Panel Recommends 5 Reporting Requirements for Standardizing AKI Definition

Inconsistent application and measure of the Kidney Disease Improving Global Outcomes definition of acute kidney injury (AKI) prompted this review, which used a measure of 75% agreement when defining consensus.

A 55-member panel has come up with 5 recommendations they believe will help to standardize reporting of acute kidney injury (AKI) for database research, according to study findings reported today in American Journal of Kidney Diseases.

Past research shows inconsistent application of the Kidney Disease Improving Global Outcomes (KDIGO) definition of AKI, the study authors noted, so hoping “to examine this variation and to achieve consensus in how AKI should be defined for research using routinely collected health care data,” they farmed Medline and Embase (inception to June 22, 2017) for studies that applied KDIGO’s creatine-based definition of the condition.

Their search produced 174 studies among which both the process to determine baseline creatinine nor recovery of kidney function were defined, at 33% and 20%, respectively. Overall, they found widely inconsistent application of KDIGO’s definition of AKI following 2 rounds of internet-based surveys by following the Delphi process. Their consensus cutoff was a 75% agreement rate with each survey round, and among the 55 respondents (mostly nephrologists and epidemiologists) from the United Kingdom, the United States, and Canada, 35 responded in round 1 and 25 in round 2.

Among the studies included in the final analysis, 15.5% did not provide detail on how they measured AKI, except that they used the KDIGO definition; 28.2% gauged AKI via urine output; 77.4% used a 48-hour measure of 0.3 mg/dL creatinine increase for AKI; and 6.9% partially implemented a definition of creatinine clearance.

In addition, 20.7% of the studies did not define baseline creatinine, and for those that did, 12.6% partially defined the measure and 67% provided a comprehensive definition (timeframe for baseline creatinine and method for selecting among multiple values). Just 19.5% defined recovery of kidney function.

Following the Delphi process, consensus was reached on the following:

  • 85.7% agreed that the KDIGO definition of stage I AKI should include a serum creatinine increase of more than 50% from baseline
  • 80.0% agreed that stage I AKI included a 0.3 mg/dL creatinine jump in 48 hours, note stage III AKI
  • 77.1% agreed on a maximum of 1 baseline creatinine measure
  • 80.0% agreed it is not a good idea to assume a prior year’s estimated glomerular filtration rate is 75 mL/min if that year’s creatine measures are unavailable
  • 76.0% agreed on using all creatine measures from the previous 7 to 30 days to define AKI, if only using that criteria to define AKI
  • 88.0% thought both inpatient and outpatient data should be used to define baseline creatinine values
  • 96.0% agreed with the statement, “If an episode of AKI is present in the baseline, then remove the peak serum creatinine and values from +/-7days of peak creatinine” when asked how they could exclude AKI when calculating baseline creatinine
  • 100% would not include patients on hemodialysis or peritoneal dialysis when asked which patient groups to exclude from routine AKI analysis
  • 80.0% would define community-acquired AKI as being present upon hospital admission
  • 85.0% would define lack of AKI biochemical recovery as having both ongoing need for renal replacement therapy and persistent peak AKI stage
  • 80.0% would not base chronic kidney disease progression on just 1 measure of CKD function in the aftermath of AKI

“We found a lack of consistency in application of the KDIGO definition for AKI to analyze routinely collected health care datasets, and a lack of transparency in reporting the definition utilized,” the authors wrote. “We have summarized areas where the panel was able to reach consensus discussed, and these have been used to make recommendations for reporting of future research.”

Going forward, the study investigators recommend the following when defining AKI via the KDIGO definition:

  1. Report whenever including that creatinine rose 0.3 mg/dL in 48 hours and if stage I, II, or III criteria are used
  2. Define timeframe for development of AKI
  3. When defining baseline kidney function include timeframe and if inpatient and outpatient creatinine data were used; if baseline creatinine was excluded, state what was used in its place
  4. State which patient groups, if any, were excluded from analysis
  5. Extend KDIGO guidelines to include a standard process for how to define AKI when using routinely collected data

“This review highlights the importance of methodological transparency in studies involving the identification of AKI from routinely collected biochemical data,” the authors concluded. “Future consensus studies would be usefully informed by research to explore the implications of different definitions on the estimation of AKI incidence, and associations with AKI outcomes of AKI identified by different definitions.”


Guthrie G, Buthrie B, Walker H, et al. Developing an AKI consensus definition for database research: findings from a scoping review and expert opinion using a Delphi process. Am J Kidney Dis. Published online July 20, 2021. doi:10.1053/j.ajkd.2021.05.019