PCI Versus CABG in Left Main and Multivessel Disease: Do We Still Have the Gordian Knot?

Abstract

The treatment of multivessel coronary artery disease (CAD) has been investigated during the last 30 years. Improvements in the surgical techniques and the evolution of several generations of bare metal stent (BMS) and drug eluting stents (DES) showed that this is a dynamic field. The continuing interest proves that there is still no absolute solution for revascularization, demonstrating the complexity (the “Gordian knot”) of the treatment of CAD. In their interesting meta-analysis, published in a previous issue of Angiology, Dr Verdoia and coworkers compared the outcomes of percutaneous coronary intervention (PCI) versus coronary artery bypass graft (CABG) in both multivessel CAD and unprotected left main disease (LMD) in the era of DES. They reported the potential superiority of CABG compared with PCI in multivessel CAD regarding major ischemic events. However, this was at the expense of an increased stroke incidence. No difference in survival was observed, especially in patients with isolated LMD. The authors included 8 randomized trials in their analysis, in which both firstand second-generation DES were used. In fact, second-generation DES were only used in 3 of the 8 studies. Thus, randomized studies with new-generation DES are warranted. In this meta-analysis, the benefit in both mortality and major adverse cardiovascular events in patients with multivessel CAD is influenced by the Future Revascularization Evaluation in Patients with Diabetes Mellitus: Optimal Management of Multivessel Disease (FREEDOM) and Coronary Atery Revascularization in Diabetes (VA CARD) studies, which only included patients with diabetes mellitus. Interestingly, the most recent trial in patients with multivessel CAD, the Randomized Comparison of Coronary Artery Bypass Surgery and Everolimus–Eluting Stent Implantation in the Treatment of Patients with Multivessel Coronary Artery Disease (BEST) trial, conducted with second-generation DES, also included a higher proportion of patients with diabetes than have most other studies investigating DES (usually <30%). However, the mortality benefit of CABG has also been confirmed by the Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery trial (SYNTAX), where only paclitaxel-eluting stents were implanted; these stents are now off the market. In the majority of randomized studies, comparing PCI versus CABG, there is no distinction between periprocedural and nonperiprocedural events in the per protocol definition of myocardial infarction (MI) and stroke. More specifically, in all but the Nordic-Baltic-British Left Main Revascularization Study trial (NOBLE), the per protocol definition of events included periprocedural MI. In this trial, a significant advantage of CABG in the incidence of MI was shown (7% vs 2%, P 1⁄4 .004). In the more recent randomized Evaluation of the Xience Everolimus Eluting Stent vs Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularization (EXCEL) study, the separate analysis of 30-day MI incidence shows a clear superiority of PCI (3.9% vs 6.2%, P 1⁄4 .02), regarding LMD revascularization, with the difference driven by the incidence of periprocedural MI (3.6% vs 5.9%, P 1⁄4 .02). Interestingly, no difference in 30-day or 3-year stroke incidence between the 2 procedures was demonstrated in this contemporary trial. Thus, more studies with these prespecified end points are needed. Another interesting point, not adequately addressed in the randomized studies, is the vascular access used in the PCI group. In most studies, except for Boudriot et al, both accesses were allowed. However, as PCIs were complex in high-risk populations, radial access was underutilized. In the EXCEL trial, only 27% of the PCIs were accomplished radially. However, recent small-scale trials show the feasibility of complex unprotected LMD procedures, through 7F sheaths, with high success rates. Taking into account the mortality benefit of radial versus femoral access recently shown in Minimizing Adverse Haemorrhagic Events by TRansradial Access Site and Systemic Implementation of angioX (MATRIX) trial, the advantages of PCI in these high-risk procedures could be underestimated.