High-dose renal replacement therapy for acute kidney injury: Systematic review and meta-analysis

Objective: To determine the effect of renal replacement therapy dose on mortality and dialysis dependence in patients with acute kidney injury. Data Sources: MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials to October 2009; PubMed "Related Articles;" bibliographies of included trials; and additional information from trial authors. Study Selection: Randomized and quasi-randomized, controlled trials in adults with acute kidney injury prescribed high-vs. standard-dose continuous renal replacement therapy (>= 30 mL/kg/hr vs. <30 mL/kg/hr), intermittent hemodialysis, or sustained low-efficiency dialysis (daily vs. alternate day, or by target biochemistry). Data Extraction: Three authors independently selected studies and extracted data on outcomes and study quality. Meta-analyses used random-effects models. Data Synthesis: Of 5416 citations, 12 trials (n = 3999) met inclusion criteria. Modalities included continuous renal replacement therapy (7 trials), intermittent hemodialysis (3 trials), sustained low-efficiency dialysis (1 trial), and all three (1 trial). Study quality was moderate-high. Meta-analyses found no effect of high-dose renal replacement therapy on mortality (risk ratio, 0.89; 95% confidence interval, 0.77-1.03; 12 trials; n = 3954) or dialysis dependence among survivors (risk ratio, 1.15; 95% confidence interval, 0.92-1.44; 8 trials with events; n = 1743). The effect on mortality was similar (all interaction p values were nonsignificant) in patients with sepsis (risk ratio, 1.02; 95% confidence interval, 0.85-1.23; 9 trials; n = 1786) vs. without sepsis (risk ratio, 0.89; 95% confidence interval, 0.75-1.05; 8 trials; n = 1955), treated exclusively with continuous renal replacement therapy (risk ratio, 0.87; 95% confidence interval, 0.71-1.06; 7 trials; n = 2462) vs. other modalities alone or in combination (risk ratio, 0.92; 95% confidence interval, 0.70-1.21; 5 trials; n = 1492), and in trials with low (risk ratio, 0.96; 95% confidence interval, 0.85-1.09; 6 trials; n = 3475) vs. higher (risk ratio, 0.76; 95% confidence interval, 0.53-1.09; 6 trials; n = 479) risk of bias. Conclusions: High-dose renal replacement therapy in acute kidney injury does not improve patient survival or recovery of renal function overall or in important patient subgroups, including those with sepsis. (Crit Care Med 2010; 38:1360-1369)