Abstract
Large animal models are a critical component of the preclinical evaluation of mechanical cardiac implants, enabling assessment of safety and performance under physiological conditions that cannot be adequately reproduced in vitro. Choosing a suitable animal model is important for both scientifically valid and ethically responsible preclinical evaluation. However, interspecies differences between animal models and humans pose significant challenges for relevant translation of preclinical findings to clinical outcomes. This narrative review provides a comprehensive overview of commonly used large animal models (sheep, goats, pigs, and calves) for the preclinical assessment of mechanical cardiac implants, including prosthetic heart valves, ventricular assist devices, and total artificial hearts. We summarize key anatomical and physiological characteristics that influence device implantation, chronic follow-up, and translational value. Emphasis is placed on three critical outcome domains for preclinical evaluation of mechanical cardiac implants: calcification, thrombogenicity, and hemodynamic performance. Species- and age-dependent differences in calcification are reviewed, identifying juvenile sheep as a worst-case model for early manifestation and detection of graft mineralization. Interspecies differences in coagulation biology are examined, showing attenuated platelet responses in sheep and closer similarity between porcine and human platelet behavior, supporting pigs as the preferred thrombogenicity model. Hemodynamic evaluation strategies in acute and chronic large-animal studies are discussed, with particular emphasis on circulatory demands influenced by somatic growth and on device adaptability under varying loading conditions. Overall, this review provides practical, outcome-driven guidance for large animal model selection and experimental design in mechanical cardiac implant research, while identifying key limitations, knowledge gaps, and the need for standardized reporting to improve the translational reliability of preclinical studies. Based on the findings presented in this review, we conclude that there is no single animal model capable of evaluating all relevant aspects of a device. Instead, different animal models provide distinct advantages depending on the outcomes of interest.