Development of thrombosis models in the rabbit.

OBJECTIVE: In an effort to decrease the probability of thrombosis at microvascular anastomotic sites, researchers have examined techniques for decreasing the probability of thrombus formation at these sites. Identification of an appropriate animal model that is relatively easy to perform, reliable, and reproducible was attempted in this study. The goal of this study is to provide other investigators with a comparison and critical appraisal of various thrombosis models allowing a more directed selection of models for their own research.
DESIGN: A prospective, nonblinded animal study was designed that compared features of thrombosis models previously described in the literature with each other and with a new model (the "tuck" model), herein described. MAIN OUTCOME MEASURES: Ease of dissection, time to thrombosis, and other technical details were recorded. Additionally, validation of clot histologic features in the tuck model was performed.
RESULTS: The inversion graft models were found to be the most highly predictable and reliably thrombogenic, although technically quite difficult. The tuck model produces clot analogous to that produced clinically and is easy to perform, although a predictable number of vessels will not thrombose.
CONCLUSIONS: The results of this study comparing various animal models of anastomotic thrombosis suggests that the tuck model is the model of choice for most thrombus prevention studies. When predictability of thrombosis is a critical factor, the inversion graft model should be chosen.