Percutaneous transluminal coronary angioplasty was developed in the late 1970s as a nonsurgical alternative for revascularization of atherosclerotic coronary arteries. It gained widespread acceptance without a controlled trial. Introduced in 1986, directional coronary atherectomy was the first of other recently developed coronary devices that sought to improve on the results of angioplasty. It was approved in 1990 by the Food and Drug Administration (FDA) on the basis of observational data. Its use expanded rapidly, reaching over 35,000 procedures in 1992, accounting for more than 10% of all interventions. After premarket approval, two major randomized trials tested the hypothesis that atherectomy would be superior to angioplasty. Their results raised a cautionary flag and stood in contrast to projections made from prior observational data. It is concluded that randomized controlled trials validate claims of relative efficacy and safety of competing medical technologies, a lesson reflected in recent changes in policy at the FDA.