IN-VITRO EVALUATION OF THE DIBIE-MUSSET V ENA-CAVAL FILTER

The Dibie-Musset (DM) vena caval filter was evaluated on a hydraulic t est bench reproducing flow conditions in the inferior vena cava : pres sure, flow, vicosity, diameter and elasticity of the conduit. The resu lts were compared with those obtained with the Greenfield filter (GF). In addition to classical measurements (captation and loss of load) we measured the velocity profile with a Doppler ultrasonic probe proxima l and distal to the filter to study flow conditions before and after e mbolisation of clots. In order to circumvent the difficulties encouter ed with the use of real thrombi, chemical gels with visco-elastic prop erties, evaluated by viscosimetry, similar to those of blood clots, we re used. Clots 45 mm long and 4 mm diameter were injected in several s eries of measurements. The DM filter.:was stable and did not migrate. In the horizontal position (flexible conduit) the DM filter was signif icantly more effective than the GF for less than 5 clots injec ted suc cessively. The filtration capacity of both filters decreased with the number of clots captured. In the vertical position (rigid conduit), wh en there are less than 5 clots injected the two devices were perfectly effective. There was no significant difference between the two filter s when 10 clots were injected. The loss of load resulting from the pre sence of the filter and clots was greater with the DM filter because o f the greater captation capacity. However, the velocity profile distal to the filter was less disturbed with the DM filter because there was a more uniform distribution of the clots captured over the surface of the filter. In addition, the risk of clot migration was less because the shear stress was smaller and the turbulence was less severe.