Different alternatives for the surgical treatment of aortic valve disease have been recently introduced. All consecutive patients who underwent aortic valve surgery between July 1988 and March 1994 were reviewed. There were 674 patients with a mean age of 32.4 years, mean preoperative functional class of 2.82, and rheumatic etiology in 59% of the cases. The patients were divided into 3 groups: Group I. standard aortic valve replacement with biological and mechanical prosthesis ( n = 313); Group II. stentless aortic valve replacement using homograft, pulmonary autograft and reconstruction with pericardium ( n = 145); and Group HI. aortic valve repair ( n = 216). The hospital mortality was 6.07% for the standard, 0.68% for the stentless, and 3.70% for the repair. Total follow-up was 1,304.75 patient years with a mean of 21.93 months. The actuarial survival at 66 months excluding hospital mortality was 85.24 ± 4.59% in the standard replacement, 92.63 ± 4.03% in the stentless, and 91.20 ± 3.02% in the repair group. The highest incidence of reoperation corresponded to the repair group with an actuarial freedom from reoperation of 74.26 ±7.03%, v. 92.52 ±4.52% in the standard and 85.11 ± 6.71% in the stentless group. There were no thromboembolic events in the isolated aortic valve survivors in both the stentless and repair groups and 1.28% patient years in the standard. We conclude that both the stentless aortic valve replacement and the aortic repair represent a good alternative v. standard replacement, especially for those young rheumatic patients in which anticoagulation and durability of the prosthesis is still a problem.