Spontaneous Myocardial Infarction After Left Main Revascularization: The EXCEL Trial.
Academic Article
Overview
abstract
BACKGROUND: Limited data are available regarding the relative rates, etiology, and long-term prognostic implications of spontaneous myocardial infarction (MI) after percutaneous coronary intervention (PCI) versus coronary artery bypass graft (CABG) surgery for left main coronary artery disease (LMCAD). METHODS: MIs after PCI and CABG for LMCAD were adjudicated from the EXCEL trial (Evaluation of Xience Versus Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularization). Cox proportional hazards regression was performed to assess the association between spontaneous (and procedural) MI and cardiovascular and all-cause mortality at 5 years. RESULTS: Among 1882 patients who underwent LMCAD revascularization, spontaneous MI during 5-year follow-up occurred in 60 (6.8%) patients after PCI and in 29 (3.4%) patients after CABG (adjusted hazard ratio [adjHR], 2.01; 95 CI, 1.29-3.15; P=0.002). By multivariable analysis, spontaneous MI (as a time-adjusted covariate) was a strong independent predictor of subsequent cardiovascular mortality (adjHR, 9.39; 95% CI, 5.22-16.87) and all-cause mortality (adjHR, 4.77; 95% CI, 2.92-7.80) within 5 years, with consistent effects after PCI and CABG (Pinteraction=0.60 and 0.78, respectively). In the same models, procedural MI as defined by extensive myonecrosis was associated with 5-year cardiovascular (adjHR, 3.02; 95% CI, 1.64-5.56) and all-cause mortality (adjHR, 2.38; 95% CI, 1.48-3.80), with consistent effects after PCI and CABG (Pinteraction=0.23 and 0.34, respectively). CONCLUSIONS: In the EXCEL trial, spontaneous MI occurred relatively infrequently within 5 years after LMCAD revascularization but at a higher rate after PCI compared with CABG. Spontaneous MI after revascularization was strongly related to subsequent cardiovascular and all-cause mortality, consistently after PCI and CABG, and was more strongly associated with mortality than was large procedural MI.
