Duke Database: CABG vs. PTCA vs. Medical Therapy for Symptomatic CAD - Duke Database: CABG vs. PTCA vs. Medical Therapy for Symptomatic CAD


This report describes outcomes of patients within a large, single-center, prospective, observational registry of patients with symptomatic coronary artery disease (CAD) referred for cardiac catheterization and treated with medical therapy, percutaneous transluminal coronary angioplasty (PTCA), or coronary artery bypass graft (CABG).


The purpose was to describe the outcomes of patients with stable CAD treated at an academic medical center with a strategy of PTCA, CABG, or medical therapy.

Study Design

Study Design:

Patients Screened: 17,309
Patients Enrolled: 9,263
Mean Follow Up: Five years
Mean Patient Age: 50-69
Female: 25

Patient Populations:

Inclusion criteria were consecutive patients referred to the Duke Heart Center for initial cardiac catheterization for evaluation of suspected ischemic heart disease.


Absence of significant (≥75% diameter stenosis) stenosis in at least one major epicardial coronary segment; ≥75% left main stenosis; 3+ or 4+ ischemic mitral regurgitation; prior PTCA; prior CABG; or primary valvular, congenital, or nonischemic cardimyopathic disease

Primary Endpoints:

The primary endpoint was survival at five years.

Drug/Procedures Used:

After catheterization, patients were followed in the Duke Cardiovascular Disease Databank at six months, one year, and then annually. Patients underwent nonrandomized treatment at the discretion of the treating physicians. Of the total 9,263 patients in the registry, within 60 days of the inital catheterization, 2,626 patients underwent PTCA and 3,080 underwent CABG.

Concomitant Medications:

Concomitant medications were at the discretion of the treating physicians.

Principal Findings:

A total of 9,263 patients were included in the analysis. Survival data were 97% complete. There were significant differences in baseline characteristics among the three groups: the PTCA group had the highest prevalence of acute myocardial infarction, and medically treated patients had the highest prevalence of congestive heart failure and other comorbidities. The distribution of CAD varied among the groups, with a lesser extent of disease in the PTCA group, and more severe disease in the CABG group. Additionally, PTCA patients had the highest and medical patients the lowest ejection fractions.

After adjustment for baseline imbalances between groups, the five-year survival for patients with single-vessel disease was 95% for PTCA, 93% for CABG, and 94% for medicine. In patients with two-vessel disease, PTCA or CABG had a slight advantage over medicine, with an adjusted five-year survival of 91% for PTCA, 91% for CABG, and 86% for medicine. In patients with three-vessel disease, CABG had a survival advantage with adjusted five-year survival of 89% for CABG, 81% for PTCA, and 72% for medicine, although only 260 patients with three-vessel disease were treated with PTCA in the overall cohort.

When patients were further classified by a CAD severity index in adjusted analyses (taking into account not only the total number of diseased vessels, but also including location of disease and stenosis severity), medical therapy was nonsignificantly better than CABG for mild disease, whereas CABG was superior for more severe disease (>95% proximal left anterior descending [LAD] lesion with another diseased vessel, and all forms of three-vessel disease). For the comparison of PTCA with medical therapy, there was a nonsignificant trend for a benefit of PTCA over medical therapy for all milder and moderate forms of disease. Finally, for the comparison of CABG vs. PTCA, in patients with less severe disease, patients treated with PTCA had lower mortality than patients treated with CABG. The two therapies were similar in severe single-vessel disease (>95% proximal LAD involvement) or moderate two-vessel disease, and CABG was superior in severe two-vessel disease and three-vessel disease.


The results of this large, prospective, observational cohort study (the first of its kind when it was published) are in line with more recent reports and randomized trial data regarding CABG vs. PTCA vs. medical therapy for symptomatic CAD. The data support the survival advantage of CABG over medical therapy for severe forms of multivessel disease, and suggest a modest survival trend for PTCA over medical therapy in patients with milder forms of disease.

Despite the authors’ efforts to carefully adjust for baseline imbalances between treatment groups and for potential bias in treatment strategies based on these imbalances, an important limitation of this analysis is the nonrandomized comparison between treatments. Additionally, as this study was completed prior to the widespread use of coronary artery stenting and far prior to the approval of drug-eluting stenting, its current applicability in the management of patients with multivessel disease with lesions amenable to PCI is limited.


Mark DB, Nelson CL, Califf RM, et al. Continuing evolution of therapy for coronary artery disease. Initial results from the era of coronary angioplasty. Circulation 1994;89:2015-25.

Keywords: Coronary Artery Disease, Myocardial Infarction, Cardiac Catheterization, Heart Failure, Comorbidity, Constriction, Pathologic, Coronary Artery Bypass, Angioplasty, Balloon, Coronary

< Back to Listings