Background and purpose — Recent advances in artificial intelligence (deep learning) have shown remarkable performance in classifying non-medical images, and the technology is believed to be the next technological revolution. So far it has never been applied in an orthopedic setting, and in this study we sought to determine the feasibility of using deep learning for skeletal radiographs.Methods — We extracted 256,000 wrist, hand, and ankle radiographs from Danderyd’s Hospital and identified 4 classes: fracture, laterality, body part, and exam view. We then selected 5 openly available deep learning networks that were adapted for these images. The most accurate network was benchmarked against a gold standard for fractures. We furthermore compared the network’s performance with 2 senior orthopedic surgeons who reviewed images at the same resolution as the network.Results — All networks exhibited an accuracy of at least 90% when identifying laterality, body part, and exam view. The final accuracy for fractures was estimated at 83% for the best performing network. The network performed similarly to senior orthopedic surgeons when presented with images at the same resolution as the network. The 2 reviewer Cohen’s kappa under these conditions was 0.76.Interpretation — This study supports the use for orthopedic radiographs of artificial intelligence, which can perform at a human level. While current implementation lacks important features that surgeons require, e.g. risk of dislocation, classifications, measurements, and combining multiple exam views, these problems have technical solutions that are waiting to be implemented for orthopedics.

Background: The usage of volar locking plate fixation for distal radial fractures has increased in older patient populations, despite the fact that surgical treatment in the elderly population has not clearly been proven to be superior to nonoperative treatment. The purpose of the present study was to compare nonoperative treatment with volar locking plate fixation with regard to clinical outcome for elderly patients with dorsally displaced distal radial fractures. Methods: In this study, 140 patients were randomly allocated to nonoperative treatment with a plaster splint (n = 72) or volar locking plate fixation (n = 68). The outcome variables were the Patient-Rated Wrist Evaluation (PRWE) score, Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire score, EuroQol-5 Dimensions (EQ-5D) score, range of motion, grip strength, radiographic outcomes, and complications. Evaluation was performed at 3 and 12 months by unblinded observers. Results: At 3 months, 122 patients were evaluated, and at 12 months, 119 patients were evaluated. At 3 months, the volar locking plate group, compared with the nonoperative treatment group, had a better median PRWE score (10.3 compared with 35.5 points; p = 0.002), DASH score (14.4 compared with 29.2 points; p = 0.016), and grip strength (71.0% of the uninjured hand compared with 53.9%; p < 0.001). Significant differences in favor of the volar locking plate group remained at 12 months; compared with the nonoperative treatment group, the volar locking plate group had a better median PRWE score (7.5 points compared with 17.5 points; p = 0.014), DASH score (8.3 points compared with 19.9 points; p = 0.028), and grip strength (96.8% compared with 80.0%; p = 0.001). Radiographic measurements favored volar locking plate fixation at 3 and 12 months. Complication rates were similar, with 11% major complications in the nonoperative group compared with 14% major complications in the volar locking plate group (p = 0.606) and 11% minor complications in the nonoperative group compared with 20% minor complications in the volar locking plate group (p = 0.197). Conclusions: The PRWE scores, DASH scores, and grip strength were better for the volar locking plate group compared with the nonoperative group at 3 and 12 months. The complication rates were similar. Our results imply that there is a benefit for the elderly patient with an unstable dorsally displaced distal radial fracture to be treated with a volar locking plate. Level of Evidence: Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.

Background Periprosthetic bone loss is a well-documented phenomenon after uncemented total hip arthroplasty (THA); however, little is known about how bone mineral density (BMD) changes after 2 years.Patients and methods 14 patients with hip arthrosis (group A) were operated with a proximally porous-and hydroxyapatite-coated stem and followed for 10 years with DEXA, radiographs and Harris hip score (HHS). Another group of 14 patients (group B) was evaluated at 6 and 14 years using the same prosthesis and protocol.Results No stem was revised and all stems were wellfixed. At final follow-up, HHS was 97 points in group A after 10 years and 94 points in group B after 14 years. Bone mineral changes in group A were greatest in Gruen zones 1 and 7, where the losses were 31% and 26%, respectively, after 2 years on the operated side. The decrease in BMD continued after 2 years and in Gruen zone 7 it was faster than the rate of bone loss on the control side. In group B, the annual change in BMD on the operated side was not significantly different from the bone loss in group A.Interpretation Up to 14 years after implantation of a tapered uncemented stem, the BMD in the calcar region continues to decrease faster than would be expected from normal ageing.

Therapeutic Level I.