Context: Acute renal failure requiring dialytic support is associated with a high risk of mortality and substantial morbidity.
Objectives: To summarize current evidence guiding provision of dialysis for patients with acute renal failure, to make recommendations for management, and to identify areas in which additional research is needed.
Data sources: Systematic searches of peer-reviewed publications in MEDLINE, EMBASE, and All EBM Reviews through October 2007.
Study selection: Randomized controlled trials (RCTs) and prospective cohort studies studying dialytic support in adults with acute renal failure that reported the incidence of clinical outcomes such as mortality, length of stay, need for chronic dialysis, or development of hypotension.
Data extraction: Quality was independently assessed by 2 reviewers using the Jadad score (RCTs) and the Downs and Black checklist (cohort studies). A single reviewer extracted data, which were independently verified by a second reviewer. Results of RCTs were pooled using a random-effects model.
Data synthesis: From 173 retrieved articles, 30 RCTs and 8 prospective cohort studies were eligible. No conclusions could be drawn about optimal indications for or timing of renal replacement. Available data comparing continuous renal replacement therapy (CRRT) with intermittent hemodialysis demonstrated no clinically relevant difference between modalities, including for all-cause mortality (relative risk [RR], 1.10; 95% confidence interval [CI], 0.99-1.23; I2 = 0%) or for the requirement for chronic dialysis treatment in survivors (RR, 0.91; 95% CI, 0.56-1.49; I2 = 0%). For patients treated with CRRT, limited data suggest that bicarbonate may be preferable to other forms of dialysate alkali and that citrate infusion may be an alternative to systemic anticoagulation in patients at high risk of bleeding. Among patients treated with continuous venovenous hemofiltration (CVVHF), the risk of death was lower at doses of 35 mL/kg per hour (RR of death compared with doses of 20 mL/kg per hour, 0.74; 95% CI, 0.63-0.88). The use of unsubstituted cellulosic membranes should be avoided in intermittent hemodialysis (RR of death compared with biocompatible membranes, 1.23; 95% CI, 1.01-1.50).
Conclusions: Based on current data, intermittent hemodialysis and CRRT appear to lead to similar clinical outcomes for patients with ARF. If CVVHF is used, a dose of 35 mL/kg per hour should be provided. Given the paucity of good-quality evidence in this important area, additional large randomized trials are needed to evaluate clinically important outcomes.