Purpose: Acute kidney injury (AKI) is associated with adverse outcomes after acute ST elevation myocardial infarction (STEMI). The recently proposed AKI network (AKIN) suggested modifications to the consensus classification system for AKI known as the risk, injury, failure, loss, end-stage (RIFLE) criteria. The aim of the current study was to compare the incidence and mortality (early and late) of AKI diagnosed by RIFLE and AKIN criteria in the STEMI patients undergoing primary percutaneous intervention (PCI).
Methods: We retrospectively studied 1,033 consecutive STEMI patients undergoing primary PCI. Recruited patients were admitted between January 2008 and November 2012 to the cardiac intensive care unit with the diagnosis of acute STEMI. We compared the utilization of RIFLE and AKIN criteria for the diagnosis, classification, and prediction of mortality.
Results: The AKIN criteria allowed the identification of more patients as having AKI (9.6 vs. 3.9 %, p < 0.001) and classified more patients with stage 1 (risk in RIFLE) (7.6 vs. 1.9 %, p < 0.001) compared with the RIFLE criteria. Mortality was higher in AKI population defined by either RIFLE (46.3 vs. 6.8 %, OR 11.9, 95 % CI 6.15-23.1; p < 0.001) or AKIN (29 vs. 6.1 %; OR 6.3, 95 % CI 3.8-10.4; p < 0.001) criteria. In a multivariable logistic regression model, AKI defined with both RIFLE and AKIN was an independent predictor of both 30-day and up to 5-year all-cause mortality. However, there was no significant statistical difference in the risk provided by these two scoring systems.
Conclusions: AKIN criteria are more sensitive in defining AKI compared with the RIFLE criteria in STEMI. However, no difference exists in the mortality risk provided by these two scoring systems.