Long-Term Outcomes Comparison Between Surgical and Percutaneous Coronary Revascularization in Patients With Multivessel Coronary Disease or Left Main Disease: A Systematic Review and Study Level Meta-Analysis of Randomized Trials

Abstract

Recent randomized trials comparing coronary artery bypass graft (CABG) with percutaneous coronary intervention (PCI) utilizing drug-eluting stents in patients with left main disease (LMD) and/or multivessel disease (MVD), reported conflicting results. We performed a study level meta-analysis comparing the 2 interventions for the treatment of LMD or MVD. Using electronic databases, we retrieved 6 trials, between January, 2010 and December, 2022. Five-years Kaplan-Meier curves of endpoints where reconstructed. Comparisons were made by cox-linear regression frailty model and by landmark analysis. A random-effect method was applied. A total of 8269 patients were included and randomly assigned to CABG (n = 4135) or PCI (n = 4134). During 5-years follow-up, PCI showed a higher incidence of all-cause mortality (hazard ratio [HR] 1.28; 95% confidence interval [CI], 1.11-1.47; P < 0.0001]), myocardial infarction (HR 1.84; 95% CI, 1.54-2.19; P < 0.0001) and repeat coronary revascularization (HR 1.96; 95% CI, 1.72-2.24; P < 0.0001). There was no long-term difference between the 2 interventions for cardiovascular death (P = 0.14) and stroke (P = 0.20), although the incidence of stroke was higher with CABG within 30-days from intervention (P < 0.0001). PCI was associated with an increased risk for composite endpoints (P < 0.0001) and major cerebral and cardiovascular events. (P < 0.0001). In conclusion, at 5-year follow-up, in patients with LMD and/or MVD there was a significant higher incidence of all-cause mortality, myocardial infarction and repeat revascularization with PCI compared to CABG. The incidence of stroke was higher with CABG during the postprocedural period, but no difference was found during 5-years follow-up. Longer follow-up is mandatory to better define outcome difference between the 2 interventions.