Development and evaluation of investigational vena caval filters: the complementary roles of in vitro and in vivo studies.

BACKGROUND: The use of animals in medical research raises both ethical and economic concerns. Validated in vitro models allow extensive device testing prior to in vivo studies which limits the numbers of animals required to reach conclusions regarding efficacy and safety.
MATERIALS AND METHODS: In vitro studies were conducted in a vena caval simulator (ECU) to evaluate characteristics of filter performance: ability to capture emboli, resistance to vertical movement from lateral forces, resistance to vertical movement from mass impact, and ability to maintain caval flow. These tests are used to select candidate filters for in vivo studies. In vivo filter studies are conducted in an ovine model with filters placed into the vena cava from a jugular or femoral vein approach. At intervals, the animals are sacrificed and both cava and device are evaluated for thrombogenicity, thrombus resolution, and mechanical integrity. Radiographs, intravascular ultrasound, and histology are used to determine outcome.
RESULTS: During the 6-year period from 1993 to 1999, 11 vena caval filter types were evaluated. In each case, it was possible to select a single prototype for in vivo study on the basis of four tests. An investigational device was developed that was comparable to the marketed devices with respect to clot capture (P = 0.001), superior to the Greenfield in resistance to movement (P = 0.001), and superior to the VenaTech and Simon Nitinol relative to volume loading (P = 0.01). Problems with metal fatigue prevented clinical development.
CONCLUSION: An integrative system of in vitro and in vivo studies allows efficient evaluation of IVC filters and identifies design characteristics associated with clinical functioning. Evaluation over time is essential to determine safety as well as efficacy.