Search

Large language models (LLMs) have shown potential in various applications, including clinical practice. However, their accuracy and utility in providing treatment recommendations for orthopedic conditions remain to be investigated. Thus, this pilot study aims to evaluate the validity of treatment recommendations generated by GPT-4 for common knee and shoulder orthopedic conditions using anonymized clinical MRI reports. A retrospective analysis was conducted using 20 anonymized clinical MRI reports, with varying severity and complexity. Treatment recommendations were elicited from GPT-4 and evaluated by two board-certified specialty-trained senior orthopedic surgeons. Their evaluation focused on semiquantitative gradings of accuracy and clinical utility and potential limitations of the LLM-generated recommendations. GPT-4 provided treatment recommendations for 20 patients (mean age, 50 years ± 19 [standard deviation]; 12 men) with acute and chronic knee and shoulder conditions. The LLM produced largely accurate and clinically useful recommendations. However, limited awareness of a patient's overall situation, a tendency to incorrectly appreciate treatment urgency, and largely schematic and unspecific treatment recommendations were observed and may reduce its clinical usefulness. In conclusion, LLM-based treatment recommendations are largely adequate and not prone to 'hallucinations', yet inadequate in particular situations. Critical guidance by healthcare professionals is obligatory, and independent use by patients is discouraged, given the dependency on precise data input., (© 2023. The Author(s).)

Autologous matrix induced chondrogenesis (AMIC) is a bone marrow stimulating technique used for the surgical management of chondral defects of the talus. The present study evaluated the clinical outcomes and imaging of AMIC as revision procedure for failed AMIC surgery for osteochondral defects of the talus. Forty-eight patients with symptomatic osteochondral defects who received a revision AMIC were evaluated after a minimum of two years follow-up. Patients with previous procedures rather than AMIC, those who required additional surgical procedures (e.g. ligament repair or deformity correction), or those who had evidence of kissing, bilateral, or multiple lesions were excluded. Outcome parameters included the Visual Analogic Scale (VAS), Tegner Activity Scale, the American Orthopedic Foot and Ankle Score (AOFAS), and the Magnetic Resonance Observation of Cartilage Repair Tissue (MOCART) score. All patients were followed by an assessor who was not involved in the clinical management. 27 patients were enrolled in the present study. The mean age of the patient was 34.9 ± 3.1 years, and the mean BMI 27.2 ± 5.1 kg/m 2 . The mean defect surface area was 2.8 ± 1.9 cm 2 . The mean follow-up was 44.3 ± 21.4 months. The mean hospital length of stay was 4.4 ± 1.4 days. At final follow-up, the mean VAS score was 4.1 ± 3.1, the mean Tegner 3.5 ± 1.6, the mean AOFAS 58.8 ± 20.6. The preoperative MOCART score was 22.1 ± 13.7 points, the postoperative MOCART score was 42.3 ± 27.9 points (+ 20.2%; P = 0.04), respectively. 30% (8 of 27 patients) experienced persistent pain and underwent a further chondral procedure. Concluding, AMIC could be a viable option as revision procedure for failed AMIC in recurrent symptomatic osteochondral defects of the talus. The PROMs indicated that patients were moderately satisfied with the procedure, and the MOCART score demonstrated a significant improvement from baseline to the last follow-up. A deeper understanding in prognostic factors and patient selection is critical to prevent failures., (© 2022. The Author(s).)