Left Main Coronary Artery Disease Revascularization According to the SYNTAX Score: Analysis from the EXCEL Trial

Background: The SYNTAX (Synergy Between Percutaneous Coronary Intervention With Taxus and Cardiac Surgery) score (SS), a measure of anatomic coronary artery disease (CAD) extent and complexity, has proven useful in past studies to determine the absolute and relative prognosis after revascularization with percutaneous coronary intervention (PCI) versus coronary artery bypass grafting (CABG). We sought to assess contemporary outcomes after PCI and CABG in patients with left main CAD according to SS and revascularization type from a large randomized trial. Methods: The EXCEL trial (Evaluation of XIENCE Versus Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularization) randomized patients with left main CAD and site-assessed SS≤32 to PCI with everolimus-eluting stents or CABG. Four-year outcomes were examined according to angiographic core laboratory-assessed SS using multivariable Cox proportional hazards regression. Results: A total of 1840 patients with left main CAD randomized to PCI (n=914) versus CABG (n=926) had angiographic core laboratory SS assessment. The mean SS was 26.5±9.3 (range 5-74); 24.1% of patients had angiographic core laboratory-assessed SS ≥33. The 4-year rate of the primary major adverse cardiac event end point of death, stroke, or myocardial infarction was similar between PCI and CABG (18.6% versus 16.7%, respectively; P=0.40) and did not vary according to SS (P_interaction=0.33). Rates of ischemia-driven revascularization rose with increasing SS after PCI, but not after CABG. As a result, the major secondary composite end point of major adverse cardiac or cerebrovascular events (major adverse cardiac event or ischemia-driven revascularization) occurred more frequently with PCI than CABG (28.0% versus 22.0%, P=0.01), a difference which rose progressively with increasing SS (P_interaction=0.03). Conclusions: In the EXCEL trial, the 4-year primary composite major adverse cardiac event end point of death, myocardial infarction, or stroke was similar after PCI with everolimus-eluting stents and CABG and was independent of the baseline anatomic complexity and extent of CAD. In contrast, the relative and absolute hazard of major adverse cardiac or cerebrovascular events with PCI compared with CABG rose progressively with the SS. These data should be considered by the heart team when deciding between PCI versus CABG for revascularization in patients with left main CAD.