Background: Treatment of patients with ULMCA (unprotected left main coronary artery disease) with percutaneous coronary intervention (PCI) has been compared with coronary artery bypass graft (CABG), without conclusive results.
Methods: All randomized controlled trials (RCTs) and observational studies with multivariate analysis comparing PCI and CABG for ULMCA were included. Major cardiovascular events (MACEs, composite of all-cause death, MI, definite or probable ST, target vessel revascularization and stroke) were the primary end points, whereas its single components were the secondary ones, along with stent thrombosis, graft occlusion and in-hospital death and stroke. Subgroup analyses were performed according to Syntax score.
Results: Six RCTs (4717 patients) and 20 observational studies with multivariate adjustment (14 597 patients) were included. After 5 (3-5.5) years, MACE rate was higher for PCI [odds ratio (OR) 1.10, 95% confidence interval (CI) 1.07-1.14], without difference in death, whereas more relevant risk of MI was because of observational studies. Coronary stenting increased risk of revascularization (OR 1.52; 95% CI 1.34-1.72). At meta-regression, performance of PCI was improved by use of intra-coronary imaging and worsened by first generation stents, whereas two arterial grafts increased benefit of CABG. For patients with Syntax score less than 22, MACE rates did not differ, whereas for higher values, CABG reduced MACE because of lower risk of revascularization. Incidence of graft occlusion was 3.24% (2.25-4.23), whereas 2.13% (1.28-2.98: all CI 95%) of patients experienced stent thrombosis.
Conclusion: Surgical revascularization reduces risk of revascularization for ULMCA patients, especially for those with Syntax score greater than 22, with a higher risk of in-hospital death. Intra-coronary imaging and use of arterial grafts improved performance of revascularization strategies.