- For left main (LM) coronary artery disease (CAD), percutaneous coronary intervention (PCI) using everolimus-eluting stents (EESs) and coronary artery bypass grafting (CABG) yield similar event rates.
- In EXCEL trial, results were similar for primary composite endpoint (death, stroke, or myocardial infarction [MI]) at 3-year follow-up in patients with and without diabetes mellitus (DM) with LMCAD and site assessed low-to-intermediate (≤32) SYNTAX scores.
Why this matters
- Whether EXCEL results apply to patients with diabetes, who typically do better with CABG vs PCI, was unclear.
- Patients (n=1905) with LMCAD and site-assessed low or intermediate CAD complexity (SYNTAX scores ≤32) randomly assigned 1:1 to PCI with EES vs CABG, stratified by the presence of diabetes (n=554 with, 1350 without).
- Funding: Abbott Vascular.
- Patients with diabetes had a significantly higher 3-year primary endpoint rate vs those without: HR, 1.60 (95% CI, 1.26-2.04).
- At 3 years, primary endpoint rate was similar after PCI and CABG in those with diabetes (1.03; 0.71-1.50) and without (0.98; 0.73-1.32).
- No significant interaction between diabetes status and revascularization type for primary endpoint (Pinteraction=.82) or secondary endpoints: death (P=.22), stroke (P=.17), MI (P=.99), repeat revascularization (P=.68), or composite of death, MI, stroke, or repeat revascularization (P=.65).
- Small subgroup sizes, short follow-up.