NEW ORLEANS — Many physicians may have received the wrong message from the major 2007 trial that compared optimal medical therapy with immediate coronary stenting in patients with stable coronary artery disease.
The message many physicians took away was that results from the Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial showed that it is safe and good practice to first test optimal medical therapy in patients with stable coronary disease, and if that doesn't work, a patient can later be sent for percutaneous coronary intervention (PCI).
But that's not how the COURAGE trial worked, Dr. Paul S. Teirstein said at the annual scientific sessions of the American Heart Association. What's often forgotten is that all patients screened for the study first underwent diagnostic angiography, and the patients who were randomized to either immediate PCI or optimal medical therapy were primarily the minority of patients with coronary lesions that did not put them at immediate risk.
“All patients [screened for COURAGE] had angiography, and many were excluded. My strong suspicion is that patients at high risk for death and myocardial infarction were excluded,” based on their high-risk angiogram results, said Dr. Teirstein, chief of cardiology and director of interventional cardiology at the Scripps Clinic in La Jolla, Calif.
The COURAGE investigators assessed more than 35,000 patients, and identified slightly more than 3,000 who met the study criteria (N. Engl. J. Med. 2007; 356:1503–16). As of late 2008, the research team had not released any data on what the angiograms looked like in the more than 35,000 patients screened, Dr. Teirstein said. He suspects that a substantial number of patients were excluded because their cardiologists weren't comfortable with not immediately stenting a life-threatening stenosis.
He reported conducting a small survey at 7 of the 50 COURAGE sites, where he asked a lead cardiologist at each site whether patients were enrolled into the randomized study if they had high-grade, proximal, left anterior descending (LAD) artery lesions. At five of the seven sites the answer was no. Patients like these were not considered safe candidates to randomize to initial medical therapy. These investigators, as well as many other cardiologists, consider proximal lesions like these to be potential killers that require immediate PCI, Dr. Teirstein said.
Another indication of the risk posed by proximal LAD lesions lies in the results from the COURAGE nuclear substudy, which used myocardial perfusion imaging in a small subset of the COURAGE patients, slightly more than 300 patients, to compare the ability of PCI and medical therapy to reduce myocardial ischemia. The results showed that PCI plus optimal medical therapy led to a 5% or greater reduction in ischemia in 33% of patients, compared with a similar reduction in just 20% of patients treated medically. The importance of this difference was underscored by a further finding: Among the 105 patients in this substudy with moderate to severe ischemia at baseline, relief of ischemia cut the subsequent rate of death or MI in half compared with patients who did not have ischemia relief (Circulation 2008;117:1283–91).
This dramatic difference in the rate of death or MI between PCI and medical treatment was not seen in the full COURAGE study. A major explanation is that in the overall study about a third of the patients had proximal LAD disease (although not necessarily high-grade lesions). But in the nuclear substudy, nearly half of the patients had such coronary lesions. By chance, the substudy included more patients with what may have been severe and dangerous stenoses, and up-front treatment with PCI led to fewer deaths and MIs, Dr. Teirstein noted.
After the COURAGE results were reported in April 2007, some defenders of the study stressed that COURAGE enrolled “high-risk” patients, based on factors such as a 34% prevalence of diabetes, a 5% prevalence of heart failure, and a 70% prevalence of multivessel disease. But Dr. Teirstein took issue with the idea that these parameters are important determinants of high risk. A high-grade, proximal LAD lesion is a much clearer indicator of high risk, and the COURAGE investigators often did not enroll these patients, he said.
Dr. Teirstein acknowledged that COURAGE was an important study and that it changed his practice. “I learned that you don't need to stent every little blockage in every little vessel,” he said. “I leave the small distal side branches alone unless the patient has recalcitrant angina.”
But, he stressed, the COURAGE results are not proof that PCI of major, proximal lesions can be safely deferred while medical therapy is tried. Patients with lesions like these were underrepresented in the study because many physicians agree that in these patients postponing PCI is too dangerous.