Prediction of Significant Creatinine Elevation in First ICU Stays with Vancomycin Use: A retrospective study through Catboost

2025-07-30

AI Summary

This study addresses the early prediction of vancomycin-associated significant serum creatinine elevation (≥0.3 mg/dL or ≥50%)—a key indicator of acute kidney injury—in ICU patients initiating vancomycin therapy. Method: Leveraging 10,288 real-world ICU admissions, we employed SelectKBest combined with random forest for feature selection and comparatively evaluated six machine learning algorithms; CatBoost achieved optimal performance (AUROC = 0.818, sensitivity = 0.800, NPV = 0.900). Contribution/Results: We innovatively integrated SHAP, Accumulated Local Effects (ALE), and Bayesian posterior sampling to quantitatively interpret nonlinear dose–response relationships of critical risk factors—including phosphate, total bilirubin, and magnesium—enhancing clinical interpretability and decision trustworthiness. The high-risk cohort exhibited a mean predicted probability of 60.5%, providing evidence-based support for individualized dosing and nephroprotective interventions.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Background: Vancomycin, a key antibiotic for severe Gram-positive infections in ICUs, poses a high nephrotoxicity risk. Early prediction of kidney injury in critically ill patients is challenging. This study aimed to develop a machine learning model to predict vancomycin-related creatinine elevation using routine ICU data. Methods: We analyzed 10,288 ICU patients (aged 18-80) from the MIMIC-IV database who received vancomycin. Kidney injury was defined by KDIGO criteria (creatinine rise >=0.3 mg/dL within 48h or >=50% within 7d). Features were selected via SelectKBest (top 30) and Random Forest ranking (final 15). Six algorithms were tested with 5-fold cross-validation. Interpretability was evaluated using SHAP, Accumulated Local Effects (ALE), and Bayesian posterior sampling. Results: Of 10,288 patients, 2,903 (28.2%) developed creatinine elevation. CatBoost performed best (AUROC 0.818 [95% CI: 0.801-0.834], sensitivity 0.800, specificity 0.681, negative predictive value 0.900). Key predictors were phosphate, total bilirubin, magnesium, Charlson index, and APSIII. SHAP confirmed phosphate as a major risk factor. ALE showed dose-response patterns. Bayesian analysis estimated mean risk 60.5% (95% credible interval: 16.8-89.4%) in high-risk cases. Conclusions: This machine learning model predicts vancomycin-associated creatinine elevation from routine ICU data with strong accuracy and interpretability, enabling early risk detection and supporting timely interventions in critical care.

Problem

• Predict vancomycin-induced kidney injury in ICU patients
• Develop machine learning model for creatinine elevation prediction
• Identify key risk factors from routine ICU data

Innovation

• CatBoost model predicts creatinine elevation accurately
• Feature selection via SelectKBest and Random Forest
• Interpretability enhanced with SHAP and ALE