAA is correlated with a higher TEA angle (Figure 3). No strong relationship was found between the FNAA and the additional distal parameters compared, in particular there was no trend between the FNAA and the geometry of the trochlea as measured by the sulcus angle and trochlea twist. Discussion. The morphology of the distal femur seems to be at least partially correlated with the proximal femur and the relationship should be studied further to assess any potential effect on THA and TKA surgery. An extension of this study should assess an increased patient sample size and further anatomical parameters

The management of proximal femoral bone loss is a significant challenge in revision hip arthroplasty. A possible solution is the use of a modular proximal femur endoprosthesis (EPR). Although the survivorship and functional outcome of megaprostheses used in tumour surgery has been well described, outcome of EPRs used in revision hip surgery has received less attention. The aim of this study was to determine the 5-year outcome following proximal femur EPR and determine factors that influence it. This was a retrospective consecutive case series of all EPRs (n=80) performed for non-neoplastic indications, by 6 surgeons, in our tertiary referral centre, between 2005–2014. Patient demographics and relevant clinical details were determined from notes. The most common indications for the use of EPRs included infection (n=40), peri-prosthetic fracture (n=12) and failed osteosynthesis of proximal femoral fractures/complex trauma (n=11). Outcome measures included complication and re-operation rates, implant survival and assessment of functional outcome using the Oxford-Hip-Score (OHS). The mean age at surgery was 69 years and mean follow-up was 4 (0 – 11) years. The mean number of previous hip operations was 2.4 (range: 0 – 17). Twenty-five patients sustained a complication (31%), the most common being infection (n=9) and dislocation (n=4). By follow-up, further surgery was required in 18 (22%) hips, 9 of which were EPR revisions. 5-yr implant survivorship was 87% (95%CI: 76 – 98%). Mean OHS was 28 (range: 4 – 48). Inferior survival and outcome were seen in EPRs performed for the treatment of infection. Infection eradication was achieved in 34/41 with the index EPR procedure and in 40/41 hips by follow-up. Limb salvage was achieved in all cases and acceptable complication- and re-operation rates were seen. EPRs for periprosthetic fractures and failed osteosynthesis had best outcome. We recommend the continued use of proximal femur EPR in complex revision surgery

Failed open reduction internal fixation (ORIF) of the proximal femur can render patients severely disabled. This study analyzed the short-term functional results and complications of total hip arthroplasty (THA) for complications of ORIF of the proximal femur. Using the Hamilton Arthroplasty Database, thirty-six patients treated with a THA for complications of ORIF of the proximal femur were compared to a matched cohort. Analysis showed that THA for complications of ORIF of the proximal femur is a successful procedure despite increased intraoperative difficulty that results in comparatively lower short-term Harris Hip Scores. No statistically significant differences in intraoperative or postoperative complications were noted. Open reduction internal fixation (ORIF) of the proximal femur is a common, successful orthopedic procedure. However, failed ORIF of the proximal femur can render patients severely disabled. The purpose of this study is to analyze and compare the short-term functional results and complications of total hip arthroplasty (THA) for complications of ORIF of the proximal femur. After ethics approval, the Hamilton Arthroplasty Registry, a prospective database, was used to identify thirty-seven patients treated with THA for complications of ORIF of the proximal femur. From September 1998 to the present a group consisting of sixteen males and twenty females, with a mean age of sixty-seven, were matched to a cohort of patients treated with a primary THA. Using Wilcoxon Test and Chi-Square Tests, the two groups were compared (p< 0.05). Initially, ORIF was used to treat thirty-six patients for proximal femur fracture. The mean follow-up was 13.5 months. The experimental group had a significantly lower (p=0.035) Harris Hip Score at the one year follow-up, however both groups showed a significant improvement from preoperative scores (p=0.0001). A significant difference was noted between the two groups in estimated blood loss (p=0.01) and operative time (p=0.01). There was no significant difference in complication rate. THA for complications of ORIF of the proximal femur is a successful procedure improving patient’s pain and functional status. This is a more complicated procedure than primary THA, at times requiring the use of a revision stem, which results in significantly lower Harris Hip Scores. Nonetheless, there appears to be no comparative increase in short-term complications

Introduction. Many uncemented femoral implant designs have had successful outcomes in total hip arthroplasty (THA). Different uncemented stem designs achieve initial and long term stability through shape, size, coating and fit. There is increasing emphasis on bone preservation, particularly in younger and more active patients. The desire to optimize load transfer has led to the development of short stems that seek to achieve fixation in the proximal femur. Short stems designed to achieve stability by engaging the metaphysis or the proximal femoral necks are currently in clinical use. The purpose of this study was to examine the extent to which five stems designed to achieve proximal fixation contact the bone in the proximal femur. Using three-dimensional CT models of 30 femurs, we assessed the fit, fill and contact of each of the five different implants. Methods. Using three-dimensional computerized templating software designed to navigate robotic surgery, pre-operative CT scans of 30 patients were analyzed. Each of five femoral implant designs (TRILOCK, ARC, ABGII, CITATION, ACCOLADE) was then optimized for size and fit based on manufacturer technique guide and design rationale. The proximal femoral metaphysis was divided into four zones in the axial plane. Five contact points were determined on the frontal plane using anatomical landmarks. Each zone was assessed for cortical contact and fill of the bone-implant interface. We graded contact from 1 to 5, with 5 being 100% contact. Results. In the 150 different templates analyzed significant variability existed in contact areas of the proximal femur depending on implant design and femoral morphology. High femoral neck resection design (ARC) had the greatest contact area in the most proximal zones (Figure 1). The ABG II and Trilock stems had comparable contact in the antero-medial zones, while the ABG II had greater fill in the sagittal plane (Figures 2 & 3). The Trilock was the only stem that consistently achieved lateral cortical contact at the distal landmarks. All stems showed a pattern of mostly posteromedial contact proximally and mostly anteromedial distally. Discussion. To our knowledge, this is the first study to examine the contact points of metaphyseal engaging stems in the proximal femur. By directly comparing implant contact points in the same femur we found significant variability in the extent of fit, fill and contact of the metaphysis. These differences in proximal femoral contact are like to have implications for fixation in bone of varying quality and for long term proximal bone remodeling

This retrospective clinical study describes our experience of the use of growing endoprostheses in children with primary malignant tumours of the proximal femur and analyses the results. Between 1983 and 1996 we treated nine children with primary bone tumors of the proximal femur by resection and proximal femoral extensible replacements. Outcomes measured were function of the limb using Musculoskeletal Tumor Society score, oncologic outcome, complications and equalization of limb length. Results: Four patients died as a result of pulmonary metastases. The remaining five patients were observed for an average follow-up period of 7. 6 years (range 11–12. 7 years). One patient had a hindquarter amputation for uncontrolled infection. In these five patients we performed an average of 10. 2 operative procedures per patient (range of 3–17 procedures) including 5 lengthening procedures (range of 1–8 procedures) and a mean total extension of 69. 7 mm per patient. Acetabular loosening and hip dislocations were the most frequent complications. Only two patients have not had a revision or a major complication. Despite this, 4 children are alive with a functioning lower limb and a mean Musculoskeletal Tumour Society functional score of 77. 6%. The limb length discrepancy was less than 1 0 mm in three of these patients. The remaining patient has a discrepancy of 50 mm and is awaiting further limb equalization procedures. Extendible endoprostheses of the proximal femur in selected children is a viable reconstructive procedure. It allows for equalization of limb length and the ability to walk without the use of mobility aids

The authors describe a new, original technique of intra-medullary nailing (originally designed for the Gamma nail system, now also suitable for other nailing systems) for the management of pertrochanteric and subtrochanteric fractures using a minimally invasive approach to the proximal femur. In this approach, the intramedullary nail is placed using a percutaneous Kirschner wire as a guide, so that the procedure has been called “Percutaneous Nailing System” (PNS). The entry portal is selected at the proximal femur using the Kirschner wire, then a series of cannulae is placed through a small cutaneous incision (15 mm). This dilatator system protects the soft tissue during the reaming procedure (usually only necessary in the proximal femur, not in the diaphysis) and the insertion of the femoral nail. From April 2001 to January 2004, 120 patients were treated with this new technique. They have been followed up and retrospectively compared to 60 patients operated with the standard technique. The comparison between the two groups was based on the surgical procedure (operation time and total blood loss) and the post-operative period (complications, length of hospitalisation). With the minimally invasive technique the operation time was on average 15 min and the blood loss, measured as the difference in pre- and postoperative haemoglobinaemia, was on average 1 point, with no need for blood transfusion: these values were less than half in comparison to the standard technique. The study shows the advantages of this minimally invasive technique, which can also be applied to fractures of the femoral diaphysis

Background: And Aims Pathological fractures of the proximal femur due to primary bone sarcomas are difficult to treat. The aim of the study was to assess the factors determining the outcomes following pathological fractures of the proximal femur due to primary bone sarcomas. Methods: 93 patients with a pathological fracture of the proximal femur due to primary bone sarcomas were studied. The patient, tumour and treatment factors in relation to overall survival were analysed. Results: There were 55 male and 38 female patients. The mean age was 47 years. The diagnoses were Chondrosarcoma -34, Osteosarcoma – 21, spindle cell sarcoma – 25, Ewing’s sarcoma -13. 74 patients had a pathological fracture at diagnosis and 19 patients had a fracture after the diagnosis. 17 patients had metastases at diagnosis. 24 patients had an intracapsular fracture. Limb salvage was possible in 60 patients (65%), 18 patients had an amputation and 15 patients had palliative treatment. 27% of the patients were referred after an unplanned surgery. The mean follow up was 49 months [range 0–302]. Twenty one patients [23%] had a local recurrence -10 patients had a diagnosis of chondrosarcoma, four patients had osteosarcoma and seven had spindle cell sarcoma. The overall five years survival was 37% [Ewing’s sarcoma 60%, Chondrosarcoma 57%, spindle cell sarcoma 28%, osteosarcoma 13% and dedifferentiated Chondrosarcoma 0% (p-0.002)]. Metastasis at diagnosis was a significant factor (p-0.04) affecting survival. Conclusion: We conclude that a pathological fracture of the proximal femur due to osteosarcoma and dedifferentiated chondrosarcoma. carry a poor prognosis

BACKGROUND. In vitro tests have shown that when a force is applied to the proximal femur within the range of directions spanned during physiological activities, the direction of principal strain vary by a very narrow angle (Cristofolini et al, 2009, J. Engng. Med.). This shows that the anatomy and the distribution of inhomogeneous and anisotropic material properties of the bone tissue make the structure of the proximal femur optimized to withstand a wide range of loading directions. The increasing use of hip resurfacing is associated with early neck fractures of the implanted femur. The aim of this study was to elucidate if such fractures could be caused by a non-physiological state of stress/strain post-implantation. While the possible role of notching at the neck-implant interface has already been elucidated, it is not know whether a resurfacing implant could make the principal strain vary in magnitude and direction in a way that could compromise integrity of the proximal femur. METHODS. The aim of this study was to measure if the direction of the principal strain in the proximal femur was affected by the presence of a resurfacing prosthesis. Seven human cadaver femurs were instrumented with 12 triaxial strain gauges to measure the magnitude and alignment of principal strains in the head-neck region. Each femur was implanted with a typical resurfacing prosthesis (BHR). All femurs were tested in vitro before and after implantation with a range of loading conditions to explore the range of loading directions during daily activity (Fig. 1). FINDINGS. Comparison of the strain distribution before and after implantation showed that: . In the natural conditions the principal tensile strain was significantly larger where the cortical bone was thinner; the compressive strain was larger where the cortical bone was thicker. This should be considered when designing a resurfacing prosthesis. The strain magnitude varied greatly between loading configurations both in the intact and implanted condition: this suggests that different loading configurations must be simulated for the preclinical validation of a resurfacing prosthesis. In the natural conditions, the direction of the principal strain varied significantly between measurement locations, but varied little between loading configurations (less than 10° when the hip force spanned a 21° cone, Fig. 2). This confirms that the anatomy and the distribution of anisotropic material properties enable the proximal femur to respond adequately to the changing direction of daily loading. In the resurfaced femurs, when the force spanned the same 21° cone, the direction of principal strain at each measurement location varied by less than 10° (Fig. 3), similar to the natural condition. In the resurfaced femurs, the direction of principal strain lied within less than 10° from the direction in the natural conditions. INTERPRETATION. Our results show that resurfacing does not disturb the alignment of principal strain in the proximal femur. In other words, the most critical directions of stress/strain after implantation stay aligned with the same direction as in the intact femur, which is the direction for which the inhomogeneous and anisotropic structure of the proximal femur is optimized

Aim. To review the results of patients who underwent fixation of complex proximal femur fractures using the Proximal Femur Locking Plates (PFP) and analyse causes of failure of PFP. Methods. Retrospective review of radiographs and case notes of PFP fixations in two hospitals between February 2008 and June 2011. Primary outcome was union at six months. Secondary outcome included post-operative complications, and need for further surgical intervention. Results. There were a total of 32 patients who underwent 34 operations. Two patient had fracture of both the proximal femur requiring bilateral PFP fixation. Mean age of the patients was 68.4 years (range 17–96 years). There were twelve males and twenty female patients. 26 (81%) of the operations were done as primary surgery for fixation of the complex proximal femur fractures. According to the AO/OTA fracture classification, there were four cases of 31-A2.2, seven cases of 31-A2.3, two cases of 31-A3.1 one case each of 31-A3.2 and 32-B1.1 and ten cases of 31 A3.3 fractures. At least six months of follow up was achieved for 30 cases. Union was achieved in 20 fixations (62%) primarily; two more cases needed bone grafting at three months which went on to union improving the total union rate to 70% at 6 months. There was failure of fixation in eight cases requiring further surgery. Varus fixation, loss of posteromedial buttress and loss of protected weight bearing were associated with fixation failure. Conclusion. Contrary to the reported literature, (1) our results are better. Analysis of the failure cases emphasises the importance of postero medial buttress restoration, avoidance of varus fixation, and protection of weight bearing till fracture unites to achieve good outcome