"CABG vs. PCI in Diabetes with Multivessel Disease".The New England Journal of Medicine. 2012. DOI: 10.1056/NEJMoa1211585.

Clinical Question

In patients with diabetes and multivessel coronary artery disease, is coronary-artery bypass grafting (CABG) superior to percutaneous coronary intervention (PCI) with drug-eluting stents in reducing the rates of death and myocardial infarction?

Bottom Line

For patients with diabetes and advanced coronary artery disease, CABG was superior to PCI in that it significantly reduced rates of death and myocardial infarction, with a higher rate of stroke.

Major Points

Guidelines

As a result of this study, guidelines may recommend CABG as the preferred revascularization strategy for patients with diabetes with multivessel coronary artery disease.

Design

- Multicenter, randomized, controlled trial
- N=1,900 patients with diabetes and multivessel coronary artery disease
- PCI with drug-eluting stents vs. CABG
- Median follow-up: 3.8 years
- Analysis: Intention-to-treat
- Primary outcome: Composite of death from any cause, nonfatal myocardial infarction, or nonfatal stroke

Population

- Inclusion Criteria: Patients with diabetes, angiographically confirmed multivessel coronary artery disease (>70% stenosis), without left main coronary stenosis
- Baseline Characteristics: Mean age 63.1 years, 29% women, 83% had three-vessel disease

Interventions

Patients were assigned to undergo PCI with drug-eluting stents or CABG following recommended medical therapies for cholesterol, blood pressure, and glycated hemoglobin control.

Outcomes

- Primary Outcomes: Higher rate of primary composite outcome in PCI (26.6%) compared with CABG (18.7%) at 5 years; P=0.005.
- Secondary Outcomes: Higher mortality in PCI group (16.3%) versus CABG group (10.9%) at 5 years; stroke was more frequent in CABG group at 5 years (PCI 2.4%, CABG 5.2%); major adverse cardiovascular and cerebrovascular events at 1 year were higher in PCI (16.8%) than in CABG (11.8%).

Criticisms

- Prespecified subgroups had very low prevalence which may affect the ability to detect interactions.
- The study was unblinded which could lead to potential treatment biases. However, outcomes were objectively and independently adjudicated.
- The trial was not powered to detect a difference in all-cause mortality.

Funding

National Heart, Lung, and Blood Institute (NHLBI). Hardware provided by Cordis, Johnson & Johnson, Boston Scientific, Eli Lilly, Sanofi-Aventis, and Bristol-Myers Squibb.

Further Reading