Coronary artery bypass grafting versus percutaneous coronary intervention for patients with multivessel coronary artery disease and advanced heart failure

Coronary artery disease (CAD) is a leading cause of heart failure (HF), including HF with severely reduced left ventricular ejection fraction (LVEF ≤ 35%). While revascularization may improve outcomes, optimal treatment strategies remain uncertain, especially in patients with multivessel disease (MVD). To compare long-term outcomes of percutaneous coronary intervention (PCI) versus coronary artery bypass grafting (CABG) in a real-world cohort of patients with MVD and HF with LVEF ≤ 35%. In this prospective observational study, 586 patients with LVEF ≤ 35% were evaluated by a multidisciplinary Heart Team and assigned to PCI (n = 394) or CABG (n = 192). Outcomes over a 6-year of follow-up included all-cause mortality, hospitalization for HF, myocardial infarction (MI), stroke and need for repeat revascularization (RR). The primary endpoint (all-cause mortality or hospitalization for HF) at 6 years was comparable between both cohorts (56.8% vs. 64.7% for PCI, p = 0.06). We found no significant differences in overall rates of death between CABG and PCI (40.1% vs. 47.4% for PCI, p = 0.09) or hospitalization for HF (27.1% vs. 32.7% for PCI, p = 0.16), however these rates were lower in CABG-cohort. Individuals who underwent PCI had higher incidences of MI and RR (18.8% vs. 9.9% for CABG, p = 0.003 and 25.9% vs. 12.5% for CABG, p = 0.0002, respectively), while higher rates of strokes (9.4% vs. 4.8% for PCI, p = 0.03) penalized CABG. Furthermore, patients who underwent surgery experienced also longer postprocedural hospital stay (9.3 (3.1) vs. 2.8 (1.4) days, p < 0.001). In-hospital mortality was nonsignificantly higher in CABG-cohort (5.2% vs. 2.3% for PCI, p = 0.06). Over a 6-year of follow-up, for real-life all-comer patients with MVD and severe LV dysfunction, we found no significant differences between surgery or percutaneous approach - all-cause mortality and hospitalization for HF were similar, rates of MI and RR favour CABG, while patients who underwent PCI were burdened with lower incidence of stroke. Nevertheless, an increased perioperative risks in this cohort highlight the need for individualized, Heart Team-driven decisions. Further randomized-controlled trials to optimize management in this high-risk population are strongly desirable.