Validation of large animal models in mechanical valve research: a histologic comparison.

Abstract

Mechanical valves still require life-long anticoagulation. Preclinical animal testing is a crucial step in the assessment of valves; however, the chosen animal model should be carefully considered, and a well-controlled animal model of mechanical valve thrombosis has not been established yet. In this study, a histopathologic comparison was performed to evaluate the representativity of pigs and sheep as large animal models in bileaflet mechanical valve thrombosis research. 10 pigs and 8 sheep were implanted with a bileaflet mechanical valve in pulmonary position. During follow-up, no anticoagulative therapy was administered. Pigs were sacrificed between 14 and 38 days for explantation and assessment of the valve. Sheep were sacrificed between 71 and 155 days. Thrombus samples were processed and (immuno)histochemical stainings were applied. A pathologist evaluated the samples morphologically and semiquantitatively and compared these samples to available slides from 3 human patients who underwent redo surgery for acute bileaflet mechanical valve thrombosis, caused by insufficient anticoagulation. All pigs showed macroscopically evident thrombi on the mechanical valve surface at sacrifice. In contrast, none of the sheep showed any sign of thrombus formation. Histology showed a high fibrin content in thrombi of both human and porcine cases (3/3 vs 8/10). Porcine thrombi showed more cellular organization (0/3 vs 6/10), more calcification (0/3 vs 9/10) and more endothelialization (0/3 vs 6/10). Inflammatory cells were present in all samples and were considered physiological. Contrary to sheep, pigs develop thrombi on their mechanical valves in the short-term if no anticoagulation is administered. Histologic comparison of human and porcine thrombi shows comparable findings. The pig model might serve interestingly for further research on valve thrombosis, if it shows not to be an overly aggressive model.