đ Abstract
PostâPCI AKI is currently measured as a binary event based on a serum creatinine (SCr) increase of â„50% or â„0.3âmg/dL.3 This approach has several shortcomings. An equal increase in SCr represents different degrees of decrease in estimated glomerular filtration rate for different baseline renal function. Further, the prognostic effect of an AKI event could vary based on the degree of AKI and on the patient's baseline renal function. In order to better characterize the prognostic implications of AKI with respect to baseline SCr, we assessed the association of AKI with 30âday allâcause mortality by levels of preprocedural baseline SCr and AKI severity using the NCDR CathPCI registry linked with Medicare claims. Our study, the first in a PCI population, finds a higher risk of mortality in stage II/III AKI than stage I AKI. More importantly, we found that the mortality risk associated with postâPCI AKI varied nonlinearly across the range of baseline SCr. Patients with normal baseline renal function had a greater prognostic effect of AKI on mortality. Our study calls for caution while using binary postâPCI AKI as a performance metric.