Objective Acute kidney injury (AKI) affects 5–7% of all hospitalized patients with a much higher incidence in the critically ill. The Acute Kidney Injury Network proposed a definition in which serum creatinine rises (>0.3mg/dl) and/or oliguria (<0.5/ml/kg/h) for a period of 6 hours are used to detect AKI. Accurate urine output measurements as well as serum creatinine values from our database were used to detect patients with AKI and calculate their corresponding mortality risk and length of stay. Design Retrospective cohort study. Setting 7 intensive care units at, a large, academic, tertiary medical center. Patients Adult patients without evidence of end stage renal disease, with more than 2 creatinine measurements and at least a 6 hours urine output recording, who were admitted to the ICU between 2001 and 2007. Interventions Medical records of all the patients were reviewed. Demographic information, lab results, charted data, discharge diagnoses, physiological data and patient outcomes were extracted from the MIMIC-II database using a SQL query. Measurements and main results From 19,677 adult patient records, 14,524 patients met the inclusion criteria. 57% developed AKI during their ICU stay. In-hospital mortality rates were: 13.9%, 16.4%, 33.8% for AKI 1, 2 and 3 respectively compared to only 6.2% in patients without AKI (p<0.0001). After adjusting for multiple covariates AKI was associated with increased hospital mortality (OR 1.4 and 1.3 for AKI1 and AKI2 and 2.5 for AKI3; p<0.0001). Using multivariate logistic regression, we found that in patients who developed AKI, urine output alone was a better mortality predictor than creatinine alone or the combination of both. Conclusions More than 50% of our critically ill patients developed some stage of AKI resulting in stage-wise increased mortality risk. However, the mortality risk associated with AKI stages 1 and 2 does not differ significantly. In light of these findings reevaluation of the AKIN staging criteria should be considered.