Contemporary Methods for Predicting Acute Kidney Injury After Coronary Intervention

Area of Science:

• Cardiology
• Nephrology
• Health Services Research

Background:

• Acute kidney injury (AKI) is a frequent complication following percutaneous coronary intervention (PCI).
• Accurate risk prediction for AKI after PCI is crucial for performance benchmarking and prospective clinical guidance.

Purpose of the Study:

• To update the 2014 National Cardiovascular Data Registry (NCDR) AKI risk model with contemporary data.
• To enhance the prediction of AKI and new dialysis requirement post-PCI.
• To develop bedside risk scores and contrast dose thresholds for clinical application.

Main Methods:

• Utilized the NCDR CathPCI Registry for 2020 PCI procedures.
• Excluded patients on dialysis or lacking postprocedural creatinine data.
• Developed logistic regression models using a derivation cohort (70%) and validated them in a separate cohort (30%).
• Defined AKI as a 0.3 mg/dL or 50% increase in creatinine from baseline.

Main Results:

• The study analyzed 455,806 PCI procedures.
• The incidence of AKI was 7.2% and new dialysis was 0.7%.
• Key predictors included baseline renal function and clinical instability.
• The final AKI model demonstrated strong predictive performance (C-statistic = 0.798) with excellent calibration in the validation cohort.

Conclusions:

• The updated NCDR AKI risk model provides refined prediction of AKI after PCI.
• This improved model supports enhanced clinical care, quality improvement initiatives, and performance benchmarking.