Applications for detection of acute kidney injury using electronic medical records and clinical information systems: workgroup statements from the 15th ADQI Consensus Conference

Table 2 Features that may influence the performance of automated AKI alerts based on the KDIGO AKI criteria

KDIGO AKI criteria	Feature
Serum creatinine	Calibration of measure according to IDMS standard
	Optimal measurement using enzymatic assay [15]
	Comparison across laboratories or measurement techniques [16]
	Relevancy of e-alert systems using estimated baseline creatinine
	If previous creatinine available, chosen definition of baseline creatinine
	Management of outliers measures
	Significance of small changes in serum creatinine in patients with low weight/body surface or with pre-existing CKD
	Performance of e-alert system in unselected population of patients.
	Management of multiple alert in a same patient
	Influence of fluid balance/dilution [17].
Urine output	Difference in measurement according to setting (ICU vs. Ward, Specificity of paediatric units, rate of Foley catheter use).
	Management of missing data
	Errors in reading [8]
	Errors related to manual entry of urinary output
	Differences related to measurement (hourly vs. by shift vs. daily)
	Recognition of the lack of specificity of oliguria [18–20]
	Cross-tabulation between serum creatinine and UO

ISSN: 2054-3581