Coronary revascularization on balance: Robert L. Frye Lecture

Abstract

The first generation of multicenter randomized controlled trials of coronary artery bypass surgery vs medical treatment in the 1970s found survival advantage only in patients with left main coronary artery disease or with multiple risk factors. Over time, these results have remained reproducible and biologically plausible and continue to be the basis for contemporary guidelines for bypass surgery. Percutaneous coronary intervention (PCI), which became available some 10 years after surgery, was compared with medical treatment in various clinical settings. Patients with stable angina receiving aggressive medical therapy had survival rates comparable to those undergoing PCI in several trials. In the presence of unstable angina, evidence suggests benefit from intervention after stabilization. In the setting of acute myocardial infarction or cardiogenic shock, PCI showed results superior to lytic therapy in centers with large PCI volume. A comparison of the 2 revascularization methods in the 1990s resulted in no overall 7-year survival difference except in patients with treated diabetes in whom bypass surgery prolonged life compared with angioplasty. None of the revascularizations had an effect on the incidence of myocardial infarctions, but importantly, the presence of bypass grafts reduced their fatal impact compared with both medical treatment and PCI. All revascularization trials reported improvement in quality of life, including symptoms, compared with less aggressive therapy. In the pursuit to explain 3 decades of steady decline of cardiovascular mortality in the United States, health economists attempted to model the decline due to availability of intervention by estimates obtained from contemporary randomized prevention and intervention trials. They concluded that, thus far, treatments have made a greater contribution to the decline than has primary prevention. Randomized trials not only contribute to evidence-based clinical practice, but also can reveal underlying biological mechanisms and provide quantitative data to model population trends. Thus, they should be regarded as the basic science of medical therapeutics and population health.