Purpose: The present observational study compares in-hospital and 12-month clinical outcomes in elderly patients with unprotected left main coronary artery disease treated either with coronary artery bypass grafting or drug-eluting stent.
Methods: From January 2004 to December 2007, 211 patients (pts) with unprotected left main coronary artery (ULMCA) stenosis, aged 75 or older, underwent coronary revascularization either with coronary artery bypass graft (CABG) (106 pts) or drug-eluting stent (DES) (105 pts). The decision to treat with CABG or percutaneous coronary intervention (PCI) was dependent on the patient's and the physician's choice. The occurrence of major adverse cardiac or cerebrovascular events (MACCE: death, nonfatal myocardial infarction, or stroke) and revascularizations was recorded after 1 year of follow-up. A multivariate logistic regression analysis was performed using a propensity score method to take potential baseline differences between groups into account.
Results: In-hospital MACCE rates were 5.7% and 3.8% in the CABG and PCI groups, respectively (P = 0.748). After 1 year of follow-up, these rates were, respectively, 13.9% and 14.9% (P = 0.841), and rates for target vessel revascularization at 12 months were 1.0% and 13.9% (P < 0.001). The PCI group was significantly associated with older age, dyslipidemia, history of cancer, high Euroscore, elevated creatininemia, single-vessel disease, fewer chronic occlusions of the left anterior descending artery, and more LMCA stenosis >or=70%. The multivariate logistic regression analysis was adjusted for age, diabetes, left ventricular ejection fraction, Euroscore, and plasma creatininemia and stratified on the score of propensity to be treated with PCI. In the subgroup below median propensity score, the adjusted odds ratio for 1-year MACCE was OR = 0.91 (95% confidence interval: 0.14 to 5.98; P = 0.924) whereas OR was 0.16 (0.04-0.69; P = 0.013) in the subgroup above median propensity score.
Conclusions: In patients with a high probability of being treated with PCI (older age, high Euroscore, high creatininemia, single-vessel disease, ...), the 1-year risk of MACCE was significantly lower in PCI- than in CABG-treated subjects. No significant difference was found in other cases.