Background and aim of the study: Modes of failure of the Perimount bioprostheses have not been characterized. Thus, the study aim was to classify their failure modes, to compare their rates of occurrence, and to relate these to patient age.
Methods: Between September 1981 and July 1985, a total of 478 patients received Carpentier-Edwards Perimount aortic valve prostheses as part of a premarket approval cohort. During yearly follow up over 19 years, 95 prostheses were explanted, 70 (74%) for reasons of structural valve deterioration (SVD). Preoperative diagnostic studies, echocardiograms, operative notes and pathology reports were reviewed to classify the primary modes of failure. Hazard function methodology established mode-specific occurrence rates and how these were modulated by patient age at implant.
Results: The pathophysiology of SVD was pure regurgitation in 23 cases, pure stenosis in 18, mixed in 20, and uncertain in nine. Primary modes of failure were calcification (n = 27; 39%), non-calcific degeneration (n = 21; 30%), fibrosis (n = 5; 7.1%), dehiscence (n = 4; 5.7%), a combination of these (n = 10; 14%), or unknown (n = 3; 4.3%). The failure rate for each mode accelerated with time, was highest and almost equal for calcification and degeneration, and was slower and almost equal for fibrosis and dehiscence. Except for dehiscence, failure modes were age-dependent (p < 0.05), and this was most prominent for degeneration and calcification.
Conclusion: Perimount bioprostheses rarely fail from design-related causes; rather, they fail at almost equal rates from calcific and non-calcific degeneration, both of which occur late and are age-dependent. Further investigation into improved methods for fixation and anti-calcification in preparing these valves is warranted.