A design modification to minimize tilting of an inferior vena cava filter does not deliver a clinical benefit

In July 2007, our group began to use a modified conical inferior vena cava filter with additional stabilizing struts designed to reduce tilting of retrievable filters. We analyzed our experience with this modified filter (Cook Medical, Bloomington, Ind) from July 1, 2007 to December 31, 2008 and compared it to our experience with the standard filter (Günther Tulip, Cook Medical, Bloomington, Ind) from January 1, 2006 through December 31, 2008 to determine if adoption of the modified filter reduced tilting and delivered a discernible clinical benefit.

The primary outcome measure was tilt angle after deployment. Secondary outcomes were change in tilt angle between deployment and retrieval (self-centering) and retrieval failure due to inability to engage the filter hook. Measurements were retrospectively determined using the anteroposterior venogram at the time of placement and removal. Tilt angle was defined by the center line of the filter relative to the center line of the inferior vena cava (IVC). Statistical significance was assumed for P ≤ .05.

During the study period, a total of 302 IVC filters were placed. Retrieval was attempted for 85 of 194 (44%) standard filters and 52 of 108 (48%) modified filters. The overall difference in tilt angle (degrees) between the standard (median [interquartile range] = 5 [3, 8]) and modified (5 [3, 8]) filters at the time of placement was not statistically significant (P = .44). Modified filters deployed through a femoral route (8 [4, 11]) had significantly greater tilt angles than modified filters deployed using jugular access (4 [2, 6]; P < .0001). At the time of retrieval, evidence of self-centering was observed more often with modified (32 of 52 [62%]) than standard (36 of 85 [42%]) filters (P = .03). Overall, there were only four failures to retrieve the filter due to excess tilting (standard, 3 of 85 [4%], modified, 1 of 52 [2%]; P = .59).

Overall, tilt angle at insertion did not differ between the modified and standard filters, although more modified filters displayed self-centering. There was no difference between the groups in retrieval failure due to excess tilting. Despite its greater tendency to self-center, we did not recognize a measurable clinical advantage of the modified filter.