New England Journal of Medicine April 12, 2007

Optimal Medical Therapy with or without PCI for Stable Coronary Disease

William E. Boden, Robert A. O'Rourke, Koon K. Teo, et al.

Bottom Line

In patients with stable coronary artery disease, an initial strategy of percutaneous coronary intervention added to optimal medical therapy did not reduce the risk of death or nonfatal myocardial infarction compared to optimal medical therapy alone.

Key Findings

1. At a median follow-up of 4.6 years, the cumulative primary event rate (all-cause death and nonfatal myocardial infarction) was 19.0% in the PCI group versus 18.5% in the medical-therapy group (Hazard Ratio [HR] 1.05; 95% CI 0.87-1.27; P=0.62).
2. There was no significant difference in the composite of death, myocardial infarction, and stroke (20.0% in the PCI group vs. 19.5% in the medical-therapy group; HR 1.05; 95% CI 0.87-1.27; P=0.62).
3. Rates of individual cardiovascular endpoints were similar, including hospitalization for acute coronary syndrome (12.4% vs. 11.8%; HR 1.07; P=0.56) and myocardial infarction alone (13.2% vs. 12.3%; HR 1.13; P=0.33).
4. PCI provided more rapid and significant relief of angina early in the trial; at 1 year, 66% of the PCI group were free of angina compared to 58% in the medical-therapy group (P<0.001).
5. The early symptomatic advantage of PCI narrowed over time and was no longer statistically significant by 5 years, at which point 74% of the PCI group and 72% of the medical-therapy group were free of angina (P=0.35).

Study Design

Design
Randomized Controlled Trial
Open-Label
Sample
2,287
Patients
Duration
4.6 yr
Median
Setting
Multicenter, US & Canada
Population Patients with stable coronary artery disease and objective evidence of myocardial ischemia on noninvasive testing.
Intervention Percutaneous coronary intervention (PCI) plus intensive pharmacologic therapy and lifestyle intervention (optimal medical therapy).
Comparator Intensive pharmacologic therapy and lifestyle intervention (optimal medical therapy) alone.
Outcome Composite of death from any cause and nonfatal myocardial infarction.

Study Limitations

The open-label design of the trial may have introduced bias, particularly regarding patient-reported angina metrics and the clinical threshold to proceed with subsequent revascularizations in the medical therapy arm.
A substantial crossover rate was observed; 32.6% of patients initially assigned to optimal medical therapy alone eventually crossed over to receive revascularization during the follow-up period [2.6].
The trial predominantly utilized bare-metal stents (drug-eluting stents were introduced only late in the enrollment phase), leading to later criticism that modern stenting techniques might yield lower restenosis rates, though this is unlikely to have altered hard endpoints like mortality or myocardial infarction.
The study rigorously excluded extremely high-risk patients, such as those with significant left main coronary artery disease, markedly depressed left ventricular ejection fraction, or highly symptomatic refractory unstable angina, limiting the generalizability to these populations.

Clinical Significance

The COURAGE trial was a watershed moment in cardiovascular medicine that fundamentally shifted the management paradigm for stable ischemic heart disease. It decisively challenged the long-held 'oculostenotic reflex'—the belief that visually apparent stenoses inherently required mechanical intervention to prevent fatal events. By demonstrating that PCI does not confer a survival advantage or prevent myocardial infarctions over optimal medical therapy in stable patients, the trial established that the primary clinical utility of PCI in this population is the relief of medically refractory angina. This guided clinical practice guidelines to heavily emphasize rigorous medical therapy and lifestyle optimization as the necessary first-line approach for stable coronary artery disease.

Historical Context

Prior to 2007, percutaneous coronary intervention was widely utilized as an initial strategy for patients with stable coronary artery disease under the prevailing pathophysiological assumption that relieving a fixed epicardial stenosis would intrinsically prevent future myocardial infarctions and improve survival. While coronary artery bypass grafting (CABG) had shown survival benefits in specific high-risk anatomical cohorts, similar robust evidence was lacking for PCI. COURAGE was the first large-scale, contemporary randomized trial to rigorously test this practice against optimized evidence-based medical therapy. Its findings generated immense debate but ultimately initiated a widespread shift toward conservative initial management. The core conclusions of COURAGE were later validated and expanded upon by the landmark ISCHEMIA trial in 2020, which re-confirmed the lack of a mortality benefit for an early invasive strategy in stable patients with moderate-to-severe ischemia.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Based on the COURAGE trial's findings, why does not opening a severe, stable focal coronary stenosis with a stent reduce the risk of a future myocardial infarction?

Key Response

Myocardial infarctions are typically caused by the sudden rupture of vulnerable, lipid-rich plaques that often do not cause severe flow-limiting stenosis prior to rupture. Optimal medical therapy stabilizes these vulnerable plaques systemically, whereas PCI only treats the local, stable flow-limiting lesion, explaining why PCI does not prevent future systemic plaque ruptures.

Resident
Resident

A 65-year-old male with stable angina and an 80% proximal RCA lesion asks if a stent will prevent him from having a heart attack. How do you apply the COURAGE trial results to counsel him regarding the primary goals of PCI versus medical therapy?

Key Response

Counsel the patient that optimal medical therapy (statins, aspirin, beta-blockers) is the cornerstone for preventing death and myocardial infarction. PCI is indicated primarily for symptom relief (angina) if medical therapy fails or is intolerable, not for reducing mortality or MI risk in stable coronary artery disease.

Fellow
Fellow

The COURAGE trial primarily evaluated patients with stable CAD and normal left ventricular function. How do the results of trials like ISCHEMIA and STICH modify our understanding of revascularization in patients with stable CAD who present with either moderate-to-severe ischemia or reduced ejection fraction?

Key Response

ISCHEMIA confirmed COURAGE findings even in patients with moderate-to-severe ischemia (excluding left main disease), showing no mortality benefit. However, STICH demonstrated a long-term mortality benefit for CABG in ischemic cardiomyopathy with reduced EF, highlighting that left ventricular dysfunction fundamentally alters the risk-benefit calculus of revascularization.

Attending
Attending

Despite the COURAGE trial being published over a decade ago, the oculostenotic reflex persists in clinical practice. What cognitive biases drive the continued overuse of PCI in asymptomatic or medically manageable stable CAD, and how can we systemically restructure cath lab workflows to mitigate this?

Key Response

Physicians struggle with action bias and the intuitive but flawed plumbing model of coronary disease. Implementing mandatory FFR/iFR physiological assessment, multidisciplinary heart team discussions for non-ACS cases, and strict adherence to appropriate use criteria (AUC) audits can curb the oculostenotic reflex and align practice with COURAGE evidence.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The COURAGE trial utilized a composite primary endpoint of death from any cause and nonfatal myocardial infarction. From a methodological standpoint, how does the inclusion of nonfatal MI, an endpoint dependent on biomarker thresholds, complicate the interpretation of an unblinded trial comparing an invasive procedure to medical therapy?

Key Response

In an unblinded trial, periprocedural MIs (Type 4) are exclusively tied to the PCI arm, potentially driving early harm signals, while spontaneous MIs might be reduced later. Blurring these distinct events into a single composite endpoint without differentiating MI types can skew the treatment effect estimate and mask temporal hazard dynamics.

Journal Editor
Journal Editor

A major criticism of COURAGE is the high rate of crossover, with nearly a third of the medical therapy group eventually undergoing revascularization. How does this crossover affect the intention-to-treat analysis, and what sensitivity analyses are critical to ensure the trial is not masking a true benefit of PCI?

Key Response

High crossover biases the intention-to-treat analysis toward the null, increasing the risk of a Type II error. A rigorous peer review would demand per-protocol or as-treated analyses, as well as time-dependent Cox proportional hazards models adjusting for crossover, to ensure the lack of mortality benefit was not simply due to the medical arm receiving the intervention.

Guideline Committee
Guideline Committee

How did the findings of the COURAGE trial shift the ACC/AHA and ESC guideline recommendations regarding the Class of Recommendation for PCI in stable ischemic heart disease (SIHD)?

Key Response

COURAGE caused a paradigm shift, leading guidelines to downgrade PCI in stable CAD without left main disease or severe LV dysfunction. Current ACC/AHA guidelines give optimal medical therapy a Class I recommendation for initial management, reserving PCI (Class I) primarily for symptom relief in patients with refractory angina despite OMT, heavily emphasizing shared decision-making.

Clinical Landscape

Noteworthy Related Trials

2009

BARI 2D Trial

n = 2,368 · NEJM

Tested

Prompt revascularization (PCI or CABG) plus intensive medical therapy

Population

Patients with type 2 diabetes and stable coronary artery disease

Comparator

Intensive medical therapy alone

Endpoint

All-cause mortality at 5 years

Key result: There was no significant difference in overall survival or major cardiovascular events between prompt revascularization and medical therapy alone in diabetic patients with stable CAD.
2017

ORBITA Trial

n = 200 · Lancet

Tested

PCI plus optimal medical therapy

Population

Patients with severe single-vessel coronary disease and stable angina

Comparator

Placebo (sham procedure) plus optimal medical therapy

Endpoint

Difference in exercise time increment

Key result: In patients with medically treated angina and severe single-vessel coronary stenosis, PCI did not increase exercise time more than a sham procedure.
2020

ISCHEMIA Trial

n = 5,179 · NEJM

Tested

Initial invasive strategy (angiography and revascularization) plus medical therapy

Population

Patients with stable coronary disease and moderate to severe ischemia

Comparator

Initial conservative strategy (medical therapy alone)

Endpoint

Composite of CV death, MI, resuscitated cardiac arrest, hospitalization for unstable angina, or heart failure

Key result: Among patients with stable coronary disease and moderate or severe ischemia, an initial invasive strategy did not significantly reduce the risk of ischemic cardiovascular events or death from any cause over a median of 3.2 years compared to a conservative strategy.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis