Journals / Muğla Sıtkı Koçman Üniversitesi Tıp Dergisi / 2026 / Volume: 13 Issue: 2

Postoperative Complications in Older Adults After Local Surgical Procedures; Risk Factor Analysis and Comparison of Machine Learning Methods

Lokal Cerrahi İşlemler Sonrası Yaşlılarda Ameliyat Sonrası Komplikasyonlar; Risk Faktörü Analizi ve Makine Öğrenme Yöntemlerinin Karşılaştırılması

Emrah Işıktekin(BALIKESIR UNIVERSITY, SCHOOL OF MEDICINE)0000-0001-6068-9899
Ali Sezgin(T.C. SAĞLIK BAKANLIĞI ANKARA ETLİK ŞEHİR HASTANESİ)0000-0003-0139-7962
Pages
—
Publication Date
2026-08-27
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

Machine-learning approaches, particularly random forest models, may improve perioperative risk stratification in this population.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Older adults frequently undergo excision of cutaneous lesions under local anesthesia and may be at increased risk of postoperative complications due to comorbid conditions and concomitant medications.
📊
Key Findings & Data:This study aimed to identify predictors of postoperative complications in patients aged 65 years and older and to compare the predictive performance of logistic regression with machine-learning models.
💡
Conclusion & Impact:Machine-learning approaches, particularly random forest models, may improve perioperative risk stratification in this population.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Cutaneous surgery · Elderly patients · Logistic regression

Abstract

Older adults frequently undergo excision of cutaneous lesions under local anesthesia and may be at increased risk of postoperative complications due to comorbid conditions and concomitant medications. This study aimed to identify predictors of postoperative complications in patients aged 65 years and older and to compare the predictive performance of logistic regression with machine-learning models. This retrospective study included 314 patients who underwent cutaneous lesion excision under local anesthesia. Demographic, clinical, and perioperative variables were collected and assessed using multivariable logistic regression. Decision tree, random forest, and gradient boosting models were developed using an 80/20 training–testing split. Missing data were imputed, and class imbalance was addressed using resampling techniques. Model performance was evaluated using discrimination, calibration, and decision curve analysis. Postoperative complications occurred in 119 patients (37.9%). Patients with complications were older, had lower body mass index, and more frequently had diabetes, coronary artery disease, and antithrombotic therapy. Logistic regression identified low body mass index, diabetes, and coronary artery disease as independent predictors but demonstrated limited discriminatory ability. Among machine-learning models, the random forest achieved the best performance, with superior discrimination and calibration. Body mass index, perioperative blood pressure, age, and procedure duration were the most influential predictors. Postoperative complications are common among older adult patients undergoing cutaneous surgery under local anesthesia. Low body mass index, cardiometabolic comorbidities, perioperative hemodynamic factors, and operative duration are key determinants of risk. Machine-learning approaches, particularly random forest models, may improve perioperative risk stratification in this population.

Özet

Yaşlı bireylerde lokal anestezi altında yapılan kutanöz lezyon eksizyonları sık uygulanmakta olup, eşlik eden komorbiditeler ve birlikte kullanılan ilaçlar nedeniyle postoperatif komplikasyon riski artmış olabilir. Bu çalışmanın amacı, 65 yaş ve üzeri hastalarda postoperatif komplikasyonların öngördürücülerini belirlemek ve lojistik regresyon ile makine öğrenmesi modellerinin öngörü performanslarını karşılaştırmaktır. Bu retrospektif çalışmaya, lokal anestezi altında kutanöz lezyon eksizyonu yapılan 314 hasta dahil edildi. Demografik, klinik ve perioperatif değişkenler toplanarak çok değişkenli lojistik regresyon analizi ile değerlendirildi. %80/%20 eğitim–test bölünmesi kullanılarak karar ağacı, random forest ve gradient boosting modelleri geliştirildi. Eksik veriler imputasyon yöntemleriyle tamamlandı ve sınıf dengesizliği yeniden örnekleme teknikleri ile düzeltildi. Model performansı ayırt edicilik, kalibrasyon ve karar eğrisi analizi kullanılarak değerlendirildi. Postoperatif komplikasyonlar 119 hastada (%37.9) gözlendi. Komplikasyon gelişen hastalar daha ileri yaşta, daha düşük vücut kitle indeksine (VKİ) sahipti ve diyabet, koroner arter hastalığı ile antitrombotik tedavi öyküsü daha sıktı. Lojistik regresyon analizinde düşük VKİ, diyabet ve koroner arter hastalığı bağımsız öngördürücüler olarak saptandı; ancak modelin ayırt edici gücü sınırlıydı. Makine öğrenmesi modelleri arasında random forest en iyi performansı göstererek daha üstün ayırt edicilik ve kalibrasyon sağladı. VKİ, perioperatif kan basıncı, yaş ve işlem süresi en etkili öngördürücüler olarak belirlendi. Lokal anestezi altında kutanöz cerrahi uygulanan yaşlı hastalarda postoperatif komplikasyonlar sık görülmektedir. Düşük VKİ, kardiyometabolik komorbiditeler, perioperatif hemodinamik faktörler ve işlem süresi komplikasyon riskinin temel belirleyicileridir. Özellikle random forest modelleri olmak üzere makine öğrenmesi yaklaşımları, bu hasta grubunda perioperatif risk sınıflandırmasını iyileştirebilir.

🔖

Cite This Article

Işıktekin, E., Sezgin, A., *, A. S. (2026). Postoperative Complications in Older Adults After Local Surgical Procedures; Risk Factor Analysis and Comparison of Machine Learning Methods. Muğla Sıtkı Koçman Üniversitesi Tıp Dergisi, Volume: 13 Issue: 2. https://doi.org/10.47572/muskutd.1881577

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (18)

  1. Chung JY, Chang WY, Lin TW, et al. An analysis of surgical outcomes in patients aged 80 years and older. Acta Anaesthesiol Taiwan. 2014;52(4):153-158.
  2. Lin HS, McBride RL, Hubbard RE. Frailty and anesthesia: risks during and post-surgery. Local Reg Anesth. 2018;11:61-73.
  3. Strickler AG, Shah P, Bajaj S, et al. Preventing and managing complications in dermatologic surgery: procedural and postsurgical concerns. J Am Acad Dermatol. 2021;84(4):895-903.
  4. Prasada S, Desai MY, Saad M, et al. Preoperative atrial fibrillation and cardiovascular outcomes after noncardiac surgery. J Am Coll Cardiol. 2022;79(25):2471-2485.
  5. Kądzioła O, Stępień K, Del Carmen Yika A, et al. CHA₂DS₂-VASc score as a mortality predictor in acute heart failure with preserved ejection fraction. Front Cardiovasc Med. 2025;12:1611825.
  6. Brydges G, Uppal A, Gottumukkala V. Application of machine learning in predicting perioperative outcomes in patients with cancer: a narrative review for clinicians. Curr Oncol. 2024;31(5):2727-2747.
  7. Theodorakis N, Nikolaou M, Hitas C, et al. Comprehensive peri-operative risk assessment and management of geriatric patients. Diagnostics (Basel). 2024;14(19):2153.
  8. Tran GS, Nguyen AP, Pham HM, et al. Impact of frailty on hospital adverse outcomes in elderly with acute coronary syndrome. Lifestyle Med. 2025;6(4):e70035.
  9. Walco JP, Rengel KF, McEvoy MD, et al. Association between preoperative blood pressures and postoperative adverse events. Anesthesiology. 2024;141(2):272-285.
  10. Cartailler J, Beaucote V, Trillat B, et al. Deciphering the explanatory potential of blood pressure variables on post-operative length of stay through hierarchical clustering: a retrospective monocentric study. PLoS One. 2024;19(9):e0308910.
  11. Cylwik J, Celińska-Spodar M, Dudzic M. Individualized perioperative hemodynamic management using hypotension prediction index software and the dynamics of troponin and NT-proBNP concentration changes in patients undergoing oncological abdominal surgery. J Pers Med. 2024;14:211.
  12. Gencer B, Eisen A, Berger D, et al. Edoxaban versus warfarin in high-risk patients with atrial fibrillation: a comprehensive analysis of high-risk subgroups. Am Heart J. 2022;247:24-32.
  13. Johnson S, Haywood C. Perioperative medication management for older people. J Pharm Pract Res. 2022;52(5):391-401.
  14. Chowdhury A, Rahman Druvo MM, Islam MS, et al. An application to detect osteoporosis using ensemble machine learning with hyperparameter tuning and model interpretability. Clin eHealth. 2025;8:177-200.
  15. Naderalvojoud B, Hernandez-Boussard T. Improving machine learning with ensemble learning on observational healthcare data. AMIA Annu Symp Proc. 2023;2023:521-529.
  16. Khan AA, Chaudhari O, Chandra R. A review of ensemble learning and data augmentation models for class imbalanced problems: combination, implementation and evaluation. Expert Syst Appl. 2024;244:122778.
  17. Napa KK, Govindarajan R, Sathya S, et al. Comparative analysis of explainable machine learning models for cardiovascular risk stratification using clinical data and Shapley additive explanations. Intell Based Med. 2025;12:100286.
  18. Hoghooghi Esfahani H, Toyonaga S, Oyibo K. The application of explainable artificial intelligence in the prediction, diagnosis, treatment, and management of chronic diseases: a systematic review. Digit Health. 2025;11:20552076251355669.