A risk score predicting unplanned renal replacement therapy after coronary catheterization

Area of Science:

• Nephrology
• Cardiology
• Medical Imaging

Background:

• Iodinated contrast media can cause acute kidney injury (AKI), potentially requiring renal replacement therapy (RRT).
• Existing prediction models for contrast-induced AKI (CI-AKI) often do not specifically identify patients needing RRT.
• There is a need for models to predict RRT risk in patients undergoing specific cardiovascular procedures.

Purpose of the Study:

• To develop and validate a preprocedural risk score model.
• To stratify patients at risk of requiring unplanned postprocedural RRT.
• To focus on patients undergoing invasive coronary angiography (ICA) or percutaneous coronary intervention (PCI).

Main Methods:

• A total of 3,469 patients were divided into development and validation cohorts.
• Multivariable logistic regression identified independent predictors of RRT.
• The model incorporated five preprocedural variables: age, serum creatinine, diabetes, hypotension, and acute myocardial infarction.

Main Results:

• Only 1.0% of patients (36/3,469) required unplanned RRT.
• The risk score model demonstrated high discrimination (C-statistic = 0.872) and good fit.
• Risk stratification identified low (~0.4%), intermediate (3.0%), and high (20.0%) RRT rates.

Conclusions:

• A validated risk score model was developed using preprocedural variables.
• This model effectively predicts the need for unplanned RRT after ICA and/or PCI.
• The score allows for hierarchical risk classification.