Epidemiology of emergency department acute kidney injury

Abstract

Aim: The epidemiology of acute kidney injury (AKI) diagnosed in the emergency department (ED) is poorly described. This study describes the incidence, demographics and outcomes of patients diagnosed with AKI in the ED (ED-AKI).

Methods: A prospective cohort study was completed in a University Teaching Hospital, (UK) between April and August 2016. In total, 20 421 adult patients attended the ED and had a serum creatinine measurement. The incident ED-AKI patient episodes were compared with a randomly selected cohort of non-AKI ED patients.

Results: A total of 572 patients had confirmed eAlert ED-AKI (548 incident cases), incidence 2.8% (of all ED attendances). ED-AKI was associated with a 24.4% in-patient mortality (non-AKI 3.2%, P < .001) of which 22.3% of deaths occurred within 24 hours and 58% within 7 days. Progression of the admission AKI stage to a higher AKI stage was associated with a 38.8% mortality compared with a 21.4% mortality in those who did not progress (P < .001). In multivariate analysis, ED-AKI was an independent risk for mortality (hazard ratio, 6.293; 95% confidence interval, 1.887-20.790, P = .003). For those discharged from hospital, 20.4% of ED-AKI patients re-attend for acute assessment within 30-days post-discharge (non-AKI 7.6%, P < .001). At 90-days post-discharge, 10.0% of ED-AKI patients died (non-AKI 1.4%, P < .001). Twelve months post-discharge 17.8% of ED-AKI patients developed CKD progression or de-novo CKD (non-AKI 6.0%).

Conclusion: The ED-AKI is an independent predictor of death. Mortality is predominantly in the early stages of hospital admission, but for those who survive to discharge have significant long-term morbidity and mortality.

Keywords: acute kidney injury; emergency departments; epidemiology.