Background: We developed risk models for predicting acute kidney injury (AKI) and AKI requiring dialysis (AKI‐D) after percutaneous coronary intervention (PCI) to support quality assessment and the use of preventative strategies.
Methods and results: AKI was defined as an absolute increase of ≥0.3 mg/dL or a relative increase of 50% in serum creatinine (AKIN Stage 1 or greater) and AKI‐D was a new requirement for dialysis following PCI. Data from 947 012 consecutive PCI patients and 1253 sites participating in the NCDR Cath/PCI registry between 6/09 and 7/11 were used to develop the model, with 70% randomly assigned to a derivation cohort and 30% for validation. AKI occurred in 7.33% of the derivation and validation cohorts. Eleven variables were associated with AKI: older age, baseline renal impairment (categorized as mild, moderate, and severe), prior cerebrovascular disease, prior heart failure, prior PCI, presentation (non‐ACS versus NSTEMI versus STEMI), diabetes, chronic lung disease, hypertension, cardiac arrest, anemia, heart failure on presentation, balloon pump use, and cardiogenic shock. STEMI presentation, cardiogenic shock, and severe baseline CKD were the strongest predictors for AKI. The full model showed good discrimination in the derivation and validation cohorts (c‐statistic of 0.72 and 0.71, respectively) and identical calibration (slope of calibration line=1.01). The AKI‐D model had even better discrimination (c‐statistic=0.89) and good calibration (slope of calibration line=0.99).
Conclusion: The NCDR AKI prediction models can successfully risk‐stratify patients undergoing PCI. The potential for this tool to aid clinicians in counseling patients regarding the risk of PCI, identify patients for preventative strategies, and support local quality improvement efforts should be prospectively tested.