Background: Large Language Models (LLMs) such as ChatGPT, Claude, and Gemini are being increasingly adopted in medicine; however, their reliability in cardiology remains underexplored. Purpose of the study: To compare the performance of three general-purpose LLMs in response to cardiology-related clinical queries. Study design: Seventy clinical prompts stratified by diagnostic phase (pre or post) and user profile (patient or physician) were submitted to ChatGPT, Claude, and Gemini. Three expert cardiologists, who were blinded to the model’s identity, rated each response on scientific accuracy, completeness, clarity, and coherence using a 5-point Likert scale. Statistical analysis included Kruskal–Wallis tests, Dunn’s post hoc comparisons, Kendall’s W, weighted kappa, and sensitivity analyses. Results: ChatGPT outperformed both Claude and Gemini across all criteria (mean scores: 3.7–4.2 vs. 3.4–4.0 and 2.9–3.7, respectively; p < 0.001). The inter-rater agreement was substantial (Kendall’s W: 0.61–0.71). Pre-diagnostic and patient-framed prompts received higher scores than post-diagnostic and physician-framed ones. Results remained robust across sensitivity analyses. Conclusions: Among the evaluated LLMs, ChatGPT demonstrated superior performance in generating clinically relevant cardiology responses. However, none of the models achieved maximal ratings, and the performance varied by context. These findings highlight the need for domain-specific fine-tuning and human oversight to ensure a safe clinical deployment.

Objective: As percutaneous mitral valve techniques become more prevalent, it is important to evaluate the contemporary outcomes of surgical mitral valve interventions. This study assessed the current results and procedural trends of mitral valve surgery for degenerative mitral regurgitation (DMR) at a regional tertiary care center. Methods: Data were analyzed from 693 consecutive DMR patients who underwent isolated mitral valve operations, with or without tricuspid valve repair and atrial fibrillation ablation between 2017 and 2024. The outcomes were defined according to MVARC criteria. The study endpoints included successful mitral valve repair, in-hospital results, and operative and long-term mortality. Logistic regression was applied to assess the impact of valve lesions and patient risk factors on the probability of valve repair. Survival was analyzed using Kaplan–Meier methodology. The follow up was 100% complete. Results: Mitral valve repair was performed in 90.9% of cases, with only 0.9% requiring the conversion to replacement due to unsuccessful repair. Posterior leaflet lesions had the highest success rate (93.4%), while anterior leaflet lesions had a lower rate (86.2%), with anterior pathology being a negative predictor of repair (OR 2.57, p = 0.02). The type of lesion (prolapse vs. flail), the commissural involvement, and the increased risk for SAM had no statistically significant impact on valve repair outcome. Less invasive transaxillary access was used in 63.2% of patients, and its adoption increased significantly (from 50.9% to 67.4% p = 0.03) over time, resulting in more frequent fast-track extubation and home discharges. The rate of in-hospital mortality was 0.6%, while the rate of 5-year survival was 95.5%. Conclusions: Contemporary surgical techniques for DMR lead to high repair rates and excellent recovery outcomes. Despite the rise in transcatheter options, our findings confirm that surgery remains the gold standard for most DMR patients.