Objective: This study aimed to comparatively evaluate the clinical knowledge generation performance of 3 widely used large language model (LLM)-based chatbots (ChatGPT, Claude, and Perplexity) in the context of septic arthritis. Methods: This cross-sectional comparative study was based on 24 scenario-based clinical questions developed in accordance with the SANJO guideline (Management of Septic Arthritis in Native Joints) of the European Bone and Joint Infection Society. Responses generated by ChatGPT (OpenAI GPT-4), Claude 2 (Anthropic), and Perplexity AI were independently assessed by 2 senior experts: 1 in orthopedic surgery and the other in infectious diseases. Each response was evaluated across 6 domains: scientific accuracy, content depth, termino logical consistency, clinical applicability, brevity, and reference support, using a 5-point Likert scale. Results: All 3 LLM-based chatbots achieved perfect scores in accuracy and terminological consistency (P = 1.000), and no significant dif ference was observed in clinical applicability (P = .912). Perplexity scored significantly lower in content depth compared to both ChatGPT (P = .001) and Claude (P = .041), whereas ChatGPT and Claude did not differ significantly (P = .807). ChatGPT produced significantly more unnecessary elaboration than Claude (P = .009) and Perplexity (P < .001), while Claude and Perplexity were comparable (P = .115). For reference support, Perplexity scored significantly higher than both ChatGPT (P < .001) and Claude (P < .001), with no difference between the latter 2 (P = 1.000). Overall, Perplexity achieved the highest total score (P < .001), followed by ChatGPT and Claude. Interrater agree ment was substantial (κ = 0.72). Conclusion: The LLM-based chat platforms demonstrated overall high performance, but their strengths differed across evaluation domains. While ChatGPT and Claude provided more comprehensive and detailed responses, Perplexity offered stronger reference sup port. These findings suggest that context-specific selection of LLMs is essential, as the optimal choice may vary depending on whether detailed explanation or robust referencing is prioritized. Cite this article as: Bayrak HC, Karagöz B, Bayrak Ö. Comparative evaluation of large language model–based chatbots in a septic arthritis scenario: ChatGPT, Claude, and Perplexity. Acta Orthop Traumatol Turc., 2025;59(6):415-420.

Background: This study aims to compare the quality, reliability, and readability of information provided by artificial intelligence-based language models, ChatGPT-5 and DeepSeek V3, regarding foot and ankle disorders. Methods: The quality, reliability, and readability of the texts generated by both AI models were analyzed using DISCERN, the Patient Education Materials Assessment Tool for Printable Materials (PEMAT-P), the Global Quality Score (GQS), and the CLEAR scoring system. DISCERN was used to assess information reliability, PEMAT-P to evaluate understandability and actionability, GQS to assess overall quality, and CLEAR to evaluate content quality and accuracy. Standardized questions were asked to both models for 35 different foot and ankle disorders, and the generated texts were evaluated by two independent orthopedic specialists using a blinded method. Readability analysis was performed using word count, the Flesch–Kincaid Grade Level (FKGL; required reading level), and the Flesch Reading Ease (FRE; ease of readability) scoring systems. Results: ChatGPT-5 scored significantly higher than DeepSeek V3 in DISCERN, PEMAT-P, GQS, and CLEAR evaluations (p < 0.05), indicating that ChatGPT-5 provides more reliable, comprehensive, and higher-quality information. DeepSeek V3 demonstrated better readability, producing simpler and more understandable content, as reflected in its lower FKGL score and higher FRE score. Conclusions: While ChatGPT-5 delivers more detailed and reliable health information, DeepSeek V3 offers simpler and more readable texts. Both models have distinct advantages for patient education. Future research should assess the impact of AI-generated health information on patient decision-making and its clinical application potential.

This study aimed to analyze the relationship between fracture type by determining data on the geometry of the proximal femur in the pre-fracture period in patients over 65 years of age who had hip fractures as a result of low-energy trauma. A total of 127 patients who were admitted to the hospital for reasons other than hip pathology within 1 year before the occurrence of hip fracture and who had an anterior-posterior pelvic X-ray were included in the study. Measurements were made to evaluate the proximal femur geometry, neck shaft angle, central edge angle, femoral head diameter, femoral neck diameter, femoral neck length, femoral offset length, femoral neck axial length, hip axis length, and femoral shaft diameter. As a result of these measurements, analyses were performed to determine the relationship between the control group and fracture types. The mean Neck shaft angle scores were significantly higher in both fracture types than in the control group (P = .034, P = .002). The mean Femoral offset length values of both fracture types were lower than those of the control group (P = .002, P = .011, respectively). Multiple logistic regression analysis revealed that the risk of collum femoris fracture increased as the Femoral head diameter value increased. (OD = 0.21, P = .002). The geometric parameters of the proximal femur play an important role in the formation of hip fracture types. Therefore, differences in proximal femur geometry in hip fracture types should be considered, and patient-focused choices should be made regarding the surgical procedures and implants to be used during fracture fixation.

Introduction: in this study, we aimed to compare the clinical and radiological results of single 4.5 mm locking compression plate (LCP), dual 3.5 mm LCP and intramedullary nailing (IMN) methods applied to the surgical treatment of humeral shaft fractures. Methods: the study included 77 patients (41 males, 36 females; mean age 46.38 years; range 18-74 years) with humeral shaft fractures treated with a single 4.5 mm LCP, dual 3.5 mm LCP and IMN between January 2016 and December 2020. Single 4.5 mm LCP (Group A) was applied to 31 (40.3% ) patients, dual 3.5 mm LCP (Group B) to 20 (26%) patients and IMN (Group C) to 26 (33.8%) patients. The preoperative and postoperative data of the patients were analyzed from the hospital registry system. A short version of the disabilities of the arm, shoulder and hand (QuickDASH) questionnaire was used to evaluate functional outcomes. Results: as a result of the comparison of the rates of nonunion between the groups, a significantly lower rate of nonunion was observed in group B patients (p=0.027). While the rate of nonunion was 14% in the cases included in the study, no cases of nonunion were encountered in group B. There was no difference between the three groups in terms of demographic data and other postoperative complications. Conclusion: dual 3.5 mm LCP method is a suitable alternative to other surgical methods in the treatment of humeral shaft fractures, due to similar functional results and lesser nonunion.

OBJECTIVE: The aim of this study was to compare the functional and radiological results of lateral, medial, and posterior surgical approaches in pediatric patients undergoing open reduction and internal fixation for supracondylar humerus fractures. METHODS: A total of 86 patients were included in the study. The clinical and radiographic results of the treatment in patients who underwent open reduction and internal fixation with lateral, medial, and posterior approaches were evaluated. Flynn's criteria were used in the evaluation of cosmetic and clinical results. Comparisons were made between the groups in terms of Baumann angle, lateral capitellohumeral angle, and post-operative complications. RESULTS: There was no statistically significant difference between the three groups in terms of complications. No statisti-cally significant relationship was observed between Flynn's criteria and surgical approaches. When the relationship among post-operative range of motion (ROM) and surgical approach was evaluated, no extension defect was found in any of the patients, but a significant relationship was found between post-operative flexion ROM and surgical approach (p=0.011). CONCLUSION: Closed reduction and percutaneous pinning are preferred in cases of pediatric supracondylar humerus fractures. However, when this method cannot be applied, lateral, medial, and posterior approaches are the possible open reduction methods, that can be safely preferred.

Background: The necessity of distal locking in proximal femoral nails (PFNs) used in intertrochanteric fracture surgery is still controversial. However, the newly designed talon distalfix PFN differs from other PFN systems with its taloned distal locking (TDL) mechanism. Aim: To compare the radiologic and functional results of this newly designed TDL system and the static distal locking (SDL) system. Methods: Between January 2018 and December 2021, 142 patients over the age of 65 who underwent PFN surgery for intertrochanteric fractures in our department and underwent SDL or TDL as the distal locking mechanism were analyzed. The patients were divided into two groups as SDL and TDL according to the type of distal locking applied. Demographic variables and intraoperative data (operative time, intraoperative bleeding amount, and fluoroscopy time) were recorded. Harris hip score and Visual Analog Scale (VAS) score results were recorded to evaluate postoperative functional outcomes. Union time and all postoperative complications were recorded. Results: Of a total of 142 patients with a mean age of 77.51 ± 11.67 years, 79 underwent SDL and 63 underwent TDL. Operative time, intraoperative bleeding amount, and fluoroscopy time were significantly reduced in patients who underwent TDL compared to those who underwent SDL. The union times were found to be significantly shorter in TDL patients. There was no significant difference between the groups in terms of postoperative Harris hip score and VAS score results and complications. Conclusions: This clinical study confirms that TDL is as effective and safe as SDL. In addition, TDL has been shown to have advantages over SDL with shorter operative time, less bleeding amount, and less fluoroscopy time.