Background: The Acute Dialysis Quality Initiative Group has published a consensus definition/classification system for acute kidney injury (AKI) termed the RIFLE criteria. The Acute Kidney Injury Network (AKIN) group has recently proposed modifications to this system. It is currently unknown whether there are advantages between these criteria.
Methods: We interrogated the Australian New Zealand Intensive Care Society (ANZICS) Adult Patient Database (APD) for all adult admissions to 57 ICUs from 1 January 2000 to 31 December 2005. We compared the performance of the RIFLE and AKIN criteria for diagnosis and classification of AKI and for robustness of hospital mortality.
Results: We included 120 123 critically ill patients, of which 27.8% had a primary diagnosis of sepsis. We found only small differences (<1%) in the number of patients classified as having some degree of kidney injury using either the AKIN or RIFLE definition or classification systems. AKIN slightly increased the number of patients classified as Stage I injury (category R in RIFLE) (from 16.2 to 18.1%) but decreased the number of patients classified as having Stage II injury (category I in RIFLE) (13.6% versus 10.1%). The area under the ROC curve for hospital mortality was 0.66 for RIFLE and 0.67 for AKIN in all patients and it was 0.65 for both in septic patients.
Conclusion: Compared to the RIFLE criteria, the AKIN criteria do not materially improve the sensitivity, robustness and predictive ability of the definition and classification of AKI in the first 24 h after admission to ICU.