Background: Large language models (LLMs) have raised both interest and concern in the academic community. They offer the potential for automating literature search and synthesis for systematic reviews but raise concerns regarding their reliability, as the tendency to generate unsupported (hallucinated) content persist. Objective: The aim of the study is to assess the performance of LLMs such as ChatGPT and Bard (subsequently rebranded Gemini) to produce references in the context of scientific writing. Methods: The performance of ChatGPT and Bard in replicating the results of human-conducted systematic reviews was assessed. Using systematic reviews pertaining to shoulder rotator cuff pathology, these LLMs were tested by providing the same inclusion criteria and comparing the results with original systematic review references, serving as gold standards. The study used 3 key performance metrics: recall, precision, and F1-score, alongside the hallucination rate. Papers were considered “hallucinated” if any 2 of the following information were wrong: title, first author, or year of publication. Results: In total, 11 systematic reviews across 4 fields yielded 33 prompts to LLMs (3 LLMs×11 reviews), with 471 references analyzed. Precision rates for GPT-3.5, GPT-4, and Bard were 9.4% (13/139), 13.4% (16/119), and 0% (0/104) respectively (P<.001). Recall rates were 11.9% (13/109) for GPT-3.5 and 13.7% (15/109) for GPT-4, with Bard failing to retrieve any relevant papers (P<.001). Hallucination rates stood at 39.6% (55/139) for GPT-3.5, 28.6% (34/119) for GPT-4, and 91.4% (95/104) for Bard (P<.001). Further analysis of nonhallucinated papers retrieved by GPT models revealed significant differences in identifying various criteria, such as randomized studies, participant criteria, and intervention criteria. The study also noted the geographical and open-access biases in the papers retrieved by the LLMs. Conclusions: Given their current performance, it is not recommended for LLMs to be deployed as the primary or exclusive tool for conducting systematic reviews. Any references generated by such models warrant thorough validation by researchers. The high occurrence of hallucinations in LLMs highlights the necessity for refining their training and functionality before confidently using them for rigorous academic purposes.

Purpose: Immediate revision refers to a reoperation that involves resetting, draping, and exchanging the implant, after wound closure in total hip arthroplasty. The purpose of this study is to investigate the impact of immediate revision after total hip arthroplasty on subsequent infection and complication rates. Methods: A total of 14,076 primary total hip arthroplasties performed between 2010 and 2020 were identified in our institutional database, of which 42 underwent immediate revision. Infection rates were determined 2 years after the index arthroplasty. The cause and type of revision, duration of primary and revision surgeries, National Nosocomial Infections Surveillance score, implant type, changes in implants, complications, and preoperative and intraoperative antibiotic prophylaxis were all determined. Results: No infections were observed within 2 years after the index arthroplasty. Leg length discrepancy (88%, n = 37) and dislocation (7.1%, n = 3) were the main causes of immediate revision. In most cases of discrepancy, the limb was clinically and radiologically longer before the immediate revision. The mean operative time was 48 ± 14 min for the primary procedure and 23.6 ± 9 min for the revision. The time between the first incision and last skin closure ranged from 1 to 3 h. None of the patients were extubated between the two procedures. Two patients had a National Nosocomial Infections Surveillance score of 2, 13 had a score of 1, and 27 had a score of 0. Conclusion: Immediate revision is safe for correcting clinical and radiological abnormalities, and may not be associated with increased complication or infection rates. Study design: Retrospective cohort study; level of evidence, 3.

Aims: After a total knee arthroplasty (TKA), up to 40% of patients suffer from chronic pain. Few studies have focused specifically on psychosocial factors as predictors of chronic pain after TKA. Our study aims to identify psychosocial risk factors of chronic pain after TKA and to identify an optimal threshold for significant scores in the multivariate analysis, using a receiver operating characteristic (ROC) curve.Methods: This was a single-centre prospective study conducted from September 2022 to September 2023, involving 204 patients. The mean age was 70 years (SD 8.59), the mean BMI was 30.49 kg/m² (SD 5.29), and 70.60% of patients were female (144/204). A form was administered before surgery to collect: Hospital Anxiety and Depression Scale (HADS), anxiety scale State-Trait Anxiety Inventory (STAI)-Y, Amsterdam Preoperative Anxiety and Information Scale (APAIS), Pain Catastrophizing Scale (PCS), and visual analogue scale (VAS). Univariate and multivariate logistic regression was used to determine the risk factors associated with chronic pain. To determine a threshold in the psychological scores associated with chronic pain, ROC curves were performed.Results: A total of 36 patients (17.60%) met the definition of chronic pain, defined as VAS ≥ 4/10 at six months postoperatively. In the multivariate analysis, to be an employee or worker (β = 1.10, 95% CI 0.20 to 2.10, p = 0.018), anxiety STAI-YB (β = 0.03, 95% CI 0.01 to 0.06, p = 0.013), and catastrophism PCS (β = 0.04, 95% CI 0.02 to 0.06, p < 0.001) were associated with chronic pain after TKA. Using ROC analysis, we identified a threshold value of 44.5 for anxiety STAI-YB (sensitivity 57.58%, specificity 74.51%) and of 21.5 for catastrophism PCS (sensitivity 56,25 %, specificity 71.90%) of developing chronic pain after TKA.Conclusion: This study shows that the psychosocial profile of patients can be linked to the occurrence of chronic pain, and underlines the importance of a multidimensional approach to patient assessment and management.Cite this article: Bone Jt Open 2025;6(7):755–763.

Irreparable large to massive rotator cuff tears (MIRCTs) are a prevalent cause of shoulder pain and dysfunction, and nonoperative treatment may not always be effective. Various surgical options exist, with isolated biceps tenotomy/tenodesis (BT) or arthroscopic partial repair with associated biceps tenotomy/tenodesis (PCR-BT) being the most common. The aim of this study was to systematically review the available data on the clinical and functional outcomes of BT and PCR-BT in patients with MIRCTs. Methods: MEDLINE, Embase, and CENTRAL databases were searched for studies on the treatment of MIRCT. We included studies with BT or PCR-BT with a minimum follow-up of 24 months. The MINORS (Methodological Index for Nonrandomized Studies) score was used to assess study quality. Outcomes included were the visual analog scale for pain, functional scores such as Constant-Murley and American Shoulder and Elbow Surgeons, range of motion, radiological measurements, and complications. Results: A total of 1101 patients (506 had a BT and 595 had a PCR-BT) from 22 studies were included (cases series = 13, case–control = 7, randomized control trial = 1, prospective cohort study = 1). The mean MINORS score was 13.2 ± 3.2. The mean age and follow-up were 67 ± 6.8 years and 4.58 ± 1.1 years (range, 2, 12), respectively. The VAS improvement showed at the last follow-up for PCR-BT (range, 1.97, 5.8) and BT (range, 4, 6.1). CMS was improved at the final follow-up for PCR-BT (range, 13, 47.6) and BT (range, 10.8, 28). Regarding the ASES, it has demonstrated significant improvements for PCR-BT (range, 31.81, 44.8) and BT (range, 30,45.8). For forward flexion, PCR-BT showed improvement (range, −14°, 59.4°), as well as the BT group (range, 2°, 27.9°). Conclusions: This systematic review demonstrated that both BT and PCR-BT improve functional outcomes and reduce pain at midterm follow-up for MIRCT. Since we know that a failed cuff repair would worsen the shoulder, it might be beneficial in terms of the risk–benefit ratio to not repair in certain patients with MIRCT.