RESUMO
OBJECTIVE: To determine the incidence of acute renal failure (ARF) in critically ill patients using the RIFLE and AKIN criteria. DESIGN: A prospective, multicenter observational study with a duration of one year from February 2010 was carried out. RIFLE and AKIN were employed using the urinary (UC) and creatinine criteria (CC) jointly and separately. SCOPE: Nine polyvalent Critical Care Units (CCUs) in Argentina. PATIENTS: A total of 627 critical patients over 18 years of age were admitted to the CCU for more than 48h. EXCLUSION CRITERIA: inability to quantify diuresis, surgical instrumentation of the urinary tract, and need for renal support therapy (RST). VARIABLES OF INTEREST: Calculated hourly diuresis (CHD) was used to apply the UC. RESULTS: The incidence of ARF was 69.4% and 51.8% according to RIFLE and AKIN, respectively. UC detected ARF in 59.5% of cases, while CC identified ARF in 34.7% (RIFLE) and 25.3% (AKIN). The mortality rate was 40.9% and 44.6% according to RIFLE and AKIN respectively, was significantly higher than in patients without ARF, and increased with disease severity (Data processing: Excel, SQL and SPSS. Levene test, comparison of means with Student t and chi-squared, with 95% confidence interval). CONCLUSIONS: RIFLE identified more cases of ARF. UC proved more effective than CC. The presence of ARF and severity levels were correlated to mortality but not to days of stay in the CCU. Implementation of the unified CHD was useful for implementing UC and achieving comparable results.