Abstract
Acute kidney injury (AKI) is a common outcome evaluated in clinical studies, often as a safety end point in a variety of cardiovascular, kidney disease, and other clinical trials. AKI end points that include modest increases in serum creatinine levels from baseline may not associate with patient-centered outcomes such as initiation of dialysis, sustained decline in kidney function, or death. Surprisingly, data from several randomized controlled trials have suggested that in certain settings, the development of AKI may be associated with favorable outcomes. AKI safety end points that are nonspecific and may not associate with patient-centered outcomes could result in beneficial therapies being inappropriately withheld or never developed for commercial use. We review several issues related to commonly used AKI definitions and suggest that future work in AKI use more patient-centered AKI end points such as major adverse kidney events at 30 days or other later time points.