The goal of the trial was to evaluate stenting compared with surgical revascularization among patients with unprotected left main coronary artery disease undergoing elective revascularization.

Study Design

Study Design:

Patients Screened: 347
Patients Enrolled: 105
Mean Follow Up: 12 months
Mean Patient Age: 61 years
Female: 68
Mean Ejection Fraction: 53% at baseline

Patient Populations:

Severe proximal or distal narrowing of unprotected left main coronary artery with or without concomitant disease of other coronary arteries suitable for PCI or CABG in patients with stable or unstable angina


Contraindication to double anti-platelet therapy with aspirin and thienopyridines due to active bleeding, bleeding diathesis, or allergy to antiplatelets; non-cardiac disease that could limit long-term survival; previous CABG or PCI; acute myocardial infarction within 3 days; total left main coronary artery occlusion; co-morbid conditions or coronary anatomic considerations making patients poor canidates for revascularization

Primary Endpoints:

12 month: 1) left ventricular ejection fraction, 2) functional capacity on treadmill stress test, and 3) angina status

Secondary Endpoints:

MACE at 30 days and 1 year, defined as cardiac death, acute MI, heart failure, stroke, or repeat revascularization; MAE at 30 days and 1 year, defined as MACE or procedure related infection, bleeding, renal or respiratory insufficiency; length of hospital stay;

Drug/Procedures Used:

Patients with narrowing of unprotected left main coronary artery but without total occlusion were randomized in an open label manner following diagnositic coronary angiography to stenting (n=52) or CABG (n=53). Patients were followed for 12 months. Repeat angiography was performed at 3-6 months post-PCI and at 12 months after PCI and CABG. Stress tests were performed at 1, 3, 6, and 12 months after the procedure.

Principal Findings:

In the PCI group, 35% of patients were treated with a drug-eluting stent, the average number of vessels dilitated was 2.3, and procedure success was 100%. In the CABG group, 79% of patients were grafted with a left internal mammary artery (LIMA), and the average number of grafts was 2.9. Baseline ejection fraction was 53% and Canadian Cardiovascular Society (CCS) angina class was 3.

By 30 days, in the CABG group there were 2 deaths, 2 MIs, 2 strokes, 4 cases of heart failure, and 1 repeat revascularizaton. Additionally, there were 7 cases of infection. In the PCI group, there was 1 MI and 1 case of heart failure by 30 days. From 30 days to 1 year, there were 2 additional deaths in the CABG group and 1 in the PCI group, along with 2 additional cases of heart failure in the PCI group. Repeat revascularization was reported in 7 patients in the CABG group (5 for left main target vessel revascularization) and 8 in the PCI group (3 for left main TVR). Any MACE occurred in 20% of the CABG group and 21% of the stent group (p=NS). There was no difference in long-term survival between groups (p=0.16).

Angina status at 12 months, one of three primary endpoints, did not differ by treatment group (p=0.38), with similar improvements from baseline in both groups. Likewise, duration of treadmill stress test showed no difference at 12 months between groups (p=NS). LVEF at 12 months was higher in the PCI group vs the CABG group (60% vs 54%, p=0.047).


Among patients with unprotected left main coronary artery disease undergoing elective revascularization, there was no difference in two of the three primary endpoints, angina status or duration on treadmill stress test, between the stent and CABG treated patients, but LVEF was slightly higher in the PCI group. However, the trial has numerous limitations, including three "primary" endpoints; standard clinical trial methodolgy entails using one primary endpoint. Additionally, there was no statistical adjustment for multiple endpoints. The rate of infection following CABG was extremely high (13% during the first month), compared with a reported rate of ~3% in recent CDC reports. The presenters did not provide information on why the rate was higher than historical data. Additionally, the rate of repeat revascularization during the first year in the surgical group was 15%, much higher than historical data of ~4%, particularly given the fact that 79% of patients in the trial were treated with LIMAs, which are normally associated with even lower occlusion rates. Finally, the trial was "open-label" and the decision to randomize was based after visualization of the coronary anatomy. As the commentator pointed out, follow-up evaluation would also not be blinded. As a result of these limitations, caution should be used in conclusions drawn from these data.


Presented by Pawel E. Buszman at TCT 2005, Washington, DC.

Clinical Topics: Cardiac Surgery, Heart Failure and Cardiomyopathies, Invasive Cardiovascular Angiography and Intervention, Noninvasive Imaging, Aortic Surgery, Cardiac Surgery and Heart Failure, Acute Heart Failure, Interventions and Imaging, Angiography, Nuclear Imaging

Keywords: Stroke, Follow-Up Studies, Coronary Angiography, Drug-Eluting Stents, Mammary Arteries, Heart Failure, Coronary Artery Bypass, Angioplasty, Balloon, Coronary, Exercise Test

< Back to Listings