Background: Large language models (LLMs) can understand natural language and generate corresponding text, images, and even videos based on prompts, which holds great potential in medical scenarios. Orthopedics is a significant branch of medicine, and orthopedic diseases contribute to a significant socioeconomic burden, which could be alleviated by the application of LLMs. Several pioneers in orthopedics have conducted research on LLMs across various subspecialties to explore their performance in addressing different issues. However, there are currently few reviews and summaries of these studies, and a systematic summary of existing research is absent. Objective: The objective of this review was to comprehensively summarize research findings on the application of LLMs in the field of orthopedics and explore the potential opportunities and challenges. Methods: PubMed, Embase, and Cochrane Library databases were searched from January 1, 2014, to February 22, 2024, with the language limited to English. The terms, which included variants of “large language model,” “generative artificial intelligence,” “ChatGPT,” and “orthopaedics,” were divided into 2 categories: large language model and orthopedics. After completing the search, the study selection process was conducted according to the inclusion and exclusion criteria. The quality of the included studies was assessed using the revised Cochrane risk-of-bias tool for randomized trials and CONSORT-AI (Consolidated Standards of Reporting Trials–Artificial Intelligence) guidance. Data extraction and synthesis were conducted after the quality assessment. Results: A total of 68 studies were selected. The application of LLMs in orthopedics involved the fields of clinical practice, education, research, and management. Of these 68 studies, 47 (69%) focused on clinical practice, 12 (18%) addressed orthopedic education, 8 (12%) were related to scientific research, and 1 (1%) pertained to the field of management. Of the 68 studies, only 8 (12%) recruited patients, and only 1 (1%) was a high-quality randomized controlled trial. ChatGPT was the most commonly mentioned LLM tool. There was considerable heterogeneity in the definition, measurement, and evaluation of the LLMs’ performance across the different studies. For diagnostic tasks alone, the accuracy ranged from 55% to 93%. When performing disease classification tasks, ChatGPT with GPT-4’s accuracy ranged from 2% to 100%. With regard to answering questions in orthopedic examinations, the scores ranged from 45% to 73.6% due to differences in models and test selections. Conclusions: LLMs cannot replace orthopedic professionals in the short term. However, using LLMs as copilots could be a potential approach to effectively enhance work efficiency at present. More high-quality clinical trials are needed in the future, aiming to identify optimal applications of LLMs and advance orthopedics toward higher efficiency and precision.

Introduction Adolescent idiopathic scoliosis (AIS) is a three-dimensional deformity of the spine. Brace treatment is effective for eligible patients with AIS and the effectiveness is significantly correlated with the average brace-wear time per day. Three-dimensional (3D) printing technology is a recent advancement that offers unique opportunities for biomedical applications, and customisation of scoliosis braces might lead to greater patient satisfaction and improved compliance. We present here the design of a randomised controlled trial on the clinical effectiveness of 3D-printed braces versus thoracolumbosacral orthoses (TLSO) for patients with AIS. Methods and analysis Patients with AIS (age 10–16 years) with Risser sign 0-II, Cobb angle of main curve of 20°−40°, premenarchal or no more than 1-year postmenarchal (for women), and no history of treatment are eligible, unless they are unable to tolerate the treatment or refuse participation. A total of 88 patients will be randomised into either the 3D group or TLSO group on a 1:1 basis. Participants in the 3D group will choose between a 3D-printed brace and TLSO, according to the Zelen’s design of the trial. Primary outcome measures will include the average brace-wear time per day, health-related quality of life and Cobb angle progression of the primary curve. Secondary outcome measures will include immediate in-brace correction of Cobb angle of the primary curve, rate of conversion to surgery and incidence of any adverse events. This study is designed as a single-centre, two-arm, superiority and open-label randomised controlled trial. The sample size is calculated with reference to the preliminary study and based on the sample size calculation formula. Ethics and dissemination This study was approved by the Peking University Third Hospital Medicine Science Research Ethics Committee (No: 2019-017-02). Results of the trial will be submitted for publication in a peer-reviewed journal and as conference presentations. Trial registration number ChiCTR1900027379, pre-results.

Background: Patients with “sandwich” fusion (concomitant C1 occipitalization and C2-C3 nonsegmentation), a subtype of Klippel-Feil syndrome, are at particular risk for developing atlantoaxial dislocation (AAD). However, the clinical and surgical characteristics of AAD in patients with sandwich fusion have not been clearly defined. Methods: A retrospective case-control study with a large sample size and a minimum 2-year follow-up was performed. From 2000 to 2018, 253 patients with sandwich AAD underwent a surgical procedure; these patients constituted the case group, and a matching number of patients with non-sandwich AAD were randomly selected to form the control group. Clinical data from electronic medical records and various imaging studies were analyzed and compared. The Japanese Orthopaedic Association (JOA) scale was used to evaluate neurological function. Results: Patients with sandwich AAD, compared with patients with non-sandwich AAD, had symptom onset at a younger age (34.8 compared with 42.8 years; p < 0.001) and had a higher likelihood for myelopathy (87.4% compared with 74.7%; p < 0.001). Patients with sandwich AAD had a higher incidence of lower cranial nerve palsy (7.9% compared with 0.0%; p < 0.001), a lower preoperative JOA score (13.4 compared with 14.2; p < 0.001), and higher incidences of accompanying Type-I Chiari malformation (20.9% compared with 1.2%; p < 0.001) and syringomyelia (21.3% compared with 1.6%; p < 0.001). Finally, patients with sandwich AAD had higher likelihoods of undergoing transoral release (28.5% compared with 5.1%; p < 0.001) and use of salvage fixation techniques (34.4% compared with 6.3%; p < 0.001), and had lower postoperative results for the JOA score (14.9 compared with 15.9; p < 0.001) and improvement rate (43.8% compared with 58.2%; p < 0.001). Conclusions: Patients with sandwich AAD demonstrated distinct clinical manifestations. Versatility involving the use of various internal fixation techniques and transoral release procedures was frequently required in the surgical management of these patients, and meticulous and personalized preoperative planning would be of paramount importance. Level of Evidence: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

The significant efforts being made towards the utilization of artificial soft materials holds considerable promise for developing tissue engineering scaffolds for bone-related diseases in clinics. However, most of these biomaterials cannot simultaneously satisfy the multiple requirements of high mechanics, good compatibility, and biological osteogenesis. In this study, an osteogenic hybrid hydrogel between the amine-functionalized bioactive glass (ABG) and 4-armed poly(ethylene glycol) succinimidyl glutarate-gelatin network (SGgel) is introduced to flexibly adhere onto the defective tissue and to subsequently guide bone regeneration. Relying on the rapid ammonolysis reaction between amine groups (-NH₂) of gelatin and ABG components and N-hydroxysuccinimide (NHS)-ester of tetra-PEG-SG polymer, the hydrogel networks were formed within seconds, offering a multifunctional performance, including easy injection, favorable biocompatibility, biological and mechanical properties (compressive strength: 4.2 MPa; storage modulus: 10⁴ kPa; adhesive strength: 56 kPa), which could facilitate the stem cell viability, proliferation, migration and differentiation into osteocytes. In addition, the integration between the SGgel network and ABG moieties within a nano-scale level enabled the hybrid hydrogel to form adhesion to tissue, maintain the durable osteogenesis and accelerate bone regeneration. Therefore, a robust approach to the simultaneously satisfying tough adhesion onto the tissue defects and high efficiency for bone regeneration on a mouse skull was achieved, which may represent a promising strategy to design therapeutic scaffolds for tissue engineering in clinical applications.