Pathogenesis of Endocarditis in Bioprosthetic Heart Valves—Re-evaluation of the Current Concept

Background: Bioprosthetic pulmonary valves have a high incidence of endocarditis. Endothelial damage (with secondary thrombus apposition) is considered the origin for these infections. In an earlier study, we found a high incidence of thrombus formation at the base of semilunar valves of pulmonary conduits without endocarditis. This led to the assumption that these fibrin collections are the possible source for later development of endocarditis. To re-evaluate this concept, we reviewed a series of explanted valved pulmonary conduits with a history of endocarditis with special regard to the histological pattern of endocarditic lesions.

Methods: We performed a thorough histopathologic work-up of valved pulmonary conduits following a standardized examination protocol (formalin fixation, hard resin embedding, sawing, and grinding). Standard staining as well as immunohistochemistry was applied.

Results: Nineteen valved pulmonary conduits explanted from patients with endocarditis were analyzed (Melody transcatheter valve n = 12, Hancock n = 3, homograft n = 2, and Contegra n = 2). Average time interval between implantation and explantation was 66 months (5–231 months). Endocarditic lesions were localized at the base of the semilunar valves in 18 of 19 specimens, at the conduit wall in 15 of 19, and at the valve edges in 10 of 19 patients. All identified endocarditic lesions consisted of thrombus material and fibrin condensations with included inflammatory cells (granulocytes and lymphocytes).

Conclusion: In a large series of explanted valved pulmonary conduits from patients with endocarditis, we demonstrate formation of typical endocarditic lesions macroscopically and histologically. Lesions at the base of semilunar valves were an almost constant finding, whereas the edges of the valves were affected much less frequent. This result further supports our assumption that the high incidence of endocarditis in bioprosthetic valves may mainly be explained by thrombus apposition at the valve base as a primary nidus for development of endocarditis. In contrast, the often supposed mechanism of endothelial injury due to shear stress at the edges of the valves may play a minor role—no evidence of endothelial damage was found in our series. Findings imply that intensified anticoagulation should be discussed for bioprosthetic valves in pulmonary position.