Randomized evidence on graft patency after off-pump versus on-pump coronary artery bypass grafting: An updated meta-analysis.

BACKGROUND: The debate between off-pump CABG (OPCAB) and on-pump CABG (ONCAB) has been ongoing for decades. We aimed to provide a comprehensive update of the current randomized controlled trials (RCTs) in evaluating the graft patency of OPCAB versus ONCAB. MATERIALS AND METHODS: A literature search was conducted in PubMed, EMBASE, and the Cochrane Library databases until April 30, 2021. All RCTs from 2003 to 2020 comparing the results of graft patency between OPCAB and ONCAB were included. We compared the overall graft occlusion between the two groups, and subgroup analyses were conducted based on different types of conduits and target territories, crossover from off-pump to on-pump rate, and the length of follow-up. RESULTS: Sixteen RCTs were identified, with 5743 grafts in the OPCAB group and 5898 in the ONCAB group. OPCAB was associated with a higher risk of occlusion in the overall graft (RR: 1.31; 95% CI, 1.17-1.46), saphenous vein graft (SVG) (RR: 1.40; 95% CI, 1.23-1.59), grafts to left anterior descending (LAD) territory (RR: 1.52; 95% CI, 1.11-2.08) and left circumflex artery (LCX) territory (RR: 1.45; 95% CI, 1.19-1.76), while no significant difference was observed between the two groups in respect of arterial conduits and grafts to right coronary artery (RCA) territory. Furthermore, the lower crossover rate and longer length of follow-up appeared to reduce the association between OPCAB and lower graft patency. CONCLUSIONS: The current meta-analysis indicates that, compared with ONCAB, graft patency is poorer with OPCAB for overall grafts, SVG grafts, grafts to LAD and LCX territories, whereas the results remain comparable for arterial conduits and grafts to RCA territory.