Author + information
- Erick Schampaert1,
- Patrick Serruys2,
- A. Pieter Kappetein3,
- Nicholas Lembo4,
- William Brown5,
- Adrian Banning6,
- Ovidiu Dressler7,
- Serge Doucet8,
- Uday Trivedi9,
- David Hildick-Smith9,
- Joseph Sabik10 and
- Gregg Stone11
- 1Hopital du Sacre-Coeur de Montreal, Montreal, Quebec, Canada
- 2Imperial College, London, United Kingdom
- 3Department of Cardiothoracic Surgery, Erasmus University Medical Center, Rotterdam, Netherlands
- 4Columbia University, New York, New York, United States
- 5Piedmont Heart Institute CardioThoracic Surgeons, Atlanta, Georgia, United States
- 6John Radcliffe Hospital, Oxford, United Kingdom
- 7Cardiovascular Research Foundation, New York, New York, United States
- 8MHI, Montreal, Quebec, Canada
- 9Royal Sussex County Hospital, Brighton, United Kingdom
- 10Department of Thoracic and Cardiovascular Surgery, The Cleveland Clinic Foundation, Cleveland, Ohio, United States
- 11Cardiovascular Research Foundation, Columbia University Medical Center/NewYork-Presbyterian Hospital, New York, New York, United States
In the large-scale international, multicenter, randomized EXCEL trial, PCI with everolimus-eluting stents (EES) was non-inferior to CABG for the treatment of pts with left main (LM) coronary artery disease and SYNTAX score ≤32. Given variability in SYNTAX score assessment, the non-LM number of diseased coronary arteries (NDV; 0, 1, 2, or 3) may be a simpler way to discriminate groups with different outcomes with PCI vs CABG.
The primary endpoint was 3-year (3Y) major adverse cardiac events (MACE), a composite of death, stroke, or MI. Major secondary endpoints included 30D MACE and 3Y ischemia-driven revascularization (IDR). A pre-specified analysis according to the angiographic core laboratory (ACL) NDV (SYNTAX criteria - lesions with ≥50% DS beyond the LM [ie, excluding the ostial LAD and ostial LCX]) was performed.
1905 LM pts were randomized to EES (n=948) or CABG (n=957); NDV was assessed by the ACL in 1852 pts; 329 (17.8%) had LM+0VD, 579 (31.3%) had LM+1VD, 608 (32.8%) had LM +2VD and 336 (18.1%) had LM+3VD, each group equally distributed between PCI and CABG. 3Y MACE rate for PCI vs. CABG were non-significantly different in each group with no interaction present (p=0.75). Results in simpler (LM+0/1VD) vs. more complex (LM+2/3VD) disease appear in the figure.
Pts with LM+0/1VD may preferentially be treated by EES. More complex LM pts with 2/3 VD should undergo heart team evaluation to carefully consider the trade-offs of reduced early MACE with PCI compared to less late revascularization after CABG.
CORONARY: PCI Outcomes