Copyright: ©Author(s) 2026. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial (CC BY-NC 4.0) license. No commercial re-use. See permissions. Published by Baishideng Publishing Group Inc.
Development and validation of an interpretable machine learning model for predicting acute kidney injury after pancreatic surgery
Wei-Bin Wang, Li-Meng Chen, Peng Xia, Meng-Hua Dai, Jun-Chao Guo, Tai-Ping Zhang, Xian-Lin Han, Qiang Xu, Chang Liu, Yao-Zong Wang, Yi-Xuan Sun, Na-Su Wang, Liang-Bo Dong, Jaeyun J Wang, Georgios A Margonis, Jia-Shu Han, Tian-Yu Li, Hua Zheng, Ren-Kui Fu, Chen Lin
Chen Lin, Ren-Kui Fu, Tian-Yu Li, Jia-Shu Han, Liang-Bo Dong, Na-Su Wang, Yi-Xuan Sun, Yao-Zong Wang, Chang Liu, Qiang Xu, Xian-Lin Han, Tai-Ping Zhang, Jun-Chao Guo, Meng-Hua Dai, Wei-Bin Wang, Department of General Surgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China
Hua Zheng, Peng Xia, Li-Meng Chen, Department of Nephrology, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China
Georgios A Margonis, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY 10065, United States
Jaeyun J Wang, Department of Surgery, University of California San Francisco, San Francisco, CA 94143, United States
Co-first authors: Chen Lin and Ren-Kui Fu.
Co-corresponding authors: Li-Meng Chen and Wei-Bin Wang.
Author contributions: Lin C, Zheng H, and Wang WB conceptualized the study and designed the methodology; Fu RK, Li TY, Han JS, Dong LB, Wang NS, Sun YX, and Wang YZ curated the data; Lin C, Fu RK, Liu C, and Li TY were involved in formal analysis; Lin C and Fu RK wrote the original draft of the manuscript; Margonis GA and Wang JJ contributed to its review and editing; Xia P, Chen LM and Wang WB supervised the study and acquired the funding; Xu Q, Guo JC, Zhang TP, Dai MH, and Han XL had access to and verified the underlying study data; all authors read and consented to the published version of the manuscript and accept the responsibility to submit the manuscript for publication; The collaborative efforts of Lin C and Fu RK were essential for the progress and successful completion of the study, underlying their merit as co-first authors; The collaboration of Chen LM and Wang WB supported the smooth and productive overall teamwork required to carry out the work and will continue beyond, underlying their roles as co-corresponding authors on this paper.
Supported by the National Natural Science Foundation of China, No. 82573412 and No. 82173074; Nonprofit Central Research Institute Fund of Chinese Academy of Medical Sciences, No. 2018PT32014; Capital’s Funds for Health Improvement and Research, No. 2024-2-4017; National Multidisciplinary Cooperative Diagnosis and Treatment Capacity Building Project for Major Diseases, No. ZK12101; and National High Level Hospital Clinical Research Funding, No. 2025-PUMCH-A-073.
Institutional review board statement: This study was approved by the Ethics Committee of the Peking Union Medical College Hospital (approval No. I-25PJ2719). This study was conducted in accordance with the Declaration of Helsinki (revised 2013).
Informed consent statement: This retrospective study used existing clinical data and was approved by the Ethics Committee with informed consent waived.
Conflict-of-interest statement: The authors declare that they have no conflict of interest.
Data sharing statement: No additional data are available.
Corresponding author: Wei-Bin Wang, MD, Professor, Department of General Surgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 1 Shuaifu Yuan, Dongcheng District, Beijing 100730, China.
wwb_xh@163.com
Received: November 7, 2025
Revised: January 4, 2026
Accepted: February 26, 2026
Published online: May 21, 2026
Processing time: 193 Days and 5.2 Hours
BACKGROUND
Acute kidney injury (AKI) is a common and serious complication of major abdominal surgery. However, predictive models specific to pancreatic surgery remain scarce.
AIM
To develop and validate an interpretable machine learning model for early prediction of postoperative AKI following pancreatic surgery.
METHODS
Adults undergoing pancreaticoduodenectomy or distal or total pancreatectomy from 2014 to 2024 were retrospectively analyzed. AKI was defined by the kidney disease Improving global outcomes creatinine-based criteria. After matching, data from 2014-2021 trained seven models using Boruta/least absolute shrinkage and selection operator selection and five-fold cross-validation. Data from 2022-2024 were utilized for validation. Model performance was evaluated by the area under the receiver operating characteristic curve (AUROC); Shapley Additive Explanations were used for interpretation, and an online calculator was developed.
RESULTS
Among the 4216 eligible patients, 230 (5.5%) developed postoperative AKI. The categorical boosting model showed the best performance in the training cohort (AUROC = 0.803) and maintained a robust prediction in the validation cohort (AUROC = 0.751). Shapley Additive Explanations analysis highlighted operative time, postoperative serum creatinine level, intensive care unit admission after surgery, postoperative white blood cell count, and intraoperative red blood cell transfusion as key features for predicting postoperative AKI.
CONCLUSION
The developed models showed satisfactory performance for predicting postoperative AKI in patients undergoing major pancreatic surgery. They may facilitate early high-risk identification and inform perioperative management strategies.
Core Tip: Using a decade-long pancreatic surgery cohort, we developed and validated an interpretable machine learning model to identify patients at risk for postoperative acute kidney injury using routinely available perioperative data. The final model achieved good discrimination and highlighted modifiable drivers of acute kidney injury, including longer operative time, early postoperative serum creatinine elevation, admission to the intensive care unit, increased postoperative white blood cell count, and greater intraoperative blood loss. An online risk calculator is provided to support bedside individualized prediction and early renoprotective management.
