In patients with small aortic annuli, the choice of prosthesis should be based on hemodynamics and valve durability. The Carpentier-Edwards pericardial valve offers good hemodynamic performance and long-term valve durability. We reviewed our 12-year experience with 121 patients who received a 19 mm prosthesis.A total of 121 patients (97 females, 24 males; mean age 73.2 +/- 9.4 years) underwent isolated aortic valve replacement with a 19 mm Carpentier-Edwards pericardial bioprosthesis in our institution between July 1984 and December 1995. Patients were followed up for an average of 4.84 years after surgery; total follow up was 581 patient-years (pt-yr).The operative mortality rate was 3.3% (4/121). At the present time, 77 patients (86%) are in NYHA class I or II with a mean gradient of 18.0 +/- 6.9 mmHg and mean effective orifice area 1.1 +/- 0.23 cm2. There were 25 late deaths. After 12 years the actuarial survival rate was 42 +/- 26%. Eight patients died of valve-related cause (one endocarditis, one structural failure, two thromboembolisms, one anticoagulant-related hemorrhage, three sudden deaths). At 12 years, the actuarial rate of freedom from valve-related death was 61 +/- 37% and from non-sudden valve-related death 88 +/- 11%. Valve-related complications included six thromboembolic episodes (1.0%/pt-yr), one endocarditis (0.17%/pt-yr), two reoperations (0.34%/pt-yr) and two structural valve failures (0.34%/pt-yr). After 12 years, freedom from reoperation was 89 +/- 11%, from valve failure 92 +/- 8%, from thromboembolic episodes 82 +/- 16% and from endocarditis 99 +/- 1%.With a low rate of structural valve failure at 12 years and a good clinical performance, the Carpentier-Edwards prosthesis is a reliable alternative for small aortic annuli.