Even Without Proof of Benefit, 1 in 6 Patients Still Given Vena Caval Filters

Despite the absence of any substantive evidence proving that patients with pulmonary embolism (PE) obtain a survival benefit from inferior vena caval filters, use of the devices has steadily increased in recent years as the number of patients hospitalized with the condition has grown.

Even Without Proof of Benefit, 1 in 6 Patients Still Given Vena Caval Filters

“The guidelines that are available from professional societies recommend use of filters only in very limited situations,” lead author Behnood Bikdeli, MD (Yale-New Haven Hospital, CT), explained to TCTMD. “However, as clinicians at bedside, not uncommonly we’ve witnessed that a lot of people have been placing filters for softer indications, if you will, for a broader cohort of patients.”

The fact that the design of these devices has changed over time to make them retrievable has potentially also made them more appealing to operators.

Bikdeli and colleagues, including senior author Harlan M. Krumholz, MD (Yale-New Haven Hospital), looked at 556,658 hospitalizations of Medicare beneficiaries treated for PE between 1999 and 2010. Overall, 16.9% underwent inferior vena caval filter placement, but patients were more likely to receive filters as the study period progressed. Between 1999 and 2010, the rate of PE hospitalizations with filter placement per 100,000 beneficiary-years increased from 19.0 to 32.5 (P < 0.001). However, while the rate of filter placement per 1,000 PE hospitalizations also rose (from 157.6 to 164.1), the difference did not reach significance (P = .11).

The study was published online this week ahead of print in the March 8, 2016, issue of the Journal of the American College of Cardiology.

While Bikdeli said that his team expected to see filters used in a “decent proportion” of patients, what they found was “probably a little bit too [high].” He indicated that his team has also compared US filter use with that in Europe—full results are will be released in a forthcoming paper—where only around 3-4% of PE patients undergo this procedure.

Mortality Not Affected by Filters

Mortality in both the short- and long-term declined in all subgroups of patients with PE over the decade studied. Among those who received filters, adjusted mortality rates declined in-hospital and at 30 days, 6 months, and 1 year (P ≤ .001 for trend for all).

Because mortality declined irrespective of filter placement, Bikdeli said, it is “probably not the case that filters save hundreds of thousands of lives based on what we can indirectly see from our cohort.” On the flip side, inappropriate procedures are costly and could result in recurrent deep vein thrombosis due to the thrombogenicity of the filter site itself, he added.

Patients aged 85 years and older saw the greatest relative increase in filter use over the course of the study, and men in general were more likely to receive filters than women. Procedural rates in black patients declined over time, but this subgroup was still most likely to receive filters than any other racial subgroup throughout the study.

Regionally, use of inferior vena caval filters varied widely, with greatest use in the South Atlantic region and lowest use in the Mountain region. The only region where filter use declined over time was New England.

The regional variability, according to Bikdeli, was more likely resultant from different practice patterns and the way practitioners “think about the condition rather than the patients being inherently different.”

Thoughts on a Randomized Trial

In an accompanying editorial, James E. Dalen, MD (University of Arizona College of Medicine, Tucson), and Paul D. Stein, MD (Michigan State University, Lansing), write that while Bikdeli et al do not “give the complete answer regarding whether [inferior vena caval filters] reduce mortality from PE,” the researchers are “on target” in suggesting that further subgroup investigation would be helpful.

Going forward, Dalen and Stein say, both older low-risk patients in stable condition and those of all ages in unstable condition need to be studied more, although it is “extremely unlikely” that a randomized trial of vena caval filters in PE patients will be performed, given past enrollment difficulties and the fact that “recruitment would be difficult if not unethical in view of the results we already have.”

Bikdeli argued that while a randomized trial is unlikely, “it’s not impossible.” More so, it’s a “question of having the will to do it,” he added, but until then, high-quality observational data will have to be enough.

For now, he said, “I hope people would be more cognizant of the slim evidence for benefit and use [filters] more judiciously.”

Bikdeli B, Wang Y, Minges KE, et al. Vena caval filter utilization and outcomes in pulmonary embolism: Medicare hospitalizations from 1999 to 2010. J Am Coll Cardiol. 2016;67:1027-1035.
Dalen JE, Stein PD. Is there a subgroup of PE patients who benefit from inferior vena cava filters? J Am Coll Cardiol. 2016;67:1035-1037.

Related Stories:


  • Krumholz reports receiving research agreements from Medtronic and Johnson &amp; Johnson and serving as chair of UnitedHealth’s scientific advisory board.
  • Bikdeli, Dalen, and Stein report no relevant conflicts of interest.

We Recommend