Integrating Pre- and Postoperative Systemic Inflammatory Markers for Acute Kidney Injury Prediction Following Radical...

Area of Science:

• Urology
• Nephrology
• Oncology

Background:

• Radical cystectomy for bladder cancer can lead to acute kidney injury (AKI).
• Dynamic changes in systemic inflammation markers during the perioperative period are implicated in AKI development.
• Predictive tools for AKI post-radical cystectomy are crucial for patient management.

Purpose of the Study:

• To investigate the association between perioperative dynamic changes in systemic inflammation markers and AKI following radical cystectomy.
• To evaluate the predictive value of these inflammatory markers for AKI using machine learning algorithms.

Main Methods:

• Retrospective analysis of 727 patients undergoing radical cystectomy (2013-2022).
• Calculation of perioperative inflammation index based on dynamic changes in peripheral blood cell counts.
• Development and evaluation of AKI prediction models using logistic regression and machine learning (XGBoost, AUROC).

Main Results:

• 20.8% of patients developed AKI post-radical cystectomy.
• Independent risk factors for AKI included lower postoperative hemoglobin and albumin, lower intraoperative fluid infusion rate, and higher perioperative inflammation index.
• The XGBoost model demonstrated the best AKI prediction performance with an AUROC of 0.801.

Conclusions:

• Perioperative dynamic changes in inflammatory markers are significantly associated with AKI after radical cystectomy.
• Integrating perioperative metrics into machine learning models allows for early AKI identification.
• These predictive models can optimize perioperative management strategies to prevent AKI.