Increased Mortality With Vena Cava Filters Stirs Controversy

July 13, 2018

The use of vena cava filters in patients with venous thromboembolism (VTE) and a contraindication to anticoagulation is associated with an increase in 30-day mortality, results of a new observational show.

"After adjustment there was a small but significant increase in all-cause mortality at 30 days in the patients who received a filter vs those that did not receive a filter.  As our study is observational and has some limitations we would describe the result as hypothesis generating, but we believe our data justifies a randomized controlled trial to answer the question," senior investigator, David L. Brown, MD, Washington University School of Medicine, St Louis, Missouri, told Medscape Medical News.

The study is published online July 13 in JAMA Network Open.

Too Many Limitations

However, authors of an accompanying commentary are not convinced by the study, noting that there are too many limitations to permit any conclusions. They also question whether a randomized trial would be feasible.

"To draw the conclusion that these devices may increase mortality is not appropriate from this trial. That conclusion goes against clinical practice and the data that is so far available," first author of the commentary, Eric A. Secemsky, MD, director of vascular intervention at Beth Israel Deaconess Medical Center, Boston, Massachusetts, told Medscape Medical News.

In the study, Brown and colleagues note that in patients with VTE and a contraindication to anticoagulation, many professional societies recommend consideration of a vena cava filter placement. The filters are widely used for this indication, although there are no data on a mortality benefit associated with these devices in this population.

An early randomized clinical trial showed a reduction in pulmonary embolism with vena cava filters, but this trial excluded patients with a contraindication to anticoagulation, which is now the most widely accepted indication for these devices, Brown and coauthors add. 

For the study, the researchers analyzed diagnostic code data on 126,030 hospitalized patients with venous thromboembolism and a contraindication to

anticoagulation from three states (California, Florida, and New York) from 2005 to 2012. In this cohort, 36.3% of patients were treated with a vena cava filter and 63.7% did not receive a filter.

Unlike in previous studies of vena cava filters, they used a time-dependent analysis to correct for the interval between hospital admission and filter placement, which they say can introduce bias and erroneously skew the results in favor of the intervention by falsely conferring a survival advantage to the treated group. Results were also adjusted by using propensity scores.

Results showed that filter placement was associated with a significantly increased hazard ratio of 30-day mortality (1.18; 95% confidence interval, 1.13 - 1.22; P < .001).

Too Many Uncertainties?

But Secemsky pointed out that relying on a claims-based database for the study introduces too many uncertainties for any conclusions to be drawn.

"You can't tell much about the patients from diagnostic code data. There will be many confounding factors that will be unaccounted for. It is likely that those who received the vena cava filter were higher risk as these are the patients we use the filters in most. 

"And we don't know if the patients were definitely not receiving anticoagulation as this information just came from diagnostic codes for conditions in which anticoagulation was likely to be contraindicated.  This is a big assumption to make. It is also more likely that those who didn't receive a filter were on anticoagulation, which would give them a benefit."

Secemsky agreed that the literature on vena cava filters is poor, but he noted that previous data lean toward a reduction in mortality. In addition,  there has never been a suggestion of an increase in mortality. 

He added that he does not believe a randomized trial is feasible.

"We use these devices in patients who have had a thromboembolism and can't take anticoagulation because it makes sense. There is a risk of another clot and these devices are designed to stop those clots going to the lungs and causing pulmonary embolism. Which clinician would risk a patient who had had a large clot being randomized to nothing? The study would end up being conducted in low-risk patients and therefore wouldn't answer the question properly," he said.

Secemsky believes a better way forward would be continued surveillance, with further observational studies using higher-quality data from a high-risk population.

He also pointed out that these filters are now retrievable and can be removed as soon as the patient can be anticoagulated.  For example, in the case of a patient who has undergone recent surgery, "That should make them safer as they will only be in place for a few weeks."

He also questioned the mechanism that may be behind an increase in mortality.

"These filters are there to protect the pulmonary artery from blood clots, so they should reduce pulmonary embolism-related deaths.  How a filter would cause death is a mystery," he said.

There can be procedural complications, but this is a very small risk, he added. In addition, Secemsky noted some thrombosis may be caused by the filter itself, but that would be very unlikely to be fatal. 

"This trial wasn't able to identify cause of death. My hunch is that this study is picking up other causes of deaths because of unaccounted-for confounding.

"My message to practicing physicians would be to be thoughtful on which patients these filters are used in and to take them out as soon as anticoagulation can be given. But I don't think we need to be too alarmed about a possible mortality increase."  

However, these objections have not dissuaded the study investigators about the need for a randomized trial.

Need for "Evidence Over Intuition"

Brown acknowledged the study had limitations.

"The reason we published is not to stop the use of these filters but to raise the possibility that they may be harmful and to provoke a discussion that could lead to a randomized trial." 

He added that vena cava filters "are not benign devices."

"There can be complications with them, and we don't have any scientific evidence to support their use in the population they are mainly being used in — those who have contraindications to anticoagulation.

"Those who say they wouldn't want to randomize a patient who can't take anticoagulation to not receiving a filter are making those statements based on belief — not on science. There are numerous examples of preconceived ideas being proven wrong in randomized trials. We need evidence over intuition."

"I think our current results are enough to justify a randomized trial. The National Heart, Lung, and Blood Institute has indicated that they would support such a trial, and if a funding agency of that stature was behind a trial I think that would convince physicians to randomize patients," he added.

This study was supported by grants from the National Center for Advancing Translational Sciences, the Agency for Healthcare Research and Quality, and the National Cancer Institute at the National Institutes of Health. Brown and Secemsky have disclosed no relevant financial relationships. Coauthor of the commentary, Robert Yeh, reports grants and personal fees from Boston Scientific and personal fees from Medtronic outside the submitted work.

JAMA Network Open.  Published online July 13.  Full text, Commentary

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.