Meta-Analysis Comparing the Risk of Myocardial Infarction Following Coronary Artery Bypass Grafting Versus Percutaneous Coronary Intervention in Patients With Multivessel or Left Main Coronary Artery Disease

Revascularization strategies in patients with multivessel coronary artery disease: a Bayesian network meta-analysis

Treatment modalities for multivessel disease have rapidly evolved, yet the preferred strategy remains controversial. This meta-analysis compared outcomes after on-pump (ONCAB), off-pump coronary artery bypass grafting (OPCAB), percutaneous coronary intervention (PCI) or hybrid coronary revascularization. A comprehensive search for observational studies and randomized controlled trials published by August 2020 was performed. A Bayesian network meta-analysis was conducted for early (<30 days) and late (>12 months) outcomes. A total of 119 studies were included (n = 700 458 patients). The main analysis was confined to 31 randomized controlled trials (n = 24 932 patients). PCI was associated with lower early mortality [odds ratio (OR) 0.50, 95% confidence interval (CI) 0.31-0.79] and stroke (OR 0.22, 95% CI 0.06-0.60) rates compared with ONCAB, whereas a reduced risk of early myocardial infarction was observed with OPCAB compared with ONCAB (OR 0.53, 95% CI 0.32-0.83). Late target vessel revascularization and major adverse cardiac and cerebrovascular events were both increased with PCI compared with ONCAB, OPCAB and hybrid coronary revascularization (by 127-203% and 59-64%, respectively), and late major adverse cardiac events were increased in PCI compared with ONCAB and OPCAB (by 64% and 59%). However, PCI was associated with a significantly lower risk of late stroke compared with ONCAB (OR 0.70, 95% CI 0.52-0.89). Sensitivity analyses (i) including observational studies and (ii) limiting to studies with recent cohorts confirmed the findings of the main analysis. Surgical approaches for revascularization remain superior to PCI in patients with multivessel disease. Hybrid coronary revascularization might be viable for some patients, although more evidence from randomized controlled trials is warranted.