Review of the COURAGE Trial

Optimal Medical Therapy with or without PCI for Stable Coronary Disease

N Engl J Med 2007;356:1503-1516

Background: By the turn of the 21st century more than 1 million coronary stent procedures were performed each year in the United States and approximately 85% were undertaken electively in patients with stable coronary artery disease.

This pattern evolved without a single clinical trial demonstrating a concrete improvement in hard endpoints with percutaneous coronary intervention (PCI) compared to optimal medical therapy (OMT) alone.

We have already reviewed several of these trials including ACME, RITA-2 and the Atorvastatin vs Angioplasty trial. Each trial was relatively small and none showed a significant benefit for revascularization compared to medical therapy on death or MI.

Previous trials involving PCI compared to standard care or OMT included less than 3,000 patients altogether, did not broadly use intracoronary stents (instead using balloon angioplasty only) and they did not employ what would be considered a contemporary standard of medical management. Thus many questions involving the efficacy and safety of PCI versus OMT alone for managing stable CAD remained unanswered.

The Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial was designed to test the hypothesis that up front PCI plus OMT would significantly reduce the risk of death and nonfatal MI compared to OMT alone in patients with stable CAD.

Patients: Eligible patients had stable CAD defined as either 1) a coronary stenosis of >/= 70% in one or more proximal epicardial coronary arteries and objective evidence of myocardial ischemia (based on the resting ECG or with exercise or pharmacologic vasodilator stress testing) or 2) a coronary stenosis of >/= 80% and classic angina without provocative testing.

Patients were excluded if they had persistent Canadian Cardiovascular Society (CCS) class IV angina, a markedly positive stress test defined by substantial ST-segment depression or hypotensive response during stage 1 of the Bruce protocol), refractory heart failure or cardiogenic shock, an EF </=30%, revascularization within 6 months, and coronary anatomy not suitable for PCI.

Baseline characteristics: The average age of patients was 61 years of age, 85% were men and 85% were white. The average EF was 61%. The median time from the first episode of angina before randomization was 5 months (median, 3 episodes per week, with exertion or at rest). Fifty eight percent of patients had CCS class II or III angina. Eighty five percent of patients underwent stress testing at baseline. Of these, 70% underwent nuclear imaging. In patients who underwent nuclear imaging, 66% had multiple reversible defects and 23% had a single reversible defect.

Angiographic data is as follows, approximately 30% of patients had single vessel disease, 40% had 2 vessel disease, and 30% had 3 vessel disease. The proportion of patients with proximal LAD disease was unbalanced between groups with the medical therapy group having a significantly higher percentage (37% vs 31%; p=0.01). Eleven percent of all patients had a previous CABG and among these participants 65% had disease in a graft.

Procedures: Patients were randomly assigned to undergo PCI and OMT (PCI group) or OMT alone (OMT group). Patients were stratified by study site and whether they had a previous CABG. All patients received antiplatelet therapy with aspirin (81 to 325 mg per day) or clopidogrel (75 mg per day), if aspirin intolerant. Patients undergoing PCI received aspirin and clopidogrel in accordance with accepted standards. Medical anti-ischemic therapy in both groups included long-acting metoprolol, amlodipine, and isosorbide mononitrate alone or in combination, along with either lisinopril or losartan as standard secondary prevention. All patients received therapy to lower LDL with simvastatin alone or in combination with ezetimibe with a target of 60 to 85 mg/dl. After the LDL target was achieved, an attempt was made to raise the HDL to above 40 mg/dl and lower triglycerides to less than 150 mg/dl with exercise, extended-release niacin, or fibrates, alone or in combination.

In patients undergoing PCI, target-lesion revascularization was always attempted, and complete revascularization was performed as clinically appropriate.

Endpoints: The primary endpoint was a composite of death from any cause and nonfatal MI. Secondary outcomes included a composite of death, MI, stroke, and hospitalization for unstable angina with negative biomarkers. The prespecified definition of MI, whether periprocedural or spontaneous, required a clinical presentation consistent with acute coronary syndrome and either 1) new abnormal Q waves in 2 or more contiguous ECG leads or 2) positive cardiac biomarkers. Silent MI, as detected by abnormal Q waves, was confirmed by a core laboratory and was also included as an outcome of MI.

Intention-to-treat analysis was performed for the primary endpoint and important secondary endpoints. The investigators estimated that a sample size of 2,270 patients, followed for 2.5 to 7 years would be needed to detect a relative 22% difference in the primary composite endpoint (16.4% in the PCI group vs 21% in the OMT group at 3 years). This would provide a power of 85% to detect the anticipated difference with a 2 sided alpha of 5%. The investigators incorporated assumptions about crossover between study groups and loss to follow-up.

Results: A total of 2,287 patients were included in the final analysis (1,149 in the PCI group and 1,138 in the OMT group), which represented 6.4% of all patients screened (N=35,539) and 8.5% of those screened who met the basic inclusion criteria (N=26,862). Among those meeting inclusion criteria, the most common reasons for exclusion were: underwent revascularization within 6 months (17%), had an inadequate EF (18%), had a contraindication to PCI (11%), had a serious coexisting illness (9%), had valvular disease (5%), had class IV angina (5%), had left main coronary stenosis >/=50% (4%).

Patients in the trial were adherent with medical interventions and motivated to improve diet and lifestyle. For all patients, BP and cholesterol were improved and activity levels increased over the course of the study. For example, in the OMT group, baseline SBP and LDL were 130 mm Hg and 102 mg/dl at baseline and declined to 122 mm Hg and 72 mg/dl at 5 years, respectively. Similar reductions were seen in the PCI group. Fifty five percent of patients performed moderate physical activity at baseline and this increased to 77% at 5 years. Dietary adherence improved across the board and smoking decreased.

Patients were followed for a median time of 4.6 years and 9% were lost to follow up with no significant difference between group. Overall, PCI + OMT did not reduce the primary endpoint of death or nonfatal MI compared to OMT alone (19.0% with PCI vs 18.5% with medical therapy; HR 1.05; 95% CI 0.87-1.27). Periprocedural MI was numerically more frequent in the PCI group whereas spontaneous MI was nearly identical between groups (statistics not provided). There were no statistically significant differences noted in any of the endpoints reported with the exception of revascularization during the follow up period, which occurred less frequently in the PCI group (20% vs 30%; p<0.001). *It is worth noting that for patients in the PCI group, 12% did not receive a stent following their randomization assignment.

Subgroup analysis yielded one potentially significant interaction for the primary endpoint. Women appeared to benefit from PCI + OMT vs OMT alone compared to men (18% vs 26% for women compared to 19% vs 18% for men; p for interaction = 0.03). Women represented only 15% of the total cohort and contributed approximately 15% of total events and thus, we suspect this represents a false positive finding due to low power since it also lacks biologic credibility.

Conclusions: In patients with stable coronary artery disease, an up-front strategy of PCI plus OMT is not superior to OMT alone for reducing the risk of death or nonfatal MI. The majority of patients with stable CAD should be treated with OMT alone, as an initial strategy, with PCI reserved for those with inadequate symptom control. One limitation of this trial when viewed through a contemporary lens is the low use of drug eluting stents for revascularization; however, we do not personally believe this detracts from the overall findings.

Some people may interpret the 30% rate of revascularization in the OMT group as a proof of efficacy of an up-front strategy of PCI; however, we disagree. These people should keep in mind that 20% of people in the PCI group were revascularized again as well. Their argument boils down to this, “you should revascularize 100% of people up front to reduce the risk of needing PCI in the future by 10%.” That rationale does not make sense on an individual or societal basis when considering cost, safety or convenience. The only people benefiting from such an approach would be the operators and hospitals.