Developing an AKI Consensus Definition for Database Research:Findings From a Scoping Review and Expert Opinion Using a Delphi Process

Rationale & Objective: The KDIGO (Kidney Disease: Improving Global Outcomes) definition of acute kidney injury (AKI) is frequently used in studies to examine the epidemiology of AKI. This definition is variably interpreted and applied to routinely collected health care data. The aim of this study was to examine this variation and to achieve consensus in how AKI should be defined for research using routinely collected health care data.Sources of Evidence and Study Design: Scoping review via searching Medline and EMBASE for studies using health care data to examine AKI by using the KDIGO creatinine-based definition. An international panel of experts formed to participate in a modified Delphi process to attempt to generate consensus about how AKI should be defined when using routinely collected laboratory data.Charting Methods and Analytical Approach: The Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) extension for scoping reviews was followed. For the Delphi process, 2 rounds of questions were distributed via internet-based questionnaires to all participants with a prespecified cutoff of 75% agreement used to define consensus.Results: The scoping review found 174 studies that met the inclusion criteria. The KDIGO definition was inconsistently applied, and the methods for application were poorly described. We found 58 (33%) of papers did not provide a definition of how the baseline creatinine value was determined, and only 34 (20%) defined recovery of kidney function. Of 55 invitees to the Delphi process, 35 respondents participated in round 1, and 25 participated in round 2. Some consensus was achieved in areas related to how to define the baseline creatinine value, which patients should be excluded from analysis of routinely collected laboratory data, and how persistent chronic kidney disease or nonrecovery of AKI should be defined.Limitations: The Delphi panel members predominantly came from the United Kingdom, the United States, and Canada, and there were low response rates for some questions in round 1.Conclusions: The current methods for defining AKI using routinely collected data are inconsistent and poorly described in the available literature. Experts could not achieve consensus for many aspects of defining AKI and describing its sequelae. The KDIGO guidelines should be extended to include a standardized definition for how AKI should be defined when using routinely collected data.