Searching for the second best graft for coronary artery bypass surgery: a network meta-analysis of randomized controlled trials

OBJECTIVES: There is a lack of unequivocal evidence basis for selecting the best second conduit in coronary artery bypass grafting (CABG). We thus aimed to perform head-to-head relative effect estimate on angiographic outcomes for second conduits, including the right internal mammary artery (RIMA), radial artery (RA), right gastroepiploic artery (RGEA) and saphenous vein graft (SVG) by means of network meta-analysis of randomized controlled trials (RCTs). METHODS: Databases were searched for RCTs comparing angiographic outcomes (>= 4 weeks) of second conduits in CABG. Odds ratios (95% confidence intervals) were computed with Markov Chain Monte Carlo simulation. RESULTS: A total of nine RCTs were identified, including 2780 patients and 1620 angiographic results available for analysis to compare RIMA (n = 145) versus RA (n = 871) versus RGEA (n = 92) versus SVG (n = 845). The mean time to angiographic follow-up ranged from 1 to 7.7 years. An SVG was significantly associated with a 4-fold (1.67-16.00) and 3-fold (0.78-22.20) increased risk of late (>= 4 years) functional graft occlusion when compared with the RIMA and RA, respectively. A RIMA was associated with a non-significant 27% absolute risk reduction for functional graft occlusion when compared with the RA. CONCLUSIONS: The present network meta-analysis consistently demonstrated an angiographic superiority of RIMA and RA over SVG. The RIMA is expected to achieve a better patency rate than the RA, but further studies are needed.