Öz

Background: Preoperative assessment and optimization are pivotal for complex surgical patients, yet traditional scores and clinician-driven workflows may miss subtle risk interactions and modifiable factors. Artificial intelligence (AI), especially machine learning (ML), can leverage large-scale electronic health record (EHR) data to improve risk stratification and support targeted optimization strategies.

Methods: We conducted a narrative review focusing on practical, clinically actionable AI applications in preoperative assessment and optimization, emphasizing recent advances and implementation considerations for high-risk surgical patients.

Results: ML-based preoperative risk models can outperform conventional calculators, exemplified by a gradient-boosted decision-tree model trained on >1.4 million surgical cases that achieved an AUROC of 0.95 for 30-day mortality and exceeded NSQIP performance, with prospective real-world deployment. Automated frailty phenotyping from structured preoperative EHR data (demographics, ASA/acuity, ICD-10/CCS diagnoses, and routine labs) has also been externally validated in older adults and shows strong stepwise associations with adverse outcomes. AI additionally supports preoperative optimization by identifying actionable targets such as anemia risk (supporting early iron/EPO pathways), penicillin allergy delabeling opportunities, and improved detection of risky alcohol use via natural language processing of clinical notes. Successful clinical impact depends on workflow integration, interpretability, and attention to privacy, bias, regulation, and prospective evidence.

Conclusions: AI-enabled preoperative assessment can enhance identification of high-risk patients and systematically surface modifiable factors for optimization, with early evidence of improved predictive performance and feasible integration into point-of-care workflows. Future work should prioritize robust external validation across diverse populations, implementation studies demonstrating patient-centered benefit, and governance frameworks ensuring safety, fairness, and clinician trust.

Anahtar Kelimeler: Artificial intelligence, management, optimization, preoperative assessment, prediction

Referanslar

  1. Jones JH, Sakai T. The Potential Impacts of Artificial Intelligence on Preoperative Optimization and Predicting Risks of Morbidity and Mortality: A Narrative Focused Review. A A Pract. 2025;19:e02061. https://doi.org/10.1213/XAA.0000000000002061
  2. Mahajan A, Esper S, Oo TH, et al. Development and Validation of a Machine Learning Model to Identify Patients Before Surgery at High Risk for Postoperative Adverse Events. JAMA Netw Open. 2023;6:e2322285. https://doi.org/10.1001/jamanetworkopen.2023.22285
  3. Lu Y, Forlenza E, Wilbur RR, et al. Machine-learning model successfully predicts patients at risk for prolonged postoperative opioid use following elective knee arthroscopy. Knee Surg Sports Traumatol Arthrosc. 2022;30:762-72. https://doi.org/10.1007/s00167-020-06421-7
  4. Corey KM, Kashyap S, Lorenzi E, et al. Development and validation of machine learning models to identify high-risk surgical patients using automatically curated electronic health record data (Pythia): A retrospective, single-site study. PLoS Med. 2018;15:e1002701. https://doi.org/10.1371/journal.pmed.1002701
  5. Fritz BA, King CR, Abdelhack M, et al. Effect of machine learning models on clinician prediction of postoperative complications: the Perioperative ORACLE randomised clinical trial. Br J Anaesth. 2024;133:1042-50. https://doi.org/10.1016/j.bja.2024.08.004
  6. Shelley B, Shaw M. Machine learning and preoperative risk prediction: the machines are coming. Br J Anaesth. 2024;133:925-30. https://doi.org/10.1016/j.bja.2024.07.015
  7. Bai C, Xiao F, Al-Ani M, Price CC, Manini TM, Mardini MT. External validation of an AI-based preoperative frailty index using real-world data. J Gerontol A Biol Sci Med Sci. 2025;80:glaf119. https://doi.org/10.1093/gerona/glaf119
  8. Park S, Chung S, Kim Y, et al. A deep-learning algorithm using real-time collected intraoperative vital sign signals for predicting acute kidney injury after major non-cardiac surgeries: A modelling study. PLoS Med. 2025;22:e1004566. https://doi.org/10.1371/journal.pmed.1004566
  9. Kolin DA, Lyman S, Della Valle AG, Ast MP, Landy DC, Chalmers BP. Predicting Postoperative Anemia and Blood Transfusion Following Total Knee Arthroplasty. J Arthroplasty. 2023;38:1262-1266.e2. https://doi.org/10.1016/j.arth.2023.01.018
  10. Murphy ZR, Muzaffar AF, Massih SA, et al. Examining cefazolin utilization and perioperative anaphylaxis in patients with and without a penicillin allergy label: A cross-sectional study. J Clin Anesth. 2024;94:111377. https://doi.org/10.1016/j.jclinane.2024.111377
  11. Jiang M, Lam A, Lam L, et al. Artificial intelligence and the potential for perioperative delabeling of penicillin allergies for neurosurgery inpatients. Br J Neurosurg. 2025;39:40-3. https://doi.org/10.1080/02688697.2023.2173724
  12. Bishara A, Chiu C, Whitlock EL, et al. Postoperative delirium prediction using machine learning models and preoperative electronic health record data. BMC Anesthesiol. 2022;22:8. https://doi.org/10.1186/s12871-021-01543-y
  13. Nair AA, Velagapudi MA, Lang JA, et al. Machine learning approach to predict postoperative opioid requirements in ambulatory surgery patients. PLoS One. 2020;15:e0236833. https://doi.org/10.1371/journal.pone.0236833
  14. Vydiswaran VGV, Strayhorn A, Weber K, et al. Automated-detection of risky alcohol use prior to surgery using natural language processing. Alcohol Clin Exp Res (Hoboken). 2024;48:153-63. https://doi.org/10.1111/acer.15222
  15. Xue B, Li D, Lu C, et al. Use of Machine Learning to Develop and Evaluate Models Using Preoperative and Intraoperative Data to Identify Risks of Postoperative Complications. JAMA Netw Open. 2021;4:e212240. https://doi.org/10.1001/jamanetworkopen.2021.2240
  16. Lee SW, Lee HC, Suh J, et al. Multi-center validation of machine learning model for preoperative prediction of postoperative mortality. NPJ Digit Med. 2022;5:91. https://doi.org/10.1038/s41746-022-00625-6
  17. Graeßner M, Jungwirth B, Frank E, et al. Enabling personalized perioperative risk prediction by using a machine-learning model based on preoperative data. Sci Rep. 2023;13:7128. https://doi.org/10.1038/s41598-023-33981-8
  18. Abraham J, Bartek B, Meng A, et al. Integrating machine learning predictions for perioperative risk management: Towards an empirical design of a flexible-standardized risk assessment tool. J Biomed Inform. 2023;137:104270. https://doi.org/10.1016/j.jbi.2022.104270
  19. Chung P, Fong CT, Walters AM, Aghaeepour N, Yetisgen M, O’Reilly-Shah VN. Large Language Model Capabilities in Perioperative Risk Prediction and Prognostication. JAMA Surg. 2024;159:928-37. https://doi.org/10.1001/jamasurg.2024.1621
  20. Ouyang D, Theurer J, Stein NR, et al. Electrocardiographic deep learning for predicting post-procedural mortality: a model development and validation study. Lancet Digit Health. 2024;6:e70-8. https://doi.org/10.1016/S2589-7500(23)00220-0

Telif hakkı ve lisans

Telif hakkı © 2026 Yazar(lar). Açık erişimli bu makale, orijinal çalışmaya uygun şekilde atıfta bulunulması koşuluyla, herhangi bir ortamda veya formatta sınırsız kullanım, dağıtım ve çoğaltmaya izin veren Creative Commons Attribution License (CC BY) altında dağıtılmıştır.

Nasıl atıf yapılır?

1.
Elmaleh Y, Moussali Y, Boyer R. Artificial intelligence in preoperative assessment and optimization: a narrative review. Turk J Intensive Care. 2026;24(1):1-10. https://doi.org/10.63729/TJIC.2026.683