Aims. The purpose of this study was to develop a convolutional neural network (CNN) for fracture detection, classification, and identification of greater tuberosity displacement ≥ 1 cm, neck-shaft angle (NSA) ≤ 100°, shaft translation, and articular fracture involvement, on plain radiographs. Methods. The CNN was trained and tested on radiographs sourced from 11 hospitals in Australia and externally validated on radiographs from the Netherlands. Each radiograph was paired with corresponding CT scans to serve as the reference standard based on dual independent evaluation by trained researchers and attending orthopaedic surgeons. Presence of a fracture, classification (non- to minimally displaced; two-part, multipart, and glenohumeral dislocation), and four characteristics were determined on 2D and 3D CT scans and subsequently allocated to each series of radiographs. Fracture characteristics included greater tuberosity displacement ≥ 1 cm, NSA ≤ 100°, shaft translation (0% to < 75%, 75% to 95%, > 95%), and the extent of articular involvement (0% to < 15%, 15% to 35%, or > 35%). Results. For detection and classification, the algorithm was trained on 1,709 radiographs (n = 803), tested on 567 radiographs (n = 244), and subsequently externally validated on 535 radiographs (n = 227). For characterization, healthy shoulders and glenohumeral dislocation were excluded. The overall accuracy for fracture detection was 94% (area under the receiver operating characteristic curve (AUC) = 0.98) and for classification 78% (AUC 0.68 to 0.93). Accuracy to detect greater tuberosity fracture displacement ≥ 1 cm was 35.0% (AUC 0.57). The CNN did not recognize NSAs ≤ 100° (AUC 0.42), nor fractures with ≥ 75% shaft translation (AUC 0.51 to 0.53), or with ≥ 15% articular involvement (AUC 0.48 to 0.49). For all objectives, the model’s performance on the external dataset showed similar accuracy levels. Conclusion. CNNs proficiently rule out proximal humerus fractures on plain radiographs. Despite rigorous training methodology based on CT imaging with multi-rater consensus to serve as the reference standard, artificial intelligence-driven classification is insufficient for clinical implementation. The CNN exhibited poor diagnostic ability to detect greater tuberosity displacement ≥ 1 cm and failed to identify NSAs ≤ 100°, shaft translations, or articular fractures. Cite this article: Bone Joint J 2024;106-B(11):1348–1360

Aims. The aim of this study was to evaluate the clinical and radiological outcomes of locking plate fixation, with and without an associated fibular strut allograft, for the treatment of displaced proximal humeral fractures in elderly osteoporotic patients. Patients and Methods. We undertook a retrospective comparison of two methods of fixation, using a locking plate without an associated fibular strut allograft (LP group) and with a fibular allograft (FA group) for the treatment of these fractures. The outcome was assessed for 52 patients in the LP group and 45 in the FA group, with a mean age of 74.3 years (52 to 89), at a mean follow-up of 14.2 months (12 to 19). The clinical results were evaluated using a visual analogue scale (VAS) score for pain, the Constant score, the American Shoulder and Elbow Surgeons (ASES) score, and the range of movement. Radiological results were evaluated using the neck-shaft angle (NSA) and humeral head height (HHH). Results. The mean forward elevation in the LP and FA groups was 125.3° (. sd. 21.4) and 148.9° (. sd. 19.8), respectively (p = 0.042), while other clinical factors showed no statistically significant differences between the groups. The changes in NSA and HHH immediately after the operation and at final follow-up were significantly better in the FA group than in the LP group (p = 0.015 and p = 0.021, respectively). Conclusion. For comminuted proximal humeral fractures in osteoporotic patients, locking plate fixation with a fibular strut allograft shows satisfactory short-term results with respect to humeral head support and maintenance of reduction, and may reduce the incidence of complications associated with fixation using a locking plate alone. Cite this article: Bone Joint J 2019;101-B:260–265

Aims

The incidence of atypical femoral fractures (AFFs) continues to increase. However, there are currently few long-term studies on the complications of AFFs and factors affecting them. Therefore, we attempted to investigate the outcomes, complications, and risk factors for complication through mid-term follow-up of more than three years.

Aims

The aim of this study was to evaluate the outcomes of a salvage procedure using a 95° angled blade plate for failed osteosynthesis of atypical subtrochanteric femoral fractures associated with the long-term use of bisphosphonates. These were compared with those for failed osteosynthesis of subtrochanteric fractures not associated with bisphosphonate treatment.

Aims

Currently, periprosthetic fractures are excluded from the American Society for Bone and Mineral Research (ASBMR) definition of atypical femoral fracture (AFFs). This study aims to report on a series of periprosthetic femoral fractures (PFFs) that otherwise meet the criteria for AFFs. Secondary aims were to identify predictors of periprosthetic atypical femoral fractures (PAFFs) and quantify the complications of treatment.

The purpose of this study was to describe the anatomical distribution and incidence of fatigue injuries of the femur in physically-active young adults, based upon MRI studies. During a period of 70 months, 1857 patients with exercise-induced pain in the femur underwent MRI of the pelvis, hips, femora, and/or knees.

Of these, 170 patients had a total of 185 fatigue injuries, giving an incidence of 199 per 100 000 person-years. Bilateral injuries occurred in 9% of patients. The three most common sites affected were the femoral neck (50%), the condylar area (24%) and the proximal shaft (18%). A fatigue reaction was seen in 57%, and a fracture line in 22%. There was a statistical correlation between the severity of the fatigue injury and the duration of pain (p = 0.001). The location of the pain was normally at the site of the fatigue injury. Fatigue injuries of the femur appear to be relatively common in physically-active patients.

Limited access surgery is thought to reduce post-operative morbidity and provide faster recovery of function. The percutaneous compression plate (PCCP) is a recently introduced device for the fixation of intertrochanteric fractures with minimal exposure. It has several potential mechanical advantages over the conventional compression hip screw (CHS). Our aim in this prospective, randomised, controlled study was to compare the outcome of patients operated on using these two devices.

We randomised 104 patients with intertrochanteric fractures (AO/OTA 31.A1–A2) to surgical treatment with either the PCCP or CHS and followed them for one year postoperatively.

The mean operating blood loss was 161.0 ml (8 to 450) in the PCCP group and 374.0 ml (11 to 980) in the CHS group (Student’s t-test, p < 0.0001). The pain score and ability to bear weight were significantly better in the PCCP group at six weeks post-operatively. Analysis of the radiographs in a proportion of the patients revealed a reduced amount of medial displacement in the PCCP group (two patients, 4%) compared with the CHS group (10 patients, 18.9%); Fisher’s exact test, p < 0.02.

The PCCP device was associated with reduced intra-operative blood loss, less postoperative pain and a reduced incidence of collapse of the fracture.

We describe a series of 20 patients with ununited fractures of the femoral neck following neglected trauma or failed primary internal fixation who were seen at a mean of 7.5 months (2 to 18) following injury. Open reduction and internal fixation of the fracture was performed in all patients, together with a myoperiosteal flap on the quadratus femoris muscle pedicle.

Union occurred at a mean of 4.9 months (2 to 10) in all patients. The mean follow-up was for 70 months (14 to 144). There was no further progression in six of seven patients with pre-operative radiological evidence of osteonecrosis of the femoral head. One patient had delayed collapse and flattening of the femoral head ten years after union of the fracture, but remained asymptomatic.

This study demonstrates the orthopaedic application of myoperiosteal grafting for inducing osteogenesis in a difficult clinical situation.

Our aim in this pilot study was to evaluate the fixation of, the bone remodelling around, and the clinical outcome after surgery of a new, uncemented, fully hydroxyapatite-coated, collared and tapered femoral component, designed specifically for elderly patients with a fracture of the femoral neck.

We enrolled 50 patients, of at least 70 years of age, with an acute displaced fracture of the femoral neck in this prospective single-series study. They received a total hip replacement using the new component and were followed up regularly for two years.

Fixation was evaluated by radiostereometric analysis and bone remodelling by dual-energy x-ray absorptiometry. Hip function and the health-related quality of life were assessed using the Harris hip score and the EuroQol-5D.

Up to six weeks post-operatively there was a mean subsidence of 0.2 mm (−2.1 to +0.5) and a retroversion of a mean of 1.2° (−8.2° to +1.5°). No component migrated after three months. The patients had a continuous loss of peri-prosthetic bone which amounted to a mean of 16% (−49% to +10%) at two years. The mean Harris hip score was 82 (51 to 100) after two years.

The two-year results from this pilot study indicate that this new, uncemented femoral component can be used for elderly patients with osteoporotic fractures of the femoral neck.

We treated 108 patients with a pertrochanteric femoral fracture using either the dynamic hip screw or the proximal femoral nail in this prospective, randomised series. We compared walking ability before fracture, intra-operative variables and return to their residence. Patients treated with the proximal femoral nail (n = 42) had regained their pre-operative walking ability significantly (p = 0.04) more often by the four-month review than those treated with the dynamic hip screw (n = 41). Peri-operative or immediate post-operative measures of outcome did not differ between the groups, with the exception of operation time. The dynamic hip screw allowed a significantly greater compression of the fracture during the four-month follow-up, but consolidation of the fracture was comparable between the two groups. Two major losses of reduction were observed in each group, resulting in a total of four revision operations.

Our results suggest that the use of the proximal femoral nail may allow a faster postoperative restoration of walking ability, when compared with the dynamic hip screw.