Purpose: To evaluate the clot trapping capacity of different temporary
vena cava filters in a vena cava model. Methods: A vena cava flow mod
el was built using PVC tubing, a hemodialysis membrane and a pulsatile
pump. Blood was imitated by a Dextran 40 solution. Five different tem
porary vena cava filters and two prototypes were tested using human th
rombi. The mechanism of clot capture was observed. Results: Decreasing
rank order according to decreasing percentage of clots captured for t
he 21-mm diameter vena cava model was Cook (C) > Angiocor (A) > Cordis
(CD) > Antheor (TF-6) > DIL for thrombi with a diameter of 3 mm and A
> C > CD > TF-6 > DIL for 5-mm thrombi. In a cava with diameter of 28
mm, decreasing rank order was C > CD = A > TF-6 > DIL and C > CD = A
> DIL > TF-6 for 3- and 5-mm thrombi, respectively. Two new prototypes
, the TF-8 and TF-10 filters, achieved better results than the TF-6 fi
lter and were in most conditions comparable to the A and CD filters. I
n most cases, thrombi were trapped between filter and cava wall. Concl
usion: The vena cava flow model demonstrates significant differences i
n rates of clot capture (range 22%-98%) depending on cava diameter, th
rombus size, and filter type.