Much of what is perceived as new and better in surgery is, in fact, a reconsideration of earlier ideas framed in the context of modern clinical practice. Since it is the inevitable tendency of surgeons to offer their patients the best of current practice, such &dquo;advances&dquo; in surgical care are often applied prior to actual scientific demonstration of their true benefit. In the past decade several examples of perhaps inappropriate enthusiasm can be cited, including the alleged superiority of the in situ vein graft, nonresective treatment of aortic aneurysm in high-risk patients, and the contention that the retroperitoneal approach is the preferred route for aortic reconstruction.A critical appraisal of the literature claiming superiority for the retroperitoneal approach reveals two distinct flaws, namely, the limitations of retrospective review and the use of historical controls. It has been our observation that a number of factors, including (but not limited to) better cardiac risk stratification, newer anesthetic techniques, and autotransfusion, have contributed to the ever-improving safety of aortic surgery at the Massachusetts General Hospital. Thus it was clear that a prospective study with randomization would be required to isolate and examine the variable of surgical approach and in this way address the contention that the retroperitoneal approach is the more &dquo;physiologic&dquo; route for aortic reconstruction. In this article we review the evolution of experience with the retroperitoneal operation and offer a critique of those individual tenets of aortic reconstruction said to be favorably influenced by use of a retroperitoneal approach. This is done from the perspective of our own randomized study of this question in which we demonstrated no important advantage for the retroperitoneal approach. BACKGROUND