The Ross operation has been performed for more than 25 years and its popularity has increased dramatically in recent years. We developed an interest in this procedure through a combination of a basic dissatisfaction with a device that requires life-long anticoagulation and the belief that a vital, autologous tissue valve with normal valve morphology and hemodynamics would prove to be superior to the mechanical valve, and that these advantages would outweigh the potential drawbacks related to the operation's technical difficulty and the risk of autograft or homograft dysfunction. From December 1992 to November 1994 40 Ross operations as total root replacements in a diverse group of patients between 5 and 72 years of age (median 32) were performed at Rigshospitalet. Seventeen (43%) of the patients had undergone at least one previous open heart operation. Eleven patients (28%) required surgery because of ongoing or previous endocarditis, and of these, nine had aortic annular destruction and cavity/pseudoaneurysm formation and five had prosthetic valve endocarditis. Three patients (8%) were operated because of mechanical valve dysfunction. One patient was treated for an ascending aortic aneurysm and aortic insufficiency. The remaining 25 patients were operated because of congenital or acquired aortic insufficiency, stenosis, or both. Ten patients (25%) underwent concomitant procedures. No mortality or serious complications occurred. Morbidity was limited to one case each of total atrioventricular (A-V) block, sternal pseudarthrosis, minor stroke, and deep vein thrombosis. Thirty-five patients had no or trivial, two patients mild, and three patients moderate autograft valve insufficiency during a median follow-up of 8 months (range 0-23 months). Two patients had pulmonary stenosis.(ABSTRACT TRUNCATED AT 250 WORDS)