PURPOSE OF REVIEW: Timing of renal replacement therapy in critically ill patients with acute kidney injury is highly subjective, and may influence outcome. We discuss renal and nonrenal criteria for timing considering the recent literature. RECENT FINDINGS: Two randomized and four nonrandomized controlled trials investigated the effects of timing on patient outcome. All but one randomized controlled trial indicated better outcome with early renal replacement therapy but had poor methodological quality. The heterogeneity of timing definition, study population and mode of therapy, however, hampered comparison among studies. SUMMARY: In the absence of large randomized controlled trials we can make no firm recommendations for timing of renal replacement therapy in acute kidney injury. Since rapid recovery of renal function is unlikely when other organ failure persists and the consequences of acute kidney injury may be more severe in critically ill patients, we suggest other organ failure is also considered. Patients with acute kidney injury, persisting shock and poorly recovering functions of other organs may benefit from early therapy. For future studies, we recommend describing renal replacement therapy timing according to the 'RIFLE' classification, as modified by the Acute Kidney Injury Network, and quantifying the severity of other organ failure. Biomarkers may refine acute kidney injury and timing definitions in the future.
PURPOSE OF REVIEW: Timing of renal replacement therapy in critically illpatients with acute kidney injury is highly subjective, and may influence outcome. We discuss renal and nonrenal criteria for timing considering the recent literature. RECENT FINDINGS: Two randomized and four nonrandomized controlled trials investigated the effects of timing on patient outcome. All but one randomized controlled trial indicated better outcome with early renal replacement therapy but had poor methodological quality. The heterogeneity of timing definition, study population and mode of therapy, however, hampered comparison among studies. SUMMARY: In the absence of large randomized controlled trials we can make no firm recommendations for timing of renal replacement therapy in acute kidney injury. Since rapid recovery of renal function is unlikely when other organ failure persists and the consequences of acute kidney injury may be more severe in critically illpatients, we suggest other organ failure is also considered. Patients with acute kidney injury, persisting shock and poorly recovering functions of other organs may benefit from early therapy. For future studies, we recommend describing renal replacement therapy timing according to the 'RIFLE' classification, as modified by the Acute Kidney Injury Network, and quantifying the severity of other organ failure. Biomarkers may refine acute kidney injury and timing definitions in the future.