Bypass Surgery or Stenting for Left Main Coronary Artery Disease in Patients With Diabetes

Milan Milojevic, Patrick W. Serruys, Joseph F. Sabik, David E. Kandzari, Erick Schampaert, Ad J. van Boven, Ferenc Horkay, Imre Ungi, Samer Mansour, Adrian P. Banning, David P. Taggart, Manel Sabaté, Anthony H. Gershlick, Andrzej Bochenek, Jose Pomar, Nicholas J. Lembo, Nicolas Noiseux, John D. Puskas, Aaron Crowley, Ioanna KosmidouRoxana Mehran, Ori Ben-Yehuda, Philippe Généreux, Stuart J. Pocock, Charles A. Simonton, Gregg W. Stone, Arie Pieter Kappetein

Research output: Contribution to journalArticlepeer-review

55 Scopus citations


Background: The randomized EXCEL (Evaluation of XIENCE versus Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularization) trial reported a similar rate of the 3-year composite primary endpoint of death, myocardial infarction (MI), or stroke in patients with left main coronary artery disease (LMCAD) and site-assessed low or intermediate SYNTAX scores treated with percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG). Whether these results are consistent in high-risk patients with diabetes, who have fared relatively better with CABG in most prior trials, is unknown. Objectives: In this pre-specified subgroup analysis from the EXCEL trial, the authors sought to examine the effect of diabetes in patients with LMCAD treated with PCI versus CABG. Methods: Patients (N = 1,905) with LMCAD and site-assessed low or intermediate CAD complexity (SYNTAX scores ≤32) were randomized 1:1 to PCI with everolimus-eluting stents versus CABG, stratified by the presence of diabetes. The primary endpoint was the rate of a composite of all-cause death, stroke, or MI at 3 years. Outcomes were examined in patients with (n = 554) and without (n = 1,350) diabetes. Results: The 3-year composite primary endpoint was significantly higher in diabetic compared with nondiabetic patients (20.0% vs. 12.9%; p < 0.001). The rate of the 3-year primary endpoint was similar after treatment with PCI and CABG in diabetic patients (20.7% vs. 19.3%, respectively; hazard ratio: 1.03; 95% confidence interval: 0.71 to 1.50; p = 0.87) and nondiabetic patients (12.9% vs. 12.9%, respectively; hazard ratio: 0.98; 95% confidence interval: 0.73 to 1.32; p = 0.89). All-cause death at 3 years occurred in 13.6% of PCI and 9.0% of CABG patients (p = 0.046), although no significant interaction was present between diabetes status and treatment for all-cause death (p = 0.22) or other endpoints, including the 3-year primary endpoint (p = 0.82) or the major secondary endpoints of death, MI, or stroke at 30 days (p = 0.61) or death, MI, stroke, or ischemia-driven revascularization at 3 years (p = 0.65). Conclusions: In the EXCEL trial, the relative 30-day and 3-year outcomes of PCI with everolimus-eluting stents versus CABG were consistent in diabetic and nondiabetic patients with LMCAD and site-assessed low or intermediate SYNTAX scores.(Evaluation

Original languageEnglish
Pages (from-to)1616-1628
Number of pages13
JournalJournal of the American College of Cardiology
Issue number13
StatePublished - 9 Apr 2019


  • SYNTAX score
  • coronary artery bypass grafting
  • diabetes
  • left main disease
  • percutaneous coronary intervention


Dive into the research topics of 'Bypass Surgery or Stenting for Left Main Coronary Artery Disease in Patients With Diabetes'. Together they form a unique fingerprint.

Cite this