Inferior vena cava filters: current best practices
- 809 Downloads
Venous thromboembolism (VTE) is a common cause of inpatient and outpatient morbidity and mortality. While anticoagulant therapy is considered the primary means of prevention and treatment of VTE, inferior vena cava filters (IVCFs) are often used as an alternative or adjunct to anticoagulation. With the advent of retrievable filters indications have liberalized, to include placement for primary prophylaxis in high-risk patients. However, this practice is based on limited evidence supporting their efficacy in preventing clinically relevant outcomes. Since indiscriminate use of IVCFs can be associated with net patient harm and increased health care costs, knowledge of the literature surrounding IVCF utilization is critical for providers to adopt best practices. In this review, we will provide an overview of the literature as it relates to specific clinical questions that arise when considering IVCF utilization in the prevention and treatment of VTE. Practice-based recommendations will be reviewed to provide the clinician with guidance on challenging clinical scenarios.
KeywordsInferior vena cava filter Deep venous thrombosis Venous thromboembolism Pulmonary embolism Anticoagulation
VTE, including deep vein thrombosis (DVT) and pulmonary embolism (PE), is a major cause of morbidity, morbidity, quality of life impairment, and health care costs. It is the most common cause of preventable hospital mortality [1, 2, 3]. Effective thromboprophylaxis is essential as the clinical diagnosis of VTE is often difficult and treatment is frequently delayed or inadequate. Anticoagulants are the standard of care for both prevention and treatment ; however, alternative mechanical prophylaxis is employed in those with active or perceived high risk of bleeding. In these scenarios, clinicians are faced with the decision of whether or not to utilize an inferior vena cava filter (IVCF) to prevent a potentially fatal PE. The purpose of this review is to provide a comprehensive overview of current best practices for IVCF utilization. Where possible, practice recommendations that are evidence-based will be provided to address common clinical questions surrounding IVCF use. However, there is clearly a lack of strong evidence in many areas related to IVCFs, in which case, expert opinion will be highlighted.
What are practice patterns of IVCF placement?
Despite the paucity of reliable data on efficacy and safety, indications for IVCFs have liberalized. Since FDA approval of the first retrievable IVCF (rIVCF), utilization has increased dramatically . The largest proportional increase in the use of rIVCFs in the United States has been in patients at risk for PE but who have neither PE nor DVT (i.e. prophylactic IVCFs) ; a trend driven in part by use for extended indications and improved filter technology, which facilitates easy percutaneous insertion and removal . While detection of VTE due to improved diagnostic imaging has increased linearly in the last decade, placement of IVCFs for primary prophylaxis increased superlinearly, increasing by 7-fold from 2003 to 2006 compared to 1985 to 2003 . IVCF insertion has increased significantly during the past 10 years in trauma patients [6, 7]. Over the last decade IVCF placement increased by 111 % in the Medicare population . Data from the National Hospital Discharge Survey indicate that 49,000 hospitalized patients received an IVCF in 1999 . Based on the most recent data from this registry, 92,000 patients received IVCFs in 2006, representing an almost 200 % increase . In 2012, vena cava filter sales were expected to exceed 250,000 of which 80 % (200,000) were expected to be retrievable filters . There is considerable national variation in use of IVCFs that cannot be explained by patient or center characteristics. Some have suggested that that medical insurance coverage and medical-legal environments may play a role in higher utilization. This is evidenced by the significantly increased Centers for Medicare and Medicaid reimbursement rates when the inpatient diagnosis related group (DRG) is modified for a patient admitted for DVT if an IVCF is placed. Outpatient retrieval is reimbursed separately and may partially explain the increasing deployment of these devices [10, 11, 12, 13, 14, 15]. Internationally, US providers place far more IVCFs compared to European providers by a staggering 25:1 IVCF placement ratio with no significant difference in annual VTE-related deaths [16, 17, 18].
Are IVCFs effective in preventing pulmonary embolism?
The sole purpose of an IVCF is to prevent a potentially fatal PE from a lower extremity DVT. However, unlike anticoagulant therapy, whose efficacy and safety have been demonstrated in prospective randomized controlled trials (RCTs), the benefit derived from IVCFs remains unproven. Unfortunately, most data on the efficacy of IVCFs originates from nonrandomized heterogeneous case series or retrospective population-based studies with varying results and obvious limitations. To date, the only RCT evaluating the incremental efficacy of IVCFs in the management of VTE is the PREPIC study [19, 20]. This trial reported mixed results. Patients with acute proximal DVT were randomized to anticoagulation versus anticoagulation and placement of a permanent IVCF. At 8 years with 35 % of patients remaining on anticoagulation, symptomatic PE was less frequent in IVCF recipients than in those treated with anticoagulation alone (6.0 vs. 15.0 %; P = 0.008). DVT was more frequent among IVCF patients (36.0 vs. 28.0 %; P = 0.042). No difference in overall survival was reported. The authors of a Cochrane Collaboration systematic review published in 2010 concluded that no recommendations could be drawn from their analysis of the literature and that further studies addressing the efficacy and safety of IVCFs are needed .
A second multicenter randomized study of IVCFs, PREPIC 2, has been published in abstract form . The purpose of this study was to assess efficacy and safety of a retrievable IVCF implanted for 3 months in patients with symptomatic unprovoked PE associated with a lower-extremity DVT. Similar to PREPIC, all patients received therapeutic anticoagulation. A total of 399 patients were recruited in 6 years. At 3-month follow-up, recurrent PE occurred in 3.0 % of the filter group compared to 1.5 % in the non-filter group (RR 2.0; 95 % CI 0.51–0.79). No difference was found in rates of DVT (0.5 % in both groups). Overall, fifteen deaths (7.5 %) occurred in the filter group and 12 (6.0 %) in the non-filter group. At 3 months, 79.2 % of patients underwent IVCF retrieval. The results of PREPIC and PREPIC 2 do not provide justification for routine placement of IVCF in patients with PE. Unfortunately, because all patients received therapeutic anticoagulation, these data offer no insight into the outcome of the typical patient who has had an IVCF placed, namely, those who have contraindications to anticoagulant therapy. Nevertheless, in the absence of a randomized comparison between IVCFs and anticoagulation for VTE, these data are the principal available means by which filter efficacy and safety can be assessed.
Additional valuable information about the efficacy of IVCFs can be found in the large population-based observational study using California patient discharge data . During a 5-year period, 3632 patients received an IVCF for VTE and 64,333 patients did not receive an IVCF. After adjusting for risk factors for recurrent VTE, recipients of IVCFs were more likely to be re-hospitalized for VTE among those that initially manifested as PE (HR 2.62; 95 % CI 2.09–3.2). In addition, filters did not seem to have a short-term protective benefit against PE as the time to recurrent PE was similar between filter recipients and non-recipients. Filter recipients were also more likely to die during follow-up than were control patients. Conversely, in an analysis of the National Inpatient Sample cohort, Stein and colleagues showed that IVCFs marginally improved in-hospital case fatality rates compared to patients without IVCFs (7.2 vs. 7.9 %, P < 0.0001) . This benefit was confined mostly in unstable rather than stable patients with VTE. The limitations of these retrospective administrative database analyses and the lack of post-discharge or long-term follow-up in the Stein study should be taken into account when interpreting these results.
What are complications of IVCFs?
IVCF placement is usually performed through common femoral or jugular vein access. Most of the data on IVCF complications comes from case reports and case series. The Manufacturer and User Facility Device Experience (MAUDE) database is a web-accessible database established by the United States Food and Drug Administration (FDA) in 1991 to compile adverse events associated with medical devices . Since reporting is voluntary, the database has obvious limitations but it does provide a “real world” source of adverse event rates associated with IVCF. From 2000 to 2010 over 842 IVCF-related complications have been reported to the MAUDE database . Complications related to IVCF placement may occur acutely or years after deployment. Immediate procedure-related complications are uncommon (<1.0 % of insertions) and include IVCF misplacement, pneumothorax, local hematoma, air embolism, carotid artery puncture, and arteriovenous (AV) fistula formation. The reported frequency of long-term complications vary widely in the literature. These include inferior vena cava (IVC) thrombosis (2–30 %), filter fracture (2–10 %), filter migration (0–18 %), IVC perforation or penetration (0–86 %) [26, 27, 28]. The most reliable data for the frequency of DVT following filter placement come from the PREPIC study which noted an 8.5 % cumulative incidence of DVT at 1 year, 20.8 % at 2 years and 35.7 % at 8 years [19, 20]. Of note, 35 % of patients in the PREPIC study were on anticoagulation for the duration of follow up. Since post-thrombotic syndrome (PTS) is a known and dreaded consequence of DVTs, it would follow that IVCFs which increase the risk of DVT would also contribute to a higher rate of PTS. A systematic review evaluated the literature for the potential association between IVCFs and development of PTS . In a pooled analysis of 1,552 patients with IVCFs and mean follow-up of 4.5 years, the weighted pooled incidence of edema was 42.9 % and chronic skin changes (including venous ulcers) was 12 %. Patients who received an IVCF for secondary VTE prevention (i.e. those with known VTE) had a higher risk of PTS than those who received an IVCF for primary prevention. One study within this analysis reported no difference in PTS according to whether anticoagulant therapy was prescribed at the time of IVCF deployment, however data on long-term anticoagulation was not reported . There were no studies that evaluated incidence of PTS in relation to use of graduated compression stockings. A recently recognized long-term complication of IVCFs is the entrapment of guidewires used to place vascular access catheters . In several instances, forceful attempts to remove guidewires have led to filter displacement [32, 33]. The best approach to avoid this complication is to prominently identify patients with IVCFs in the medical record, use straight-tipped rather than J-tipped guidewires when placing central venous catheters in IVCF patients and limit the insertion of guidewires to 18 cm (the mean distance to the SVC: right atrial junction) .
What is the appropriate practice for IVCF retrieval?
The rationale for using rIVCFs is to offer mechanical protection against PE during the limited high-risk period when anticoagulation may be contraindicated. More than half of IVCFs are placed for temporary prophylaxis and therefore are candidates for retrieval [35, 36, 37, 38]. However, a systematic review showed an average retrieval rate of only 34 % . Therefore, attention should be directed to improving retrieval rates. The overall retrieval rate is a product of the clinical and procedural rate . If attempted, ~85–90 % of IVCF retrievals are successful [41, 42]. Procedural factors associated with retrieval failure include prolonged dwelling time, advanced age, filter head position, and filter design [43, 44]. The true time frame for successful retrieval of IVCFs remains undefined, although retrieval is typically most successful within 9–12 weeks of insertion before filter adherence to the caval wall [41, 42]. On the other hand, manufacturer literature reports safe retrieval of some filters up to 300 days after deployment . Clinical factors that influence whether or not IVCF retrieval is attempted include comorbidities, concurrent anticoagulation, primary indication for placement, and documented plans for removal at the time of insertion or when the patient is discharged [15, 46, 47]. On August 8, 2010 the FDA issued a Safety Alert encouraging timely removal of IVCFs in order to avoid long-term complications . Therefore improved IVCF retrieval rates are mandatory and methods to increase retrieval rates are needed. Experts have suggested that an effective system geared towards improving retrieval rates must consist of a multi-faceted approach including education of the patient and family, a tracking system to minimize patient loss to follow-up, and dedicated personnel to oversee the process [40, 49]. In 2007, a multidisciplinary consensus panel deemed studies focusing on IVCF retrieval as a top clinical research priority .
Is the presence of an IVCF an indication for anticoagulation?
While IVCFs may protect against PE in the short-term, prolonged indwelling IVCFs are associated with thrombotic complications, most commonly DVT and IVC thrombosis. Therefore, it has been postulated that patients with long-term indwelling IVCFs should remain on prophylactic anticoagulation. This is an especially important question given the growing number of young “high-risk” trauma patients that are receiving IVCFs for primary VTE prophylaxis and subsequently not having the filter removed. Although anticoagulant therapy has been reported to reduce prothrombin activation in filter recipients , its impact on long-term clinically relevant outcomes in patients with filters remains to be seen. In the PREPIC study, 35 % of patients in both the IVCF group and non-filter group remained on anticoagulation during the entire 8-year follow-up. While 15 % of patients in the IVCF group experienced major bleeding events, there was no difference in mortality between the IVCF and non-filter group . Unfortunately, the authors did not report whether or not anticoagulant therapy affected clinical outcomes such as recurrent PE, DVT, IVC thrombosis or PTS. Therefore, it remains unclear whether the benefits of indefinite anticoagulant therapy would have exceeded the associated risks.
Billett and colleagues evaluated outcomes in patients diagnosed with DVT between 1997 and 2004 who received an IVCF with or without concurrent anticoagulation . Five-year mortality significantly favored the filter group that received anticoagulation, however, when adjusted for age this difference was no longer noted. This retrospective study suggests that filters with anticoagulation do not add benefit over IVCF alone in patients with DVT, however other confounders such as severity of illness, indication for IVCF placement, comorbidities, and time in therapeutic range or duration of anticoagulation were not considered.
Another study compared anticoagulation versus no anticoagulation after IVCF placement. In this large retrospective single center cohort study between 2002 and 2006, a total of 706 patients had an IVCF placed. Forty percent of these patients were treated with concurrent anticoagulation . All-cause mortality favored anticoagulation (37.7 vs. 56.0 %, P < 0.001). However, post-filter placement, VTE occurred in more patients on anticoagulation (26.8 vs. 14.1 %, P < 0.001). Specifically, DVT (20.3 vs. 11.1 %, P = 0.001), PE (7.3 vs. 3.5 %, P = 0.0270) and IVC thrombosis (6.9 vs. 2.1 %, P = 0.002) were more common in patients who were treated with anticoagulation. The authors concluded that anticoagulation may not protect patients from thrombotic complications associated with IVCF placement.
A 2008 meta-analysis addressed the need for anticoagulation following IVCF placement . Data from nine studies indicated that anticoagulation resulted in a trend towards decreased VTE following filter placement, however this did not achieve statistical significance (OR 0.639; 95 % CI 0.351–1.159, P = 0.14). Limitations of this meta-analysis include the significant heterogeneity of studies included and the lack of individual studies reporting important confounders such as duration of anticoagulation, time in therapeutic range for anticoagulation, stratification based on filter indication, and clear indication of clinical and imaging follow-up. The authors concluded that routine long-term anticoagulation following IVCF placement can be neither supported nor refuted.
In considering long-term anticoagulation in patients with an IVCF, clinicians must weigh the risks (bleeding) and benefits (prevention of thrombotic events) of anticoagulation. The annual incidence of major bleeding on vitamin K antagonists is 2 % per year, with an absolute fatal bleeding risk of 0.2–0.9 % per year . In contrast, according to the PREPIC study, the maximal excess risk of thrombotic mortality is 0.2 % per patient-year . Therefore, even with optimal anticoagulant therapy control, it is doubtful that indefinite anticoagulation for filter recipients would result in net benefit. Furthermore, because the risk of recurrent VTE declines over time while the risk of bleeding with anticoagulant therapy remains constant, the risk:benefit ratio of long-term anticoagulation likely worsens as time passes [55, 56]. This conclusion may change with the emergence of newer oral anticoagulants that appear at least as effective as warfarin in the treatment of VTE with potentially less bleeding [57, 58, 59, 60, 61]. Until additional data are available on the long-term efficacy of these agents, the duration of anticoagulation for patients with IVCFs should be determined on an individualized assessment of the risk of bleeding and thrombosis and not solely by the presence of a filter . Further prospective studies are needed to determine the role of anticoagulation in the setting of IVCF placement in the prevention of VTE. The impact of anticoagulation post-filter placement on clinical outcomes is uncertain.
What are the potential indications for IVCF placement?
Professional Medical Society Practice-Based Guidelines for IVCF indications
Acute VTE and contraindication to AC
Yes (grade B, level III)
Yes (class 1, level B)
Yes (grade 1B)
Yes (class IIa, level C)
Failure of AC
Consider (grade C, level IV)
Yes (class IIa, level C)
Yes (class IIa, level C)
Preoperatively if recent acute VTE (<30 days) and must have AC interrupted for surgery
Yes (grade C, level IV)
As an adjunct to therapeutic AC in acute VTE
No (grade A, level 1b)
No (class III, level C)
No (grade 1B)
Free-floating proximal DVT
No (grade B, level III)
Massive PE or proximal DVT undergoing thrombolysis
No (grade C, level IV)
Consider (class IIb, level C)
Primary prophylaxis in high-risk surgical patients (e.g. trauma, orthopedic, or spinal)
Yes, if PP CI (level III)
Consider if PP CI
No (grade 2C)
Contraindication to or complication from anticoagulation
Although the vast majority of patients with an acute VTE can be managed with anticoagulation, in a small subset of patients, anticoagulant therapy is contraindicated, usually due to active or high-risk of bleeding or the need for urgent major surgery. Because patients with an acute episode of VTE are at high risk for recurrence in the absence of anticoagulation (40 % in the first month) [68, 69] and contraindications to anticoagulation are usually temporary, IVCFs should be considered for this population. In patients with a recent acute DVT, e.g. within 30 days, who require major surgery where therapeutic anticoagulation is contraindicated, a perioperative retrievable IVCF is reasonable. If more than 30 days have passed since the thrombotic event, patients can usually be managed with bridging anticoagulation postoperatively, initially at prophylactic then therapeutic doses, instead of an IVCF. The 2012 ACCP guidelines recommend placement of an IVCF in patients with acute proximal lower extremity DVT in whom anticoagulation is contraindicated (Grade 1B) . Once bleeding risk resolves, a conventional course of anticoagulation therapy should be administered (Grade 2B) . Similarly, other professional society guidelines concur with these recommendations [62, 63, 64, 65, 66, 67].
Failure of anticoagulation
Failure of anticoagulation is often cited as a justification for IVCF placement. However, true anticoagulant failure is an uncommon cause of recurrent VTE. RCTs of anticoagulation for VTE demonstrate 95 % reduction in recurrent VTE. Further, the PREPIC study provided evidence that IVCFs do not add incremental benefit to standard anticoagulation. Therefore, any patient with recurrent VTE despite anticoagulation should be carefully evaluated for other causes of recurrent VTE prior to considering IVCF placement, e.g. nonadherence to anticoagulation therapy, inadequate dosing of anticoagulant therapy, or misdiagnosis as acute recurrent VTE when in fact it is chronic. Instead of labeling these as failures of anticoagulation, efforts should be redoubled to maintain the patient in the therapeutic range or use an alternative anticoagulant, rather than place an IVCF that may predispose to further thrombosis and promote morbidity if subtherapeutic anticoagulant therapy continues. If the patient has been clearly therapeutic, investigation for hypercoagulable syndromes such as antiphospholipid syndrome, Trousseau’s syndrome, or heparin-induced thrombocytopenia, should be undertaken [70, 71, 72, 73, 74]. Because these syndromes are due to a systemic hypercoagulable state, regional approaches to prevent thromboembolism such as IVCFs are never adequate and perhaps may exacerbate thrombotic morbidity and mortality. Instead these conditions may require more intensive or alternative forms of anticoagulation. Recurrent events in the same location despite therapeutic anticoagulation should prompt consideration of anatomic abnormalities such as iliac vein compression May-Thurner syndrome or thoracic outlet Paget-Schröetter syndrome.
Massive pulmonary embolism
Patients with massive or submassive PE with hemodynamic compromise are often considered for IVCF placement. The concern is that despite therapeutic anticoagulation, even a recurrent small PE could lead to fatal outcomes due to limited cardiopulmonary reserve. A retrospective analysis of the data collected in the International Cooperative PE Registry found that in patients with massive PE (defined as hemodynamic instability on presentation) treated with therapeutic anticoagulation, IVCF insertion reduced 90-day mortality compared to those that did not receive an IVCF . However, only 10 % of patients with massive PE received an IVCF and two-thirds did not receive thrombolysis. Therefore, conclusive evidence in support of IVCFs in this vulnerable population does not exist. The ACCP 2012 guidelines acknowledge that their firm recommendations against IVCFs in patients with acute VTE that can be treated with anticoagulation may not apply to this subpopulation with massive PE and hemodynamic compromise .
Thrombolysis of acute ilio-caval venous thrombosis
The principal complications of DVT are PE and PTS. While catheter-directed thrombolysis does not reduce the risk of PE, studies have shown that thrombolysis decreases the incidence of PTS [76, 77, 78, 79, 80]. Further, thrombolysis for limb-threatening proximal DVT may be associated with fatal and non-fatal PE [81, 82]. Therefore, implantation of an IVCF has been suggested in patients undergoing systemic or catheter-directed thrombolysis to prevent embolization of thrombotic fragments to the pulmonary circulation. The evidence to support this practice is conflicting and based on mostly retrospective data. Some studies have shown that despite IVCF placement prior to thrombolysis, risk of fatal and non-fatal PE still exists . Conversely, other studies noted no episodes of PE with routine use of an IVCF prior to thrombolysis [83, 84, 85]. In a multicenter registry of catheter directed thrombolysis, only one fatal PE (0.3 %) occurred without routine filter use . These conflicting data indicate that the value of preemptive deployment an IVCF during thrombolysis of DVT remains unproven. Therefore, clinicians should consider filters on a case-by-case basis after reviewing the patient’s risk for embolization (e.g., poorly adherent IVC, iliac thrombi) and mortality from PE (e.g., patients with concomitant PE, those with limited cardiopulmonary reserve).
Major trauma patients
Without pharmacologic prophylaxis, VTE is a common complication of major trauma and can occur in up to 58 % with 18 % being proximal vein DVTs . Guidelines recommend routine VTE thromboprophylaxis with either low-dose unfractionated heparin or low-molecular weight heparin over no prophylaxis (Grade 2C) [63, 87]. In some instances, use of pharmacologic thromboprophylaxis is contraindicated due to active or high risk of bleeding from traumatic injury [87, 88, 89]. Mechanical prophylaxis with graduated compressions stockings or intermittent pneumatic compression devices can be used in trauma patients at risk for excessive bleeding complications. However, many trauma patients have injuries that preclude the use of lower extremity mechanical devices. With the introduction of retrievable IVCFs, their use, particularly for primary thromboprophylaxis in a major trauma patient at high-risk for VTE and bleeding, has gained significant popularity . The rationale for using retrievable IVCFs is to offer mechanical protection against PE during the limited high-risk period when anticoagulation may be contraindicated. However, in a large retrospective review of data from the largest trauma center in New England, Sarosiek et al. found that many IVCFs placed after trauma were inserted when the highest risk of bleeding had resolved and thromboprophylaxis could have been resumed .
More than two-dozen nonrandomized cohort studies have assessed the usefulness of IVCFs in the prevention of PE in high-risk trauma patients, many with conflicting results. A recent meta-analysis evaluated the evidence for prophylactic IVCFs in trauma patients without known VTE. The authors found no RCTs addressing the role of prophylactic IVCFs in any patient population. Among observational studies in trauma patients, there was a statistically significant decrease in PE with prophylactic IVCF placement compared to matched controls (OR 0.21, 95 % CI 0.09–0.49). However, these studies had significant limitations, most notably the lack of contemporary pharmacologic prophylaxis. Thus, no firm recommendations either for or against the routine use of prophylactic IVCFs in this population could be made . Another meta-analysis confirmed these findings; the authors rated the strength of evidence as low to support the reduction of non-fatal and fatal-PE, insufficient to support a reduction in mortality, and insufficient to determine an increase or decrease in DVT in trauma patients receiving IVCFs versus no IVCFs .
In 2002, the Eastern Association for the Surgery of Trauma (EAST) cited class III evidence (retrospective data, expert opinion, or case reports) in support of the use of prophylactic IVCFs in very high-risk trauma patients who have an injury rendering them immobilized for a prolonged period of time and have a contraindication to thromboprophylaxis . However, these guidelines have not been updated in over a decade and were primarily based on data using permanent IVCFs. On the other hand, the 2012 ACCP guidelines recommend against the use of prophylactic IVCFs as primary prophylaxis in trauma patients (Grade 2C) based on nearly the same body of evidence . This dichotomy in clinical recommendations may explain differences in practice patterns across trauma centers in North America . A recent cost analysis found that adherence to the EAST guidelines for placement of prophylactic IVCFs in high-risk trauma patients was not cost-effective when considering outcomes during the inpatient hospitalization or long-term after hospital discharge. Instead, the more restrictive ACCP 2012 approach to using therapeutic IVCFs for known VTE and contraindication to anticoagulation is cost-effective .
The majority of filters in the trauma population is for primary prophylaxis and therefore are candidates for retrieval [35, 36, 37, 38]. The presumed advantage of retrievable IVCFs avoiding long-term complications is predicated on their removal in a timely manner when bleeding or thrombotic risk subsides. This is particularly true in the young trauma patient who has no other risk factors for thrombosis outside of their injury. Unfortunately, retrieval rates in trauma patients are dismal; studies have shown average retrieval rates are only ~20 % [6, 93, 94, 95, 96, 97] with the primary reason for non-retrieval being patient lost to follow-up, an expected finding in the transient trauma population. Further sustainable strategies to improve retrieval rates are needed to prevent long-term complications associated with indwelling filters.
Cancer is a well-documented independent risk factor for the development of VTE. Thrombosis is the second most common cause of death in cancer patients and is predictive of worse short-term and long-term survival [98, 99, 100]. Pharmacologic prophylaxis and treatment of VTE, specifically with low-molecular weight heparin, is the preferred method based on Level 1 evidence [101, 102, 103]. However, cancer patients with VTE being treated with anticoagulant therapy are at particular high risk for recurrent VTE (hazard ratio of 3.2; 95 % CI 1.9–5.4) and major bleeding compared to patients without cancer (hazard ratio of 2.2; 95 % CI 1.2–4.1) . Interestingly, this increased risk of recurrence thrombosis and major bleeding does not seem to be related to over- or under-anticoagulation, but rather to stage of cancer. Further, certain cancers such as gliomas or secondary metastatic brain tumors may confer inherently higher spontaneous bleeding risk with or without systemic anticoagulation [105, 106, 107]. Therefore, IVCFs have been suggested as an alternative or complimentary strategy to anticoagulation.
While case series have reported efficacy and safety of IVCFs in cancer patients with VTE, pooled analysis of comparative studies of filters versus anticoagulation found an increased risk of symptomatic PE, DVT, and IVC thrombosis in filter recipients and no difference in overall survival between groups . However, the pooled rate of major bleeding was increased in the anticoagulant treated patients. A large retrospective study designed to report practice patterns of IVCF utilization and IVCF-related complications in cancer patients showed that significant differences in the indication for IVCF placement exist between cancer and non-cancer patients . In patients with cancer, IVCFs were more likely to be placed outside of guideline supported indications, e.g. for primary VTE prophylaxis. In addition, cancer patients were more likely to experience negative clinical outcomes such as shorter time to DVT, PE, IVC thrombosis, and death. Further, retrieval of IVCFs was less likely in cancer patients compared to non-cancer patients (28 vs. 42 %, P < 0.001), presumably due to shorter life expectancy. The effectiveness of IVCFs as an adjunct to therapeutic anticoagulation was studied in a single-center randomized controlled trial . Sixty-four cancer patients with acute DVT were randomized to therapeutic doses of fondaparinux with or without an IVCF. At 3-year follow-up there was no difference between the two groups in terms of recurrent DVT, PE, or IVCF thrombosis. Although a trend towards decreased survival was seen in the IVCF group, this may have been due to differences in patient characteristics at baseline, e.g. more patients with poor prognosis cancers in the IVCF group.
The hypercoagulable state associated with cancer affects all vascular beds. It would therefore ensue that regional approaches to preventing recurrent VTE, such as IVCFs, would not provide adequate protection in these patients and may actually be harmful . Therefore, despite the greater hemorrhagic morbidity of long-term anticoagulant therapy in patients with cancer [104, 112], the risk:benefit profile of anticoagulation appears to be more favorable than that associated with IVC filter use. Taken together, these data indicate that pharmacologic approaches, not mechanical approaches, will continue to be the preferred therapy for most patients with cancer who have VTE. To reflect this approach, recent evidence-based guidelines on the prevention and management of VTE in cancer patients have recommended against the routine use of IVCFs for prevention and treatment of VTE [113, 114, 115, 116].
High-risk orthopedic patients
Patients undergoing orthopedic surgery such as total knee arthroplasty (TKA) or total hip arthroplasty (THA) are considered to be at very high risk for the development of VTE because of a number of factors that contribute to venous stasis, such as older age, position on the operating table, use of thigh tourniquets during knee arthroplasty, long periods of preoperative and postoperative recumbency, as well as vascular injury. The prevalence of DVT without pharmacologic prophylaxis in this group of patients is as high as 40–60 % . With contemporary pharmacologic VTE prophylaxis, the rate of symptomatic DVT is estimated at 0.8 % and PE at 0.35 % . However, in certain orthopedic populations such as spinal surgery cases, the incidence of PE has remained as high as 4 % with appropriate pharmacologic prophylaxis . Some have suggested that bleeding (e.g. epidural hematoma) in the operative field due to anticoagulation has detrimental effects for post-operative recovery. Therefore, IVCFs are an attractive option for adjunctive VTE prophylaxis in these high-risk patients [119, 120, 121, 122]. Unfortunately, RCTs in this patient population are likely not feasible and therefore clinical practice is based on observational studies alone. However, limitations of this low quality evidence include small sample size, use of historical controls, variable definitions for high-risk patients, different follow-up periods, and inconsistent pharmacologic prophylaxis. In a recent observational study involving more than 9000 orthopedic patients, 90 patients (0.96 %) received an IVCF, 55 (0.6 %) for primary prophylaxis of PE. Interestingly, only 42 % of those that received an IVCF had a contraindication to anticoagulation. Only 51 % of retrievable filters were removed at 6 months and 2 patients (2.2 %) encountered complications with the filter . Limited data suggest that DVT may be more frequent in spinal cord injury patients receiving a prophylactic IVCF versus those that do not receive an IVCF (21 vs. 5.2 %) with similar incidence of PE (2 %) in both groups . Glotzbecker and colleagues performed a systematic review of observational studies evaluating the rates of postoperative DVT and PE with different forms of pharmacologic and mechanical prophylaxis in spinal surgery patients. They reported the highest rate of DVT (22 %) with use of prophylactic IVCFs compared with other forms of VTE prophylaxis or no prophylaxis. The authors concluded that there is insufficient evidence to support the routine use of IVCFs in this population outside of consensus indications, i.e. acute VTE and contraindication to anticoagulation .
In the era of new targeted oral anticoagulants recently shown to be at least as effective as traditional pharmacologic prophylaxis in prevention of VTE after high-risk orthopedic surgery, the minimal benefit of prophylactic IVCFs is diminishing. Given the low quality evidence available, potential for adverse events with IVCF placement and non-retrieval, and the efficacy and safety of contemporary pharmacologic prophylaxis, the routine use of IVCFs for primary prevention of PE in major orthopedic surgery is not recommended (GRADE 2C) .
Patients undergoing bariatric surgery are at increased risk for VTE due to underlying obesity, added risk of surgery, comorbidities such as obesity hypoventilation syndrome and obstructive sleep apnea, and a possible under-dosing with standard fixed-doses of thromboprophylaxis. PE is considered the leading cause of perioperative death in bariatric surgical patients [125, 126, 127, 128]. The reported incidence of DVT is 1–3 % and PE is 0.3–2 %; however, mortality in patients with PE may be as high as 30 % . Although the risk of VTE is highest during the first 2 months after surgery , the optimal drug, timing, dose, frequency, and duration of thromboprophylaxis in bariatric surgery are not established.
A recent survey of the members of the American Society of Metabolic and Bariatric Surgery found that 94 % of respondents routinely use chemical thromboprophylaxis and 98 % use mechanical prophylaxis . Interestingly, 55 % of respondents considered placement of an IVCF for thromboprophylaxis compared with only 7 % in a survey published by the same authors in 2000 [131, 132]. Few data exist to support the widespread use of IVCFs for primary PE prevention. An analysis from a prospective national registry including data of 6376 patients undergoing bariatric surgery reported that approximately 8.5 % of patients have an IVCF placed pre-operatively, 65 % of which are for primary VTE prophylaxis . Despite this aggressive strategy for VTE prevention, filter patients did not have reduced rates of postoperative VTE [odds ratio (OR) 1.28; 95 % CI 0.51–3.21], serious complications (adjusted OR 1.40; 95 % CI 0.91–2.16), or death/permanent disability (adjusted OR 2.49; 95 % CI 0.99–6.26). In fact, 57 % of patients with major perioperative complications could be attributed directly to the IVCF. Subgroup analyses did not identify any patient group for whom IVCF were beneficial. A recent systematic review of IVCFs in bariatric surgery identified no randomized controlled trials and seven observational studies . Use of IVCFs was associated with a threefold increased risk of postoperative DVT (RR 2.81; 95 % CI 1.33–5.97, P = 0.007) and death (RR 3.27; 95 % CI 0.78–13.64, P = 0.1). No benefit in terms of risk of PE was found in recipients of IVCFs (RR 1.02; 95 % CI 0.31–3.77, P = 0.09). The authors noted significant heterogeneity across studies and concluded that the routine placement of IVCFs in high-risk bariatric surgery patients should be discontinued due lack of evidence for efficacy and potential for harm. However, others suggest that prophylactic IVCFs should be considered in very high-risk bariatric surgery patients, defined as a BMI >50 kg/m2, history of VTE, immobility, hypercoagulable state, pulmonary insufficiency and hypertension, and chronic venous stasis. Given the poor retrieval rates of IVCFs and lack of strong evidence for efficacy, IVCFs in this population should be restricted to the universally agreed upon indications for IVCFs, namely patients with acute VTE and contraindication to anticoagulation. Efforts focused on identifying the optimal regimen for pharmacologic prophylaxis may invalidate any indication for IVCFs in this population.
Knowledge gaps in IVCF utilization
IVCF practices have become engrained into our culture based on legacy rather than empiric science over the last four decades since the introduction of the Greenfield filter. Despite close to 250,000 filters being placed in the United States alone in 2012, there is a paucity of convincing evidence that shows filters improve any clinically relevant outcomes. In fact, while more articles have been published from 2001 to 2012 compared to a preceding period 1975 to 2000, not a single new prospective randomized trial designed to test efficacy of IVCFs has emerged . Instead, current standard practice surrounding IVCFs is based on conclusions drawn from heterogeneous non-randomized studies with significant limitations in methodology or on expert opinion. Research in the field of IVCFs is challenging given the lack of standardized indications for placement and retrieval of these devices, geographical practice pattern variations, differences in the devices themselves, and the heterogeneous population in which these devices are deployed. In 2007, a multidisciplinary panel of IVCF experts convened to establish and prioritize a research agenda for IVCFs. To address the fundamental gaps in knowledge of IVCF efficacy, safety, and cost-effectiveness, the panel determined the clinical priority topics in which well-designed clinical trials are urgently needed . These areas of research and clinical interest include questions such as: Do IVCFs improve mortality in patients with acute VTE and contraindication to anticoagulation? Do IVCFs add any benefit in addition to contemporary pharmacologic prophylaxis? Are there any populations that would benefit from primary VTE prophylaxis with IVCFs? What strategies can maximize removal of retrievable IVCFs? What is the ideal removal window for retrievable IVCFs? Several studies have been published to address the clinical questions highlighted by this expert panel. For example, a recent randomized pilot study evaluated the feasibility of recruiting and randomizing trauma patients to prophylactic IVCF versus no filter. At 1-year interim analysis feasibility objectives were met and there was no significant difference in PE or DVT between both groups, although the study was not powered to detect a difference in efficacy . Other studies have successfully increased retrieval rates by focusing on increased clinician oversight, novel technical aspects of retrieval, and streamlining systems-based approaches [40, 47, 95, 137, 138, 139, 140, 141, 142].
While a well-designed RCT testing the efficacy and safety in each subpopulation that IVCFs are currently employed (e.g. patients with acute VTE and contraindication to anticoagulation or for primary prophylaxis in patients that are high-risk for VTE) would be ideal, this would require large sample sizes, significant funding, and formal collaboration with multiple centers and disciplines to engage all stakeholders in the area. Instead, a more practical approach to answering topics of clinical equipoise would include large prospective cohort studies designed to evaluate multiple clinically relevant outcomes of efficacy, safety, and cost-effectiveness to determine which subgroup of patients, if any, may benefit from IVCFs without excessive complications. Further, quality improvement studies that evaluate the effectiveness of institutional protocols aimed at decreasing inappropriate use of IVC filters would provide useful information on how to implement evidence-based guidelines on a local level. Until such studies are carried out a more restrictive approach to IVCF utilization should be adopted to prevent undue harm from these devices.
VTE is a significant health care concern for both inpatients and outpatients. Standard treatment is anticoagulant therapy for both the prevention and treatment of VTE. With the introduction of new anticoagulants over the last few decades, prevention and treatment of VTE has been revolutionized. However, despite these innovations there will always remain a small group of medical and surgical patients that cannot receive anticoagulation. For these patients, it remains to be seen whether IVCFs truly improve clinically relevant outcomes. Until then, clinicians are forced to assimilate the vast low-quality evidence on IVCFs to make case-by-case decisions on IVCF utilization in their patients.
Conflict of interest
Anita Rajasekhar has funding from the American Society of Hematology. She has past involvement on physician advisory boards for Octapharma, Baxter, Bayer and Alexion. She has no other financial interests to disclose.
- 9.Smouse B, Johar A (2010) Is market growth of vena cava filters justified? Endovasc Today 74–77Google Scholar
- 10.Dossett LA, Adams RC, Cotton BA (2011) Unwarranted national variation in the use of prophylactic inferior vena cava filters after trauma: an analysis of the National Trauma. Databank J Trauma 70(5):1066–1070Google Scholar
- 16.European Union: European Commission; 2010 [updated 27 July 2010]. Available from: http://ec.europa.eu/eurostat. Accessed 17 December 2014
- 17.Heit JA, Cohen AT, Anderson FA Jr; on Behalf of the VTE Impact Assessment Group (2005) Estimated annual number of incident and recurrent, non-fatal and fatal venous thromboembolism (VTE) events in the US. ASH Annual Meeting Abstracts 106(11): Abstract 910 Google Scholar
- 20.PREPIC Study Group (2005) Eight-year follow-up of patients with permanent vena cava filters in the prevention of pulmonary embolism: the PREPIC (Prevention du Risque d’Embolie Pulmonaire par Interruption Cave) randomized study. Circulation 112(3):416–422Google Scholar
- 21.Young T, Tang H, Hughes R (2010) Vena cava filters for the prevention of pulmonary embolism. Cochrane Database Syst Rev, Issue 2Google Scholar
- 22.Mismetti P, Ennezat PV, Quere I, Jeannot S, Barral FG, Couturaud F, Elias A, Teinturier C, Laporte S, Meyer G. Prevention of pulmonary embolism recurrences by retrievable vena cava filter: results of the randomized multicenter trial PREPIC 2 (2013) International Society of Thrombosis and Hemostasis Annual Meeting, Volume 11. Supplement 2 Abstract 18.2Google Scholar
- 25.United States Food and Drug Administration. Manufacturer and user facility device experience database. http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfMAUDE/search.cfm. Accessed 1 December 2014
- 41.Uberoi R, Chalmers N, Kinsman R, Walton P (2011) The first BSIR inferior vena cava filter registry report. Dendrite Clinical Systems Ltd, OxfordshireGoogle Scholar
- 45.G2 Filter System IFUs [package insert]. Tempe AZ: Bard Peripheral Vascular, Inc; October 2009. http://www.bardpv.com/_international/in-prod-g2filter-ifu.php. Accessed 17 December 2014
- 48.http://www.fda.gov. Accessed 17 December 2014
- 49.Goei AD, Josephs SC, Kinney TB et al (2011) Improving the tracking and retrieval of inferior vena cava filters. Semin Interven Radiol 28(1):118–127Google Scholar
- 51.Billett HH, Kornblum N, Jacobs L, Gargiulo N, III (2004) A preliminary analysis of inferior vena cava filters with and without anticoagulation vs anticoagulation only: readmission and mortality rates. ASH Annual Meeting Abstracts 104(11):1776Google Scholar
- 52.Narayan A, Streiff MB, Hong K, Lee A, Kim H (2009) Anticoagulation does not prevent venous thromboembolism following IVC filter placement: a retrospective study of 702 patients. ASH Annual Meeting Abstracts 114(22):2104Google Scholar
- 53.Ray CE, Prochazka A (2008) The need for anticoagulation following inferior vena cava filter placement: a systematic review. Cardiovasc Interven Radiol 31(2):316–324Google Scholar
- 58.Investigators EINSTEIN, Bauersachs R, Berkowitz SD et al (2010) Oral rivaroxaban for symptomatic venous thromboembolism. N Engl J Med 363(26):2499–2510Google Scholar
- 59.Investigators EINSTEIN-PE, Buller HR, Prins MH et al (2012) Oral rivaroxaban for the treatment of symptomatic pulmonary embolism. N Engl J Med 366(14):1287–1297Google Scholar
- 61.Investigators Hokusia-VTE (2013) Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism. N Engl J Med 369:1406–1415Google Scholar
- 64.British Committee for Standards in Haematology Writing Group, Baglin TP, Brush J, Streiff M (2006) Guidelines on use of vena cava filters. Br J Haematol 134(6):590–595Google Scholar
- 65.Kaufman JA, Kinney TB, Streiff MB, Sing RF, Proctor MC, Becker D, Cipolle M, Comerota AJ, Millward SF, Rogers FB, Sacks D, Venbrux AC (2006) Guidelines for the use of retrievable and convertible vena cava filters: report from the Society of Interventional Radiology Multidisciplinary Consensus Conference. J Vasc Interv Radiol 17:449–459PubMedGoogle Scholar
- 66.Jaff MR, McMurty S, Archer SL, Cushman M, Goldenberg N, Goldhaber SZ, Jenkins S, Kline JA, Michaels AD, Thistlethwaite P, Vedantham S, Whilte RJ, Zierler BK (2011) Management of massive and submassive pulmonary embolism, iliofemoral deep vein thrombosis, and chronic thromboembolic pulmonary hypertension. Circulation 123:1788–1830PubMedGoogle Scholar
- 67.Konstantinides S, Torbicki A, Agnelli G, Danchin N, Fitzmaurice D, Galiè N, Gibbs JSR, Huisman M, Humbert M, Kucher N, Lang I, Lankeit M, Lekakis J, Maack C, Mayer E, Meneveau N, Perrier A, Pruszczyk P, Rasmussen LH, Schindler TH, Svitil P, Vonk Noordergraaf A, Zamorano J, Zompatori M (2014) 2014 ESC Guidelines on the diagnosis and management of acute pulmonary embolism. Eur Heart J 35(43):3033–3073PubMedGoogle Scholar
- 71.Ruiz-Irastorza G, Cuadrado MJ, Ruiz-Arruza I et al (2011) Evidence-based recommendations for the prevention and long-term management of thrombosis in antiphospholipid antibody-positive patients: report of a task force at the 13th International Congress on antiphospholipid antibodies. Lupus 20(2):206–218PubMedGoogle Scholar
- 72.Sack GH Jr, Levin J, Bell WR (1977) Trousseau’s syndrome and other manifestations of chronic disseminated coagulopathy in patients with neoplasms: clinical, pathophysiologic, and therapeutic features. Medicine (Baltimore) 56(1):1–37Google Scholar
- 74.Warkentin TE, Greinacher A, Koster A, Lincoff AM, American College of Chest Physicians (2008) Treatment and prevention of heparin-induced thrombocytopenia: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest 133(6 Suppl):340S–380SPubMedGoogle Scholar
- 87.Gould MK, Garcia DA, Wren SM et al (2012) Prevention of VTE in nonorthopedic surgical patients: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 141(2 Suppl):e227S–e277SPubMedCentralPubMedGoogle Scholar
- 90.Rajasekhar A, Lottenberg R, Lottenberg L, Liu H, Ang D (2011) Pulmonary embolism prophylaxis with inferior vena cava filters in trauma patients: a systematic review using the meta-analysis of observational studies in epidemiology (MOOSE) guidelines. J Thromb Thrombolysis 32(1):40–46PubMedGoogle Scholar
- 97.Rosenthal D, Wellons ED, Lai KM, Bikk A (2005) Retrievable inferior vena cava filters: early clinical experience. J Cardiovasc Surg (Torino) 46(2):163–169Google Scholar
- 101.Lee AY, Levine MN, Baker RI et al (2003) Low-molecular-weight heparin versus a coumarin for the prevention of recurrent venous thromboembolism in patients with cancer. N Engl JMed 349:146–153Google Scholar
- 103.Lorio A, Guercini F, Pini M (2003) Low-molecular-weight heparin for the long-term treatment of symptomatic venous thromboembolism: meta-analysis of the randomized comparisons with oral anticoagulants. J Thromb Haemost 1:1906–1913Google Scholar
- 110.Barginear MF, Gralla RJ, Bradley TP, Ali SS, Shapira I, Greben C, Nier-Shoulson N, Akerman M, Lesser M, Budman DR (2012) Investigating the benefit if adding a vena cava filter to anticoagulation with fondaparinux sodium in patients with cancer and venous thromboembolism in a prospective randomized clinical trial. Support Care Cancer 20:2865–2872PubMedGoogle Scholar
- 112.Hutten BA, Prins MH, Gent M, Ginsberg J, Tijssen JG, Buller HR (2000) Incidence of recurrent thromboembolic and bleeding complications among patients with venous thromboembolism in relation to both malignancy and achieved international normalized ratio: a retrospective analysis. J Clin Oncol 18(17):3078–3083PubMedGoogle Scholar
- 117.Falck-Ytter Y, Francis CW, Johanson NA et al (2012) Prevention of VTE in orthopedic surgery patients: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 141(2 Suppl):e278S–e325SPubMedCentralPubMedGoogle Scholar
- 135.Girard P, Meyer G, Parent F, Mismetti P (2014) Medical literature, vena cava filters and evidence of efficacy. Thromb Haemost 111:1–9Google Scholar
- 137.Jeet Minocha, Idakoji I, Riaz A et al (2010) Improving inferior vena cava filter retrieval rates: impact of a dedicated inferior vena cava filter clinic. J Vasc Interv Radiol 21:1847–1851Google Scholar