Timing of renal replacement therapy initiation in acute renal failure: a meta-analysis.

BACKGROUND Some studies have suggested that early institution of renal replacement therapy (RRT) might be associated with improved outcomes in patients with acute renal failure (ARF). STUDY DESIGN A systematic review and meta-analysis of randomized controlled trials and cohort comparative studies to assess the effect of early RRT on mortality in patients with ARF. SETTING & POPULATION Hospitalized adult patients with ARF. SELECTION CRITERIA FOR STUDIES We searched several databases for studies that compared the effect of "early" and "late" RRT initiation on mortality in patients with ARF. We included studies of various designs. INTERVENTION Early RRT as defined in the individual studies. OUTCOMES The primary outcome measure was the effect of early RRT on mortality stratified by study design. The pooled risk ratio (RR) for mortality was compiled using a random-effects model. Heterogeneity was evaluated by means of subgroup analysis and meta-regression. RESULTS We identified 23 studies (5 randomized or quasi-randomized controlled trials, 1 prospective and 16 retrospective comparative cohort studies, and 1 single-arm study with a historic control group). By using meta-analysis of randomized trials, early RRT was associated with a nonsignificant 36% mortality risk reduction (RR, 0.64; 95% confidence interval, 0.40 to 1.05; P = 0.08). Conversely, in cohort studies, early RRT was associated with a statistically significant 28% mortality risk reduction (RR, 0.72; 95% confidence interval, 0.64 to 0.82; P < 0.001). The overall test for heterogeneity among cohort studies was significant (P = 0.005). Meta-regression yielded no significant associations; however, early dialysis therapy was associated more strongly with lower mortality in smaller studies (n < 100) by means of subgroup analysis. LIMITATIONS Paucity of randomized controlled trials, use of variable definitions of early RRT, and publication bias preclude definitive conclusions. CONCLUSION This hypothesis-generating meta-analysis suggests that early initiation of RRT in patients with ARF might be associated with improved survival, calling for an adequately powered randomized controlled trial to address this question.