PurposeEarly renal replacement therapy (RRT) initiation should theoretically influence many physiological disorders related to acute kidney injury (AKI). Currently, there is no consensus about RRT timing in intensive care unit (ICU) patients.MethodsWe performed a retrospective analysis of all critically ill patients who received RRT in our ICU during a 3 year-period. Our goal was to identify mortality risk factors and if RRT initiation timing had an impact on survival. RRT timing was calculated from the moment the patient was classified as having acute kidney injury in the RIFLE classification.ResultsA hundred and ten patients received RRT. We identified four independent mortality risk factors: need for mechanical ventilation (OR = 12.82 (1.305 - 125.868, p = 0.0286); RRT initiation timing >16 h (OR = 5.66 (1.954 - 16.351), p = 0.0014); urine output on admission <500 ml/day (OR = 4.52 (1.666 - 12.251), p = 0.003); and SAPS II on admission >70 (OR = 3.45 (1.216 - 9.815), p = 0.02). The RRT initiation <16 h and RRT initiation >16 h groups presented the same baseline characteristics, except for more severe gravity scores and kidney failure in the early RRT group.ConclusionsEarly RRT in ICU patients with acute kidney injury or failure was associated with increased survival.