Off-pump coronary artery bypass surgery and acute kidney injury: a meta-analysis of randomized and observational studies.

BACKGROUND Acute kidney injury (AKI) after coronary artery bypass grafting (CABG) is associated with significant morbidity and mortality. Controversy exists regarding whether an off-pump technique can reduce post-CABG renal injury. STUDY DESIGN Systematic review and meta-analysis. SETTING & POPULATION Adult patients undergoing CABG. SELECTION CRITERIA FOR STUDIES MEDLINE, EMBASE, Cochrane Renal Library, and Google Scholar were searched in May 2008 for randomized controlled trials (RCTs) and observational studies comparing off-pump CABG (OPCAB) with conventional CABG (CAB) for renal outcomes. Studies involving patients on long-term renal replacement therapy (RRT) were excluded. INTERVENTION OPCAB. OUTCOMES Primary outcomes were overall AKI and AKI requiring RRT. RESULTS 22 studies (6 RCTs and 16 observational studies) comprising 27,806 patients met the inclusion criteria. The pooled effect from both study cohorts showed a significant reduction in overall AKI (odds ratio [OR], 0.57; 95% confidence interval [CI], 0.43 to 0.76; P for effect < 0.001; I(2) = 67%; P for heterogeneity < 0.001) and AKI requiring RRT (OR, 0.55; 95% CI, 0.43 to 0.71; P for effect < 0.001; I(2) = 0%; P for heterogeneity = 0.5) in the OPCAB group compared with the CAB group. In RCTs, overall AKI was significantly reduced in the OPCAB group (OR, 0.27; 95% CI, 0.13 to 0.54); however, no statistically significant difference was noted in AKI requiring RRT (OR, 0.31; 95% CI, 0.06 to 1.59). In the observational cohort, both overall AKI (OR, 0.61; 95% CI, 0.45 to 0.81) and AKI requiring RRT (OR, 0.54; 95% CI, 0.40 to 0.73) were significantly less in the OPCAB group. RCTs were noted to be underpowered and biased toward recruiting low-risk patients. Sensitivity analysis restricted to good-quality studies showed a significant reduction in AKI. LIMITATIONS Lack of uniform AKI definition in the included studies, heterogeneity for overall AKI outcome. CONCLUSIONS Analysis of the current evidence suggests a reduction in AKI using the OPCAB technique; however, studies lack consistency in defining AKI. Available RCTs are underpowered to detect a difference in AKI requiring RRT; evidence from observational studies suggests a reduction in RRT requirement. Future studies should apply a standard definition of AKI and target a high-risk population.