guidelines filters

8/3/2019 Guidelines Filters

1/11

Guidelines for the Use of Retrievable andConvertible Vena Cava Filters: Report from theSociety of Interventional RadiologyMultidisciplinary Consensus Conference

John A. Kaufman, MD, Thomas B. Kinney, MD, Michael B. Streiff, MD, Ronald F. Sing, DO, Mary C. Proctor, MS,Daniel Becker, MD, MPH, Mark Cipolle, MD, PhD, Anthony J. Comerota, MD, Steven F. Millward, MD,Frederick B. Rogers, MD, David Sacks, MD, and Anthony C. Venbrux, MD

EDITORS NOTE: Endorsed by the American Venous Forum.

J Vasc Interv Radiol 2006; 17:449459

Abbreviations: DVT deep vein thrombosis, IVC inferior vena cava, PE pulmonary embolism, VTE venous thromboembolism

RATIONALE FOR THECONSENSUSCONFERENCE

REMOVABLE vena cava filters fortemporary protection from pulmonaryembolism (PE) were first proposed in

1967 (1). In 2003 and 2004, the UnitedStates Food and Drug Administrationapproved changes to the instructionsfor use of three existing permanent fil-ters to allow percutaneous retrieval(2). Filter retrieval was added to theinstructions without modification ofthe indications for placement or theaddition of indications for retrieval.

The instructions for use provide phy-sicians with minimal guidance for theuse of these devices as removable fil-ters.

Many medical specialties are in-volved in requesting and placing fil-

ters as well as subsequent patientmanagement. The overall use of venacava filters may be increased by theavailability of nonpermanent devices(3). However, there is a paucity ofmedical literature on these filters (4).On January 14 and 15, 2005, the Soci-ety of Interventional Radiology (SIR)convened a multidisciplinary confer-

ence to address the clinical applicationof nonpermanent vena cava filters.Representatives from interventionalradiology, trauma surgery, vascularsurgery, and internal medicine partic-ipated. The goal of the consensus con-ference was to develop a documentthat would provide clinical guidancefor all physicians who use these venacava filters. Specifically, we sought toaddress the indications for placementof a nonpermanent filter, the manage-ment of patients with such a filter insitu, the conditions for discontinuationof caval filtration, the evaluation ofpatients before discontinuation, andpatient management after discontinu-ation. Filter design, performance, re-porting standards, quality assurance,and recommendations for specific de-vice selection were not included in thediscussions (57).

PURPOSE OF THISDOCUMENT

The intent of this document is toprovide suggestions for the clinical ap-plication of nonpermanent vena cavafilters. These suggestions can beadapted to conform to local practices.More specifically, the document ad-dresses two types of optional filtersavailable or about to become available

From the Dotter Interventional Institute (J.A.K.), Or-egon Health & Science University, Mail Code L-605,

3181 Southwest Sam Jackson Park Road, Portland,Oregon 97239; Department of Radiology (T.B.K.),University of California San Diego Medical Center,San Diego, California; Department of Medicine(M.B.S.), Johns Hopkins University, Baltimore,Maryland; Department of Surgery (R.F.S.), CarolinasMedical Center, Charlotte, North Carolina; Depart-ment of Surgery (M.C.P.), University of MichiganMedical Center, Ann Arbor, Michigan; Departmentof Medicine (D.B.), University of Virginia HealthSystems, Charlottesville, Virginia; Department ofSurgery (M.C.), Lehigh Valley Hospital, Allentown;Department of Radiology (D.S.), The Reading Hos-pital, West Reading, Pennsylvania; Jobst VascularCenter (A.J.C.), Toledo, Ohio; Department of Sur-gery (F.B.R.), University of Vermont, Burlington,Vermont; Department of Radiology (A.C.V.),

George Washington University Medical Center,Washington, DC; and Department of Radiology

(S.F.M.), Peterborough Regional Health Center,Omemee, Ontario, Canada. Received December 17,2005; accepted December 20. Address correspon-dence to J.A.K.; E-mail: [email protected]

Supported by unrestricted educational grants fromanonymous corporate donor, CR Bard, Boston Scien-tific, Cook Group, Cordis Endovascular, and Terumo.

This article will also appear in the March/April 2006issue of Surgery for Obesity and Related Diseases andthe World Journal of Surgery.

J.A.K., T.B.K., R.F.S., and F.B.R. have identified aconflict of interest.

SIR, 2006

DOI: 10.1097/01.RVI.0000203418-39769.0D

Special Communications

449


2/11

in the United States, retrievable andconvertible (as described later). Thedocument does not represent an im-plied, suggested, or legal standard ofcare.

BASIS OFRECOMMENDATIONS

A standardized approach based ontwo types of recommendations exists

for the generation of guidelines to fa-cilitate clinical treatment decisions fortypical patients (8,9). The first is basedon the tradeoff between benefits oftreatment and risks, harm, and costs.The second is based on the method-ologic quality of the underlying evi-dence. Randomized controlled trialswith consistent results provide the ba-sis for the strongest recommendations.Unfortunately, the overwhelming ma-

jority of the published data on venacava filters have come in the form ofobservational studies (10). Published

data were considered insufficient topermit anything more than recom-mendations based on a consensus ofopinions by the writing group. A sum-mary of the major points of consensusis presented in Table 1.

INDICATIONS FORPLACEMENT OF OPTIONALVENA CAVA FILTERS

Venous Thromboembolism

There are numerous risk factors for

venous thromboembolism (VTE), in-

cluding malignancy, thrombophilias,recent major surgery or trauma, in-creased age, acute major medical ill-ness, previous VTE, and morbid obe-sity (1115). Some of these risk factorsare transient, such as major surgeryand trauma, whereas others are per-manent and may even increase in im-portance over time. Careful evaluationof each patient for the presence of riskand duration of the period at risk for

VTE is important when selecting aprevention or treatment strategy.The primary means of prevention

and therapy of VTE is pharmacologic,such as systemic administration of an-ticoagulant agents (12,1620). Antico-agulant therapy usually entails an ac-ceptably small risk of bleeding andother complications (2024). For pre-vention of VTE, there are additionalroles for external mechanical adjunctsincluding but not limited to lower-ex-tremity sequential compression de-vices (12,16,18). Thrombolytic therapy

of established VTE is useful in patientswith severe symptoms or hemody-namic instability (25). Specific proto-cols for pharmacologic prophylaxisand therapy of VTE are beyond thescope of this document but can befound in numerous published docu-ments (12,1625).

Role of Vena Cava Filters inTreatment and Prophylaxis of VTE

The sole function of vena cava fil-ters is to prevent clinically significant

PE by trapping venous emboli. Vena

cava filters do not prevent or treat ve-nous thrombosis. In general, the use ofvena cava filters is indicated when pri-mary therapy cannot be started, must

be stopped, or is insufficient to protect

patients from clinically significant PEwho are at high risk (6,16,17,26,27).When used in patients at risk of devel-oping VTE but who do not yet have it,the purpose of the filter is to preventclinically significant PE should deepvein thrombosis (DVT) occur. Filtersplaced for so-called prophylactic indi-cations do not provide prophylaxis fordevelopment of DVT.

Permanent Vena Cava Filters

Permanent vena cava filters have

been commercially available for morethan 35 years. The overall use of per-manent vena cava filters has increaseddramatically in the United States dur-ing the past 20 years (28,29). The exactpercentage of patients with objectivelyconfirmed VTE who require vena cavafilters is not known, but it is estimatedthat filters are placed in 3%11% ofcases (3,28,30,31). The use of filters inpatients without documented VTE (ie,for prophylactic indications) has in-creased substantially during the pasttwo decades (28,29).

A large body of clinical experienceand published data are available forpermanent filters and are summarizedin several comprehensive reviews (3239). Unfortunately, with few excep-tions, the literature on permanent fil-ters is comprised of uncontrolledretrospective observational series, casereports, and single-center experiences(10). Although vena cava filters aregenerally accepted as safe devices,short- and long-term complicationsmay occur (36,3943). Concerns have

been raised regarding the long-term

benefits and risks of the use of thesedevices in some patients (44).

Nonpermanent Caval FiltrationDevices

There are currently two types of fil-ters that can provide nonpermanentprotection from PE (2,45,46). Optionalfilters are devices that may remainpermanently, whereas temporary fil-ters must be removed as a result oftheir design constraints. There are twosubtypes of optional filters, retrievable

and convertible.

Table 1Key Points of Consensus

1. The primary means of therapy and prophylaxis of VTE are pharmacologic2. No unique indications for optional vena cava filters exist that are distinct from

permanent vena cava filters3. Some patients with indications for vena cava filters have limited periods of risk

of clinically significant PE and/or contraindication to anticoagulation and maynot require permanent protection from PE with a vena cava filter

4. Patients with filters in situ should be managed with pharmacologic methodsaccording to their VTE status and risk of anticoagulation as soon as safe andfeasible

5. There are no absolute indications for discontinuation of filtration unless the filteritself is a source of documented major morbidity that will be relieved by retrievalor conversion

6. Discontinuation of filtration should only occur when the risk of clinicallysignificant PE is reduced to an acceptable level and is estimated to be less thanthe risk of leaving the filter in situ

7. The quality of literature on optional vena cava filters is not sufficient to supportevidence-based recommendations at this time

450 Guidelines on the Use of IVC Filters March 2006 JVIR


3/11

Retrievable filters are defined inthis document as permanent devicesdesigned for imaging-guided percuta-neous removal during device-specifictime windows of retrievability with

use of catheter-based technologies.These filters can be placed with orwithout the intent to be retrieved de-pending on the indications and clinicalcircumstances. Like all permanent fil-ters, retrievable devices initially affixto the wall of the vena cava by meansof hooks, barbs, or radial pressure.Over time, some filter elements be-come adherent to the wall of the venacava as a result of endothelial over-growth (4753). Filters that are not re-trieved function as permanent filters.

Convertible devices are defined in

this document as permanent filtersthat can be altered structurally afterimplantation to no longer function asfilters. These filters can be placed withor without the intent to be convertedto a nonfiltration state depending onthe indications and clinical circum-stances. These devices usually main-tain their position in the vena cavawith hooks, barbs, or radial pressure.Filter elements are subject to endothe-lial overgrowth. The filtering capacityof the device can be eliminated duringa percutaneous imaging-guided cath-

eter-based procedure. After conver-sion, some or all of the filter remains inthe patients vena cava but does notprovide protection from PE. Filtersthat are not converted provide perma-nent filtration.

Temporary filters are defined inthis document as devices that are notdesigned for permanent placement.These are not currently available in theUnited States. Frequently, these de-vices do not have hooks, barbs, or anyother means for fixation to the wall ofthe vena cava, but are supported in

place by tethers or catheters. The sup-porting elements frequently traverselong segments of the central venoussystem, are externalized, and occupy avenous access site. Removal of thesedevices is required within severalweeks before the filter and/or tether

becomes adherent to the venous walls(46,54,55). Removal is usually an imag-ing-guided procedure. Permanent filtra-tion requires removal of the temporaryfilter and placement of a different de-vice.

Although the general principles in

this document can be applied to op-

tional and temporary filters, the focushenceforth will be on optional filters.

The available data on optional de-vices are even more limited than onpermanent filters (4). The few pub-

lished clinical reports (4,5661) sug-gest that optional filters, as a group,are associated with equivalent out-comes to permanent devices. How-ever, there has never been a clinicalstudy to directly compare optionaland permanent filters. Percutaneousremoval of retrievable filters appearsto be a safe procedure with few imme-diate or late complications (56,57,60,61). Published human data on con-vertible filters are currently lacking.The benefit of discontinuation of filtra-tion has never been proven, but is in-

ferred from the data available on per-manent devices (62). Until more dataare available, permanent caval filtra-tion should always be consideredwhen a filter is indicated.

Indications for Optional Filters

Placement of an optional filter withthe intent to discontinue filtrationthrough retrieval or conversion shouldfollow the same indications used forpermanent vena cava filters. The deci-sion to use an optional filter rather

than a permanent filter should bebased on the anticipated required du-ration of protection against clinicallysignificant PE and/or risk of pharma-cologic therapy (Fig 1). There are nonew unique indications for optionalvena cava filters distinct from perma-nent filters.

In general, indications for all venacava filters have been divided into ab-solute, relative, and prophylactic cate-gories (Table 2) (5,6,27,36,39,63). Pa-tients with absolute indications havedocumented VTE, are at high risk of

clinically significant PE, and have acontraindication to or complication orfailure of pharmacologic therapy(5,6,17,27,36,39,63). In some patientswith contraindications to or complica-tions of anticoagulation, the periodduring which anticoagulant therapycannot be used may be temporary ortransient (22). Optional vena cava fil-ters can be considered in these situa-tions.

Patients with relative indicationsfor vena cava filters have VTE and areconsidered to be at continued high risk

of clinically significant PE despite pri-

mary therapy, at increased risk ofcomplications of anticoagulation, ornoncompliant with medications (Ta-ble 2). In general, the data supportingthe clinical value of filters for relative

indications are more sparse than thosefor absolute indications. When the pe-riod of increased risk for clinically sig-nificant PE or complications of antico-agulation is of short duration inpatients with relative indications forfilters, optional filters may be consid-ered.

Patients with prophylactic indica-tions for filters do not have VTE, butare at increased risk for developmentof clinically significant PE and unableto undergo primary prophylaxis. Al-though placement of vena cava filters

for these indications is common, thedata supporting the practice are weak.The one possible exception is filterplacement in a setting of trauma in apatient deemed to be at high risk(64,65). Additional specific clinical sce-narios in which the use of prophylacticfilters have been reported (but are ofunproven benefit) include critically illpatients with a history of VTE andcontraindication to anticoagulation(66,67), perioperative settings in pa-tients with a history of VTE and con-traindication to anticoagulation (2),

and patients undergoing bariatric sur-gery (68,69). When the period of in-creased risk for clinically significantPE is of short duration, optional filtersmay be considered.

MANAGEMENT OF PATIENTSWITH AN OPTIONAL VENACAVA FILTER IN SITU

Patients with optional vena cava fil-ters require tracking and routine fol-low-up for several reasons. The win-dow of retrievability varies for each

device, patient conditions may changesuch that discontinuation of the filteris no longer desired or safe, and pri-mary physicians may require guid-ance on the timing of filter discontin-uation. Preferably, the physician whoplaces the filter should perform thisfollow-up.

Patients with VTE and a vena cavafilter should undergo primary therapyat the first safe opportunity regardlessof the presence of a vena cava filter(Fig 2). Evidence-based reviews oftherapy for VTE are available to guide

therapy (16,17,20,7175). In patients

Kaufman et al 451Volume 17 Number 3


4/11

with VTE and a vena cava filter placedfor a contraindication to anticoagula-tion, the patient should receive antico-agulation as soon as the period of ex-cess hemorrhagic risk has passed. Inthe subgroup of patients with VTEwhose contraindication to anticoagu-lation is major surgery, data from ran-

domized clinical trials are lacking, but

several reviews are available to informclinical decision-making regarding re-sumption of primary therapy (72,74).For patients with VTE and a vena cavafilter placed for a major hemorrhagiccomplication associated with antico-agulation, the physician may restartanticoagulation when the clinical situ-

ation dictates that it is safe. This deci-

sion must be made on a case-by-case basis after the risks and benefits ofanticoagulation are weighed. No evi-dence-based guidelines exist to assistclinical decision-making. However,small case series suggest that anticoag-ulation may be reinstated when thecausal anatomic lesions have been de-finitively corrected (74,76). Neverthe-less, these patients appear to be athigher risk for subsequent episodes ofmajor bleeding (76). Tighter control ofanticoagulation management (eg, morefrequent International Normalized Ra-tio determinations) and laboratory sur-

veillance for recurrent hemorrhage are

Figure 1. Algorithm for placement of vena cava filters. PE Pulmonary embolism; AC anticoagulation; O optional vena cava filter (in patients with limited life expectancy[6 months], there may be limited benefit from retrieval/conversion of a filter [2,45,70]);P permanent vena cava filter.

Table 2Indications and Contraindications forAll Vena Cava Filters

Absolute Indications (Proven VTE)Recurrent VTE (acute or chronic)

despite adequate anticoagulationContraindication to anticoagulationComplication of anticoagulationInability to achieve/maintain

therapeutic anticoagulationRelative Indications (Proven VTE)

Iliocaval DVTLarge, free-floating proximal DVTDifficulty establishing therapeutic

anticoagulationMassive PE treated with

thrombolysis/thrombectomyChronic PE treated with

thromboendarterectomyThrombolysis for iliocaval DVT

VTE with limited cardiopulmonaryreserve

Recurrent PE with filter in placePoor compliance with anticoagulant

medicationsHigh risk of complication of

anticoagulation (eg, ataxia,frequent falls)

Prophylactic Indications (No VTE,primary prophylaxis notfeasible*)

Trauma patient with high risk ofVTE

Surgical procedure in patient at highrisk of VTE

Medical condition with high risk ofVTEContraindications to Filter Placement

No access route to the vena cavaNo location available in vena cava

for placement of filter

* Primary prophylaxis not feasible as aresult of high bleeding risk, inability tomonitor the patient for VTE, etc.



5/11

probably worthwhile additions to rou-tine clinical management.

In patients with filters placed forprophylactic indications, primary VTEprophylaxis should be initiated basedon the underlying risk factors at thefirst safe opportunity regardless of thepresence of a vena cava filter. Evi-dence-based guidelines on VTE pro-phylaxis are available to guide deci-

sion-making (16). Bleeding risk should be formally assessed by the patientcare team on a daily basis with thegoal to institute appropriate anticoag-ulant VTE prophylaxis. In addition,mechanical forms of VTE prophylaxis(eg, graduated compression stockings,sequential/pneumatic compression de-vices) should be used whenever pos-sible. The filter does not provide pro-phylaxis for venous thrombosis, butrather only for PE should DVT develop.For patients at high risk who are receiv-ing suboptimal VTE prophylaxis, du-

plex ultrasonographic (US) surveillancefor DVT should be considered. In theevent that a patient with a filter placedfor prophylactic indications developsVTE, appropriate primary therapyshould be initiated as soon as is safe.

WHEN TO CONSIDERDISCONTINUATION OFVENA CAVA FILTRATION

General

The sole purpose and function of a

vena cava filter is to prevent clinically

significant PE. For this reason, the fun-damental clinical criterion for discon-tinuation of caval filtration is an ac-ceptably low risk of PE (Fig 3). In mostinstances, this will occur when the pa-tient is receiving satisfactory treatmentwith primary therapy or has passedthe period of risk for VTE (11,14,16,17,34,35,77). When a patient is con-sidered to have an acceptably low risk

of PE, the presumed risks of the filtermust be weighed against the esti-mated future risk of recurrent PE. As aresult of the inadequacies of publisheddata on filters and the complexity ofreal clinical situations, the balance ofrisks cannot be quantified at thepresent time and remain a matter ofphysician judgment. The decision todiscontinue filtration must be individ-ualized in each case. In the clinicalcontext, the decision to not retrieve orconvert a satisfactorily functioning fil-

ter is acceptable.When the local standard of care isto prescribe lifelong anticoagulationsimply because of the presence of avena cava filter, discontinuation of fil-tration should be considered. Routineanticoagulation of patients with venacava filters to prevent recurrent DVTand other thrombotic complications iscontroversial (78,79). Long-term anti-coagulation is associated with a small

but definable incidence of complica-tions, usually hemorrhagic (22). Dis-continuation of filtration to avoid life-

long anticoagulation after the patient

no longer requires treatment of VTEmay be warranted.

Recommendations beforeDiscontinuation of Filtration for All

Patients

Vena cava filtration may be discon-tinued in the following clinical scenar-ios:

1. An indication for a permanentfilter is not present. Patients chroni-cally at high risk of clinically signifi-cant PE irrespective of managementwith primary therapy, short life ex-pectancies (6 months), or noncom-pliance with primary therapy or fol-low-up appointments should have apermanent filter. All retrievable and

convertible filters can provide perma-nent protection from PE.2. The risk of clinically significant

PE is acceptably low as a result ofachievement of sustained appropriateprimary treatment (therapy or pro-phylaxis) or change in clinical status.Appropriate primary treatment willvary depending on the VTE statusof the patient (16,17,19,22,23,65,73,8082). Patients should demonstrate theability to tolerate sustained primarytreatment before discontinuation of fil-tration. The period of time required

will vary based on the patients VTEstatus and local standards of care (asdescribed later).

3. The patient is not anticipated toreturn to a high-risk state for PE be-cause of interruption of primary treat-ment, change in clinical management,or change in clinical condition. For ex-ample, filtration should not be discon-tinued when a patient will shortly un-dergo additional procedures thatwould normally require placementof a vena cava filter. Although it isnot always possible to identify which

patients will return to a high-risk sta-tus, careful review of all aspects ofthe patients care should be con-ducted before filter retrieval or con-version.

4. The life expectancy of the patientis long enough that the presumed ben-efits of discontinuation of filtrationcan be realized. Limited evidence sug-gests that some suspected complica-tions of filters take years to manifest(42,43). Patients not anticipated to sur-vive more than 6 months are unlikelyto derive any discernible benefit from

filter retrieval or conversion.

Figure 2. Algorithm for management of patients with filters in situ. AC anticoagula-tion; DUS duplex ultrasound scan.



6/11

5. The filter can be safely retrievedor converted. Patients with unmanage-able issues related to contrast agents (eg,allergy or renal insufficiency) or coagu-lopathies should not undergo the proce-dure until adequate prophylaxis or cor-rection has been performed. Suitable

venous access for the procedure shouldbe available. Last, filters that the treatingphysician judges cannot be safely re-trieved or converted without causingunacceptable injury to the vena cavashould not be manipulated.

6. The patient or consenting guard-ian agrees to have the filter removedor converted. Patients who desire tocontinue caval filtration permanentlyshould be allowed to so.

Specific Recommendations forPatients with VTE

Patients with filters and establishedVTE should be treated with primarytherapy as soon as possible (17). Thepresence of the filter should not alterthe intensity or duration of anticoagu-lation. The period of highest risk forPE in patients undergoing therapeuticanticoagulation for VTE cannot be pre-cisely defined because of the heteroge-neity of published studies. Severalstudies (42,83,84) suggest that symp-tomatic PE is most likely to occurwithin 23 weeks of the initial episode

of VTE. The duration of highest risk

for clinically significant PE during pri-mary therapy should be estimated in-dividually for each patient. Patientswho require laboratory monitoring foranticoagulation should have stablemeasurements with no evidence of

bleeding for at least 7 days before the

discontinuation procedure.Patients with established VTE shouldnot have clinical or objective evidence offailure or a complication of primarytherapy before filter retrieval. These sit-uations warrant continued caval filtra-tion, perhaps permanently (39).

Specific Recommendations forPatients without VTE

Patients in whom filters are placedfor prophylaxis (ie, those at risk for

but without a diagnosis of VTE)

should be treated with primary pro-phylaxis as soon as possible accordingto published guidelines (16,19,65). Dis-continuation of filtration should notoccur until adequate primary prophy-laxis has been consistently achievedand can be maintained or the high riskof clinically significant PE has abated.In some patients, the period of risk fordevelopment of VTE may end beforediscontinuation of vena cava filtration.At all times, the underlying conditionof the patient should determine theneed and type of VTE prophylaxis.

When a filter has been placed for

prophylactic indications, there shouldbe no clinical or objective evidence ofinterval development of VTE beforediscontinuation of filtration. A patientwho develops VTE while a prophylac-

tic filter is in place should be treatedwith primary therapy for VTE, as out-lined earlier (17,73,80). The patientshould be at an acceptably low risk ofclinically significant PE before discon-tinuation of filtration is considered.

Unique Situations

When a retrievable filter has be-come a source of severe morbidity oris no longer protective for PE as a re-sult of change in filter position or lossof structural integrity, removal may be

indicated. These cases are very rare,but examples include maldeployment,unmanageable pain related to perfora-tion, filter instability, and dislodge-ment during other procedures (40,41,85). Conclusive objective evidence offilter-related morbidity should besought before filter removal for thisindication. These occurrences should

be reported according to standard de-vice-tracking practices. Placement ofanother filter may be necessary whenretrieval of an in situ device is indi-cated despite the presence of a con-

tinuing indication for a vena cava filter.

PATIENT EVALUATIONBEFORE DISCONTINUATIONOF FILTRATION

Preprocedural Evaluation

The goals of patient evaluation be-fore filter retrieval or conversion are toensure that the risk of clinically sig-nificant PE is acceptably low and thatthe retrieval or conversion procedurecan be performed safely (Fig 4). A fo-

cused history and physical examina-tion should be performed to assess forsigns of VTE. Patients suspected of hav-ing new, recurrent, or progressive VTEon the basis of symptoms or clinicalfindings should undergo diagnostic im-aging procedures to resolve this ques-tion before proceeding. Routine coag-ulation measurements and complete

blood counts are appropriate for theassessment of patients receiving ther-apeutic anticoagulation. Renal func-tion assessment should be consideredin patients at risk for contrast agent

nephropathy.

Figure 3. Algorithm for patient selection for discontinuation of vena cava filtration.*Risk of PE determined by patients current venous thromboembolic disease status,underlying conditions, and tolerance of primary therapy or prophylaxis. **Requiresconsideration of life expectancy and need for caval filtration in near future.



7/11

Patients who have documentedVTE should have laboratory values inthe appropriate therapeutic range

based on laboratory monitoring (17).Laboratory monitoring is not neces-sary for the majority of patients who

are receiving low molecular weightheparins. Patients who have under-gone anticoagulation and require lab-oratory monitoring should have stablemeasurements with no evidence of

bleeding for at least 7 days. In thepresence of acute VTE, a delay of 23weeks after initiation of primary ther-apy should be considered before dis-continuation of vena cava filtration(42,83,84). When coagulation studiesare warranted for anticoagulationmonitoring, they should be performedthe day of the discontinuation proce-

dure to ensure that the measurementsare within the appropriate therapeuticranges. It is neither necessary nor de-sirable to interrupt anticoagulation forthe retrieval or conversion procedure.

Patients with VTE who have re-ceived full anticoagulation and are instable condition while receiving pri-mary therapy, and who have no new,recurrent, or progressive symptoms orclinical findings of VTE, do not requireadditional imaging of the extremityveins or pulmonary arteries before dis-continuation of vena cava filtration. If

imaging performed for clinical indica-

tions reveals new or progressive VTE,repeat evaluation of the anticoagulationregimen and delay of filter discontinua-tion should be considered. These pa-tients may be reassessed at a later datefor filter retrieval or conversion or the

filter may be considered a permanentdevice.

Patients without a known diagnosisof VTE should undergo imaging of thelower-extremity veins such as duplexvenous US before discontinuation offiltration. A finding of DVT in thesepatients requires postponement of fil-ter discontinuation and initiation ofprimary therapy if not contraindicated(17). The patient should exhibit stan-dard therapeutic levels of anticoagula-tion for at least 23 weeks before filterdiscontinuation is reconsidered. Alter-natively, the filter may remain in placeas a permanent device.

The physician responsible for remov-ing or converting the filter should con-firm the appropriateness of discontinu-ation of filtration before the procedure.During the consent process, the follow-ing should be discussed in addition tothe usual elements of informed consent:(i) the rationale for discontinuation offiltration, (ii) the possibility that filterretrieval or conversion may not be pos-sible, and (iii) the voluntary nature of

discontinuing caval filtration.

Imaging of the Filter and Vena Cava

Imaging of the vena cava and im-planted filter can be performed at thetime of the discontinuation procedurewith catheter-based techniques orwithin the preceding 24 hours with atechnique that permits evaluation ofthe entire filter and the vena cava suchas contrast material enhanced com-puted tomography (CT), magnetic res-onance venography, or US. In patientswith known VTE, identification ofthrombus in the filter requires an as-sessment of the risk of clinically signif-icant PE at the time of filter discontin-uation or subsequent to the procedure(Fig 5). There are no published data tosupport guidelines for making this as-sessment; an individualized decision

by the physician will be required ineach case. For example, substantial fill-ing defects within the filter present animmediate risk of PE during filter re-trieval or conversion and may indicatean ongoing embolic risk. Conversely,subcentimeter filling defects adherentto filter elements may pose little risk ofPE during filter retrieval or conversionand may imply a resolved embolicrisk. In some cases, based on the com-

bined judgment of the physicians withprimary responsibility for retrieval or

conversion of the filter and manage-ment of anticoagulation, the patientcan return after a period of weeks forrepeat imaging and reconsideration ofdiscontinuation of filtration.

In patients without known VTE be-fore the filter retrieval or conversionprocedure, identification of trappedthrombus within the filter constitutesa new diagnosis of VTE regardless ofthe absence of clinical symptoms ornormal results of lower-extremity ve-nous US. The filter discontinuationprocedure should be terminated and

appropriate primary therapy should be instituted unless contraindicated.Reassessment for discontinuation offiltration may be considered at a latertime, as described earlier.

MANAGEMENT AFTERDISCONTINUATION OFFILTRATION

During the Procedure

After retrieval or conversion of afilter, imaging of the vena cava may be

performed as part of the procedure to

Figure 4. Algorithm for patient evaluation before discontinuation of vena cava filtration.



8/11

assess for evidence of caval trauma orthrombus. Caval imaging is stronglyrecommended after difficult or pro-longed procedures or when a patientreports significant pain during and/orafter the procedure. Findings sugges-tive of caval injury such as intimalflaps or extravasation of contrast me-dium should be managed according tothe degree of the abnormality and lo-cal standards of care. There are few

data in the published literature toguide management in these cases, al-though minor abnormalities on cavalimaging have been followed expect-antly or with anticoagulation withoutclinical sequelae (86). The presence ofobstructive intimal flaps or thrombusmay require additional interventionsto preserve patency of the vena cava.Patients with suspected retroperito-neal bleeding after discontinuation offiltration may require evaluation withCT, serial hematocrit measurements,and close clinical observation.

Retrieved or converted filters

should be inspected for integrity (di-rectly or with imaging) by the physi-cian performing the procedure. If afilter is incomplete, careful evaluationof the retrieval or conversion catheterand imaging of the patients abdomenand chest should be performed to lo-calize the missing components anddocument their position. There are nogenerally accepted guidelines for thetreatment of patients with retained

fractured or detached filter elements.Retroperitoneal and intrapulmonaryfragments are rarely symptomatic. In-tracardiac fragments should be man-aged in consultation with a cardiacspecialist.

After the Procedure

Patients should be treated accord-ing to their VTE status and underlyingconditions after retrieval or conver-sion of a filter (16,17,22,65,7275,80).There are no specific additional thera-

pies required after discontinuation of

vena cava filtration. Patients with VTEshould continue to receive primarytherapy for the full duration suggestedin published practice guidelines or ac-cording to local standards of care (17).

Patients without VTE should undergoprophylaxis with use of standard tech-niques appropriate for any underlyingconditions (16). All patients should bemonitored for new, recurrent, or pro-gressive DVT and/or PE, and if diag-nosed, managed accordingly.

FUTURE RESEARCH

Background

For nearly 40 years, vena cava filterresearch has focused on three areas: in

vitro simulations of performance, invivo tests of function, and patient out-comes. The initial goal of this workwas to acquaint physicians with thepurpose of these devices and to dem-onstrate their advantages comparedwith earlier methods of caval interrup-tion. Over time, this approach becamethe primary means of evaluating newdesigns and comparing competitivedevices (87).

In vitro studies often compare twoor more devices and focus on charac-teristics such as clot trapping ability,

resistance to movement, and flow dy-namics. Device manufacturers useproprietary in vitro testing routines toevaluate prototypes and obtain infor-mation before the initiation of animaltesting. Some tests are validatedagainst an objective standard such asthe performance of earlier devices,whereas others rely on face validity(8892).

In vivo studies of inferior vena cavafilters are typically conducted in ani-mal models with an inferior vena cavasize similar to that of humans (47,

49,52). The purpose of animal studiesis to assess some aspects of device per-formance in a biologic system. Cur-rently, animal studies are used to testdevice placement, biocompatibility,stability, impact on caval wall and sur-rounding structures, thrombus trap-ping, and removal. Testing in animalmodels does not evaluate all aspects ofperformance, especially filter efficacyin a clinical setting, and long-term out-comes.

Human research in filters has beennotably devoid of randomized pro-

spective studies. Almost all large

Figure 5. Algorithm for management of thrombus found in a filter before or during adiscontinuation procedure. VTE venous thromboembolism such as deep vein throm-bosis or pulmonary embolism. *The determination of the volume and age of thrombuspresent in a filter is made by the physician performing the retrieval or conversionprocedure, as is the clinical significance of the thrombus. Whenever uncertainty prevailsabout the volume, age, or clinical significance of thrombus in the filter, the filter shouldremain in place.



9/11

studies are observational in natureand often lack a control group. Deter-mination of outcomes and complica-tions of filters has been based on datathat are highly variable in quality, ob-

jectivity, and reliability. Althoughstandardized reporting guidelines forstudies of filters have been developed,they have not been uniformly adopted

by authors or journals (5,7). Significantcomplications from IVC filters are in-frequent. Recognition of these rareevents in small observational studiesis an unreasonable expectation (57,93,94). Similarly, large randomized tri-als before commercialization of newdevices are also unreasonable becausethe number of subjects needed and thecost of the required studies would pre-

vent new devices from coming to themarket.

Research Imperatives

General.The introduction of op-tional and temporary vena cava fil-ters has created renewed interest inexisting questions about filters andraised new questions that need to beinvestigated. These questions rangefrom the long-term clinical benefitsof filters to the clinical outcomes af-ter removal of a filter from the vena

cava. Additional levels of testing arenecessary at the in vivo, in vitro, andclinical levels for devices that un-dergo manipulation or removal afterinitial placement (49,52,89,91,92). Ta-ble 3 lists several basic questions re-garding performance of IVC filters.

In vitro testing.A significant con-cern with current in vitro testing isthe lack of standardization for per-manent and optional devices. Re-search is needed to develop uni-formly accepted in vitro tests that ac-curately predict in vivo function in

terms of vena cava flow dynamics,the filters clot trapping ability, filterstability, filter durability, and theability to remove or convert the filter.These tests need to be sufficiently ro-

bust to allow for differences in filterdesign characteristics and still be ableto reliably measure performance pa-rameters. Computerized modelingand simulations may provide rapidmethods to investigate multiple filterdesigns. Standardized in vitro testingregimens should be developed byrepresentatives from industry, aca-

demia, and the Food and Drug Ad-

ministration. The tests could then beapplied before in vivo evaluation andthe results required as part of the ap-proval process for marketing applica-tions.

In vivo testing.In vivo testinglacks standardization for permanent

and optional devices. As suggestedfor in vitro testing, research isneeded to develop a uniformly ac-cepted animal model for testing fil-ters. The extrapolation from animalmodels to human experience remainsinferential. Correlation of hemody-namic, pathologic, and clinical out-comes in animal models with subse-quent human clinical experience isnecessary. The short- and long-termimpact of removable and convertiblefilters on the vena cava and its envi-rons has been studied only in ani-

mals, yet this remains a central con-cern in the human application ofthese devices.

Clinical testing.A national filterregistry has been suggested by in-dustry, professional organizations,and academic centers at varioustimes during the past 15 years. Ques-tions regarding the role of tempo-rary, permanent, and optional venacava filters in prevention of PE needto be addressed. With the use ofsound design principles, it is possibleto prospectively collect sufficient

data to study these issues and to

guide physician practice. Such a reg-istry would provide insight into theuse of vena cava filters, an under-standing of the types of patients be-ing treated, and the long-term effects

of different filter designs. Changes inpractice can be observed and studied.Major professional organizations in-volved with vascular interventionalprocedures and the treatment of pa-tients with VTE should jointly orga-nize and create such a registry.

Ideally, large randomized prospec-tive studies will ultimately be per-formed that will address fundamentalunanswered questions about filters,such as the validity of the current in-dications for these devices, impact onmortality from PE after filter place-

ment in specific patient populations,and long-term outcomes. Optional,temporary, and future filter technolo-gies should be studied in the samemanner.

New filter materials, devices that can be separated or collapsed, absorbablefilters, and drug-eluting filters are cur-rently being studied. The cost of re-search can be controlled while the paceof advancement increases through pro-grams that integrate in vitro and in vivostudies and use the resulting data todirect clinical development (95). Knowl-

edge of the strengths and liabilities ofthe present generation of low-profile op-tional filters can provide the foundationfor future filter development.

References1. Williams R, Schenk W. A removable

intracaval filter for prevention of pul-monary embolism: early experiencewith the use of the Eichelter catheter inpatients. Surgery 1970; 68:9991008.

2. Kaufman J. Retrievable vena cava fil-ters. Tech Vasc Interv Radiol 2004;7:96104.

3. Quirke T, Ritota P, Swan K. Inferiorvena caval filter use in U.S. traumacenters: a practitioner survey. J Trauma1997; 43:333337.

4. Stein P, Alnas M, Skaf E, et al. Out-come and complications of retrievableinferior vena cava filters. Am J Cardiol2004; 94:10901093.

5. Greenfield L, Rutherford R. Recom-mended reporting standards for venacaval filter placement and patient fol-low-up. Vena Caval Filter ConsensusConference. J Vasc Interv Radiol 1999;10:10131019.

6. Grassi C, Swan T, Cardella J, et al.Quality improvement guidelines for

percutaneous permanent inferior vena

Table 3Suggested Research Topics

Validated standardized in vitro testsfor filters

Validated standardized animal modelsfor filters

Multiinstitutional clinical registry offilters and/or randomizedprospective clinical trials to:

Validate the current relative andprophylactic indications for filters

Compare clinical performance ofoptional filters with permanentfilters

Compare the clinical performance ofindividual devices

Identify patient populations suitablefor short-term caval filtration

Determine criteria fordiscontinuation of filtration

Determine clinical outcomes of filterremoval or conversion

Determine cost effectiveness ofdiscontinuation of filtration



10/11

cava filter placement for the preven-tion of pulmonary embolism. SCVIRStandards of Practice Committee. JVasc Interv Radiol 2001; 12:137141.

7. Millward S, Grassi C, Kinney T, et al.Reporting standards for inferior vena

caval filter placement and patient fol-low-up: supplement for temporary andretrievable/optional filter. J Vasc In-terv Radiol 2005; 16:441443.

8. Guyatt D, Schunemann H, Cook D, etal. Grades of recommendations forantithrombotic agents. Chest 2001;119(suppl 1):S3S7.

9. Sackett D. Rules of evidence and clin-ical recommendations on the use of an-tithrombotic agents. Arch Intern Med1986; 146:464465.

10. Girard P, Stern J, Parent F. Medicalliterature and vena cava filters: so farso weak. Chest 2002; 122:963967.

11. Anderson F, Spencer F. Risk factorsfor venous thromboembolism. Circula-tion 2003; 107:I9I16.

12. Haas S. Venous thromboembolic riskand its prevention in hospitalized med-ical patients. Semin Thromb Hemost2002; 28:577584.

13. Hansson P, Sorbo J, Eriksson H. Re-current venous thromboembolism afterdeep vein thrombosis: incidence andrisk factors. Arch Intern Med 2000; 160:769774.

14. Heit J. Risk factors for venous throm-boembolism. Clin Chest Med 2003; 24:112.

15. Kearon C. Epidemiology of venousthromboembolism. Semin Vasc Med2001; 1:726.

16. Geerts W, Pineo G, Heit J, et al. Pre-vention of venous thromboembolism:the Seventh ACCP Conference on An-tithrombotic and Thrombolytic Ther-apy. Chest 2004;126(suppl):338S400S.

17. Buller H, Agnelli G, Hull R, et al. An-tithrombotic therapy for venousthromboembolic disease: the SeventhACCP Conference on Antithromboticand Thrombolytic Therapy. Chest2004;126(suppl):401S428S.

18. Gerotziafas G, Samama M. Prophy-laxis of venous thromboembolism inmedical patients. Curr Opin Pulm Med2004; 10:356365.

19. Kearon C. Duration of venous thromboembolism prophylaxis after surgery.Chest 2003;124(suppl):386S392S.

20. McRae S, Ginsberg J. Initial treatmentof venous thromboembolism. Circula-tion 2004; 110:I3I9.

21. Warkentin T, Greinacher A. Hepa-rin-induced thrombocytopenia: recog-nition, treatment, and prevention: theSeventh ACCP Conference on Anti-thrombotic and Thrombolytic Therapy.Chest 2004;126(suppl):311S337S.

22. Levine M, Raskob G, Beyth R, et al.

Hemorrhagic complications of antico-

agulant treatment: the Seventh ACCPConference on Antithrombotic andThrombolytic Therapy. Chest 2004;126(suppl):287S310S.

23. Kearon C. Long-term management ofpatients after venous thromboembo-

lism. Circulation 2004; 110:I10I18.24. Hirsh J, Bates S. Clinical trials that have

influenced the treatment of venousthromboembolism: a historical perspec-tive. Ann Intern Med 2001; 134:409417.

25. Goldhaber S, Bounameaux H. Throm-bolytic therapy in pulmonary embolism.Semin Vasc Med 2001; 1:213220.

26. Proctor M. Indications for filter place-ment. Semin Vasc Surg 2000; 13:194198.

27. Levy J, Duszak R, Akins E, et al. In-ferior vena cava filter placement.American College of Radiology. ACRAppropriateness Criteria. Radiology2000;215(suppl):981997.

28. Stein P, Kayali F, Olson R. Twenty-one-year trends in the use of inferiorvena cava filters. Arch Intern Med2004; 164:15411545.

29. Athanasoulis C, Kaufman J, Halpern E,et al. Inferior vena caval filters: re-view of a 26-year single-center clinicalexperience. Radiology 2000; 216:5466.

30. Anderson F, Wheeler H. Physicianpractices in the management of venousthromboembolism: a community-widesurvey. J Vasc Surg 1992; 16:707714.

31. Kazmers A, Jacobs L, Perkins A. Pul-monary embolism in Veterans AffairsMedical Centers: is vena cava interrup-

tion underutilized? Am Surg 1999; 65:11711175.32. Grassi C. Inferior vena caval filters:

analysis of five currently available de-vices. AJR 1991; 156:813821.

33. Becker D, Philbrick J, Selby J. Inferiorvena cava filters: indications, safety, ef-fectiveness. Arch Intern Med 1992; 152:19851994.

34. Velmahos G, Kern J, Chan L, et al.Prevention of venous thromboembo-lism after injury: an evidence-based re-portpart I: analysis of risk factorsand evaluation of the role of vena cavalfilters. J Trauma 2000;49:132138.

35. Velmahos G, Kern J, Chan L, et al.Prevention of venous thromboembo-lism after injury: an evidence-based re-portpart II: analysis of risk factors andevaluation of the role of vena cavalfilters. J Trauma 2000; 49:140144.

36. Streiff M. Vena caval filters: a com-prehensive review. Blood 2000;95:36693677.

37. Girard T, Philbrick J, Angle JF, BeckerD. Prophylactic vena cava filters fortrauma patients: a systematic review ofthe literature. Thromb Res 2003; 112:261267.

38. Kercher K, Sing R. Overview of cur-rent inferior vena cava filters. Am Surg

2003; 69:643648.

39. Kinney T. Update on inferior venacava filters. J Vasc Interv Radiol 2003;14:425440.

40. Ray C, Kaufman J. Complications ofinferior vena cava filters. Abdom Im-aging 1996; 21:368374.

41. Savin M, Panicker H, Sadiq S, et al.Placement of vena cava filters: factorsaffecting technical success and imme-diate complications. AJR 2002; 179:597602.

42. Decousus H, Leizorovicz A, Parent F,et al. A clinical trial of vena caval fil-ters in the prevention of pulmonaryembolism in patients with proximaldeep-vein thrombosis. Prevention duRisque dEmbolie Pulmonaire par In-terruption Cave Study Group. N EnglJ Med 1998; 338:409415.

43. White R, ZhouH, KimJ, et al. A pop-ulation-based study of the effective-

ness of inferior vena cava filter useamong patients with venous thrombo-embolism. Arch Intern Med 2000; 160:20332041.

44. Stein P, Hull R, Raskob G. Withhold-ing treatment in patients with acutepulmonary embolism who have a highrisk of bleeding and negative serialnoninvasive leg tests. Am J Med 2000;109:301306.

45. Linsenmaier U, Rieger J, SchenkF, et al.Indications, management, and compli-cations of temporary inferior vena cavafilters. Cardiovasc Intervent Radiol1998; 21:464469.

46. Millward S. Temporary and retriev-able inferior vena cava filters: currentstatus. J Vasc Interv Radiol 1998; 9:381387.

47. Reekers J, Hoogeveen Y, Wijnands M,et al. Evaluation of the Retrievabilityof the OptEase IVC Filter in an AnimalModel. J Vasc Interv Radiol 2004; 15:261267.

48. Pavcnik D, Uchida B, Keller F, et al.Retrievable IVC square stent filter: ex-perimental study. Cardiovasc InterventRadiol 1999;22:239245.

49. Neuerburg J, Handt S, Beckert K, et al.Percutaneous retrieval of the Tulipvena cava filter: feasibility, short- andlong-term changesan experimentalstudy in dogs. Cardiovasc InterventRadiol 2001; 24:418423.

50. Dibie A, Musset D, Heissler M, et al.In vivo evaluation of a retrievable venacava filterthe Dibie-Musset filter: ex-perimental results. Cardiovasc Inter-vent Radiol 1998; 21:151157.

51. de Gregorio M, Gimeno M, Tobio R, etal. Animal experience in the GuntherTulip retrievable inferior vena cava fil-ter. Cardiovasc Intervent Radiol 2001;24:413417.

52. Brountzos E, Kaufman J, Venbrux A, etal. A new optional vena cava filter:

retrieval at 12 weeks in an animal



11/11

model. J Vasc Interv Radiol 2003; 14:763772.

53. Vesely T, Krysl J, Smith S, et al. Pre-liminary investigation of the Irie infe-rior vena caval filter. J Vasc Interv Ra-diol 1996; 7:529535.

54. Watanabe S, Shimokawa S, MoriyamaY, et al. Clinical experience with tem-porary vena cava filters. J Vasc Surg2001; 35:285290.

55. Lorch H, Welger D, Wagner V, et al.Current practice of temporary venacava filter insertion: a multicenter reg-istry. J Vasc Interv Radiol 2000; 11:8388.

56. Millward S, Oliva V, Bell S, et al.Gunther Tulip retrievable vena cavafilter: results from the Registry of theCanadian Interventional RadiologyAssociation. J Vasc Interv Radiol 2001;12:10531058.

57. Asch M. Initial experience in humanswith a new retrievable inferior venacava filter. Radiology 2002; 225:835844.

58. Pieri S, Agresti P, Morucci M, et al.Optional vena cava filters: preliminaryexperience with a new vena cava filter.Radiol Med (Torino) 2003; 105:5662.

59. Hoff W, Hoey B, Wainwright G, et al.Early experience with retrievable infe-rior vena cava filters in high-risktrauma patients. J Am Coll Surg 2004;199:869874.

60. Morris C, Rogers F, Najarian K, et al.Current trends in vena caval filtrationwith the introduction of a retrievablefilter at a level I trauma center. JTrauma 2004; 57:3236.

61. Terhaar O, Lyon S, Given M, et al. Ex-tended interval for retrieval of gunthertulip filters. J Vasc Interv Radiol 2004;15:12571262.

62. PREPIC. Eight-year follow-up of pa-tients with permanent vena cava filtersin the prevention of pulmonary embo-lism: the PREPIC (Prevention du Ris-que dEmbolie Pulmonaire par Inter-ruption Cave) randomized study.Circulation 2005;112:416422.

63. Jacobs D, Sing R. The role of venacaval filters in the management of ve-

nous thromboembolism. Am Surg2003; 69:635642.

64. Pasquale M, Fabian T. Practice man-agement guidelines for trauma fromthe Eastern Association for the Surgeryof Trauma. J Trauma 1998; 44:941956.

65. Rogers F, Cipolle M, Velmahos G, et al.Practice management guidelines forthe prevention of venous thromboembolism in trauma patients: the EASTpractice management guidelines workgroup. J Trauma 2002; 53:142164.

66. Rohrer M, Scheidler M, Wheeler H, etal. Extended indications for place-ment of an inferior vena cava filter. J

Vasc Surg 1989; 10:4450.

67. Legere B, Dweik R, Arroglia A. Ve-nous thromboembolism in the inten-sive care unit. Clin Chest Med 1999;20:367384.

68. Sapala J, Wood M, Schuhknecht M, etal. Fatal pulmonary embolism after

bariatric operations for morbid obesity:a 24-year retrospective analysis. ObesSurg 2003; 13:819825.

69. Ferrell A, Byrne T, Robison J. Place-ment of inferior vena cava filters inbariatric surgical patientspossible in-dications and technical considerations.Obes Surg 2004; 14:738743.

70. Rosen M, Porter D, Kim D. Reassess-ment of vena cava filter use in patientswith cancer. J Vasc Interv Radiol 1994;5:501506.

71. Lee A. Treatment of venous thrombo-embolism in cancer patients. ThrombRes 2001; 102(suppl):V195V208.

72. Ansell J, Hirsh J, Poller L, et al. Thepharmacology and management of thevitamin K antagonists: the SeventhACCP Conference on Antithromboticand Thrombolytic Therapy. Chest2004; 126(suppl):204S233S.

73. Bates S, Greer I, Hirsh J, et al. Use ofantithrombotic agents during pregnan-cy: the Seventh ACCP Conference onAntithrombotic and ThrombolyticTherapy. Chest 2004; 126(suppl):627S644S.

74. Douketis J. Perioperative anticoagu-lation management in patients who arereceiving oral anticoagulant therapy: apractical guide for clinicians. ThrombRes 2003; 108:313.

75. Hirsh J, Raschke R. Heparin and low-molecular-weight heparin: the SeventhACCP Conference on Antithromboticand Thrombolytic Therapy. Chest2004; 126(suppl):188S203S.

76. Ananthasubramaniam K, Beattie J,RosmanH, et al. How safely and forhow long can warfarin therapy bewithheld in prosthetic heart valve pa-tients hospitalized with a major hem-orrhage? Chest 2001; 119:478484.

77. Martinelli I. Risk factors in venousthromboembolism. Thromb Haemost2001; 86:395403.

78. Jones B, Fink J. A prospective com-parison of the status of the deepvenous system after treatment with in-tracaval interruption versus anticoagu-lation. J Am Coll Surg 1994; 178:220222.

79. Ortega M, Gahtan V, Roberts A, et al.Efficacy of anticoagulation post-infe-rior vena caval filter placement. AmSurg 1998; 64:419423.

80. Monagle P, Chan A, Massicotte P, et al.Antithrombotic therapy in children:the Seventh ACCP Conference on An-tithrombotic and Thrombolytic Ther-apy. Chest 2004;126(suppl):645S687S.

81. Olin J. Pulmonary embolism. Rev

Cardiovasc Med 2002;3(suppl 2):S68S75.

82. Kearon C. Natural history of venousthromboembolism. Circulation 2003;107:I22I30.

83. Carson J, Kelley M, Duff A, et al. Theclinical course of pulmonary embo-lism. N Engl J Med 1992; 326:12401245.

84. Douketis J, Foster G, Crowther M, et al.Clinical risk factors and timing of re-current venous thromboembolism dur-ing the initial 3 months of anticoagu-lant therapy. Arch Intern Med 2000;160:34313436.

85. Kaufman J, Geller S, Rivitz S, et al.Operator errors during percutaneousplacement of vena cava filters. AJR1995; 165:12811287.

86. Binkert C, Bansal A, Gates J. Inferiorvena cava filter removal after 317-dayimplantation. J Vasc Interv Radiol2005; 16:395398.

87. Food and Drug Administration. Guid-ance for cardiovascular intravascular fil-ter 510(K) submissions. Rockville, MD:FDA, 1999. Available at: www.fda.gov/cdrh/ode/24.pdf.

88. Lorch H, Dallmann A, Zwaan M, et al.Efficacy of permanent and retrievablevena cava filters: experimental studiesand evaluation of a new device. Car-diovasc Intervent Radiol 2002; 25:193199.

89. Bruckheimer E, Judelman A, Bruckhei-mer S, et al. In vitro evaluation of aretrievable low-profile nitinol venacava filter. J Vasc Interv Radiol 2003;14:469474.

90. Xian Z, Roy S, Hosaka J, et al. Multi-ple emboli and filter function: an invitro comparison of three vena cavafilters. J Vasc Interv Radiol 1995; 6:887893.

91. Schroeder T, Elkins R, Greenfield L.Entrapment of sized emboli by theKMA-Greenfield intracaval filter. Sur-gery 1978; 83:435439.

92. Irie T, Yamauchi T, Makita K, et al.Retrievable IVC filter: preliminary in

vitro and in vivo evaluation. J VascInterv Radiol 1995; 6:449454.93. Rousseau H, Perreault P, Otal P, et al.

The 6-F nitinol TrapEase inferior venacava filter: results of a prospective mul-ticenter trial. J Vasc Interv Radiol 2001;12:299304.

94. Cho K, Greenfield L, Proctor M, et al.Evaluation of a new percutaneousstainless steel Greenfield filter. J VascInterv Radiol 1997; 8:181187.

95. Proctor M, Cho K, Greenfield L. De-velopment and evaluation of investiga-tional vena caval filters: the comple-mentary roles of in vitro and in vivo

studies. J Surg Res 2003; 110:241254.


guidelines filters

Documents