Objectives: The aim of this study was to investigate the long-term clinical and hemodynamic outcomes after aortic valve replacement with the Carpentier-Edwards Perimount bioprosthesis (Edwards Lifesciences, Irvine, Calif), which has been used in our institution since 1984.
Methods: From January 1984 to December 1995, the Carpentier-Edwards pericardial bioprosthesis was used for aortic valve replacement in 254 patients (male/female ratio 117:137) with a mean age of 71 years (range 25-87 years). Before the operation, 216 patients (85%) were in New York Heart Association functional class III or IV. The predominant diagnosis was aortic stenosis (n = 219, 86%). Associated surgical procedures included coronary artery bypass grafting in 130 cases (51%), mitral valve replacement in 11 cases (4%), and tricuspid or mitral valve repair in 12 cases (5%). Previous cardiac operations had been performed in 36 cases (14%). Follow-up was 100% complete at a mean of 60 +/- 31 months. Univariate estimates of time-related cumulative probabilities were calculated by the Kaplan-Meier method. Multivariable adjustment was performed by Cox proportional hazards regression. Echocardiography was performed in 61% of long-term survivors.
Results: There were 11 early deaths (4%) and 58 late deaths. Actuarial survivals at 5, 10, and 12 years were 80% +/- 3%, 50% +/- 8%, and 36% +/- 9%, respectively. At 12 years the freedom from cardiac death was 73% +/- 7%, the freedom from valve-related death was 84% +/- 11%, the freedom from valve reoperation was 83% +/- 9%, the freedom from primary tissue failure was 86% +/- 9%, the freedom from thromboembolism was 67% +/- 13%, and the freedom from endocarditis was 98% +/- 1%. Echocardiography was performed on long-term survivors (mean follow-up 67 +/- 25 months) and showed that transvalvular peak and mean pressure differences measured with Doppler echocardiography were 23.2 +/- 9.6 and 12.3 +/- 4.8 mm Hg, respectively. Aortic regurgitation was found by Doppler echocardiography to be none or trivial, mild, moderate, and severe in 64%, 30%, 3%, and 1% of patients, respectively. Mean left ventricular mass index was 107.2 +/- 35.3 g/m(2) (118.9 +/- 40.2 g/m(2) in men and 98.8 +/- 28.8 g/m(2) in women) at late follow-up. One third of all patients, regardless of sex (n = 26/64 women and n = 14/45 men), had evidence of left ventricular hypertrophy. However, our analyses indicate that the residual left ventricular hypertrophy was not caused by valve mismatch but was probably multifactorial.
Conclusion: The Carpentier-Edwards Perimount bioprosthesis has provided satisfactory clinical and hemodynamic outcome. However, at long-term follow-up about one third of the patients being investigated still had left ventricular hypertrophy examined by echocardiography.