Abstract: Objective: To investigate the occurrence of acute kidney injury（AKI）in hospitalized patients, and to compare the detection rate by real-time electronic alert system with the results by manual diagnosis. Methods: A total of 2 563 hospitalized patients in Zhongshan Hospital, Fudan University from November 3rd, 2014 to November 9th, 2014 were screened by real-time electronic alert system and the results of manual diagnosis were compared. Results: Detection rates by electronic alert system and manual diagnosis were 10.2% and 3.59%, respectively. The median age of AKI patients was (63±16) years, and among them 176 cases (67.4%) were male and 17 cases（6.5%） received renal replacement therapy; the hospital mortality of AKI patients was 5.7%. Of them 93.5% of patients were not from Department of Nephrology, and only 9.8% received consultation by nephrologists. Only 1.5% had discharge diagnosis of AKI related to acute kidney injury. The Departments with the high incidence of AKI were Cardiac Surgery(39.0%), nephrology(33.0%) and Liver Surgery (19.0%). The incidence of community acquired AKI (CA-AKI) and hospital-acquired AKI (HA-AKI) were 26.1% and 73.9%, respectively. The AKI group had significantly longer hospital stay and higher expenses than the non-AKI group (P<0.05). Of AKI patients, AKI stage(KDIGO criteria) and age were independent risk factors of in-hospital mortality according to the results of multivariate logistic regression. Conclusions: The majority of AKI patients are not from Department of Nephrology and has high missed diagnosis rate and low consultation rate by nephrologists. The establishment of AKI electronic alerting system can significantly increase the recognition rate of AKI patients, raise doctors’ awareness of AKI, and may help to improve the prognosis of AKI patients.

Key words: Acute kidney injury, Incidence, Electronic alert system, Manual statistics