{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T17:12:58Z","timestamp":1788541978666,"version":"build-2803163510"},"reference-count":20,"publisher":"MDPI AG","issue":"17","license":[{"start":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T00:00:00Z","timestamp":1788480000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background: Perioperative antithrombotic management is a complex and high-risk aspect of anesthesia practice, requiring a careful balance between bleeding and thromboembolic risks. Large language models (LLMs) are increasingly explored for clinical decision support, but their reliability in complex perioperative antithrombotic scenarios remains uncertain. Methods: In this blinded, scenario-based study, 100 elective surgical cases, including grey-zone scenarios, were developed to reflect clinically relevant perioperative antithrombotic decisions. The European Society of Anaesthesiology and Intensive Care\/European Society of Regional Anesthesia and Pain Therapy (ESAIC\/ESRA) 2022 guideline was incorporated into the standardized prompt as a common framework for neuraxial safety and relevant antithrombotic interruption intervals. Four LLMs (GPT-5.4 Thinking, Claude Opus 4.5, DeepSeek v3.2 Reasoning and Qwen3.5-Plus) were evaluated using a standardized prompt. Model responses were presented without model identity and independently assessed by two experienced anesthesiologists, neither of whom was an author of the present study, across five clinical domains on a 5-point Likert scale, with a separate clinical applicability assessment. Results: Using a uniform four-domain composite across all 100 scenarios, overall expert-rated performance differed across models (Friedman \u03c72 = 28.75, p &lt; 0.001; Kendall\u2019s W = 0.096). Claude had the highest median primary composite score, although its difference from DeepSeek was not statistically significant. Leave-one-domain-out sensitivity analyses showed that between-model separation was particularly sensitive to the clinical-rationale domain; exclusion of this domain reduced Kendall\u2019s W to 0.030. A secondary five-domain analysis restricted to the 51 common-proceed scenarios, all of which were standard rather than grey-zone cases, also showed an overall between-model difference (Friedman \u03c72 = 55.13, p &lt; 0.001; Kendall\u2019s W = 0.360). Because each scenario was queried only once, fine-grained differences in model ordering should be considered provisional. Although all models performed well in drug discontinuation decisions, greater variability was observed in discontinuation timing, anesthesia choice and clinical reasoning. Postponement behaviour varied substantially across models, with DeepSeek and Qwen recommending postponement more often than ChatGPT and Claude. Conclusions: LLMs show promise as supportive tools in perioperative decision-making, but their performance remains variable, especially in complex situations. At present, they should be used cautiously and always under expert supervision, rather than as independent decision-makers.<\/jats:p>","DOI":"10.3390\/diagnostics16172848","type":"journal-article","created":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T16:07:32Z","timestamp":1788538052000},"page":"2848","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Large Language Models in Perioperative Antithrombotic Management: A Blinded Scenario-Based Expert Evaluation"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1627-1986","authenticated-orcid":false,"given":"\u0130rem","family":"Durmu\u015f","sequence":"first","affiliation":[{"name":"Department of Anesthesiology and Reanimation, Kartal Dr. L\u00fctfi K\u0131rdar City Hospital, University of Health Sciences Turkey, 34865 Istanbul, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1809-0791","authenticated-orcid":false,"given":"Merve","family":"Bulun Yediy\u0131ld\u0131z","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology and Reanimation, Kartal Dr. L\u00fctfi K\u0131rdar City Hospital, University of Health Sciences Turkey, 34865 Istanbul, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,9,4]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Sciahbasi, A., Minardi, S., Salvi, N., Infusino, F., and Granatelli, A. (2025). Antithrombotic therapy in the elderly with cardiovascular disease: Walking the tightrope between efficacy and bleeding risk\u2014A narrative review. J. Clin. Med., 14.","DOI":"10.3390\/jcm14207340"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"e207","DOI":"10.1016\/j.chest.2022.07.025","article-title":"Perioperative management of antithrombotic therapy: An American College of Chest Physicians clinical practice guideline","volume":"162","author":"Douketis","year":"2022","journal-title":"Chest"},{"key":"ref_3","first-page":"211","article-title":"Neuraxial and perineural bleeding after neuraxial techniques: An overview of the last year","volume":"52","author":"Correale","year":"2020","journal-title":"Eurasian J. Med."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"263","DOI":"10.1097\/AAP.0000000000000763","article-title":"Regional anesthesia in the patient receiving antithrombotic or thrombolytic therapy: American Society of Regional Anesthesia and Pain Medicine evidence-based guidelines (Fourth Edition)","volume":"43","author":"Horlocker","year":"2018","journal-title":"Reg. Anesth. Pain Med."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"172","DOI":"10.1038\/s41586-023-06291-2","article-title":"Large language models encode clinical knowledge","volume":"620","author":"Singhal","year":"2023","journal-title":"Nature"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Kung, T.H., Cheatham, M., Medenilla, A., Sillos, C., De Leon, L., Elepa\u00f1o, C., Madriaga, M., Aggabao, R., Diaz-Candido, G., and Maningo, J. (2023). Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models. PLoS Digit. Health, 2.","DOI":"10.1371\/journal.pdig.0000198"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"100","DOI":"10.1097\/EJA.0000000000001600","article-title":"Regional anaesthesia in patients on antithrombotic drugs: Joint ESAIC\/ESRA guidelines","volume":"39","author":"Kietaibl","year":"2022","journal-title":"Eur. J. Anaesthesiol."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1038\/s41591-024-03425-5","article-title":"The TRIPOD-LLM reporting guideline for studies using large language models","volume":"31","author":"Gallifant","year":"2025","journal-title":"Nat. Med."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"\u00c7elik, E., Turgut, M.A., Aydo\u011fan, M., K\u0131l\u0131n\u00e7, M., Tokta\u015f, \u0130., and Akelma, H. (2025). Comparison of AI applications and anesthesiologist\u2019s anesthesia method choices. BMC Anesthesiol., 25.","DOI":"10.1186\/s12871-024-02882-2"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"ooaf055","DOI":"10.1093\/jamiaopen\/ooaf055","article-title":"Comparative analysis of large language models in clinical diagnosis: Performance evaluation across common and complex medical cases","volume":"8","author":"Dinc","year":"2025","journal-title":"JAMIA Open"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Tulgar, S., Aksu, C., Selvi, O., Sultan, P., Dogan, A.T., Y\u00f6r\u00fcko\u011flu, H.U., Thomas, D.T., and Ahiskalioglu, A. (2026). A comparative evaluation of the quality of responses provided by different large language model chatbots to frequently asked questions regarding nerve blocks. BMC Anesthesiol., 26.","DOI":"10.1186\/s12871-025-03596-9"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"\u00d6nc\u00fc, S., Torun, F., and \u00dclk\u00fc, H.H. (2026). Comparison of large language models for clinical scenario generation in medical education: A mixed-methods study. BMC Med. Educ., 26.","DOI":"10.1186\/s12909-026-08821-8"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Taskin, K., and Yilmaz Ak, H. (2026). Artificial intelligence vs. human evaluation of anesthesia education videos: A comparative analysis using validated quality scales. Front. Med., 13.","DOI":"10.3389\/fmed.2026.1752664"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"661","DOI":"10.1136\/rapm-2023-104868","article-title":"Danger, Danger, Gaston Labat! Does zero-shot artificial intelligence correlate with anticoagulation guidelines recommendations for neuraxial anesthesia?","volume":"49","author":"Hurley","year":"2024","journal-title":"Reg. Anesth. Pain Med."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"81","DOI":"10.1186\/s44158-025-00296-1","article-title":"Bridging complexity: AI-guided regional anaesthesia in anticoagulated trauma patients","volume":"5","author":"Pascarella","year":"2025","journal-title":"J. Anesth. Analg. Crit. Care"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1007\/s10916-025-02247-3","article-title":"Performance of large language models in complex anesthesia decision-making: A comparative study of four LLMs in high-risk patients","volume":"49","author":"Ruan","year":"2025","journal-title":"J. Med. Syst."},{"key":"ref_17","first-page":"1584","article-title":"Evaluating large language models in cardiovascular antithrombotic care: Performance, accuracy, and implications for clinical practice","volume":"41","author":"Antiperovitch","year":"2025","journal-title":"Can. J. Cardiol."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"100949","DOI":"10.1016\/j.landig.2025.100949","article-title":"Mapping the susceptibility of large language models to medical misinformation across clinical notes and social media: A cross-sectional benchmarking analysis","volume":"8","author":"Omar","year":"2026","journal-title":"Lancet Digit. Health"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"249","DOI":"10.4103\/sja.sja_955_23","article-title":"Applications of artificial intelligence in anesthesia: A systematic review","volume":"18","author":"Kambale","year":"2024","journal-title":"Saudi J. Anaesth."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Blease, C., Bernstein, M.H., Gaab, J., Kaptchuk, T.J., Kossowsky, J., Mandl, K.D., Davis, R.B., and DesRoches, C.M. (2018). Computerization and the future of primary care: A survey of general practitioners in the UK. PLoS ONE, 13.","DOI":"10.1371\/journal.pone.0207418"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/17\/2848\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T16:17:11Z","timestamp":1788538631000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/17\/2848"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,4]]},"references-count":20,"journal-issue":{"issue":"17","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["diagnostics16172848"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16172848","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,4]]}}}