Research Appraisaldiagnostic

Machine learning-based risk prediction model for postoperative acute kidney injury in surgical patients with chronic kidney disease (CKD): development, validation, and SHAP-based explainability

Renal failureWu, Min, Liu, Rong, Jiang, Ming et al.1 Dec 2026DOI

Clinical Snapshot

70CEBM
Evidence: Moderatediagnostic

PICO Framework

P — PopulationSurgical patients with pre-existing chronic kidney disease (n=3,851 from MIMIC-IV database)
I — InterventionMachine learning-based risk prediction models (XGBoost, Random Forest, LightGBM, and 6 others)
C — ComparatorTraditional clinical assessment (implied comparison across different ML algorithms)
O — OutcomesPostoperative acute kidney injury as defined by KDIGO criteria

Bottom Line

This study developed machine learning models to predict postoperative AKI in surgical patients with CKD, with XGBoost achieving good discrimination (AUC 0.844, sensitivity 0.79). While the large sample size and use of standard AKI criteria are strengths, the study lacks external validation and confidence intervals for key metrics. The 24.2% AKI rate reflects the high-risk nature of this population. Key predictors included SOFA score, eGFR, blood pressure, albumin, and phosphorus. Before clinical implementation, external validation in diverse populations and demonstration of clinical utility are essential. The model shows promise for perioperative risk stratification but requires further validation to establish real-world effectiveness and safety.

Evidence: Moderate

Key Findings

  • P Value: Not reported

  • Effect Size: AUC 0.844 for XGBoost model

  • Primary Outcome: Postoperative acute kidney injury (KDIGO criteria)

  • Nnt Or Sensitivity: Sensitivity 0.79, specificity not reported

  • Confidence Interval: Not reported

Clinical Application

Requires implementation of ML infrastructure and real-time data collection systems Could inform TGA medical device approval pathways for AI-based clinical decision support; relevant for ANZICS perioperative guidelines Surgical patients with pre-existing CKD undergoing major procedures

Abstract

This study aimed to develop machine learning models to predict postoperative acute kidney injury (AKI) in surgical patients with pre-existing chronic kidney disease (CKD), a serious complication associated with high morbidity and accelerated disease progression. Using data from 3,851 surgical patients with CKD from the MIMIC-IV database, nine supervised machine learning models were developed to predict AKI, as defined by the KDIGO criteria. Postoperative AKI occurred in 24.2% (932/3,851) of the patients. The XGBoost model demonstrated the best predictive performance, achieving an AUC of 0.844, accuracy of 0.791, sensitivity of 0.79, and F1-score of 0.73, followed by Random Forest and LightGBM. The calibration curve showed a good agreement between the predicted and observed risks. For interpretability, SHAP analysis was used to identify SOFA score, estimated glomerular filtration rate (eGFR), systolic blood pressure, albumin, and phosphorus as the most influential predictors. In conclusion, machine learning models, particularly XGBoost, provide accurate and interpretable predictions of postoperative AKI in this high-risk population. The integration of explainable AI facilitates transparent risk stratification, which may support personalized perioperative care and the development of real-time clinical decision support systems.

References

  1. 1.Wu, M., Liu, R., Jiang, M., Gui, X., Yuan, J., & He, Y. (2026). Machine learning-based risk prediction model for postoperative acute kidney injury in surgical patients with chronic kidney disease (CKD): development, validation, and SHAP-based explainability. Renal Failure. https://doi.org/10.1080/0886022X.2026.2646411
Share:XLinkedIn

Related Research

Systematic Review
15CEBM

International urology and nephrology

Acute kidney injury prediction and prognostication using machine learning

1 Aug 2026

Systematic Review
60CEBM

Journal of robotic surgery

Robot-assisted versus open kidney transplantation: an umbrella review of systematic reviews and meta-analyses

27 July 2026

Systematic Review
55CEBM

Journal of robotic surgery

Transperitoneal versus retroperitoneal single-port robot-assisted partial nephrectomy: systematic review and meta-analysis of perioperative and functional outcomes

21 July 2026

This content is for educational purposes for healthcare professionals only and does not constitute clinical advice. Clinical decisions should be based on individual patient assessment, current guidelines, and appropriate specialist consultation. Editorial Standards · Privacy Policy · Terms of Service