Thrombogenicity Testing for Blood-Contacting Medical Devices in an In Vitro Human Blood-Loop

Thrombogenicity testing continues to be a critical requirement for regulatory approval of blood-contacting medical devices and the ISO guidelines have recently been updates [1]. This new guideline ascribes value to both in vivo and in vitro testing including both the non-anticoagulated venous implant (NAVI) model, and the new methods for in vitro testing. One challenge with the animal-blood-based in vitro assays that have been validated and used for submissions is that they still may not accurately translate to clinical safety or predict the risk for thrombogenic potential in humans. We have previously described a model using minimally heparinized ovine blood and are continuing to improve the overall methodology [2,3]. In addition, we have transferred these methods to a human blood assay which therefore has enhanced potential for prediction of clinical risk. As with the ovine model, the key characteristics of a successful in vitro method include fresh blood, low levels of anticoagulation, flow conditions and minimization of air/blood interfaces. This human model integrates freshly harvested human blood containing minimal levels of heparin with variable flow from a unidirectional peristaltic pump and unlike many of the human blood assays, it can accommodate larger devices and higher flow rates than previously described [1,4]. Control materials which were optimized in the ovine model were also used to reproducibly elicit positive and negative thrombogenic responses. We feel that this model can be used for validation of the ovine model with cross comparisons of a number of legally marketed comparator devices. Alternatively, if the human blood methodology can be streamlined and performed cost effectively on a regular and basis, this assay could supplant the current ovine model and allow a highly predictive preclinical test for thrombogenicity of devices.