Techniques in Orthopaedics® 19(4):327–336 © 2004 Lippincott Williams & Wilkins, Inc., Philadelphia Use of Vena Cava Filters Leoncio L. Kaw, Jr., M.D. and Ralph B. Dilley, M.D., F.A.C.S. Summary: Venous thromboembolism remains a significant cause of morbidity and mortality in the United States. Although anticoagulation is considered the mainstay of treatment, placement of vena cava filters is generally indicated in circumstances when there is a contraindication, complication, or documented failure of anticoagulation therapy. A number of devices are available and approved by the U.S. Food and Drug Administration for use. Each of these devices has distinct designs and deployment mechanisms and, more importantly, have variable data on safety and effectiveness. It is paramount for physicians to be familiar with the specific features of each device. This article addresses the indications, techniques of deployment, and complications associated with vena cava filters. Unique features and comparison data of each available device is presented. Key Words: Vena cava filters—Deep vein thrombosis— Pulmonary embolism. Venous thromboembolism (VTE) continues to be a significant cause of morbidity and mortality in the United States. It is projected that at least 201,000 new cases of VTE occur annually.63 Pulmonary embolism (PE) is the most severe complication of VTE. The estimated case fatality rate of PE is 7.7 deaths per 100 patients diagnosed with the condition.65 According to the National Center for Health Statistics, PE was the underlying or contributing cause of death in 572,773 decedents during a 20-year period (1979 –1998).33 Anticoagulation is currently the standard treatment for acute PE; however, approximately 3.9% to 7.9% of patients who receive therapeutic anticoagulation have been shown to experience a recurrence.4,16,21 Furthermore, bleeding complications occur in up to 10.5% of patients.4,21 In these instances, interruption of the inferior vena cava (IVC) has emerged as an effective alternative for the prevention of PE. This article provides an overview of currently available vena caval filters (VCF) approved by the U.S. Food and Drug Administration (FDA). Techniques of deployment, results of studies on VCF, and complications associated with insertion and use are addressed. In addition, established and controversial indications are discussed. HISTORICAL BACKGROUND Lower extremity deep venous thrombosis (DVT) as a source of pulmonary emboli was established by John Homans in the 1930s when he demonstrated during autopsy that the end of a thrombus taken from the pulmonary artery matched with the residual clot in the popliteal vein.32 Interruption of venous outflow by femoral, iliac, or IVC ligation was then performed in patients with lower extremity venous thrombi to prevent PE.31,51 Unfortunately, these methods were frequently associated with significant operative mortality, a high rate of recurrent PE, and the disabling sequelae of chronic venous stasis.3,28,49 A different approach to venous interruption was to allow some blood flow and still prevent passage of large emboli to the lungs by partial occlusion of the IVC. Various techniques such as caval plication with sutures or staples and use of ingenious extravascular occlusive devices were reported by numerous authors.1,44,48,64 Although the recurrent PE rate was considerably reduced with these approaches, operative mortality and long-term complications from venous stasis remained significant.48 Consequently, the need for a procedure or device that allowed effective filtration of emboli while maintaining IVC patency, and which could From the Division of Cardiothoracic and Vascular Surgery, Scripps Clinic and Scripps Green Hospital, La Jolla, California. Address correspondence and reprint requests to Ralph B. Dilley, MD, FACS, Division of Cardiothoracic and Vascular Surgery, Scripps Clinic/Scripps Green Hospital, 10666 N. Torrey Pines Road, La Jolla, CA 92037. E-mail: firstname.lastname@example.org 327 328 L. L. KAW AND R. B. DILLEY TABLE 1. Specifications of Current FDA-Approved Vena Cava Filters Device Simon Nitinol (Bard, Covington, GA) Vena Tech LP (B. Braun, Evanston, IL) Vena Tech LGM (B. Braun, Evanston, IL) Stainless steel Greenfield (Boston Scientific, Watertown, MA) Titanium Greenfield (Boston Scientific, Watertown, MA) Gianturco-Roehm Bird’s Nest (Cook, Bloomington, IN) TrapEase (Cordis Endovascular, Warren, NJ) Recovery (Bard, Covington, GA) Günther Tulip (Cook, Bloomington, IN) OptEase (Cordis Endovascular, Warren, NJ) Sheath Size (ID/OD) Length (mm) Diameter (mm) Jugular/Femoral/ antecubital Jugular/femoral 7/9 F 38 28 Permanent 7/9 F 43 40 Permanent Conical Jugular/femoral 12/14 F 38 30 Permanent 316 L stainless steel Conical Jugular/femoral 12/14 F 49 32 Permanent Titanium Conical Jugular/femoral 12/14 F 47 38 Permanent Stainless steel Multiplane Jugular/femoral 12/14 F variable 40 Permanent Nitinol Double basket Jugular/femoral/ antecubital 6/8 F 50 35 Permanent Nitinol Femoral 7/9 F 40 32 Permanent/retrievable Conichrome Bilevel, conical Conical Jugular/femoral 8.5/10 F 45 30 Permanent/retrievable Nitinol Double basket Jugular/femoral/ antecubital 6/8 F 58 30 Permanent/retrievable Material Design Access Phynox Bilevel, domeconical Conical Phynox Nitinol Intended Use ID/OD ⫽ inner/outer diameter. be performed under local anesthesia to minimize operative risk, became obvious. Initial efforts in the late 1960s included the Eichelter catheter,71 which was a removable device, and the Hunter balloon,34 designed for permanent occlusion of the vena cava. The most popular device was the Mobin-Uddin umbrella, which was introduced in 1967.47 This device was an umbrella design and consisted of 6 metallic spokes radiating from a central hub. It was covered on both sides by a circular sheet of heparin-impregnated silastic membrane with 1.5-mm perforations to allow continuous blood flow. The umbrella was inserted through a venotomy performed under local anesthesia and fluoroscopic guidance with the apex pointing inferiorly. In their first report on the clinical application of this device, no recurrence of PE was noted at 8 to 10 months follow up.46 However, thrombotic occlusion of the IVC was observed in all instances. The development of the Greenfield filter represented a major improvement over the Mobin-Uddin umbrella. The filter was made of 6 stainless steel strands fitted at an angle of 35° to an apical hub. Fixation was provided by fine recurved hooks, which grasp the vena caval wall without penetrating it. The unique conical-shaped design of the Greenfield filter had the advantage of allowing trapped emboli to fill and occlude 70% of the length of the cone without reducing the cross-sectional area more than 50%, thereby allowing continued blood flow.24 As a Techniques in Orthopaedics®, Vol. 19, No. 4, 2004 result, a much lower rate of caval occlusion was observed when compared with prior methods of clot filtration. Initial animal and clinical studies on the Greenfield filter showed maintenance of caval patency in all subjects during the period of observation.24,25 Like the Mobin-Uddin umbrella, the Greenfield filter was inserted through a venotomy using a 28 French introducer sheath. Over the next 30 years, several versions of IVC filters have been developed and marketed. IVC filter insertion has evolved from an open to a percutaneous procedure, with introducer sheaths as small as 8 French. In addition, retrievable filters are now available for patients requiring temporary protection from thromboembolic events. Although these filters share many similarities in design and deployment mechanisms, familiarity with their specific characteristics and features is mandatory for their safe and appropriate use. DESCRIPTION OF AVAILABLE DEVICES Currently, there are 7 permanent and 3 retrievable IVC filters approved by the FDA. Table 1 summarizes the physical characteristics and specifications of each. Simon Nitinol Filter The Simon Nitinol filter is constructed from a nickel– titanium alloy that has unique thermal–mechanical prop- USE OF VENA CAVA FILTERS 329 FIG. 1. Four of 6 FDA-approved permanent vena cava filters. From left, Bird’s Nest, stainless steel Greenfield, TrapEase, and Simon Nitinol filters. erties. The filter is pliable at room temperature and reforms into its predetermined shape when warmed to body temperature. This flexibility, along with its small delivery system, allows placement through alternative sites such as the antecubital or left jugular veins. The filter design consists of a dome of 7 overlapping wires below which 6 struts with hooks at the base diverge to form a cone (Fig. 1). This configuration, in effect, provides 2 levels of filtration. The insertion procedure requires iced normal saline to be continually infused through the delivery system. The filter is advanced rapidly by a feeder pump and subsequently discharged from the storage catheter. Once in the IVC, the filter then configures to its predetermined shape. The dome of the filter permits itself to align with the IVC on deployment. Vena Tech LGM and LP Filters The Vena Tech LGM filter is fabricated from Phynox, a nonferromagnetic alloy with excellent magnetic resonance imaging (MRI) compatibility. It has a conical configuration with 6 stabilizing sidebars on each limb designed to center and maintain orientation of the filter on deployment. These sidebars each contain anchoring barbs, some oriented superiorly and others inferiorly. The filter comes in a single kit that can be used for either the femoral or jugular approach. A guidewire is used to position the sheath into the IVC. The filter is then advanced through the entire length of the sheath and released by quick withdrawal of the sheath. The Vena Tech LP filter is a more recent, low-profile version that has 8 Phynox wires forming the conical configuration instead of 6. Greenfield Filter The original Greenfield filter is no longer commercially available and has since been replaced by 2 varieties: the over-the-wire stainless steel and modified-hook titanium filters (Fig. 1). Both share the similar conical design of the original filter but differ in that the stainless steel version can be placed over a guidewire. In addition, the stainless steel filter has a more flexible carrier system, which allows insertion through a left femoral approach. The initial design of the titanium filter was made with hooks similar to the original Greenfield filter. This, however, was noted to have a tendency to migrate, tilt, and perforate the IVC. Subsequent modifications resulted in a recurved hook design with an 80° angulation. The recurved hook serves as a barrier to penetration beyond the axis of the filter limb, and the 80° hook limits both upward and downward force vectors that may induce migration.23 The Greenfield filter has one of the simplest deployment mechanisms among the currently available filters. An intravascular sheath is inserted through which the Techniques in Orthopaedics®, Vol. 19, No. 4, 2004 330 L. L. KAW AND R. B. DILLEY FIG. 2. The OptEase (left) and Gunther Tulip (right) retrievable filters. carrier system is advanced to the desired level of placement. The locking mechanism on the control handle is then released and the carrier gradually retracted until the filter springs open in the vena cava. Separate color-coded kits are available for jugular and femoral approaches. Bird’s Nest Filter The Gianturco-Roehm Bird’s Nest filter uses a fine wire, multiplane filtering system consisting of 4 stainless steel wires, each 25 cm in length and 0.18 mm in diameter, attached to 2 pairs of V-shaped struts (Fig. 1). The tips of each strut have hooks designed to engage the caval wall for fixation. During insertion, a pusher is used to position the first set of hooks into the IVC. The 4 wires are then extruded from the filter catheter through a series of controlled steps, which allows the operator to alter placement and accommodate anatomic variations. The advantage of this is the avoidance of the need to center the filter in the caval lumen and the flexibility to be placed into vena cavae up to 40 mm in diameter (all other filters are limited to cavae of 30 mm or less). On the other hand, optimal deployment requires approximately 7 cm in caval length and therefore is not useful in patients with short vena cavae. The Bird’s Nest filter generates the largest MRI artifact because of its stainless steel construction. Techniques in Orthopaedics®, Vol. 19, No. 4, 2004 TrapEase and OptEase Filters The TrapEase filter has a symmetric, double-basket design that is laser cut from a single tube of nitinol (Fig. 1). The proximal and distal baskets each consist of 6 diamond-shaped struts oriented opposite to each other and connected by 6 straight side struts. The side struts contain proximal and distal hooks for fixation within the IVC. This unique configuration essentially provides 2 levels of filtering and allows a single filter to be placed from either direction. Moreover, the filter comes with the lowest profile delivery system available on the market and therefore can be inserted by bilateral femoral, jugular, or antecubital approaches. The filter is self-centering and is deployed by simply extruding it from the sheath. The OptEase filter is a retrievable version of the TrapEase and shares its basic design and deployment system. Unlike the TrapEase, the hooks on the side struts of the OptEase are located only on its proximal aspect and are longitudinally oriented (Fig. 2). The OptEase also has a hook on its inferior aspect for retrieval using a loop snare through a femoral approach. Proper orientation during insertion is therefore mandatory. Recovery Filter The Recovery filter has a unique conical design composed of 12 nitinol wires extending from a nitinol sleeve. The filter has 6 arms and 6 legs, which provide 2 levels USE OF VENA CAVA FILTERS of filtration. Insertion is through either femoral vein, and deployment is similar to the technique used for the Simon Nitinol filter. Filter removal is performed with a retrieval cone, constructed with 9 metal claws covered by a urethane cover, and inserted through the jugular vein. The retrieval cone is then advanced through a 10 French sheath and docked with the filter tip so that the filter may be retracted into the sheath and removed. Günther Tulip Filter The Günther Tulip filter is made of an alloy consisting of cobalt, nickel, chromium, and other trace metals (commonly called conichrome). It has a conical design with 4 main struts and 4 secondary struts (Fig. 2). The secondary struts loop around three fourths of the length of main struts forming a shape that resembles the petals of a tulip, hence its name. The primary struts have barbs on their distal end to provide secure attachment to the caval wall. The top of the filter has a small hook for insertion and retrieval through a jugular approach. The filter can be placed through femoral or jugular routes, with separate kits available for each. The femoral filter comes preloaded, whereas the jugular filter has to be loaded into the introducer before the procedure. Like with most filters, manual unsheathing is required for deployment. When inserted through the jugular approach, the filter can be repositioned even after deployment simply by pushing the sheath over the filter to collapse it. The whole system can then be moved to the desired position and redeployed by withdrawing the sheath again. Retrieval is through the jugular approach using an 80-cm loop snare made of braided platinum. TECHNIQUES IN FILTER PLACEMENT Access to the IVC for filter placement is generally obtained through the right femoral or jugular approach using the modified Seldinger technique. The choice of insertion site is arbitrary and normally depends on anatomy, location of the thrombus, type of filter available, and operator preference. Depending on the clinical condition of the patient, VCF insertion can be performed safely in the operating room, angiography suite, or at the bedside, and as an outpatient or inpatient procedure. Whatever the approach, evaluation of the IVC is mandatory before filter placement. The length and diameter of the infrarenal IVC should be measured, the location and number of renal veins determined, the presence and location of the thrombus ascertained, and the existence of anatomic variations such as circumaortic renal vein or duplication of the IVC must be defined. Inadequate 331 imaging has been shown to be a major factor for operator errors during percutaneous VCF placements.37 A retrospective review by Savin et al.59 demonstrated significantly increased filter misplacement rate when placement was not preceded by cavography (43% vs. 2.5%, P ⬍ 0.0001). One patient whose filter was misplaced in the iliac vein developed PE from an existing DVT in the contralateral limb. Inadequate imaging can also lead to errors in filter selection. Placement of a filter too small for the caval diameter, for instance, can lead to filter migration and embolism. Currently, only the Bird’s Nest filter is approved for use in an IVC of more than 30 mm in diameter. In patients with megacava, therefore, a Bird’s Nest filter should be used (as long as the IVC diameter does not exceed 40 mm), or 2 filters should be deployed, 1 in each common iliac vein.9 The preferred method of imaging is contrast cavography. Iodinated contrast agents, carbon dioxide, or gadolinium may be used depending on the patient’s renal function, comorbidity, prior contrast load, and/or the need for further contrast imaging. Although not routine in our practice, some authors advocate selective venography before VCF placement.13,30 In their experience, selective venography depicted aberrant findings not seen with standard cavography in 23% to 26% of patients. A significant number of these patients required alterations in placement or selection of the VCF. Duplex and intravascular ultrasound (IVUS) are alternative imaging modalities and have been shown to be effective and safe.10,12,17,69 These can be done at the bedside and are particularly advantageous in the multiply injured or critically ill patient when transport to the operating room or angiography suite can be complex and potentially hazardous. Secondary benefits include costeffectiveness and avoidance of contrast agent use. With duplex ultrasound, visualization of the IVC–right renal vein junction is performed first and used as the anatomic reference for VCF placement. Diameter measurements of the IVC in 2 dimensions and assessment of the patency of the proposed cannulation site are then determined. The femoral approach is chosen preferentially for logistic reasons because preparation of a sterile field cephalad to the bed as needed for jugular access may be extremely inconvenient. Once the filter is positioned in the proper location, deployment with duplex ultrasound guidance is accomplished. Postprocedure plain abdominal radiographs are obtained selectively to confirm filter position. Adequate duplex imaging of the vena cava before filter placement is possible in up to 98.5% of nontrauma patients18 and 90% to 92% of patients with multiorgan trauma.10,12 If the duplex scan imaging results are inadTechniques in Orthopaedics®, Vol. 19, No. 4, 2004 332 L. L. KAW AND R. B. DILLEY equate, then fluoroscopic or IVUS-directed filter placement should be carried out. The ability of IVUS to delineate relevant arterial and venous anatomy with excellent sensitivity and specificity is well known. Using selective venography as the “gold standard,” Ashley et al.6 compared the accuracy of the anatomic information obtained by both IVUS and contrast venography for VCF placement in trauma patients. In this study, IVUS was observed to be significantly more accurate in localizing the renal veins and measuring IVC diameter than contrast venography; however, the disadvantage of using IVUS for VCF placement is that actual filter deployment is done blindly. IVUS is used to define the anatomy and identify the VCF landing zone, which is marked on the patient. The IVUS catheter is then removed and exchanged for the VCF catheter, which is subsequently deployed. In recognition of this limitation, Wellons et al.69 described an innovative technique in which continuous real-time ultrasonography is performed to ensure precise filter deployment. Two punctures 1 cm apart are made on the femoral vein, 1 for the IVUS and another for the IVC filter sheath. Ninetyfour percent of their patients underwent successful VCF placement. With this technique, the authors emphasized the use of IVC filters with low-profile introduction systems (TrapEase and Simon Nitinol) to minimize complications brought about by 2 venipunctures. Following device deployment, the sheath is removed and gentle manual compression is applied for 3 to 5 minutes. Prolonged, aggressive compression is discouraged as a result of significant incidence of insertion site thrombosis. In patients in whom the jugular approach is used, the head of the bed is elevated up to 60° before sheath removal to reduce venous pressure and facilitate hemostasis. We do not routinely anticoagulate patients after VCF placement. Patient factors, however, must be taken into consideration in the decision-making. Although VCFs prevent PE, they have no effect on the prevention and treatment of DVT, and they cannot prevent an extension, recurrence, or the sequelae of postthrombotic syndrome. There is some evidence that anticoagulants lessen the complication of postthrombotic syndrome. INDICATIONS FOR USE Anticoagulation remains the cornerstone for treatment of acute VTE, but in patients with absolute contraindications to or documented failures of anticoagulation therapy, insertion of a VCF is generally indicated. With improved ease of percutaneous insertion provided by reduced and more flexible delivery systems, and the reportedly low complication rates associated with their Techniques in Orthopaedics®, Vol. 19, No. 4, 2004 TABLE 2. Indications and Contraindications for Vena Cava Filter Placement Absolute indications DVT/PE with failure of anticoagulation DVT/PE with complications from anticoagulation DVT/PE with contraindication to anticoagulation (severe gastrointestinal bleeding, intracerebral and retroperitoneal hemorrhage, need for major surgery) Relative indications DVT/PE with pulmonary hypertension and cor pulmonale Free-floating iliofemoral or IVC thrombus Prophylaxis for high-risk trauma and surgical patients DVT/PE in patients with cancer Recurrent PE with IVC filter in place History of DVT in patients for hip/knee replacement and gastric bypass surgery Contraindications Chronically thrombosed IVC No access route to place filter DVT ⫽ deep vein thrombosis; PE ⫽ pulmonary embolus; IVC ⫽ inferior vena cava. use, an expansion of the traditional indications for VCF placement has been observed. In a large populationbased study in the state of California, White et al.70 noted that most patients with VTE treated with an IVC filter had neither of the 2 widely accepted indications for filter use. Indeed, an increasing number of patients are having IVC filters inserted for prophylaxis, broadly interpreted to include patients with or without documented DVT or PE and those in whom the filter was used as an adjunct to anticoagulation. Data from the Michigan Filter Registry revealed that 46% of filters placed in 1999 were for prophylaxis, a dramatic increase from 14% as of 1988.26 Unfortunately, there is little level I data to support these expanded indications. Girard et al.20 reviewed the literature on VCFs published between 1975 and 2000 and concluded that the evidence for the need of VCF placement is weak. Of the 568 references analyzed, only 1 was a randomized, controlled trial, and the few large prospective series were so heterogenous as to preclude relevant comparison and analysis. Nonetheless, guidelines and reporting standards have been formulated for IVC filter insertion.22,27 Strict adherence to these guidelines is important for institutional quality assurance programs and provides physicians the necessary foundation to develop a practical approach to decision-making. Table 2 lists the current indications for IVC filter placement modified from these guidelines. Relatively controversial indications are discussed subsequently. Free-Floating Thrombus Identification of free-floating thrombus (FFT) is often considered an indication for VCF placement based on the reported incidence of PE as high as 60%50; however, the available data are conflicting and inconclusive. A recent USE OF VENA CAVA FILTERS prospective, nonrandomized study reported no higher risk for PE from FFT compared with occlusive thrombus when treated with standard anticoagulation.53 Baldridge et al.8 evaluated the fate of FFT using duplex ultrasound and noted that up to 87% do not embolize but rather become adherent to the vein wall, decrease in size, or resolve. We agree with most authors that anticoagulation should be initiated first and VCF placement be reserved for complications or failures of conservative therapy. Gastric Bypass Surgery Pulmonary embolism is the leading cause of perioperative death in bariatric surgical patients.11,54 Although the reported incidence of PE in this patient population is low (1–2%), PE accounts for nearly one third to one half of deaths after gastric bypass surgery.43,54 This does not appear to be altered by routine use of perioperative DVT prophylaxis. Perioperative VCF placement has been suggested for prophylaxis. In a 24-year retrospective review that looked at specific risk factors for developing fatal PE, Sapala et al.58 identified venous stasis, body mass index of ⬎60 kg/m2, truncal obesity, and hypoventilation syndrome/ obstructive sleep apnea as significant risk factors for VTE. In the presence of a combination of these factors, the authors recommended prophylactic IVC filter placement. Patients With Cancer Since Trousseau’s first observation in 1865, the association between cancer and thrombosis has been well established. The annual incidence of VTE among patients with cancer is estimated to be 1 in 200.41 Chemotherapy increases the risk 6.5-fold.29 Furthermore, a recent prospective, randomized trial comparing the efficacy of dalteparin versus coumadin in preventing recurrent thrombosis in patients with cancer revealed a 13% combined probability of recurrence at 6 months.42 This has led some investigators to adopt an aggressive approach to VTE prophylaxis in these patients with a liberal policy toward the use of IVC filters. On the other hand, the limited life expectancy of patients with cancer may not justify the cost or potential risk for thromboembolic complications. In a retrospective review of patients with malignant disease and VTE, the 1-year survival rate for those treated with an IVC filter was 35%, whereas major recurrent thromboembolic complications developed in 17% of patients.35 Jarrett et al.36 reported a similar, dismal 26.8% survival rate at 1 year in 116 patients undergoing active treatment for malignant disease and who had IVC filters placed for established indications. They concluded that prophylactic filter placement may be of little benefit in this group. 333 High-Risk Trauma Patients Deep vein thrombosis and PE are common complications of trauma. Endothelial damage, changes in the coagulation cascade, and prolonged immobilization occur frequently in multiply-injured patients, rendering them at high risk for VTE. Using serial impedance plethysmography and contrast venography for diagnosis, a prospective evaluation of patients with major trauma (injury severity score ⱖ9) revealed a 58% incidence of lower extremity DVT, 18% of which involved popliteal or more proximal veins.19 In addition, Schultz et al.60 observed a 24% incidence of asymptomatic PE diagnosed by chest computed tomographic scan. Effective, aggressive but safe prophylactic measures are therefore warranted. These generally include anticoagulation therapy, sequential compression devices (SCDs), and a DVT screening protocol with serial venous duplex. There are, however, 2 main issues associated with this approach. First, it has been shown that up to 35% of trauma patients have physical impediments to application of SCDs such as plaster immobilizers, external fixators, complex wounds, or traction.62 A significant number of these patients, especially those with solid organ or head injuries which are managed conservatively, cannot be given anticoagulation. Second, a metaanalysis of existing evidence clearly showed that there is no evidence to support any 1 method of posttraumatic thromboprophylaxis as superior to another method or to no prophylaxis at all.67 Thus, many trauma surgeons have opted for the prophylactic use of VCFs to manage this risk. There is no question that VCFs are efficacious. Numerous studies and a metaanalysis have convincingly shown that patients with a prophylactically inserted VCF have a lower incidence of PE compared with concurrently managed patients without a VCF or historical controls without a VCF.38,40,68 Nonetheless, there is considerable controversy regarding this approach, primarily as a result of the relatively young age of the patient population and the lack of long-term data. The evidence would indicate, however, that the risk– benefit ratio is favorable for prophylactic VCF insertion in high-risk trauma patients. To define high risk, the Eastern Association for the Surgery of Trauma Practice Management Guidelines on VTE identified patients with spinal cord injuries or spinal fractures (level I recommendation), older age, increased injury severity score, and blood transfusion (level II) as risk factors for VTE after injury.55 Patients with multiple long bone fractures, complex pelvic fractures associated with long bone fractures, and severe closed head injuries (Glasgow coma score ⬍8) are likewise considered by many as high-risk injury patterns for PE.38,40,56,72 With the advent of retrievable Techniques in Orthopaedics®, Vol. 19, No. 4, 2004 334 L. L. KAW AND R. B. DILLEY TABLE 3. Summary of Comparison Data Device* No. Simon Nitinol Vena Tech LP Vena Tech LGM Stainless steel Greenfield Titanium Greenfield Gianturco-Roehm Bird’s Nest TrapEase Recovery 319 30 1050 599 511 1426 189 32 Günther Tulip 91 Duration of Follow Up (months) 16.9 2.3 12 26 5.8 14.2 4.2 223 (days) (4–522) 85 (days) (7–420) Recurrent PE (%) IVC Thromboses (%) 3.8 0 3.4 2.6 3.1 2.9 0.5 0 0 DVT Retrieval Success (%) Mean Implantation Period (Days) 7.7 0 11.2 1.7 6.5 3.9 1.5 0 8.9 10.3 32 7.3 22.7 6 1.5 0 — — — — — — — 100 5 0 98 — — — — — — — 53 (5–134) 9 (2–25) *Data for the OptEase filter not available. PE ⫽ pulmonary embolus; IVC ⫽ inferior vena cava; DVT ⫽ deep vein thrombosis. filters, we foresee more widespread application in this patient population. COMPARISON OF INFERIOR VENA CAVA FILTERS The effectiveness of IVC filters is generally measured in terms of its ability to prevent PE while maintaining caval patency. Unfortunately, there is no prospective, randomized study comparing the effectiveness and complications associated with currently available IVC filters. Retrospective studies abound but direct comparison and analysis of pooled data are difficult as a result of wide variations in the population studied, indications, associated treatment, and follow up. Presently, no one filter appears to be superior to the next, but one may have distinct attributes that would be ideal for a particular application. Table 3 summarizes the comparison data of currently available IVC filters based on reviews by Streiff,66 Kinney,39 and individual case series.5,26,45,61 Temporary or Retrievable Filters Concerns regarding the long-term safety of VCFs and the report by Decousus et al.,15 demonstrating only short-term (12 days) benefit of IVC filter insertion in preventing PE, have sparked considerable interest in the use of removable filters. Historically, there are 2 types of removable filters: temporary filter and retrievable filter. Temporary filters are designed for mandatory removal and therefore remain attached to an accessible transcutaneous catheter or wire. Retrievable filters are similar to permanent filters but with modifications to facilitate subsequent removal. These filters may also be left in place indefinitely. If temporary filtration is indicated, it is recommended that these filters be removed within 14 days. Although successful retrieval have been reported Techniques in Orthopaedics®, Vol. 19, No. 4, 2004 up to 134 days,5 significant neointimal proliferation at the attachment sites may preclude safe removal beyond the recommended time period. However, prolongation of implantation time is feasible with repositioning of the device to a different site.14,52 Although indications for use of retrievable filters have not been established, potential use include prophylaxis for high-risk trauma patients with a contraindication to anticoagulation and in patients undergoing gastric bypass surgery or total hip arthroplasty in whom the risk for PE is highest in the perioperative period. We anticipate further expansion of these indications as clinical experience for retrievable IVC filters continues to grow. Removal of retrievable filters can be challenging and fraught with risks. Major complications include retrieval failure, IVC perforation, PE (from captured emboli), and filter embolization. These risks must be weighed against the long-term complications of permanent filter placement. Complications Major complications from IVC filter placement occur infrequently. A large single-center review comprising 1765 VCF insertions reported a major procedural complication rate of only 0.3%.7 Likewise, VCF-related death is a rare event and is usually the result of fatal PE. Other uncommon causes include exsanguination from IVC perforation, pericardial tamponade or fatal arrhythmia from filter migration to the heart, and the serious limb complications of IVC thrombosis. These complications can occur during insertion, shortly after the procedure, or months to years later. The majority of adverse events reported in the literature are associated with longterm use of the device and include recurrent PE and IVC thrombosis. The incidence of recurrent PE and IVC thrombosis among current VCFs is summarized in Table 3. Note that there is a considerable variation in compli- USE OF VENA CAVA FILTERS cation rates between devices and this difference reflects the lack of prospective, randomized data. Short-term complications may be insertion- or deployment-related. Insertion-related complications are associated with obtaining central venous access and include bleeding, infection, pneumothorax, air embolism, arterial injury, and arteriovenous fistula formation. Deploymentrelated complications are usually a function of technique and are largely avoidable with proper imaging and familiarity of the devices used. For instance, incorrect orientation of a filter can occur when the filter is improperly loaded or is placed through an inappropriate access route. All currently available filters with the exception of the Recovery filter can be inserted from the jugular or femoral routes. Moreover, all filters with the exception of the Bird’s Nest and the TrapEase filters need to be oriented correctly depending on the access route chosen. The efficacy and stability of inverted filters are unknown and therefore care must be taken to avoid this complication. Improper placement of a filter can result from inadvertent extrusion of the device during deployment or incorrect estimation of its intended position in the IVC as a result of inadequate imaging or the presence of a variant anatomy, which was missed. This can lead to placement of the filter either higher (suprarenal IVC) or lower (iliac veins) than expected or in another vessel along the access route (renal vein, right atrium). Again, the importance of preprocedural imaging cannot be overemphasized. Full-thickness penetration of the filter through the caval wall is often seen with imaging but is of unknown clinical significance. The majority of cases are most likely asymptomatic; however, erosion of the device into contiguous structures have been reported with disastrous consequences.2,57 CONCLUSION Inferior vena cava filters are effective in providing protection from life-threatening PE. A number of devices are available and provide comparable efficacy and safety profiles. The ease of insertion and low complication rate of currently available devices have expanded the indications for filter placement. Caution, however, should be exercised in their liberal use. The relative lack of solid scientific evidence to determine appropriate indications and the absence of long-term data should be taken into account. Retrievable filters are appealing and may take a broader role in thromboprophylaxis in the future. REFERENCES 1. Adams JT, DeWeese JA. Partial interruption of the inferior vena cava with a new plastic clip. Surg Gynecol Obstet 1966;123:1087–1088. 335 2. Al Zahrani HA. Bird’s nest inferior vena cava filter migration into the duodenum: a rare cause of upper gastrointestinal bleeding. J Endovasc Surg 1995;2:372–375. 3. Amador E, Li TK, Crane C. Ligation of inferior vena cava for thromboembolism: clinical and autopsy correlations in 119 cases. JAMA 1968;206:1758 –1760. 4. Arcelus JI, Caprini JA, Monreal M, et al. The management and outcome of acute venous thromboembolism: a prospective registry including 4011 patients. J Vasc Surg 2003;38:916 –922. 5. Asch MR. Initial experience in humans with a new retrievable inferior vena cava filter. Radiology 2002;225:835– 844. 6. Ashley DW, Gamblin TC, Burch ST, et al. Accurate deployment of vena cava filters: comparison of intravascular ultrasound and contrast venography. J Trauma 2001;50:975–981. 7. Athanasoulis CA, Kaufman JA, Halpern EF, et al. Inferior vena caval filters: review of a 26-year single-center clinical experience. Radiology 2000;216:54 – 66. 8. Baldridge ED, Martin MA, Welling RE. Clinical significance of free-floating venous thrombi. J Vasc Surg 1990;11:62– 67. 9. Baron HC, Klapholz A, Nagy AA, et al. Bilateral iliac vein filter deployment in a patient with megacava. Ann Vasc Surg 1999;13: 634 – 636. 10. Benjamin ME, Sandager GP, Cohn EJ, et al. Duplex ultrasound insertion of inferior vena cava filters in multitrauma patients. Am J Surg 1999;178:92–97. 11. Byrne TK. Complications of surgery for obesity. Surg Clin North Am 2001;81:1183–1193. 12. Conners MS, Becker S, Guzman RJ, et al. Duplex scan-directed placement of inferior vena cava filters: a five-year institutional experience. J Vasc Surg 2002;35:286 –291. 13. Danetz JS, McLafferty RB, Ayerdi J, et al. Selective venography versus nonselective venography before vena cava filter placement: evidence for more, not less. J Vasc Surg 2003;38:928 –934. 14. de Gregorio MA, Gamboa P, Gimeno MJ, et al. The Gunther Tulip retrievable filter: prolonged temporary filtration by repositioning within the inferior vena cava. J Vasc Interv Radiol 2003;14:1259–1265. 15. Decousus H, Leizorovicz A, Parent F, et al. A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis. N Engl J Med 1998;338:409 – 415. 16. Douketis JD, Kearon C, Bates S, et al. Risk of fatal pulmonary embolism in patients treated with venous thromboembolism. JAMA 1998;279:458 – 462. 17. Ebaugh JL, Chiou AC, Morasch MD, et al. Bedside vena cava filter placement guided with intravascular ultrasound. J Vasc Surg 2001; 34:21–26. 18. Friedland M, Kazmers A, Kline R, et al. Vena cava duplex imaging before caval interruption. J Vasc Surg 1996;24:608 – 612. 19. Geerts WH, Code KI, Jay RM, et al. A prospective study of venous thromboembolism after major trauma. N Engl J Med 1994;331: 1601–1606. 20. Girard P, Stern J, Parent F. Medical literature and vena cava filters: so far so weak. Chest 2002;122:963–967. 21. Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet 1999;353:1386 –1389. 22. Grassi CJ, Swan TL, Cardella JF, et al. Quality improvement guidelines for percutaneous permanent inferior vena cava filter placement for the prevention of pulmonary embolism. J Vasc Interv Radiol 2001;12:137–141. 23. Greenfield LJ. Evolution of venous interruption for pulmonary thromboembolism. Arch Surg 1992;127:622– 626. 24. Greenfield LJ, McCurdy JR, Brown PP, et al. A new intracaval filter permitting continued flow and resolution of emboli. Surgery 1973;73:599 – 606. 25. Greenfield LJ, Peyton MD, Brown PP, et al. Transvenous management of pulmonary embolic disease. Ann Surg 1974;180:461– 468. 26. Greenfield LJ, Proctor MC. The percutaneous Greenfield filter: outcomes and practice patterns. J Vasc Surg 2000;32:999 –993. 27. Greenfield LJ, Rutherford RB, et al. Recommended reporting Techniques in Orthopaedics®, Vol. 19, No. 4, 2004 336 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. L. L. KAW AND R. B. DILLEY standards for vena caval filter placement and patient follow-up: vena caval filter consensus conference. J Vasc Interv Radiol 1999;10:1013–1019. Gurewich V, Thomas DP, Rabinov KR. Pulmonary embolism after ligation of the inferior vena cava. N Engl J Med 1966;274:1350 –1354. Heit JA, Silverstein MD, Mohr DN, et al. Risk factors for deep vein thrombosis and pulmonary embolism: a population-based case– control study. Arch Intern Med 2000;160:809 – 815. Hicks ME, Malden ES, Vesely TM, et al. Prospective anatomic study of the inferior vena cava and renal veins: comparison of selective renal venography with cavography and relevance in filter placement. J Vasc Interv Radiol 1995;6:721–729. Homans J. Deep quiet venous thrombosis in the lower limb: preferred levels for interruption of the veins: iliac section or ligation. Surg Gynecol Obstet 1944;79:70 – 82. Homans J. Thrombosis of the deep veins of the lower leg causing pulmonary embolism. N Engl J Med 1934;211:993–997. Horlander KT, Mannino DM, Leeper KV. Pulmonary embolism mortality in the United States, 1979 –1998: an analysis using multiplecause mortality data. Arch Intern Med 2003;163:1711–1717. Hunter JA, Sessions R, Buenger R. Experimental balloon obstruction of the inferior vena cava. Ann Surg 1970;171:315–320. Ihnat DM, Mills JL, Hughes JD, et al. Treatment of patients with venous thromboembolism and malignant disease: should vena cava filter placement be routine? J Vasc Surg 1998;28:800 – 807. Jarrett BP, Dougherty MJ, Calligaro KD. Inferior vena cava filters in malignant disease. J Vasc Surg 2002;36:704 –707. Kaufman JA, Geller SC, Rivitz SM, et al. Operator errors during percutaneous placement of vena cava filters. AJR Am J Roentgenol 1995;165:1281–1287. Khansarinia S, Dennis JW, Veldenz HC, et al. Prophylactic Greenfield filter placement in selected high-risk trauma patients. J Vasc Surg 1995;22:231–236. Kinney TB. Update on inferior vena cava filters. J Vasc Interv Radiol 2003;14:425– 440. Langan EM, Miller RS, Casey WJ, et al. Prophylactic inferior vena cava filters in trauma patients at high risk: follow-up examination and risk/benefit assessment. J Vasc Surg 1999;30:484 – 490. Lee AY, Levine MN. Venous thromboembolism and cancer: risks and outcomes. Circulation 2003;107:I17–21. Lee AY, Levine MN, Baker RI, et al. Low-molecular-weight heparin versus a coumarin for the prevention of recurrent venous thromboembolism in patients with cancer. N Engl J Med 2003;349:146 –153. Melinek J, Livingston E, Cortina G, et al. Autopsy findings following gastric bypass surgery for morbid obesity. Arch Pathol Lab Med 2002;126:1091–1095. Miles RM, Richardson RR, Wayne L, et al. Long-term results with the serrated Teflon vena cava clip in the prevention of pulmonary embolism. Ann Surg 1969;169:881– 891. Millward SF, Oliva VL, Bell SD, et al. Gunther Tulip retrievable vena cava filter: results from the registry of the Canadian Interventional Radiology Association. J Vasc Interv Radiol 2001;12: 1053–1058. Mobin-Uddin K, McLean R, Jude JR. A new catheter technique of interruption of inferior vena cava for prevention of pulmonary embolism. Am Surg 1969;35:889 – 894. Mobin-Uddin K, Smith PE, Martinez LD, et al. A vena cava filter for the prevention of pulmonary embolus. Surg Forum 1967;18:209 –211. Moretz WH, Still JM, Griffin LH, et al. Partial occlusion of the inferior vena cava with a smooth Teflon clip: analysis of long-term results. Surgery 1972;71:710 –719. Nasbeth DC, Moran JM. Reassessment of the role of inferior vena cava ligation in thromboembolism. N Engl J Med 1965;273:1250 – 1253. Norris CS, Greenfield LJ, Herrmann JB. Free-floating iliofemoral thrombus: a risk of pulmonary embolism. Arch Surg 1985;120: 806 – 808. Techniques in Orthopaedics®, Vol. 19, No. 4, 2004 51. Ochsner A, Ochsner JL, Saunders HS. Prevention of pulmonary embolism by caval ligation. Ann Surg 1970;171:923–938. 52. Offner PJ, Hawkes A, Madayag R, et al. The role of temporary inferior vena cava filters in critically ill surgical patients. Arch Surg 2003;138:591–595. 53. Pacouret G, Alison D, Pottier J, et al. Free-floating thrombus and embolic risk in patients with angiographically confirmed proximal deep venous thrombosis. Arch Intern Med 1997;157:305–308. 54. Podnos YD, Jimenez JC, Wilson SE, et al. Complications after laparoscopic gastric bypass: a review of 3464 cases. Arch Surg 2003;138:957–961. 55. Rogers FB, Cipolle MD, Velmahos G, et al. Practice management guidelines for the prevention of venous thromboembolism in trauma patients; the EAST practice management guidelines work group. J Trauma 2002;53:142–164. 56. Rogers FB, Shackford SR, Ricci MA, et al. Routine prophylactic vena cava filter insertion in severely injured trauma patients decreases the incidence of pulmonary embolism. J Am Coll Surg 1995;180:641– 647. 57. Rozin L, Perper JA. Spontaneous fatal perforation of aorta and vena cava by Mobin-Uddin umbrella. Am J Forensic Med Pathol 1989;10:149 –151. 58. Sapala JA, Wood MH, Schuhknecht MP, et al. Fatal pulmonary embolism after bariatric operations for morbid obesity: a 24 year retrospective analysis. Obes Surg 2003;13:819 – 825. 59. Savin MA, Panicker HK, Sadiq S, et al. Placement of vena cava filters: factors affecting technical success and immediate complications. AJR Am J Roentgenol 2002;179:597– 602. 60. Schultz DJ, Brasel KJ, Washington L, et al. Incidence of asymptomatic pulmonary embolism in moderately to severely injured trauma patients. J Trauma 2004;56:727–733. 61. Schutzer R, Ascher E, Hingorani A, et al. Preliminary results of the new 6F TrapEase inferior vena cava filter. Ann Vasc Surg 2003; 17:103–106. 62. Shackford SR, Davis JW, Hollingsworth-Fridlund P, et al. Venous thromboembolism in patients with major trauma. Am J Surg 1990;159:365–369. 63. Silverstein MD, Heit JA, Mohr DN, et al. Trends in the incidence of deep vein thrombosis and pulmonary embolism. Arch Intern Med 1998;158:585–593. 64. Spencer FC, Quattlebaum JK, Quattlebaum JK Jr, et al. Plication of the inferior vena cava for pulmonary embolism. Ann Surg 1962; 155:827. 65. Stein PD, Kayali F, Olson RE. Estimated case fatality rate of pulmonary embolism, 1979 –1998. Am J Cardiol 2004;93:1197–1199. 66. Streiff MB. Vena caval filters: a comprehensive review. Blood 2000;95:3669 –3677. 67. Velmahos GC, Kern J, Chan LS, et al. Prevention of venous thromboembolism after injury: an evidence-based report—Part I: analysis of risk factors and evaluation of the role of vena caval filters. J Trauma 2000;49:132–139. 68. Velmahos GC, Kern J, Chan LS, et al. Prevention of venous thromboembolism after injury: an evidence-based report—Part II: analysis of risk factors and evaluation of the role of vena caval filters. J Trauma 2000;49:140 –144. 69. Wellons ED, Matsuura JH, Shuler FW, et al. Bedside intravascular ultrasound-guided vena cava filter placement. J Vasc Surg 2003; 38:455– 457. 70. White RH, Zhou H, Kim J, et al. A population-based study of the effectiveness of inferior vena cava filter use among patients with venous thromboembolism. Arch Intern Med 2000;160:2033–2041. 71. Williams RW, Schenk WG. A removable intracaval filter for prevention of pulmonary embolism: early experience with the use of the Eichelter catheter in patients. Surgery 1970;68:999 –1008. 72. Winchell RJ, Hoyt DB, Walsh JC, et al. Risk factors associated with pulmonary embolism despite routine prophylaxis: implications for improved protection. J Trauma 1994;37:600 – 606.