The Use of Automated Electronic Alerts in Studying Short-Term Outcomes Associated with Community-Acquired Acute Kidney Injury

Background/Aims: The use of electronic alerts (e-alerts) may increase the detection rate of acute kidney injury (AKI) since they are sensitive to small changes in serum creatinine. Our aim was to follow-up a cohort of patients presenting to hospital from the community with AKI (community-acquired AKI [c-AKI]), detected through the use of e-alerts, and describe their short-term outcomes regardless of whether they were subsequently admitted to hospital. Methods: Blood samples for all hospital attenders from the community either to the Accidents and Emergency department or one of the acute care areas of the hospital during a 6-month period (November 1, 2013-April 30, 2014) were screened for presence of c-AKI using a locally developed e-alerts system based on Kidney Disease: Improving Global Outcomes criteria. Follow-up data were obtained for a period of 3 months. Results: A total of 1,277 c-AKI episodes were identified in 1,185 patients (incidence 579 per 100,000 persons). Episodes that lead to hospitalization (n = 1,096 [86%]) were associated with a median length of hospital stay of 6.6 days; a graded increase in duration of stay was noted with increasing severity of AKI. Acute dialysis was needed during 21 (1.6%) episodes. For mortality rates, only the first AKI episode was considered. There were 298 deaths within 30 days of diagnosis irrespective of admissions status (30-day mortality rate: 25%). Conclusions: Using e-alerts in acute care settings to detect c-AKI is novel and may be used to stage and follow-up AKI using existing diagnostic criteria. c-AKI is relatively common and leads to significant mortality.