Refining Safe Contrast Limits for Preventing Acute Kidney Injury After Percutaneous Coronary Intervention

Area of Science:

• Cardiology
• Nephrology
• Medical Informatics

Background:

• Contrast-associated acute kidney injury (CA-AKI) is a significant complication of percutaneous coronary intervention (PCI).
• Current tools for determining safe contrast volumes during PCI are limited.
• Developing accurate risk models is crucial for preventing CA-AKI.

Purpose of the Study:

• To develop and validate novel risk models for predicting CA-AKI during PCI.
• To tailor safe contrast volume limits based on individual patient risk.
• To compare the performance of new models against existing safe contrast equations.

Main Methods:

• Logistic regression models were developed using data from 20,579 PCIs across 18 hospitals.
• Models included simpler versions using easily derivable electronic health record data.
• Prospective validation was performed on 5,423 PCIs, comparing AUC with preexisting models.

Main Results:

• Preexisting safe contrast limits showed poor accuracy (53.7% and 56.6%) in predicting CA-AKI.
• New multivariate models demonstrated significantly better accuracy (up to 73.1%) and AUC (up to 0.80).
• Applying safe contrast limits could reduce CA-AKI risk in approximately 25% of patients.

Conclusions:

• New multivariate models are substantially more accurate than preexisting equations for predicting CA-AKI.
• These models can identify patients who would benefit most from contrast volume limitation during PCI.
• Implementation into electronic health records can provide actionable insights to enhance PCI safety.