‘Disconnect’ blamed for potential overuse of inferior vena cava filters

Inferior vena cava filters have become an increasingly popular tool to reduce risk for recurrent pulmonary emboli since their introduction 50 years ago.

The retrievable, umbrella-like devices — inserted into the inferior vena cava, just below the kidney — are designed to capture pulmonary emboli that break loose from deep veins in the legs before they reach the heart and lungs, where they can be fatal.

Physicians implanted inferior vena cava (IVC) filters in more than 1.1 million Americans between 2005 and 2014.

The devices often are used for patients with cancer, whose risk for venous thromboembolism is six times greater than the general population.

However, they are not without risks. The FDA reviewed reports about hundreds of filter-associated adverse events — including thrombosis, device migration and fracture — and issued a safety communication that recommended physicians remove the devices from patients when they no longer needed PE protection.

Nearly 70% of patients with cancer who receive IVC filters do not have them removed, increasing the potential for complications. It also is unclear whether IVC filters improve survival.

Photo courtesy of UC Davis Health.

“There is some debate as to their overall efficacy,” Ted Wun, MD, FACP, director of the UC Davis Clinical and Translational Science Center and chief of the division of hematology and oncology at UC Davis School of Medicine, told HemOnc Today. “There is still this notion that patients with cancer who are sicker and more susceptible to dying of a PE are better candidates for IVC filters simply because they can’t tolerate another PE. So, in addition to anticoagulation, putting in a filter might prevent death. But there has never been proof of that and, even in retrospective studies, the data are not very robust.”

HemOnc Today spoke with researchers and hematologists about which patients benefit most from IVC filters, the effect the FDA recommendations have had on their use, why device use and retrieval remain inconsistent, and whether new therapies may be safer alternatives for patients with elevated PE risk.

Birth of retrievable filters

Anticoagulants such as low-molecular-weight heparin, vitamin K antagonists, and activated factor X and thrombin inhibitors have long been standard treatment for thromboembolic diseases.

However, those agents carry a high risk for serious adverse events, such as intracranial hemorrhage and gastrointestinal bleeding. Anticoagulation is contraindicated for many patients because of these risks, and therapy is ineffective for others.

“The most appropriate use of an IVC filter would be in a patient who already has a blood clot and cannot get therapeutic anticoagulation,” Alok A. Khorana, MD, professor of medicine at Cleveland Clinic Lerner College of Medicine, vice chair for clinical services at Taussig Cancer Institute and director of the gastrointestinal malignancies program at Cleveland Clinic, told HemOnc Today.

“Let’s say a patient has a deep vein thrombosis in the lower extremity but also massive bleeding from a primary gastric cancer or a bleed into the central nervous system and can’t get heparin or one of the oral anticoagulants because of the possibility of bleeding,” Khorana added. “In that setting, it’s very appropriate to put in an IVC filter to at least reduce the risk for PE without anticoagulation.”

The bird’s nest IVC filter — a tangled device made of stainless steel — was introduced in the mid-1980s as a permanent method to intercept blood clots. Other permanent devices, some made of titanium, entered clinical practice shortly thereafter and were inserted percutaneously under fluoroscopic control through a femoral or jugular vein.

“We’d place these permanent devices in the inferior vena cava and they’d be in place forever,” Robert Lewandowski, MD, FSIR, associate professor of radiology at Northwestern University Feinberg School of Medicine, told HemOnc Today.

An IVC filter — which more closely resembles a trap — “theoretically catches clots as they’re flying up to the lungs,” reducing the risk for potentially fatal PEs, Wun said.

“But the filter itself is thrombogenic because any foreign material put into a person’s blood forms clots,” he added. “Clots can be formed on the pointy side of the umbrella and break off into the lungs. So, whether they truly decrease the incidence of pulmonary emboli as compared to not putting them in has not been definitely shown.”

The number of patients with a permanent IVC filter increased from 2,000 in 1979 to 49,000 in 1999, according to data from Stein and colleagues.

However, 8-year follow-up data from the randomized PREPIC study showed the addition of a permanent IVC filter to standard anticoagulation reduced PE incidence (6.2% vs. 15.1%; P = .008) but increased DVT incidence (35.7% vs. 27.5%; P = .042).

“IVC filters prevented PE in the short term, but they had no significant impact on mortality, which is really what we are trying to prevent with these devices,” Lewandowski said. “And, patients with IVC filters were at higher risk for deep vein thrombosis in the legs. This spawned the advent of retrievable IVC filters, which provided the benefit of mechanical prophylaxis against PE in the short term while potentially mitigating the risk for DVT in the long term.”

The FDA approved the use of retrievable IVC filters in 2003. After that, the placement rate increased 22.2% in 5 years, from 45.2 per 100,000 individuals in 2005 to 55.1 per 100,000 in 2010 (P < .001), according to data from Reddy and colleagues published this year in a research letter in JAMA Internal Medicine.

Uncertain cancer role

VTE is the second most common cause of mortality among patients with cancer. It also is the leading cause of mortality within 30 days of cancer surgery, accounting for 46.3% of such deaths.

“Many of the interventions we undertake in patients with cancer include major surgeries, chemotherapy and blood transfusions, all of which are known to increase the risk for thrombosis,” Khorana said. “A patient already is at a pre-existing high risk because they have cancer, and then all the things we do to them further increases the risk.”

Despite the need in this population, the role of IVC filters is questionable due to a lack of study data.

“Unfortunately, the majority of IVC filter trials have specifically excluded patients with cancer,” John Moriarty, MD, director of the IVC filter clinic and associate professor of radiology and medicine at David Geffen School of Medicine at UCLA, told HemOnc Today. “The only real evidence we have for use in these patients comes from retrospective registries and small studies. Because the level of evidence is relatively little, we do not have a great handle on when exactly we should be inserting IVC filters and when we should not.”

In a study published in 2016 in Thrombosis Research, Wun and colleagues reported that — from 2005 through 2009 — 19.6% of 14,000 patients with cancer and VTE received IVC filters. Higher rates occurred among those with brain cancer (OR = 4.6; 95% CI, 3.7-5.6), those who underwent major surgery (OR = 4.9; 95% CI, 3.9-6.1) or those with bleeding (OR = 2.7; 95% CI, 2-3.5).

Still, a survival benefit is unclear.

In a study published in 2017 in Thrombosis Research, Wun and colleagues analyzed data from 2,747 patients with cancer hospitalized with acute DVT or PE.

Results showed use of IVC filters placed between 2005 and 2009 did not improve 30-day mortality (HR = 1.12; 95% CI, 0.99-1.26) or the adjusted 180-day risk for subsequent PE or DVT (HR = 0.81; 95% CI, 0.52-1.27). Filter use increased the adjusted 180-day risk for recurrent DVT (HR = 2.1; 95% CI, 1.53-2.89).

A retrospective study by Barginera and colleagues, published in Clinical and Applied Thrombosis/Hemostasis, examined the effect of IVC filters on survival among 206 patients with cancer and VTE. Results showed significantly longer median OS among patients treated with anticoagulation (13 months) than those treated with IVC filters (2 months) or a combination of anticoagulation and IVC filters (3.25 months; P < .0001).

“Nobody has done a study to show a mortality benefit in patients with cancer but, if I had to guess, I would say there is not a reason for them to be ineffective in every other population but suddenly effective in this one,” Ido Weinberg, MD, MSc, FACC, assistant professor of medicine at Massachusetts General Hospital, told HemOnc Today. “Despite the widespread use of these devices and the fact they make a lot of clinical sense — if you stop the clot, you’re going to save lives — that is not how the data pan out.”

Effect of FDA warning

The use of IVC filters decreased considerably following the FDA’s 2010 device safety warning.

Over a 5-year period, the FDA received reports about 921 filter-associated adverse events. They included device migration; filter fracture; embolization, or movement of the entire filter or fractured fragments to the heart or lungs; perforation of the IVC; and difficulty removing the device. These risks are associated with long-term use of retrievable IVC filters, which are intended for short-term use.

In their research letter, Reddy and colleagues reported a 29% decrease in the rate of IVC filter placements following the advisory — from 55.1 per 100,000 individuals in 2010 to 39.1 per 100,000 in 2014 (P < .001).

The advisory also contributed to increased litigation.

“A decade ago, people would say part of the reason for the increase in IVC filters was medical-legal,” Wun said. “If somebody had a PE after a surgery and the surgeon did not put one in, they would be faulted for not doing so. It’s somewhat the opposite now. Internet ads try to get people who have had IVC filters to take legal action.”

The FDA updated its safety communication in 2014, citing long-term risks for lower limb DVT and IVC occlusion. The agency again recommended that physicians and clinicians responsible for the ongoing care of patients with retrievable IVC filters consider removing them as soon as possible.

The FDA communication included data from Morales and colleagues — published in Journal of Vascular Surgery: Venous and Lymphatic Disorders — which identified 29 to 54 days after implementation as the appropriate timeframe for filter removal.

ASH — as part of its participation in American Board of Internal Medicine Foundation’s Choosing Wisely initiative, a campaign intended to eliminate unnecessary procedures and reduce expenses without sacrificing quality care — recommended against routine use of IVC filters in patients with acute VTE. When IVC filters are used, ASH recommends retrievable instead of permanent filters, with removal as soon as the risk for PE has resolved or a patient can resume anticoagulation.

Despite this guidance and the FDA safety communication, research suggests filters still are used improperly. Although timely removal is key to reducing risks, data suggest they routinely are left in place.

“There is certainly a disconnect,” Lewandowski said. “Not every clinician understands IVC filters, and there are many physicians who still don’t know there is a difference between permanent and retrievable devices. There are also physicians who don’t understand these devices can and should be removed when no longer indicated.”

Sarosiek and colleagues conducted a review of 679 patients implanted with retrievable filters. Results showed only 58 (8.5%) were successfully removed. Retrieval attempts failed in 13 of 71 patients (18.3% of attempts) because the filter embedded in the IVC (n = 8), protruded through a blood vessel (n = 3), had an abnormal position (n = 2) or had a clot (n = 1). Median retrieval time after placement was 122 days (range, 2-1,931 days).

An analysis by Abtahian and colleagues, published in 2014 in American Journal of Medicine, showed patients with cancer were less likely to have filters retrieved (28% vs. 42%; P < .001).

“These people are between a rock and a hard place,” said Weinberg, senior researcher on that study. “Unfortunately, patients with cancer already have a high risk for bleeding even on anticoagulation, and many times you can’t retrieve these devices because they may be needed long term. The fact is, many of the filters that need to come out don’t come out.”

Still, it is proposed that IVC filters can prevent between 97% and 99% of PEs, and they can remain effective at significant PE prevention for as long as they are in the body, Moriarty said.

However, these benefits must be balanced with the risks that increase with longer use.

“Patients who are recovered and have had an IVC filter for several months or several years are often told they cannot be removed because it’s been in for years, and that is not true,” Moriarty said. “That filter can and should be removed, and the success rates are very good. However, the longer they are within the body, the rate of DVT increases, as does the rate of filter fracture and embolization. That’s why we should try to remove it to prevent any damaging problems as soon as the patient no longer requires the IVC filter.”

Because of these risks, as well as cost — which is about $3,000 per filter before considering radiology costs — appropriate use is essential.

“These devices are often inappropriately utilized, and that is a really big concern,” Khorana said. “They are expensive and it’s an invasive procedure.”

Despite efforts to curb inappropriate use, the placement of IVC filters in the United States in 2012 was 25 times greater than in Europe (224,700 vs. 9,070), according to a study published in Annals of Medicine.

One reason for the disparity could be more liberal use of pharmacological prophylaxis in Europe, where surgeons often will administer anticoagulants prior to surgeries, Wun said.

“I’m not a zealot against filters because there are appropriate places for their use,” he said. “But we tend to be prejudiced toward invasive procedures and doing something active, which is an American trait in general. That’s not necessarily bad, but we should question our practices when essentially the rest of the world that has the ability to [use filters] is choosing not to.”

Permanent vs. retrievable filters

Because of the risks associated with retrievable filters, permanent IVC filters are still used for certain patients.

“What we have seen historically is very low device retrieval rates combined with time-dependent device complications,” Lewandowski said. “A lot of the current consternation over IVC filters has to do with retrievable filters that are not removed.”

However, determining which filter is appropriate remains a challenge, particularly in the cancer setting.

Permanent IVC filters — made of titanium, stainless steel or a nickel titanium called nitinol — are designed to be placed in the IVC and remain there, developing more wall contact and scar tissue within the vena cava. Retrievable devices generally are less robust in structure, and hence they tend to move and fracture more easily than permanent devices, Moriarty said.

Although most oncologists recommend having IVC filters retrieved within 3 weeks of insertion, there are cases in which filters can serve a purpose for extended periods.

“Knowledge of life expectancy is very important for patients with cancer,” Moriarty said. “In patients where the likelihood is that they will not recover from their cancer, we should use a permanent filter. Alternatively, if a patient has a very good expected outcome or is expected to recover completely, we should use a temporary filter and retrieve it as soon as the patient’s risk for PE has reduced.”

When patients require an IVC filter, a multidisciplinary discussion among clinicians is imperative to assess the necessity for mechanical PE prophylaxis and determine if a permanent or retrievable filter is indicated. If a retrievable device is chosen, it is ideal to develop a target time for when to take the device out, Lewandowski said.

A cancer patient with venous thromboembolic disease and a good prognosis undergoing surgery who needs temporary mechanical prophylaxis against PE would be a good candidate for a retrievable filter, as long as the filter is removed as soon as feasible, he added.

“It’s more challenging when you have end-of-life scenarios and you have patients with bad prognoses who develop DVT or PE,” Lewandowski said. “These patients are hypercoagulable because of their cancer and many are more likely to develop blood clots because of immobility. How do you manage those patients? An IVC filter could be placed in someone with advanced cancer and they pass away a day or 2 later. Was that the appropriate thing to do?

“That may be a societal question and the literature is not so clear,” he added.

According to a study by Jarrett and colleagues, published in Journal of Vascular Surgery, 46% of 91 patients with stage IV cancer who received an IVC filter died within 6 weeks.

A study by Schunn and colleagues, published in Vascular and Endovascular Surgery, showed 23.6% of 55 patients with stage III or IV cancer who received an IVC filter survived less than 30 days. However, the filter prevented PE among 94.5% of patients; complications occurred in 7.3%.

“At least in our group, almost no one puts permanent filters in anymore and, among patients with cancer, a very low percentage of temporary filters are ultimately removed,” Wun said. “Oftentimes, patients with advanced cancer may get to a point where it’s not felt to be worth it to expose them to another procedure, or they may have died before a scheduled time to take it out.”

Even in the noncancer setting, such as after trauma, patients with retrievable filters do not come back for follow-up and never have the filter removed, Wun added.

“Researchers have advocated that, because so few temporary filters get removed, there is no point in putting those in, and you might as well put in a permanent filter,” he said.

More data expected

To learn more about the safety and efficacy of IVC filters, the Society of Interventional Radiology and the Society of Vascular Surgery are conducting PRESERVE, a 5-year study of more than 1,800 patients from approximately 60 U.S. centers.

About 300 patients will be assigned an IVC filter from one of six different manufacturers. Patients will undergo evaluation every 6 months for up to 24 months or until filter retrieval.

Weinberg said he is anxious to see the results of that trial, particularly whether the introduction of a new device called the Angel Catheter (Bio2 Medical) — an IVC filter that can be inserted and removed through a central venous catheter — might reverse what he considers a trend away from the use of filter devices for patients with cancer.

Although he acknowledges there is some “misuse” of IVC filters in the United States, Khorana said better communication among all specialties is essential.

“Often, it’s not the oncologists’ fault,” he said. “Surgeons may not be as well versed of all the appropriate indications and pitfalls of IVC filters. If there is overuse, and many people think there is, then it’s due largely because of non-hem/onc specialists.

“It’s important to have all of the disciplines being involved instead of one person making the decision,” he added. “We have to consider the big picture as to what else is going on with the cancer and, if it is placed, make sure there is a plan for retrieving it at an appropriate time.”

Many oncologists agree that for patients with cancer who have underlying lung or heart disease or are facing brain, ovarian or gastric bypass surgery, a case can be made for the prophylactic use of IVC filters.

But, with the promise of anticoagulants such as heparin and vitamin K antagonists like warfarin, Wun said the use of IVC filters could decline dramatically.

“Compared with filters, there will be large, randomized studies in the use of pharmacological treatment and prophylaxis in patients with cancer,” Wun said. “I believe those studies are going to show that, although the adverse events might be higher, pharmacological prophylaxis in most patients with cancer is still the main therapy.” – by Chuck Gormley

Click here to read the , “Does an indication exist to place IVC filters in patients with cancer who can receive anticoagulation?”

References:

Barginear MF, et al. Clin Appl Thromb/Hemo. 2008;doi:10.1177/1076029608315165.

Brunson A, et al. Thromb Res. 2017;doi:10.1016/j.thromres.2017.

Brunson A, et al. Thromb Res. 2016;doi:10.1016/S0049-3848(16)30112-8.

FDA. Inferior vena cava (IVC) filters: initial communication: risk of adverse events with long-term use. Available at: wayback.archive-it.org/7993/20161022180008/http:/www.fda.gov/medicaldevices/safety/alertsandnotices/ucm221676.htm. Accessed July 31, 2017.

Jarrett BP, et al. J Vasc Surg. 2002;doi:10.1067/mva.2002.127958.

Morales JP, et al. J Vasc Surg Venous Lymphat Disord. 2013;doi:10.1016/j.jvsv.2013.05.005.

PRECIP study group. Circulation. 2005;112:416-422.

Reddy S, et al. JAMA Intern Med. 2017;doi:10.1001/jamainternmed.2017.2719.

Sarosiek S, et al. JAMA Intern Med. 2013;doi:10.1001/jamainternmed.2013.343.

Schunn C, et al. Vasc Endovascular Surg. 2006;doi:10.1177/1538574406291821.

Stein PD, et al. Arch Intern Med. 2004;164:1541-1545.

Wang SL, et al. Annals Med. 2013;doi:10.3109/07853890.2013.832951.

For more information:

Alok A. Khorana, MD, can be reached at khorana@ccf.org.

Robert Lewandowski, MD, FSIR, can be reached at r-lewandowski@northwestern.edu.

John Moriarty, MD, can be reached at jmoriarty@mednet.ucla.edu.

Ido Weinberg, MD, MSc, FACC, can be reached at iweinberg@mgh.harvard.edu.

Ted Wun, MD, FACP, can be reached at twun@ucdavis.edu.

Disclosures: Khorana, Lewandowski, Moriarty, Weinberg and Wun report no relevant financial disclosures.