Background-The American College of Cardiology and the American Heart Association have issued guidelines for the use of coronary artery bypass graft surgery (CABG) and percutaneous coronary interventions (PCI) for many years, but little is known about the impact of these evidence-based guidelines on referral decisions. Methods and Results-A cardiac catheterization laboratory database used by 19 hospitals in New York State was used to identify treatment (CABG surgery, PCI, medical treatment, or nothing) recommended by the catheterization laboratory cardiologist for patients undergoing catheterization with asymptomatic/mild angina, stable angina, and unstable angina/non-ST-elevation myocardial infarction between January 1, 2005, and August 31, 2007. The recommended treatment was compared with indications for these patients based on American College of Cardiology/American Heart Association guidelines. Of the 16 142 patients undergoing catheterization who were found to have coronary artery disease, the catheterization laboratory cardiologist was the final source of recommendation for 10 333 patients (64%). Of these 10 333 patients, 13% had indications for CABG surgery, 59% for PCI, and 17% for both CABG surgery and PCI. Of the patients who had indications for CABG surgery, 53% were recommended for CABG and 34% for PCI. Of the patients with indications for PCI, 94% were recommended for PCI. For the patients who had indications for both CABG surgery and PCI, 93% were recommended for PCI and 5% for CABG surgery. Catheterization laboratory cardiologists in hospitals with PCI capability were more likely to recommend patients for PCI than hospitals in which only catheterization was performed. Conclusions-Patients with coronary artery disease receive more recommendations for PCI and fewer recommendations for CABG surgery than indicated in the

For patients without ACS/prior CABG, only 1% of patients undergoing CABG surgery who could be rated were found to be inappropriate for the procedure according to the ACCF appropriateness criteria, but 14% of the PCI patients who could be rated were found to be inappropriate, and 28% lacked enough noninvasive test information to be rated.